Dementia, depression and delirium in the very...

Dementia, depression and delirium in the very old Prevalences and associated factors

Johan Mathillas

Department of Community Medicine and

Rehabilitation, Geriatric Medicine

Umeå 2013

Till mormor, i åminnelse

Responsible publisher under swedish law: the Dean of the Medical Faculty

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7459-718-9

ISSN: 0346-6612

Elektronic version available at http://umu.diva-portal.org/

Tryck/Printed by: Print & Media, Umeå University, Umeå, Sweden 2013

Cover Illustration: At Eternity’s Gate by Vincent Willem Van Gogh. From Wikipedia, the free encyclopedia.

Table of contents

Abstract Abbreviations Notes for this thesis List of original papers Summary in Swedish Introduction 1 Demographics 1 Dementia 2 Epidemiology 3 Pathophysiology 3 Signs and symptoms 3 Prevention and treatment 6 Depressive disorders 6 Epidemiology 7 Pathogenesis 7 Signs and symptoms 9 Prevention and treatment 9 Delirium 13 Background 13 Epidemiology 13 Pathogenesis 15 Signs and symptoms 15 Prevention and treatment 17 Dementia, depression and delirium 18 Rationale for this thesis 20 Aims of this thesis 21 Specific aims 21 Methods 22 The Umeå85+/GErontological Regional DAtabase 22 Geography 22 Umeå 23 Dorotea, Malå, Storuman, Vilhelmina and Sorsele 23 Korsholm and Vasa 24 Participants 25 Contact and consent 26 Non-participation analysis 26 Procedure 33 The assessment procedure 33 The Mini Mental State Examination 33 The Geriatric Depression Scale 34

The Montgomery-Åsberg Depression Rating Scale 34 The Organic Brain Syndrome scale 34 The Philadelphia Center Geriatric Morale Scale 35

The Life Orientation Scale 35 The Barthel Activities of Daily Living Index 36 Other assessments 36 Diagnosis of dementia, depression and delirium 37 Statistics 38

Ethics 39 Results 40 Paper 1 40 Paper 2 44

Paper 3 49

Paper 4 53 General Findings 57 Discussion 59

Summary of results 59 Dementia 59 Prevalence 59 Increasing prevalence 60 Depressive disorders 62 Prevalences 62 Increasing prevalences 64 Associated factors in univariate analysis 65 Associated factors in multivariate analysis 66 Delirium 67 Prevalences 67 Other studies 67 Delirium in stroke, dementia and in old people with depression 68 The 3Ds 69 Interlinked pathogenesis 69 The Hypothalamus-Pituitary-Adrenal axis in dementia, depression and delirium 71 Methodological considerations 72 Clinical and social implications 75 Implications for future research 76 Conclusions 77 Acknowledgments 78 References 80 Papers 1-4 List of dissertations

Abstract

Emotional suffering in old age is largely caused by a variety of psychiatric disorders which are often, however, undertreated and underrecognized. This leads to reduced quality of life and functional status and increased morbidity and mortality.

Dementia, delirium and depression are common disorders in the very old, and are similar in several ways. All have multiple causes and are diagnosed by means of symptomatic criteria, are challenging to diagnose and difficult to differentiate from each other in the very old. They often coexist in the same individual, and may have common risk factors.

The overall aim of this thesis was to add to our knowledge about threats to successful aging, by investigating the prevalences of dementia, depression and delirium, and factors associated with these three disorders. Further aims were to measure change over time in the prevalence of dementia and depression, respectively, and to investigate the risk factors for incident depression.

This thesis uses the population-based Umeå85+/GErontological Regional DAtabase (GERDA) material. In 2000-2002, every other 85-year-old, all 90-year-olds and all people ≥95 years living in six municipalities in Västerbotten, Sweden were invited to participate, and in 2005-2007 the process was repeated, with the additional inclusion of the corresponding populations of two municipalities in Österbotten, Finland. A third data collection was carried out in 2010-2012. Trained assessors carried out assessments in the form of structured interviews during one or more home visits, recorded current medication and reviewed medical records obtained from general practitioners and hospitals.

In 2000-2002 the prevalence of dementia was 17% among 85-year-olds, 24% among 90-year-olds and 46% among those aged ≥95 years. In 2005-2007 dementia prevalence reached 28% among 85-year-olds, 40% among 90-year-olds and 45% among those aged ≥95 years. The prevalence of dementia in the total sample was 27% in 2000-2002 and 37% in 2005-2007 (p=0.001).

In 2000-2002 the prevalence of depressive disorders was 24% among 85-year-olds, 34% among 90-year-olds and 31% among those aged ≥95 years. In 2005-2007 the prevalence of depressive disorders was 33% among 85-year-olds, 39% among 90-year-olds and 38% among those aged ≥95 years. The prevalence od depressive disorders in the total sample was 29% in 2000-2002 and 37% in

2005-2007 (p=0.025). Among participants not depressed at baseline, 26% had developed depression after five years. Factors independently associated with new cases of depression at follow-up were hypertension, a history of stroke and a higher score on the 15-item Geriatric Depression Scale at baseline. The thirty-day prevalence of delirium in 2005-2007 was 17% among 85-year-olds, 21% among 90-year-olds and 39% among participants aged ≥95 years. Delirium prevalence among individuals with dementia was higher than among those without dementia (52% vs. 5%, p<.001). Factors independently associated with delirium superimposed on dementia in a multivariate logistic regression model were depression, heart failure, living in an institution and prescribed antipsychotics.

There was a high prevalence of dementia, depression and delirium in the papers comprising this thesis, and 55% had at least one of the three disorders. The prevalence of dementia and depression also increased between 2000-2002 and 2005-2007, after controlling for age and sex. Dementia and depression were important associated factors for delirium and half of those with dementia were depressed. The increasing age-specific prevalence of depression seems to be associated with the increasing age-specific prevalence of dementia.

The papers presented are among the first to report a significantly increasing age-specific prevalence of dementia and depression among very old people. More knowledge about associated factors and risk factors concerning these disorders may be helpful for carers and decision-makers, as well as providing reference values for studies in other regions. Further efforts are needed in both care and research to better prevent, screen for, diagnose and treat dementia, depression and delirium, especially considering the growing number of very old people.

Abbreviations

(the) 3Ds Dementia, depression and delirium

ACE Angiotensin Converting Enzyme

ADL Activities of Daily Living

ASA Acetylsalicylic Acid

ATC Anatomical Therapeutical Chemical (classification system)

BMI Body Mass Index

CASI-short Cognitive Abilities Screening Instrument, short version

CI Confidence Interval

CPRS Comprehensive Psychopathological Rating Scale

CT Computer Tomography

DBP Diastolic Blood Pressure

DSM-III Diagnostic and Statistical Manual of Mental Disorders,

third edition

DSM-III-R Diagnostic and Statistical Manual of Mental Disorders,

third edition- Revised

DSM- IV Diagnostic and Statistical Manual of Mental Disorders,

fourth edition

DSRS Dementia Severity Rating Scale

FLAIR FLuid Attenuation Inversion Recovery

GDS-15 Geriatric Depression Scale, 15-item version

GERDA GErontological Regional Database

HPA Hypothalamic-Pituitary-Adrenal (axis)

IBM International Business Machines (Corporation)

LOS Life Orientation Scale

MADRS Montgomery-Åsberg Depression Rating Scale

mmHg Millimetre of mercury

MMSE Mini Mental State Examination

MNA Mini Nutritional Assessment

MONICA Multinational mONItoring of trends and determinants in CArdiovascular disease

MRI Magnetic Resonance Imaging

NMDA N-Methyl-D-Aspartate

NSAID Non-Steroidal Anti-Inflammatory Drug

OBS Organic Brain Syndrome (scale)

OR Odds Ratio

OSAS Obstructive Sleep Apnea Syndrome

PGCMS Philadelphia Geriatric Center Morale Scale

SBP Systolic Blood Pressure

SD Standard Deviation

SPE Standardized Psychiatric Inventory

SPSS Statistical Product and Service Solutions

WMH White Matter Hyperintensity

Notes for this thesis

Tables, figures and boxes are numbered x-y where x refers to the paper they belong to and y their number in this paper. If x=0 the table, figure or box belongs to the introduction.

Original papers

1. Mathillas J, Lövheim H, Gustafson Y. Increasing prevalence of dementia among very old people. Age and Ageing. 2011; 40(2):243-249.

2. Mathillas J, Olofsson B, Lövheim H, Allard P, Gustafson Y. Prevalence of depressive disorders among very old people in 2000-2002 and 2005-2007; the Umeå 85+/GERDA study. In manuscript.

3. Petersson S, Mathillas J, Wallin K, Olofsson B, Allard P, Gustafson Y. Risk factors for depression in very old age: a population-based cohort study with a five-year follow-up. In manuscript.

4. Mathillas J, Olofsson B, Lövheim H, Gustafson Y. Thirty-day prevalence of delirium among very old people: a population-based study of very old people living at home and in institutions. Archives of Gerontology and Geriatrics. 2013; 57(3):298-304.

Papers are reprinted with the kind permission of the respective pulisher.

Svensk sammanfattning

(Summary in Swedish)

I takt med att antalet och andelen mycket gamla personer ökar i Sverige, kommer också sjukdomar som är vanliga i dessa åldersgrupper öka. Då det över tid kan förekomma förändringar bland faktorer kopplade till sådana sjukdomar kan det även vara så att den åldersspecifika förekomsten av sjukdomar vanliga bland mycket gamla personer förändras.

Demens, depression och delirium är vanliga psykiska störningar hos mycket gamla personer, och ger ofta kraftig försämring av funktionsförmåga och livskvalitet. De tre förekommer ofta tillsammans, och kan ha gemensamma riskfaktorer.

Forskning på demens, depression och delirium är sparsamt utförd bland mycket gamla människor, vilket leder till att det finns begränsat med kunskap om tillstånden i dessa åldersgrupper. Många studier är också gjorda på speciella populationer såsom personer på äldreboenden eller sjukhus.

Syftet för denna avhandling var att studera förekomsterna av demens, depression och delirium samt associerade faktorer för dessa tre tillstånd. För demens och depression var även målet att undersöka om den åldersspecifika förekomsten av dessa två tillstånd ökade eller minskade över tid. Ytterligare ett mål var att undersöka riskfaktorer för att utveckla depression i mycket hög ålder.

Denna avhandling baseras på Umeå85+/GERDA (GERontologisk Regional Databas) – materialet, en befolkningsbaserad databas från Botniaregionen. Under 2000-2002 bjöds varannan 85-åring, alla 90-åringar och alla personer 95 år och äldre boende i Umeå, Storuman, Malå, Sorsele, Vilhelmina och Dorotea in att delta, varefter strukturerade datainsamlingar gjordes i deltagarnas hem, kompletterat med uppgifter från anhöriga, vårdpersonal och sjukvårdsjournaler. Under 2005-2007 och 2010-2012 upprepades processen, dessa gånger med ytterligare inklusion av 2 kommuner i Österbotten, Finland (Vasa och Korsholm).

Under 2000-2002 var förekomsten av demens (alla former) 17 % bland 85-åringar, 24 % bland 90-åringar och 46 % bland ≥95-åringar. Under 2005-2007 var förekomsten 28 % bland 85-åringar, 40 % bland 90-åringar och 45 % bland

≥95-åringar. I hela den undersökta befolkningen ökade förekomsten av demens från 27 % till 37 % (p=0,001).

Under 2000-2002 var förekomsten av depression 24 % bland 85-åringar, 34 % bland 90-åringar och 31 % bland ≥95-åringar. Under 2005-2007 var förekomsten 33 % bland 85-åringar, 39 % bland 90-åringar och 38 % bland ≥95-åringar. I hela den undersökta befolkningen ökade förekomsten av depression från 29 % till 37 % (p=0,025). Över fem år utvecklade en fjärdedel av deltagarna depressioner. Faktorer kopplade till att ha utvecklat depression efter fem år var hypertoni, tidigare stroke och poäng på Geriatric Depression Scale (GDS-15) vid den första undersökningen.

En fjärdedel av deltagarna hade delirium eller hade haft delirium under de närmast föregående trettio dagarna, med signifikant högre förekomster hos deltagare med demenssjukdom. Hos de med delirium men utan demens hade merparten depressioner. Oberoende faktorer kopplade till att ha delirium bland personer med demenssjukdom var depression, hjärtsvikt, boende på sjukhem eller äldreboende och behandling med antipsykotiska läkemedel (neuroleptika).

Över hälften av deltagarna hade minst en av de tre sjukdomarna. Demens och depression samförekom i hög utsträckning med varandra och delirium återfanns nästan uteslutande hos personer med åtminstone en av de två andra diagnoserna.

Fynden från denna avhandling antyder att den åldersspecifika förekomsten av demens och depression ökar bland mycket gamla. Att mäta förekomster av dessa sjukdomar samt undersöka kopplade faktorer och riskfaktorer för dessa sjukdomar bidrar med jämförelsedata för andra liknande studier, kan hjälpa till att lägga grund för ytterligare kohortstudier samt interventionsstudier där man provar att behandla associerade faktorer. Fynden kan även vara av vikt vid planering för framtida vårdinsatser. Då det verkar som att befolkningsökningen bland mycket gamla personer, både i absoluta och relativa tal kommer att fortgå blir dessa kunskaper än viktigare.

1

Introduction

Demographics

During the early part of the new millennium countries worldwide will have to cope with a substantial aging among their populations, a process expected to continue in the future (1). As a result of falling fertility and increasing life expectancy, the proportion of older to younger age groups will increase. In absolute numbers, the global population aged ≥80 years is projected to increase by 270% between 2011 and 2050 (2). The growth rate will be even higher in older age groups. The term very old has several different definitions in terms of years, but in this thesis it will be used mainly to refer to those aged ≥85 years.

Sweden has seen declining rates and better outcomes for infectious diseases since the end of the 18th century. Since around 1940, there has also been declining mortality from chronic disease (3), more recently accompanied by a reduced incidence of and increased survival after coronary events such as myocardial infarction (4).

The effects of these developments are not certain. One theory is that as the impact of fatal disease is reduced, more people will live close to the maximum life span physiologically possible for humans. Further advances in healthcare combined with improved living standards will compress the period lived with disease and disability to the final years of life because of later disease onset. This idea of compression of morbidity and disability was introduced in 1980 (5).

Others argue that despite increase in life expectancy, disease incidence rates will not decrease, resulting in higher disease prevalence rates and disability levels in the population (6). For example, between 1947 and 1957, the Lundby study found a doubling of dementia prevalence among individuals aged >60 years. An increasing survival among those with dementia was also found in 1957 compared to 1947. As this coincides with the advent of antibiotics, it may have been due to a lower mortality from infections (6). More recent research has also found an increased proportion of very old people surviving with chronic disease (7), as well as increased survival after ischemic heart disease (4) and stroke (8).

Ultimately, because of the increasing number of very old people, any disease or disorder common among the very old will increase in total health impact, unless prevention and treatment advances even more rapidly.

2

This thesis is on the subject of dementia, depression and delirium; three common psychiatric disorders with negative effects on quality of life and ability levels. They are often diagnosed on symptomatic criteria, and may be difficult to differentiate in a clinical situation. All criteria for dementia, depressive disorders and delirium presented are adapted from the Diagnostic and Statistical Manual of psychiatric disorders, fourth edition (DSM-IV) (9), and are presented in Boxes 0-6.

Dementia

Dementia, from the Latin word demens (de- (without) and mens (mind)), meaning madness, denotes a group of disorders, all of which are the final consequence of an even larger number of mechanisms of brain injury, including different types of neurodegeneration and vascular lesions. The common denominator is the development of chronic and progressive multiple cognitive deficits, always including memory impairment. Of the existing subtypes, Alzheimer’s dementia is regarded as the most common followed by vascular dementia (10). Other less common types include Lewy body dementia, frontal lobe dementia and dementia due to Parkinson’s disease or alcohol.

Reported risk factors for dementia are old age, female gender (11), hypertension (12) and stroke (13). Mean blood pressure (14) and stroke mortality (8) have declined over time in Sweden, possibly affecting dementia incidence and prevalence.

Disease burden is a measure of the impact a disease makes in the form of disability or premature death. In high-income countries such as Sweden, dementia is reported as the fourth largest cause of burden of disease, surpassed only by ischemic heart disease, cerebrovascular disease and depressive disorders (15). On an individual level, dementia is associated with a significantly increased mortality (16) reduced quality of life and impaired activities of daily living (ADL) functionality (17).

3

Epidemiology

In 2010 the number of worldwide cases of dementia was estimated to be approximately 35.6 million people, a figure expected to double every 20 years (18). Table 0-1 shows prevalence studies with age groups matching those in Paper 1. A few studies have also investigated age-specific variation of dementia rates over time (23-26). These studies will be presented in the discussion.

Pathophysiology

Dementia is a cognitive disorder chiefly affecting memory and constitutes the final result of various mechanisms of injury to the brain. In vascular dementia the mechanism is cerebrovascular insults while in in the case of Alzheimer’s disease the mechanism is unknown but accompanied by deposition of neurofibrillary tangles and amyloid plaques and neuronal death. Despite being separate entities in diagnostic manuals such as the DSM-IV, those with dementia aged ≥85 years are unlikely to have Alzheimer’s dementia or vascular dementia in their pure forms (27). Instead, this patient category on autopsy often has lesions typical of both diseases.

Signs and symptoms

Dementia, in addition to impaired memory, presents with aphasia, apraxia, agnosia or disturbances in executive functioning in any combination, and the deficits are serious enough to affect occupation or social function. A higher level of functioning must have been present before dementia onset to warrant the diagnosis (28). Boxes 1 and 2 present the DSM-IV criteria for Alzheimer’s dementia and vascular dementia.

The memory impairment is either forgetting material that was already learned or difficulty in forming new memories. Recognition of close friends and relatives or orientation in familiar environments may become impaired in advanced dementia. Sleep disturbance, suspicion of others, delusions, anxiousness, hallucinations (primarily visual), mood disturbances, aggressive behaviour, slurred speech, aphasia or unnecessary repetition of words and phrases as well as motor or gait disturbances leading to falls are often present (28).

4

5

Box 1. DSM-IV Diagnostic criteria for Dementia of the Alzheimer’s type

A. the development of multiple cognitive deficits manifested by both

(1) memory impairment

(2) one or more of the following cognitive disturbances:

(a) aphasia

(b) apraxia

(c) agnosia

(d) disturbance in executive functioning

B. The cognitive deficits in Criteria A1 and A2 each cause significant impairment in social or occupational functioning and represent a significant decline from a previous level of functioning.

C. The course is characterized by gradual onset and continuing cognitive decline.

D. The cognitive deficits in Criteria A1 and A2 are not due to any of the following:

(1) other central nervous system conditions that cause progressive deficits in memory and condition

(2) systemic conditions that are known to cause dementia

(3) substance-induced conditions

E. The disturbances do not occur exclusively during the course of a delirium

F. The disturbance is not better accounted for by another axis 1 disorder

Box 2. DSM-IV Diagnostic criteria for Vascular Dementia

A. the development of multiple cognitive deficits manifested by both

(1) memory impairment

(2) one or more of the following cognitive disturbances:

(a) aphasia

(b) apraxia

(c) agnosia

(d) disturbance in executive functioning

B. The cognitive deficits in Criteria A1 and A2 each cause significant impairment in social or occupational functioning and represent a significant decline from a previous level of functioning.

C. Focal neurological signs and symptoms or laboratory evidence indicative of cerebrovascular disease that are judged to be etiologically related to the disturbance.

D. The cognitive deficits in Criteria A1 and A2 are not due to any of the following:

(1) other central nervous system conditions that cause progressive deficits in memory and condition

(2) systemic conditions that are known to cause dementia

(3) substance-induced conditions

E. The disturbances do not occur exclusively during the course of a delirium

F. The disturbance is not better accounted for by another axis 1 disorder

6

Prevention and treatment

Although the disorder is irreversible, both pharmacological (29) and nonpharmacological (30) treatments may improve function and wellbeing. Management goals depend on disorder severity, and choice of treatment depends on the symptoms and functional limitations of the patient. For example, physical exercise has been shown to improve depressive symptoms, physical function and quality of life in old people with dementia (31), and cognitive function may be improved with cholinesterase inhibitors (32), sometimes with addition of a NMDA receptor antagonist, i.e. Memantine (33). Pharmacological treatment should be continuously evaluated (34). Early dementia treatment should focus on improving or stabilizing cognitive ability and mood, maintaining or re-establishing independence and promoting autonomy, as well as educating and supporting carers (35). As the disorder progresses, there should be more focus on facilitating mental and physical stimulation, managing behavioural disturbances, and providing practical care for patients and respite for non-professional carers.

Depressive Disorders

Depressive disorders are not simply passing blue moods, but are characterized by persistent and all-pervasive sadness, lack of interest and pleasure in activities. As with dementia and delirium, the cause is probably multifactorial, with both psychosocial and physiological components.

The condition “melancholia” was described by Hippocrates as a condition of “aversion to food, despondency, sleeplessness, irritability and restlessness”. Over the centuries several other synonyms for or disorders related to depression have been described, such as acedia, tristitia and ennui (36).

Depression is the third leading cause of disease burden worldwide, the largest cause of disease burden in middle- and high-income countries and is expected to be the most important global cause of disease burden in 2030 (15). While there is a general lack of studies depression in the very old is reported as underdiagnosed (37), showing low rates of remission (38) poor response (39) or poor long-term response to treatment (40). It is also disabling, (40) reduces quality of life and increases mortality (41, 42).

Since depression in late life has been associated with cerebrovascular risk factors such as hypertension (43) and stroke (44) and the prevalence of and

7

survival after these may have changed over time, depression prevalence may also change over time. Given the momentous impact depression has on society and healthcare, information about such a prevalence change over time would be useful.

Epidemiology

In psychiatric research in general, the very old are underrepresented (40) and the study of depression is no exception, with few prevalence studies existing. Table 0-2 shows studies with at least one age group matching the age groups in Paper 2. Depression incidence has been reported to be higher in the very old than in the younger old (45, 46).

Reported nursing and residential home depression rates range from 19.9% to 40% (47, 48).

Conflicting results have been reported regarding change over time in depression prevalence and incidence (49-51). This will be treated further in the discussion.

Pathogenesis

The cause of major depression is probably heterogeneous- it is not certain that there is a single pathway to the disorder (56). Genetic, neurochemical and psychosocial factors may interact. Depression, like dementia, may be considered a chain of effects with an uncertain starting point, with psychosocial factors, neurochemistry and other factors all influencing each other. For example, the monoamine deficiency in depression targeted by antidepressant medication may be a downstream effect of other neurochemical deficits (56).

Reported factors associated with depression among the very old include ADL impairment, (57, 58) loneliness, (37) and preceding cognitive decline (59). Depression has been shown to be both highly prevalent after stroke and also associated with the presence of cerebrovascular risk factors (44). This, along with findings of silent cerebral infarctions (60) and white matter lesions in the depressed, has given rise to a vascular depression hypothesis, stating that vascular-based damage to the brain is a contributor to age-related depressive symptoms (61).

8

9

Signs and symptoms

Since the very old represent a select group of individuals who have survived disease and other causes of death for a number of contributory reasons, such as heredity and lifestyle, factors associated with depression in this group may differ from younger age groups. The clinical expression of depression in the very old is also different from that in younger adults, with less sadness and more somatisation, apathy and sub-threshold depressive symptoms that do not meet the criteria for major depressive disorder (62-64). Boxes 3-7 present the diagnostic criteria for the different depressive disorders included in the depression “umbrella diagnosis” in this thesis.


Accepted treatments among adults are pharmacotherapy with antidepressants and psychotherapy, with supportive measures such as light therapy in the case of seasonal affective disorder (65). Physical exercise reportedly also has some positive effect on depressive symptoms (66). For adults with severe depression electroconvulsive therapy in combination with antidepressants is regarded should be considered (67). In older populations and with increasing comorbidity, there is less evidence of the effect of both psychotherapy (68) and antidepressants (69). Further studies are also needed on the effect of electroconvulsive therapy in the elderly (70), although this therapy is often used in clinical practice with good anecdotal effect (71). There is inconclusive evidence of antidepressant effect in elderly nursing home residents (72), and when there is concomitant dementia effect is uncertain (73, 74). In the very old, there are too few high-quality studies of antidepressant response (75) or the effectiveness of psychotherapy (68) to give a recommendation on treatment (40), and while exercise seems to have an effect among community-dwelling elderly, there is limited knowledge of effect on the very old, and there seems to be no effect among people living in residential care facilities (76).

10

Box 3. DSM-IV Diagnostic criteria for major depressive disorder

A. Five (or more) of the following symptoms have been present during the same two-week period and represent a change from previous functioning; at least one of the first two symptoms is present.

1. Depressed mood most of the day, nearly every day, as indicated by either subjective report or observation made by others.

2. Markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day.

3. Significant weight loss when not dieting or weight gain, or decrease or increase in appetite nearly every day.

4. Insomnia or hypersomnia nearly every day.

5. Psychomotor agitation or retardation nearly every day.

6. Fatigue or loss of energy nearly every day.

7. Feelings of worthlessness or excessive or inappropriate guilt nearly every day.

8. Diminished ability to think or concentrate, or indecisiveness, nearly every day.

9. Recurrent thoughts of death, recurrent suicidal ideation, or a suicide attempt or a specific plan for committing suicide.

B. The symptoms do not meet the criteria for a mixed episode.

C. The symptoms cause clinically significant distress or impairment in social, occupational or other important areas of functioning.

D. The symptoms are not due to the direct physiological effects of a substance or a general medical condition.

E. The symptoms are not better accounted for by bereavement.

Box 4. DSM-IV Diagnostic criteria for minor depressive disorder

A. Two to four of the following symptoms have been present during the same two-week period and represent a change from previous functioning; at least one of the first two symptoms is present.

1. Depressed mood most of the day, nearly every day, as indicated by either subjective report or observation made by others.

2. Markedly diminished interest or pleasure in all, or almost all, activities most of the day, nearly every day.

3. Significant weight loss when not dieting or weight gain, or decrease or increase in appetite nearly every day.

4. Insomnia or hypersomnia nearly every day.

5. Psychomotor agitation or retardation nearly every day.

6. Fatigue or loss of energy nearly every day.

7. Feelings of worthlessness or excessive or inappropriate guilt nearly every day.

8. Diminished ability to think or concentrate, or indecisiveness, nearly every day.

9. Recurrent thoughts of death, recurrent suicidal ideation, or a suicide attempt or a specific plan for committing suicide.

B. There has never been a major depressive episode and criteria are not met for dysthymic disorder.

C. The symptoms cause clinically significant distress or impairment in social, occupational or other important areas of functioning.

D. The symptoms are not due to the direct physiological effects of a substance or a general medical condition.

E. The symptoms are not better accounted for by bereavement.

F. There has never been a mixed, manic or hypomanic episode and criteria are not met for cyclothymic disorder

E. The mood disturbance does not occur exclusively during schizophrenia, schizophreniform disorder, delusional disorder or psychotic disorder not otherwise specified.

11

Box 5. DSM-IV Diagnostic criteria for mood disorder due to general medical condition

A. A prominent and persistent disturbance in mood predominates in the clinical picture and is characterized by either or both of the following:

(1) Depressed mood or markedly diminished interest or pleasure in all, or almost all activities

(2) Elevated, expansive, or irritable mood

B. The symptoms do not meet the criteria for a mixed episode.

C. The disturbance is not better accounted for by another mental disorder

D. The disturbance does not occur exclusively during the course of a delirium.

E. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

Box 6. DSM-IV Diagnostic criteria for dysthymic disorder

A. Depressed mood, for most of the day, for more days than not, as indicated either by subjective account or observation by others, for at least two years.

B. Presence, while depressed, of two or more of the following:

(1) poor appetite or overeating

(2) insomnia or hypersomnia

(3) low energy or fatigue

(4) low self-esteem

(5) poor concentration or difficulty making decisions

(6) feelings of hopelessness

C. During the two-year period of the disturbance, the person has never been without the symptoms in criteria A and B for more than two months at a time.

D. No major depressive episode has been present during the first two years of the disturbance.

E. There has never been a Manic Episode, a Mixed Episode, or a hypomanic episode, and criteria have never been met for cyclothymic disorder.

F. The disturbance does not occur exclusively during the course of a chronic Psychotic Disorder, such as Schizophrenia or Delusional Disorder.

G. The symptoms are not due to the direct physiological effects of a substance or a general medical condition

H. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

12

Box 7. DSM-IV Diagnostic criteria for Substance-Induced Mood Disorder

A. A prominent and persistent disturbance in mood predominates in the clinical picture and is characterized by either (or both) of the following:

(1) depressed mood or markedly diminished interest or pleasure in all, or almost all, activities

(2) elevated, expansive or irritable mood

B. There is evidence from the history, physical examination, or laboratory findings of either (1) or (2):

(1) The symptoms in Criterion A developed during, or within a month of, Substance Intoxication or Withdrawal

(2) medication use is etiologically related to the disturbance

C. The disturbance is not better accounted for by a Mood Disorder that is not substance induced. Evidence that the symptoms are better accounted for by a Mood Disorder that is not substance induced might include the following: the symptoms precede the onset of the substance use (or medication use); the symptoms persist for a substantial period of time (e.g., about a month) after the cessation of acute withdrawal or severe intoxication or are substantially in excess of what would be expected given the type or amount of the substance used or the duration of use; or there is other evidence that suggests the existence of an independent non-substance-induced Mood Disorder

D. The disturbance does not occur exclusively during the course of a delirium.

E. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

13

Delirium

Delirium is a disorder characterized by an acute impairment of attention and cognition, and a fluctuating course. It is by definition caused by one or more precipitating insults, such as infection or intoxication. Because reliable biomarkers are lacking, diagnosis is primarily clinical.

Background

“Phrenitis” was an acute disorder, usually associated with fever, featuring cognitive and behavioural disturbances as well as disruption of sleep. The opposite condition was “lethargus” with symptoms such as sleepiness, listlessness, inertia, memory loss and dulling of the senses (77). These two states were described by Hippocrates in 500 B.C. and correspond well to the hyperactive and hypoactive forms of delirium described below. Celsus used the term delirium in the first century as a synonym for Phrenitis. The disorder has more recently had several synonyms, such as acute confusional state, toxic encephalopathy, toxic confusion and acute organic brain syndrome (78).

Epidemiology

Table 0-3 shows population-based studies or studies including participants with and without dementia. Few studies have been made on a population-based sample containing very old people with and/or without dementia living independently and in institutions.

Delirium prevalence in general hospital settings is reported as being 14-53% (83, 84) and postoperative delirium prevalence as 24-44% (85, 86) among elderly patients. Institutional prevalence has been reported as 47% (87).

14

15

Pathogenesis

Delirium, or acute brain failure, occurs as a result of predisposing factors and precipitating factors exceeding the cognitive reserve capabilities in the individual. Although the definitive pathogenesis of delirium remains unclear, several mechanisms have been proposed, including neurotransmitter imbalance, more specifically dopamine excess and acetylcholine deficiency, dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis (88) and disturbance of the sleep-wake cycle, reduced cerebral blood flow and metabolism or the effects of increased cytokine production on cerebral function, i.e. inflammation (89). Two or more of these pathways may interact, especially considering that several subtypes of delirium have been described.

Predisposing factors for delirium include increasing age, history of ischemic strokes, cognitive impairment, dementia and depressive disorders while precipitating factors for delirium include surgical procedures, infections, trauma and drugs (90-93). However, associations in a population-based sample of very old people with dementia have not been extensively studied.

Signs and symptoms

Delirium is characterized by the presence of disturbances of awareness and attention, alongside with other neuropsychiatric symptoms. Onset is acute or subacute and the course is fluctuating (80). Box 8 presents the DSM-IV criteria for delirium.

16

Box 8. DSM-IV Diagnostic criteria for Delirium

A. Disturbance of consciousness with reduced ability to focus, sustain or shift attention.

B. A change in cognition or the development of a perceptual disturbance that is not better accounted for by a pre-existing, established or evolving dementia.

C. The disturbance develops over a short period of time (usually hours to days) and tends to fluctuate over the course of the day.

AND EITHER:

D. There is evidence from the history, physical examination, or laboratory findings that the disturbance is caused by the direct physiological consequences of a general medical condition

OR:

D. There is evidence from the history, physical examination, or laboratory findings of either (1) or (2):

(1) the symptoms in criteria A and B developed during Substance Intoxication

(2) medication use is etiologically related to the disturbance

OR:

D: There is evidence from the history, physical examination, or laboratory findings that the symptoms in Criteria A and B developed, during, or shortly after, a withdrawal syndrome.

OR:

D: There is evidence from the history, physical examination, or laboratory findings that the delirium has more than one etiology.

17

Hyperactive and hypoactive delirium

Zbigniew Lipowski classified delirium into hyperactive, hypoactive or mixed subtypes. Hyperactive delirium is characterized by restlessness, agitation and hyperalertness, and hypoactive delirium by lethargy, drowsiness and psychomotor retardation (94). Mixed delirium combines symptoms from both.

Emotional and psychotic delirium

Delirium has also been subclassified into subtypes with primarily emotional or psychotic symptoms, or a mix of the two types (95). Emotional and psychotic deliria may require different medical interventions. While the hyperactive-hypoactive subclassification is based on motoric activity level the emotional-psychotic subclassification is based on the psychological symptoms experienced by the individual, allowing for different combinations of the two to exist, but this has not yet been studied.


Some multi-approach interventions have been successful, with outcomes including reduced delirium incidence and severity, decreased injury rate and reduced length of hospital stay (96-99). When delirium is manifested, the identification and treatment of the precipitating cause or causes should be the first priority. When the precipitating factors have been treated it is important to create the best possible prerequisites for the brain to recover, through for example review of current medication. There is little evidence relating to the treatment of delirium symptoms with psychotropic drugs since few high-quality studies have been published (100).

18

Dementia, depression and delirium

Dementia is common in very old age. Depression often co-occurs with both Alzheimer’s and vascular dementia (101, 102), but depression has also been reported as a risk factor for (103-105) or as a prodromal stage of (59, 106) dementia. Others have hypothesized dementia as a risk factor for depression (106, 107). Some of the depressions co-occurring with dementia are probably untreated or recurrent early-onset depressions, shown to be associated with loss of volume of the hippocampus (108), an important part of the brain for consolidating memories, which could be a link to dementia. Subcortical ischemic lesions are more common among those elderly with depression than among similarly aged individuals without depression (109), and these changes are also common in dementia.

Both dementia (110) and depression (111) are associated with increased risk of delirium. Delirium might also be a risk factor for dementia in the very old (112).

In clinical practice, the three disorders are often difficult to differentiate, and special attention should be given to onset, course and medical history. Table 0-4 presents key clinical features for the three disorders.

19

Table 0-4. A comparison of the clinical features of delirium, dementia and depression.

Clinical feature Dementia Depression Delirium

Onset Slow, over months/years

Varies Acute

Course Progressive deterioration, lifelong

Variable, weeks to years

Fluctuating, days to weeks

Consciousness Clear except in severe cases

Clear Altered

Attention Initially normal Generally normal Impaired, fluctuates

Orientation Normal or impaired Selective disorientation

Generally impaired

Memory Recent and remote impaired

Selective impairment Recent and immediate impaired

Thinking Difficulty with abstraction, thoughts impoverished

Intact with themes of hopelessness and helplessness

Disorganized, incoherent

Perception Misperceptions usually absent

Intact Impaired: illusions, hallucinations and delusions

Psychomotor behaviour Generally normal Variable Variable: hypokinetic, hyperkinetic, mixed

Assessment Struggles with assessment to find appropriate reply

Generally lacks motivation, frequent “don’t know” answers

Distracted from task, numerous errors

Sleep May be disturbed, individual pattern

Disturbed, early morning awakening or hypersomnia

Disturbed but no set pattern

20

Rationale for this thesis

In order to manage an aging population in the best possible way, it is important to add to our knowledge about the threats to successful aging. In the study of psychiatric disorders, very old people are a largely unexplored area of research. Population-based prevalence studies focusing on the very old are also rare. Thus there is a lack of research performed on a reciprocally associated group of disorders which have a great impact on health, quality of life and functionality in a part of the population which is growing rapidly. Dementia, depression and delirium are three disorders common among the very old that have serious consequences for the individual, and knowledge about prevalences and prevalence trends for these disorders would be of value in planning health care and services for the elderly, while knowledge of associated factors would be valuable in longitudinal and intervention studies. If the age-specific prevalences for these disorders were to change, it would have a multiplicative effect on the impact of these disorders. Further, dementia, depression and delirium may vary regarding risk factors, associated factors and prevalence both in different age groups but also in different parts of Sweden, and studies of the three disorders among the very old in Västerbotten are scarce, if not completely lacking. Contributing to an increase in the knowledge about associated factors and risk factors for these disorders may improve knowledge and awareness of them among carers and decision-makers, as well as provide reference values for studies in other regions, and material for further longitudinal and intervention studies.

21

Aims of this thesis

The overall aim of this thesis was to investigate the prevalences of dementia, depression and delirium, and factors associated with these three disorders among the very old. Further, aims were to measure change over time in the prevalence of depression and dementia, and to investigate risk factors for developing depression among the very old.

Specific aims

Paper 1

To evaluate whether the age-specific prevalence of dementia among the very old changed over time in two cross-sectional population-based samples five years apart.

Paper 2

To evaluate whether the age-specific prevalence of depression in a sample of very old people changed over time in two cross-sectional population-based samples five years apart, and investigate factors associated with any change in depression prevalence.

Paper 3

To investigate risk factors associated with the development of depression in very old age.

Paper 4

To measure the thirty-day prevalence of delirium in a population-based sample of very old individuals and to study factors associated with delirium in people with and without dementia.

22

Methods

The Umeå85+/GErontological Regional DAtabase

This thesis uses the population-based Umeå85+/GErontological Regional DAtabase (GERDA) material, a demographic cohort study of very old people started in 2000 (44, 113, 114). The purpose of the database is to investigate factors associated with successful aging and threats to successful aging in a representative sample of very old people in northern Sweden and Finland. The participants were interviewed and assessed in their homes, either in the community or in an institution. The Umeå 85+/GERDA material now encompasses over 2500 in-depth quantitative interviews and additional qualitative interviews. It also includes over 15000 postal surveys including all municipalities in Västerbotten and Österbotten collected since the beginning of the year 2000.

Geography

Six municipalities in Sweden (Umeå, Dorotea, Malå, Storuman, Vilhemina and Sorsele, Figure 0-1) and two in Finland (Vasa and Korsholm, Figure 0-2) were included in the GERDA database. The Swedish municipalities are located in the county of Västerbotten in the north of Sweden, and Vasa and Korsholm are located in Österbotten, the county closest to Västerbotten on the Finnish side of the Gulf of Bothnia.

23

Umeå

The municipality of Umeå includes the largest city in the county and in 2011 had a population of 116 500, with a mean age of 38.3 years. The municipality is classed as an urban area (50.3 inhabitants/km2) (115).

Dorotea, Malå, Storuman, Vilhelmina and Sorsele

These are rural municipalities (0.4-2.0 inhabitants/km2) making up the northwestern part of Västerbotten county. The total population in 2011 was 21 900, with a mean age of 45.7 years (115).

Figure 0-1. The county of Västerbotten. Shaded municipalities were included in

Umeå85+/GERDA. Map adapted from Wikipedia Commons.

24

Korsholm and Vasa

These Finnish municipalities are part of the county of Österbotten which is separated from Västerbotten by the Baltic Sea. Vasa is the central city of the county and the municipality had a total of 60 900 inhabitants in 2012 (180 inhabitants/km2)(116, 117). Korsholm, a more rural municipality surrounding Vasa, had 19 000 inhabitants the same year (22 inhabitants/km2)(116, 117). Vasa has a mainly Finnish-speaking and Korsholm a mainly Swedish-speaking population.

Figure 0-2. The county of Österbotten. Shaded municipalities were included

in Umeå85+/GERDA. Map adapted from Arttu Pahrlati, areal data from www.stat.fi

25

Participants

Data collections were made in Umeå in 2000-2001 (the first urban sample), 2005-2006 (second urban sample) and 2010-2011 (third urban sample), in Dorotea, Malå, Storuman, Vilhelmina and Sorsele in 2002 (first rural sample), 2007 (second rural sample) and 2012 (third rural sample) and in Vasa and Korsholm in 2005-2006 and 2010-2011. The data collections included half of the 85-year-olds, all the 90- year-olds and all those aged ≥95 years living in the above designated areas. The names, addresses and civil registration numbers were collected from the National Tax Board in Sweden and the Finnish population register. The 85-year-olds were alternately assigned to participation or non-participation from a randomised starting point in the lists. No additional inclusion or exclusion criteria were set. Any survivors from an earlier data collection were included again in the subsequent data collection in the appropriate age group if they were still living in the catchment areas. The selection procedure and inclusion criteria were identical for all data collections.

Papers 1 and 2 include the participants from Umeå assessed in 2000-2001 and 2005-2006 and the participants from Dorotea, Malå, Storuman, Vilhelmina and Sorsele assessed in 2002 and 2007. Papers 1 and 2 are cross-sectional studies with two measuring points five years apart.

Paper 3 includes the participants from Umeå assessed in 2000-2001 and 2005-2006, previously assessed participants from Umeå who were assessed again in 2010-2011, participants from Dorotea, Malå, Storuman, Vilhelmina and Sorsele assessed in 2002 and assessed participants from these areas previously assessed in 2007. Paper 3 is a longitudinal study. The very few individuals participating in all three Umeå cross-sections were excluded from the last cross-section, since they already had a baseline and follow-up assessment.

Paper 4 includes the participants from Umeå assessed in 2005-2006, from Dorotea, Malå, Storuman, Vilhelmina and Sorsele assessed in 2007, and from Finland assessed in 2005-2006. Paper 4 is a cross-sectional study.

Figure 0-3 shows the cohorts of the Umeå85+/GERDA study, with footnotes concerning which samples were used in which paper.

26

Contact and consent

The eligible participants were contacted by letter where information about the study was given and again a few weeks later by telephone so that they could give their informed consent to participation. If they were living in institutions the caring staff was asked to evaluate the respondent’s cognitive capabilities and informed consent was then obtained either from the participant or, in the case of significant cognitive impairment, first from their next of kin and then the participant. In some cases, only partial consent was given; for example consent to access medical records and to interview caring staff and relatives.

Non-participation analysis

Paper 1

Out of 527 people, 44 (8.3%) died before being asked to participate in 2000-2002 and 47 out of 610 (7.7%) in 2005-2007 (p=0.690) (Figure 1-1). Fifty-two people (10.8%) declined to participate in 2000-2002 and 97 (17.2%) in 2005-2007 (p=0.003). Sex did not differ significantly between the decliner groups. When decliners are compared to participants, a significantly lower proportion of the former were female in 2000-2002, but their mean age did not differ

Figure 0-3. The Umeå85+/GERDA cohorts. 1= included in paper 1. 2= included in paper 2.

3= included in paper 3. 4= included in paper 4.

27

significantly, while in 2005-2007, the mean age was significantly lower when compared to those who agreed to participate, but the proportion of females did not differ significantly.

Among those who agreed to participate, 1 (0.3%) participant in the first urban sample and 1 (0.3%) in the later urban sample lacked sufficient data for a decision to be made whether or not the participant fulfilled the DSM-IV criteria for dementia. These individuals were excluded from the study. All persons in the rural sample had sufficient data for dementia assessment.

Among those asked to participate, participation rate was 89.0% in 2000-2002 and 82.6% in 2005-2007.

Paper 2

Out of 527 people 45 (8.5%) died before being approached in 2000-2002 and 47 out of 610 (7.7%) in 2005-2007 (p=0.607) (Figure 2-1). Of the participants approached in 2000-2002, 63 (13.1%) persons declined to participate and of the participants approached in 2005-2007 97 persons declined to participate (17.2%) (p=0.061). In 2000-2002, a significantly lower proportion of decliners were female, but their mean age did not differ significantly when compared to participants. In 2005-2007, the mean age of the decliners was significantly lower when compared to those who agreed to participate, but the proportion of females did not differ significantly. While the second cross-section had a significantly higher proportion of female decliners than the first (p=0.016), mean age did not differ between the two decliner groups.

Among those who agreed to participate, 21 (9.7%) participants in the first urban sample, 12 (8.6%) in the first rural sample, one (0.3%) in the second urban sample and none (0%) in the second rural sample lacked sufficient data for a decision to be made whether or not the participant fulfilled the DSM-IV criteria for depressive disorders. These individuals were excluded from the study.

Among those asked to participate, participation rate was 80.0% in 2000-2002 and 83.6% in 2005-2007.

28

Paper 3

Of the total number of persons selected (n=831), 78 individuals died prior to being approached, 121 declined to participate and 65 could not be evaluated for depression due to missing data, which left a final baseline sample of 567 participants. At follow-up five years later, 339 had died and one had moved. Of the 227 asked to participate in the follow-up, 20 declined to participate and one could not be evaluated for depression, leaving a final follow-up sample of 206 individuals (Figure 3-1). Of these, 49 had baseline depression and 157 did not. Non-participants did not significantly differ from participants with regard to age or sex. Those who died before the follow-up had a lower mean Mini Mental State Examination (MMSE) score (p<0.001), a lower mean Barthel ADL index score (p<0.001), lower mean systolic and diastolic blood pressure (p<0.001 and p=0.007) and a higher mean age (p<0.001). They were also more likely to have depression (p=0.002) or dementia (p<0.001). Among those asked to participate, participation rate was 74.5%.

Paper 4

Out of 963 eligible persons, 77 persons (8.0%) died before being approached (Figure 4-1). Of the participants approached, 177 (20.0%) declined to participate. Decliners were more often female (p=0.035) but their mean age did not differ from that of the participants.

Among those who agreed to participate, three (0.4%) participants could not be assessed for delirium due to lack of data. These individuals were excluded from the study. Participation rate, calculated from those asked to participate, was 79.9%.

29

Figure 1-1. Flow chart of participants in paper 1.

30


31


32


33

Procedure

The assessment procedure

Structured interviews were conducted with the participants including, but not limited to, several different assessment scales for measuring various aspects of health. Trained assessors- physicians, nurses, physical therapists or medical students- performed the assessments during one or more home visits. Questions about the participant's current health status, social data, and medical history were included in the interviews. The investigators also reviewed medical records from general practitioners and hospitals, recorded current medication, and collected social and medical information from relatives and carers. The oldest participants were assessed first.

Investigators attempted to assess all the participants, even those with impaired cognition, using all the scales, as long as the participants were deemed to be able to comprehend what they were being asked. All scales were interviewer-administered. Depression scales were used directly with the participants and not filled in by proxy. The Organic Brain Syndrome (OBS) scale (which registers symptoms of organic brain disease over the preceding month) and Montgomery-Åsberg Depression Rating Scale (MADRS) registered the assessor’s clinical impression of participants in addition to direct questions. Further, the OBS scale also registered information about the participant obtained from carers or relatives.

If carers or relatives were available, they provided information for the ADL scales if the assessor thought more reliable answers would be gained that way. There is no data on the proportion of information collected by proxy.

The Mini Mental State Examination

The MMSE was used to assess cognitive function (19, 20). This scale was developed by Folstein et al. as a screening test to quantitatively assess the level of cognitive impairment in individuals and to document cognitive changes that occur over time. The score ranges from 0 to 30 with a higher score indicating better cognition.

34

The Geriatric Depression Scale

The Geriatric Depression Scale, 15-item version (GDS-15) was used to screen for depression (118). The scale consists of 15 yes-no questions, and is designed to avoid bias from somatic ailments. The scale ranges from 0 to 15 with a higher score indicating increased risk of depression. The scale has been found to have high sensitivity and specificity for detecting depression in very old people (119), and has been shown to be a useful instrument for assessing depression in individuals with MMSE scores ≥10 (120, 121). The Swedish Council on Health Technology Assessment recommends use of the GDS-15 for depression screening among the elderly (122).

The Montgomery-Åsberg Depression Rating Scale

The 30-point version of the MADRS (123) was used to assess level of depressive symptoms. The scale is designed to assess severity of depression and is commonly used to evaluate the effects of antidepressant treatment. The scale was applied on a subsample of participants. In the first urban sample, the participants' GDS-15 score determined whether they would be assessed using the MADRS. A specialist in geriatric medicine assessed individuals with a GDS-15 score of ≥ 5 using the MADRS during a follow-up visit approximately two weeks after the first visit. In all the later assessments, the MADRS was used during the initial visit if the assessor was a physician or medical student trained to use the scale, irrespective of GDS-15 score. A higher score on the MADRS indicates more severe symptoms of depression.

When comparing the scale to other depression rating scales in post-stroke populations, the MADRS compared favourably (124) and another study reported good validity for the scale (125) in this setting.

The Organic Brain Syndrome Scale

The OBS scale was developed to assess symptoms that appear in cases of delirium, dementia and other organic brain disorders (21, 22). The scale is subdivided into OBS 1, which is a questionnaire measuring the individual’s awareness and orientation to their own data, and OBS 2 which is based on observation of the person and describes a wide spectrum of psychopathology, such as suspiciousness, emotional reactions, delusions, hallucinations, as well as disturbances in speech, spatial orientation, recognition, physical and

35

practical abilities, and variations in the person’s clinical state. In the Umeå 85+/GERDA study, only OBS 2 was used since OBS 1 provides similar information to that obtained from the MMSE. The scale was used to register symptoms experienced during the month prior to assessment by means of observation and interviews with the participants, carers and relatives.

The OBS scale played a central role in providing information for a delirium diagnosis. The scale is not based on the operationalization of any particular DSM version but registers symptoms and fluctuations of symptoms that may be used to establish a diagnosis according to the assessor’s choice of DSM version. The OBS scale applied on DSM-IV criteria has been validated in thoracic surgery patients against the Confusion Assessment Method (126) with high levels of agreement (127). Regarding validity, the OBS scale has been reported to satisfy the requirements set up by the Scientific Advisory Commitee of the Medical Outcomes Trust (128, 129). The assessors were trained in the use of the OBS scale but there was no inter-assessor validation during the data collection period.

The Philadelphia Geriatric Center Morale Scale

Participants were assessed using the Philadelphia Geriatric Center Morale Scale (PGCMS) (130). This scale is designed to measure subjective or psychological wellbeing in the old and is graded from zero to seventeen with higher scores indicating greater wellbeing. The British Geriatric Society and the Royal College of Physicians have recommended the scale for assessment of subjective wellbeing among old people (131).

The Life Orientation Scale

A modified version of the Life Orientation Scale (LOS) (132) was used to assess personal values and positive life orientation.

36

The Barthel Activities of Daily Living Index

The Barthel ADL Index was used to assess dependency in ADL. (133, 134) The maximum score is 20, indicating total independence in personal ADL (135). The scale was completed by proxy if the participant had significant cognitive impairment.

Other assessments

Assessors measured height and weight and calculated the Body Mass Index (BMI) as kg/m2.

Feelings of loneliness were assessed by asking the question “Do you ever feel lonely?”

Unimpaired vision was defined as being able to read a 4-5 mm font from normal reading distance with or without glasses. Unimpaired hearing was defined as being able to hear a conversation of normal loudness one meter away with or without hearing aids.

The number of social visits participants made and received in a normal week were registered, based on information obtained from the participants, relatives, and/or carers.

Prescribed drugs were registered based on information from the participant and medical records. The participant was also asked to present medication lists and/or containers for prescription and non-prescription drugs taken for confirmation. The Anatomical Therapeutical Chemical (ATC) classification system was used to classify prescribed drugs. (136)

Blood pressure was measured manually with a stethoscope and a sphygmanometer calibrated against a mercury manometer. A single measurement was taken in the supine position after five minutes of rest. A minority of participants who could not lie down sat instead during measurement.

Hypertension was regarded as present if blood pressure was ≥160/95 or if the person had a hypertension diagnosis in their medical records and current treatment against hypertension.

37

The participants were considered to have undergone heart surgery if they had had any surgery performed on the heart (such as valve surgery) or coronary arteries (such as coronary artery bypass graft), including any endovascular interventions (such as percutaneous coronary intervention).

Diagnosis of dementia, depression and delirium

An experienced specialist in geriatric medicine established the diagnoses according to the criteria set out in the DSM-IV (28) by carefully reviewing and evaluating the collected material including medical history and assessment scales (the MADRS, OBS, GDS-15, PGCMS and LOS for depression and the OBS and MMSE for dementia). The diagnostic procedure was the same for all participants in all cohorts in both Sweden and Finland. For both dementia and depression diagnoses, the overall clinical picture from the respective sources was considered. Criteria for having a major depressive episode, dysthymic disorder, minor depression or depression due to general medical condition had to be fulfilled for depressive disorder to be diagnosed, and criteria for any DSM-IV dementia diagnosis had to be met for a dementia diagnosis. A subclassification of Alzheimer’s dementia and vascular dementia was also made based on the criteria in the DSM-IV, or based on established diagnoses in medical records.

Treatment with antidepressants on an explicit depressive disorder indication noted in medical records was considered as a current depressive disorder, and if the person had a GDS-15 score <5 they were regarded as having responded to the treatment.

During collection of the first sample, assessors referred a few uncertain dementia cases for a complete dementia assessment at the Geriatric Centre in Umeå, but no diagnosis was changed as a result.

Other disorders and diseases were established by the same specialist in geriatric medicine based on information from participants, carers/relatives and medical records.

38

Statistics

All papers

Differences in proportions between groups were analysed using the Pearson Chi-square test, or 2-sided Fisher exact test if the expected cell count was below 5. Differences in the means of continuous variables between groups were analysed using the independent samples t-test. A p-value of <0.05 was regarded as statistically significant. The International Business Machines (IBM) Statistical Product and Service Solutions (SPSS) version 19 for Mac was used for data analysis.

To adjust for the skewed selection of participants adjusted prevalences of dementia, depression and delirium for people aged 85 years or older were calculated. First the respective prevalence for ages between the selected age groups was extrapolated, assuming that the disorders increased linearly between the sampled age groups. Then an adjusted total prevalence was calculated based on the background population statistics for the same geographical area.

Paper 1

A logistic regression analysis was performed, including age and sex, to control for different case mix between the two cross-sectional samples.

Paper 2

To investigate whether depressive disorder prevalence changed between the cross-sections adjusted for confounders, variables that both differed between cross-sections (a p-value of ≤0.15 in table 2-1) and were associated with depressive disorders (a p-value of ≤0.15 in both positions in table 2-2) were included, together with age, sex, and a dichotomous variable representing participation in either the 2000-2002 or the 2005-2007 sample, in a binary multivariate logistic regression analysis. One variable, MMSE score, was excluded in favor of the dementia variable due to high correlation (>0.4 Spearman rank correlation coefficient). The dementia variable was chosen over the MMSE variable since it could be argued that the MMSE score decreased as a direct effect of depression while dementia is a separate diagnosis.

39

Paper 3

Variables associated with incident depression (p-value <0.15) in Table 3-2, as well as age and sex, were included in a manual stepwise multivariate logistic regression analysis. Only the significant variables along with age and sex were included in the final regression model (Table 3-3).

Paper 4

A logistic regression analysis was performed on the subsample of participants with dementia. Factors associated with delirium in earlier studies were included in the regression; these can be found in Table 4-3. Any variables in this table with a p-value <0.15 in univariate analysis were included in the regression. The least significant variables were then removed stepwise to obtain the final model with only significant variables.

Ethics

The Regional Ethical Review Board in Umeå (§99-326 and §05-063M) and the Ethics Committee of Vaasa Central Hospital approved the study (05-87).

40

Results

Paper 1

In 2000-2002 it was possible to assess 430 individuals (71% women) with a mean age of 89.5 years for dementia. In 2005-2007 the corresponding figures were 465 individuals (70% women) with a mean age of 90.2 years.

The prevalence of dementia was 17.4% in 2000-2002 and 28.2% in 2005-2007 in the 85-year-old group (p=0.017). The corresponding figures for 90-year-olds were 24.0% and 39.8% (p=0.003), and for ≥95-year-olds 46.1% vs. 45.0% (p=0.874)(Table 1-2). The prevalence of dementia in the total sample was 26.5% in 2000-2002 and 37.2% in 2005-2007 (p=0.001)(Table 1-1). The MMSE score was also significantly lower in the later sample.

Controlling for differences in age and sex between the samples using logistic regression, the prevalence of dementia increased between the two samples (odds ratio 1.587, 95% confidence interval (CI) 1.185-2.127, p=0.002).

In univariate analysis, more people had had heart surgery and were treated with ß-blockers, calcium channel blockers, ACE (Angiotensin Converting Enzyme) inhibitors, antilipemic agents and cholinesterase inhibitors in the later sample (Table 1-1).

In 2005-2007 significantly more of the 85-year-olds had had heart surgery and were treated with anti-hypertensive medication such as ACE-inhibitors and ß-blockers compared to 2000-2002 (Table 1-2). There was also an increase in the use of antilipemic agents in this age group. The later sample of 85-year-olds also had significantly lower systolic blood pressure and MMSE scores.

The 2005-2007 group of 90-year-olds had lower MMSE scores, higher proportion treated with ß-blockers and cholinesterase inhibitors compared to 2000-2002 but their blood pressure was not significantly lower than that of the earlier cross-section.

The 2005-2007 ≥95-year-old group showed no significant difference in MMSE score but a larger proportion with heart diseases and a larger proportion of participants treated with ß-blockers, diuretics and calcium channel blockers when compared to the earlier cross-section.

41

Among women in the total sample, prevalence was 30.9% in the first cross-section, and 41.1% in the second (p=0.008). Among men in the total sample, the prevalence of dementia was 19.5% in 2000-2002 and 28.1% in 2005-2007 (p=0.106).

When divided geographically, the urban municipality had a dementia prevalence of 28.6% and 39.9% five years later (p=0.004) while the prevalence in the rural areas was 26% in the first cross-section and 32% in the second (p=0.166).

The proportion of people with Alzheimer’s disease to vascular dementia was 47 to 26 (64.4%) in the first and 84 to 45 (65.1%) in the subsequent cross-section (p=0.917).

Additional Results

Adjusting for the sampling method, the prevalences in the two cross-sections would correspond to a prevalence of dementia of 23% in 2000-2002 and of 35% in 2005-2007 among ≥85-year-olds.

42

Table 1-1. Characteristics of participants in Paper 1.

2000-2002 2005-2007 N=430

n (%) N=465 n (%)

p-value

Age group 85 178 (41.4) 163 (35.1) 90 150 (34.9) 171 (36.8) ≥95 102 (23.7) 131 (28.2)

Dementia 114 (26.5) 173 (37.2) 0.001 Women 307 (71.4) 326 (70.1) 0.672 Living alone 338 (86.2) 360 (82.0) 0.098 Living in an institution 168 (39.1) 181 (41.0) 0.571 Heart surgery 9 (2.1) 22 (4.7) 0.031 Stroke 81 (18.8) 103 (22.2) 0.220 Heart disease 246 (57.3) 294 (63.4) 0.066 Diabetes 51 (11.9) 62 (13.3) 0.507 Warfarine 11 (2.6) 14 (3.0) 0.681 ASA 158 (36.7) 192 (41.3) 0.164 Diuretics 188 (43.7) 230 (49.5) 0.085 ß-blockers 82 (19.1) 178 (38.3) <0.001 Calcium channel blockers 37 (8.6) 59 (12.7) 0.049 ACE inhibitors 41 (9.5) 76 (16.3) 0.003 Antilipemic agents 1 (0.2) 23 (4.9) <0.001 Antidepressants 66 (15.3) 82 (17.6) 0.358 Antipsychotics 61 (14.2) 64 (13.8) 0.855 Benzodiazepines 114 (26.5) 136 (29.2) 0.362 Cholinesterase inhibitors 8 (1.9) 24 (5.2) 0.011 NSAIDs 36 (8.4) 37 (8.0) 0.821

Mean ±SD Mean ±SD Number of medications 5.99 ±4.7 6.65 ±4.1 0.026 Barthel’s ADL Index score 15.6 ±6.3 15.6 ±6.0 0.870 GDS-15 score 3.8 ±2.6 3.7 ±2.7 0.648 SBP (mmHg) 147.7 ±24.1 144.4 ±23.3 0.058 BMI (kg/m2) 24.7 ±4.7 24.8 ±4.1 0.773 Years in school 6.2 ±1.9 6.8 ±2.0 <0.001 MMSE score 21.6 ±8.0 19.8 ±7.8 0.002

Individual variables may have lower denominators than the dementia variable. Abbreviations: Acetylsalicylic acid (ASA), Angiotensin converting enzyme (ACE), Non steroidal anti inflammatory drug (NSAID), Standard Deviation (SD), Activities of daily living (ADL), Geriatric Depression Scale, 15 item version (GDS-15), Systolic Blood Pressure (SBP), Body Mass Index (BMI), Mini Mental State Examination (MMSE).

43

44

Paper 2

Depressive disorder prevalence was 24.0% in 2000-2002 and 33.1% in 2005-2007 among the 85-year-olds (p=0.060), 33.8% in 2000-2002 and 38.6% in 2005-2007 among the 90-year-olds (p=0.321) and 31.2% in 2000-2002 and 38.2% in 2005-2007 among the people aged ≥95 years (p=0.308). The prevalence of depressive disorders in the total sample was 29.3% (95% CI 24.7-33.8%) in 2000-2002 and 36.6% (95% CI 32.2-41.0%) in 2005-2007 (p=0.025). The mean GDS-15 score for the total sample was 3.8±2.6 in 2000-2002 and 3.7 ±2.7 in 2005-2007 (p=0.65). The mean GDS-15 score among individuals with depressive disorders was 6.2±2.8 in 2000-2002 and 6.0±2.7 in 2005-2007 (p=0.580).

Among men, depressive disorder rates were 21/108 (19.4%) in 2000-2002 and 44/139 (31.7%) in 2005-2007 (p=0.031). Among women, depressive disorder rates were 92/278 (33.1%) in 2000-2002 and 126/326 (38.7%) in 2005-2007 (p=0.156).

In the urban subsample, depressive disorder rates were 72/258 (27.9%) in 2000-2002 and 122/306 (39.9%) in 2005-2007 (p=0.003). In the rural subsample, depressive disorder rates were 41/128 (32.0%) in 2000-2002 and 48/159 (30.2%) in 2005-2007 (p=0.737).

Of the individuals with a depressive disorder diagnosis in 2000-2002, 65/111 (58.6%) were being treated with antidepressants, compared to 79/169 (46.7%) in 2005-2007 (p=0.053). Of the individuals with completed GDS-15 scales in this subsample, 23/45 (51.1%) had responded to treatment (GDS-15<5) in 2000-2002 and 26/48 (54.2%) in 2005-2007 (p=0.768).

Among individuals with dementia, depressive disorder prevalence was 46/106 (43.4%) in 2000-2002 and 89/173 (51.4%) in 2005-2007 (p=0.236), while among individuals without dementia, depressive disorder prevalence was 67/290 (23.1%) in 2000-2002 and 81/292 (27.7%) in 2005-2007 (p=0.298).

The MMSE scores of the total sample were 21.9±7.7 in 2000-2002 and 19.8±7.8 in 2005-2007 (p<0.001).

Factors associated with depressive disorders in the two samples are presented in Table 2-2. In both samples, dementia, institutional living, feeling lonely, impaired vision, constipation, low MMSE scores, low Barthel scores, and high GDS-15 scores were associated with depressive disorders in the univariate analysis. In addition, the variables female sex, loss of a child, and heart failure

45

were associated with depressive disorders in 2000-2002 and living alone and having a history of stroke in 2005-2007.

In the multiple logistic regression analysis, dementia, participation year, and living alone were significantly associated with depressive disorders, as presented in Table 2-3.

Additional results

Adjusting for the oversampling of persons aged ≥95 years, population depressive disorder prevalence among people aged ≥85 years, calculated from Paper 2, would be 28% in 2000-2002 and 36% in 2005-2007.

46

Table 2-1. Characteristics of participants in Paper 2.

2000-2002 2005-2007 N=386

n (%) N =465 n (%)

p-value

Age, mean ±SD 89.6 ±4.61 90.2 ±4.38 0.053 Age group

85 90 ≥95

154 (39.9) 139 (36.0) 93 (24.1)

163 (35.1) 171 (36.8) 131 (28.2)

0.177

Depressive disorders 85 90 ≥95

113 (29.3) 37 (24.0) 47 (33.8) 29 (31.2)

170 (36.6) 54 (33.1) 66 (38.6) 50 (38.2)

0.025 0.060 0.321 0.308

Women 278 (72.0) 326 (70.1) 0.732 Living in an institution 158 (40.9) 181 (41.0) 0.996 Living alone 324 (86.6) 360 (82.0) 0.072 Have children 326 (84.9) 379 (85.2) 0.913 Dead children 61 (16.9) 80 (20.0) 0.279 Feeling lonely 150 (49.5) 175 (53.5) 0.314 History of hip fracture 77 (19.9) 87 (18.8) 0.659 Heart failure 99 (25.7) 131 (28.2) 0.411 History of stroke 76 (19.7) 103 (22.2) 0.380 Dementia 106 (27.5) 173 (37.2) 0.002 Constipation 155 (40.3) 202 (43.5) 0.336 Impaired vision 83 (23.2) 80 (21.2) 0.509 Impaired hearing 75 (20.5) 81 (20.8) 0.694 Analgesics 216 (60.5) 299 (64.6) 0.231 Antidepressants 66 (17.6) 82 (17.7) 0.967 Benzodiazepines 111 (29.6) 136 (29.4) 0.943 Antipsychotics 58 (16.2) 64 (14.0) 0.381

Mean ±SD Mean ±SD BMI (kg/m2) 24.83 ±4.60 24.84 ±4.19 0.982 MMSE score 21.71 ±7.84 19.79 ±7.81 0.001 Barthel’s ADL Index score 15.76 ±6.20 15.57 ±6.00 0.699 GDS-15 score 3.75 ±2.63 3.66 ±2.67 0.648 Number of visits each week 3.00 ±2.99 2.74 ±2.38 0.243

Individual variables may have lower denominators than the depressive disorders variable. Variables in the tables were predefined and either factors known to be associated with depression from the literature, deemed important in describing the sample or hypothesized to be associated with depression. Abbreviations: Standard Deviation (SD), Body Mass Index (BMI), Mini Mental State Examination (MMSE), Activities of daily living (ADL), Geriatric Depression Scale, 15 item version (GDS-15)

47

Table 2-2. Characteristics of participants in Paper 2 and association with depression. 2000-2002 2005-2007

Depressive disorder Depressive disorder Yes

N=113 n (%)

No N=273 n (%)

p-value

Yes N=170 n (%)

No N=295 n (%)

p-value

Age, mean ±SD 90.01 ±4.43 89.43 ±4.67 0.263 90.54 ±4.44 90.00 ±4.35 0.29 Age group

85 90 ≥95

37 (32.7) 47 (41.6) 29 (25.7)

117 (42.9) 92 (33.7) 64 (23.4)

0.166 54 (31.8) 66 (38.8) 50 (29.4)

109 (36.9) 105 (35.6) 81 (27.5)

0.528

Women 92 (81.4) 186 (68.1) 0.008 126 (74.1) 200 (67.7) 0.152 Living in an institution 63 (55.8) 95 (34.8) <0.001 93 (56.0) 88 (31.9) <0.001 Living alone 101 (91.8) 223 (84.5) 0.057 145 (88.4) 215 (78.2) 0.007 Have children 97 (86.6) 229 (84.2) 0.548 140 (85.4) 239 (85.1) 0.929 Dead children 25 (24.0) 36 (14.1) 0.022 30 (19.9) 50 (20.1) 0.959 Feeling lonely 55 (68.8) 95 (42.6) <0.001 84 (70.6) 91 (43.8) <0.001 History of hip fracture 25 (22.1) 52 (19.0) 0.491 33 (19.4) 54 (18.4) 0.781 Heart failure 38 (34.9) 61 (21.8) 0.022 53 (31.2) 73 (24.9) 0.144 History of stroke 26 (23.0) 50 (18.3) 0.291 55 (32.4) 50 (16.9) <0.001 Dementia 46 (40.7) 60 (22.0) <0.001 89 (52.4) 84 (28.5) <0.001 Constipation 67 (59.3) 88 (32.4) <0.001 100 (58.8) 102 (34.7) <0.001 Impaired vision 34 (33.3) 49 (19.2) 0.004 40 (27.6) 40 (17.2) 0.017 Impaired hearing 24 (23.1) 51 (19.5) 0.440 39 (25.7) 42 (17.7) 0.06 Analgesics 72 (27.3) 45 (40.5) 0.011 92 (66.7) 102 (41.3) <0.001 Antidepressants 65 (57.5) 1 (0.4) <0.001 79 (46.5) 3 (1.0) <0.001 Benzodiazepines 47 (42.3) 64 (24.2) <0.001 65 (38.2) 71 (24.1) 0.001 Antipsychotics 25 (22.5) 36 (13.6) 0.010 37 (22.2) 27 (9.3) <0.001

Mean ±SD Mean ±SD Mean ±SD Mean ±SD BMI (kg/m2) 24.86 ±4.94 24.82 ±4.47 0.944 25.07 ±4.56 24.69 ±3.84 0.418 MMSE score 19.03 ±8.30 22.70 ±7.44 <0.001 17.59 ±7.93 21.14 ±7.43 <0.001 Barthel’s ADL Index score 13.63 ±6.75 16.61 ±5.75 <0.001 13.64 ±6.42 16.82 ±5.37 <0.001

GDS-15 score 6.18 ±2.76 2.90 ±1.97 <0.001 5.96 ±2.71 2.29 ±1.41 <0.001 Number of visits each week 2.58 ±3.05 3.14 ±2.96 0.170 2.50 ±2.55 2.87 ±2.27 0.197

Individual variables may have lower denominators than the depressive disorders variable. Variables in the tables were predefined and either factors known to be associated with depression from the literature, deemed important in describing the sample or hypothesized to be associated with depression. Abbreviations: Standard Deviation (SD), Body Mass Index (BMI), Mini Mental State Examination (MMSE), Activities of daily living (ADL), Geriatric Depression Scale, 15 item version (GDS-15).

48

Table 2-3. Logistic regression of factors associated with the increased prevalence of depressive disorders in Paper 2.

Odds ratio 95% CI p-value

Female sex 1.32 0.93-1.88 0.12 Age 1.00 0.96-1.03 0.89 Living alone 1.65 1.02-2.66 0.040 Dementia 2.47 1.87-3.54 <0.001 2000-2002 or 2005-2007 1.37 1.01-1.86 0.044

Confounding factors included in the model were those associated with depressive disorders that also increased between the cross-sections. Nagelkerke R square = 0.088. Abbreviations: Confidence Interval (CI)

49

Paper 3

Sociodemographic factors and clinical characteristics at baseline are presented in Table 3-1. Of the 567 participants, 183 (32%) were diagnosed with depression at baseline (Figure 3-1). The age-specific prevalence of depression was 29.5% for the 85-year-olds, 34.6% for the 90-year-olds and 34.4% for those aged ≥95 years. Of the participants with baseline depression, 55.6% were treated with antidepressants (Table 3-1). Of the 384 participants not depressed at baseline, 157 were possible to evaluate for depression at follow-up, and of those 40 (25.5%) had developed depression. Of the non-depressed aged 85 years, 90 years and ≥95 years at baseline who could be evaluated for depression at follow-up, 24.7% (24/97), 23.4% (11/47) and 38.5% (5/13) respectively had developed depression after five years. There were no significant differences in mean age (p=0.489) or sex (p=0.851) between those who developed depression and those who did not. Of the 49 participants with baseline depression who were evaluated for depression at follow-up, 38 (77.6%) were still depressed after five years. Of those depressed at baseline 68.9% died before the follow-up, compared with 55.5% (p=0.002) of the non-depressed at baseline (Figure 3-1). Baseline characteristics univariately associated with new cases of depression five years later are presented in Table 3-2. Thirteen variables were significantly associated with depression at baseline (3-1), but not with incident depression after five years (Table 3-2). Poor self-rated health and the GDS-15 score were the only factors significantly associated with both baseline depression and new cases of depression at follow-up.

In the final multivariate logistic regression analysis, presented in Table 3-3, the only baseline factors independently associated with the incident depressive disorders after five years were the GDS-15 score at baseline (p=0.007), hypertension (p=0.034) and a history of stroke (p=0.031).

50

Table 3-1. Characteristics of participants with or without depressive disorders at baseline in paper 3.

Total sample

(N=567) n (%)

Depressive disorder (N=183)

n (%)

No depressive disorder (N=384)

n (%)

p-value

Sociodemographic factors: Age, mean ±SD 89.5 ±4.5 89.8 ±4.4 89.3 ±4.5 0.272 Women 399 (70.4) 138 (75.4) 261 (68.0) 0.070 Years of education, mean ±SD 6.5 ±2.1 6.5 ±2.2 6.5 ±2.1 0.897 Living alone 451 (91.1) 160 (93.6) 291 (89.8) 0.163 Living in an institution 222 (39.4) 100 (54.6) 122 (32.1) <0.001 Feeling lonely 220 (50.1) 90 (70.3) 130 (41.8) <0.001 3 or more visits per week 227 (54.3) 49 (40.5) 178 (59.9) <0.001 Poor self-rated health 168 (40.3) 71 (58.2) 97 (32.9) <0.001 Clinical characteristics: Antidepressants 102 (18.4) 100 (55.6) 2 (0.5) <0.001 Analgesics, ASA excluded 272 (49.0) 115 (63.9) 157 (41.9) <0.001 Anxiolytics 37 (6.7) 21 (11.7) 16 (4.3) 0.001 Neuroleptics 87 (15.7) 45 (25.0) 42 (11.2) <0.001 Hypnotics 175 (31.5) 74 (41.1) 101 (26.9) 0.001 History of stroke 108 (19.0) 42 (23.0) 66 (17.2) 0.102 Dementia 173 (30.5) 83 (45.4) 90 (23.4) <0.001 Delirium during preceding month 116 (20.5) 66 (36.1) 50 (13.0) <0.001 Diabetes Mellitus 70 (12.3) 22 (12.0) 48 (12.5) 0.871 Hypertension 324 (57.1) 101 (55.2) 223 (58.1) 0.517 Heart failure 150 (26.5) 62 (33.9) 88 (23.0) 0.006 History of myocardial infarction 49 (8.6) 18 (9.8) 31 (8.1) 0.485 History of hip fracture 117 (20.6) 45 (24.6) 72 (18.8) 0.108 History of thyroid disease 66 (11.8) 25 (13.9) 41 (10.8) 0.283 Impaired vision 100 (19.6) 44 (27.2) 56 (16.0) 0.003 Impaired hearing 102 (19.5) 36 (21.4) 66 (18.5) 0.436

Mean ±SD Mean ±SD Mean ±SD BMI (kg/m2) 24.8 ±4.5 24.8 ±4.9 24.8 ±4.3 0.987 SBP (mmHg) 147.1 ±23.9 143.7 ±22.3 148.7 ±24.4 0.031 DBP (mmHg) 74.6 ±10.9 73.6 ±10.6 75.0 ±11.0 0.166 MNA score 23.1 ±4.6 21.2 ±4.7 24.1 ±4.2 <0.001 MMSE score 21.4 ±8.0 18.8 ±8.5 22.6 ±7.4 <0.001 Barthel's ADL index score 15.8 ±6.1 13.7 ±6.8 16.9 ±5.5 <0.001 GDS-15 score 3.7 ±2.6 6.0 ±2.6 2.7 ±1.9 <0.001 Number of medications 6.7 ±4.4 8.8 ±4.7 5.7 ±3.9 <0.001

Individual variables may have lower denominators than the depressive disorders variable. Abbreviations: Standard Deviation (SD), Acetylsalicylic acid (ASA), Body Mass Index (BMI), Systolic Blood Pressure (SBP), Diastolic Blood Pressure (DBP), Mini Nutritional Assessment (MNA), Mini Mental State Examination (MMSE), Activities of daily living (ADL), Geriatric Depression Scale, 15 item version (GDS-15)

51

Table 3-2. Baseline characteristics associated with new cases of depressive disorders at five-year follow-up in paper 3.

New cases of depressive disorder

(N=40) n (%)

Still no depressive disorder (N=117)

n (%)

p-value

Sociodemographic factors: Age, mean ±SD 88.0 ±4.1 87.6 ±3.2 0.489 Women 29 (72.5) 83 (70.9) 0.851 Years of education, mean ±SD 6.7 ±2.0 6.5 ±1.9 0.757 Living alone 29 (90.6) 87 (84.5) 0.562 Living in an institution 6 (15.0) 16 (13.7) 0.835 Feeling lonely 17 (45.9) 45 (40.5) 0.564 3 or more visits per week 22 (62.9) 66 (60.0) 0.763 Poor self-rated health 16 (43.2) 25 (23.1) 0.019 Clinical characteristics: Anxiolytics 2 (5.3) 0 (0) 0.060 Neuroleptics 3 (7.9) 7 (6.1) 0.710 Hypnotics 13 (34.2) 28 (24.3) 0.234 Analgesics, ASA excluded 15 (39.5) 39 (33.9) 0.534 History of stroke 10 (25.0) 12 (10.3) 0.020 Dementia 3 (7.5) 9 (7.7) 1.000 Delirium during preceding month 3 (7.5) 2 (1.7) 0.105 Diabetes Mellitus 6 (15.0) 9 (7.7) 0.213 Hypertension 31 (77.5) 72 (61.5) 0.067 Heart failure 5 (12.5) 18 (15.4) 0.656 History of myocardial infarction 1 (2.5) 8 (6.8) 0.450 History of hip fracture 3 (7.5) 19 (16.2) 0.169 History of thyroid disease 4 (10.3) 12 (10.3) 1.000 Impaired vision 3 (7.9) 10 (8.8) 1.000 Impaired hearing 7 (18.4) 9 (8.0) 0.123

Mean ±SD Mean ±SD BMI (kg/m2) 25.2 ±4.2 25.2 ±4.0 0.986 SBP (mmHg) 154.8 ±25.9 153.7 ±22.0 0.794 DBP (mmHg) 77.7 ±11.9 76.2 ±10.7 0.468 MNA score 25.1 ±3.2 25.8 ±2.4 0.168 MMSE score 25.9 ±3.0 26.1 ±3.2 0.789 Barthel's ADL index score 19.0 ±2.5 19.3 ±2.0 0.355 GDS-15 score 3.3 ±2.2 2.4 ±1.5 0.004 Number of medications 5.7 ±3.5 4.5 ±3.4 0.062

Individual variables may have lower denominators than the incident depressive disorders variable. Abbreviations: Standard Deviation (SD), Acetylsalicylic acid (ASA), Body Mass Index (BMI), Systolic Blood Pressure (SBP), Diastolic Blood Pressure (DBP), Mini Nutritional Assessment (MNA), Mini Mental State Examination (MMSE), Activities of daily living (ADL), Geriatric Depression Scale, 15 item version (GDS-15)

52

Table 3-3. Multivariate logistic regression model displaying factors independently associated with new cases of depressive disorders in Paper 3.

Odds ratio 95% CI p-value

Age 1.01 0.90-1.14 0.831 Female sex 1.09 0.43-2.82 0.852 GDS-15 score 1.39 1.09-1.76 0.007 History of stroke 3.25 1.12-9.44 0.031 Hypertension 2.83 1.08-7.42 0.034

Factors included are variables from Table 2 with a p-value of <0.15, along with age and gender. Nagelkerke R2=0.162. Abbreviations: Confidence Interval (CI), Geriatric Depression Scale, 15 item version (GDS-15)

53

Paper 4

The prevalence of delirium was 43/251 (17.1%) among those aged 85 years, 53/249 (21.3%) among those aged 90 years and 80/208 (38.5%) among those aged ≥95 years (p<0.001). The prevalence of delirium in the total sample was 24.9%.

Among men, delirium prevalence was 19.1% and among women 27.2% (p=0.025). In the total sample, delirium rates among individuals without dementia were much lower than among those with dementia (4.9% vs. 52.0%, p<0.001)(Table 4-1). Delirium prevalence was 7.6% among individuals living independently (n=393) vs. 49.2% among those living in institutions, (n=295, p<0.001).

Characteristics of the total sample as well as the division into dementia diagnosis or no dementia diagnosis can be found in Table 4-1.

Delirium subtype prevalence in the total sample, and subgrouping by dementia status can be found in Table 4-2. Significantly more individuals with concomitant dementia and delirium had any hyperactive symptoms (55.1% vs. 20%, p=0.003) and any psychotic symptoms (60.3% vs. 30.0%, p=0.01) compared to those without dementia but with delirium. Half of all those with delirium (n=176) had any hyperactive symptoms and two thirds had any emotional symptoms. There were no significant differences in symptom profiles between the sexes regarding either motor level subtypes or emotional and psychotic subtypes (data not shown).

In the subsample without dementia (n=409, Table 4-1), the group with delirium (n=20) were prescribed a larger mean number of drugs (8.9 vs. 6.0, p=0.002) while mean Barthel scores (15.2 vs. 18.8, p=0.023), mean MMSE scores (21.4 vs. 24.8, p=0.017) and mean MNA scores (21.1 vs. 24.8, p<0.001) were lower compared to those without delirium (n=389). In the group with delirium there were also higher rates of depressive disorders (70.0% vs. 26.5% p<0.001), institutionalization rates (45.0% vs. 19.8%, p= 0.007) and prescription rates for antipsychotic medication (21.1% vs. 7.0%, p=0.024).

Characteristics of the subsample with dementia with and without delirium can be found in Table 4-3. The final regression model showed that depressive disorders (OR=2.0, p=0.011), heart failure (OR=2.1 p=0.006), institutional living (OR=4.4, p<0.001) and being prescribed antipsychotic medication (OR=3.0, p=0.002) were independently associated with delirium amongst subjects with dementia (Table 4-4).

54

Additional results

The dementia prevalence among participants living in institutions was 72% compared to 21% among participants living independently (p<0.001).

Table 4-1. Characteristics of the Sample in paper 4. Total

(N=708)

n (%)

Without dementia (N=409)

n (%)

With dementia (N=299)

n (%)

p-value

Age, mean ±SD 90.3 ±4.6 89.4 ±4.3 91.5 ±4.7 <0.001 Women 504 (71.2) 269 (65.8) 139 (78.3) <0.001 Delirium 176 (24.9) 20 (4.9) 156 (52.0) <0.001 Living in institution 295 (42.9) 83 (21.1) 212 (71.9) <0.001 Impaired hearing 154 (24.6) 55 (15.5) 99 (36.4) <0.001 Impaired vision 111 (18.6) 41 (11.6) 70 (28.7) <0.001 Depressive disorders 257 (36.3) 117 (28.6) 140 (46.8) <0.001 Stroke 156 (22.2) 78 (19.1) 78 (26.1) 0.026 Urinary tract infection last month 102 (14.4) 34 (8.3) 68 (22.7) <0.001 Heart failure 218 (30.8) 108 (26.5) 110 (36.8) 0.003 Diabetes 101 (14.3) 56 (13.7) 45 (15.1) 0.61 Antipsychotics 94 (13.3) 31 (7.6) 63 (21.1) <0.001 Hypnotics and sedatives 212 (30.0) 117 (28.7) 95 (31.8) 0.39 Antidepressants 133 (18.8) 51 (12.5) 82 (27.4) <0.001 Analgesics 272 (38.5) 123 (30.2) 149 (49.8) <0.001 Drugs with anticholinergic properties 142 (20.1) 72 (17.7) 70 (23.4) 0.061

Mean ±SD Mean ±SD Mean ±SD Number of medications 6.6 ±4.0 6.2 ±4.1 7.1 ±3.8 0.001 MNA score 22.4 ±4.5 24.6 ±3.2 19.6 ±4.3 <0.001 MMSE score 19.4 ±8.2 24.7 ±3.4 12.4 ±7.5 <0.001 GDS-15 score 3.6 ±2.6 3.3 ±2.5 4.3 ±2.9 <0.001 Barthel ADL index score 15.1 ±6.4 18.6 ±2.9 10.6 ±6.9 <0.001 Individual variables may have lower denominators than the delirium variable. Abbreviations: Standard Deviation (SD), Mini Nutritional Assessment (MNA), Mini Mental State Examination (MMSE), Geriatric Depression Scale, 15 item version (GDS-15), Activities of Daily Living (ADL)

55

Table 4-2. Delirium subtypes in paper 4.

Total

(N=176) n (%)

Concomitant dementia

(N=156)

n (%)

No concomitant

dementia (N=20) n (%)

p-value Motor Level Pure hyperactive 34 (19) 32 (21) 2 (10) 0.37

Pure hypoactive 31 (18) 27 (17) 4 (20) 0.76 Mixed hyper-hypoactive 56 (32) 54 (35) 2 (10) 0.026 Unclassifiable 55 (31) 43 (28) 12 (60) 0.003 Any hyperactive symptomsa 90 (51) 86 (55) 4 (20) 0.003 Any hypoactive symptomsa 87 (49) 81 (52) 6 (30) 0.065

Emotional/ psychotic

Pure emotional 40 (23) 33 (21) 7 (35) 0.17 Pure psychotic 23 (13) 23 (15) 0 0.080 Mixed 77 (44) 71 (46) 6 (30) 0.19 Unclassifiable 36 (21) 29 (19) 7 (35) 0.13 Any emotional symptomsa

117 (67) 104 (67) 13 (65) 0.88 Any psychotic symptomsa 100 (57) 94 (60) 6 (30) 0.010

a: Any = Pure + Mixed

56

Table 4-3. Characteristics of participants with and without delirium among those with dementia.

Delirium (N=156)

n (%)

No delirium (N=143)

n (%)

p-value Age, mean ± SD 91.8 ±4.6 91.1 ±4.8 0.19 Women 121 (78) 113 (79) 0.76 Living in institution 136 (87.7) 76 (54.3) <0.001 Impaired hearing 56 (38.4) 43 (34.1) 0.47 Impaired vision 46 (35.4) 24 (21.1) 0.014 Depressive disorders 89 (57.1) 51 (35.7) <0.001 Stroke 39 (25.0) 39 (27.3) 0.66 Urinary tract infection last month 42 (26.9) 26 (18.2) 0.072 Heart failure 70 (44.9) 40 (28.0) 0.002 Diabetes 23 (16.1) 22 (14.1) 0.63 Antipsychotics 49 (31.4) 14 (9.8) <0.001 Hypnotics and sedatives 51 (32.7) 44 (30.8) 0.72 Antidepressants 50 (32.1) 32 (22.4) 0.061 Analgesics 83 (53.2) 66 (46.2) 0.22 Drugs with anticholinergic properties 48 (30.8) 22 (15.4) 0.002

Mean ±SD Mean ±SD Number of drugs prescribed, mean ± SD 7.6 ±3.9 6.6 ±3.6 0.022 Barthel ADL index, 9.3 ±6.0 12.2 ±7.5 0.001 MNA, mean ± SD 19.0 ±3.8 20.4 ±4.9 0.016 MMSE, mean ± SD 11.2 ±7.2 13.6 ±7.6 0.009 GDS-15, mean ± SD 4.5 ±2.8 4.0 ±2.9 0.33

Individual variables may have lower denominators than the delirium variable. Abbreviations: Standard Deviation (SD), Activities of Daily Living (ADL), Mini Nutritional Assessment (MNA), Mini Mental State Examination (MMSE), Geriatric Depression Scale, 15 item version (GDS-15)

Table 4-4. Logistic regression model of delirium among people with dementia in paper 4.

Odds Ratio 95% CI p-value

Depressive disorders 2.0 1.2-3.3 0.011 Heart failure 2.1 1.2-3.7 0.006 Living in institution 4.4 2.4-8.2 <0.001 Antipsychotics 3.0 1.5-6.0 0.002

Model Cox and Snell R square 0.203. Abbreviations: Confidence Interval (CI)

57

General findings

In the Swedish material, the total number of eligible participants increased from 527 in the 2000-2002 cross-section to 610 in the second, an increase of 16%. Not only do the age-specific prevalences of dementia and depression seem to increase but also the number of very old in the underlying population.

Another finding from the collected data was the large overlap of dementia, depressive disorders and delirium, as seen in the figure below. In total, from the 2005-2007 sample, 257 out of 465 (55%) eligible participants had either dementia, a depressive disorder or delirium.

Fifty-six percent of those with dementia and 40% of those with depressive disorders also had delirium. Fifty-one percent of those with dementia had

Figure 0-3. Overlap of dementia, depressive disorders and

delirium in the 2005-2007 sample. a= dementia (n=174), b=

depressive disorders (n=170), c =delirium (n=113).

58

depressive disorders and 52% of those with depressive disorders had dementia. Only three people with delirium during the preceding 30-day period did not have either dementia or a depressive disorder.

59

Discussion

Summary of results

Dementia, depressive disorders and delirium were common among the very old in these studies. One in three participants had dementia, the same proportion had depressive disorders, and one in four had experienced delirium during the preceding thirty-day period. The prevalence of dementia and depressive disorders increased between 2000-2002 and 2005-2007, after controlling for age and sex. In Paper 3, a quarter of those without depressive disorders at baseline had developed depressive disorders after five years. Factors independently associated with new cases of depressive disorders at follow-up were hypertension, a history of stroke and GDS-15 score at baseline. Dementia and depressive disorders were associated with having delirium, half of the participants with dementia had a depressive disorder and almost half of those with depressive disorders had dementia.

Dementia

In Paper 1, the prevalence of dementia was one in six in 2000-2002 and one in four in 2005-2007 in the 85-year-old group. The corresponding figures for the 90-year-olds were one in four and two in five and for the ≥95-year-olds almost half of the participants had dementia at each cross-section. Adjusting for the sampling method, the prevalence of dementia in the total sample was calculated as one in four in 2000-2002 and one in three in 2005-2007 among ≥85-year-olds.

Prevalences

The prevalences of dementia found in the present studies are comparable to findings reported earlier (20-22). A study from Gothenburg by Skoog et al reported similar findings among 85-year-olds 10 years before the 2005-2007 cross-section in this thesis (19). On the whole, the dementia prevalences from the 2005-2007 cross-section appear to be more in line with other studies than do the 2000-2002 prevalences.

60

The high participation rate in the Umeå85+/GERDA study increases the probability that the sample accurately represents the same age groups in the population. The sample contains individuals from both urban and rural areas, living both independently and in institutions.

Increasing prevalences

In Paper 1, the prevalence of dementia increased significantly among the 85-year-olds and 90-year-olds but not among the ≥95-year-olds from 2000-2002 to 2005-2007. The prevalence of dementia also increased significantly in the total sample. The MMSE score was also significantly lower in the later total sample, and in the age groups where dementia prevalence increased.

Most other studies have not shown any significant increase in age-specific prevalence (24) or incidence (137) of dementia in the elderly. Lobo et al. did not find any significant change in the prevalence of dementia among ≥65-year-olds in a Spanish region when comparing one cross-section from 1988-1989 with another from 1994-1996 (23), nor did Hall et al. among African Americans aged ≥70 years between 1992 and 2001 (138). While dementia prevalence increased between 1947 and 1957 in the Lundby study, Rorsman et al. reported no statistically significant changes in the prevalence or incidence of dementia in the longer run between 1947 and 1972 among people aged ≥60 years (26).

In a recently published study Qiu et al. found stable prevalences of dementia from 1987-1989 to 2001-2004 in people aged ≥75 years living in central Stockholm (24).

Another recently published study from Rotterdam reported a trend towards declining dementia incidence from 1990 to 2000, and fewer white matter lesions and a larger brain volume was found in the later cohort subgroup examined using Magnetic Resonance Imaging (MRI). However, the authors also report increased use of antilipemic and antithrombotic drugs, in line with our findings. (139).

Two studies provide some support for increasing dementia rates. Kokmen et al found an increasing incidence of dementia among people aged ≥85 years between 1960 and 1982 in Minnesota (25), and Taylor et al. found an increasing age-adjusted prevalence of Alzheimer’s disease in a sample of Medicare beneficiaries from 1991 to 1999 (140).

61

By and large, previous findings do not support ours, but the disparity of time periods, place, assessment and diagnostic methods makes possible the consolidation of our findings with theirs. Also, few of the studies exclusively focus on the very old.

Depression is a possible prodrome (62) or risk factor for (141, 142) dementia. In Paper 2 we report an increase in depressive disorder prevalence between 2000-2002 and 2005-2007. This could mean that an increase in depressive disorder prevalence would result in a subsequent increase in dementia prevalence. However, dementia might also be an important risk factor for depression (143). An increase in one of the contributory factors the two have in common, such as silent cerebrovascular lesions, is also possible (43, 144).

As presented above, the prevalences in the 2005-2007 cross-section are more in line with results from other studies. It could be that the dementia prevalence has historically been lower in the county of Västerbotten than in other regions, and the increase has brought the prevalence up to conformity. Cardiovascular diseases have historically been more common in the county and better survival after these, perhaps achieved earlier in other parts of Sweden, might have been brought to bear during the observation period.

The proportion of very old people prescribed ß-blockers, calcium channel blockers, antilipemic agents and mean number of prescribed drugs increased between cross-sections indicating more extensive treatment of cardiovascular risk factors such as hypertension and hyperlipidemia. Further, in Paper 1 we report an increasing proportion of people who had had heart surgery in 2005-2007 compared to 2000-2002. Peltonen et al. reported increased survival in Västerbotten after stroke between 1985 and 1994 in the MONICA (Multinational MONItoring of trends and determinants in CArdiovascular disease) study (8) and a similar trend has been reported among men for myocardial infarction (145). Possibly more aggressive treatment of risk factors and increased attention to coronary disease in the very old might contribute to increased survival among people with cardiovascular disease. Cardiovascular disease is a risk factor for dementia, and if more people survive with increasing levels of morbidity then more people will be at risk, as well as being at higher risk, of developing dementia. This could contribute to increased prevalences of both depressive disorders and dementia.

Similarly, improved care for those with dementia would increase survival and therefore increase dementia prevalence, if incidence remains unchanged. The study by Qui et al. from Stockholm found an increased survival rate among people with dementia between 1987-1989 and 2001-2004 (but no increase in dementia prevalence)(24). The study from Rotterdam mentioned above (139)

62

also reported lower mortality rates in the later cohort, which could indicate longer survival among people with dementia.

The last decade has seen greater awareness of dementia among physicians, as measured by the number of specialist referrals (146), and more research on dementia, as measured by the steep increase in the number of articles on the subject. The advent of drugs such as cholinesterase inhibitors targeting the symptoms of Alzheimer’s dementia could have been a contributing factor for this development (147). Cholinesterase inhibitors were indeed also used increasingly between 2000-2002 and 2005-2007, which could indicate more awareness of and interest in dementia among general practitioners, as well as greater familiarity with the drugs. A recent study by Nordström et al. reported that prescribed anticholinergic drugs are associated with a lower risk of myocardial infarction and death (148), which might contribute to the increased survival of people with dementia, resulting in an increased prevalence of the disorder. Further, greater awareness of the disorder among general practitioners could lead to more dementia diagnoses in the medical records. Our diagnoses are partly based on information from these records, possibly contributing to the increased prevalence of dementia seen in Paper 1.

Depressive disorders

Prevalences

In the 2005-2007 cross-section one in three 85-year-olds and almost two out of five 90-year-olds and ≥95-year-olds had depressive disorders.

There are two papers on the topic of depressive disorders included in this thesis: Papers 2 and 3. Both present depressive disorder prevalence rates, and as does Paper 4. The 2005-2007 prevalence in Paper 2 will be used for comparison with other studies. The prevalences presented in this cross-section are considerably higher than other reported prevalences from the studies presented in Table 0-2. (52-55). The study by Skoog et al. reports a prevalence of depressive disorders of 22%, which is close to our findings. This is the only study of the five to use an "umbrella diagnosis" (major depression, dysthymia and depression Not Otherwise Specified) of depressive disorders, the other four assessed major depression, "suspected severe depression" or simply “depression” based on a GDS cutoff.

63

There are some possible reasons for the higher prevalences of depression in our studies. Our depression diagnosis includes major depressive disorder, dysthymic disorder, substance-induced mood disorder with depressive features, mood disorder with depressive features due to a general medical condition and minor depression. Many others report only the prevalence of major depression. The reason for this “umbrella diagnosis” is that it is difficult to differentiate depressive disorders in the very old. One advantage of choosing this approach is that fewer very old with clinically significant depressive symptoms will be missed, a limitation is that the broad definition of depression limits comparison with other studies using narrower criteria.

The high participation rate compared to other studies (52) may also lead to the inclusion of more people with psychiatric comorbidities, especially in institutional care.

When conducting epidemiological studies on depressive disorders in Sweden, midwinter insomnia and seasonal affective disorder should also be considered, as these disorders may lead to some degree of diagnostic overlap and influence the data, depending on the season when the study is undertaken (149). However, the rural part of the Umeå85+/GERDA material was mainly collected from May to October (the urban part was collected year-round), and midwinter insomnia and seasonal affective disorder are most prevalent in winter, which would result in lower depression prevalences. Moreover, the seasonal effect on mood among the elderly has been reported as being small (150), and, although insomnia prevalence increases with age (151), to the author’s knowledge midwinter insomnia has not been investigated among the very old, which makes their impact on prevalences uncertain.

It is possible that there is a genetic component in the high rates of depressive disorders in this area. When investigating single nucleotide polymorphism not only does the population in northern Sweden differ genetically from the rest of the country, Västerbotten also differs from the rest of northern Sweden (152). However, to the knowledge of this author, there are no studies reporting a higher prevalence of depressive disorders in Västerbotten than other Swedish counties, which would have provided a potential link from genetics to disease prevalence. In fact, the suicide prevalence, which may reflect the prevalence of depressive disorders has in the county of Västerbotten during the 21st century been one of the lowest in Sweden (153). The low suicide rate could, however, also reflect the strong religious tradition in the county, or successful suicide prevention strategies from the healthcare.

64

Increasing prevalences

The prevalence of depressive disorders increased from 2000-2002 to 2005-2007 in the total sample. The mean GDS-15 score in the total sample did not increase correspondingly, but nor did the mean GDS-15 score among participants with depressive disorders change. Depressive disorders did not increase significantly in any of the three age groups, probably due to small sample size for the separate age groups.

One international study of depression rates among adults (n=39000) spanning 1974 to 1989 found an increase in cumulative lifetime rates of major depression in almost all cohorts (50), and a recently published paper on the morbidity levels in a representative sample of old people in Sweden reported worsening self-rated levels of psychiatric symptoms between 1992 and 2002 (154). However, others have not found any increase. The Stirling county study found that the prevalence of major depressive episodes remained around 5% over three measurements (1952, 1970 and 1992) (51), and Mattison et al. found decreasing incidence in the Lundby Study from 1947-1972 to 1972-1997 (49).

Dementia is associated with depression. The proportion of participants with dementia who had depressive disorders did not change significantly between 2000-2002 and 2005-2007, nor did the proportion of participants without dementia who had depressive disorders (Paper 2, results). This may indicate that the increasing prevalence of depressive disorders corresponds to the increase in dementia prevalence.

Measuring depressive disorder rates over time using different studies is difficult. Even if the method used by the studies is similar, the diagnostic criteria may be updated, and over long time periods social paradigms concerning depressive disorders may shift, for example the stigma attached to mental illness may decline. Even if the same scale is used for depression assessment over time, the items in the scale may change their meaning for the participants. However, this is unlikely to have happened over five years in a sample of very old people.

Depressive disorders in this thesis include not only major depression but also minor depression, dysthymia, substance-induced mood disorder with depressive features and mood disorder with depressive features due to a general medical condition. The method did not allow separation of these diagnoses and they could contribute unequally to the increase in total depressive disorder prevalence.

65

During the first urban cross-section, the methodology for depression assessment differed from the other cross-sections: the participants' GDS-15 score determined whether they would be assessed with the MADRS. One of two specialists in geriatric medicine assessed individuals with a GDS-15 score of ≥5 using the MADRS during a follow-up visit approximately two weeks after the first visit. In all the later assessments, irrespective of the participant’s GDS-15 score, the assessor performed the MADRS during the initial visit if he or she was a physician or medical student trained to use the scale. The different procedure for assessment using the MADRS in the first urban sample coincides with the lowest subsample prevalence of depressive disorders, and is thus the largest contribution to the increase in prevalence of depressive disorders in Paper 2. A follow-up visit by a physician with the sole purpose of assessing depression including administering of the MADRS may have led to a higher rate of negative diagnoses in the first cross-section. Although the follow-up visit was only done on a subsample of the first urban cross-section, this methodological issue must be taken into account when evaluating the difference in depressive disorder prevalence between the two samples.

Paper 1 reports an increase between 2000-2002 and 2005-2007 in the prescription of ß-blockers, calcium blockers, ACE inhibitors and antilipemic agents, all drugs that have depression as a possible side effect (155). This could contribute to increasing depression prevalences in the very old, who often have more than one of these drugs prescribed, although the extent of this effect is uncertain since evidence is limited (156).

As might have been the case with the dementia diagnosis, there is a possibility that there was a change in documentation habits and procedures for depressive disorders in the medical records between the cross-sections. It is also possible that between cross-section periods general practitioners improved their awareness of depressive disorders among the very old, which might in turn have resulted in higher detection rates.

Associated factors in univariate analysis

In both studies on depressive disorders living in an institution, higher disability levels, more cognitive impairment, dementia and higher GDS-15 scores were associated with depressive disorders in univariate analysis. These results are not surprising and have been reported by others. That a higher GDS-15 score is associated with depressive disorders is a logical closed circle since the depressive disorder diagnoses in the present studies was partly based on the GDS-15, but also since the GDS was constructed to measure depressive

66

symptoms which naturally should correlate with depression.

The associations listed above may all be bidirectional (except the GDS-15 score). For example, a history of depression (157) or current depression (142) reportedly increases the risk of developing dementia later on. This could be through the direct adverse effect of having depression, similarly to depressive symptoms increasing the risk of cardiovascular disease (158). Depression in old age also often adversely affects cognition directly through depressive symptoms such as apathy (159). In the reverse direction, since the pathological changes in the brain leading to dementia (or at least Alzheimer’s dementia) can appear decades before the disorder is manifested (160), it is difficult to rule out depressive symptoms as a prodrome of dementia. Depression as a reaction to an incipient reduction in cognitive capabilities has also been suggested (161).

It is important to remember that both dementia and depression (and delirium) are disorders based on diagnostic criteria, and can have a multitude of causes, either singly or in combination. These may partly explain the observed associations between these conditions.

Associated factors in multivariate analysis

Decreasing blood pressure in people with previous hypertension has recently been reported to be associated with depressive symptoms (and therefore possibly depression) (162). Also in line with our findings, other studies have reported a univariate association between lower systolic pressure and depression among elderly people (163, 164). In Paper 3 a multivariate analysis was performed, where hypertension was independently associated with the development of depressive disorders. This association has also been reported earlier (165) but is not well-established. Here the opposite has also been reported, low blood pressure as a risk factor for depression (166). This could be due to increased antihypertensive treatment. If lower systolic pressure and depression were associated this change would be reflected in an increase in the prevalence of depressive disorders, as observed in Paper 2. The continuous blood pressure baseline variable was not associated with incident depression.

A history of stroke was associated with developing depressive disorders. Although the association between a history of stroke and incident depression has been described earlier (167), very few studies have been made on lifetime depression risk in stroke survivors. Vascular brain lesions could be a common denominator.

67

A higher GDS-15 score at baseline was associated with incident depressive disorders at follow-up. It should be noted that since the analysis only included individuals without depression at baseline the GDS scores were generally low (mean 2.7). Earlier studies have likewise shown a strong association between depressive symptoms at baseline and the development of depression (38, 168-170). The association is to be expected. It could be due to those developing depressive disorders having depressive prodromal symptoms, or people misdiagnosed as not having depression then being diagnosed correctly at follow-up. It could also be that the mild depressive symptoms exhibited by these people are a discrete state, analogous to mild cognitive impairment for dementia.

Delirium

Prevalences

The prevalence of delirium was one in six among participants aged 85 years, one in five among participants aged 90 years and two in five among participants aged ≥95 years. Adjusting for the sampling method, this would correspond to a prevalence of delirium of one in five among people aged 85 years or older. Total prevalence in Paper 4 cannot be directly compared to any other studies, since the few studies available differ in methodology and sample characteristics.

Other studies

To my knowledge, there are no studies directly comparable to paper 4. Andrew et al. reported a delirium point-prevalence to ≤0.5% in a Canadian study on people aged ≥65 years without dementia (n=1793) (79). Our thirty-day prevalence among people without dementia (n=409) was 5%; hence the findings in Paper 4 seem to be in line with these findings. Rahkonen et al. reported delirium incidence in a Finnish study on people aged ≥85 years (n=199) without dementia as 10% over three years. A comparison of an incidence study with a prevalence study is impossible for a disorder such as delirium where mean duration is uncertain.

Another subgroup available for comparison is the elderly living in institutions. Boorsma et al. reported a delirium point-prevalence of 8% among people (n=2193) living in institutions in the Netherlands (171) and, in a Canadian

68

study, McCusker et al. reported a delirium point-prevalence of 7% among people living in institutions aged ≥65 years (n=279) (172). Sandberg et al. reported a one-week delirium prevalence of 47% among people aged ≥75 years living in institutions (n = 398) (87). Our thirty-day delirium prevalence among participants living in institutions was 49%. Apart from the difference in assessment periods, these three studies had lower mean ages than Paper 4, possibly indicating lower levels of comorbidity and higher levels of functionality. Level of disability and morbidity required for placement in institutional living as well as the actual definition of the term may also vary across the studies. When comparing dementia prevalence among those living in institutions, the Boorsma study found a considerably lower dementia prevalence (34%) than that reported in Paper 1 (72%) while the other two studies had only slightly lower prevalences (66%).

Delirium in stroke, dementia and in old people with depression

Delirium is common after stroke and in people with dementia and/or depression. Delirium precipitated by stroke is well described (173, 174), and dementia is a strong predisposing factor for delirium, as well as being reported as a potential outcome of having delirium (175, 176). Depressive symptoms have been reported to predict delirium (177).

Depressive disorders were common among those with delirium (with and without dementia) in this sample (although a potential confounder is psychotropic medication and opioids, common in persons with depression, which sometimes have anticholinergic properties increasing susceptibility to delirium). Depressive symptoms are a reported risk factor for the development of delirium among elderly people (177, 178).

In Paper 4, thirty-day delirium prevalence among people with dementia was 52% and 5% among those without dementia. Dementia is known to multiply delirium risk (179), and delirium is reported to increase risk of subsequent dementia development (91). Since the two conditions are closely associated and dementia prevalences were high in the Umeå85+/GERDA study, this will also affect delirium prevalence.

Since those living at home have a more independent lifestyle less information from sources other than the participants themselves was available than is the case with individuals living in a residential care facilities, increasing the risk of

69

recall or response bias. Thus, delirium prevalence among people living independently may actually be higher than is reported in Paper 4.

The 3Ds

In clinical practice dementia, depression and delirium often coexist, and they may be difficult to distinguish from each other. In clinical practice and theory, the three disorders are sometimes referred to as “the 3D’s” and educational materials often focus on how to differentiate between them in a clinical setting (17, 180). This makes sense from a clinical perspective. Since they are so common, a health professional working with geriatric patients will frequently come into contact with the disorders, making knowledge of symptoms and signs valuable, since misdiagnosis might lead to incorrect treatment, or no treatment when treatment is necessary.

Interlinked pathogenesis

In Paper 3, diagnosis of hypertension and a history of stroke were associated with the development of depression over five years, although both those who had and those who had not developed depression at follow-up had similar systolic pressure at baseline. A univariate association between hypertension and depression has been reported by others (181), and studies have shown that depression is highly prevalent after stroke besides being associated with the presence of stroke risk factors (61, 109). Earlier hypertension has been suggested as a strong risk factor for dementia (182, 183), and decreasing blood pressure in persons who formerly had hypertension has been reported to be associated with incident depression (162). In addition to this, hypertension is also a risk factor for later development of ischemic white matter hyperintensities (WMHs)(184). WMHs are radiological correlates of depression and dementia which can be seen on MRI examination of the brain as unspecific areas of increased T2 and FLuid Attenuated Inversion Recovery (FLAIR) signal intensity in the subcortical white matter. Due to the association with cardiovascular risk factors and cerebrovascular ischemia, it has been proposed that the WMHs represent brain lesions of vascular origin (43, 185). The extent of WMHs correlates with increasing age (186), delirium (187), geriatric depression (60, 188) and dementia (144). Older people with late-onset depression will, to a greater degree, have MRI hyperintensities representing

70

possible cerebrovascular disease than controls or persons with early-onset depression (189).

The vascular depression hypothesis posits that cerebrovascular disease may predispose, precipitate or perpetuate some geriatric depressive syndromes (43).

In Paper 3 78% of surviving participants with depressive disorders at baseline still had depressive disorders at follow-up. If depressive disorders in the very old are largely caused by a progressive vascular brain disease, this could partly explain the chronic course of depression in this age group. WMHs in depression and concomitant dementia, at least of the Alzheimer type, may reduce the effect of antidepressant treatment (74, 190). These states are indicative of structural brain damage and possibly reduced brain perfusion against which antidepressants are ineffective (43). In the very old, high levels of comorbidity, malnutrition or insufficient follow-up and adjustment of treatment may be other causes, as well as lack of social support.

The relation between depression, dementia and blood pressure in the very old seems to be complex, and needs to be further investigated. In Paper 3, baseline depressive disorders were significantly associated with a lower measured systolic blood pressure. Lower systolic pressure could give lower cerebral perfusion, leading to vascular brain lesions leading to vascular depression. In Papers 1 and 2 there was a trend towards decreasing blood pressure between cross-sections. Other studies have linked lower MMSE score (191, 192) and Alzheimer’s disease (193) to low systolic blood pressure, and reported low systolic blood pressure as a risk factor for dementia (194). Molander reported that, in association with an increased prescription of antihypertensive drugs, systolic blood pressure had declined over time in the elderly (195), which would support the increase in dementia and depressive disorders.

In the present thesis it was found that the treatment with antihypertensives had almost doubled in the very old. If the lowering of blood pressure in this age group can contribute to an increasing incidence of depressive disorders and dementia this needs to be evaluated in randomized treatment studies as well as studies of the discontinuation of antihypertensive treatment.

In Paper 3, despite being a risk factor for the development of depressive disorders over five years, hypertension was not associated with depressive disorders at baseline. This could be due to low systolic blood pressure acting as a confounder. In longitudinal analysis baseline hypertension was associated with the development of depressive disorders after five years, possibly since low blood pressure was associated with increased mortality. In the same vein, a history of stroke was not associated with baseline depressive disorders. If

71

hypertension and stroke are associated, and low blood pressure is associated with increased mortality, then the situation might be that there is selective attrition of “non-vascular depressions” (those not associated with hypertension and history of stroke), leaving mostly those with depressive disorders associated with stroke and hypertension alive at follow-up.

The Hypothalamic-Pituitary-Adrenal axis in dementia, depression and delirium

There is dysregulation of the HPA axis in all three disorders, suggesting an interlinked pathogenesis. People with depression show, more often than normal, chronic HPA axis hyperactivity coupled with an inability for this system to return to normal functioning following a stressor (196, 197). HPA axis hyperactivity may lead to abnormally elevated levels of serum glucocorticoids, and high cortisol levels also increase the hippocampus’s vulnerability to global ischemia more easily leading to neuronal death (198), suggesting depression as a risk factor for dementia. Hippocampal changes are also seen in almost all AD patients as well as in many persons with vascular dementia (161), possibly linking these disorders to depression. Hypercortisolism is also associated with delirium (199). One possible mechanism is via inflammation. Even if cortisol is an anti-inflammatory steroid, under special circumstances, such as prolonged stress, cortisol increases inflammation (200), leading to neurodegeneration and increased susceptibility to delirium.

Studies report that chronic inflammation has a central role in all three disorders (201, 202). Higher serum levels of inflammatory markers have been shown to be associated with both an increased incidence of delirium and accelerated cognitive decline (203), and the systemic inflammation as seen in depression (204) has been shown to cause neurodegeneration (205), increasing susceptibility to delirium. Acute inflammation as seen in systemic infection is a well-described trigger of delirium (206), especially among those with dementia. Cholinesterase inhibitors have been reported to decrease inflammation (207) and could, therefore, inhibit neurodegeneration, which is a possible pathway to the reduction of dementia symptoms, apart from cholinergic augmentation. In short, inflammation may be another part of the interlinked pathogenesis of the three disorders.

Obstructive Sleep Apnea Syndrome (OSAS) is a medical disorder reportedly associated with delirium (208), depression (209) (210) and dementia (211). Since OSAS may be treated this disorder should be considered in persons with dementia, depression or delirium.

72

Methodological considerations

The Umeå85+/GERDA study is one of the largest population-based studies on the very old in Europe, and includes participants living independently and in institutions from both rural and urban areas in the investigated geographical area, which stretches over two countries. There were high participation rates; in the 2000-2002 cross-section, the participation rate was 80% and in the 2005-2007 cross-section the corresponding figure was 76%. Most participants were investigated in person, in their homes, and since independent investigators and not caring staff made the assessments assessors were not distracted by other worktasks and every effort was made to collect complete data. However, the Umeå85+/GERDA study was designed to investigate, among other things, prevalences of a large number of different diseases and disorders. This means that it was not designed with only dementia, depression and delirium in focus. This led to some limitations, such as no follow-up assessments, which might have further increased diagnostic accuracy.

The Swedish National Tax Board lists and the Finnish Population Register lists are very accurate allowing all eligible participants to be easily identified, approached and asked for their consent to participate. The study encompasses a broad sample drawn from the population, and comprises individuals both living independently and in institutions, thus forming representative samples of 85-, 90- and ≥95-year-olds, respectively. However, in order to form three age group of approximately equal size, the selection process included half of those aged 85 years, all individuals aged 90 years, and everyone aged ≥95 years, meaning that the 90-year-olds and participants aged ≥95 years are overrepresented compared to their proportion in the population aged ≥85 years. When considering total prevalences across all age groups, any variables positively correlating with age will be higher than the true value and the reverse is true for any variables negatively correlating with age. This sampling method should be kept in mind when making comparisons with studies including all individuals aged ≥85 years. However, each age group is representative of the corresponding age group in the underlying population, and although the sampling method, with three like-sized age groups and the advanced age of the sample, limits the number of studies available for comparison it also facilitates direct comparison with other studies that include samples of 85-, 90- or ≥95-year-olds.

Although different assessors collected the information upon which diagnoses were based, in both countries the same experienced researcher and geriatrician judged whether the data collected was sufficient to decide if the participants fulfilled the DSM-IV criteria for delirium, depressive disorders or dementia.

73

The diagnoses were established on the basis of all available data, including applicable assessment scales, medical records and information from carers/relatives, according to the same criteria for all participants, in all samples. This approach, compared to scale cutoff diagnoses or other methods may have both advantages and disadvantages. One advantage of a less rigid method may be that incorrectly diagnosing depression in the presence of depressive symptoms due to recent negative life events can be avoided. One possible limitation would be diagnostic shift over time.

Depression, delirium and dementia can be difficult both to diagnose and to distinguish from one another among the elderly, because of a large overlap in diagnoses, comorbidity and similar symptomatology. Establishing a diagnosis in a psychiatric epidemiological study can be done in various ways. A scale cut-off can be used with values above a certain level indicating a case (as can be done with the GDS-15) or a diagnostic system can be applied to either a single scale (as with the CAM when assessing delirium (126)) or to a more extensive assessment. In the present studies the DSM-IV criteria for the mental disorders studied were applied to all the information collected through direct assessment and data gathering from carers and medical records. The strengths of this approach might include a higher sensitivity, dissimulation could be made more difficult than with an assessment scale, limitations could include limited reproducibility since different information sources were available for different participants and since establishing diagnoses “manually” is unlikely to be done in the exactly the same fashion by different physicians.

The Diagnostic and Statistical Manual published by the American Psychiatric Association attempts to provide a common language and standard criteria for the classification of mental disorders. The diagnostic criteria for each disorder are offered as guidelines for establishing diagnoses in a clinical setting. The DSM-IV, despite its flaws, represents a great effort to facilitate cross-national, cross-cultural and cross-paradigm research.

In Paper 2, the large difference between the two cross-sections in the number of individuals available to assess for depressive disorders may also be due to the fact that, with the gradual advent of computerized medical records there was more available data for more individuals. In particular, the data collected in the first urban sample sometimes lacked medical records from general practices, as these were not yet digitalized. This is a potential source of error in Paper 2 and may also in part contribute to the lower proportion of depressive disorders in the first urban sample. However, the mean GDS-15 score of individuals with a depressive disorder diagnosis did not differ between the cross-sections, which would have been an indicator of diagnostic shift. In addition, such increased awareness should lead to better care, possibly leading to increased survival

74

among persons with dementia or depressive disorders, and through this higher prevalences. Further, regarding dementia, the decrease in mean MMSE scores supports a true increase in prevalence. The proportion of depressed persons treated with antidepressants did not increase significantly either, which would have indicated increased attention to depressive disorders in the very old.

Another possible limitation is the clinical diagnosis, which might be less accurate than a comprehensive dementia examination including brain imaging, or in the case of depressive disorders, repeated assessments by a suitable specialist physician. However, during the first urban cross-section some participants with cognitive impairment were referred for a complete dementia assessment and in no case did this change the diagnoses.

The cross-sectional design of Papers 1, 2 and 4 prevents the drawing of conclusions about cause and effect. It is also not known how many of those who declined to participate did so because of dementia or depressive disorders, and thus if our sample differs from the population. It could be speculated that individuals who declined participation may have had depressive disorders or dementia to a greater extent than participants, leading to an underestimation of the prevalence of these disorders.

One limitation of Paper 3 is that only 40 individuals, of the original baseline sample without depression (n=384), were diagnosed with depressive disorders at follow-up, which is a small sample to analyse using multivariate logistic regressions. The attrition rate is unfortunate but not surprising due to the high mortality in this age group and probably since being at risk of incident depression also means being at increased risk of death. Since only current depressive disorders were measured at follow-up and since depressive disorders may in some cases have short durations, the lengthy time-span between baseline and follow-up could result in depressive disorders present between baseline and follow-up going into remission, and thus not being registered at follow-up.

The first and second urban and the first rural sample have been investigated by Bergdahl et al. in a series of studies. One of these studies investigating the prevalence of depression, factors associated with depression as well as the impact of depression on well-being and mortality on a partly overlapping sample of the Umeå85+/GERDA database sample, reported data on the prevalence of depression in the 2000-2002 sample (212). Since the aim of Paper 2 was to investigate change in depression prevalence over time, these data were reported again here.

75

Clinical and social implications

During recent decades, an increasing proportion of the population have survived to reach a higher age, at least in part because of better treatment of cardiovascular risk factors such as hypertension, hyperlipidemia, and diabetes. Reduced mortality from cardiovascular disease, such as myocardial infarction and stroke, has been reported, (2, 27) as well as increased survival after stroke, probably due to improved secondary prevention and rehabilitation (27). Furthermore, coronary surgery, such as coronary artery bypass graft (CABG), has also been made available to older people, which applied particularly to the 85-year-olds in Paper 1. This might indicate that a larger proportion of people with cardiovascular risk factors survive to reach a greater age and are, therefore, exposed to a higher risk of developing dementia or depressive disorders, leading to an increased incidence of these disorders. These changes could at least partially explain the increasing proportion of individuals with dementia and depressive disorders in this thesis. Another possible explanation could be that people with dementia or depressive disorders, perhaps due to better care and drug treatment, survive longer with their dementia or depressive disorders. In the Umeå85+/GERDA material, a larger proportion of participants with dementia were treated with cholinesterase inhibitors in the groups from 2005-2007, indicating a more active treatment approach, and possibly increased survival among persons with dementia. The increase in depressive disorder prevalence may partly be driven by the increase in dementia.

We have not found any support for the compression of morbidity and disability hypothesis described in the introduction. The increases in the prevalences of dementia and depressive disorders were controlled for age.

The finding that a large proportion of participants had dementia or depressive disorders or had had delirium recently raises the question of how many of these cases could have been detected by optimal screening. Although it is unknown due to practical limitations how many of our cases that were registered in medical records, one indication could be the low treatment rates of depressive disorders. The suspected difference in detection between the present studies and clinical practice may also arise because of different diagnostic criteria: dysthymic disorder and minor depression are perhaps not diagnosed very extensively in geriatric clinical practice, or by general practitioners.

Papers 1 and 2 seem to be among the first to report a significantly increasing age-specific prevalence of dementia and depressive disorders among very old people. The high prevalence of delirium found in the population may also be a reason for action.

76

Since both the number and proportion of very old is increasing, the number of very old individuals suffering from these disorders will probably increase even further in the future. A possible age-specific increase in dementia would further increase the prevalence of delirium and depression. This represents a challenge to the clinician, since identifying and treating depression among the very old is difficult (213, 214), due among other things to the heterogeneity of symptoms and high levels of comorbidity. Regarding risk factors for depressive disorders, those with mild depressive symptoms as indicated by only a few points on the GDS-15 scale should be the focus of watchful waiting, since this could ultimately lead to depression.

Implications for future research

An increased prevalence of a disorder can be due either to increased survival with the disorder or increased incidence. In order to know why the prevalence of dementia and depressive disorders increased between 2000-2002 and 2005-2007 further longitudinal studies are needed.

Further, the increasing prevalences may be simply a chance finding which would not be confirmed if prevalence was investigated over a longer time. New studies on the Umeå85+/GERDA material including later cross-sections could address this.

Paper 3 is one of the largest population-based longitudinal studies of depressive disorders in the very old. However, few of the participants were still alive after five years, resulting in a small study sample. Larger study samples are needed, to investigate risk factors for depression in this age group. Since there are some etiological differences between depression in the very old and in younger ages, and since some studies indicate that the condition shows a poor response to treatment among the very old (37, 39, 69), the efficacy of antidepressant treatment in the very old should be further investigated. Also, though we do not have data on this, the general impression was that virtually none of the participants with depression were receiving psychotherapy. Since there is little information available on the effect of psychotherapy in the oldest old, as with antidepressants, randomized controlled trials should be initiated in order to determine whether psychotherapy is effective in this age group. Whether primary and secondary prevention of cerebrovascular disease could prevent depression from developing should also be investigated. However, the relationship between blood pressure and depression in the very old is too complex to suggest a more extensive anti-hypertensive treatment, especially since those who died before the follow-up were more likely to have low blood

77

pressure. This needs to be further investigated before any treatment recommendations can be made.

The thesis identifies several important factors associated with dementia, depressive disorders and delirium which could possibly be targets for prevention or treatment. In the case of univariate analysis associations, longitudinal studies could be of use in determining causality. In the case of longitudinal associations randomized intervention studies are needed.

Conclusions

We found that depressive disorders, dementia and delirium are common and often coexist among very old people. As indicated by the age-specific prevalence increases found in this thesis, dementia and depressive disorders may be an increasing problem among the very old, even regardless of population aging. Important baseline risk factors or predictors for development of depressive disorders over five years are hypertension, a history of stroke and higher GDS-15 score. When the 2010-2012 data collection is ready for analysis it will be interesting to see the prevalence levels from this data. Higher identification rates of these three disorders would be beneficial, but effective treatment of depression and dementia in old age is still out of reach. Investigating and minimizing risk factors may be the only way to lower prevalences in the foreseeable future.

78

Acknowledgements

This work was carried out at the department for Community Medicine and Rehabilitation, Geriatric Medicine, Sweden.

Financial support

The Umeå85+/GERDA study and this work was supported by funds from the Interreg IIIA Kvarken-Mittskandia Program (2005-2007) and the Bothnia-Atlantica Program (2008-2011), both funded by the European Union and the European Regional Development Fund; The Regional Council of Ostrobothnia, the Municipality of Umeå; the County Administrative Board in Västerbotten; Äldrecentrum Västerbotten; the University of Umeå; Novia University of applied sciences; the University of Vasa and Åbo Akademi University, and was additionally supported by grants from the Vårdal research foundation, the Swedish Research Council (K2005-27VX-15357-01A); the King Gustav V’s and Queen Viktoria’s Foundation of Freemasons; the Umeå University Foundations for Medical Research; Umeå University (ALF); the Erik and Anne-Marie Detlof’s foundation at Umeå University; the Swedish Dementia Association; the Borgerskapet in Umeå Research Foundation and the Alzheimer Foundation.

Personal acknowledgements

To the women and men who generously participated in the Umeå85+/GERDA study. A special thank you to those I personally met for memorable moments and stories.

To Yngve Gustafson, my supervisor, for believing in my capabilities when I didn’t and for endless support. This thesis would never have happened without you.

To Hugo Lövheim, my co-supervisor, who always had a good answer when I needed one.

To my Co-authors: Sofia Petersson, Karin Wallin, Per Allard and Birgitta Olofsson. Thank you for good feedback and attention to detail.

79

To my great colleagues and friends at the Geriatric Centre in Umeå: Bodil Weidung, Maine Carlsson, Håkan Littbrand, Peter Nordström, Carl Hörnsten, Annica Toots, Nina Lindelöf, Robert Wiklund and Magdalena Domellöf. Thank you for the seminar input, the fika and the laughs.

To the assessors involved in data collection for the Umeå85+/GERDA study. Thank you for your efforts. A special thank you to those I worked alongside in the nooks and crannies of Western Bothnia.

To Karin Gladh and Annica Dahlberg, for valuable help with a wide range of things.

To Jon Brännström for the skiing and the Nissesfika.

To Lena Molander, for invaluable help with tables and layout during the writing of this thesis, and for being a great person.

To Mia Conradsson for the dancing, the laughs, and good advice.

To Martin Gustafson for time well spent together.

To Gustaf Boström for links to useful studies and for patience and enthusiasm in the rehearsal room.

To Lisa, Hans and Rut Nemeczek, for being my friends.

To other helping friends: Fredrik Toss and Mattias Vågberg. You really went above and beyond the call of duty.

To Lisa Olsson for the support.

To my parents. I love you guys.

80

References

1. United Nations Department of Economic and Social Affairs

Population Division. World Population to 2300. New York: United Nations; 2004.

2. Bloom DE. 7 billion and counting. Science. 2011;333(6042):562-9.

3. Lundström H. Kohortdödligheten i Sverige: dödlighets-utvecklingen sedan år 1861: SCB; 2010.

4. Tunstall-Pedoe H, Kuulasmaa K, Mahonen M, Tolonen H, Ruokokoski E, Amouyel P. Contribution of trends in survival and coronary-event rates to changes in coronary heart disease mortality: 10-year results from 37 WHO MONICA project populations. Monitoring trends and determinants in cardiovascular disease. Lancet. 1999;353(9164):1547-57. Epub 1999/05/20.

5. Fries JF. Aging, natural death, and the compression of morbidity. N Engl J Med. 1980;303(3):130-5. Epub 1980/07/17.

6. Gruenberg EM. The failures of success. The Milbank Memorial Fund quarterly Health and society. 1977;55(1):3-24. Epub 1977/01/01.

7. Rosen M, Haglund B. From healthy survivors to sick survivors—implications for the twenty-first century. Scandinavian Journal of Public Health. 2005;33(2):151-5.

8. Peltonen M, Stegmayr B, Asplund K. Time Trends in Long-term Survival After Stroke: The Northern Sweden Multinational Monitoring of Trends and Determinants in Cardiovascular Disease (MONICA) Study, 1985–1994. Stroke. 1998;29(7):1358-65.

9. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th edition. Washington, DC: American Psychiatric Association; 1994.

81

10. Lobo A, Launer LJ, Fratiglioni L, Andersen K, Di Carlo A, Breteler MM, et al. Prevalence of dementia and major subtypes in Europe: A collaborative study of population-based cohorts. Neurologic Diseases in the Elderly Research Group. Neurology. 2000;54(11 Suppl 5):S4-9. Epub 2000/06/15.

11. Launer LJ, Andersen K, Dewey ME, Letenneur L, Ott A, Amaducci LA, et al. Rates and risk factors for dementia and Alzheimer's disease: results from EURODEM pooled analyses. EURODEM Incidence Research Group and Work Groups. European Studies of Dementia. Neurology. 1999;52(1):78-84. Epub 1999/01/28.

12. Nagai M, Hoshide S, Kario K. Hypertension and dementia. Am J Hypertens. 2009;23(2):116-24. Epub 2009/11/21.

13. Ivan CS, Seshadri S, Beiser A, Au R, Kase CS, Kelly-Hayes M, et al. Dementia after stroke: the Framingham Study. Stroke. 2004;35(6):1264-8. Epub 2004/05/01.

14. Svanborg A. The health of the elderly population: results from longitudinal studies with age-cohort comparisons. Research and the ageing population. 1988:3-16.

15. The global burden of disease : 2004 update. Mathers C, Fat DM, Boerma JT, World Health O, editors. Geneva, Switzerland :: World Health Organization; 2008.

16. Todd S, Barr S, Roberts M, Passmore AP. Survival in dementia and predictors of mortality: a review. International Journal of Geriatric Psychiatry. 2013.

17. Downing LJ, Caprio TV, Lyness JM. Geriatric psychiatry review: differential diagnosis and treatment of the 3 D's - delirium, dementia, and depression. Current psychiatry reports. 2013;15(6):365. Epub 2013/05/03.

18. Prince M, Bryce R, Albanese E, Wimo A, Ribeiro W, Ferri CP. The global prevalence of dementia: a systematic review and metaanalysis. Alzheimer's & dementia : the journal of the Alzheimer's Association. 2013;9(1):63-75 e2. Epub 2013/01/12.

19. Skoog I, Nilsson L, Palmertz B, Andreasson L-A, Svanborg A. A population-based study of dementia in 85-year-olds. New England Journal of Medicine. 1993;328(3):153-8.

82

20. Corrada M, Brookmeyer R, Berlau D, Paganini-Hill A, Kawas C. Prevalence of dementia after age 90 Results from The 90+ Study. Neurology. 2008;71(5):337-43.

21. von Strauss E, Viitanen M, De Ronchi D, Winblad B, Fratiglioni L. Aging and the occurrence of dementia: findings from a population-based cohort with a large sample of nonagenarians. Arch Neurol. 1999;56(5):587-92. Epub 1999/05/18.

22. Riedel-Heller SG, Busse A, Aurich C, Matschinger H, Angermeyer MC. Prevalence of dementia according to DSM-III-R and ICD-10: results of the Leipzig Longitudinal Study of the Aged (LEILA75+) Part 1. Br J Psychiatry. 2001;179:250-4. Epub 2001/09/05.

23. Lobo A, Saz P, Marcos G, Dia JL, De-la-Camara C, Ventura T, et al. Prevalence of dementia in a southern European population in two different time periods: the ZARADEMP Project. Acta Psychiatr Scand. 2007;116(4):299-307. Epub 2007/09/07.

24. Qiu C, von Strauss E, Backman L, Winblad B, Fratiglioni L. Twenty-year changes in dementia occurrence suggest decreasing incidence in central Stockholm, Sweden. Neurology. 2013;80(20):1888-94. Epub 2013/04/19.

25. Kokmen E, Beard CM, O'Brien PC, Offord KP, Kurland LT. Is the incidence of dementing illness changing? A 25-year time trend study in Rochester, Minnesota (1960-1984). Neurology. 1993;43(10):1887-92. Epub 1993/10/01.

26. Rorsman B, Hagnell O, Lanke J. Prevalence and incidence of senile and multi-infarct dementia in the Lundby Study: a comparison between the time periods 1947-1957 and 1957-1972. Neuropsychobiology. 1986;15(3-4):122-9. Epub 1986/01/01.

27. Langa KM, Foster NL, Larson EB. Mixed dementia: emerging concepts and therapeutic implications. JAMA. 2004;292(23):2901-8. Epub 2004/12/16.

28. Diagnostic and Statistical Manual of Mental Disorders, 4th edition. Washington, DC: American Psychiatric Association; 1994.

29. Schwarz S, Froelich L, Burns A. Pharmacological treatment of dementia. Curr Opin Psychiatry. 2012;25(6):542-50. Epub 2012/09/21.

83

30. Yaffe K, Hoang T. Nonpharmacologic treatment and prevention strategies for dementia. Continuum (Minneap Minn). 2013;19(2 Dementia):372-81. Epub 2013/04/06.

31. Littbrand H, Stenvall M, Rosendahl E. Applicability and effects of physical exercise on physical and cognitive functions and activities of daily living among people with dementia: a systematic review. American journal of physical medicine & rehabilitation / Association of Academic Physiatrists. 2011;90(6):495-518. Epub 2011/03/25.

32. Birks J. Cholinesterase inhibitors for Alzheimer's disease. Cochrane Database Syst Rev. 2006(1):CD005593. Epub 2006/01/27.

33. Raina P, Santaguida P, Ismaila A, Patterson C, Cowan D, Levine M, et al. Effectiveness of cholinesterase inhibitors and memantine for treating dementia: evidence review for a clinical practice guideline. Ann Intern Med. 2008;148(5):379-97. Epub 2008/03/05.

34. Rodda J, Carter J. Cholinesterase inhibitors and memantine for symptomatic treatment of dementia. BMJ. 2012;344:e2986. Epub 2012/05/03.

35. Olazaran J, Reisberg B, Clare L, Cruz I, Pena-Casanova J, Del Ser T, et al. Nonpharmacological therapies in Alzheimer's disease: a systematic review of efficacy. Dement Geriatr Cogn Disord. 2010;30(2):161-78. Epub 2010/09/15.

36. Kleinman A, Good B. Culture and depression : studies in the anthropology and cross-cultural psychiatry of affect and disorder. Berkeley: University of California Press; 1985. xi, 535 p. p.

37. Stek ML, Gussekloo J, Beekman AT, van Tilburg W, Westendorp RG. Prevalence, correlates and recognition of depression in the oldest old: the Leiden 85-plus study. Journal of Affective Disorders. 2004;78(3):193-200. Epub 2004/03/12.

38. Stek ML, Vinkers DJ, Gussekloo J, van der Mast RC, Beekman AT, Westendorp RG. Natural history of depression in the oldest old: population-based prospective study. British Journal of Psychiatry. 2006;188:65-9. Epub 2006/01/03.

39. Bergdahl E, Gustavsson JM, Kallin K, von Heideken Wågert P, Lundman B, Bucht G, et al. Depression among the oldest old: the

84

Umeå 85+ study. International Psychogeriatrics. 2005;17(4):557-75. Epub 2005/09/28.

40. Blazer DG. Psychiatry and the oldest old. American Journal of Psychiatry. 2000;157(12):1915-24. Epub 2000/12/01.

41. Van der Weele GM, Gussekloo J, De Waal MW, De Craen AJ, Van der Mast RC. Co-occurrence of depression and anxiety in elderly subjects aged 90 years and its relationship with functional status, quality of life and mortality. International Journal of Geriatric Psychiatry. 2009;24(6):595-601. Epub 2008/11/26.

42. Vinkers DJ, Stek ML, Gussekloo J, Van Der Mast RC, Westendorp RG. Does depression in old age increase only cardiovascular mortality? The Leiden 85-plus Study. International Journal of Geriatric Psychiatry. 2004;19(9):852-7. Epub 2004/09/08.

43. Taylor WD, Aizenstein HJ, Alexopoulos GS. The vascular depression hypothesis: mechanisms linking vascular disease with depression. Molecular psychiatry. 2013. Epub 2013/02/27.

44. Hornsten C, Molander L, Gustafson Y. The prevalence of stroke and the association between stroke and depression among a very old population. Archives of Gerontology and Geriatrics. 2012. Epub 2012/06/01.

45. Weyerer S, Eifflaender-Gorfer S, Wiese B, Luppa M, Pentzek M, Bickel H, et al. Incidence and predictors of depression in non-demented primary care attenders aged 75 years and older: results from a 3-year follow-up study. Age and Ageing. 2013;42(2):173-80.

46. Djernes JK. Prevalence and predictors of depression in populations of elderly: a review. Acta Psychiatr Scand. 2006;113(5):372-87. Epub 2006/04/11.

47. Jongenelis K, Pot AM, Eisses AM, Beekman AT, Kluiter H, Ribbe MW. Prevalence and risk indicators of depression in elderly nursing home patients: the AGED study. J Affect Disord. 2004;83(2-3):135-42. Epub 2004/11/24.

48. Mann AH, Schneider J, Mozley CG, Levin E, Blizard R, Netten A, et al. Depression and the response of residential homes to physical health needs. International Journal of Geriatric Psychiatry. 2000;15(12):1105-12.

85

49. Mattisson C, Bogren M, Nettelbladt P, Munk-Jorgensen P, Bhugra D. First incidence depression in the Lundby Study: a comparison of the two time periods 1947-1972 and 1972-1997. J Affect Disord. 2005;87(2-3):151-60. Epub 2005/06/28.

50. The changing rate of major depression. Cross-national comparisons. Cross-National Collaborative Group. JAMA. 1992;268(21):3098-105. Epub 1992/12/02.

51. Murphy JM, Laird NM, Monson RR, Sobol AM, Leighton AH. A 40-year perspective on the prevalence of depression: the Stirling County Study. Arch Gen Psychiatry. 2000;57(3):209-15. Epub 2000/03/11.

52. Skoog I. The prevalence of psychotic, depressive and anxiety syndromes in demented and non-demented 85-year-olds. International Journal of Geriatric Psychiatry. 1993;8(3):247-53.

53. Collerton J, Davies K, Jagger C, Kingston A, Bond J, Eccles MP, et al. Health and disease in 85 year olds: baseline findings from the Newcastle 85+ cohort study. BMJ: British Medical Journal. 2009;339.

54. Forsell Y, Jorm AF, von Strauss E, Winblad B. Prevalence and correlates of depression in a population of nonagenarians. Br J Psychiatry. 1995;167(1):61-4. Epub 1995/07/01.

55. Baltes PB, Mayer KU. The Berlin aging study: Aging from 70 to 100: Cambridge University Press; 2001.

56. Hasler G. Pathophysiology of depression: do we have any solid evidence of interest to clinicians? World Psychiatry. 2010;9(3):155-61. Epub 2010/10/27.

57. Broe GA, Jorm AF, Creasey H, Grayson D, Edelbrock D, Waite LM, et al. Impact of chronic systemic and neurological disorders on disability, depression and life satisfaction. Int J Geriatr Psychiatry. 1998;13(10):667-73. Epub 1998/11/18.

58. Penninx BW, Leveille S, Ferrucci L, van Eijk JT, Guralnik JM. Exploring the effect of depression on physical disability: longitudinal evidence from the established populations for epidemiologic studies of the elderly. Am J Public Health. 1999;89(9):1346-52. Epub 1999/09/04.

86

59. Vinkers DJ, Gussekloo J, Stek ML, Westendorp RG, van der Mast RC. Temporal relation between depression and cognitive impairment in old age: prospective population based study. BMJ. 2004;329(7471):881. Epub 2004/09/04.

60. Fujikawa T, Yamawaki S, Touhouda Y. Incidence of silent cerebral infarction in patients with major depression. Stroke. 1993;24(11):1631-4. Epub 1993/11/01.

61. Alexopoulos GS, Bruce ML, Silbersweig D, Kalayam B, Stern E. Vascular depression: a new view of late-onset depression. Dialogues in Clinical Neuroscience. 1999;1(2):68-80. Epub 1999/09/01.

62. Rodda J, Walker Z, Carter J. Depression in older adults. British Medical Journal. 2011;343:d5219. Epub 2011/10/01.

63. Skoog I. Psychiatric disorders in the elderly. Canadian Journal of Psychiatry. 2011;56(7):387-97. Epub 2011/08/13.

64. Blazer DG. Depression in late life: review and commentary. J Gerontol A Biol Sci Med Sci. 2003;58(3):249-65. Epub 2003/03/14.

65. Bschor T, Adli M. Treatment of depressive disorders. Deutsches Arzteblatt international. 2008;105(45):782-92. Epub 2009/07/07.

66. Rimer J, Dwan K, Lawlor DA, Greig CA, McMurdo M, Morley W, et al. Exercise for depression. Cochrane Database Syst Rev. 2012;7:CD004366. Epub 2012/07/13.

67. Åsberg M. Behandling av depressionssjukdomar: en systematisk litteraturöversikt: Sammanfattning och slutsatser: SBU; 2004.

68. Wilson KC, Mottram PG, Vassilas CA. Psychotherapeutic treatments for older depressed people. Cochrane Database Syst Rev. 2008(1):CD004853. Epub 2008/02/07.

69. Wilkinson P, Izmeth Z. Continuation and maintenance treatments for depression in older people. Cochrane Database Syst Rev. 2012;11:CD006727. Epub 2012/11/16.

70. Stek M, Hoogendijk W, Beekman A. Electroconvulsive therapy (ECT) for depression in elderly people. 2009.

87

71. Sherman FT. Life-saving treatment for depression in elderly. Always think of electroconvulsive therapy (ECT). Geriatrics. 2009;64(4):8, 12. Epub 2009/04/30.

72. Boyce RD, Hanlon JT, Karp JF, Kloke J, Saleh A, Handler SM. A review of the effectiveness of antidepressant medications for depressed nursing home residents. J Am Med Dir Assoc. 2012;13(4):326-31. Epub 2011/10/25.

73. Rosenberg PB, Drye LT, Martin BK, Frangakis C, Mintzer JE, Weintraub D, et al. Sertraline for the treatment of depression in Alzheimer disease. Am J Geriatr Psychiatry. 2010;18(2):136-45. Epub 2010/01/21.

74. Banerjee S, Hellier J, Dewey M, Romeo R, Ballard C, Baldwin R, et al. Sertraline or mirtazapine for depression in dementia (HTA-SADD): a randomised, multicentre, double-blind, placebo-controlled trial. The Lancet. 2011;378(9789):403-11.

75. Roose SP, Sackeim HA, Krishnan KR, Pollock BG, Alexopoulos G, Lavretsky H, et al. Antidepressant pharmacotherapy in the treatment of depression in the very old: a randomized, placebo-controlled trial. Am J Psychiatry. 2004;161(11):2050-9. Epub 2004/10/30.

76. Conradsson M, Littbrand H, Lindelöf N, Gustafson Y, Rosendahl E. Effects of a high-intensity functional exercise programme on depressive symptoms and psychological well-being among older people living in residential care facilities: A cluster-randomized controlled trial. Aging Ment Health. 2010;14(5):565-76. Epub 2010/05/25.

77. Adamis D, Treloar A, Martin FC, Macdonald AJ. A brief review of the history of delirium as a mental disorder. History of Psychiatry. 2007;18(4):459-69.

78. Gustafson Y. Acute confusional state (delirium) : clinical studies in hip-fracture and stroke patients. Umeå: Univ.; 1991.

79. Andrew MK, Freter SH, Rockwood K. Prevalence and outcomes of delirium in community and non-acute care settings in people without dementia: a report from the Canadian Study of Health and Aging. BMC Med. 2006;4:15. Epub 2006/06/27.

80. Vilalta-Franch J, Llinas-Regla J, Lopez-Pousa S, Garre-Olmo J. Prevalence and evolution of delirium in a community population

88

of 70 years and older. Actas espanolas de psiquiatria. 2009;37(1):27-33. Epub 2008/03/19. Prevalencia y evolucion del delirium en una muestra comunitaria de 70 anos y mas.

81. Folstein MF, Bassett SS, Romanoski AJ, Nestadt G. The epidemiology of delirium in the community: the Eastern Baltimore Mental Health Survey. International Psychogeriatrics. 1991;3(2):169-76.

82. Rahkonen T, Eloniemi-Sulkava U, Halonen P, Verkkoniemi A, Niinisto L, Notkola IL, et al. Delirium in the non-demented oldest old in the general population: risk factors and prognosis. Int J Geriatr Psychiatry. 2001;16(4):415-21. Epub 2001/05/03.

83. White S, Calver BL, Newsway V, Wade R, Patel S, Bayer A, et al. Enzymes of drug metabolism during delirium. Age Ageing. 2005;34(6):603-8. Epub 2005/11/04.

84. Bucht G, Gustafson Y, Sandberg O. Epidemiology of delirium. Dement Geriatr Cogn Disord. 1999;10(5):315-8. Epub 1999/09/04.

85. Galanakis P, Bickel H, Gradinger R, Von Gumppenberg S, Förstl H. Acute confusional state in the elderly following hip surgery: incidence, risk factors and complications. International Journal of Geriatric Psychiatry. 2001;16(4):349-55.

86. Berggren D, Gustafson Y, Eriksson B, Bucht G, Hansson L-I, Reiz S, et al. Postoperative confusion after anesthesia in elderly patients with femoral neck fractures. Anesthesia & Analgesia. 1987;66(6):497-504.

87. Sandberg O, Gustafson Y, Brännström B, Bucht G. Prevalence of dementia, delirium and psychiatric symptoms in various care settings for the elderly. Scand J Soc Med. 1998;26(1):56-62. Epub 1998/04/04.

88. Mu DL, Wang DX, Li LH, Shan GJ, Li J, Yu QJ, et al. High serum cortisol level is associated with increased risk of delirium after coronary artery bypass graft surgery: a prospective cohort study. Crit Care. 2010;14(6):R238. Epub 2011/01/05.

89. Flacker JM, Lipsitz LA. Neural mechanisms of delirium: current hypotheses and evolving concepts. The Journals of Gerontology Series A: Biological Sciences and Medical Sciences. 1999;54(6):B239-B46.

89

90. Guenther U, Theuerkauf N, Frommann I, Brimmers K, Malik R, Stori S, et al. Predisposing and precipitating factors of delirium after cardiac surgery: a prospective observational cohort study. Annals of surgery. 2013;257(6):1160-7. Epub 2013/02/22.

91. Witlox J, Eurelings LS, de Jonghe JF, Kalisvaart KJ, Eikelenboom P, van Gool WA. Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: a meta-analysis. JAMA. 2010;304(4):443-51. Epub 2010/07/29.

92. de Lange E, Verhaak PF, van der Meer K. Prevalence, presentation and prognosis of delirium in older people in the population, at home and in long term care: a review. Int J Geriatr Psychiatry. 2013;28(2):127-34. Epub 2012/04/20.

93. MacLullich AM, Beaglehole A, Hall RJ, Meagher DJ. Delirium and long-term cognitive impairment. Int Rev Psychiatry. 2009;21(1):30-42. Epub 2009/02/17.

94. Lipowski ZJ. Delirium: acute confusional states: Oxford University Press New York; 1990.

95. Sandberg O, Gustafson Y, Brännström B, Bucht G. Clinical profile of delirium in older patients. Journal of the American Geriatrics Society. 1999;47(11):1300-6.

96. Inouye SK, Bogardus ST, Jr., Charpentier PA, Leo-Summers L, Acampora D, Holford TR, et al. A multicomponent intervention to prevent delirium in hospitalized older patients. N Engl J Med. 1999;340(9):669-76. Epub 1999/03/04.

97. Gustafson Y, Brännström B, Berggren D, Ragnarsson JI, Sigaard J, Bucht G, et al. A geriatric-anesthesiologic program to reduce acute confusional states in elderly patients treated for femoral neck fractures. J Am Geriatr Soc. 1991;39(7):655-62. Epub 1991/07/01.

98. Marcantonio ER, Flacker JM, Wright RJ, Resnick NM. Reducing delirium after hip fracture: a randomized trial. J Am Geriatr Soc. 2001;49(5):516-22. Epub 2001/06/15.

99. Björkelund KB, Hommel A, Thorngren KG, Gustafson L, Larsson S, Lundberg D. Reducing delirium in elderly patients with hip fracture: a multi-factorial intervention study. Acta Anaesthesiol Scand. 2010;54(6):678-88. Epub 2010/03/20.

90

100. Meagher DJ, McLoughlin L, Leonard M, Hannon N, Dunne C, O'Regan N. What Do We Really Know About the Treatment of Delirium With Antipsychotics? Ten Key Issues for Delirium Pharmacotherapy. Am J Geriatr Psychiatry. 2013. Epub 2013/04/10.

101. Bergdahl E, Allard P, Gustafson Y. Depression among the very old with dementia. Int Psychogeriatr. 2010:1-8. Epub 2011/01/06.

102. Potter GG, Steffens DC. Contribution of depression to cognitive impairment and dementia in older adults. The neurologist. 2007;13(3):105-17. Epub 2007/05/15.

103. Andersen K, Lolk A, Kragh-Sorensen P, Petersen NE, Green A. Depression and the risk of Alzheimer disease. Epidemiology. 2005;16(2):233-8. Epub 2005/02/11.

104. Leinonen E, Santala M, Hyotyla T, Santala H, Eskola MN, Salokangas RK. Elderly patients with major depressive disorder and delusional disorder are at increased risk of subsequent dementia. Nord J Psychiatry. 2004;58(2):161-4. Epub 2004/06/19.

105. Modrego PJ, Ferrandez J. Depression in patients with mild cognitive impairment increases the risk of developing dementia of Alzheimer type: a prospective cohort study. Arch Neurol. 2004;61(8):1290-3. Epub 2004/08/18.

106. Garre-Olmo J, Lopez-Pousa S, Vilalta-Franch J, Turon-Estrada A, Hernandez-Ferrandiz M, Lozano-Gallego M, et al. Evolution of depressive symptoms in Alzheimer disease: one-year follow-up. Alzheimer Dis Assoc Disord. 2003;17(2):77-85. Epub 2003/06/10.

107. Mitchell AJ. Depression as a risk factor for later dementia: a robust relationship? Age Ageing. 2005;34(3):207-9. Epub 2005/05/03.

108. Sapolsky RM. Depression, antidepressants, and the shrinking hippocampus. Proceedings of the National Academy of Sciences of the United States of America. 2001;98(22):12320-2. Epub 2001/10/25.

109. Krishnan KR, McDonald WM, Doraiswamy PM, Tupler LA, Husain M, Boyko OB, et al. Neuroanatomical substrates of

91

depression in the elderly. European Archives of Psychiatry and Clinical Neuroscience. 1993;243(1):41-6. Epub 1993/01/01.

110. Robinson TN, Raeburn CD, Tran ZV, Angles EM, Brenner LA, Moss M. Postoperative delirium in the elderly: risk factors and outcomes. Annals of surgery. 2009;249(1):173-8. Epub 2008/12/25.

111. Greene NH, Attix DK, Weldon BC, Smith PJ, McDonagh DL, Monk TG. Measures of executive function and depression identify patients at risk for postoperative delirium. Anesthesiology. 2009;110(4):788-95. Epub 2009/03/28.

112. Davis DH, Muniz Terrera G, Keage H, Rahkonen T, Oinas M, Matthews FE, et al. Delirium is a strong risk factor for dementia in the oldest-old: a population-based cohort study. Brain : a journal of neurology. 2012;135(Pt 9):2809-16. Epub 2012/08/11.

113. von Heideken Wågert P, Gustavsson JM, Lundin-Olsson L, Kallin K, Nygren B, Lundman B, et al. Health status in the oldest old. Age and sex differences in the Umea 85+ Study. Aging Clin Exp Res. 2006;18(2):116-26. Epub 2006/05/17.

114. Mathillas J, Lövheim H, Gustafson Y. Increasing prevalence of dementia among very old people. Age Ageing. 2011;40(2):243-9. Epub 2011/01/25.

115. SCB. Sveriges befolkning, kommunala jämförelsetal, 31 december 2011. 2012; Available from: http://www.scb.se/Pages/ProductTables____25795.aspx.

116. Statistikcentralen. Preliminär folkmängd efter område, oktober 2012. 2012; Available from: http://www.stat.fi/.

117. Institutet för hälsa och välfärd. SOTKAnet. Available from: http://uusi.sotkanet.fi/portal/page/portal/etusivu.

118. Sheikh JI. Geriatric Depression Scale (GDS): recent evidence and development of a shorter version. Clinical Gerontologist. 1986;5(1/2):165.

119. de Craen AJ, Heeren TJ, Gussekloo J. Accuracy of the 15-item geriatric depression scale (GDS-15) in a community sample of the oldest old. International Journal of Geriatric Psychiatry. 2003;18(1):63-6. Epub 2002/12/24.

92

120. Smalbrugge M, Jongenelis L, Pot AM, Beekman AT, Eefsting JA. Screening for depression and assessing change in severity of depression. Is the Geriatric Depression Scale (30-, 15- and 8-item versions) useful for both purposes in nursing home patients? Aging Ment Health. 2008;12(2):244-8. Epub 2008/04/05.

121. Conradsson M, Rosendahl E, Littbrand H, Gustafson Y, Olofsson B, Lövheim H. Usefulness of the Geriatric Depression Scale 15-item version among very old people with and without cognitive impairment. Aging & mental health. 2013. Epub 2013/01/24.

122. Swedish Council on Health Technology Assessment. Diagnostik och uppföljning av förstämningssyndrom- en systematisk litteraturöversikt. 2012.

123. Montgomery SA, Asberg M. A new depression scale designed to be sensitive to change. The British Journal of Psychiatry. 1979;134(4):382-9.

124. Agrell B, Dehlin O. Comparison of six depression rating scales in geriatric stroke patients. Stroke. 1989;20(9):1190-4. Epub 1989/09/01.

125. Kang HJ, Stewart R, Kim JM, Jang JE, Kim SY, Bae KY, et al. Comparative validity of depression assessment scales for screening poststroke depression. J Affect Disord. 2012. Epub 2012/11/22.

126. Inouye SK, van Dyck CH, Alessi CA, Balkin S, Siegal AP, Horwitz RI. Clarifying confusion: the confusion assessment method. A new method for detection of delirium. Ann Intern Med. 1990;113(12):941-8. Epub 1990/12/15.

127. Eriksson M, Samuelsson E, Gustafson Y, Aberg T, Engstrom KG. Delirium after coronary bypass surgery evaluated by the organic brain syndrome protocol. Scand Cardiovasc J. 2002;36(4):250-5. Epub 2002/08/31.

128. Lohr KN, Aaronson NK, Alonso J, Burnam MA, Patrick DL, Perrin EB, et al. Evaluating quality-of-life and health status instruments: development of scientific review criteria. Clin Ther. 1996;18(5):979-92. Epub 1996/09/01.

129. Björkelund KB, Larsson S, Gustafson L, Andersson E. The Organic Brain Syndrome (OBS) scale: a systematic review. Int J Geriatr Psychiatry. 2006;21(3):210-22. Epub 2006/01/28.

93

130. Lawton MP. The Philadelphia Geriatric Center Morale Scale: a revision. Journal of Gerontology. 1975;30(1):85-9. Epub 1975/01/01.

131. Dickinson EJ. Standard assessment scales for elderly people. Recommendations of the Royal College of Physicians of London and the British Geriatrics Society. Journal of epidemiology and community health. 1992;46(6):628-9. Epub 1992/12/01.

132. Fagerström L. Positive life orientation–an inner health resource among older people. Scandinavian journal of caring sciences. 2010;24(2):349-56.

133. Collin C, Wade DT, Davies S, Horne V. The Barthel ADL Index: A reliability study. Disability and Rehabilitation. 1988;10(2):61-3.

134. Mahoney FI, Barthel DW. Functional Evaluation: The Barthel Index. Md State Med J. 1965;14:61-5. Epub 1965/02/01.

135. Wade DT. Measurement in neurological rehabilitation. Current Opinion in Neurology. 1992;5(5):682-6.

136. WHO Collaborating Centre for Drug Statistics Methodology. Guidelines for ATC classification and DDD assignment 2013. Oslo 2012.

137. Rocca WA, Petersen RC, Knopman DS, Hebert LE, Evans DA, Hall KS, et al. Trends in the incidence and prevalence of Alzheimer's disease, dementia, and cognitive impairment in the United States. Alzheimer's & dementia : the journal of the Alzheimer's Association. 2011;7(1):80-93. Epub 2011/01/25.

138. Hall KS, Gao S, Baiyewu O, Lane KA, Gureje O, Shen J, et al. Prevalence rates for dementia and Alzheimer's disease in African Americans: 1992 versus 2001. Alzheimer's & dementia : the journal of the Alzheimer's Association. 2009;5(3):227-33. Epub 2009/05/12.

139. Schrijvers EM, Verhaaren BF, Koudstaal PJ, Hofman A, Ikram MA, Breteler MM. Is dementia incidence declining?: Trends in dementia incidence since 1990 in the Rotterdam Study. Neurology. 2012;78(19):1456-63. Epub 2012/05/04.

140. Taylor DH, Jr., Sloan FA, Doraiswamy PM. Marked increase in Alzheimer's disease identified in medicare claims records

94

between 1991 and 1999. J Gerontol A Biol Sci Med Sci. 2004;59(7):762-6. Epub 2004/08/12.

141. Ownby RL, Crocco E, Acevedo A, John V, Loewenstein D. Depression and risk for Alzheimer disease: systematic review, meta-analysis, and metaregression analysis. Arch Gen Psychiatry. 2006;63(5):530-8. Epub 2006/05/03.

142. Wallin K, Bostrom G, Kivipelto M, Gustafson Y. Risk factors for incident dementia in the very old. Int Psychogeriatr. 2013:1-9. Epub 2013/04/12.

143. Steinberg M, Shao H, Zandi P, Lyketsos CG, Welsh-Bohmer KA, Norton MC, et al. Point and 5-year period prevalence of neuropsychiatric symptoms in dementia: the Cache County Study. Int J Geriatr Psychiatry. 2008;23(2):170-7. Epub 2007/07/04.

144. Barber R, Scheltens P, Gholkar A, Ballard C, McKeith I, Ince P, et al. White matter lesions on magnetic resonance imaging in dementia with Lewy bodies, Alzheimer's disease, vascular dementia, and normal aging. J Neurol Neurosurg Psychiatry. 1999;67(1):66-72. Epub 1999/06/17.

145. Peltonen M, Lundberg V, Huhtasaari F, Asplund K. Marked improvement in survival after acute myocardial infarction in middle-aged men but not in women. The Northern Sweden MONICA study 1985–94. Journal of Internal Medicine. 2000;247(5):579.

146. Banerjee S, Chan J. Organization of old age psychiatric services. Psychiatry. 2008;7(2):49-54.

147. Rodda J, Walker Z. Ten years of cholinesterase inhibitors. Int J Geriatr Psychiatry. 2009;24(5):437-42. Epub 2009/04/16.

148. Nordström P, Religa D, Wimo A, Winblad B, Eriksdotter M. The use of cholinesterase inhibitors and the risk of myocardial infarction and death: a nationwide cohort study in subjects with Alzheimer's disease. Eur Heart J. 2013. Epub 2013/06/06.

149. Vaeroy H, Merskey H. The prevalence of current major depression and dysthymia in a Norwegian general practice. Acta Psychiatr Scand. 1997;95(4):324-8. Epub 1997/04/01.

95

150. Eagles JM, McLeod IH, Douglas AS. Seasonal changes in psychological well-being in an elderly population. Br J Psychiatry. 1997;171:53-5. Epub 1997/07/01.

151. Pallesen S, Nordhus IH, Nielsen GH, Havik OE, Kvale G, Johnsen BH, et al. Prevalence of insomnia in the adult Norwegian population. Sleep. 2001;24(7):771-9. Epub 2001/10/31.

152. Salmela E, Lappalainen T, Liu J, Sistonen P, Andersen PM, Schreiber S, et al. Swedish Population Substructure Revealed by Genome-Wide Single Nucleotide Polymorphism Data. PLoS ONE. 2011;6(2):e16747.

153. Centre for Epidemiology of the National Board of Health and Welfare. Statistics-Health and Diseases. Causes of death 2000. 2002.

154. Fors S, Lennartsson C, Agahi N, Parker MG, Thorslund M. Äldre har fått fler hälsoproblem, men klarar vardagen bättre. Läkartidningen. 2013;110(32-33):1403-5.

155. Läkemedelsindustriföreningens Service AB. FASS.se för förskrivare. 2013; Available from: http://www.fass.se/LIF/home/index.jsp?UserTypeID=0&clickable=0.

156. Rogers D, Pies R. General medical with depression drugs associated. Psychiatry (Edgmont). 2008;5(12):28-41. Epub 2009/09/03.

157. Jorm AF. History of depression as a risk factor for dementia: an updated review. The Australian and New Zealand journal of psychiatry. 2001;35(6):776-81. Epub 2002/05/07.

158. Slomka JM, Piette JD, Post EP, Krein SL, Lai Z, Goodrich DE, et al. Mood disorder symptoms and elevated cardiovascular disease risk in patients with bipolar disorder. J Affect Disord. 2012;138(3):405-8. Epub 2012/02/24.

159. Schillerstrom JE, Royall DR, Palmer RF. Depression, disability and intermediate pathways: a review of longitudinal studies in elders. J Geriatr Psychiatry Neurol. 2008;21(3):183-97. Epub 2008/10/08.

96

160. Braak H, Braak E. Frequency of stages of Alzheimer-related lesions in different age categories. Neurobiol Aging. 1997;18(4):351-7. Epub 1997/07/01.

161. Korczyn AD, Halperin I. Depression and dementia. Journal of the neurological sciences. 2009;283(1-2):139-42. Epub 2009/04/07.

162. Siennicki-Lantz A, Andre-Petersson L, Elmståhl S. Decreasing Blood Pressure Over Time is the Strongest Predictor of Depressive Symptoms in Octogenarian Men. Am J Geriatr Psychiatry. 2013. Epub 2013/04/10.

163. Barrett-Connor E, Palinkas LA. Low blood pressure and depression in older men: a population based study. BMJ. 1994;308(6926):446-9. Epub 1994/02/12.

164. Henderson AS, Korten AE, Jacomb PA, Mackinnon AJ, Jorm AF, Christensen H, et al. The course of depression in the elderly: a longitudinal community-based study in Australia. Psychol Med. 1997;27(1):119-29. Epub 1997/01/01.

165. Garcia-Fabela L, Melano-Carranza E, Aguilar-Navarro S, Garcia-Lara JM, Gutierrez-Robledo LM, Avila-Funes JA. Hypertension as a risk factor for developing depressive symptoms among community-dwelling elders. Revista de investigacion clinica; organo del Hospital de Enfermedades de la Nutricion. 2009;61(4):274-80. Epub 2009/10/24.

166. Paterniti S, Verdier-Taillefer MH, Geneste C, Bisserbe JC, Alperovitch A. Low blood pressure and risk of depression in the elderly. A prospective community-based study. Br J Psychiatry. 2000;176:464-7. Epub 2000/07/27.

167. Whyte EM, Mulsant BH, Vanderbilt J, Dodge HH, Ganguli M. Depression after stroke: a prospective epidemiological study. J Am Geriatr Soc. 2004;52(5):774-8. Epub 2004/04/17.

168. Harris T, Cook DG, Victor C, DeWilde S, Beighton C. Onset and persistence of depression in older people--results from a 2-year community follow-up study. Age Ageing. 2006;35(1):25-32. Epub 2005/11/24.

169. Wilson K, Mottram P, Sixsmith A. Depressive symptoms in the very old living alone: prevalence, incidence and risk factors. International Journal of Geriatric Psychiatry. 2007;22(4):361-6. Epub 2006/10/18.

97

170. Horwath E, Johnson J, Klerman GL, Weissman MM. Depressive symptoms as relative and attributable risk factors for first-onset major depression. Archives of General Psychiatry. 1992;49(10):817-23. Epub 1992/10/01.

171. Boorsma M, Joling KJ, Frijters DH, Ribbe ME, Nijpels G, van Hout HP. The prevalence, incidence and risk factors for delirium in Dutch nursing homes and residential care homes. Int J Geriatr Psychiatry. 2012;27(7):709-15. Epub 2011/09/16.

172. McCusker J, Cole MG, Voyer P, Monette J, Champoux N, Ciampi A, et al. Prevalence and incidence of delirium in long-term care. Int J Geriatr Psychiatry. 2011. Epub 2011/01/29.

173. Juliebø V, Krogseth M, Neerland BE, Watne LO, Wyller TB. Delirium–a common condition associated with negative outcome in the elderly. Norsk epidemiologi. 2012;22(2).

174. Gustafson Y, Olsson T, Eriksson S, Asplund K, Bucht G. Acute confusional states (delirium) in stroke patients. Cerebrovasc Dis. 1991;1(5):257-64.

175. Fick DM, Agostini JV, Inouye SK. Delirium superimposed on dementia: a systematic review. J Am Geriatr Soc. 2002;50(10):1723-32. Epub 2002/10/09.

176. Lundström M, Edlund A, Bucht G, Karlsson S, Gustafson Y. Dementia after delirium in patients with femoral neck fractures. J Am Geriatr Soc. 2003;51(7):1002-6. Epub 2003/07/02.

177. McAvay GJ, Van Ness PH, Bogardus ST, Jr., Zhang Y, Leslie DL, Leo-Summers LS, et al. Depressive symptoms and the risk of incident delirium in older hospitalized adults. J Am Geriatr Soc. 2007;55(5):684-91. Epub 2007/05/12.

178. Leung JM, Sands LP, Mullen EA, Wang Y, Vaurio L. Are preoperative depressive symptoms associated with postoperative delirium in geriatric surgical patients? J Gerontol A Biol Sci Med Sci. 2005;60(12):1563-8. Epub 2006/01/21.

179. Elie M, Cole MG, Primeau FJ, Bellavance F. Delirium risk factors in elderly hospitalized patients. J Gen Intern Med. 1998;13(3):204-12. Epub 1998/04/16.

98

180. Edwards N. Differentiating the three D's: delirium, dementia, and depression. Medsurg nursing : official journal of the Academy of Medical-Surgical Nurses. 2003;12(6):347-57; quiz 58. Epub 2004/01/17.

181. Chang-Quan H, Xue-Mei Z, Bi-Rong D, Zhen-Chan L, Ji-Rong Y, Qing-Xiu L. Health status and risk for depression among the elderly: a meta-analysis of published literature. Age Ageing. 2010;39(1):23-30. Epub 2009/11/12.

182. Sharp SI, Aarsland D, Day S, Sonnesyn H, Ballard C. Hypertension is a potential risk factor for vascular dementia: systematic review. Int J Geriatr Psychiatry. 2011;26(7):661-9. Epub 2011/04/16.

183. Skoog I, Lernfelt B, Landahl S, Palmertz B, Andreasson LA, Nilsson L, et al. 15-year longitudinal study of blood pressure and dementia. Lancet. 1996;347(9009):1141-5. Epub 1996/04/27.

184. van Dijk EJ, Breteler MM, Schmidt R, Berger K, Nilsson LG, Oudkerk M, et al. The association between blood pressure, hypertension, and cerebral white matter lesions: cardiovascular determinants of dementia study. Hypertension. 2004;44(5):625-30. Epub 2004/10/07.

185. Debette S, Markus HS. The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis. BMJ. 2010;341:c3666. Epub 2010/07/28.

186. O'Brien J, Desmond P, Ames D, Schweitzer I, Harrigan S, Tress B. A magnetic resonance imaging study of white matter lesions in depression and Alzheimer's disease. Br J Psychiatry. 1996;168(4):477-85. Epub 1996/04/01.

187. Soiza RL, Sharma V, Ferguson K, Shenkin SD, Seymour DG, Maclullich AM. Neuroimaging studies of delirium: a systematic review. J Psychosom Res. 2008;65(3):239-48. Epub 2008/08/19.

188. Herrmann LL, Le Masurier M, Ebmeier KP. White matter hyperintensities in late life depression: a systematic review. J Neurol Neurosurg Psychiatry. 2008;79(6):619-24. Epub 2007/08/25.

99

189. Tupler LA, Krishnan KR, McDonald WM, Dombeck CB, D'Souza S, Steffens DC. Anatomic location and laterality of MRI signal hyperintensities in late-life depression. J Psychosom Res. 2002;53(2):665-76. Epub 2002/08/10.

190. Gunning-Dixon FM, Walton M, Cheng J, Acuna J, Klimstra S, Zimmerman ME, et al. MRI signal hyperintensities and treatment remission of geriatric depression. J Affect Disord. 2010;126(3):395-401. Epub 2010/05/11.

191. Paran E, Anson O, Reuveni H. Blood pressure and cognitive functioning among independent elderly. Am J Hypertens. 2003;16(10):818-26. Epub 2003/10/14.

192. Molander L, Gustafson Y, Lövheim H. Low blood pressure is associated with cognitive impairment in very old people. Dement Geriatr Cogn Disord. 2010;29(4):335-41. Epub 2010/04/15.

193. Morris MC, Scherr PA, Hebert LE, Bennett DA, Wilson RS, Glynn RJ, et al. The cross-sectional association between blood pressure and Alzheimer's disease in a biracial community population of older persons. J Gerontol A Biol Sci Med Sci. 2000;55(3):M130-6. Epub 2000/05/05.

194. Ruitenberg A, Skoog I, Ott A, Aevarsson O, Witteman JC, Lernfelt B, et al. Blood pressure and risk of dementia: results from the Rotterdam study and the Gothenburg H-70 Study. Dement Geriatr Cogn Disord. 2001;12(1):33-9. Epub 2000/12/23.

195. Molander L. Blood pressure in advanced age : with focus on epidemiology, cognitive impairment and mortality. Umeå: Umeå Universitet; 2010.

196. Guerry JD, Hastings PD. In search of HPA axis dysregulation in child and adolescent depression. Clinical child and family psychology review. 2011;14(2):135-60. Epub 2011/02/04.

197. Swaab DF, Bao AM, Lucassen PJ. The stress system in the human brain in depression and neurodegeneration. Ageing research reviews. 2005;4(2):141-94. Epub 2005/07/06.

198. Sapolsky RM, Pulsinelli WA. Glucocorticoids potentiate ischemic injury to neurons: therapeutic implications. Science. 1985;229(4720):1397-400. Epub 1985/09/27.

100

199. Gustafson Y, Olsson T, Asplund K, Hägg E. Acute confusional state (delirium) soon after stroke is associated with hypercortisolism. Cerebrovasc Dis. 2010;3(1):33-8.

200. Sorrells SF, Caso JR, Munhoz CD, Sapolsky RM. The stressed CNS: when glucocorticoids aggravate inflammation. Neuron. 2009;64(1):33-9. Epub 2009/10/21.

201. Byers AL, Yaffe K. Depression and risk of developing dementia. Nature reviews Neurology. 2011;7(6):323-31. Epub 2011/05/04.

202. Simone MJ, Tan ZS. The role of inflammation in the pathogenesis of delirium and dementia in older adults: a review. CNS neuroscience & therapeutics. 2011;17(5):506-13. Epub 2010/06/18.

203. Holmes C, Cunningham C, Zotova E, Woolford J, Dean C, Kerr S, et al. Systemic inflammation and disease progression in Alzheimer disease. Neurology. 2009;73(10):768-74. Epub 2009/09/10.

204. Schiepers OJ, Wichers MC, Maes M. Cytokines and major depression. Progress in neuro-psychopharmacology & biological psychiatry. 2005;29(2):201-17. Epub 2005/02/08.

205. Cunningham C, Campion S, Lunnon K, Murray CL, Woods JF, Deacon RM, et al. Systemic inflammation induces acute behavioral and cognitive changes and accelerates neurodegenerative disease. Biol Psychiatry. 2009;65(4):304-12. Epub 2008/09/20.

206. van Gool WA, van de Beek D, Eikelenboom P. Systemic infection and delirium: when cytokines and acetylcholine collide. Lancet. 2010;375(9716):773-5. Epub 2010/03/02.

207. Rosas-Ballina M, Tracey KJ. Cholinergic control of inflammation. J Intern Med. 2009;265(6):663-79. Epub 2009/06/06.

208. Mirrakhimov AE, Brewbaker CL, Krystal AD, Kwatra MM. Obstructive sleep apnea and delirium: exploring possible mechanisms. Sleep & breathing = Schlaf & Atmung. 2013. Epub 2013/04/16.

101

209. Harris M, Glozier N, Ratnavadivel R, Grunstein RR. Obstructive sleep apnea and depression. Sleep medicine reviews. 2009;13(6):437-44. Epub 2009/07/15.

210. Sandberg O, Franklin KA, Bucht G, Gustafson Y. Sleep apnea, delirium, depressed mood, cognition, and ADL ability after stroke. J Am Geriatr Soc. 2001;49(4):391-7. Epub 2001/05/12.

211. Kielb SA, Ancoli-Israel S, Rebok GW, Spira AP. Cognition in obstructive sleep apnea-hypopnea syndrome (OSAS): current clinical knowledge and the impact of treatment. Neuromolecular medicine. 2012;14(3):180-93. Epub 2012/05/10.

212. Bergdahl E, Allard P, Lundman B, Gustafson Y. Depression in the oldest old in urban and rural municipalities. Aging Ment Health. 2007;11(5):570-8. Epub 2007/09/21.

213. Cole MG, Yaffe MJ. Pathway to psychiatric care of the elderly with depression. International Journal of Geriatric Psychiatry. 1996;11(2):157-61.

214. Reynolds CF, 3rd, Frank E, Dew MA, Houck PR, Miller M, Mazumdar S, et al. Treatment of 70(+)-year-olds with recurrent major depression. Excellent short-term but brittle long-term response. Am J Geriatr Psychiatry. 1999;7(1):64-9. Epub 1999/01/27.

Date post:	27-Jun-2020
Category:	Documents
Upload:	others
View:	0 times
Download:	0 times

Dementia, depression and delirium in the very...

Documents