Title

Validation of known risk factors and socioeconomic status associated with Parkinson’s

disease: A retrospective nationwide 11-year population-based cohort study in South Korea

Hyoung Seop Kim M.D. 4Jong Moon Kim M.D.1, Junbeom Kwon, M.D. 2, Hyunsun Lim

Ph.D. 3,

4 Department of Physical Medicine and Rehabilitation, National Health Insurance Service

Ilsan Hospital

1Department of Rehabilitation Medicine, CHA Bundang Medical Center, CHA University

Rehabilitation and Regeneration Research Center, CHA University

2 Department of Rehabilitation Medicine and Research Institute of Rehabilitation Medicine,

Yonsei University College of Medicine & Severance Hospital, Seoul, Korea.

3 Research and Analysis Team, National Health Insurance Service Ilsan Hospital, Goyang,

Gyeonggi-do, Korea.

Corresponding author: Hyoung Seop Kim

E-mail address: rehappydoc@gmail.com

Departments of Physical Medicine and Rehabilitation, National Health Insurance Ilsan

Hospital, 100 Ilsan-roilsan-donggu, Goyang10444,

Korea Tel : +82-31-900-0137

Fax : +82-31-900-0343

Disclosure of Conflicts of Interest: Level 0 (no funding)

1

Author Contributions: Dr. Jong Moon Kim and Hyoung Seop Kim had full access to all the

data in the study and take responsibility for the integrity of the data and the accuracy of the

data analysis. Dr. Junbeom Kwon performed data analysis and interpretation. Hyun Sun Lim

provided statistical working.

Conflict of Interest: None of authors has conflicts of interest to report.

2

Abstract

Objective: To validate various known risk factors of Parkinsonism and to establish basic

information to present public health policy with 10-year follow-up cohort model.

Methods: This population based nation-wide study was performed in the republic of South

Korea using the database of regular health check in 2003 and 2004 of National Health

Insurance Database of reimbursement claim of Health Insurance Review and Assessment

Service of South Korea with 10 years follow up.

Results: The authors identified 7,746 patients with Parkinsonism. The old age, hypertension,

diabetes, depression, anxiety, taking statin medication, high body mass index, non-smoking,

non-alcohol drinking and low socioeconomic status were each associated with increase in risk

of Parkinsonism (fully adjusted Cox proportional hazards model: hazard ratio (HR) 1.259,

95% confidence interval (CI) 1.194 to 1.328 for hypertension, HR 1.255, 95% CI 1.186 to

1.329 for diabetes, HR 1.554, 95% CI 1.664 to 1.965 for depression, HR 1.808, 95% CI 1.462

to 1.652 for anxiety, and HR 1.157, 95% CI 1.072 to 1.250 for taking statin).

Conclusions: In our study, old age, depression, anxiety, and nonsmoker were found to be risk

factors of Parkinsonism, coincided with previous studies. However, gender, hypertension,

diabetes, taking statin medication, non-alcohol drinking, and lower socioeconomic status

differ from some previous studies and need to be verified in subsequent studies.

Key word: Parkinsonism, Socioeconomic Status, Risk factor, National Health Programs

3

Introduction

Parkinsonism is one of the most common neurodegenerative diseases and affects more than

1% of the elderly population.1 As the mean age of the population worldwide increases

dramatically, the prevalence and incidence of Parkinsonism also have been increasing.2

Therefore, the impact of Parkinsonism on global society is a major concern and this situation

is no exception in South Korea. On the strength of these trends, in recent years, many genetic

factors have been discovered. 3 However, Parkinsonism is a complex of variable etiology

including both genetic and environmental factors and genetic factors only accounts for a

minor portion of Parkinsonism. Therefore, Majority of etiologies is still unknown and

investigating its distribution and characteristics within the nation has occurred to be important

to find new etiologic factors and plan the public health systems.4

The South Korea medical service delivery system is National health insurance. In South

Korea, most people are obligated to enroll in the national health insurance service. All

medical health insurance related data have been being collected into database centrally. In

addition, Koreans will receive regular health checkups with the support of the National

Health Insurance Corporation after age 40. There have been many studies on the relationship

between the occurrence of Parkinsonism and some of the probable risk factors.5-8 Well known

risk factors of Parkinsonism include increasing Age5, alcohol use disorder6, urbanization and

exposure to pesticides9, cardiovascular physiologic change7, Depression10, Anxiety11 but

Smoking has been proven to a preventive factor to the development of Parkinsonism.12, 13

However, the relationship between socioeconomic status and the onset of Parkinsonism is

unknown.

However, we can verify the relationship between known risk factors, associated problems,

and the onset of Parkinsonism using both databases from the regular health checkup and

health insurance of South Korea. Therefore, we aimed to validate various known risk factors

4

of Parkinsonism and to establish basic information to present public health policy with 10-

year follow-up cohort model.

Methods

Statement of Ethics

This research project was approved by the national health insurance service of Korea (The

research management number is NHIS-2017-2-542). This study was reviewed and approved

by the Institutional Review Board of our hospital adhered to the tenets of the Declaration of

Helsinki. Informed consent was waived. All authors contributed to the study design,

interpretation of the results, and made the decision to submit the manuscript for publication.

No commercial support for this study was obtained.

Database

The Management of National Health Insurance Service is divided into two independent

institutions; national health insurance corporation (NHIC) and Health Insurance Review and

Assessment Service of Korea (HIRA). NHIC has accumulated data (National Health

Insurance Database; NHID) of the information of insured person, premium imposition,

regular health check-up and HIRA has accumulated data on health insurance claims, which

are accompanied by data regarding diagnostic codes, procedures, prescription drugs, personal

information, information about the hospital, the direct medical costs of both inpatient and

outpatient care and dental services.14 Therefore, we utilized the combined data from the

database of regular health check in 2003 and 2004 of NHIC and the database of

reimbursement claim of HIRA from 2003 to 2013.

5

Blue collar worker can take regular health checkup every year, and white collar worker and

self-employed insurance holders or beneficiaries such as independent businessmen, farmers,

fishermen housewives, and retiree every other year.

The following are regular health checkup items;

1. Body index measurements: height, body weight, waist circumference, Body mass index

(BMI); Based on the Asian standard, BMI is classified into five grades; below 18.5 as

underweight, 18.5-22.9 as normal, 22.9-24.9 as overweight, 25-29.9 as moderate

obesity, and 30-35 severe obesity.

2. History taking: smoking, alcohol history, medication; Smoking history is classified into

a non-smoker, ex-smoker, current smoker. Drinking history is divided into the five

grades as no drinking, 2-3 times a week, 1-2 times a week, 3-4 times a week, drinking

almost every day. Medication history includes the drugs of hypertension, diabetes, and

hyperlipidemia.

3. Screening test; an eye test, hearing test, oral hygiene test, laboratory tests (total

cholesterol, aspartate aminotransferase, alanine aminotransferase, gamma- guanosine

triphosphate, fasting blood glucose, urine protein, plasma creatinine, hemoglobin),

chest X-ray

After the first health checkup, a second health check-up will be done for patients suspected of

hypertension and Diabetes.

Study Sample

KNHIS and HIRA use the Korean Classification of Diseases (KCD) as disease classification

codes, which are modified from International Classification of Diseases. The 5,147,950

people who received regular health checkup from 2002 to 2003, of whom randomly selected

10%, 514,795 subjects were included in this study.

6

Based on data from HIRA, we defined patients with Parkinsonism who have diagnostic

codes (KCD); Parkinsonism or Parkinson disease (Not otherwise specific, idiopathic,

primary) (G20) But we excluded the codes of secondary parkinsonism (G21) and

parkinsonism in disease classified elsewhere (G22).

Definition of Risk factors

We defined risk factors as age, hypertension, diabetes mellitus (DM), alcohol drinking,

smoking, socioeconomic status, statin medication, depression, and anxiety. The presence or

absence of the disease or risk factor was determined by the presence of codes in health

insurance claims; hypertension (I10 to I15), DM (E10 to E14), depression (F32, F33) and

anxiety (F40 or F41). Statin medication was defined if there were drug codes of statins in

health insurance claims

Based on the Asian standard, BMI is classified into the five grades; below 18.5 as

underweight, 18.5-22.9 as normal, 23-24.9 as overweight, 25-29.9 as moderate obesity, and

30-35 severe obesity. Smoking history is classified into a non-smoker, ex-smoker, and current

smoker. Alcohol drinking is classified into five grades; no drinking, 2-3times/month, 1-

2times/week, 3-4times/week and Daily.

Classification of socioeconomic status by health insurance premium rate

We assumed that the premium rate reflects the socioeconomic status of subjects. Health

insurance premiere payment system is divided into three types: “self-employed insured”,

“employed insured” and Medicaid. Medicaid provides medical care for the old or the disabled

who have a little income or property. The population on Medicaid pays a part or none of their

medical bills, although there are regulations and legal limits to the use of the medical system.

The premium rate of employees is set by the standard based on monthly salary but the income

7

of self-employee could not be detected exactly, the self-employed insured premium rate is set

by the conversion points which include the insurance holder’s property such the cost of

house, possession of a car or not, economic activity by age and gender, and total income.

Health insurance premium rate is divided into ten quantiles in each type of premium

payment. We divided socioeconomic status into the nine groups according to the premium

payment system (self-employed, employed, Medicaid), premium rate (four grades) (Table 1).

Data processing and Statistical analysis

As previously mentioned, we created a 10-year follow-up cohort model for random sampling

of 10% of the total subjects enrolled in regular health checkup in 2002 and 2003 (Figure). We

defined the onset of Parkinsonism if there were the Parkinson’s disease diagnostic code

(G.20) of the main or second disease in health insurance claims.

Descriptive statistics of the study populations are presented, and chi-square tests were

performed to examine the association of risk factors with Parkinsonism. To identify the

hazard associated with Parkinsonism, adjusted hazard ratios (HRs) and 95% confidence

intervals (CI) were calculated via multivariate Cox proportional hazard regression was

performed. A significant level of 0.05 was selected. The statistical packages SAS system for

Windows, version 9.2 (SAS Inc, Cary, NC) was used to perform the analyses in this study.

Results

Table 2 shows the demographic data and the incidence of Parkinsonism according to each

group in the subjects who received regular health checkup in 2002 and 2003. The HRs of the

subjects aged 50-59, 60-69, 70-79, and 80 years or older were 3.101, 8.958, 14.709, and

16.797, respectively, and statistically significant (p < 0.0001). In comparison between both

8

genders, the HR of females was 0.971, which was statistically insignificant (p = 0.3273)

(Table 3). In the case of hypertension, diabetes, depression, and anxiety, the hazards ratios

were 1.259, 1.255, 1.554 and 1.808, respectively, (p < 0.0001). Also, the HR of the group

taking Statins was 1.157, which was higher than that of the group not taking Statins at the

time of diagnosis.

The risk of Parkinsonism was higher in the highest BMI group than in the normal weight

group (p <0.0001). The HRs of ex-smokers and current smokers rate were same 0.920, but

statistically significant for only current smokers. (p < 0.0287) The HRs of all groups in

Alcohol drinking were less than 1 and statistically significant (p < 0.0001), except for daily

alcohol drinking group (p = 0.6530). As Medicaid was set as the standard, HR was less than 1

in all groups (p < 0.05), indicating that socioeconomic status and Parkinsonism were closely

related.

Discussion

The results of our study revealed that increasing age, hypertension, DM, depression, anxiety,

extreme overweight, statin medication, non-smoker, non-alcohol drinker, and the lowest

socioeconomic class were statistically related to the onset of Parkinsonism.

Parkinsonism is the second most neurodegenerative disease, which has been reported with an

increasing age in all studies.4 In our study, there was no difference in the incidence of

Parkinsonism between both genders, but one review article reported that males have a 1.5 to

2 fold increased risk of developing Parkinsonism than that of females.4 However, some large

population-based cohort studies on the gender and Parkinsonism have been inconsistent. One

study reported that men are more likely to have Parkinsonism than women5, and another did

the opposite15, whereas the other found no association between both genders.5, 15, 16 Therefore,

9

a meta-analysis including large population-based cohort studies is necessary to identify the

incidence of Parkinsonism according to gender.

Hypertension can cause ischemic cerebrovascular lesions which involve dopaminergic or

non-dopaminergic subcortical structures. It also can cause hypertensive vasculopathy in basal

ganglia, which may injury the dopaminergic cells in the pars compacta and develop

Parkinsonism by breaking the neuron connection between substantia nigra and putamen or

decreasing the expression of the -2, -4 subunit of the nicotinic acetylcholine receptor that

activates the dopaminergic pathway.17 However, most of the studies related to hypertension

and incidence of Parkinsonism were contrary to our findings7, 13, 18, 19 or showed no

relationship between hypertension and developing Parkinsonism.20 This is considered to be

related to the pathophysiology and progress of Parkinsonism.  Autonomic nervous

dysfunction in Parkinsonism is a very common feature.19 Loss of sympathetic cardiac

innervation in Parkinsonism causes changes in cardiovascular physiology and it may precede

a diagnosis of Parkinsonism.7, 21, 22 Because the Lewy body pathology involves the dorsal

motor nucleus of the vagus 23, the parasympathetic tone become out of control, leading to

orthostatic hypotension and decrease of heart rate variability, which are typical symptoms of

autonomic dysfunction in Parkinsonism.24

In our study, DM was found to be a risk factor for the development of Parkinsonism.

Previous cross-sectional studies have a limitation to prove causality but the results of them

did not show coincidence. In one study, the patients with Parkinsonism were more likely to

have DM, but the other study showed opposite results.19, 25 However, one large 18-year

prospective cohort study reported that type 2 DM was a risk factor for Parkinsonism

development so it was consistent with our findings.26 Although the relationship between

diabetes and Parkinsonism is unclear, animal or in vitro studies have shown that insulin and

brain dopaminergic activity are interrelated in brain tissue.27 So insulin dysregulation and the

10

change of insulin action are assumed to affect pathophysiology and clinical symptoms of

Parkinsonism. However, hypertension and DM also increase with aging likewise

Parkinsonism, except for type 1 DM in childhood and secondary hypertension in the young.

For this reason, we assume that hypertension and DM can be risk factors for developing

Parkinsonism.

Depression and anxiety are well known as common non-motor symptoms of Parkinsonism.10

Depression is known to be an early marker of Parkinsonism pathogenesis and a significant

involvement of dopaminergic neurons in the substantia nigra10, 28 and anxiety is an early

symptom of Parkinsonism and is known to involve noradrenergic and serotonergic neurons in

the brainstem.29

Previous studies have shown that depression and anxiety are a significant association with

Parkinsonism.9, 10, 18, 20, 28 However, except for one cohort study involving anxiety and

Parkinsonism onset30, the others were all cross-sectional9, 10, 18, 20, 28, which are difficult to

distinguish between risk factors for the onset of Parkinsonism or non-motor symptoms of

Parkinsonism. But our study proved that depression and anxiety are the risk factors for

Parkinsonism onset.

Cholesterol is abundant in the central nervous system and plays an important protective role

in early Parkinsonism development. Higher serum cholesterol and serum triglyceride have

been reported to reduce the risk of Parkinsonism.31 Statin is a 3-hydroxy-3-methylglutaryl-

coenzyme A reductase inhibitor that inhibits the synthesis of cholesterol.32 Similar to the

result of our study, there is also a prospective study that the use of statin significantly

increased the risk of Parkinsonism.

However, the studies about statin medication and development of Parkinsonism have been

also inconsistent.33, 34 Most of the studies were cross-sectional and retrospective, making it

difficult to identify causal relationships, but recent meta-analysis reported that statin

11

medication reduced the onset of Parkinsonism.35

 Obesity is a well-known risk factor for metabolic and vascular disorders such as type 2

diabetes, coronary heart disease, and stroke.36 So obesity has been suspected as a risk factor

for Parkinsonism and this study showed that severe obesity elevated the risk of Parkinsonism

significantly. The suggested mechanism is that obese persons have lower availability of

dopamine D2 receptor in the striatum than non-obese control.11 However, one meta-analysis

reported that higher BMI did not increase Parkinsonism risk36, which is still controversial in

relation to obesity and the risk of Parkinsonism.37

Smoking is a powerful risk factor for hypertension, atherosclerosis, ischemic and

hemorrhagic stroke, but is well established as a preventive factor for Parkinsonism. In our

study, the same results were obtained. Non-Smokers easily developed Parkinsonism

compared to ex-smokers and current smokers.12, 13, 18, 22, 28, 37 The hypotheses about the

relationship between Smoking and Parkinsonism are as follows. Smoking increases dopamine

activity by reducing MAO-B activity.12, 28 Cytochrome P-450 families are responsible for the

metabolism and detoxification of environmental toxin to cause dopaminergic neural damage.

Smoking induces cytochrome P-450 as a result of polycyclic hydrocarbons such as

benzopyrene in cigarette smoke.12, 38

In our study, the risk of Parkinsonism statistically significantly was reduced in the alcohol

drinking groups except for the everyday alcohol drink group, compared with the non-alcohol

drinker group.39 One study reported alcohol drinking was associated with increasing the risk

of Parkinsonism37, other studies have reported the same results with our study.13, 18 One

review article concluded the weak protective association between alcohol drinking and

Parkinsonism risk was found in studies at greater risk of selection and recall bias. One study

found that the risk of Parkinsonism was elevated in patients with alcohol abuse disorder6 and

our study showed that there was no effect of lowering the risk of Parkinsonism in the

12

everyday alcohol drinking group. Therefore, considering previous studies as well as ours, we

could expect that proper alcohol drinking is associated with lowering Parkinsonism.

One study reported that the incidence of Parkinsonism was not affected by socioeconomic

status.5 However, our study showed that Parkinsonism risk was reduced in all health

insurance payment groups, compared to Medicaid. This is probably due to that the standards

of the socioeconomic status of the two studies were different and Medicaid subscribers in

South Korea mainly composed of the old with no economic ability.

The strong points of our study are the nationwide 10-year follow-up cohort model with a

population over 500,000, and the quality of the data is relatively objective and accurate using

the regular health checkup data and the disease diagnostic codes from HIRA. HIRA reviews

claims with disease codes to determine whether the reimbursement is clinically valid or

unnecessary medical services were provided. So, HIRA can maintain the quality of health

care and provide the standard medical service guideline about each disease. There is little

chance that medical record of the duplicated or omitted because all Korean residents have

received a unique identification number at birth.

The relationship between the onset of Parkinsonism, various comorbidities, body indexes,

and various known risk factors could be verified at once. It is pointed out as a strong point

that socioeconomic status by the health premium payment methods was utilized. Our study is

also a retrospective cohort model and not a cross-sectional, so it is more useful for

identification of causality.

The limitations can be pointed out as follows; the subjects under 40 years of age are not

included in the study because the subjects for regular health checkup are in their 40s or older.

In a general multicenter prospective cohort study, the same diagnostic criteria are set and the

data are collected. However, since the nationwide data cohort model uses the health insurance

related database, it is very likely that the same diagnostic criteria were not applied. Smoking,

13

BMI, and premium payment form are not fixed but may change during the follow-up period.

For example, if an employee becomes unemployed or starts his or her business, the payment

system will change from Employed insured to Self-insured, and in extreme cases, if the

insurance holder has no earned income or property, premium payment form may change into

Medicaid.

Conclusion

In our study, increasing age, depression, anxiety, and nonsmoker were found to be risk factors

of Parkinsonism, coincided with previous studies. However, gender, hypertension, DM, statin

medication, alcohol drinking, and lower socioeconomic status differ from some previous

studies and need to be verified in subsequent studies. This study was the first to verify the

known risk factors of Parkinsonism and socioeconomic status using NHID of South Korea.

By being compared with the results of studies in other countries, it will be helpful to

understand the pathophysiology and epidemiology of Parkinsonism and to present Korean

public health policy. In order to verify more risk factors of Parkinsonism, it is necessary to

carry out studies combined with NHID and genetic and environmental factors in near future.

14

Reference

Uncategorized References1. Muangpaisan W, Hori H Fau - Brayne C, Brayne C. Systematic review of the prevalence and incidence of Parkinson's disease in Asia.2. Yamawaki M, Kusumi M Fau - Kowa H, Kowa H Fau - Nakashima K, Nakashima K. Changes in prevalence and incidence of Parkinson's disease in Japan during a quarter of a century.3. Polymeropoulos MH, Lavedan C, Leroy E, et al. Mutation in the α-synuclein gene identified in families with Parkinson's disease. science 1997;276:2045-2047.4. Tan LC. Epidemiology of Parkinson's disease. Neurology Asia 2013;18:231-238.5. Caslake R, Taylor K Fau - Scott N, Scott N Fau - Gordon J, et al. Age-, gender-, and socioeconomic status-specific incidence of Parkinson's disease and parkinsonism in northeast Scotland: the PINE study.6. Eriksson AK, Lofving S Fau - Callaghan RC, Callaghan Rc Fau - Allebeck P, Allebeck P. Alcohol use disorders and risk of Parkinson's disease: findings from a Swedish national cohort study 1972-2008.7. Jain S, Ton Tg Fau - Perera S, Perera S Fau - Zheng Y, et al. Cardiovascular physiology in premotor Parkinson's disease: a neuroepidemiologic study.8. Qiu C, Hu G Fau - Kivipelto M, Kivipelto M Fau - Laatikainen T, et al. Association of blood pressure and hypertension with the risk of Parkinson disease: the National FINRISK Study.9. Dick FD, De Palma G Fau - Ahmadi A, Ahmadi A Fau - Scott NW, et al. Environmental risk factors for Parkinson's disease and parkinsonism: the Geoparkinson study.10. Fang F, Xu Q Fau - Park Y, Park Y Fau - Huang X, et al. Depression and the subsequent risk of Parkinson's disease in the NIH-AARP Diet and Health Study.11. Ishihara-Paul L, Wainwright Nw Fau - Khaw KT, Khaw Kt Fau - Luben RN, et al. Prospective association between emotional health and clinical evidence of Parkinson's disease.12. Allam MF, Campbell Mj Fau - Hofman A, Hofman A Fau - Del Castillo AS, Del Castillo As Fau - Fernandez-Crehuet Navajas R, Fernandez-Crehuet Navajas R. Smoking and Parkinson's disease: systematic review of prospective studies.13. Paganini-Hill A. Risk factors for parkinson's disease: the leisure world cohort study.14. Lee J, Lee JS, Park S-H, Shin SA, Kim K. Cohort profile: The national health insurance service–national sample cohort (NHIS-NSC), South Korea. International journal of epidemiology 2016;46:e15-e15.15. Winter Y, Bezdolnyy Y Fau - Katunina E, Katunina E Fau - Avakjan G, et al. Incidence of Parkinson's disease and atypical parkinsonism: Russian population-based study.16. Linder J, Stenlund H Fau - Forsgren L, Forsgren L. Incidence of Parkinson's disease and parkinsonism in northern Sweden: a population-based study.17. Tohgi H, Utsugisawa K Fau - Yoshimura M, Yoshimura M Fau - Nagane Y, Nagane Y Fau - Mihara M, Mihara M. Alterations with aging and ischemia in nicotinic acetylcholine receptor subunits alpha4 and beta2 messenger RNA expression in postmortem human putamen. Implications for susceptibility to parkinsonism.18. Noyce AJ, Bestwick Jp Fau - Silveira-Moriyama L, Silveira-Moriyama L Fau - Hawkes CH, et al. PREDICT-PD: identifying risk of Parkinson's disease in the community: methods and baseline results.19. Scigliano G, Musicco M Fau - Soliveri P, Soliveri P Fau - Piccolo I, Piccolo I Fau - Ronchetti G, Ronchetti G Fau - Girotti F, Girotti F. Reduced risk factors for vascular disorders in Parkinson disease patients: a case-control study.20. Huang YF, Cherng YG, Hsu SP, et al. Risk and adverse outcomes of fractures in patients with Parkinson's disease: two nationwide studies.21. Goldstein DS. Dysautonomia in Parkinson's disease: neurocardiological abnormalities.

15

22. Vikdahl M, Backman L, Johansson I, Forsgren L, Haglin L. Cardiovascular risk factors and the risk of Parkinson's disease.23. Braak H, Ghebremedhin E Fau - Rub U, Rub U Fau - Bratzke H, Bratzke H Fau - Del Tredici K, Del Tredici K. Stages in the development of Parkinson's disease-related pathology.24. Valappil RA, Black Je Fau - Broderick MJ, Broderick Mj Fau - Carrillo O, et al. Exploring the electrocardiogram as a potential tool to screen for premotor Parkinson's disease.25. Powers KM, Smith-Weller T, Franklin GM, Longstreth W, Swanson PD, Checkoway H. Diabetes, smoking, and other medical conditions in relation to Parkinson's disease risk. Parkinsonism & related disorders 2006;12:185-189.26. Hu G, Jousilahti P Fau - Bidel S, Bidel S Fau - Antikainen R, Antikainen R Fau - Tuomilehto J, Tuomilehto J. Type 2 diabetes and the risk of Parkinson's disease.27. Craft S, Watson GS. Insulin and neurodegenerative disease: shared and specific mechanisms.28. Taylor CA, Saint-Hilaire Mh Fau - Cupples LA, Cupples La Fau - Thomas CA, et al. Environmental, medical, and family history risk factors for Parkinson's disease: a New England-based case control study.29. Lin CH, Lin JW, Liu YC, Chang CH, Wu RM. Risk of Parkinson's disease following anxiety disorders: a nationwide population-based cohort study.30. Lin CH, Wu Rm Fau - Chang HY, Chang Hy Fau - Chiang YT, Chiang Yt Fau - Lin HH, Lin HH. Preceding pain symptoms and Parkinson's disease: a nationwide population-based cohort study.31. Bykov K, Yoshida KA-Ohoo, Weisskopf MG, Gagne JJ. Confounding of the association between statins and Parkinson disease: systematic review and meta-analysis. LID - 10.1002/pds.4079 [doi].32. Huang X, Alonso A Fau - Guo X, Guo X Fau - Umbach DM, et al. Statins, plasma cholesterol, and risk of Parkinson's disease: a prospective study.33. Becker C, Meier CR. Statins and the risk of Parkinson disease: an update on the controversy.34. Gao X, Simon Kc Fau - Schwarzschild MA, Schwarzschild Ma Fau - Ascherio A, Ascherio A. Prospective study of statin use and risk of Parkinson disease.35. Bai S, Song Y, Huang X, et al. Statin Use and the Risk of Parkinson's Disease: An Updated Meta-Analysis.36. Wang YL, Wang YT, Li JF, Zhang YZ, Yin HL, Han B. Body Mass Index and Risk of Parkinson's Disease: A Dose-Response Meta-Analysis of Prospective Studies.37. Saaksjarvi K, Knekt P Fau - Mannisto S, Mannisto S Fau - Lyytinen J, et al. Reduced risk of Parkinson's disease associated with lower body mass index and heavy leisure-time physical activity.38. Allam MF, Del Castillo As Fau - Navajas RFC, Navajas RF. Parkinson's disease, smoking and family history: meta-analysis.39. Bettiol SS, Rose TC, Hughes CJ, Smith LA. Alcohol Consumption and Parkinson's Disease Risk: A Review of Recent Findings.

16

Table 1. Classification of socioeconomic class by health insurance premium standard (The

mean values from 2002 to 2003 years)

ClassificationIncome Quantile

Employed Insured KRW (US Dollar)

Self-employed InsuredKRW (US Dollar)

Group 1 10 843,945 (703.29) 982,721 (818.93)Group 2 9 569,525 (474.60) 590,686 (492.23)

8 384,443 (320.36) 366,403 (305.34)7 259,505 (216.25) 227,161 (189.30)

Group 3

6 182,225 (151.85) 140,893 (117.41)

5 128,145 (106.79) 93,892 (78.24)

4 83,720 (69.77) 64,603 (53.84)

Group 4

3 49,345 (41.12) 40,541 (33.78)

2 24,625 (20.52) 29,120 (247.27)

1 9,945 (8.29) 8,531 (7.11)

*1$ = 1200 Korean Won Rate

17

Table 2. Demographic data and the incidence of ParkinsonismGroup Subjects with

Parkinsonism

Subjects without

Parkinsonism

Incidence (%)

Age

40-49 648 224,862 0.13

50-59 1,478 144,470 0.29

60-69 3,447 101,583 0.67

70-79 2,088 34,265 0.41

>80 85 1,322 0.02

Total 7,746 506,502 1.51

Gender

Men 3,574 275,276 0.69

Women 4,172 231,226 0.81

Hypertension

Yes 3,917 132,388 0.76

No 3,829 374,114 0.74

Diabetes

Yes 2,096 67,667 0.41

No 5,650 438,835 1.10

Depression

Yes 1,200 24,302 0.23

18

No 5,258 399,874 1.24

Anxiety

Yes 1,863 51,437 0.36

No 5,883 455,065 1.14

Statin

Yes 938 32,537 0.18

No 6,808 473,965 1.32

BMI

<18.5 181 9,659 0.04

18.5-22.9 2,108 149,450 0.50

23.0-24.9 1,681 115,366 0.40

25.0-29.9 2,007 133,134 0.47

30.0-39.9 188 11,463 0.04

Smoking History (the numbers of missing data=21,740)

Non-Smoker 5,691 324,584 1.16

Ex-smoker 532 43,134 0.11

Current smoker 1,217 117,350 0.25

Alcohol History (the numbers of missing data =21,740)

No 5,393 280,322 1.07

2-3times/month 735 76,330 0.15

1-2times/week 703 82,950 0.14

19

3-4times/week 344 35,314 0.07

Daily 390 22,069 0.08

Socioeconomic status

Medicaid 39 1,006 0.01

Self-employed insured grade 1 1,136 38,514 0.22

Self-employed insured grade 2 1,004 51,036 0.20

Self-employed insured grade 3 1,009 70,598 0.20

Self-employed insured grade 4 448 29,787 0.09

Employed-insured Grade 1 839 75,401 0.16

Employed-insured Grade 2 767 65,357 0.15

Employed-insured Grade 3 1,649 113,355 0.32

Employed-insured Grade 4 855 614,448 0.17

20

Table 3. HRs of Known Risk Factors of Parkinsonism

21

22

Variables HR 95% Hazard Ratio Confidence Limits P Value

Age

40s 1

50s 3.101 2.808-3.423 <.0001

60s 8.958 8.167-9.826 <.0001

70s 14.709 13.319-16.244 <.0001

80s 16.797 13.112-21.518 <.0001

GENDER

Male 1

Female 0.971 0.914-1.030 0.3273

Comorbidities

Hypertension 1.259 1.194-1.328 <.0001

DM 1.255 1.186-1.329 <.0001

Depression 1.808 1.664-1.965 <.0001

Anxiety 1.554 1.462-1.652 <.0001

Statin 1.157 1.072-1.250 0.0002

BMI

18.5-22.9 0.933 0.808-1.078 0.3458

<18.5 1.017 0.956-1.083 0.5906

23.0-24.9 1.005 0.946-1.067 0.8739

25.0-29.9 0.932 0.806-1.078 0.3452

30.0-39.9 1.878 1.746-2.021 <.0001

Smoking History

Non-Smoker 1

Ex-smoker 0.920 0.831-1.018 0.1062

Current Smoker 0.920 0.854-0.991 0.0287

Alcohol Drinking History

Figure. Flowchart of the study design

23

24

Figure 1. HRs of Known Risk Factors of Parkinsonism

Variables HR 95% Hazard Ratio Confidence Limits

Pr > ChiS

q

Age

40s 1

50s 3.101 2.808-3.423 <.0001

60s 8.958 8.167-9.826 <.0001

70s 14.70

9

13.319-16.244 <.0001

80s 16.79

7

13.112-21.518 <.0001

GENDER

Male 1

Female 0.971 0.914-1.030 0.3273

Comorbidities

Hypertension 1.259 1.194-1.328 <.0001

DM 1.255 1.186-1.329 <.0001

Depression 1.808 1.664-1.965 <.0001

Anxiety 1.554 1.462-1.652 <.0001

Statin 1.157 1.072-1.250 0.0002

BMI

18.5-22.9 0.933 0.808-1.078 0.3458

<18.5 1.017 0.956-1.083 0.5906

23.0-24.9 1.005 0.946-1.067 0.8739

25.0-29.9 0.932 0.806-1.078 0.3452

30.0-39.9 1.878 1.746-2.021 <.0001

Smoking History

Non-Smoker 1

Appendix

General name of

DrugDrug codes

atorvastatin 668100590, 653601490, 670000590, 646801490, 660700860,

648101430, 641800580, 642800510, 649600920, 656000700,

651202710, 652900780, 657305300, 648201020, 665001250,

693900570, 650201360, 643304610, 653801370, 671800570,

642903770, 642702290, 651600840, 642505070, 642100980,

643504090, 641602890, 649101310, 653100600, 648900250,

655400910, 647300180, 641700370, 657804910, 653402570,

647803660, 653005080, 698502310, 693200800, 654004200,

651500480, 621802290, 670606360, 652603190, 663605250,

646201280, 661904110, 670103330, 657502390, 643802110,

654304570, 642800520, 653401650, 648201030, 671800580,

653801380, 642902850, 642500640, 643503650, 642100990,

641602880, 643300660, 648900260, 646802530, 641805210,

657804900, 650203780, 653100610, 647803670, 648506810,

644308440, 641704470, 643802120, 641604610, 642103630,

650202940, 642705320, 642903070, 643300670, 642500650,

653803460, 643506130, 648900270, 642506260, 643304640,

643505820, 648900280, 643504570, 643504580, 643504590,

643504600, 693900390, 654303810, 670102330, 668901530,

652601750, 642801910, 649000590, 647802900, 650202790,

651202630, 655402390, 671702270, 645701430, 694000120,

651601040, 642704450, 642402000, 642001620, 653102060,

25

660702180, 658602010, 648102590, 671804140, 651902500,

670001000, 643304820, 654002230, 649505040, 643201470,

657804480, 648900960, 642001660, 654303800, 642801920,

652601760, 643305090, 660702190, 651902510, 654002240,

643201480, 670000990, 648900970, 648900980, 648902110

fluvastatin668100350, 653600300, 668100360, 653600310, 653600290,

668100340

pravastatin642401670, 640001680, 640003820, 642401700, 669904240,

641902070, 642401680, 640003800, 640003810, 642401690

simvastatin 693200510, 648502500, 621801960, 644700550, 657801200,

642702150, 642800410, 649401280, 665000420, 640001030,

642101760, 643501460, 642503380, 661900760, 669801800,

669500030, 643200170, 640900330, 670000210, 649801210,

651500560, 645700650, 645201540, 670601380, 670101290,

643301480, 656000680, 642900890, 652300990, 647602090,

649601420, 654700360, 646201110, 657302400, 671700660,

657500690, 646001610, 654301730, 654000580, 644001090,

663601030, 644301880, 647300740, 650201300, 648201180,

649502510, 655401320, 696600020, 665503800, 650301130,

655500690, 645601800, 644802220, 645302620, 698500090,

649101780, 643103350, 641805120, 643501450, 643301490,

654000590, 651803130, 655500700, 643501470, 641902730,

642503390, 640001040, 662501500, 642702160, 664600740,

648502510, 679800660, 661900770, 655602740, 642702140,

26

655500360, 655500090

rosuvastatin

663604300, 654304420, 646003490, 649505880, 641604600,

671804830, 645603790, 657304470, 653401910, 642506150,

640902230, 655904150, 670001520, 657202150, 644307830,

650102270, 643305080, 641804750, 650302670, 643902810,

652602840, 651203240, 670303190, 654003750, 645702150,

643505410, 665001300, 653102310, 650203410, 644703330,

651903240, 644501730, 642002590, 648506270, 643703260,

641904780, 645103630, 652903110, 698001120, 653804410,

650700520, 652602910, 642506250, 641604590, 671804840,

645603800, 670001550, 650102280, 643305700, 641804760,

650302680, 643902820, 670303180, 645702160, 643505440,

650203390, 644703340, 642002600, 641905360, 645103640,

653804400, 650700530, 652602920, 642506240, 641604580,

643305640, 643902800, 645702140, 650203400, 641905390,

645103620, 650700540

pitavastatin

641605320, 671805100, 645905310, 643902990, 669804070,

657804710, 665001510, 643103410, 645702360, 670701620,

642306190, 660702460, 648203480, 642403280, 642705740,

641704160, 670605810, 651203600, 648506560, 643703390,

657202320, 654304480, 654003970, 650102410, 652903280,

642904230, 646003600, 653701940, 671703330, 652104440,

643505980, 645304570, 643306060, 644900800, 644900790,

644913470

27

simvastatin

642103990, 642705520, 642306120, 641703770, 643703320,

657804500, 643103210, 643505720, 643305870, 657305470,

654304400, 641905490, 655500070, 641703780, 643703310,

657804510, 643103220, 643505730, 642104000, 643305860,

657305480, 654304410, 671804920, 641905500, 642705530,

642306110, 655500080

lovastatin

644300970, 643300490, 642300910, 654300450, 645900360,

643200120, 643500570, 641801440, 660700360, 651900300,

669800860, 669901090, 642900310, 657200320, 648501110,

653801530, 656700710, 656202260, 698400260

28