2017 Intermountain Healthcare Annual Research Meeting

Andrew J Knighton PHD CPAIntermountain Institute for Healthcare Delivery Research

Social determinants, health and healthcare outcomes

“Adversity is not randomly distributed: instead it tends to cluster and to accumulate present on top of past disadvantage”

David Blane, MSc MD

Social determinants of health

• People with a higher standard of living have better health outcomes (Marmot, 2006)

• The majority of health is driven by non-care delivery factors – genetic, social, environmental, behavioral

• Conditions in the places where people live, learn, work and play affect a wide range of health risks and outcomes (CDC, 2015)

• No agreed-upon measurement methods have been adopted (Knighton, 2016; Phillips 2016)

Healthy People 2020

Intermediary vs structural determinantsIntermediary determinants of health - immediate

• Material circumstances (including access to care)

• Behaviors and biologic factors

• Psychosocial factors

Structural determinants of health - upstream• Neighborhood living conditions

• Opportunities for learning and capacity for development

• Employment opportunities and community development

• Prevailing norms, customs and processes

• Social cohesion, civic engagement, and collective efficacy

• Health promotion , disease prevention and healthcare opportunities

World Health Organization

What is the Singh Area Deprivation Index (ADI)?

Geographic, area-based measure of the relative average socio-economic position of a particular neighborhood

Factor and principal components analysis used to develop and validate index by Singh for the United States (2003)

Index based upon 17 census measures in four categories linked to all-cause US mortality: income, living conditions, employment and education

Developed at the census block group level for the state of Utah (Knighton et al, 2016)

Patient assigned an ADI score (Mean: 100; Range -40 to 150) based upon the census block group they live in

Profiling disparities by quintile

6Knighton et al., 2016

Census Indicator Q1 Q5

Observed

odds ratio

Median family income, $ 105,000$ 42,000$ 0.4

Population below 150% poverty level, % 11.2% 39.8% 3.6

Single parent household, % 5.6% 13.0% 2.3

Owner-occupied housing, % 83.0% 53.8% 0.6

More than 1 person per room, % 1.4% 6.4% 4.6

Median home value, $ 383,380$ 126,620$ 0.3

Unemployment rate, % 5.5% 10.4% 1.9

High school graduation rate, % 97.1% 81.9% 0.8

<9th grade education, % 0.8% 7.0% 8.8

Census mean values by ADI quintile

What does an ADI measure?

• Estimates the combined effect of both individual and neighborhood deprivation exposures (both compositional + contextual effects)

• Strong causal link between individual socio-economic status and the neighborhood we live in (Cutrona, 2006; Cierda, 2010; Bikdeli, 2014)

• Neighborhood-level contextual exposures have an independent effect on health (Bikdeli, 2014; Chi, 2016)

• High degree of persistence in neighborhood deprivation exposure over time despite mobility (Solon, 1999; Kunz, 2001; Van Ham, 2014; Vartanian, 2007;

Knighton, 2017)

Case Study – Do social determinants predict higher levels of subsequent utilization in already high-utilizing patients?

Goal – Stratification of patients with increased risk of future utilization

Opportunity - Better stratification could support identification of patients and interventions designed to address effects

Study Population: 5158 adult patients identified as high-cost in 2014 given classification in the top 10% in the prior year and top 15% two of last three years.

Excluded costs for deceased patients, transplant costs, chemo, dialysis therapy and orthopedic replacements.

Adjustment factors include age, sex, ethnicity, race, marital status, Charlson comorbidity score, Medicaid payer status (*p<.05)

(n=5158)

*

*

**

*

*

Adjustment factors include age, sex, ethnicity, race, marital status, Charlson comorbidity score, Medicaid payer status (*p<.05)

(n=5158)

*

*

*

* *

*

*

Case Study – What role does community play?

Neighborhood

material

deprivation

Patient

Outcome

-Neighborhood

material

deprivation

Patient-

reported faith

identification

or urban

residence

Patient

Outcome

Social

isolation

+

-

(n=6065)

OR 0.35 (95% CI: 0.14-0.87); p=.03

(n=6065)

OR 0.29 (95% CI: 0.09-0.98); p=.05

Case Study –Collaborating for health in pre-natal and post-natal care

Goal –Nurse-based home visits for a high-risk expectant mother can significantly improve the life course for both the mother and child

Opportunity – Collaboration between IH and Salt Lake County Department of Health (SLCDH) developed to increase IH patient participation

Approach –

• Use clinical judgment supported by risk stratification to better identify those most likely to benefit from program engagement

• Patient invitation and warm referral provided by the clinical team

• Patient feedback shared between IH clinical team and SLCDH nurse

Performance Baseline – Patient invitations to participate

Risk stratification and prediction

Initial Criteria

• First pregnancy

• Low income

Revised Criteria

• First pregnancy

• <29 weeks gestation

• >12 weeks gestation

• Any patient living in a neighborhood with ADI quintile=5

• Any patient on Medicaid living in neighborhoods with an ADI of 3-5

Introduction into the clinical workflow

Preliminary results

• Introduction of a warm referral into the OB/GYN visit is a natural extension of the clinical workflow

• Observed enrollment rates are higher in communities using a warm handoff

• Observing an increase in the percentage of patients referred from more deprived neighborhoods

Evaluating generalizability of collaborative approach within the pediatric asthma clinical program

Summary

• ADI provides a low-cost, accessible measure of relative patient socio-economic position

• ADI appears useful in risk stratification and prediction• Focus has been on its effects within the delivery system

• Measure can support assessment of structural determinants as we look “upstream”

• Identification of vulnerable patients can guide the use of scarce resources to tailor interventions to improve outcomes

ReferencesBikdeli B et al. Place of residence and outcomes of patients with health failure: analysis from the telemonitoring to improve heart failure outcomes trial. Circ Cardiovasc Qual Outcomes. 2014; 7:749-756.

Chi GC et al. Individual and neighborhood socio-economic status and the association between air pollution and cardiovascular disease. Environmental Health Perspectives 2016; 124(12): 1840-1847.

Cierda M, et al. The relationship between neighborhood poverty and alcohol use: estimation by marginal structural models. Epidemiology, 2010. 21(40: 482-489.

Cutrona CE et al. Neighborhood characteristics and depression an examining of stress processes. Curr Dir Psychol Sci. 2006; 15(4): 188-192.

Galea S. Contextual determinants of drug use risk behavior: a theoretic framework. Journal of Urban Health 2003; 80(4):iii50-iii58.

Kind AJ, et al. Neighborhood socio-economic disadvantage and 30-day re-hospitalization: a retrospective cohort study. Ann Intern Med. 2014; 161(11):765-74.

Knighton AJ, et al.. Introduction of an area deprivation index measuring patient socio-economic status in an integrated health systems: implications for population health. eGEMS. 2016; 4(3):1238.

Knighton AJ et al. Measuring the effects of social determinants on patient outcomes: a systematic literature review. Submitted manuscript, 2017.

Knighton AJ. Is a patient’s current address of record a reasonable measure of neighborhood deprivation exposure? A case for point in time measures of residence in clinical care. Submitted 2017.

Kunz J et al. Are Single-Year Measures of Neighborhood Characteristics Useful Proxies for Children’s Long-Run Neighborhood Environment? Economic Letters 2001;79:231-7.

Loignon C, et al. Perceived barriers to healthcare for persons living in poverty in Quebec, Canada: the Equi-healthy Project. Int J Equity Health. 2015; 14:4.

Marmot M, Wilkinson R. Social determinants of health. Oxford: Oxford University Press; 2006.

Phillips RL et al. How other countries use deprivation indices – and why the United States desparately needs one. Health Affairs 2016; 35(11): 1991-1998.

Singh G. Area Deprivation and Widening Inequalities in US Mortality, 1969–1998. Am J Public Health. 2003;93(7):1137-1143.

Solon G. Intergenerational mobility in the labor market. In Shenfelter OC, Card D, eds. Handbook of Labor Economics, Vol 3A. Amsterdam: North-Holland. 1999:1761-1800.

Tarlov AR. Public Policy Frameworks for Improving Population Health. Annals of the New York Academy of Sciences, 1999;896:281-293.

Van Ham M et al. Intergenerational transmission of neighborhood poverty: an analysis of neighbourhood histories of individuals. Institute of British Geographers 2014; 29:402-407.

Vartanian TP et al. Intergenerational neighborhood-type mobility: examining differences between blacks and whites. Housing Studies 2007;22:833-856.

Villanueva C, et al. The association between neighborhood socio-economic status and clinical outcomes among patients 1 year after hospitalization for cardiovascular disease. J Community Health 2013. 38(4): 690-697

Winkleby M, et al. Effect of cross-level interaction between individual and neighborhood socio-economic status on adult mortality rates. American Journal Public Health 2006; 96(12): 2145-2153..