Including Socioeconomic Status in Coronary Heart Disease Risk Estimation

Peter Franks; Daniel J. Tancredi; Paul Winters; Kevin Fiscella

doi:10.1370/afm.1167

Article Figures & Data

Figures

Tables
Additional Files

Download figure
Open in new tab
Download powerpoint
Figure 1.
Kaplan-Meier and Framingham risk score–adjusted survival curves for coronary heart disease event, by individual-based and block group–based hybrid SES.
FRS=Framingham risk score; SES=socioeconomic status.

View popup

Table 1.

Sociodemographic Characteristics of the Study Sample

Characteristic	Value
Notes: Eligible sample includes participants without baseline coronary heart disease stroke, peripheral vascular disease, or diabetes, and no missing Framingham risk score elements.
Eligible sample, No.	12,945
Age, mean years (SD, range)	53.9 (5.7, 45–64)
Female, %	56.2
Black, %	23.5
<12 years education, %	21.1
<$12,000 income, %	13
Quartiles of national median house-hold zip code income, %
<$20,230	8.9
$20,230–<$25,200	14.0
$25,200–<$32,266	30.6
>$32,266	46.5
Quartiles of national median house-hold block group income, %
<$20,463	16.4
$20,463–<$27,857	16.7
$2,7857–<$38,250	32.1
>$38,250	34.8

View popup

Table 2.

Cox Survival Analyses of Coronary Heart Disease Risk Prediction Using Framingham Risk Scoring SES Adjustment and With Individually Based and Hybrid SES Adjustment

	Individual-Based SES (n=12,393)		Block Group–Based Hybrid SES (n=12,032)		Zip Code–Based Hybrid SES (n=12,921)
Survival Analysis	Hazard Ratio (95% CI)	P Value	Hazard Ratio (95% CI)	P Value	Hazard Ratio (95% CI)	P Value
SES=socioeconomic status.
Note: Framingham risk score is complementary log-log transformed: log[–log (1–Framingham risk score)].
^a Individually based, block group–based hybrid, or zip code–based hybrid SES adjustment.
Framingham risk score alone
Framingham risk score	2.34 (2.12–2.57)	<.01	2.29 (2.08–2.52)	<.01	2.33 (2.12–2.55)	<.01
Framingham risk score + SES
Framingham risk score	2.30 (2.09–2.53)	<.01	2.25 (2.04–2.47)	<.01	2.30 (2.09–2.52)	<.01
Lower SES^a	1.60 (1.34–1.92)	<.01	1.58 (1.32–1.89)	<.01	1.42 (1.18–1.70)	<.01

Additional Files

Figures
Tables

Supplemental Appendix and Tables

Supplemental Appendix. Analyses; Supplemental Table 1. Cox Survival Analyses of Sex-Specific Coronary Heart Disease Risk Prediction Using Framingham Risk Scoring With Individually Based and Hybrid Socioeconomic Status (SES) Adjustment; Supplemental Table 2. Calibration Analyses of Coronary Heart Disease Risk Prediction Using Framingham Risk Scoring Alone; Supplemental Table 3. Calibration Analyses of Coronary Heart Disease Risk Prediction Using Both Framingham Risk Scoring and SES Adjustment

Files in this Data Supplement:

Supplemental data: Appendix and Tables - PDF file, 5 pages, 123 KB
The Article in Brief

Including Socioeconomic Status in Coronary Heart Disease Risk Estimation

Peter Franks , and colleagues

Background There is a growing socioeconomic disparity in coronary heart disease. This study explored alternative approaches to assessing risk based on socioeconomic status (SES) (using the patient's place of residence) and derived an approach to incorporating SES risk into treatment guidelines.

What This Study Found Adding two measures of socioeconomic status into coronary heart disease risk estimation, the authors contend, can reduce the biases inherent in the widely used Framingham risk score, which currently results in undertreatment of patients of lower socioeconomic status. Modifying the score to include a geographic area-based measure of income (using block group US Census data) and individual education level eliminated the significant socioeconomic bias observed using the Framingham risk score alone. The revised approach suggests more aggressive cholesterol treatment thresholds for those with low socioeconomic status--thresholds of 10 percent and 20 percent should be lowered to 6 percent and 13 percent for low SES persons.

Implications

The authors assert the proposed modifications should be easy for clinicians to accommodate.