Self-Rated Health Trajectories in the African American Health Cohort

Padmaja Ayyagari,Mario Schootman,J Philip Miller,Fredric D Wolinsky,Elena M Andresen,Fred Ullrich,Theodore K Malmstrom,Douglas K Miller,Antony Bayer

doi:10.1371/journal.pone.0053278

Abstract

BackgroundSelf-rated health taps health holistically and dynamically blends prior health histories with current illness burdens and expectations for future health. While consistently found as an independent predictor of functional decline, sentinel health events, physician visits, hospital episodes, and mortality, much less is known about intra-individual changes in self-rated health across the life course, especially for African Americans.Materials/MethodsData on 998 African American men and women aged 50–64 years old were taken from a probability-based community sample that was first assessed in 2000–2001 and re-assessed 1, 2, 3, 4, 7, and 9 years later. Using an innovative approach for including decedents in the analysis, semi-parametric group-based mixture models were used to identify person-centered group trajectories of self-rated health over time. Multivariable multinomial logistic regression analysis was then used to differentiate the characteristics of AAH participants classified into the different group trajectories.ResultsFour self-rated health group trajectories were identified: persistently good health, good but declining health, persistently fair health, and fair but declining health. The main characteristics that differentiated the self-rated health trajectory groups from each other were age, education, smoking, morbidity (angina, congestive heart failure, diabetes, and kidney disease), having been hospitalized in the year prior to baseline, depressive symptoms, mobility limitations, and initial self-rated health.ConclusionsThis is the first study to examine self-rated health trajectories separately among African Americans. Four qualitatively distinct self-rated health group trajectories were identified that call into question the accuracy of prior reports that a single, average self-rated health trajectory for African Americans adequately captures their within-group heterogeneity.

Highlights

For four decades, self-rated health has been an important indicator capturing information above and beyond that reflected in objective health assessments and physician examinations [1,2,3]
Four qualitatively distinct self-rated health group trajectories were identified that call into question the accuracy of prior reports that a single, average self-rated health trajectory for African Americans adequately captures their within-group heterogeneity
Self-rated health has been an important indicator capturing information above and beyond that reflected in objective health assessments and physician examinations [1,2,3]. The reason for this is that self-rated health taps a more holistic perspective of health and dynamically blends prior health histories with current illness burdens as well as expectations for future health [4,5,6,7]

Summary

Introduction

Self-rated health has been an important indicator capturing information above and beyond that reflected in objective health assessments and physician examinations [1,2,3]. Some studies have shown that self-rated health is stable over time [15], other studies have shown that it improves over time [16], and still other studies have shown non-linear declines over time [17,18] The divergence in these findings may in part be attributable to differences in the ‘‘point of reference’’ used in the self-rated health question [19,20], which can be either global (as in the traditional question shown above), self-comparative (where the question is prefaced by ‘‘Compared to your health one [or two] year[s] ago...’’), or agecomparative (where the question is prefaced by ‘‘Compared to other persons your age...’’). While consistently found as an independent predictor of functional decline, sentinel health events, physician visits, hospital episodes, and mortality, much less is known about intra-individual changes in self-rated health across the life course, especially for African Americans. Multivariable multinomial logistic regression analysis was used to differentiate the characteristics of AAH participants classified into the different group trajectories

Methods

Results

Conclusion