A pilot study among older adults of the concordance between their self-reports to a health survey and spousal proxy reports on their behalf

Abstract

BackgroundProxy respondents are frequently used in health surveys, and the proxy is most often the spouse. Longstanding concerns linger, however, about the validity of using spousal proxies, especially for older adults. The purpose of this pilot study was to evaluate the concordance between self-reports and spousal proxy reports to a standard health survey in a small convenience sample of older married couples.MethodsWe used the Seniors Together in Aging Research (STAR) volunteer registry at the University of Iowa to identify and consent a cross-sectional, convenience sample of 28 married husband and wife couples. Private, personal interviews with each member of the married couple using a detailed health survey based on the 2012 Health and Retirement Study (HRS) instrument were conducted using computer assisted personal interviewing software. Within couples, each wife completed the health survey first for herself and then for her husband, and each husband completed the health survey first for himself and then for his wife. The health survey topics included health ratings, health conditions, mobility, instrumental activities of daily living (IADLs), health services use, and preventative services. Percent of agreement and prevalence and bias adjusted kappa statistics (PABAKs) were used to evaluate concordance.ResultsPABAK coefficients indicated moderate to excellent concordance (PABAKs >0.60) for most of the IADL, health condition, hospitalization, surgery, preventative service, and mobility questions, but only slight to fair concordance (PABAKs = −0.21 to 0.60) for health ratings, and physician and dental visits.ConclusionsThese results do not allay longstanding concerns about the validity of routinely using spousal proxies in health surveys to obtain health ratings or the number of physician and dental visits among older adults. Further research is needed in a nationally representative sample of older couples in which each wife completes the health survey first for herself and then for her husband, each husband completes the health survey first for himself and then for his wife, and both spouses’ Medicare claims are linked to their health survey responses to determine not just the concordance between spousal reports, but the concordance of those survey responses to the medical record.

Comparison of Patient and Proxy Assessment of Patient-Centeredness in the Care of Coronary Heart Disease: A Cross Sectional Survey Using the PACIC-S11.1

“…Other studies have also produced mixed results in terms of correlation between patient and proxy ratings, with some studies reporting better agreement on health status and symptom-related questions, while others report better agreement regarding frequency, quality and/or organization of care. 25 - 27 Overall, the differences in ratings highlight the need to encourage patients’ relatives and/or friends to become closely involved in disease management and care planning at an early stage in practice. If the patient’s health status deteriorates, the person who takes on the role of their legal representative will need to be aware of the patient’s wishes and preferences when making decisions on their behalf.…”

Can caregivers report their care recipients’ post-stroke hospitalizations and outpatient visits accurately? Findings of an Asian prospective stroke cohort

“…Furthermore, we illustrate a novel approach for quantifying agreement across different data sources using the original data (e.g., counts or discrete events), without dichotomizing or categorizing our data. In the past, some studies limited their results to reporting single agreement measures such as kappa or the traditional ICC [ 8 , 10 , 12 – 18 ], while others opted for a second analysis, reporting factors associated with over- or under-reporting [ 11 , 19 – 26 ]. Often with healthcare data, count variables are binarized to facilitate the use of kappa, compromising on the information available, which potentially results in a loss of power, and our study addressed this gap in the literature.…”

“…In addition to a dearth of research on the validity of proxy-reported data for stroke, there is a general methodological issue regarding how data in such studies are treated. Typically, healthcare utilization agreement studies have attempted to either quantify agreement based on traditional measures, such as Cohen’s kappa or the intraclass correlation coefficient (ICC) [ 8 , 10 , 12 – 18 ], or used a two-step process to extend the analysis to describing the reporting patterns of utilizers [ 11 , 19 – 26 ]. Studies often end up dichotomizing or categorizing their healthcare service utilization data (e.g., counts or discrete events), which could potentially result in a loss of power.…”

Proportional likelihood ratio mixed model for discrete longitudinal data

“…Follow-up interviews include survey questions on health ratings (5-point scale), overall health and memory since the previous interview (3-point scale), sums of self-reported yes-no chronic disease questions, activities of daily living, instrumental activities of daily living, plus cognitive function tests, such as, immediate word recall, delayed word recall and mental status, each on a 0-to 10-point scale. [5][6][7] The two predominantly used methods for the analysis of longitudinal or clustered data are generalized linear mixed models (GLMM) 8,9 and generalized estimating equation (GEE) models. 10 As an extension to the GLM, the GLMM is a likelihood method which incorporates random effects in the linear predictor and makes full distributional assumptions for the response variable given the random effects, whereas the GEE method is a quasi-likelihood-like approach which depends on the correct specification of the first and second marginal moments of the response variable.…”

“…As a second example, a number of studies here have used the Study on Assets and Health Dynamics among the Oldest Old (AHEAD), a nationally representative cohort of individuals aged 70 and older, in which data collection started in 1993 (baseline) with biennial follow‐ups. Follow‐up interviews include survey questions on health ratings (5‐point scale), overall health and memory since the previous interview (3‐point scale), sums of self‐reported yes‐no chronic disease questions, activities of daily living, instrumental activities of daily living, plus cognitive function tests, such as, immediate word recall, delayed word recall and mental status, each on a 0‐ to 10‐point scale 5‐7 …”