Understanding non-performance reports for instrumental activity of daily living items in population analyses: a cross sectional study

BMC Geriatrics

Research Article

Debra Saliba¹  Dawei Xie²  Qiang Pan²  Jibby E. Kurichi²  Margaret G. Stineman³  Joel E. Streim⁴  Sophia Miryam Schüssler-Fiorenza Rose⁵ 

[1] Anna and Harry Borun Chair in Geriatrics and Gerontology at UCLA, RAND, Santa Monica, CA, USA;Research Physician, VA GLAHS GRECC, RAND, Santa Monica, CA, USA;UCLA/JH Borun Center for Gerontological Research, RAND, Santa Monica, CA, USA;RAND Health, RAND, Santa Monica, CA, USA;Department of Biostatistics and Epidemiology, The Center for Clinical Epidemiology and Biostatistics, Perelman School of Medicine, University of Pennsylvani, 423 Guardian Drive, 617 Blockley Hall, 19104, Philadelphia, PA, USA;Department of Physical Medicine and Rehabilitation, Perelman School of Medicine, University of Pennsylvania, 19104, Philadelphia, PA, USA;Geriatric Psychiatry Section of the Department of Psychiatry, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA;VISN 4 Mental Illness Research Education & Clinical Center, Corporal Michael J. Crescenz Veterans Affairs Medical Center, Philadelphia, PA, USA;Spinal Cord Injury Service, Veterans Affairs Palo Health Care System, 3801 Miranda Ave, 94304, Palo Alto, CA, USA;Department of Neurosurgery, Stanford University School of Medicine, Stanford, CA, USA;

关键词: Disparities;    Function;    Self-rated health;    Survey design;   

DOI  :  10.1186/s12877-016-0235-0

 received in 2015-09-11, accepted in 2016-02-26,  发布年份 2016

来源: Springer

【 摘 要 】

BackgroundConcerns about using Instrumental Activities of Daily Living (IADLs) in national surveys come up frequently in geriatric and rehabilitation medicine due to high rates of non-performance for reasons other than health. We aim to evaluate the effect of different strategies of classifying “does not do” responses to IADL questions when estimating prevalence of IADL limitations in a national survey.MethodsCross-sectional analysis of a nationally representative sample of 13,879 non-institutionalized adult Medicare beneficiaries included in the 2010 Medicare Current Beneficiary Survey (MCBS). Sample persons or proxies were asked about difficulties performing six IADLs. Tested strategies to classify non-performance of IADL(s) for reasons other than health were to 1) derive through multiple imputation, 2) exclude (for incomplete data), 3) classify as “no difficulty,” or 4) classify as “difficulty.” IADL stage prevalence estimates were compared across these four strategies.ResultsIn the sample, 1853 sample persons (12.4 % weighted) did not do one or more IADLs for reasons other than physical problems or health. Yet, IADL stage prevalence estimates differed little across the four alternative strategies. Classification as “no difficulty” led to slightly lower, while classification as “difficulty” raised the estimated population prevalence of disability.ConclusionsThese analyses encourage clinicians, researchers, and policy end-users of IADL survey data to be cognizant of possible small differences that can result from alternative ways of handling unrated IADL information. At the population-level, the resulting differences appear trivial when applying MCBS data, providing reassurance that IADL items can be used to estimate the prevalence of activity limitation despite high rates of non-performance.

【 授权许可】

CC BY   
© Stineman et al. 2016