Objective: In the United States, the prevalence of noncompliance to prescribed treatment regimens is 50%. The consequences for this health behavior include escalating costs associated with healthcare and altered health outcomes. Ongoing research for successful adherence intervention strategies requires an examination of variables associated with poor, as well as successful treatment adherence. Current findings indicate patients perceive memory, specifically "forgetting," as a common reason why they do not take prescribed medications and further suggest a need for the examination of self-reported everyday memory. Therefore, this study examined self-reported everyday memory failures as a predictor of medication adherence.
Design: This descriptive study of medication adherence in chronic disease used a sample diagnosed with rheumatoid arthritis (RA) as the disease model to examine the relationship of everyday memory and electronically measured medication compliance. Prior findings indicated a relationship exists between measures of everyday memory and depressive symptoms, but not with electronically measured adherence across a population that included a wide spectrum of medication adherence rates from 0 to 100%. However, the typical distribution of adherence scores is not normally distributed and raises question for subgroup differences with respect to adherence and covariates. Therefore, this study proposed to examine the relationships of everyday memory, depression, and electronic measured adherence among three subgroups of subjects grouped by adherence rate (1=0 to 39%, 2=40 to 79%, and 3=80 to 100%). Everyday memory was not expected to be associated with adherence among subjects attaining rates between 40 and 100%.
Population, Sampling, Setting, Years: The sample consisted of 307 subjects recruited from rheumatology practices for participation in a RCT of adherence intervention strategies. On average, patients were women (78%), in their latter midlife years (M=59, SD=11 years), educated with a high school diploma or beyond (86%), married (73%), White (94%), and with longstanding diagnosis of RA (M=14, SD=9.5 years). These demographic findings did not differ between the three subgroups of subjects.
Concept or Variables Studied Together: The Cognitive Failures Questionnaire (CFQ) and a one-item self-rated global measure of memory (GM) were the instruments selected to operationalize self-reported everyday memory. Increasing scores on the CFQ reflect recognition of greater everyday memory failures. Low scores on the GM reflect greater everyday memory failures. The Beck Depression Inventory (BDI) was used as a measure of depressive symptoms. Medication adherence was ascertained with electronic event monitors (EEM) adherence.
Methods: Following descriptive analysis, baseline measures for the CFQ, GM, BDI, and EEM were examined for association with correlational analyses (Pearson and Spearman).
Findings: Average CFQ scores by subgroup were not significantly different (1: M=35, SD=14; 2: M=37, SD=14; 3: M=34, SD=14). GM scores also did not differ by subgroup (1: Md=5, 2: Md=5, 3: Md=5), as well as BDI scores (1: M=11, SD=14; 2: M=10, SD=7; 3: M=9, SD=7). Rate of adherence by EEM per group averaged (1: M=10, SD=13, n=110; 2: M=63, SD=13, n=96; 3: M=93, SD=6, n=100). Although CFQ, GM, and BDI scores were significantly correlated in each subgroup, the CFQ (r=.249, p=.009, n=110) was the positively associated with only EEM among subjects with poor adherence (0 - 40%). The CFQ, GM, and BDI were not associated with EEM among subjects in the subgroups with higher EEM adherence (40 - 79% and 80 - 100%).
Conclusions: The positive association between EEM adherence and CFQ scores suggests subjects with the lowest rates of adherence are less likely to self-report difficulty with everyday memory.
Implications: These findings suggest patients with poor adherence may be at risk for lack of recognition of everyday memory errors, potentially for forgetting of omission of medication administration. Additional research, including assessment of true memory problems though neuropsychological testing, is needed to confirm these findings and to consider whether subgroup differences exist within EEM adherence levels. Further research is needed to examine whether patients with poor baseline adherence would benefit from adherence interventions identifying their self-reported memory difficulties.
Back to Predictors of Adherence in Chronic Disease
Back to The Advancing Nursing Practice Excellence: State of the Science