Is formal long-term care services utilization determined by caregivers?

Abstract

Introduction: The well-established Andersen and Newman Framework of Health Service Utilization posits that care recipient and service provider factors influence utilization of various health services. Evidence in the past decade suggests that caregiver factors may also play an important role in the utilization of formal long-term care services (LTCS). However, no systematic quantitative review examining the contribution of caregiver factors on LTCS utilization has been conducted. A meta-analytic review will be conducted to summarize the knowledge on this literature, and propose a model of factors for LTCS service utilization. Method: The first author conducted electronic searches of Medline, PubMed, Web of Science, PsycInfo, PsycArticles, and ProQuest. He also hand-searched the following journals: Journal of American Geriatrics Society, Journal of the American Medical Association, BMC Health Services Research, American Journal of Public Health, Age and Ageing, and Psychological Bulletin. The first author also asked experts about relevant studies. Studies were eligible for inclusion if they (i) used the Andersen’s Model to guide in the selection of variables, (ii) collected data directly from caregivers, and (iii) had service utilization of one or more LTCS as the outcome variable(s). Studies that did not report quantitative data were excluded. The first author conducted electronic searches of Medline, PubMed, Web of Science, PsycInfo, PsycArticles, and ProQuest. He also hand-searched the following journals: Journal of American Geriatrics Society, Journal of the American Medical Association, BMC Health Services Research, American Journal of Public Health, Age and Aging, and Psychological Bulletin. The first author also asked experts about relevant studies. Studies were eligible for inclusion if they (i) used the Andersen’s Model to guide in the selection of variables, (ii) collected data directly from caregivers, and (iii) had service utilization of one or more LTCS as the outcome variable(s). Studies that did not report quantitative data were excluded. Data on the participants recruited, methods used and the association between caregiver factors and LTCS utilization will be extracted from eligible studies by the first and second authors independently. The effect size of interest is operationally defined as the difference in the independent variables between LTCS users and non-users. Odds ratios (ORs) and other effects and their sizes will be calculated for each study and pooled using appropriate effects models for each study. Data extraction is in progress. World Congress on Integrated Care 2014, Sydney, November 23-26, 2014. International Journal of Integrated Care – Volume 14, 8 December – URN:NBN:NL:UI:10-1-116671– http://www.ijic.org/ Preliminary Results: 13 studies reporting 12 independent datasets were identified. Of these 12 datasets, 11 were reported in journal articles, and one was reported in a dissertation. 9 of the datasets were from North America, 1 from Europe, 1 from Australia and 1 from Asia. Overall, there were 2,649 caregivers whose care recipients were LTCS users, and 1,985 caregivers whose care recipients were non-users. The range of the age means was 42 to 60 years. The caregivers were predominantly women (68%). All except one of the datasets were obtained from cross-sectional studies. There was only dataset that was obtained from a longitudinal study. Discussion: The implications of this meta-analysis on theoretical and practical development will be discussed. Where plausible, the Andersen’s model will be enhanced. Recommendations to practice will be put forth. Limitations: The number of studies involved in this analysis was quite small, so any lack of statistical significance may have been due to a lack of statistical power. Only English articles known to the authors, whether published or unpublished, will be included in the analysis. The observed point-biserial correlations for dissertations and articles will be compared to check publication bias. The Duval and Tweedie trim and fill procedure will be conducted to examine the robustness of findings relative to all possible sources of data censoring.