Determinants of Healthcare Use Based on the Andersen Model: A Systematic Review of Longitudinal Studies

Abstract

The aim was to give an overview of longitudinal observational studies investigating the determinants of healthcare use explicitly using the Andersen model. To this end, three electronic databases (Medline, PsycINFO and CINAHL) were searched (and an additional hand search was performed). Longitudinal observational studies examining the determinants of healthcare use (outpatient physician services and hospital stays) based on the Andersen model were included, whereas disease-specific samples were excluded. Study quality was evaluated. The selection of studies, extraction of data and assessment of the studies were conducted by two reviewers. The following determinants of healthcare use were displayed based on the (extended) Andersen model: predisposing characteristics, enabling resources, need factors and psychosocial factors. In sum, n = 10 longitudinal studies have been included in our systematic review. The included studies particularly showed a longitudinal association between increased needs and higher healthcare use. Study quality was rather high. However, several studies did not conduct robustness checks or clarify the handling of missing data. In conclusion, this systematic review adds to our current understanding of the factors associated with healthcare use (mainly based on cross-sectional studies). It showed mixed evidence with regard to the association between predisposing characteristics, enabling resources and healthcare use longitudinally. In contrast, increased need factors (in particular, self-rated health and chronic conditions) were almost consistently associated with increased healthcare use. This knowledge may assist in managing healthcare use. Since most of the studies were conducted in North America or Europe, future longitudinal studies from other regions are urgently required.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 38 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.