A systematic review of case-mix models for home health care payment: Making sense of variation

• Published in: Health policy (Amsterdam) 2020-02, Vol.124 (2), p.121-132
• Main Authors: van den Bulck, Anne O.E., de Korte, Maud H., Elissen, Arianne M.J., Metzelthin, Silke F., Mikkers, Misja C., Ruwaard, Dirk
• Format: Article
• Language: English
• Subjects: Attributes; Casemix; Classification; Criteria; Diagnosis-Related Groups - economics; Health administration; Home care services; Home Care Services - economics; Home health care; Humans; Literature reviews; Prospective Payment System; Reimbursement Mechanisms; Systematic review; Websites
• ISSN: 0168-8510; 1872-6054
• DOI: 10.1016/j.healthpol.2019.12.012

•Different case-mix models exist for prospective home health care (HHC) payment.•These models explain 14%–54.3% of variance in HHC utilization/costs.•There is a considerable variation in included case-mix predictors across models.•Common predictors relate to physical functioning, daily functioning, and health service use.•Few of the identified case-mix models are implemented in HHC practice. Case-mix based payment of health care services offers potential to contain expenditure growth and simultaneously support needs-based care provision. However, limited evidence exists on its application in home health care (HHC). Therefore, this study aimed to synthesize available international literature on existing case-mix models for HHC payment. We performed a systematic review of scientific literature, supplemented with grey literature. We searched for literature using six scientific databases, reference lists, expert consultation, and targeted websites. Data on study design, case-mix model attributes, and conclusions were extracted narratively. Of 3303 references found, 22 scientific studies and 27 grey documents met eligibility criteria. Eight case-mix models for HHC were identified, from the US, Canada, New Zealand, Australia, and Germany. Three countries have implemented a case-mix model as part of a HHC payment system. Different combinations of in total 127 unique case-mix predictors are included across models to predict HHC use. Case-mix models also differ in targeted services, operationalization, and outcome measures and predictive power. Case-mix based payment is not yet widely used within HHC. Multiple varieties were found between HHC case-mix models, and no one best form of a model seems to exist. Even though varieties are partly inevitable due to country-specific contexts, developing a shared vision in case-mix model attributes would be key to achieving efficient, needs-based HHC.