OBJECTIVES

: In health-economic evaluations, productivity-loss is commonly assessed by human capital approach (HCA) or friction cost approach (FCA). The objective of this study was to summarize the rationales of both approaches, to discover trends in their application and to identify potential alternative handlings in future studies.

METHODS

: We conducted a systematic literature review (SLR) in 5 databases (Medline, Embase, Cochrane, CRD, Prospero), ISPOR and HTA webpages and searched for HCA or FCA-related terms in title and abstract without restrictions of time and population. Study languages included English, German and Spanish. Reviewers screened titles, abstracts and full-texts to identify relevant studies. Included studies were clustered in (i) methodological, (ii) disease-specific studies applying HCA and FCA, (iii) SLRs, or (iv) guidelines.

RESULTS

: We identified 80 papers ((i)17; (ii)21; (iii)15; (iv)21). Based on the identified evidence, arguments can be found for both approaches to measure costs for productivity-loss. Accordingly, we identified in (i) that FCA takes the employer’s perspective limited to the short-term consequences of a disease (i.e. friction-period) or death. Furthermore, the FCA assumes that vacancy can be covered with previously unemployed or already employed workers. This may result in a job-shift causing further vacancies (i.e. vacancy-chain). Whereas, the HCA considers the whole professional-life (including leisure-time or informal-care) and therefore long-term consequences. Disease-specific studies (ii) focusing on long-term consequences are more likely using HCA. In (iii), we identified descriptions, comparisons and concomitant cost impacts when using FCA or HCA. Over 76% of included guidelines (iv) postulate the use of HCA, whereas 19% recommend FCA. Belgium guideline describes a combination of both, HCA and FCA.

CONCLUSIONS

: We identified that costs for productivity-loss regarding short-term consequences and death are sufficiently considered with FCA, whereas long-term consequences are sufficiently considered with HCA. Therefore, to sufficiently estimate those costs, we suggest a combination of both approaches.