OBJECTIVES:

United States payers have expressed concerns with budget impact models submitted to them for formulary review and/or pricing negotiations. The objective of this study was to review and develop lessons from published budget impact models for United States public and private payers.

METHODS:

A systematic literature review was conducted using PubMed and Medline for full text budget impact publications for United States public and private payers. The data were extracted for model methods: design, assumptions, time horizon, incorporation of adverse events, resource use, sensitivity analyses and payer perspective. Insights and lessons for new model development were generated from this review.

RESULTS: Sixty seven (67) full text budget impact model publications for United States were identified and reviewed. Nearly half of the model publications claimed savings (30 of 67) or low/small budget impact (4 of 67). Adverse events were included in 21 models, and resource use (hospital, emergency room and nursing time) was included in 17 models. Cost-Effectiveness model structure was used in 10 models. Sensitivity analyses were reported in 33 publications (all one-way sensitivity analysis, no probabilistic analyses were reported). Most models focussed on commercial payer or Medicare perspective (37), while only 2 models were developed for Medicaid. The time horizons varied across publications, with an even mix of 3 and 5 years. The hypothetical plan size also varied across models (1, 5, 10 million), though 1 million was the most common plan size (24 models). Overall, models show a wide range of heterogeneity which could make it challenging for payers to accept and compare model results.

CONCLUSIONS: US budget impact models show wide range of heterogeneity in methodology and assumptions. There is a need for a consistent modeling approach to improve acceptability by payers.