Article Abstract

Online ISSN: 1099-176X    Print ISSN: 1091-4358
The Journal of Mental Health Policy and Economics
Volume 29, Issue 2, 2026. Pages: 57-64
Published Online: 1 June 2026

Copyright © 2026 ICMPE.

Early Impacts of a Medicaid Value-Based Payment Policy on Quality of Care in a Large Population with Serious Mental Illness

Michael William Flores,¹ Benjamin Lê Cook,¹ Paloma Luisi,² Lindsay Cogan,³ Jia Ren,⁴ Elizabeth Nichols,³
Sharon-Lise T. Normand,⁵ K. John McConnell,⁶ Catherine Holmes,⁷ Gareth Parry,¹ Ana Progovac,⁸ Akhil Reddy,⁷ and Marcela Horvitz-Lennon^9* 

¹PhD,MPH, Health Evaluation Research Lab; & Cambridge Health Alliance, Cambridge, MA; & Harvard Medical School, Boston, MA, USA.
²MPH, New York State Department of Health, Albany, NY, USA.
³PhD, New York State Department of Health, Albany, NY, USA.
⁴MA, New York State Department of Health, Albany, NY, USA.
⁵PhD, Harvard Medical School, Boston, MA; & Harvard TH Chan School of Public Health, Boston, MA, USA
⁶PhD, Oregon Health & Science University, Portland, OR, USA.
⁷BA, Health Evaluation Research Lab; & Cambridge Health Alliance, Cambridge, MA, USA.
⁸PhD, Health Evaluation Research Lab; & Cambridge Health Alliance,  Cambridge, MA & New York University, New York, NY, USA.
⁹MD,MPH, Health Evaluation Research Lab; & Cambridge Health Alliance, Cambridge, MA; & Harvard Medical School, & RAND, Boston, MA, USA.

Background: In the US, several states including New York State (NYS) have adopted Medicaid value-based payment (VBP) policies to improve healthcare quality and reduce spending. However, the impacts of these policies on vulnerable Medicaid beneficiaries with high behavioral health needs including those with serious mental illness (SMI) are unknown.

Aims of the Study: To inform policy efforts to improve the value of Medicaid-funded care by examining the early impacts of NYS’s VBP policy implemented in 2017 on the quality of healthcare received by adult beneficiaries with SMI.

Methods: We conducted a retrospective cohort study to examine the association of the VBP policy with healthcare quality among adult Medicaid beneficiaries with SMI. We estimated difference-in-differences models comparing VBP-exposed and unexposed beneficiaries during the period preceding VBP implementation (2014-2016) and in 2018, a full year following implementation. Quality was assessed with 10 validated and SMI-relevant measures of mental healthcare quality (e.g., adequate Adherence to Antipsychotic Medications for Individuals with Schizophrenia, and Follow-up after Hospitalization for Mental Illness, 30-day), physical healthcare quality (e.g., Diabetes Screening for People with Schizophrenia or Bipolar Disorder who are using Antipsychotic Medications; Comprehensive Diabetes Care, Eye Exam), and overall healthcare quality (All-Cause Readmissions). Models for all outcomes adjusted for concurrent policies, and models for acute care measures also adjusted for need variables.

Results: Our diverse cohort included 172,420 person-years with SMI, with 28.5% VBP-exposed. The largest diagnostic groups were schizophrenia and bipolar disorder (42.2% and 41.3%, respectively). The VBP-exposed and unexposed beneficiaries differed on most characteristics, with the unexposed being slightly younger, less diverse, and healthier relative to the VBP-exposed. After adjusting for covariates, VBP-exposure was associated with changes in only two of the 10 quality measures compared with the VBP-unexposed group: a 1.8- percentage-point (pp) increase in adequate Adherence to Antipsychotic Medications for Individuals with Schizophrenia (95% CI, 0.5-3.2) and a 1.4-pp decrease (95% CI, -2.6 to -0.1) in Diabetes Screening for People with Schizophrenia or Bipolar Disorder who are using Antipsychotic Medications. 

Discussion: A year following its implementation, a Medicaid VBP policy that involved varying degrees of provider risk-bearing had few quality impacts, one positive and one negative, among the state’s beneficiaries with SMI. Our study, the first to examine VBP quality impacts with several validated measures of mental, physical, and overall healthcare quality, has some limitations, including that we cannot rule out time-varying unmeasured confounding, nor can we isolate the effects of type of risk arrangement (one-sided versus two-sided) or the mix of measures potentially associated with quality of SMI care included in the VBP contracts. Additionally, our study may not have captured the full impacts of the VBP policy due to our relatively short observation period.

Implications for Health Policies and Further Research: Policymakers seeking to use VBP to improve quality of SMI care may consider VBP policy features likely to facilitate achieving this goal, for example, incentives for delivering clinically integrated care and inclusion in contracts of SMI-relevant quality measures. Future studies should be designed to confirm and extend our findings.