Related Experiment Video
Updated: Mar 21, 2026
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
Published on: December 23, 2025
Motivational Interviewing Improves Medication Adherence: a Systematic Review and Meta-analysis
Ana Palacio1,2,3,4, Desiree Garay5, Benjamin Langer5
1Division of Population Health and Computational Medicine, Department of Medicine, University of Miami, Miami, FL, USA. apalacio2@med.miami.edu.
Background:
Randomized clinical trials (RCTs), mostly conducted among minority populations, have reported that motivational interviewing (MI) can improve medication adherence.
Objectives:
To evaluate the impact of MI and of the MI delivery format, fidelity assessment, fidelity-based feedback, counselors' background and MI exposure time on adherence.
Data Sources:
We searched the MEDLINE database for studies published from 1966 until February 2015.
Study Eligibility Criteria:
We included RCTs that compared MI to a control group and reported a numerical measure of medication adherence.
Data Synthesis:
The main outcome was medication adherence defined as any subjective or objective measure reported as the proportion of subjects with adequate adherence or mean adherence and standard deviation. For categorical variables we calculated the relative risk (RR) of medication adherence, and for continuous variables we calculated the standardized mean difference (SMD) between the MI and control groups.
Results:
We included 17 RCTs. Ten targeted adherence to HAART. For studies reporting a categorical measure (n = 11), the pooled RR for medication adherence was higher for MI compared with control (1.17; 95 % CI 1.05- 1.31; p < 0.01). For studies reporting a continuous measure (n = 11), the pooled SMD for medication adherence was positive (0.70; 95 % CI 0.15-1.25; p < 0.01) for MI compared with control. The characteristics that were significantly (p < 0.05) associated with medication adherence were telephonic MI and fidelity-based feedback among studies reporting categorical measures, group MI and fidelity assessment among studies reporting continuous measures and delivery by nurses or research assistants. Effect sizes differed in magnitude, creating high heterogeneity.
Conclusion:
MI improves medication adherence at different exposure times and counselors' educational level. However, the evaluation of MI characteristics associated with success had inconsistent results. Larger studies targeting diverse populations with a variety of chronic conditions are needed to clarify the effect of different MI delivery modes, fidelity assessment and provision of fidelity based-feedback.
