Clinic Visit Timing, Prescribing Patterns, and Antihypertensive Medication Adherence: Evidence From Linked U.S....

Medication adherence is critical for hypertension management but remains suboptimal. Most interventions target patient-level barriers, whereas routine care factors, such as the timing of clinic visits, remain underexamined. Time-of-day effects on clinical decision making are documented in acute care, but implications for chronic disease management are less understood.

A longitudinal analysis of 936,502 U.S. adults with hypertension receiving 4,027,607 antihypertensive prescriptions (2016-2024) linked electronic health records with pharmacy dispensing data. Associations between appointment hour and prescription filling and 90-day medication adherence (proportion of days covered ≥80%) were estimated. Secondary outcomes included visit duration, days' supply, and fixed-dose combination prescribing. Models adjusted for patient, clinician, and visit characteristics with month and year fixed effects. Analyses were conducted in 2025.

The 5PM visits had 10.6% lower odds of adherence (AOR=0.894; 95% CI=0.854, 0.935) and 8.8% lower odds of prescription filling (AOR=0.912; 95% CI=0.871, 0.954) than 7AM visits. Each hour later was associated with 1.0% lower odds of adherence (AOR=0.990; 95% CI=0.988, 0.992) and 0.9% lower odds of filling (AOR=0.991; 95% CI=0.989, 0.993). Later visits were associated with shorter visit durations, fewer days' supply per fill, and less fixed-dose combination prescribing. Time-of-day patterns appeared only for filled prescriptions, consistent with an association linked to prescribing-dispensing processes rather than patient characteristics.

Later visit times were associated with lower prescription filling and adherence as well as shorter visits, reduced medication supply, and less fixed-dose combination prescribing. These patterns suggest that prescribing and dispensing processes vary by time of day. Attention to consistent medication management practices throughout the clinic day may help reduce this variation.