Editorial
May 12, 2025
Morning or Nighttime Medication Dosing—Does It Matter in the Treatment of Hypertension?
Sandra J. Taler
JAMA. Published online May 12, 2025. doi:10.1001/jama.2025.7286
Normal circadian rhythm of blood pressure (BP), defined as a BP decline of 10% to 20% at night, is an important predictor of cardiovascular health and an indicator of cardiovascular disease (CVD) risk when absent. For treatment of hypertension, it has been standard practice to dose antihypertensive medications in the morning to address higher daytime BP levels; the effects of altering timing of drug dosing on circadian rhythm have not been convincingly delineated.
In 2010, a report from the Spanish group MAPEC shocked the BP world by reporting dramatic benefits from transferring BP treatment to bedtime rather than morning dosing.1 In this study, 2156 patients with hypertension were randomized to ingest all of their prescribed antihypertensive medications on awakening or to take at least 1 medication at bedtime. All underwent 48-hour ambulatory BP monitoring combined with actigraphy by wrist monitor to verify awake and sleep times, repeated annually. Although BP was similar between groups at the start of the study, those taking medication at bedtime were reported to have lower sleep BP, greater nocturnal BP decline, less nondipping of BP at night, and overall greater BP control. After 5.6 years, the bedtime dosing group had a remarkable 61% lower risk of total CVD events compared with the morning group. A second trial reported by the same investigators in 2020, the Hygia Chronotherapy Trial, randomized 19 084 patients to take all of their antihypertensive medications at bedtime or on awakening and to undergo serial 48-hour ambulatory BP monitoring.2 Over 6.3 years, patients assigned to bedtime dosing had 45% lower rates of the primary outcome (myocardial infarction, coronary revascularization, heart failure, ischemic stroke, hemorrhagic stroke, and CVD death) compared with morning dosing. Again, bedtime dosing was associated with improved BP control and a greater percentage of patients with normal nocturnal BP decline.
The extent of the differences between groups, not from changing medications or dosage but only from a change in timing of dosing, was so extreme that the hypertension community waited for additional validation from other studies. Editors of the European Heart Journal published a special article in the same issue with the Hygia results discussing their concerns about the validity of the results and efforts to verify them.3 A second commentary on Hygia by noted hypertension experts highlighted the dramatic relative risk reductions of 45% for the composite and 56% for CVD mortality, raising concerns that inappropriate data collection may have contributed to these results.4Doubts were raised regarding the power analysis and randomization process illustrated by differences in treatment group baseline characteristics. Continuing the trial beyond detection of significant differences was considered questionable from an ethical perspective, as such marked benefits would be expected to trigger early study termination. Although no evidence for misconduct was found, an ethics review committee advised verification of the source data before the results could be applied to clinical practice.
Additional studies were conducted in the United Kingdom to examine the timing of BP medication dosing. The TIME study randomized 21 104 participants to take all of their antihypertensive medications in the morning between 6 am and 10:00 am or in the evening between 8:00 pm and midnight, with follow-up for clinical events over 4 years.5 They found no differences in clinical end points and concluded that patients should take their regular antihypertensive medications at a convenient time that minimized any undesirable effects, and that evening dosing was not different from morning dosing with regard to major cardiovascular outcomes. A small randomized crossover study, HARMONY, recruited 103 adults with hypertension (BP ≤150/≤90 mm Hg while receiving stable therapy with 1 or more agents) to take their antihypertensive medications in the morning or the evening for 12 weeks then to cross over to the opposite schedule.6 Clinic and 24-hour ambulatory BP measurements were obtained at baseline, 12 weeks, and 24 weeks. For the 95 who completed all parts of the trial, mean 24-hour systolic and diastolic BP did not differ between daytime and evening dosing, and there was no difference in mean daytime and nighttime ambulatory BP or in clinic levels and no differences by age or sex.
Conflicts between MAPEC and Hygia and the TIME and HARMONY results left the issue of BP medication timing in limbo,7,8 with most experts and clinicians not convinced of the benefits of night dosing but with limited data to counter this finding. The BedMed studies now strengthen that evidence.9,10 BedMed enrolled primary care patients from 5 Canadian provinces who were randomized to take all of their antihypertensive medications in the evening or in the morning, with follow-up for mortality and CVD outcomes and potential evidence of harm. The study design was similar to that of the TIME study for inclusion criteria and outcomes of interest and was conducted similarly in a publicly funded health care system. For 3357 patients followed up for a median of 4.6 years, primary outcomes did not differ between those in the bedtime vs morning dosing groups. Furthermore, there were no differences for individual outcomes or safety outcomes between groups, including rates of falls, fractures, new glaucoma diagnoses, or cognitive decline. These findings demonstrated no differences in outcomes even while data on ambulatory BP monitoring, performed for 151 participants from each study group, indicated lower mean nighttime systolic and diastolic BP and higher daytime BP control rates in the bedtime dosing group.
An added concern in an older population relates to the potential negative consequences of bedtime dosing, which could excessively lower nighttime pressures, exacerbating other conditions, such as glaucoma or orthostatic hypotension. The BedMed-Frail trial10 focused specifically on a cohort of 776 Canadian nursing home residents with hypertension recruited from 13 facilities (median age, 88 years, with 85.6% having some degree of dementia) using the same design as the larger BedMed trial, but with an opt-out approach and with shorter follow-up at a median of 1.1 years. Even with this short follow-up, they were able to reach target end points due to high death rates in people of very advanced age. Again, there were no differences in deaths, which were the main outcome events in this older cohort, and no differences in falls, fractures, ulcers, or cognition problems related to timing of drug dosing.
What can we conclude from these efforts? First, if something sounds too good to be true, it is probably not true. After the initial dramatic results observed in the MAPEC and Hygia trials, attempts by other research groups to corroborate the initial findings were not able to do so, raising serious questions. Although the use of 48-hour ambulatory BP monitoring and actigraphy may be valuable for mechanistic studies, these techniques are not practical for use in large-scale clinical trials or for routine patient care. The most important finding, that BP medication administration at night did not affect cardiovascular event rates or survival, is clear. And the use of these additional tools is unlikely to change those outcomes.
Second, the combined data also disprove the maxim that medications must be dosed in the morning. The inclusion of a frail subpopulation in BedMed-Frail who did not experience adverse effects from bedtime dosing and careful reporting on adverse events, including new glaucoma diagnoses, vision loss, and reduced cognition, in the larger BedMed trial provide additional reassurance of the safety of bedtime dosing. Loosening the rules on time of dosing may facilitate better supervision by caregivers, who may be able to provide assistance or oversight later in the day due to other responsibilities. At the end of the day, timing of medications doesn’t matter as much as consistency in taking them. Regular dosing and use of long-acting medications should be emphasized and may better address concerns related to BP variability.
Finally, we have more to learn about circadian BP rhythm and drug dosing. How much do medications alter circadian BP levels, and is this consistent over time? Why in HARMONY were there no differences in mean daytime and nighttime BP even when drug dosing times were reversed, while in BedMed, lower BP was observed in the ambulatory BP monitoring substudy?
The BedMed trials support the earlier UK TIME and HARMONY trials and should help to resolve questions on optimal dose timing. The addition of a subgroup of very advanced age with attention to risks of overtreatment adds additional reassurance that time of dosing is not a safety issue. We can all sleep better at night whether medications are dosed before or after bedtime hours.
