HRV Biofeedback Went With Less Substance Use in One Trial

In one phase 2 trial, adding wearable HRV biofeedback to usual outpatient treatment went with lower craving and less substance use. Of 120 randomized adults with substance use disorder, those given biofeedback had significant reductions in craving and negative affect over 8 weeks. Craving and negative affect rose in the usual-care group. Among the 120 adults, the odds of a substance use day were 0.36 times as high with biofeedback (95% credible interval 0.25 to 0.54). The authors call for phase 3 trials.

Did HRV biofeedback reduce cravings in the trial?

It went with lower craving. The trial recruited treatment-seeking adults with substance use disorder online across the US. Of 260 people assessed, 120 were randomized to biofeedback plus treatment as usual, or treatment as usual only, for 8 weeks of outpatient care.

Among 115 participants in the analysis (60% female, mean age 46.2), biofeedback went with significant reductions in negative affect and craving. Both reductions were significant (craving z = -4.60, P < .001). The usual-care group had increases in both. Positive affect did not differ.

Outcomes came from repeated momentary check-ins in daily life, not one clinic questionnaire.

Did it reduce actual substance use?

It went with fewer days of use. Biofeedback was linked to a lower proportion of alcohol and other drug use days. The odds ratio was 0.36 (95% credible interval 0.25 to 0.54). The authors describe this as a 64% reduction against controls.

The trial also looked at what happens after craving. Among people receiving biofeedback, craving was less likely to be followed by use (odds ratio 0.84, 0.73 to 0.97).

The authors say biofeedback appears to work in part by loosening the link between craving and later use. They describe the results as supporting biofeedback as a potentially efficacious treatment and call for phase 3 trials.

Why would heart rhythm relate to cravings at all?

Because chronic substance use affects the autonomic nervous system. A review of 33 articles on HRV in alcohol use found resting HRV was reduced in people with alcohol use disorder. It improved only after at least four months of abstinence.

Reactive HRV, the response to a stressor, was higher in those with alcohol use disorder. Higher reactivity went with more craving, faster relapse and more negative mood. The authors say HRV may be a sensitive marker of craving and relapse.

That does not prove biofeedback fixes it. A separate review of acute alcohol effects on HRV said there is almost no research data supporting HRV response to alcohol as a marker of susceptibility to alcoholism. A 2020 review called HRV biofeedback research in substance use disorders nascent.

Does a low HRV number on a wearable say anything about craving risk?

Nothing in this evidence shows it does. HRV varies with things unrelated to substance use. In 149,205 people in the general population, age and sex together explained 17.4% of the variance in RMSSD. Adding lifestyle and psychosocial factors explained less than 0.5% more. Stress, social well-being and neuroticism were not significantly associated.

The craving result came from a structured 8-week biofeedback program. It did not come from watching a daily HRV score. For how wearables produce the number, see how wearables measure HRV and what counts as a good HRV. Related trial evidence is in whether HRV biofeedback reduces anxiety and HRV and anxiety.

What does this not settle?

It is one phase 2 trial. The comparison group received treatment as usual and no comparable activity. The abstract describes no sham control, so an effect of attention or routine cannot be separated from the biofeedback itself.

Participants were adults seeking outpatient treatment, recruited virtually. The results do not show what happens over the long term or outside treatment.

Common questions