Bright light for weight loss: results of a controlled crossover trial

Abstract

Objective: To investigate whether bright light treatment can reduce body mass in overweight subjects irrespective of their seasonal (= light) dependence.

Methods: A crossover, placebo-controlled, randomized clinical trial was performed between November and April in Novosibirsk, Russia (55° N). The trial comprised a 3-week in-home session of morning bright light treatment using a device of light-emitting diodes and a 3-week placebo session by means of a deactivated ion generator, separated by an off-protocol period of at least 23 days. The number of placebo and light sessions was matched with respect to season. Data were obtained from 34 overweight women, aged 20-54 years, 10 were seasonal-dependent according to the Seasonal Pattern Assessment Questionnaire. Weekly measures included body weight, percentage body fat by bioimpedancemetry, and subjective scores (appetite, mood, energy levels).

Results: Motivation and expectation towards weight loss were similar for the two intervention sessions. With light, compared to the placebo session, weight did not reduce significantly, but percentage fat, fat mass, and appetite were significantly lower (average fat reduction 0.35 kg). The latter two results remained significant after excluding seasonal-dependent subjects from the analysis. Irrespective of the type of intervention, seasonal-dependent subjects had greater weight and fat mass changes during treatment (decline p < 0.036) or between sessions (regain p < 0.003). Photoperiod (p = 0.0041), air temperature to a lesser extent (p = 0.012), but not sunshine (p = 0.29) was associated with the weight change (greater weight reduction if the second session was in spring).

Conclusion: Morning bright light treatment reduces body fat and appetite in overweight women and may be included in weight control programs.

Trial registration: ClinicalTrials.gov NCT00406770.

Fig. 1

Dynamics of body mass and subjective scores following light and placebo interventions in 34 women wishing to lose excess weight. The week 0 value is assigned to 100%. Difference between corresponding values at light and placebo sessions: ★p ℋ 0.05, ★★p ℋ 0.01, ★★★p ℋ 0.001, by either paired Student t-test (for body mass) or paired sign test (for subjective scores), based on absolute values (see column ‘∆∆’ in table 3).

Fig. 2

Difference in weight reduction between the 2nd and the 1st trial sessions depending on the time of year (downward bars indicate that the 2nd session was more efficacious than the 1st one). The difference was greater when the 2nd session occurred in spring than in winter (first two vs. last two bars combined, effect of season). However, the light session was consistently more efficacious towards weight loss than placebo session (open vs. dark bars combined, effect of intervention). Two subjects who performed the study over two different winters, were not included in the diagram; both were non-seasonals. The effect of seasons is obviously attributable to a change (∆) in photoperiod, not ambient light (as January and February happened to be very sunny during both years).