Effects of a multimodal lifestyle intervention on body mass index in patients with bipolar disorder: a randomized controlled trial

Abstract

Objective: Patients with bipolar disorder are at increased risk of weight gain, which in turn increases the risk for somatic disease and nonadherence to maintenance therapy. Therefore, interventions addressing weight gain are expedient for the management of this disorder. We set out to evaluate the effects of a lifestyle intervention on body mass index (BMI) and cardiovascular and metabolic parameters in patients with bipolar disorder undergoing mood-stabilizing pharmacologic treatment.

Method: Fifty outpatients with bipolar disorder undergoing mood-stabilizing treatment participated in a randomized controlled trial (waiting control group: n = 24 and multimodal lifestyle intervention group: n = 26). Groups consisted of 2 cohorts (cohort 1: March 2005-February 2006; cohort 2: September 2005-August 2006). The intervention lasted 5 months and consisted of 11 group sessions and weekly fitness training. BMI and body weight as well as cardiovascular and metabolic parameters were determined at 3 assessment points: at pretreatment baseline, at 5 months (end of treatment), and at 11 months (6-month follow-up).

Results: Intention-to-treat analyses showed that the intervention significantly reduced BMI over time (P = .03), with significant and stable mean differences in BMI change between groups of 0.7 kg/m² (95% CI, 0.2-1.3) at 5 months and 0.8 kg/m² (95% CI, 0.1-1.6) at 11 months' follow-up assessment. The lifestyle intervention had no significant effect on cardiovascular and metabolic parameters (all nonsignificant). The BMI reduction was only seen in female patients (P = .003).

Conclusions: BMI in patients with bipolar disorder can be reduced with a long-lasting effect by a multimodal lifestyle intervention. However, this effect was only seen in female participants, indicating the need for gender-specific interventions.

Trial registration: clinicaltrials.gov Identifier: NCT00980863.

Figures

Figure 1

Flowchart for the Multimodal Lifestyle Intervention Trial

Figure 2

Effects of the Intervention on Body Mass Index Levelsa,b aBody mass index levels (mean/SD) over time between intervention (open squares) and control (black squares) groups in the total sample (N = 50, intervention group: n = 26, control group: n = 24) and in women (intervention group: n = 13, control group: n = 10) and men (intervention group: n = 13, control group: n = 14). bThe intervention led to a reduction between baseline to 11-month follow-up assessment of 0.3 kg/m2 in the intervention group, whereas the control group experienced an increase of 0.4 kg/m2 (intervention by time effect, P = .03; reanalysis for gender: women, P = .003 and men, P = .74).

Figure 2

Effects of the Intervention on Body Mass Index Levelsa,b aBody mass index levels (mean/SD) over time between intervention (open squares) and control (black squares) groups in the total sample (N = 50, intervention group: n = 26, control group: n = 24) and in women (intervention group: n = 13, control group: n = 10) and men (intervention group: n = 13, control group: n = 14). bThe intervention led to a reduction between baseline to 11-month follow-up assessment of 0.3 kg/m2 in the intervention group, whereas the control group experienced an increase of 0.4 kg/m2 (intervention by time effect, P = .03; reanalysis for gender: women, P = .003 and men, P = .74).

Figure 2

Effects of the Intervention on Body Mass Index Levelsa,b aBody mass index levels (mean/SD) over time between intervention (open squares) and control (black squares) groups in the total sample (N = 50, intervention group: n = 26, control group: n = 24) and in women (intervention group: n = 13, control group: n = 10) and men (intervention group: n = 13, control group: n = 14). bThe intervention led to a reduction between baseline to 11-month follow-up assessment of 0.3 kg/m2 in the intervention group, whereas the control group experienced an increase of 0.4 kg/m2 (intervention by time effect, P = .03; reanalysis for gender: women, P = .003 and men, P = .74).