Very low calorie diet (VLCD) followed by a randomized trial of corset treatment for obesity in primary care

Abstract

Objective: The primary objective was to investigate the feasibility and cost-effectiveness of weight reduction using very low calorie diet (VLCD) in groups. The secondary objective was to investigate whether subsequent corset treatment could maintain the weight reduction long term.

Design: Participants, consecutively included in groups of 8-14 subjects, underwent three months of VLCD with lifestyle advice at group meetings. Subjects attaining > or = 8 kg reduction were randomized to corset (A) or no corset (B) treatment for nine months. Weight was registered at all meetings and after 24 months. Costs were calculated using current salaries and anti-obesity drug prices as at 2008.

Settings: Primary care in Skaraborg, Sweden. Subjects. A total of 26 men and 65 women aged 30-60 years with BMI > or = 30-< 45 kg/m2.

Main outcome measures: Weight changes and costs of treatment.

Results: VLCD (dropout n = 14) resulted in a mean weight reduction of 20.1+/-6.6 kg (20 men) and 15.7+/-4.7 kg (57 women). These 77 subjects were randomized to treatment A (n = 39) or B (n = 38). Compliance with corset was only 20% after three months. After one year (dropout n = 17) weight loss was 11.7+/-8.1 kg (A) and 9.3+/-6.9 kg (B), p = 0.23 and after two years (dropout n = 22) 6.1+/-7.0 kg and 4.4+/-7.3 kg respectively, p = 0.94. Serum glucose and lipids were altered favourably. The cost per participant of treatments A and B was SEK 4440 and SEK 1940 respectively.

Conclusions: VLCD in groups was feasible and reduced weight even after one year. The cost of treatment was lower than drug treatment. Corset treatment suffered from poor compliance and could therefore not be evaluated.

Figure 1.

Flow chart over study subjects during very low calorie diet, corset treatment, and follow-up. Notes. VLCD = very low calorie diet. Group A: Corset and lifestyle advice. Group B: No corset but lifestyle advice. The 14 subjects dropping out during VLCD did so very soon (within 2 weeks) after the start of the study.

Figure 2.

Weight change of the completers during the two years of study. Notes: Initially, very low calorie diet for all participants for three months, subsequent corset (treatment A) and no corset (treatment B) for nine months and measurement of weight after 24 months. The value at −1 month was the mean weight at inclusion (n = 91), the value at 0 month shows the mean weight at VLCD start. The standard deviations of weight changes in A and B were at one month 2.9 vs. 2.0 kg, at 2 months 5.1 vs. 4.1 kg, at 3 months 5.7 vs. 4.8 kg, at 5 months 5.9 vs. 5.5 kg, at 8 months 6.2 vs. 5.3 kg, at 12 months 7.9 vs. 7.4 kg, and at 24 months 8.4 vs. 7.6 kg, respectively. The compliance with corset treatment (A) was 100% at start after VLCD completion, 33% after two months, 20% after three months, and 15% and 10% after five and nine months respectively.