Background: Weight Loss Maintenance (WLM) although necessary for the preservation of the health benefits derived from weight loss, is difficult to attain. Research investigating the perceived barriers and predictors of WLM success as well as the impact of WLM success on cardiovascular related biomarkers are usually limited to short-term post weight loss periods not exceeding two years. Studies on commercial weight loss programmes are mostly limited to short-term efficacy studies with little known about the long-term outcomes. Additionally, investigations on WLM so far, are mostly carried out in developed countries with no data on WLM outcomes of commercial weight loss programmes in developing countries such as Ghana, and the associated factors of WLM success. Given the global importance of commercial programmes for weight loss, there is the need to gain insight into their long-term outcomes in developing countries including Ghana, and determine the perceived barriers, predictors, and the associated biomarkers of WLM success. Methods: A retrospective cohort study involving 230 participants who formerly enrolled in a commercial weight loss programme between 2008 and 2016 was undertaken in Accra, Ghana. The prevalence of WLM success was determined using a definition of successful WLM as achieving ≥ 5.0% weight loss below starting weight for a period of at least six months post weight loss intervention and unsuccessful WLM as otherwise. An interviewer-administered questionnaire was used to obtain information on the demographic, behavioural, psychosocial and programme based characteristics of participants as well as the perceived barriers to WLM success. A sub-sample of 112 individuals were selected for biochemical analysis on fasting blood lipids and glucose parameters. Categorical Principal Component Analysis (CATPCA) was employed to determine the various components of perceived healthy eating and physical activity related barriers to WLM success. Multivariate binary logistic regression analysis was conducted to identify the predictors of WLM success. Multiple linear regression was used to determine the association between WLM and the concentrations of lipids/glucose biomarkers. Results: The prevalence of WLM success was 23.9%. The healthy eating related barriers to WLM success were locational, cost of healthy eating, food craving, emotional factors and lack of social support and accounted for 66.9% of the variance in data. The physical activity related barriers to WLM success were the cost of physical activity, environmental factors and personal factors and accounted for 65.6% of the variance in data. Achieving ≥ 10% weight loss at the end of the weight reduction programme increased the odds of WLM success compared to not achieving ≥10.0% weight loss (AOR = 6.72, 95.0% C.I = 3.15-14.31). Similarly, logging physical activity compared to not logging (AOR = 3.52, 95.0% C.I = 1.49-8.32), limiting food portions at meal times compared to not limiting (AOR = 3.51, 95.0% C.I = 1.07-11.57), and a good/excellent perceived competence in carrying out behaviours for weight maintenance compared to a poor/disappointing perceived competence (AOR = 5.93, 95.0% C.I = 1.74-20.19), increased the odds of successful WLM. Being out of the weight loss treatment for three years or more as at the time of this study reduced the odds of WLM success compared to being out of treatment for less than three years (AOR = 0.46, 95.0% C.I = 0.22-0.97). Successful Weight Loss Maintainers (SWLM) compared to Unsuccessful Weight Loss Maintainers (UWLM) had significantly lower concentrations of adjusted serum total cholesterol (TC) (5.69±0.24mmol/L versus 6.26±0.18mmol/L, respectively, P<0.013) and triglycerides (TG) (1.11±0.10mmol/L versus 1.34±0.07mmol/L, respectively, P<0.021). SWLM with 10.0% or more weight loss had significantly lower concentrations of adjusted LDL compared to UWLM (3.86±0.31mmol/L versus 4.46±0.18mmol/L, respectively, P = 0.046). The adjusted serum High Density Lipoprotein (HDL), Fasting Blood Glucose (FBG) and glycosylated haemoglobin (HbA1c) concentrations did not differ between SWLM and UWLM.

show more