Main Article Content

Predictors of weight reduction in a Nigerian family practice setting


Adetola M. Ogunbode
Mayowa O. Owolabi
Olayinka O. Ogunbode
Lawrence A. Adebusoye
Adesola Ogunniyi

Abstract

Objectives: This study identified the predictors of weight reduction among adult obese patients in a Family Practice Setting and developed a statistical model to predict weight reduction.
Design: A prospective cohort design.
Setting: The Family Practice Clinic, University College Hospital, Ibadan, Nigeria
Participants and study tools: Obese adults were recruited into a three-month weight reduction program. Patient Information Leaflets were used for counselling, while questionnaires were administered to obtain socio-demographic and lifestyle factors. Potential predictors were assessed using the Multidimensional Scale of Perceived Social Support, Zung Depression Scale, Rosenberg Self-Esteem scale, Garner’s Eating Attitude Test-26 (EAT-26), 24-hour dietary recall and International Physical Activity Questionnaire-short form. Anthropometric indices, blood pressure and Fast-ing Lipid Profile were assessed. Descriptive and inferential statistics were used for analysis with a significance set at α0.05.
Results: Most 99(76.2%) of the 130 participants achieved weight reduction and had a median weight change of -2.3kg (IQR-4, -0.5), with 66 (66.7%) out of 99 attaining the weight reduction target of 10%. The regression model showed predictors of weight reduction to be Total Cholesterol [TC] (p=0.01) and Low-Density Lipoprotein Cholesterol [LDL-C] (p=0.03). The statistical model derived for Weight reduction = 0.0028 (LDL-C) -0.029 (TC)-0.053 (EAT-26) +0.041(High-Density Lipoprotein Cholesterol). The proportion of variance of the model tested was R2 = 0.3928 (ad-justed R2 = 0.2106).
Conclusion: Predictors of weight reduction among patients were eating attitude score, Total Cholesterol, Low-Density Lipid and High-Density Lipoprotein Cholesterol levels. A statistical model was developed for managing obesity among patients.


Journal Identifiers


eISSN:
print ISSN: 0016-9560