This is a simple and well-reported study. However:
(1) Patients were not randomly assigned to groups, which may introduce bias.
(2) More details about the sensitivity analysis would have been useful.
(3) The impact of side-effects was considered neither on the final benefits nor on total costs.
Moreover, the authors recognised the problems associated with data collection,the setting of the study, and the failure to investigate the effects of variables such as diet and exercise.