The results of this review demonstrate that small improvements in body weight, BMI and QOL are possible through psycho educational and/or behavioral interventions targeting PA and/or eating habits in persons with SMD.
In the intervention groups, weight loss and a decrease of BMI is observed in 11 studies. In the control groups, a decrease in mean body weight and BMI is reported in only four studies. The difference in weight change between intervention and control groups is statistically significant in nine studies. Differences in mean endpoint BMI between intervention and control groups are statistically significant in eight studies.
Beside the significance of results it is also important to give attention to the clinical relevance of these results. According to the UK Department of Health  reductions in body weight of 5.0% or more are considered to greatly reduce the risks of physical health problems. In the included trials in our review, no study achieved this target. Moreover, only in a limited number of studies the weight loss resulted in changes of BMI classification (e.g. from obesity to overweight or from overweight to normal weight). In this sense, participants in the studies included in the review may lose weight following a lifestyle intervention but still remain in the overweight or obesity class. It can thus be questioned to what extent the BMI is useful to identify the risk for developing CVD. It appears that, according to the results of several studies, the measurement of the waist circumference and the waist-hip ratio is more appropriate than measuring the BMI to estimate the risk for future cardiovascular events [71, 72]. It is important to note that in all 14 trials participants were taking atypical antipsychotics. It is well known that these drugs cause weight gain . Furthermore, there is conclusive evidence that persons with SMD are more likely having sedentary lifestyles, making poor dietary choices and are more likely to smoke .
In eight studies included in the review, no information was provided about the weighing process (time of weighing, scale, clothing). The remaining six studies provided, however not always detailed, information about the weighing process. To minimize weighing errors, patients should be weighed at the same time of day using the same scale and in light clothing without shoes . Including this information in the methods section of studies is important to enable an appropriate assessment of the results. According to the results of a study of weight measurement protocols no advantage was found about measurement of weight on two separate days compared with measurement on a single day . However, as far as we know, literature on this topic is scarce and further studies evaluating weight measurement procedures are required.
There are some limitations to acknowledge. For this review, only references from the period 01/03/1990 until 01/03/2010 were included. In this way, it is possible that we miss relevant papers from the period before 1990.
The trials included in this review are frequently limited in terms of small sample sizes, short intervention periods and absence of long-term follow up. This raises questions about the generalization of the results to wider populations with mental disorders. In their review, Lowe & Lubos concluded that the current literature on weight reduction interventions appears to provide limited evidence on the effectiveness either of psycho educational or of programs including educational and exercise components. They also concluded that more research is needed, with larger sample sizes as well as standardized outcome measures to determine and compare the effectiveness of these kinds of interventions.
The focus of the review considered 'healthy living interventions' in persons with SMD. This created a broad spectrum of 'healthy living interventions' topics and approaches. From a research perspective it may have been more appropriate only investigating one type of intervention. However, people with SMD share many risk factors, so it appears that general interventions will be more beneficial . Weight management programs through healthy eating, exercise and tobacco cessation should be integrated into mental health care . In this sense, it was decided to include randomized and non-randomized controlled trials with focus on health promotion interventions targeting PA and eating habits with primary outcomes changes in weight and BMI. Beside 'healthy living interventions' also pharmacological interventions to control weight gain in persons with MD have been evaluated. In a review of pharmacological and non-pharmacological interventions to control weight in persons with schizophrenia, the authors concluded that non-pharmacological interventions are preferable. The promising results in the non-pharmacological studies must however be tempered by weaker designs and small sample sizes used in these studies . According to the results of a review on the mechanisms and management of antipsychotic weight gain in schizophrenia, it was concluded that pharmacological agents like orlistat and sibutramine have not been sufficiently evaluated in antipsychotic weight gain 
Despite the limitations, it is promising that small decreases of body weight and BMI in this population are possible. It appears that health promotion interventions targeting PA and eating habits in persons with SMD may be useful for prevention of weight gain. It is yet important to note that there may be some patients (outliers) that may derive tremendous benefit, but that the mean changes observed in the group are modest. Categorical outcomes would enable the calculation of the number needed to treat for the observation of a clinically significant benefit for the intervention. Only one study  reported confidence intervals around the changes in weight or BMI. Data on confidence intervals could however give insight in the ranges and thus in the solidness of weight changes.
Furthermore, persons with mental health problems usually want to learn more about healthy lifestyles and background theories of lifestyle interventions [78, 79]. In a study of perceptions of barriers to and benefits of PA among patients with SMD, participants saw exercise as positive and desirable, with benefits for both physical and mental health . This suggests that persons with SMD are prepared to participate in health promotion interventions.
Such findings support the integration of health promotion interventions targeting PA and eating habits into mental health care, whereby patients should be motivated to follow these kinds of interventions. When health promotion becomes a part of daily care, mental health professionals could play an important role in motivating their patients to participate. According to patients' perceptions mental health professionals can provide support, motivation, and structure and they feel comfortable with this support .
Elements of QOL were only investigated in five studies, providing no homogeneity of the effectiveness of 'healthy living interventions' on QOL and general health. Yet, improvements in QOL were, although not always statistically significant, observed. This is important because weight gain is associated with perceptions of poorer QOL and general health .
As far as known to the authors, this is the first systematic review of 'healthy living interventions' targeting PA and eating habits in persons with SMD, in which special attention was given if any of the included studies also examined the cost-effectiveness of these interventions. There is a growing need on health economic research in health care and health policy. Especially, attention is given to health economic evaluations of medicines and technologies. Recently, more attention is given to health economic evaluations of preventive health care. In general populations, research on cost-effectiveness of 'healthy living interventions' produces no conclusive evidence [12, 13], which is likely explained by wide differences in program contents. In persons with SMD, no studies examining the cost-effectiveness of 'healthy living interventions' targeting PA and eating habits were found. Yet, such research has a great social value. Prevention has an economic cost, but it can also save money because diseases and complications can be avoided. Finally, prevention can produce healthy life expectation.
Eight of 14 studies included in this review were conducted in the USA. Because of differences in e.g. the management of health care, health insurance systems, and access to hospital care further studies, especially in European countries examining both effectiveness and cost-effectiveness of lifestyle interventions targeting PA and eating habits in persons with SMD are required to assist in the development of new health promotion interventions in this population. Concerning cost-effectiveness of interventions the viewpoint for the analysis (e.g. point of view of society, the Ministry of Health, the patient) should be carefully considered. Emphasis should also be put on long-term effects of these kinds of interventions.