Prevention of weight gain is an important step in reducing the impact of obesity on population health, yet there are few reports of successful interventions to date. Here we have reported on a low intensity intervention specifically to reverse the current trend of weight gain seen in Australian women.
Self-reported weight is commonly used in interventions due to the convenience and low cost. While it is considered to under report actual weight, it has been suggested self-report may accurately reflect changes in weight . We did not weigh the control group at 4 months to avoid any unplanned intervention effect of weighing as the longer term study was still in progress. While we cannot be sure the self-reported weight changes described are not due to under-reporting, both groups were weighed at baseline and this may have contributed to more accurate self-reporting at 4 months.
It appears a similar small weight loss has occurred in both the comparison group and the intervention group. The weight change itself is of limited clinical importance because of the short term nature of the study, but it does point to interesting short term behavioral change in women with young children who are difficult to engage in health interventions. Previous reports including our own suggest the majority of women are dissatisfied with their weight even those within a healthy BMI range, and frequently use weight loss behaviors [23, 24]. It is interesting then that even a single education session in the control group resulted in small weight loss in the short term. The program may have influenced early motivational changes in women as described in the Health Education Model, or influenced women at the contemplation and action stage as described in the Trans Theoretical Model. As participation was voluntary there is potential for a more motivated group to be recruited in both groups and this may partly explain the weight loss. However the women were invited to attend a lifestyle program, not a weight loss program and we anticipated we might recruit those who were already active, consuming a healthy diet and maintaining their weight. However the recruited participants included those who were overweight and obese as well as normal weight and the proportion who were inactive was similar to population levels as we have reported previously. So whilst some women may have attended with the intention of losing weight, we do not believe the majority of women volunteered with this intention. If in fact we did recruit a sample that was already engaging in healthy lifestyle behaviors we would expect to see an attenuated effect among the respondents, and hence the effect size is an under-estimate of the effect which would be seen in all women. This suggests that being weighed and some simple general guidelines provided in a community setting may be sufficient to assist short term behavior change leading to better weight control in some women, at least in the short term. The greater weight loss reported in overweight and obese women suggests that even a single education session will encourage significant weight loss in some, but not all, overweight and obese women. Previous studies suggest this is not likely to be sustained long term. It is possible some women respond well to community interventions and it may be important to determine how these women can be identified to improve outcomes. Of concern is the large number of women who reported weight gain in each BMI category even in this short period.
The intervention group not only achieved weight loss, but there was a trend toward small, but potentially important changes to health related behaviors such as physical activity. Accurately measuring small changes to dietary intake and physical activity is difficult due to the lack of sensitive tools. We were unable to determine a difference between groups in fat intake behaviors in the short term. Both groups reported replacing high fat foods and snacks with lower fat alternatives such as fruit and vegetables, which potentially reduces the overall energy density of the diet and possibly contributed to the small energy deficit and weight loss. Some of the lower fat behaviors such as avoiding frying food and using low fat meats are well known messages throughout the community. Some women may have already been practicing these behaviors resulting in little further change attributed to the intervention. However, the small improvements in physical activity and diet and the associated changes in energy balance are consistent with that expected to prevent population weight gain long term, estimated to be as little as 60 kJ per day . The weight change reported here is important if sustained long term in populations . The longer term results of this study will help determine the clinical importance and sustainability of the small energy deficit and weight change.
Self-weighing appears to facilitate weight loss when included as part of a behavior change intervention. There is only one report of using self-weighing in interventions specifically to prevent weight gain, although there are reports of the successful use of weight monitoring in obesity treatment and the subsequent prevention of weight re-gain . Our results confirm those seen in the 'Pound of Prevention' a large community prevention intervention, where at least weekly weighing was associated with weight loss, although they did not show a difference between intervention and control. The study showed no detrimental effect of self weighing . Simple methods of monitoring progress such as self-weighing may be a viable population recommendation although potential detrimental effects would need to be fully evaluated first.
Self-efficacy has been reported to be a predictor of successful behavior change. In this study, we were not able to show a correlation between self-efficacy and weight change despite positive correlations in other studies . Self-efficacy declined over this short period, particularly in the control group. Intervention components such as outcome expectancy discussions, social reinforcement and modeling may have helped maintain self-efficacy in the intervention participants. The decline in self-efficacy for physical activity cannot be explained easily, although Bandura theorized that past successful experience is a powerful influence on self-efficacy and is supported by a study by Hofstetter . Declining self-efficacy for physical activity in women with children has also been demonstrated previously . This group may therefore need additional support to maintain self-efficacy for physical activity behaviors. Possibly, failure to employ the planned physical activity may have led to a decline in confidence to participate in future activity. This is explained partly through research on mediators of self-efficacy .
This study is important as it is the first to target mothers of young children specifically for the prevention of weight gain. This intervention recruited a representative sample of community dwelling women with children in Australia which along with other baseline characteristics has been reported previously . Overall approximately 11% of invited participants responded. In the context of population reach the school setting gave access to all available women with children, unselected in terms of health risks. Reaching and changing the behavior in 11% of women with children translates into a substantial health promotion intervention
We were not able to show a significant difference between the groups in weight change and some specific lifestyle behaviors. The inability to detect these differences may reflect the inadequacy of the available self-report questionnaires to detect small changes. We cannot eliminate the possibility that self-reported improvements in diet and physical activity may also be the result of positive response bias. It is also possible seasonal variation contributed to the weight change although this was limited by having a rolling period of data collection. The study was powered to detect a 600 g difference between groups in the long term, and not adequately powered to detect the small difference observed here at four months. The intervention aimed for small consistent changes to lifestyle and this small energy deficit is difficult to measure with available tools. Despite this, the weight change and the behavior change seen in the intervention group is encouraging for the longer term success of this intervention.