Ethnicity contributes to overweight related preventive behaviour for almost all outcome measures of this study. The outdoor play of Turkish children is less and snack intake is higher compared to native Dutch children. This is as expected, considering the large difference in prevalences of overweight and obesity in these ethnic groups. Dutch parents have stronger intentions to monitor outdoor play and snack intake than Turkish parents. Remarkably, for the child’s snacking behaviour an association with ethnicity was absent when controlled for the parental cognitive TPB factors. The TPB determinants are relevant for all the physical activity and diet outcomes of this study. With regard to the parental intention to monitor the child’s outdoor play and snack intake, attitude and social norm are the main contributing factors. With regard to the child’s outdoor play behaviour, the parent’s perceived behavioural control is an important contributor. For the child’s snacking behaviour, perceived behavioural control is a significant factor, especially in the group of parents who consider their child as overweight.
As most associations between the TPB-factors and overweight related preventive behaviour are not modified by ethnicity, it may be assumed that a deep structure in the cultural and psychological processes influencing overweight related behaviour specific for the ethnic groups is absent
. Only in one instance, effect modification by ethnicity was found. The intention of Turkish parents to monitor outdoor play is less dependent on the perceived social norm than in Dutch parents. In the qualitative interviews prior to this study mothers of Turkish children indicated that they received little support from their family in promoting overweight preventive behaviour, which may explain their independent position regarding their child’s outdoor play (unpublished data). An earlier study of dietary behaviour in ethnic groups also found an absence of cultural differences in parental perceived barriers or self-efficacy
. Furthermore, the parents’ beliefs about the role of genetic factors in overweight do not contribute significantly to the parental intentions and child’s behaviours with regard to physical activity and diet.
The inclusion of TPB determinants explained the largest proportion of variance in the regression models, however the sizes of theses proportions are still limited. Especially more insight is needed into factors determining the child’s snacking behaviour. Literature shows that parenting styles and environmental factors, that were not included in this study, are also relevant for obesity prevention
[19–21, 38]. Moreover the child’s individual attitudes with regard to playing outdoors and snacking will play an increasing role when they grow older. These factors may add to the understanding of the child’s snacking behaviour.
Although overweight and obesity were three times as high in the children of the Turkish response group, parental beliefs in this group about overweight prevention and management did not reflect a sense of urgency about changes in their child’s behaviour. A number of other studies in native and ethnic minority children of normal weight and overweight found that parents underestimate their children’s weight and are relatively unconcerned about childhood obesity
[16, 17]. In this study, child overweight status, as established objectively with weight and height scores (BMI) was also not important in terms of parents’ intentions to prevent or manage overweight, but parents’ subjective perceptions relating to a child’s weight were.
A limitation of this study is the cross-sectional study design. A longitudinal design could shed more light on the causal determinants of child behaviour and parental intention. The difference between the Dutch native and one large ethnic minority group is examined, which limits the generalizability of this study. However we were interested in relevance of cultural aspects for dietary and physical activity behaviour in general. Further testing of the relationship with behavioural cognitions in other minority groups is advised. Another limitation of this study is that questions were asked about parental intentions to encourage their child to play outdoors or monitor snack intake, while we asked for their child’s actual behaviour. The moderate correlation in our study between the intention to pay attention to the snack intake and the scale monitoring unhealthy eating behaviour from the Child Feeding Questionnaire is an indication of an association between parental intention and their monitoring behaviour. We were not familiar with a similar validated instrument on parental monitoring behaviour of the child’s physical activities, therefore actual parental monitoring behaviours were not measured. A third limitation may be the influence of the order of items and scales of the questionnaire on the respondents’ answering on certain types of questions, such as on perceived child weight and the child’s behaviour. These questions were placed at the beginning of the questionnaire to prevent answers influenced by preceding questions as much as possible, however possible response effects as a result of question order were not systematically pre-tested in this study. A final limitation of this study is its restriction to outdoor play and snacking behaviour. Other overweight related preventive behaviours, such as daily breakfast or sports participation, were not examined.
A strength of this study is the inclusion of an ethnic minority population in which parents of overweight children are overrepresented. This enabled us to explore beliefs and intentions in a population faced with a high prevalence of overweight children. The characteristics of the Dutch and Turkish response groups did not differ from the general populations in terms of educational level of the parents, and overweight and obesity, indicating the representativeness of our study results for these populations. The BMIs of the children of non-respondents could not be compared with those of respondents since we obtained informed consent to use register data from respondents only, and not from non-respondents.
The instruments and scales used in this study to measure intentions and behaviours were tailored to the ethnic minority response group using results from qualitative interviews. This meant that the number of items had to be limited, but the words and expressions used in the questionnaire are familiar to the Turkish respondents. The further development of robust instruments and scales for ethnic minority groups is recommended.
Several implications of the study results for preventive practice can be mentioned. Special attention has to be paid to the ethnicity of the target population when addressing parental overweight related preventive behaviours, given the differences between ethnic groups with regard to the behavioural outcomes of this study. The finding in literature that it is important to use a cultural approach to prevention, appropriate for norms and customs of the ethnic group, also applies to overweight and obesity behaviour
[15, 28, 29]. Adaption of interventions to places and locations where the ethnic groups can be found, language, values and norms with regard to food quality and intake, and other cultural aspects are important to improve their reach among the targeted audiences
. As for most outcomes no interaction could be found between TPB-factors and ethnicity, it is hypothesized that preventive interventions targeting parents who contemplate or intend to change their behaviour, may follow a general theory-based approach to the social norms and attitudes with regard to overweight prevention or management in children, regardless of the ethnicity of the target population. However, for the intention to monitor outdoor play, the perceived social norm in Turkish parents could be strengthened, which implies that the perceived lack of support from the family with regard to overweight prevention has to be targeted also. When change of the actual behaviour of the child is aimed, the parent’s perceived behavioural control with respect to monitoring the physical activity and diet of the child appears an option for intervention. Furthermore, educational programmes and screening programmes should take into account the parent’s perception of a child’s weight status. Further research is advised into behavioural determinants regarding children’s dietary behaviour. Moreover, more insight has to be gained into behavioural determinants of dietary and physical behaviours in other ethnic groups. It is generally recommended that interventions to prevent overweight and obesity should start at an early child age. Our finding that parents intend to monitor physical activity and eating behaviour when the child is still young supports this recommendation.