Background
Obesity is an important risk factor for chronic non-communicable diseases globally [
1]. In South Africa, like many other developing countries, adult women are the most vulnerable group, with a markedly higher prevalence of obesity than men [
2]. There is substantial evidence associating body image to women's response to weight changes and attitudes towards weight control [
3‐
5]. In South Africa for example, women are less likely to see themselves as overweight, irrespective of body size [
6,
7]. In their study Puoane et al. [
6] found that only 22.1% of South African women of all races perceived themselves as being overweight, when in fact 56.6% of women interviewed were classified as overweight and obese. These findings were influenced by ethnicity in that only 27% of overweight or obese black women correctly perceived themselves as overweight compared to 65% of mixed ancestry and 100% of white overweight or obese women. Similar results have previously been observed in urban South African adolescent girls in that black adolescent participants were less dissatisfied with their body size and shape, and were also less likely to desire a smaller body size than mixed ancestry and white girls [
5,
8].
Similar results have been observed in America in that, low-income African American women of different age groups tend to have fewer weight-related body image concerns than their white counterparts [
9‐
12]. Similarly, Flynn and Fitzgibbon [
12] showed that normal weight black adolescent girls have a preference for a larger body size compared to white adolescent American girls. Accordingly, one may surmise that the black adolescent population is less motivated than the white adolescent population to engage in behaviors that would prevent the development of obesity. In accordance with these findings, Stevens et al. [
10], in their study comparing overweight white and black adult American women, found that black women were 40% less likely to feel guilty after eating, 2.5 times more likely to be satisfied with their weight and 2.7 times more likely to consider themselves attractive than white women. In this regard, the tolerance of a bigger body size status in black women appears to be the modifier of high self-esteem and positive body image. Moreover, in the same study, among those women who were not overweight, white American women perceived themselves to be larger in body size and reported lower ideal body weight compared to black American women [
10].
While the previous studies argue strongly in favour of ethnic-specific differences in body size tolerance in women, the extent to which these differences may be attributed to intra-familial effects is not clear. Indeed, various international studies have previously shown that the family environment, particularly the mother-child relationship, has an influence on obesity risk [
11‐
15]. In these studies, the relationship between mothers and daughters has been found to be stronger than that between mothers and sons. Further, the mother-daughter relationship has been shown to influence body image, with mothers unintentionally modelling both positive and negative body image to their daughters [
13,
15].
There is little research directed at exploring the socio-cultural factors influencing body size perceptions and attitudes between mothers and daughters in relation to obesity in countries undergoing epidemiological transition, such as South Africa. This may be considered in the context of the co-existence of maternal over-nutrition and childhood under-nutrition, which has been described in Africa [
16].
In our previous study of 204 South African women and 333 pre-adolescent girls from different ethnic groups, we showed that overall, participants had appropriate perceptual body size, in that positive correlations were found between the silhouettes chosen by the participants to represent their 'feel' and their actual/measured body size and body fat percentage [
3]. However, these relationships were altered by ethnicity, particularly with respect to body size tolerance in the girls, in that black girls had less body size dissatisfaction at higher actual BMI than white and mixed ancestry girls [
3].
Therefore, the aim of this study was to identify the extent to which family status (presented as mother-daughter resemblance) and ethnicity, might explain differences or similarities in the body image and perceptions of South African mothers and their pre-adolescent daughters.
Discussion
This study provides novel insight into the important and respective roles of family membership or maternal modelling and ethnicity on different dimensions of body image such as body size perception, and body size dissatisfaction in women and their daughters from South Africa. Intra-familial resemblance for perceived body size, ideal body size and body size dissatisfaction were found when the potential confounding of maternal body size was removed. In addition, mothers and daughters from black families demonstrated an overall greater body size tolerance than their white and mixed ancestry counterparts. To our knowledge, this is the first time in South Africa that these two constructs, family membership and ethnicity, have been compared in the same population, with respect to different dimensions of body image in relation to obesity. These results corroborate those of international studies [
29,
30], suggesting intra-familial resemblances in body image between mothers and their pre-adolescent daughters. Further, these results corroborate both local and international results that have shown a greater tolerance for a bigger body size in women of African descent, compared to other cultures [
4,
7,
8,
10‐
12,
26,
31].
Striking findings from this study were that black girls seemed to differ from their mothers in terms of body size preference, which contrasts to the findings of Hill and Bhatti [
30]. In this study, black girls were leaner, yet they preferred a larger silhouette. On the other hand, the majority of black mothers were obese, yet they preferred leaner silhouettes. What is of concern is that South Africa studies suggest that overweight black women are resistant to adopt health behaviours, despite knowing that they are at an increased risk for non-communicable diseases [
28,
32‐
34]. This resistance may be partly endorsed by the weight-loss stigma associated with the HIV/AIDS wasting syndrome [
34], and the fact that, being overweight in the black culture is a symbol of wealth, autonomy, attractiveness and happiness [
16,
27,
28,
32,
33].
Indeed, black women participating in the current research were more likely to regard fatness as a sign of health, beauty, respect and happiness. Similarly, other South African studies [
16,
27,
28,
32,
33] have shown that in the black communities of South Africa, being overweight is desirable as it denotes beauty, happiness and affluence. Further, Matoti-Mvalo [
34] reported that these beliefs are now further exacerbated by the idea that being thin can be equated with HIV/AIDS virus infection. The following quote typifies this: 'If you are thin, people think that you are sick -- you may have TB or HIV/AIDS' [
32,
34]. Based on these beliefs structures, black South African women may be more reluctant to lose weight than women of different ethnic origin. Interestingly, white and mixed ancestry families participating in the current research only associated a leaner body size with beauty, health and happiness. Indeed, Brink [
35] has highlighted that in the Western culture, thinness does not just mean the size of the body, but thinness is associated with qualities of being healthy, attractive and in control. In contrast, a fat body is viewed as a sign of poor health, laziness, sloppiness and lack of personal will [
35,
36].
Research internationally has also shown that the social stigma associated with being obese is more prevalent in women than men, and women are more likely to be discriminated against because of their weight [
37]. Some of the social stigmas attached to obesity relate to the attributes and cultural emphasis placed on appearance and especially women's body size [
37]. In the current research it has been clearly shown that girls perceived that their friends were more tolerant of their body size than their mothers. The most likely explanation for these differences may be that the girls, on average, were within the normal range for expected weight for height, compared to the maternal cohort, who had a mean BMI of 27.8 kg/m
2. This is in line with studies which show that self-esteem tends to be higher and body size dissatisfaction lower in children who are within the normal range of body weight [
38,
39]. In accordance with other studies [
8,
39] ethnicity modulated these effects, with the perception of body size tolerance amongst friends and peers being greater in black girls and their mothers, compared to other ethnic groups. Young-Hyman et al. [
40] also suggested that body size acceptance by the primary caregivers of children may also be influenced by cultural differences. Indeed, Puoane et al. [
5] recently highlighted the ambiguity concerning overweight and obesity in urban South African girls from black families in which two-thirds of those surveyed associated obesity with happiness and wealth and the remainder were ambivalent. Thus, in terms of societal norms, current body size status and culture may influence both one's individual body size satisfaction, as well as one's perception of how they are viewed by others.
In summary, this study has shown that despite South African preadolescent girls being significantly less dissatisfied about their body size compared to their mothers, intra-familial resemblances for body size dissatisfaction existed when the potential confounding differences in maternal body size were removed. Further, this study has highlighted that society and culture mediates body size dissatisfaction. These results have important implications for the development of obesity in South Africa, given the high prevalence of obesity in women, which also differs between ethnic groups. This strongly suggests that in South Africa health promotion needs to be ethnic-specific, and should always involve families, not individuals. Strategies and interventions should be directed at increasing the awareness of a healthy body size status and maintaining it in an attempt to prevent obesity. This is highly important for those populations that are at the highest risk of becoming overweight and obese, but may not be bothered when they are overweight, black South African women in particular.
There are a number of limitations to this study. Firstly, the mothers and their daughters' eating and exercising behaviours and attitudes were not analysed, as this was beyond the scope of this paper. However, this is of relevance as international studies have shown that family environment and parental modelling influence a child's dietary intake, diet quality and participation in exercise activities [
11,
12,
41]. Secondly, we only included preadolescent girls. The inclusion of adolescent girls may have provided further insight regarding children's attitudes and perceptions towards body image, which has been shown to be influenced by children's sexual maturity. Future studies including this group are encouraged.
Conclusions
To conclude, the findings from the this study suggest that South African researchers, educators and health promoters should consider the effects of family environment and ethnicity on body image, when developing intra-personal and targeted interventions for the prevention and management of obesity. Most particularly, health education should not only be directed to the affected (the overweight/obese), but also include the whole family, so as to help dispel the myth and stereotypes suggesting "big" to be beautiful, healthy and respected. More focus is to be directed to those under-served and vulnerable communities, young black South African pre-adolescent children, in particular, who are at risk of developing to be overweight adults.
Competing interests
The authors declare that they have no competing interests.
Authors' contributions
ZJM - conducted literature search, conceptualized the information in the paper, produced the first draft of the paper, made all the changes resulting to editing by co-authors and finalised the paper; JHG -edited the paper; VEL- helped in the conceptualization of the paper, the statistical analysis, the review of the literature and edited all the drafts of the paper.
All authors have read and approved the final manuscript.