Body image of children and adolescents with chronic illness: A meta-analytic comparison with healthy peers

Abstract

This meta-analysis integrates results from 330 studies on differences between body image of children and adolescents with and without chronic physical illness. Young people with a chronic illness had a less positive body image than their healthy peers although the average size of differences was small (g = −.30 standard deviation units). A comparison of diseases showed that young people with obesity (g = −.79), cystic fibrosis (g = −.50), scoliosis (g = −.41), asthma (g = −.37), growth hormone deficits (g = −.35), spina bifida (g = −.23), cancer (g = −.20), and diabetes (g = −.17) evaluated their body less positively than their healthy peers. Furthermore, levels of body dissatisfaction varied by age at onset of the disease, method for assessing body image, ethnicity, year of publication, and comparison group. Recommendations are stated for reducing effects of chronic illness on the body image of people with chronic illness.

Introduction

The body image (BI) encompasses body-related self-perceptions and self-attitudes, such as body (dis-)satisfaction (Cash, 2004). The subjective evaluation of one's appearance is a central aspect of the BI (Smolak & Thompson, 2009). A negative BI is associated with depressive symptoms, social anxiety and avoidance of social situations, low self-esteem, disordered eating, and other forms of less than optimum self-care (e.g., Laurouche and Chin-Peukert, 2006, Neumark-Sztainer et al., 2006, Smolak and Thompson, 2009, Varni et al., 1995).

Epidemiological studies across the world indicate that 15% of children and adolescents have a chronic health condition (van der Lee, Mokkink, Grootenhuis, Heymans, & Offringa, 2007). Chronic physical illness and disability are risk factors for developing a negative BI. First, many chronic diseases have negative effects on appearance (e.g., cleft lip, obesity, scoliosis) and body function (e.g., cancer, epilepsy, spina bifida). Thus, children with chronic illness and disability may feel that they look different from healthy peers (e.g., looking sick or unattractive) and become dissatisfied with their body (Laurouche and Chin-Peukert, 2006, Lee et al., 2012). Second, body-altering side effects of treatment may adversely affect BI. For example, aggressive treatments of cancer affect appearance due to hair loss, changes in body weight, or even amputation (Laurouche & Chin-Peukert, 2006). Third, given the high importance of physical attractiveness in Western societies (Wertheim & Paxton, 2011), it could be expected that children and adolescents with visible physical differences often experience discrimination. In fact, children with physical diseases and disabilities are, on average, less liked by their peers than healthy children (Latner & Stunkard, 2003). Children with visible disfiguration are at higher risk for experiencing stigmatization than other children (Masnari et al., in press), and appearance-related teasing has been found to be particularly detrimental to the development of a positive BI (Eisenberg, Neumark-Sztainer, Haines, & Wall, 2006).

However, differences between the BI of children with and without chronic illness and disability may be smaller than expected. For example, young people with chronic illness and disability may compare themselves to other chronically ill children or adolescents, which could result in feeling similar to or even being better off than some other patients (Melzer & Rouke, 2005). In line with this consideration, average associations between severity of disfigurement and BI of children and adolescents tend to be small (r = −.25 to −.34; Lawrence et al., 2007, Sheerin et al., 1995).

A negative BI has consistently been found in young people with obesity. For example Herbozo and Thompson (2009) reviewed 6 comparative studies on children with overweight and/or obesity showing elevated levels of body dissatisfaction. Similarly, Neumark-Sztainer (2011) reported that findings from quantitative and qualitative studies on children and adolescents indicate that overweight or obese youth is at increased risk for body image concerns.

However, results of studies and narrative reviews on other chronic illness or disabilities have often been inconclusive. Fan and Eiser (2009) reviewed 11 studies on BI of young cancer patients that allowed for comparisons with healthy peers. No consistent evidence was found for elevated levels of body dissatisfaction in children with cancer. However, a narrative review of 8 qualitative studies found increased body dissatisfaction in children with cancer (Lee et al., 2012). In addition, a review by McQuaid, Barton, and Campbell (2000) on body image issues of children with burns only detected a lack of studies that compare the BI of these children with that of their healthy peers.

Two illness-specific meta-analyses have been published. Shields, Murdoch, Loy, Dodd, and Taylor (2006) found no systematic differences between perceived appearance of children with cerebral palsy and their healthy peers, but only three of the included studies provided relevant information. However, another meta-analysis of seven studies found that young people with spina bifida evaluated their appearance more negatively than their healthy peers although differences were, on average, small (d = −.26 standard deviation units; Shields, Taylor, & Dodd, 2008).

Because comparative meta-analyses of BI of children with chronic physical illness and disability are lacking for most chronic physical illnesses, the goal of the present meta-analysis was to integrate the available research across a broad range of diseases. With regard to the first research question, we expected that children and adolescents with chronic physical illness and/or disability have, on average, a more negative BI than their healthy peers. Given the heterogeneity of results on individual diseases (Shields et al., 2006, Shields et al., 2008), the second research question asks which chronic diseases are associated with elevated levels of body dissatisfaction. We compare 19 diseases of which 17 have not been addressed in previous meta-analyses on BI of children and adolescents. In the final research question, we ask whether differences between the BI of young people with and without chronic illness vary by other study characteristics, such as the mean age or representativeness of the sample.

Extent of disfigurement. There is variability regarding whether and to what extent individual chronic illnesses and their treatments affect appearance. Greater dissatisfaction with one's appearance could be expected if the disease has stronger negative effects on appearance, for example due to higher risk of appearance-related teasing (Cattarin and Thompson, 1994, Masnari et al., in press). Small associations between extent of disfigurement and body dissatisfaction of children and adolescents with chronic illness have been reported by Lawrence et al. (2007) and Sheerin et al. (1995).

Kind of illness. With regard to social consequences, Latner and Stunkard (2003) found that obese children were more likely to be stigmatized than children in a wheel chair and children with facial disfigurement. Thus, it could be expected that obese children would be most dissatisfied with their appearance. Elevated levels of body dissatisfaction could also be expected for other disfiguring conditions (e.g., burns, cerebral palsy, cleft lip; e.g., Lawrence et al., 2007) and in cases where side effects of therapy affect appearance (e.g., cancer chemotherapy, treatment of arthritis and asthma with steroids). For example, a qualitative study by Wallace, Harcourt, Rumsey, and Foot (2007) found that concerns around an altered appearance were widespread during cancer treatment.

Age at onset/duration of illness. It has been suggested that an earlier onset and a longer duration of the disease allow more time for coping with the disease and psychological adaptation (Newell, 2000). Thus, differences between the BI of children with and without chronic illness may decline with earlier onset and longer duration of the illness.

Age. Adolescence is an important and sometimes vulnerable period in the development of BI because the physical and psychological changes associated with adolescence increase the focus on physical appearance (Wertheim & Paxton, 2011). Thus, negative effects of (visible) diseases on BI may be stronger in adolescence than in childhood because of negative reactions by peers in regard to physical attractiveness and their suitability as a friend or mate (Rumsey & Harcourt, 2011).

Gender. Girls appear to be more susceptible to body dissatisfaction than boys (Wertheim & Paxton, 2011). This might cause stronger negative effects of chronic diseases on the BI of girls as the result of greater emphasis placed on female appearance by society. In line with this, Meissner (1997) found that the greatest impact on body image occurred in girls with physical problems of high impact and visibility.

Country. Scarce resources put effective treatments of chronic illness beyond the reach of many children and adolescents from developing countries (e.g., surgical treatment of cleft lip; Mossey, Little, Munger, & Dixon, 2009). This may cause stronger negative effects of chronic illness on BI and mental health in these countries. In fact, a recent meta-analysis found stronger differences between self-esteem of children with and without chronic illness in developing/threshold countries than in industrialized countries (Pinquart, in press). The present study tests whether similar differences are found with regard to BI.

Publication status. As nonsignificant results may be less likely to be published than significant results (the so-called “file-drawer problem”; Rosenthal, 1979), the present study tests whether this is also the case in studies on BI.

Year of publication/presentation. Latner and Stunkard (2003) found that the stigmatization of children with obesity and impaired mobility by other children was stronger in the early 2000s than in the early 1960s, probably indicating an increased importance of appearance. This may cause more adverse aversive effects of chronic illness on BI in more recent studies.

Target of comparison. Psychological scales have often been normed on school-based samples (e.g., Harter, 1985). Because about 15% of the students have a chronic health condition (van der Lee et al., 2007) it is very likely that some members of the norm population have a chronic illness. Thus, between-group differences may be weaker in studies that use test norms rather than a healthy control group as a basis for comparison.

Exploratory analysis of other moderators. Exploratory tests were computed with regard to moderating effects of method of assessing BI, race/ethnicity, representativeness of the sample (community-based probability samples versus convenience samples of patients attending a hospital), and sociodemographic matching of the patient and control group because no specific expectations could be stated. The goal of these tests was to identify additional sources of heterogeneity.