Anorectal malformations (ARM) and Hirschsprung disease (HD) are congenital, chronic bowel diseases. Children and adolescents with ARM or HD must cope with substantial functional problems, such as severe constipation and fecal incontinence, which affect quality of life (1–4). The ultimate goal of treatment has moved from mere survival to alleviating symptoms and to improving quality of life (QL).
Although definitions of QL vary widely, there is consensus about 2 central aspects. First, QL should be assessed from the patient's perspective whenever possible (5). Second, QL should be regarded as a multidimensional construct incorporating at least 3 broad domains that can be affected by one's disease or treatment, including physical, mental, and social functioning (6,7). QL of children only recently was established as an endpoint in paediatric research (8). Many researchers and clinicians have long been reluctant to include children's self-assessments into QL studies because they thought the results would be unreliable. Because a child's vocabulary, language, and perception of health and illness are still developing, it is indeed difficult to measure QL in children. The reading and developmental levels of the child determine whether the patient is old enough to complete the instrument on his or her own. Most studies agree that children from the age of 8 are able to provide reliable QL data (9).
Most studies examining the QL of patients focus on separate disease categories. However, it is not clear whether it is appropriate to classify the QL of patients by specific defects or by the broader effects of illness on the patient (10). For instance, patients with ARM and HD have comparable disease-specific problems. Besides, patients with both ARM and HD are born with a chronic disease, and both patient groups have had surgical correction in early childhood. This means that patients with ARM or HD should have much in common with respect to QL.
To provide tailored care, knowledge is needed about the factors affecting patients' QL, and the relations between those factors and QL. Some attempts have been made to clarify the relations among factors affecting QL (11,12), and a few studies have tested theoretical QL models on a patient population (13–16). To study which factors explain the QL of children and adolescents with ARM or HD, we have used an explanatory model, based on the Wilson and Cleary model (17) (Fig. 1). We hypothesized that QL is affected by background characteristics, such as clinical factors (eg, disease severity) and demographic factors (eg, age), but that these effects are mediated by the functioning of the patient, such as global disease-specific functioning (eg, fecal incontinence) and perceived self-competence (eg, school attitude). This means that these background characteristics do not affect QL directly, but do so through the effects of global disease-specific functioning and the level of perceived self-competence. The model posits that the disease (ARM/HD) and its severity (mild/severe) precede the perception of the disease (global disease-specific functioning) and coping with life (perceived self-competence), and that global disease-specific functioning and perceived self-competence precede patients' QL.
In the present study, first we compared QL and perceived self-competence between children (8–11 years) and adolescents (12–16 years) with ARM or HD, with those of the reference groups. Then, we tested whether patients' QL was explained by the clinical and demographic background characteristics via effects of global disease-specific functioning and perceived self-competence.
PATIENTS AND METHODS
In December 1998 all 722 known patients between 8 to 16 years old with ARM or HD were traced from the 6 Dutch paediatric surgical centers and from the ARM and HD patient societies. A total of 231 (32%) patients did not participate in the study for reasons including lack of basic proficiency in Dutch, mental retardation and/or Down syndrome, or having a cloaca. In addition, some of these patients were untraceable or had died. A set of questionnaires was sent to 491 patients (276 with ARM, 215 with HD), which was completed and returned by 316 (64%) of them (164 with ARM, 59%; 152 with HD, 71%). Informed consent to retrieve data also was obtained through the questionnaire. The medical ethics committees of all 6 paediatric surgical centers approved the study.
Most instruments in QL research can be classified as either generic or disease-specific. The generic instruments are designed to measure all aspects of health and well-being regardless of the underlying disease, and enable comparisons across different disease groups and with healthy reference groups. However, the generic instruments are limited in that they do not measure aspects that are of particular relevance to specific disease groups, such as disease symptoms. Conversely, disease-specific measures examine the symptoms of specific disease groups and how they function (18,19). As is common in QL research, we decided to include both types of questionnaires.
Quality of Life
QL of patients with ARM or HD was measured with the Netherlands Organization for Applied Scientific Research Academic Medical Center (TNO-AZL) Child Quality of Life Questionnaire (TACQOL) (20,21). The TACQOL is a generic Dutch instrument that has been shown to be a reliable and valid instrument for measuring QL in children and adolescents between the ages of 8 and 15 (20–23). The TACQOL was selected for this study because it measures not only health status problems but also the perception of the health status problem. It offers the respondents the possibility to differentiate between their functioning and the way they feel about it. We used 2 different versions of the TACQOL, a children's form (8–11 years) (20) and an adolescent's form (12–16 years) (21). Both versions consist of 56 identical items, only slightly differing in wording, making up 7 scales of QL. Five health-related functioning scales include pain and symptoms (Cronbach α in our sample is 0.73), basic motor functioning (α = 0.81), autonomy (α = 0.70), cognitive functioning (α = 0.82), and interaction with parents and peers (α = 0.70). Two scales represent experience of positive emotions (α = 0.82) and experience of negative emotions (α = 0.70). The respondent could rate whether a specific problem had occurred in the past few weeks, with 3 response options (“never,” “occasionally,” and “often”) for each item of the health-related functioning scales. If a problem occurred, then the respondent was asked how he or she felt about this problem: “fine,” “not so good,” “quite bad,” and “bad.” For each item, both responses (functioning and how they feel about it) were combined into a single item score ranging from 0 to 4 (never = 4; occasionally or often combined with fine = 3, with not so good = 2, with quite bad = 1, and with bad = 0). With the emotion scales, respondents indicated on a Likert scale whether the feeling presented had occurred in the past few weeks (never, occasionally, often). Item scores for the 2 emotion scales ranged from 0 to 2. With all 7 scales, higher scores correspond to better QL. The most recent manual (21) advised omitting the autonomy scale and replacing the original interaction with parents and peers scale with a new peers-only scale because of poor reliability. In our study, the autonomy scale did indeed appear to be insufficiently reliable for some of the subgroups, with some Cronbach α about 0.50. However, the interaction with parents and peers scale appeared more reliable (Cronbach α = 0.67–0.88) than the peers scale (Cronbach α = 0.55–0.96) for different disease groups and different age groups. Therefore, we decided to remove the autonomy scale as advised but to keep the interaction with parents and peers scale.
To reduce the number of outcome variables for the analysis for explaining QL, we used the principal component analysis with all of the items to summarize TACQOL scale scores by 2 component scores, interpreted as physical QL (pain and symptoms, basic motor functioning, cognition) and mental QL (interaction with parents and peers, experience of positive and negative emotions, cognition). We decided to use oblique rotation, befitting the idea that resulting components are correlated. Indeed, after rotation, it appeared that the resulting physical and mental components have a clear interpretation. Apparently, cognitive functioning is indicative for both physical and mental functioning.
Disease-specific functioning was measured with the Hirschsprung disease/Anorectal malformation Quality of Life questionnaire (HAQL) (24). The HAQL questionnaire was developed by our research group for patients with ARM or HD. We used 38 identical items for children and adolescents, which only slightly differed in wording. The item scores range from 1 to 4 (“never” = 1, “sometimes” = 2, “often” = 3, and “very often” = 4). Scores were recoded so that higher scores indicate a higher level of functioning. The 38 items of the HAQL were combined to form 9 subscales. Almost half of the disease-specific subscales appeared to be unreliable (Cronbach α ≤0.70) because of lack of variance (24,25). We only selected those scales with reasonable score distribution, which were presence of diarrhoea (α = 0.68), faecal continence (α = 0.83), physical symptoms (α = 0.69), emotional functioning (α = 0.79), and body image (α = 0.77). We summed all 5 of these HAQL scales item scores into a single global disease-specific functioning component covering all of the aspects that are likely to be affected by ARM or HD (ie, fecal incontinence, constipation, and disease-specific psychosocial functioning), accounting for more than 40% of all variance. Cronbach α ranged from 0.79 to 0.91 for different disease groups and different age groups. Moreover, this single factor appeared to be substantially correlated with the other research variables in the study, indicating sufficient validity.
A Dutch translation (26,27) of the Harter Self-Perception Profile for Children and Adolescents (SPPC and SPPA, respectively) (28,29) has been shown to be a reliable and valid instrument to assess children's and adolescents' perceptions of competence (26–30). We used 36 identical items for children and adolescents, only slightly differing in wording, which were formulated as pairs of opposite assertions; children and adolescents first determine which of 2 statements is more like them. For example, “other children/adolescents would like to look different” with 2 response options: “completely true” or “hardly true.” The opposite assertion is formulated as “some children/adolescents are satisfied with their appearance,” with the same response options. Each answer was scored between 1 = most competent (completely true) and 4 = least competent (completely true for the opposite pair). Scores, ranging from 1 to 4, were coded so that higher scores indicated a higher level of functioning. The 36 identical items made up 6 scales of perceived competence in the following areas: social acceptance (α = 0.81), behavioural conduct (α = 0.74), scholastic competence (α = 0.79), athletic competence (α = 0.87), physical appearance (α = 0.78), and global self-worth (α = 0.78). Through principal component analysis, we summarized the 6 competence scales into 3 components: self-esteem (physical appearance, global self-worth), athletic competence (athletic competence, social acceptance), and school attitude (scholastic competence, behavioural conduct, social acceptance). Social acceptance appeared to be indicative of both athletic competence and school attitude.
Clinical variables were extracted from medical records and included disease severity (mild vs severe), presence of additional congenital anomalies, presence of a permanent stoma, and surgery. For ARM, mild versus severe referred to low defects (bucket handle, covered anus, anterior displaced anus, perineal or vestibular fistula) versus more complex defects (urethral, vesical, or vaginal fistula, or no fistula). In addition, with ARM we frequently observed the VACTERL association (the mnemonic for vertebral, anorectal, cardial, tracheo-esophageal, renal, and limb defects) (31). For HD, mild disease was defined as aganglionosis of a common (or usual) segment (rectum or sigmoid colon), and severe disease referred to aganglionosis of a larger segment (descending colon, transverse colon, ascending colon, or ileum). Demographic characteristics included age and sex, which were assessed with 2 additional items in the questionnaire.
Comparisons With Reference Groups
The TACQOL mean scores of the patient groups were compared with reference values from a Dutch general population, children ages 8 to 11 (n = 878) and adolescents ages 12–15 (n = 931), as reported by Vogels et al (20,21). Mean scores of the SPPC/SPPA of the patient groups also were compared with reference values from a Dutch general population of children ages 8 to 12 (n = 361) and adolescents ages 12 to 18 (n = 1386), as reported by Veerman et al (26) and Treffers et al (27). TACQOL and SPPC/SPPA reference groups both consist of 50% boys versus 50% girls, whereas in our samples sex was distributed as 65% boys versus 35% girls for children with ARM, 69% boys versus 31% girls for adolescents with ARM, 78% boys versus 22% girls for children with HD, and 82% boys versus 18% girls for adolescents with HD (Table 1). Therefore, we weighted our samples according to the 50/50 sex distribution of the reference groups.
Explaining Patients' QL
To check whether the different samples of children and adolescents with ARM or HD could be combined for the analysis of QL, first we tested whether the relations between the research variables for these groups were similar or not by comparing the correlation matrices using structural equation modelling (32). Subsequently, we tested whether the explanatory model fitted the data by examining whether the covariance structure that follows from the proposed model fitted to the observed covariances. The fit was evaluated through a chi-square test and through the root mean square error of approximation (RMSEA). According to a generally accepted rule of thumb (33), RMSEA values lower than 0.08 indicate reasonable fit, and values lower than 0.05 indicate close fit. In addition to overall goodness of fit, component fit was evaluated by inspecting standardized discrepancies (between observed and expected correlations) and the computer program modification indices of LISREL (34). Relations between background characteristics, global disease-specific functioning, perceived self-competence, and QL were expressed as standardized regression coefficients and tested through Wald tests (32). In our study, 0.1, 0.3, and 0.5 can be considered as small, medium, and large standardized regression coefficients (35).
The total sample consisted of 316 patients, including 164 patients with ARM (92 children and 72 adolescents) and 152 patients with HD (76 children and 76 adolescents). Patient characteristics for disease groups (ARM/HD) and age groups (children/adolescents) are given in Table 1.
Comparing Patients With Reference Groups
Means and standard deviations of the QL scales and the perceived self-competence scales are depicted in Table 2. Although on average no QL differences were found between children and adolescents with ARM or HD, standard deviations revealed considerable variation between patients. Children in both patient groups scored significantly lower on all perceived competency scales compared with the reference groups (P < 0.001 for almost all of the subscales). The perceived competency comparisons between the adolescents of both ARM or HD and the reference groups also resulted in significant differences in the scores of the subscale physical appearance, with the patients doing better (P < 0.01 and P < 0.001, respectively). Also, adolescents with ARM scored better on social acceptance and athletic competencies (P < 0.05), and adolescents with HD scored better on global self-worth (P = 0.01).
Explaining Patients' QL
Structural equation modelling showed that the correlation matrices of the age groups (RMSEA = 0.072, 90% confidence interval [CI] 0.046–0.096) and the 2 disease groups (RMSEA = 0.034, 90% CI 0.000–0.110) were essentially equivalent, so we combined age and disease groups for the analysis for explaining QL with the additional advantage of a larger sample size (increased power and higher precision). Despite equivalent relations between research variables, patients with different diseases (ARM or HD) and patients with different ages (children or adolescents) may have different levels of QL, global disease-specific functioning, and perceived self-competence. To control for effects of type of disease or age, we included type of disease (ARM/HD) and age among the predictor variables.
The explanatory model described in Figure 1 was fitted to the correlation matrix. The chi-square distribution with x degrees of freedom, or CHISQ (x), measure of overall goodness-of-fit was 25.42 (CHISQ (10), P = 0.00), and the hypothesis of exact fit was rejected. The RMSEA was 0.069, and the 90% CI ranged from 0.035 to 0.10, which was considered a satisfactory fit. However, inspection of the component fit suggested 2 additional direct effects of disease severity on physical QL and of age on mental QL. The addition of these direct effects resulted in a modified model with good fit (CHISQ (8) = 10.12; P = 0.26; RMSEA = 0.028; 90% CI 0.00–0.075). Table 3 presents the standardized regression coefficients of the modified model.
Of the background characteristics, female sex had a negative effect on athletic competence (β = −0.27), older age negatively affected self-esteem (β = −0.29), athletic competencies (β = −0.17), and school attitude (β = −0.16). A severe form of the disease had a negative effect on global disease-specific functioning (β = −0.16) (Table 3). In addition, older age and a severe disease form had direct positive effects on mental QL (β = 0.10) and physical QL (β = 0.14), respectively (Table 3). In turn, almost all of the mediating variables had significant effects on QL; that is, global disease-specific functioning positively affected physical QL (β = 0.26) and mental QL (β = 0.16), self-esteem positively affected mental QL (β = 0.21), athletic competencies positively affected physical QL (β = 0.19) but negatively affected mental QL (β = −0.09), and school attitude positively affected physical QL (β = 0.36) and mental QL (β = 0.49). The last section of Table 3 shows that the modified model explained 39% of variance in physical QL and 46% in mental QL (Table 3).
The 2 patient groups were combined a priori in 1 study because both patients with ARM and HD have to deal with similar physical problems. These similarities imply considerable commonality in the impact of these diseases on patients' daily functioning and QL. To examine whether patients with ARM or HD indeed reported similar QL problems, we compared them separately to the reference groups. On average, children and adolescents in both patient groups did not report impaired QL as compared with healthy peers. However, we were interested not only in the group as a whole but also in individual variation. It appeared that within our groups, some of the ARM or HD patients reported high levels of QL, whereas others indicated low levels of QL.
From other studies, it appeared that the age of ARM patients has direct influence on the impact of QL (36,37). Two crucial stages have been identified, before the commencement of primary education and immediately before adolescence (37). Moreover, because developmental stages require their own adaptation, children and adolescents may report different levels of QL, which may call for age-related care. The QL and perceived self-competencies of the children and adolescents cannot be measured with exactly the same instruments, rendering comparisons of the child populations on the one hand and the adolescent populations on the other hand impossible. Because patients in different age groups necessarily need age-specific questionnaires, problems with comparisons are inherent to examining children and adolescents in 1 study. Consequently we used 2 age-related versions of the TACQOL and the SPPC/SPPA, for children (20,26) and adolescents (21,27), respectively. Because the child and the adolescent versions of the used questionnaires contain overlapping items, which only slightly differ in wording, we were able to form comparable subscales. Therefore, we were allowed to compare levels of QL domains of children with ARM or HD with those of adolescents with ARM or HD to see whether they differ. We found that adolescents' QL and perceived self-competence was poorer than that of children, but this was also true for healthy adolescents (data not shown). We also compared the QL and perceived competence of children with ARM or HD and that of adolescents with ARM or HD separately to the reference groups. Remarkably, compared with the reference groups, we found that adolescents performed better than children did. Because level of QL and perceived self-competence varies between age groups also in healthy populations, it can be argued that healthy peers provide the best standard for comparisons because they provide more valid information regarding patients' QL and perceived self-competence. Therefore, we only showed the results of the comparisons with healthy peers, and concluded that adolescents performed better than children did.
More specifically, children with ARM and HD scored lower on almost all of the perceived competency scales. Conversely, adolescents with ARM or HD perceived some of their competencies as even better when compared with healthy peers (social acceptance, athletic competencies, physical appearance, and global self-worth). It may be that compared with healthy adolescents, adolescents with ARM or HD have developed stronger psychosocial competency and coping skills, probably because they have had to learn to live with chronic functional problems. An alternative explanation is that adolescents have a strong wish to be “normal” and therefore completed the questionnaires with as positive of an outlook as possible (ie, in a socially desirable way).
To provide tailored care for the specific needs of children and adolescents with ARM or HD, knowledge about the factors that affect QL is needed. Therefore, our second question addressed the factors associated with QL. Because relations between the research variables between the 2 disease groups appeared to be equivalent, combining ARM and HD patients in the analysis to explain QL was justified. Still, despite equivalent relations between research variables, ARM patients and HD patients may have different levels of QL. To investigate and check for such differences, we included type of disease (ARM/HD) as a predictor variable in the analyses for explaining QL.
We found that the observational data fitted the explanatory model satisfactorily, indicating that effects of background characteristics on QL are indeed mediated by global disease-specific functioning and psychosocial competence. This means that the background characteristics do not affect QL directly, but do so through the effects of global disease-specific function and perceived self-competence. The model explained 39% to 46% of physical and mental QL, respectively, of children and adolescents with ARM or HD.
Effects of background characteristics on global disease-specific functioning and perceived self-competence showed that those with female sex, older age, or more severity of the disease reported lower levels of perceived self-competence and global disease-specific functioning. Notice that the classification of disease severity is arbitrary. We chose to classify the severity of ARM and HD by the length of the affected part of the bowel because longer affected segments (or with ARM, more complex forms) are usually associated with worse physical outcome/more symptoms than short (or usual) segments of the bowel. It appeared that patients with a severe form of the disease reported more chronic defecation problems than patients with a mild form. Furthermore, the results of our longitudinal study with children and adolescents with ARM or HD (38) show that the severely affected patients did not improve in global disease-specific functioning, whereas the less severely affected patients improved. The direct relation between disease severity and (change in) global disease-specific functioning confirmed that our classification of disease severity was justified. However, in future research it would be useful to include more objective parameters on faecal functioning, such as anorectal manometer or dynamic rectal imaging to enable further disentangling of the interesting relations between objective and subjective parameters of faecal continence problems and QL.
Effects of global disease-specific functioning and perceived self-competence on QL showed that the strongest effects were found for perceived self-competence on QL, in particular for school attitude. School attitude refers to school performance, behavioural conduct, and social acceptance. Results showed that global disease-specific functioning also explained the QL of these children and adolescents. In contrast, the results in our studies with the adult populations showed that constipation and faecal incontinence had almost no influence on QL (39) or changes in QL (40), whereas the level of self-esteem and perceived self-competence (eg, body image, sexual interest, emotional functioning) were the most important factors affecting QL and changes in QL. The results of the adult populations indicate that neither fewer symptoms nor an improvement in global disease-specific functioning automatically imply better QL or an improvement in all QL aspects. However, because perceived self-competence and global disease-specific functioning both appear to be consistent predictors of QL of children and adolescents with ARM or HD, we can conclude that increasing perceived self-competence complementary to alleviating symptoms may improve QL for those patients with ARM or HD who are in need of extra care.
In the proposed model, we took into consideration that relations between variables may be complex, and we specified not only causal effects between background characteristics and mediating factors, and between mediating factors and outcome factors, but also correlations among the variables within the background characteristics, the mediating factors, and the outcome factors. This way possible confusion between variables was accounted for. With LISREL, causal models with causal effects are fitted to the data. However, if a model fits the data, the effect in the model is causal, but whether the relation is actually causal and in which direction can only be examined in an experiment with random assignments to several conditions. Nevertheless, we believe in the causality of the model. We believe that QL is affected by the global perception of the disease and by perceived self-competence. Most of all, we believe in the strength of the Wilson and Cleary model in practical terms. After all, disease severity cannot be treated directly, and neither can QL. Ideally, to decrease the negative impact of the disease and to improve patients' QL, treatment of children and adolescents should be directed at the mediating factors, that is, at both alleviating symptoms (faecal incontinence and constipation) and improving perceived self-competence (eg, feeling of embarrassment, lack of positive self-esteem, school problems). For instance, global disease-specific functioning may be improved by medical treatment, which could include surgical interventions, anal dilatations, prescription of various laxatives and enemas, necessary information about diet and toilet habits, medication, and provision of incontinence material (41–43). To improve psychosocial functioning, medical treatment should be combined with paramedical treatment such as physiotherapy or dietary advice, and psychosocial treatment such as psychotherapy or counselling. By directing treatment at both global disease-specific and psychosocial functioning, it may be expected that patients with ARM or HD improve their QL. Our results also are in concordance with other studies that show that nonmedical treatment is considered a useful supplement to the standard medical treatment (42,43). Peňa and Hong (44) found that 95% of their patients improved their QL dramatically and remained clean most of the time by implementing a well-instructed bowel management program, showing that a combination of medical and nonmedical care can be successful. Therefore, we recommend that health care providers who treat patients with ARM or HD should be aware of psychosocial problems and should therefore inquire not only about patients' physical well-being but also about patients' emotional and social well-being.
A limitation of the study was that the response rate was lower for ARM patients (59%) than for HD patients (71%), which may be because of the simultaneous study of Poley et al (1). Perhaps the overlapping inclusion of patients resulted in loss of patients. However, our sample size is still sufficiently large (n = 316). For reasons of confidentiality, we were not able to assess the characteristics of nonresponding patients from the first measurement, and to what extent selection bias may have played a role was not easy to determine. Only 64% of the approached patients returned the questionnaire, which may constitute a selection bias. However, the sex distribution, male–female ratio in relation to severity of the disease, and the incidence of additional congenital diseases was similar to those of other studies, which suggest a representative sample (45–47). The only exception was the percentage of adolescent ARM patients with congenital anomalies (38% vs 50%), giving an underrepresentation of adolescents with ARM with additional congenital anomalies.
Recommendations for Treatment
First, health care providers should be especially alert to children and adolescents who are female, with a severe form of the disease or with additional congenital diseases, because they reported lower levels of perceived self-competence and global disease-specific functioning. Furthermore, because the results of the present study showed that the QL of these patients is highly dependent on perceived self-competence, nonmedical assistance to increase self-competence would be a valuable addition to the standard medical care. Because clinical practice normally is directed at minimizing symptoms and not at improving perceived self-competence, increasing awareness of the nonmedical problems and applying psychosocial treatment is an important addition to treatment for paediatric surgeons and other health care providers because it may be expected that patients with ARM and HD will achieve better control of their stool problems and can improve their QL. Ideally, there should be a screening tool to make clinicians aware of individual physical and psychosocial problems, preferably in childhood. The Dutch proverb “learnt young is done old” reinforces the maxim that if you learn something at an early age, you become quite good at it when you are older. This bit of folk wisdom has been shared by other cultures—similar English proverbs include “learn young, learn fair” and “what's learnt in the cradle lasts till the tomb”—and we suggest that the results of our study may have similar adaptability.
1. Poley MJ, Stolk EA, Tibboel D, et al. Short term and long term health related quality of life after congenital anorectal malformations and congenital diaphragmatic hernia. Arch Dis Child 2004; 89:836–841.
2. Diseth TH, Emblem R. Somatic function, mental health, and psychosocial adjustment of adolescents with anorectal anomalies. J Pediatr Surg 1996; 31:638–643.
3. Goyal A, Williams JM, Kenny SE, et al. Functional outcome and quality of life in anorectal malformations. J Pediatr Surg 2006; 41:318–322.
4. Athanasakos E, Starling J, Ross F, et al. An example of psychological adjustment in chronic illness: Hirschsprung's disease. Pediatr Surg Int 2006; 22:319–325.
5. Eiser C, Mohay H, Morse R. The measurement of quality of life in young children. Child Care Health Dev 2000; 26:401–414.
6. Aaronson NK, Meyerowitz BE, Bard M, et al. Quality of life research in oncology—past achievements and future priorities. Cancer 1991; 67:339–343.
7. World Health Organization. Preamble to the Constitution of the World Health Organization as Adopted by the International Health Conference. Geneva: World Health Organization; 1948.
8. Ravens-Sieberer U, Bullinger M. Assessing health-related quality of life in chronically ill children with the German KINDL: first psychometric and content analytical results. Qual Life Res 1998; 7:399–407.
9. Eiser C, Morse R. A review of measures of quality of life for children with chronic illness. Arch Dis Child 2001; 84:205–211.
10. Stein REK, Jessop DJ. What diagnosis does not tell—the case for a noncategorical approach to chronic illness in childhood. Soc Sci Med 1989; 29:769–778.
11. Ong NT, Beasley SW. Long-term continence in patients with high and intermediate anorectal anomalies treated by sacroperineal (Stephens) rectoplasty. J Pediatr Surg 1991; 26:44–48.
12. Ludman L, Spitz L, Kiely EM. Social and emotional impact of fecal incontinence after surgery for anorectal abnormalities. Arch Dis Child 1994; 71:194–200.
13. Smith KW, Avis NE, Assmann SF. Distinguishing between quality of life and health status in quality of life research: a meta-analysis. Qual Life Res 1999; 8:447–459.
14. Amir M, Roziner I, Knoll A, et al. Self-efficacy and social support as mediators in the relation between disease severity and quality of life in patients with epilepsy. Epilepsia 1999; 40:216–224.
15. Rosen RC, Contrada RJ, Gorkin L, et al. Determinants of perceived health in patients with left ventricular dysfunction: a structural modeling analysis. Psychosomat Med 1997; 59:193–200.
16. Sullivan MD, Kempen GI, Van Sonderen E, et al. Models of health-related quality of life in a population of community-dwelling Dutch elderly. Qual Life Res 2000; 9:801–810.
17. Wilson IB, Cleary PD. Linking clinical variables with health-related quality of life. A conceptual model of patient outcomes. JAMA 1995; 273:59–65.
18. Testa MA, Simonson DC. Assessment of quality of life outcomes. N Engl J Med 1996; 334:835–840.
19. Patrick DL, Deyo RA. Generic and disease-specific measures in assessing health status and QL. Med Care 1989; 27:s217–s232.
20. Vogels T, Verrips GHW, Koopman HM, et al. TACQOL Manual Parent and Child Form. Leiden, the Netherlands: Leiden Center for Child Health and Paediatrics; 2000.
21. Vogels T, Bruil J, Koopman HM, et al. TACQOL CF 12-15 Manual. Leiden, the Netherlands: Leiden Center for Child Health and Paediatrics; 2004.
22. Vogels T, Verrips GH, Verloove-Vanhorick SP, et al. Measuring health-related quality of life in children: the development of the TACQOL parent form. Qual Life Res 1998; 7:457–466.
23. Verrips GHW, Vogels TGC, Verloove-Vanhorick SP, et al. Health related quality of life measure for children—the TACQOL. J Appl Ther 1998:357–360.
24. Hanneman MJG, Sprangers MA, De Mik EL, et al. Quality of life in patients with anorectal malformations or Hirschsprung's disease: the development of a disease-specific questionnaire. Dis Colon Rectum 2001; 44:1650–1660.
25. Trajanovska M, Catto-Smith AG. Quality of life measures for fecal incontinence and their use in children. J Gastroenterol Hepatol 2005; 20:919–928.
26. Veerman JW, Straathof MAE, Treffers DA, et al. Manual of the Self-perception Scale for Children [in Dutch]. Lisse, the Netherlands: Swets & Zeitlinger; 1997.
27. Treffers DA, Goedhart AW, Veerman JW, et al. Manual of the Self-perception Scale for Adolescents [in Dutch]. Lisse, the Netherlands: Swets & Zeitlinger; 2002.
28. Harter S. Manual for the Self-perception Profile of Children. Denver: University of Denver; 1985.
29. Harter S. Manual for the Self-perception Profile of Adolescents. Denver: University of Denver; 1986.
30. Aasland A, Diseth TH. Can the Harter Self-Perception Profile for Adolescents (SPPA) be used as an indicator of psychosocial outcome in adolescents with chronic physical disorders? Eur Child Adolesc Psychiatry 1999; 8:78–85.
31. Hassink EA, Rieu PN, Hamel BC, et al. Additional congenital defects in anorectal malformations. Eur J Pediatr 1996; 155:477–482.
32. Bollen KA. Structural Equations with Latent Variables. New York: Wiley; 1989.
33. Browne MW, Cudeck R. Alternative ways of assessing model fit. Sociol Methods Res 1992; 21:230–258.
34. Joreskog KG, Sorbom D. LISREL 8 User's Guide. Chicago: Scientific Software International; 1996.
35. Cohen J. Statistical Power Analysis for the Behavioral Sciences. Mahwah, NJ: Lawrence Erlbaum Associates; 1988.
36. Bai Y, Yuan Z, Wang W, et al. Quality of life of children with fecal incontinence after surgically corrected anorectal malformations. J Pediatr Surg 2000; 35:462–464.
37. Ditesheim JA, Templeton JM Jr. Short-term versus long-term quality of life in children following repair of high imperforate anus. J Pediatr Surg 1987; 22:581–587.
38. Hartman EE, Oort FJ, Aronson DC, et al. Explaining change in quality of life of children and adolescents with anorectal malformations or Hirschsprung disease. Pediatrics 2007; 119:e374–e383.
39. Hartman EE, Oort FJ, Aronson DC, et al. Critical factors affecting quality of life of adult patients with anorectal malformations or Hirschsprung's disease. Am J Gastroenterol 2004; 99:907–913.
40. Hartman EE, Oort FJ, Visser MR, et al. Explaining change over time in quality of life of adult patients with anorectal malformations or Hirschsprung's disease. Dis Colon Rectum 2006; 49:96–103.
41. Loening-Baucke V. Fecal incontinence in children. Am Fam Physician 1997; 55:2229–2238.
42. Van Kuyk EM, Brugman-Boezeman AT, Wissink-Essink M, et al. Defecation problems in children with Hirschsprung's disease: a prospective controlled study of a multidisciplinary behavioral treatment. Acta Paediatr 2001; 90:1153–1159.
43. Van Kuyk EM, Wissink-Essink M, Brugman-Boezeman AT, et al. Multidisciplinary behavioral treatment of defecation problems: a controlled study in children with anorectal malformations. J Pediatr Surg 2001; 36:1350–1356.
44. Pena A, Hong A. Advances in the management of anorectal malformations. Am J Surg 2000; 180:370–376.
45. Kiely EM, Pena A. Anorectal malformations. In: O'Neill JA, editor. Pediatric Surgery. 5th ed St Louis: Mosby-Year Book; 1986. pp. 1425–1448.
46. Zaslavsky C, Loening-Baucke V. Anorectal manometric evaluation of children and adolescents postsurgery for Hirschsprung's disease. J Pediatr Surg 2003; 38:191–195.
47. Teitelbaum DH, Coran AG, Weitzman JJ, et al. Hirschsprung's disease and related neuromuscular disorders of the intestine. In: O'Neill JA, Rowe MI, Grosfield JL, et al., editors. Pediatric Surgery. 5th ed St Louis: Mosby-Year Book; 1998. pp. 1381–1424.
© 2008 Lippincott Williams & Wilkins, Inc.