Behavioral Functioning and Treatment Adherence in Pediatric Inflammatory Bowel Disease

Abstract

The purpose of this article is to review the existing research pertaining to behavioral functioning and treatment adherence in children and adolescents with inflammatory bowel disease (IBD), discuss potential effects of behavioral dysfunction on adherence, and provide clinical recommendations for assessment and treatment options. An updated literature review in pediatric IBD is presented, and research across chronic conditions supports the plausibility of negative consequences of patient and family behavioral problems on disease management. Clinical recommendations include use of multimethod assessment of treatment adherence, routine screening for behavioral difficulties, patient-provider discussion of behavioral issues and disease management during clinic visits, and increased attention to the process of gradually transitioning responsibility of disease management from parents to adolescents as patients approach adulthood.

Inflammatory bowel disease (IBD) affects 71 in 100,000 children in the United States (43 per 100,000 children with Crohn's disease and 28 per 100,000 children with ulcerative colitis),¹ and approximately 25% of all individuals with IBD are diagnosed during childhood or adolescence. The chronic, intermittent, and unpredictable symptoms of IBD (eg, diarrhea, weight loss, fatigue, delayed puberty, growth failure, abdominal pain), as well as undesirable side effects of some medications, including cushingoid appearance, weight gain, pancreatitis, and increased risk of cancer, can negatively impact cognitive and emotional adjustment to IBD. Indeed, children and adolescents with IBD are at an increased risk for significant psychosocial maladjustment.^2–4 In addition, the treatment regimens of these patients are often complex, involving multiple medications with varying dosing schedules and pill quantities, dietary modifications, infusions, clinic visits, and surgery. These regimen and psychosocial factors likely combine to make treatment adherence challenging for pediatric patients and their families.⁵ Several studies have been published on behavioral and family functioning in pediatric IBD, but few studies have focused directly on treatment adherence in IBD. Moreover, although the relationship between behavioral dysfunction and treatment adherence has been examined in other pediatric chronic illness populations, it has been neglected in pediatric IBD. The primary aim of this article is to provide an updated review of the literature on behavioral functioning and treatment adherence in children and adolescents with IBD, discuss the plausible effects of behavioral dysfunction on adherence using studies in representative pediatric populations as models, and provide clinical recommendations for assessment and treatment options for patients who demonstrate behavioral dysfunction and/or treatment nonadherence in IBD.

Behavioral Functioning in Children and Adolescents With Inflammatory Bowel Disease

Comprehensive reviews of psychosocial and family functioning in pediatric IBD patients have been recently provided by Mackner and colleagues.^4,6,7 The following summarizes the pertinent research on behavioral functioning in IBD.

Internalizing disorders such as depression and anxiety have been the focus of several studies in pediatric IBD patients. Lifetime prevalence and current prevalence of depressive disorders have been documented in 25% and 10%, respectively, of pediatric IBD patients, whereas lifetime and current prevalence of anxiety disorders have been observed in 11% and 4%, respectively.⁸ In a separate study, Burke and associates⁹ reported a prevalence of 14% for major depression and a prevalence of 28% for an anxiety disorder in recently diagnosed IBD patients. Higher rates of internalizing and overall behavior problems compared to healthy controls have also been observed by Engstrom^10–12 and Engstrom and Lindquist.¹³ Similarly, overall behavior problems and internalizing symptoms have been documented in IBD patients compared to a sibling comparison group.¹⁴ Externalizing disorder symptomatology (eg, aggressive behavior, oppositional behavior) has not been shown to be significantly different in IBD patients compared to healthy controls or other chronic illness groups^10–13; however, this issue is an understudied area of behavioral functioning in this population.

Recent well-controlled studies have provided additional information regarding behavioral factors in children and adolescents with IBD. Mackner and Crandall¹⁵ assessed psychological functioning in 50 adolescents with IBD 1 year after diagnosis and 42 healthy controls. Although most adolescents with IBD in this study reported normal behavioral functioning, approximately 20% exhibited clinically elevated scores on behavioral measures, which was consistent with the healthy control adolescents. Social functioning in this sample was also similar to the healthy control group. Another study² reported that approximately one third of adolescents with IBD demonstrated clinically elevated internalizing and overall behavioral problems. Moreover, these patients demonstrated increased depressive and anxiety symptoms, as well as greater social problems and poorer social competency compared to healthy controls.

Family functioning and parental emotional functioning have been examined in several studies. One study reported greater family dysfunction among the families of IBD patients compared to the families of diabetes patients or healthy controls.¹² Wood and coworkers¹⁶ reported greater family dysfunction in the families of Crohn's disease patients compared to the families of ulcerative colitis patients or patients with recurrent abdominal pain. In addition, Burke and colleagues¹⁷ reported that IBD patients who were depressed had families who were more disengaged and exhibited more conflict than nondepressed IBD patients. In contrast, Mackner and Crandall² found that family functioning among IBD patients was similar to that in a healthy adolescent control group. There is also some evidence that family functioning worsens with increased IBD disease severity.^16,18 Research examining parental distress and psychological dysfunction has reported increased maternal rates of lifetime and current depression diagnosis similar to those of mothers of cystic fibrosis patients,¹⁹ as well as increased psychiatric symptoms in general.^10,12 One study investigated siblings of IBD patients and found that siblings of Crohn's disease patients demonstrated elevated behavioral problems compared to siblings of ulcerative colitis patients and the normative sample for the measure.²⁰

The link between disease activity and behavioral factors is unclear. For example, Burke and associates⁹ found that depression was associated with a greater number of stressful life events yet better disease functioning. Additionally, one study reported that, although greater disease severity was correlated with obsessive-compulsive symptoms in children with Crohn's disease, an inverse correlation was observed in children with ulcerative colitis.²¹ A negative effect has been associated with subjective, but not objective, disease severity.²² Other studies have shown no relationship between behavioral factors and disease severity indices such as growth parameters and relapse rates.^1,2,14,15,23

Research has shown that children and adolescents with IBD have lower health-related quality of life (HRQOL) in areas such as emotional functioning, treatment concerns, body image, and somatic complaints.^24,25 The majority of HRQOL research in IBD has used the IMPACT questionnaire,^26,27 which is a disease-specific measure for use with IBD patients. Although prior research has produced mixed results with respect to the relationship between disease severity and HRQOL,^24,28 a recent study²⁹ using a large sample reported significantly poorer HRQOL as disease severity increased. The sensitivity of this measure was also demonstrated by improved HRQOL scores over the course of 1 year of treatment immediately following diagnosis.

There are several notable methodologic limitations to the research conducted on behavioral factors in pediatric IBD, particularly in some of the earlier studies. These limitations include modest sample sizes, a lack of appropriate control groups, inadequate documented training of interviewers or inadequate administration of assessments, and use of nonstandardized assessments for some behavioral domains.⁴ Nevertheless, there is evidence that children and adolescents with IBD are at an increased risk for behavioral and family dysfunction.

In summary, children and adolescents with IBD demonstrate increased behavioral difficulties, particularly internalizing disorders (eg, depression, anxiety disorders), social problems, and social competency. The families of pediatric IBD patients exhibit more conflict and disengagement, and their parents are more distressed than those of healthy controls. The relationships between disease severity and behavioral factors and disease severity and HRQOL show mixed results across studies, but emerging research indicates an inverse relationship. Research in adults with IBD has yielded similar findings regarding behavioral factors,^30,31 which highlight the importance of facilitating adaptive functioning in adolescents with IBD to minimize future behavioral and disease management difficulties.

Adherence in Children and Adolescents With Inflammatory Bowel Disease

There have been only a few published studies that have directly measured treatment adherence in pediatric IBD. One study examined medication adherence in adolescents with asthma, HIV, or IBD and found no differences in the prevalence of nonadherence (defined as <80% of medication taken) among the 3 groups; the prevalence of IBD nonadherence was 50%.³² In another study, Mackner and Crandall³³ reported medication nonadherence rates of 52% and 62%, according to adolescent and parent reports, respectively. This study also indicated that family dysfunction and poor patient coping were associated with medication nonadherence. Using pharmacy refill record data, Oliva-Hemker and coworkers³⁴ reported nonadherence rates of 50% for immunomodulators and 66% for mesalamine. Greater adherence in this study was related to greater healthcare utilization. Each of these studies, however, utilized a unimodal assessment of adherence (ie, self-report questionnaire, semistructured interview, or pharmacy refill records), which limits the conclusions and generalizability of the findings. A recent study by Hommel and colleagues³⁵ indicated that, although self-reported nonadherence was likely underestimated by adolescent patients at 7% for 6-mercaptopurine (6-MP)/azathioprine and 3% for 5-aminosalicylic acid (5-ASA) medications, the prevalence of pill count nonadherence was 38% for 6-MP/azathioprine and 48% for 5-ASA medications. Additionally, bioassay data indicated that approximately 80% of patients had subtherapeutic 6-thioguanine nucleotide (6-TGN) levels, though adequacy of dosing was not examined. Together, these studies provide initial data indicating that nonadherence is a significant behavioral health issue in the pediatric IBD population, with prevalence rates ranging from 38% to 66%.

Potential Effects of Behavioral Dysfunction on Adherence

Several studies in the pediatric literature have examined patient and family behavioral factors as correlates of treatment nonadherence. Studies across pediatric populations have demonstrated poorer adherence to long-term medical regimens in children and adolescents who exhibit behavioral or emotional dysfunction.^36–41 The same pattern has been shown in organ transplant patients who are children or adolescents.^42,43 Moreover, behavioral factors associated with nonadherence are a significant problem in adolescents,^42,44 placing them at a greater health risk than younger children. This risk is particularly concerning given that as children move into adolescence, parental monitoring of medication adherence may significantly decrease.^45–47 Indeed, a cross-sectional study showed that as children enter adolescence, they experience a concomitant increase in disease management responsibilities and a decrease in treatment adherence.⁴⁸

Behavioral factors associated with nonadherence are not limited to the patients. Several studies have demonstrated that parental stress in the form of maladaptive coping, depression, and anxiety is related to patient nonadherence.^41,49,50 Poor global family functioning and greater parental or family distress or conflict have also been associated with patient nonadherence in several studies.^43,51–55 Specific factors such as poor problem solving and poor communication among family members are also associated with nonadherence.^44,54 One study, which utilized videotaped observation of family interactions, suggested that the quality of a relationship among family members was related to regimen adherence.⁵⁶ Notably, although family dysfunction may significantly contribute to poor adherence in children and adolescents, treatments targeting global family functioning have demonstrated mixed results.^57–59 Thus, treatments targeting specific components of family dysfunction related to nonadherence (eg, parental monitoring, parent-child conflict, communication, problem solving^42,45) may be more effective.⁶⁰ Research has also shown that parental involvement is positively correlated with adherence.^61,62 Although adolescents appear to require some supervision to ensure adequate adherence,⁴⁵ increasing parental monitoring at a time when adolescents are attempting to gain autonomy may be a particular challenge.

Although patient behavioral problems and family dysfunction cannot be considered causal influences in non-adherence due to the correlational design of these studies,⁵⁴ Rapoff⁴⁵ has proposed a risk profile for treatment nonadherence in pediatric chronic illness, in which he suggests that patients at risk for nonadherence tend to have greater adjustment and coping difficulties, less disease management education, and greater responsibility for disease management with little to no parental supervision. Families at risk for nonadherence tend to have dysfunctional behavioral patterns and fewer resources or, conversely, more resources and activities that take time and supervision away from disease management, and these parents tend to have more adjustment and coping difficulties.

Thus, several factors, including patient behavioral problems, family dysfunction, and parental distress appear to contribute to nonadherence in pediatric chronic illnesses. Moreover, these behavioral and family factors appear to be particularly challenging for adolescents with chronic conditions. Although these conclusions are based upon research in other disease groups, it is plausible that these data adequately represent the challenges faced by IBD patients and families given the similarities in chronic disease course, intermittency of symptoms, complexities of treatment regimens, increased risk for behavioral comorbidity, and developmental levels of patients across disease conditions.

Clinical Recommendations for Assessment and Treatment

Assessment of nonadherence in general is a challenging issue due to the fact that there is no gold standard of measurement. Each method of adherence assessment has inherent strengths and weaknesses.⁵ Although patient, parent, and provider reports are cost-effective and feasible clinic-based assessments, they have poor reliability and generally overestimate adherence. Electronic monitoring provides a plethora of data but is an expensive method that is not very feasible in clinic settings. Other objective measures such as direct observation and pharmacy record data are easily skewed and are not always feasible. Pill counts are objective and easily conducted in clinic settings, but they can be manipulated and can present logistical issues with bringing medications to appointments. Finally, bioassays are a valuable measure of drug concentration in blood, but they are subject to pharmacokinetic and pharmacodynamic variation, can vary as a result of inadvertent underdosing, and can be manipulated by patient-initiated dosing changes prior to appointments. Thus, the most appropriate and advantageous assessment for clinicians to use may be one that consists of a combination of measures. The most feasible clinic-based assessments include self-reports (patient or parent), pill counts, and bioassays. Use of a combination of these measures would likely capitalize on the strengths while countering the limitations of each measure. For example, although 6-TGN bioassays can confirm consumption of 6-MP/azathioprine, pill counts can provide data on the estimated number of doses taken. Although a self-report is likely to be overestimated, using a pill count as an objective index of the report can provide more reliability as well as an opportunity to discuss the discrepancy (ie, perceived vs actual adherence) between the two measurements with patients and their families.

Assessment of behavioral dysfunction is also challenging in that the majority of valid and reliable assessment measures require administration by a healthcare provider with specific training in behavioral assessment and interpretation (eg, a psychologist). Some practices have begun to utilize a multidisciplinary treatment approach that routinely incorporates psychologists into patient visits. In these cases, mental health assessments can be performed regularly. For practices that do not use this approach, assessment options are restricted. However, broad assessment measures such as HRQOL questionnaires can be effectively used to screen for suspected emotional and social dysfunction as well as school- and health-related difficulties. These measures are not diagnostic tools and provide limited information with respect to significant behavioral dysfunction; however, clinicians can use the questionnaires as starting points for conversations about how well patients are coping with their illness. Even without the use of screening measures, providers can obtain valuable information by regularly discussing patient adjustment during clinic visits. Often, patients will discuss emotional and behavioral difficulties with a trusted provider if they are given the opportunity. Because of the association between behavioral dysfunction and treatment nonadherence, data obtained via assessments or conversations with patients or families can be used as indicators of risk for disease management problems. Consequently, appropriate referrals to psychologists or psychiatrists can be made to provide treatment for patients' psychological and disease management difficulties.

The issue of nonadherence should also be discussed at every routine clinic appointment. This discussion serves two functions: it helps patients understand that treatment nonadherence is common in patients, and it presents an opportunity to discuss nonadherence as a modifiable health risk. In addition, the language used by providers can heavily influence the reliability of the information obtained and the productivity of the discussion. For instance, the statement “Based upon your 6-TGN assay, it appears that you have not been taking your medication” has a negative connotation compared to the statements “I am concerned about the results of your 6-TGN assay. Taking medication can be difficult for many people and issues such as difficulty swallowing pills or after school activities can keep people from taking all their medicine. Tell me how you have been doing with taking your medication and what challenges you are having.” Use of this type of nonjudgmental language can be very helpful in facilitating both patients' and providers' understanding of issues regarding treatment adherence.

Finally, one particularly important issue regarding treatment adherence is the transition of responsibility from parents to adolescents for completing treatments. The vast majority of children rely on their parents appropriately for support in adhering to regimens. Although adolescents should gradually assume more responsibility for managing their disease, they often assume more responsibility than is developmentally appropriate. Moreover, there is great variability in the age at which adolescents assume responsibility for their treatments. This variability is likely influenced by a number of factors, including but not limited to parental beliefs of their child's readiness, maturity of the adolescent, severity of the disease, psychological comorbidity in the patient or family, complexity of treatments, and the adolescent's demonstrated responsibility for other tasks. Unfortunately, there are no established guidelines to support parents in initiating and completing this important transition. Furthermore, there is no known research that provides an empirical basis for this process from which guidelines could be developed. Despite this limitation in the existing literature, clinicians can provide general, albeit nonempirically supported, guidance regarding this issue. Adolescent patients and their parents or caregivers should be encouraged to base the timing of the transition on the demonstration of the adolescent's developmental readiness. In other words, showing that they can reliably perform tasks that are not disease-related (eg, chores, babysitting) may serve as an indicator that it is time to begin transitioning. The transition should be made slowly and gradually, beginning with the treatments that are easiest to complete and progressing toward more difficult treatments as each one is mastered. Furthermore, transitioning treatment responsibility should not be perceived as a linear process that progressively develops without regression in skill or motivation. When patients have more difficulty managing their disease, parents will need to monitor their adolescent's adherence and assist more with treatments. This process may take several months up to several years, depending upon the individual patient and family. If behavioral problems arise as a function of the transition of responsibility, a referral to a psychologist or another mental health provider may be warranted.

Summary

Pediatric IBD patients are at an increased risk for behavioral and family dysfunction, as well as nonadherence to treatment regimens. Although evidence is lacking, it is plausible, based upon evidence from similar disease populations, that behavioral difficulties impact treatment adherence in IBD. These empirical issues, as well as others (listed in Table 1), require examination to determine the precise relationships among these variables and the most significant behavioral risk factors that impact adherence in IBD patients. Discussion of behavioral issues and disease management skills should be incorporated into clinic visits, and assessment of treatment adherence should consist of a combination of measures. Finally, transition of responsibility for disease management from parents to patients should be gradual and monitored closely by parents and providers to ensure acquisition of skills.

Table 1.

Research Areas for the Future Study of Behavioral Functioning and Adherence in Pediatric Inflammatory Bowel Disease

Area	Specific Issue(s)
Patient/family functioning	Impact of internalizing disorders and family functioning on adherence Academic and peer functioning impact on adherence Transition of treatment responsibility from parents to adolescents
Adherence measurement	Utility and feasibility of objective and subjective clinic-based adherence assessment Benchmarking of adherence with clinical outcome
Patient/family-level intervention	Treatment of nonadherence/self-management support in clinic settings and effect on clinical outcome
Provider-level intervention	Standardization of adherence assessment and intervention approaches