|Home | About | Journals | Submit | Contact Us | Français|
ObjectiveKnowledge of factors impacting adolescents’ ability to adhere to their inflammatory bowel disease (IBD) regimen is limited. The current study examines the collective impact of barriers to adherence and anxiety/depressive symptoms on adolescent adherence to the IBD regimen.MethodsAdolescents (n=79) completed measures of barriers to adherence, adherence, and anxiety/depressive symptoms at one of two specialty pediatric IBD clinics.ResultsMost adolescents reported barriers to adherence and 1 in 8 reported borderline or clinically elevated levels of anxiety/depressive symptoms. Anxiety/depressive symptoms moderated the relationship between barriers to adherence and adherence. Post hoc probing revealed a significant, additive effect of higher anxiety/depressive symptoms in the barriers–adherence relationship, with adherence significantly lower among adolescents with higher barriers and higher anxiety/depressive symptoms.ConclusionsIn order to optimize adherence in adolescents, interventions should target not only barriers to adherence but also any anxiety/depressive symptoms that may negatively impact efforts to adhere to recommended treatment.
Crohn's disease and ulcerative colitis (collectively known as inflammatory bowel disease [IBD]) are chronic conditions characterized by gastrointestinal inflammation and unpredictable periods of disease activity and remission (Dubinsky, 2008). Occurring at an incidence rate of 71 cases per 100,000 in youth below 17 years of age, common symptoms of IBD include abdominal pain, diarrhea, delayed growth/puberty, and weight loss (Auvin et al., 2005). Management of IBD symptoms may involve a complex, potentially burdensome, treatment regimen consisting of multiple medication doses per day, dietary modifications, and in some cases, surgery (Kappelman et al., 2007). Although dietary modifications are individually tailored, youth may be asked to abstain from foods that commonly cause gastrointestinal discomfort such high fiber (e.g., nuts, raw vegetables, popcorn), high fat (e.g., fried food), dairy, and spicy food (e.g., pepperoni; Crohn's & Colitis Foundation of America, 2011).
Adhering to IBD treatment may be particularly difficult during the adolescent years, where up to a quarter of IBD cases have an onset (median age of diagnosis=15 years; Kappelman et al., 2007). In a recent review of the pediatric IBD literature, Hommel and colleagues (Hommel, Mackner, Denson, & Crandall, 2008) documented adherence rates ranging from 16% to 62%. Other studies have reported nonadherence prevalence rates of 50–88% (Hommel, Davis, & Baldassano, 2008, 2009; Mackner & Crandall, 2005b; Oliva-Hemker, Abadom, Cuffari, & Thompson, 2007).
Although a link between nonadherence and health outcome has not been firmly established in adolescents with IBD, research from the adult literature suggests that consequences of nonadherence to the IBD regimen can be severe. Adults who are nonadherent are up to 5.5 times more likely to experience a relapse in symptoms (Kane, Huo, Aikens, & Hanauer, 2003) and generate 12.5% greater annual health care costs than adherent patients (Higgins, Rubin, Kaulback, Schoenfeld, & Kane, 2009). Considering these consequences, and that many lifelong positive and negative health management behaviors become established during adolescence (Holmbeck, 2002a), understanding factors which impact adolescents’ ability to adhere to their IBD regimen can play a critical role in the design of interventions to improve lifelong IBD management.
Characteristics unique to the clinical course of IBD, its treatment, and the developmental period of adolescence are likely to increase the challenge of being adherent. Many adolescents struggle with managing a complex oral medication regimen and dietary modifications and cite the time consuming nature of IBD treatment as a major barrier to adherence (Greenley, Stephens, Doughty, Raboin, & Kugathasan, 2010; Ingerski, Baldassano, Denson, & Hommel, 2010). Medication-related side effects, another commonly reported barrier (Greenley et al., 2010), may also impact adherence, particularly when those side-effects (e.g., weight gain, facial swelling, emotional lability) have undesirable social implications or when they occur in the absence of IBD symptoms due to its unpredictable, fluctuating course. These challenges, combined with developmental factors such as adolescents’ increased independence, the transition of treatment responsibility from parent to adolescent, and increased academic and social demands all have the potential to negatively impact adherence in adolescents with IBD. Indeed, the relationship between increased barriers to adherence and decreased adherence has been well-documented in other pediatric populations (Logan, Zelikovsky, Labay, & Spergel, 2003; Modi & Quittner, 2006; Simons, McCormick, Devine, & Blount, 2010; Zelikovsky, Schast, Palmer, & Meyers, 2008) as well as among adolescents with IBD (Greenley et al., 2010; Ingerski et al., 2010).
In addition to having problems with treatment adherence, adolescents with IBD are also at an increased risk for problems in emotional functioning (Mackner & Crandall, 2006; Mackner, Crandall, & Szigethy, 2006; Mackner, Sisson, & Crandall, 2004). They have a 4.6 times greater odds of having clinically significant symptoms of anxiety/depression than healthy peers (Mackner & Crandall, 2006) and 20% of youth report clinically elevated emotional and behavioral symptoms (Mackner & Crandall, 2005a). On the Children's Depression Inventory, 18.5–24.5% of youth exceed the clinical cut-off for depression (Hommel, Davis, et al., 2008; Szigethy et al., 2004).
The lack of prospective, well-controlled research has led to significant controversy regarding the interplay between anxiety/depressive symptoms and the onset and clinical course of IBD (for a review, see Micocka-Walus et al., 2007). There is no evidence that emotional problems play a role in the development of IBD, but they have been observed as an associated outcome of the disease that further exacerbates disease activity. Research among other pediatric populations suggests that illness-associated physical, academic and social impairments, and feelings of helplessness, dependency, and self-consciousness, combined with difficulties managing demanding lifelong treatment regimens may play a role in the development of adjustment problems (Bennett, 1994). Moreover, IBD-specific factors such as difficulties coping with the unpredictable waxing and waning course of IBD symptoms and medication-related side-effects may also contribute to psycho-emotional vulnerability. Collectively, these symptoms may have direct implications for youth's ability to adhere to their treatment regimen.
Research from other pediatric illness populations, such as asthma, HIV, and diabetes (Bender, 2006; Gonzalez et al., 2008; Murphy et al., 2005; Williams et al., 2006), suggests a link between poorer adolescent emotional functioning and suboptimal adherence. However, the relationship between emotional functioning and adherence has not been thoroughly examined in pediatric IBD. In the one known study that examined this relationship, emotional functioning (as measured by the Total Problems score on the Child Behavior Checklist) approached, but did not reach, a significant association with self-reported adherence in a sample of 50 adolescents with IBD (Mackner & Crandall, 2005b). The unestablished link between emotional functioning and adherence in IBD may be due to methodological, as well as conceptual, shortcomings of the emerging IBD literature. Published studies have been largely descriptive, comprised primarily from single-site research, and limited by small sample sizes. Additionally, predictive statistical models, which account for more than one influencing factor, have rarely been incorporated into research, limiting our ability to consider the interactive impact of multiple factors on adherence.
To date, no known study has examined the additive effect of anxiety and depressive symptoms on the relationship between barriers to adherence and adolescent medication adherence in IBD. However, there is evidence that youth with IBD frequently experience barriers to adherence, are at greater risk for anxiety/depression, and struggle with adhering to their treatment regimen. Thus, the current study moves the IBD literature forward by examining the relationship between barriers to adherence, anxiety/depressive symptoms, and adherence in a multi-site sample of adolescents with IBD. Anxiety/depressive symptoms were hypothesized to moderate the relationship between barriers to adherence and medication adherence such that the relationship between these two variables would be strongest among youth with greater anxiety/depressive symptoms.
The current study is a part of a larger two-site project examining treatment adherence and health outcome in adolescents with IBD. Institutional review board approval was obtained at each study site. Participants were 79 adolescents (ages 13–17 years) receiving treatment for either Crohn's disease or ulcerative colitis at one of two hospital-based specialty IBD clinics located in the Midwest (n=34) and Northeastern (n=45) United States, and their accompanying legal guardian. Study eligibility included: (a) adolescent between ages of 13 and 17 years, (b) diagnosis of IBD, and (c) current prescription of oral 5-aminosalycylic acid (5-ASA) and/or 6-mercaptopurine (6-MP)/azathioprine medication (the two most commonly prescribed medications in pediatric IBD). Presence of a neurocognitive disorder (e.g., mental retardation, autism), other chronic illness, limited English literacy, or corticosteroid treatment prescribed at greater than 1mg/kg/day precluded study participation. This latter criterion was established due to higher risk of treatment-associated behavioral and psychiatric side-effects (Kayani & Shannon, 2002; Soliday, Grey, & Lande, 1999).
Eligible participants were recruited by a member of the research team while in private patient rooms during a regularly scheduled medical appointment or via telephone following their appointment. Consent and assent were obtained prior to data collection and each family was compensated $25 for completion of the study measures. With the exception of the adherence interview, all measures were paper-based and independently completed by the adolescent or their parent. Of 109 adolescent–parent pairs approached by study personnel, 21 declined study participation, citing either lack of time, lack of interest, or not wanting a blood draw for disease severity assessment as primary reasons for refusal. Participants with incomplete behavioral data (n=9) were excluded from the study, resulting in a final sample of 79.
Parents completed a brief questionnaire designed for this study to obtain information regarding family annual income, caregiver marital status, and child age, race, and gender.
Adolescents completed the Youth Self-Report (YSR) version of the Child Behavior Checklist (Achenbach, 1988). This well-validated measure assesses behavioral and emotional functioning in adolescents by asking them to rate the frequency with which they have experienced various problem behaviors/symptoms over the past six months. Participant responses are then scored and compared to age and gender-normed data to produce t-scores. T-scores below 65 are classified as “normal,” 65–69 are considered “Borderline” elevated, and scores above 69 indicate “Clinical” levels of symptoms. The YSR yields eight subscale scores of emotional/behavioral functioning and two broad measures of functioning: Internalizing and Externalizing symptoms. The anxiety/depression subscale was used in the current study. This subscale was purposefully chosen over the more global Internalizing syndrome scale which, being partly comprised of the Somatic Complaints scale, may inappropriately produce elevated scores for youth with chronic health conditions (Perrin, Stein, & Drotar, 1991). The anxiety/depression scale of YSR has good test-retest reliability (r=.74) and internal consistency (α=.84) and significantly discriminates between referred and nonreferred youth (Achenbach & Rescorla, 2001).
Adolescent barriers to adherence and adherence to prescribed oral medications were assessed using the Medication Adherence Measure (MAM) (Zelikovsky & Schast, 2008). Designed to be administered during standard clinical care, the MAM is a semi-structured interview conducted jointly with the adolescent and their parent to assess knowledge of their medical regimen, adherence, organization, and barriers to adherence. The MAM contains multiple self-contained modules that can be administered individually or collectively based on the aims of the clinician, clinical population, or purpose of research. For the current study, global assessments of barriers to adherence and adherence were used as this approach is most similar to what is typically done in a clinical setting (La Greca & Bearman, 2003; Rapoff, 2010). Specifically, barriers to adherence were assessed with the question “What are some reasons you miss taking your medications?” After recording spontaneously reported barriers, the interviewer provides the opportunity to endorse any additional barriers by reading the barriers listed in the MAM that were not previously identified. The number of barriers endorsed is summed to generate a barriers score, with higher scores indicating greater number of barriers to adherence. Adolescent self-report of adherence to their treatment regimen was assessed with the question “On a scale of 0 (hardly ever take my medications; usually miss) to 10 (always take my medications; rarely miss), how would you rate how well you take your medications, on average?” Research has shown that patient global report of self-management, such as the approach adopted in this current study, is a more accurate predictor of adherence behavior than more specific assessment methods such as adolescent report of number of pills missed over the past week (Greenley et al., in press). The MAM demonstrates adequate convergent validity (electronic monitoring r=.40, p<.05), test–retest reliability (r=.89, p <.05), and is predictive of long-term health outcome in other pediatric populations (r=0.62, p<0.001; Zelikovsky, 2007; Zelikovsky et al., 2008).
Disease activity was collected for descriptive purposes using either the Pediatric Crohn's Disease Activity Index (PCDAI; Hyams et al., 1991) for youth with Crohn's disease or the Lightiger Colitis Activity Index (LCAI; Lichtiger et al., 1994) for youth with ulcerative colitis. Both measures provide an objective rating of disease activity for their respective IBD subtypes using patient medical data. Each item response is associated with a numerical value and scores are summed to generate a disease activity score. Scores on the PCDAI range from 0 to 100. The PCDAI has strong convergent validity with other measures of disease activity (physician global assessment r=.80, p<.001; modified Harvey–Bradshaw index score r=.81, p<.001) and is able to discriminate between categories of disease activity (Hyams et al., 1991). Scores on the LCAI range from 0 to 21. The LCAI has been used extensively in clinical trials, is sensitive to changes in disease activity in response to treatment, and discriminates well between patients adequately responding to treatment and those requiring surgical intervention (Lichtiger et al., 1994). Reliability in the current sample for both the PCDAI (α=.83) and the LCAI (α=.79) was good.
All analyses were conducted using SPSS 17.0 software (Chicago, IL, USA). Prior to combining samples, t-tests and chi-squared tests were conducted to ensure that participant demographic data (i.e., age, gender, ethnicity, family income, diagnosis) did not differ by recruitment site. All variables were used in their continuous form and were centered prior to analyses. Pearson product correlations between disease severity, barriers to adherence, anxiety/depressive symptoms, and adherence were calculated. Hypothesized moderating relationships and post-hoc probing of significant moderator effects were examined according to the guidelines proposed by Holmbeck (Holmbeck, 1997, 2002b). Per these guidelines, the interaction term (barriers to adherence×anxiety/depressive symptoms) was entered into a regression predicting the outcome variable (adherence) after controlling for the individual predictor variables (barriers to adherence and anxiety/depressive symptoms) and other variables of interest (disease severity). If the interaction term is significantly associated with the outcome variable, moderation is suggested. Although this initial analysis suggests that the association between the predictor and the outcome significantly varies across different levels of the moderator, it does not provide any information about the specific conditions in which this relationship significantly varies. Post hoc probing clarifies this situation through the use of post hoc regressions in which the simple slopes of conditional levels of the moderator (defined as 1 SD above or below the mean of anxiety/depressive symptoms) are computed and individually evaluated for statistical significance. Results of post hoc probing can then be visually depicted by substituting high (1 SD above the mean) and low (1 SD below the mean) values of the independent variable (barriers to adherence) into the post hoc regression equations to illustrate the conditions in which the relationship between the barriers to adherence and adherence vary by level of anxiety/depressive symptoms.
Independent samples t-tests and chi-square tests revealed no significant differences in demographic variables by recruitment site; thus, samples were combined for all analyses. Demographic data for the combined sample are presented in Table I. Of the 80% of adolescents diagnosed with Crohn's disease, 58.7% had inactive disease (PCDAI score≤10), 38.1% had mild disease activity (PCDAI=11–29), and 3.2% had moderate-to-severe disease activity (PCDAI≥30; Hyams et al., 2005). Among participants with ulcerative colitis, 50% had disease activity in the quiescent range (LCAI score≤2), 43.8% were in the mild range and responding to therapy (LCAI=3-9), and 6.3% had active disease that was not responding to therapy (LCAI≥10; Fanjiang, Russell, & Katz, 2007).
Almost all patients (96.2%) endorsed one or more barriers to adherence (M=2.33 barriers, SD=1.47). Most commonly reported barriers included: forgetting (endorsed by 84.8% of sample), being away from home (43%), and medication interfered with an activity (34.2%; Table II). Out of a possible range from 0 (hardly ever take my medications; usually miss) to 10 (always take my medications; rarely miss), mean adolescent self-report of adherence was 8.63 out of 10 (SD=1.32). Table III presents means, standard deviations, and intercorrelations between anxiety/depressive symptoms, barriers to adherence, and self-report of adherence. All relationships were statistically significant (p<.05) and of small to moderate magnitude. Most adolescents (87.3%) had an anxiety/depression t-score within the normal range. A smaller percentage reported anxiety/depression levels in the Borderline range (6.3%) or Clinical range (6.3%).
The moderating role of anxiety/depressive symptoms on the relationship between barriers to adherence and medication adherence was examined via hierarchical regression. Disease severity was entered into the first block of the regression due to its association with adherence. Centered, continuous form versions of the independent variable (barriers to adherence) and the proposed moderator (anxiety/depressive symptoms) were entered into the second block. The interaction of these terms (barriers to adherence×anxiety/depressive symptoms) was entered into the third block. Both the overall model, F(4,78)=8.21, p<.001, and the interaction term, t=−2.26, p<.05, were significant, supporting the hypothesis that anxiety/depressive symptoms moderated the relationship between barriers to adherence and medication adherence (Table IV).
Post hoc probing was conducted to examine the extent to which anxiety/depressive symptoms moderated this relationship. As seen in Figure 1, adherence generally declined as barriers increased. However, this relationship was only significant among youth who had higher levels of anxiety/depressive symptoms (b=−.43, p<.001). By converting the scale of patient report into percentages (i.e., reported numeric value of adherence/10), findings illustrate that an increase in barriers results in an overall decrease of 2% in adherence among patients with lower levels of anxiety/depressive symptoms. However, among patients with higher levels of anxiety/depressive symptoms, adherence decreases by 12.6% as barriers increase. These results suggest that the presence of anxiety/depressive symptoms intensifies the negative impact of barriers to adherence on adolescent's ability to adhere to their regimen.
The current study is the first and only known report of the negative additive impact of anxiety/depressive symptoms on the relationship between barriers to adherence and adherence among adolescents with IBD. Although prior research has linked poorer adherence with increased barriers to adherence, our study is the first to demonstrate that the extent to which this relationship occurs is affected by the presence of anxiety/depressive symptoms. Thus, our findings suggest that anxiety/depressive symptoms may be a critical risk factor in adherence among youth with IBD. Improved understanding of this risk factor may play a vital role in the design of timely, effective adherence interventions. Ultimately, helping adolescents establish positive health management behaviors may help prevent nonadherence and its associated medical and financial consequences in adulthood.
Results indicate that barriers to adherence are normative, rather than the exception, among youth with IBD (barriers endorsed by 96.2% of sample). Additionally, the reasons why adolescents do not take their medication vary greatly. In our study sample, forgetting, being away from home, and treatment interfering with an activity were the most commonly reported barriers to adherence. These barriers in particular illustrate the many difficulties adolescents experience when attempting to manage their IBD treatment in the context of their busy everyday lives. However, when working with adolescents to reduce barriers to adherence, it is important not to assume a one-size-fits-all approach as the underlying causes of the most commonly endorsed barriers may vary greatly and thus require different intervention approaches. For example, when dealing with the barrier of “forgetting,” it is important to determine what factors cause the adolescent to forget to take their medicine. If poor planning and organization are the underlying cause, working with the adolescent to better integrate their medications into their normal routine may be helpful. Other adolescents, who forget to take their medicine when away from home, may benefit most from reminder tools (e.g., cell phone alarm, reminder text messages). However, either of these two approaches would likely be unsuccessful when working with an adolescent whose primary contributors to forgetting are a lack of parental support/monitoring or low motivation to adhere to treatment. In many cases, addressing barriers to adherence is not a straightforward process and in depth problem solving, planning, and support from family, friends, and medical providers is often needed to promote change. Thus, adolescent report of barriers should primarily be considered a starting point in an attempt to understand the complex manner in which barriers negatively impact adherence.
In order to overcome barriers to adherence, it is important to first recognize when problems with adherence occur. Although adolescents generally assume greater treatment responsibility over time, it is essential that parents continue to play an active role in their adolescent's care by monitoring their adherence (Shaw, 2001). This can be done overtly (e.g., checking in with adolescent to see if dose was taken) or covertly (e.g., checking pill box, counting pills). Ideally, the method of monitoring should be a shared decision between the adolescent and their parent to minimize the potential for declines in adherence due to parent-teen conflict (Hauser et al., 1990). Regardless of the method used, monitoring of adherence is essential in first identifying barriers to adherence, then taking steps to address them.
However, addressing barriers to adherence may be more challenging among adolescents with poorer emotional functioning. The current study is the first and only report of higher anxiety/depressive symptoms having an additive effect on the barriers–adherence relationship. Among adolescents with lower anxiety/depressive symptoms, increasing barriers did not significantly impact adherence. However, among those with high anxiety/depressive symptoms, adherence was significantly lower (12.6 percentage points) than those with fewer barriers, suggesting that the presence of anxiety/depressive symptoms amplifies the negative impact of barriers on adherence in youth with IBD.
Barriers to adherence and anxiety/depressive symptoms may combine to impact adherence in several ways. The presence of problematic emotional functioning may undermine adolescents’ ability to recognize declines in adherence, identify barriers, and actively take steps to overcome them. Adolescents with poorer emotional functioning may also struggle with soliciting needed support from family and friends to maintain good adherence. Additionally, those with emotional problems may mistakenly interpret the physiological symptoms of anxiety (e.g., nausea, abdominal distress) and depression (e.g., fatigue) as symptoms of IBD, erroneously conclude that treatment is not working (treatment efficacy barrier), and stop taking their medication. Such health beliefs have been consistently linked with poorer adherence (Brownlee-Duffeck et al., 1987; Bucks et al., 2009; Janz & Becker, 1984).
The detrimental combined impact of barriers to adherence and emotional functioning symptoms on adherence suggests that solely assessing for barriers to adherence among adolescents with IBD is inadequate. In order to optimize improvements in adherence, clinicians must routinely screen for barriers to adherence and symptoms of anxiety and depression. This can easily be done by administering brief, well-validated screeners of anxiety (e.g., Beck Anxiety Inventory [BAI]; Beck & Steer, 1990), depression (e.g., Center for Epidemiological Studies Depression for Children [CES-DC]; Weissman, Orvaschel, & Padian, 1980), and barriers to adherence (e.g., barriers module of Medication Adherence Measure; Zelikovsky & Schast, 2008) during clinic visits and using patient responses on these measures to guide discussion of factors impacting their health and their medical care. Adolescents endorsing clinically elevated levels of anxiety or depression should be referred to a psychologist or other mental health professional for additional care.
Findings from the current study should be considered in the context of its strengths and limitations. The current study moves the pediatric IBD literature forward in several ways. First, this is one of the few known studies to examine multiple factors impacting adherence through the use of predictive, rather than descriptive, statistical analyses. Another strength is the use of multi-site data collection, which allowed for the recruitment of a sample larger than typically included in pediatric IBD research. The use of a well-validated, norm-referenced measure of anxiety/depressive symptoms and the assessment of barriers to adherence and adherence as it is typically done in clinical practice are additional strengths. Limitations of the current study include the cross-sectional nature of data, a relatively homogeneous sample of youth with IBD, the use of self-report to assess adherence, and high self-report of adherence. Although our sample was relatively homogeneous, it is demographically similar to samples previously reported in pediatric IBD studies (Mackner & Crandall, 2007). Self-report, the most commonly used method to measure adherence, has a tendency to overestimate adherence compared to more objective measures such as pill counts, electronic monitoring, or blood assays (Hommel et al., 2009; La Greca, 1995). However, this approach was used for two reasons. Compared to these more objective approaches, which may not be feasible due to time and budgetary constraints, our assessment approach is consistent with what is typically done in clinical practice and is therefore more generalizable. Second, compared to our global assessment approach, objective approaches are less able to capture the multifaceted nature of adherence. Blood assay data, which were available for some participants, were not used for this reason as these data are easily influenced by recent adherence and, as they are medication-specific, do not capture overall adherence. Recent research suggests that adopting a more global approach to adherence assessment is more predictive of actual adherence behaviors than more specific approaches such as asking patients to report the number of pills missed over the past week (Greenley et al., in press). High self-report of adherence is another limitation as this restriction of range may have deflated our correlation coefficients. Given this, it is possible that the combined impact of barriers to adherence and poorer emotional functioning on adherence has been underestimated in our study.
Additional adherence research is needed among adolescents with IBD. Specifically, prospective studies examining the dynamic, interactive relationship between barriers to adherence, emotional functioning, and adherence are needed. Such research would greatly inform the design of effective, well-timed interventions among youth with IBD and would aid in clinical decision making when problems with adherence or anxiety/depressive symptoms become known. Research examining the role of parental monitoring on adherence is also needed as there is little guidance on the optimal type or amount of parental involvement to protect against declines in adherence among adolescents with IBD. Additionally, adolescent beliefs regarding treatment efficacy and the benefits and consequences of nonadherence should be examined as improved understanding of the adolescent's perspective can inform the approach clinicians and parents take to effectively engage the adolescent in a collaborative discussion of problems with adherence.
Although nonadherence is linked with poor health outcome and increased health expenditures among adults with IBD, little is known regarding the consequences of nonadherence to the IBD regimen among pediatric populations. Our cross-sectional data suggest a relationship between adherence and disease activity. However, longitudinal assessment is critically needed to determine the extent to which adherence affects health outcome. Due to our limited knowledge of this issue, it is currently unknown what constitutes a clinically significant increase in adherence or to what extent a minimum adherence threshold exists. Thus, it is difficult to determine the extent to which a 12.6% difference in adherence found among adolescents with high or low barriers to adherence and high anxiety/depressive symptoms corresponds to long term differences in health outcome. These are a critical gaps in the literature that need to be addressed as the pharmacokinetics of IBD medications in children may vary significantly from adults due to children's continuous state of physiological development (Rylance, 1981).
Ultimately, the current study is an important first step toward our understanding of additional factors that negatively impact adherence among adolescents with IBD. Our findings suggest that anxiety/depressive symptoms play a critical role in whether or not barriers to adherence will impact adherence and findings may guide the development of risk profiles to help identify and preventatively intervene with patients at risk for problematic adherence. Continued research in this area would greatly move the pediatric IBD literature closer to providing optimal disease management support for adolescents with IBD.
Research supported in part by NIDDK K23 DK079037, Procter and Gamble Pharmaceuticals, and Institutional Clinical and Translational Science Award NIH/NCRR Grant Number 1UL1RR026314.
Conflicts of interest: None declared.