Longitudinal Stability of Specific Barriers to Medication Adherence

Jennifer L. Lee; MS; Cyd Eaton, BS; Ana M. Gutiérrez-Colina, BA; Katie Devine, PhD; Laura E. Simons, PhD; Laura Mee, PhD; Ronald L. Blount, PhD


J Pediatr Psychol. 2014;39(7):667-676. 

In This Article

Abstract and Introduction


Objective Higher levels of barriers are related to lower medication adherence and negative medical outcomes in pediatric transplant recipients. Although total number of barriers appears to be stable over time, it is unclear whether the same is true for specific barriers. This study examined the frequency of endorsement and the stability of specific barriers over 18 months.

Method Participants included 63 parents and 51 adolescents and young adults. Transplant types included 39 kidneys, 16 livers, 10 hearts, and 1 double lung. Participants completed measures of perceived barriers to adherence at Time 1 (T1) and Time 2 (T2).

Results The majority of parent- and adolescent-reported specific barriers showed a positive relationship from T1 to T2. Few specific barriers showed significant differences in the level of endorsement between time points.

Conclusion Specific barriers to medication adherence tend to be stable over time. Patients' specific barriers appear unlikely to change without targeted intervention.


Pediatric solid organ transplantation has become an increasingly common treatment for children with a variety of severe medical conditions, including liver, kidney, and heart disease. Medical progress in the domain of pediatric transplantation, along with advancements in the safety and efficacy of immunosuppressive therapy, has significantly increased the life expectancy of children with an organ transplant (Agarwal & Pescovitz, 2006). However, along with increased life expectancy comes the responsibility of lifelong adherence to a complex medical regimen necessary to maintain the viability of the transplanted organ and to ensure the health of the recipient after surgery (Griffin & Elkin, 2001). Adherence to medical advice and regimens may be burdensome for patients and their families, and nonadherence is common. A recent systematic review of the literature indicated that, on average, 43% of children with organ transplants are nonadherent to their immunosuppressant medications (Dobbels et al., 2010).

The consequences of nonadherence are significant. Failure to follow prescribed medical recommendations can result in organ rejection, death, decreased patient quality of life, and increased cost to the individual (Butler, Roderick, Mullee, Mason, & Peveler, 2004; Fredericks et al., 2008; Fredericks, Lopez, Magee, Shieck, & Opipari-Arrigan, 2007; Pinsky et al., 2009). Specifically, medication nonadherence in pediatric populations is related to greater numbers of visits to the emergency department and inpatient hospitalizations, both of which are the most costly types of health care (McGrady & Hommel, 2013). The U.S. Government Accountability Office (2007) revealed that treatment cost for Medicare beneficiaries who experienced organ loss was $50,398 per year. In contrast, treatment costs for patients who maintained functioning transplants was only $8,550 per year. Therefore, maintaining a healthy organ graft may not only improve medical and psychosocial status, but also help reduce economic burden on the health-care system.

According to the Health Belief Model, perceived barriers represent potentially difficult or unpleasant aspects of health behaviors (e.g., side effects; Janz & Becker, 1984) that may interfere with adherence to prescribed treatment recommendations (Rapoff, 2010). A substantial body of literature in pediatric populations has shown that greater levels of barriers are associated with increased difficulty in following prescribed medical regimens (Marhefka et al., 2008; Modi & Quittner, 2006; Rapoff, 2010). In adolescent transplant recipients, higher numbers of barriers are associated with greater risk for experiencing negative medical outcomes such as rejection episodes, hospitalizations, and/or death (Simons, McCormick, Devine, & Blount, 2010). Barriers to adherence have similarly been associated with negative psychosocial outcomes, including less family cohesion, less emotional expression, and greater conflict among family members (Simons & Blount, 2007). The types of barriers associated with poor medication adherence in pediatric patients include cognitive factors (e.g., forgetting, poor planning), aversive medication properties or difficulties ingesting medication (e.g., hard to swallow, tastes bad), and voluntary resistance (e.g., defiance) toward medication taking (Hommel & Baldassano, 2010; Simons, McCormick, Mee, & Blount, 2009; Zelikovsky, Schast, Palmer, & Meyers, 2008). Low family adjustment (Logan, Zelikovsky, Labay, & Spergel, 2003), as well as high adolescent responsibility and low parent supervision (Modi, Marciel, Slater, Drotar, & Quittner, 2008; Zelikovsky et al., 2008), have also been shown to be associated with poor adherence.

Our previous work with pediatric transplant recipients demonstrated that the total overall number of barriers and barriers subscale scores for the entire sample are stable over time. This study also reported significant associations between specific barriers at baseline assessment and nonadherence and negative health outcomes at long-term follow-up. As these data are part of a larger study from which the current article is derived (Simons et al., 2010), we will review selected results below to support the established relationship between individual barriers to medication adherence and medical outcomes. The relationship between specific barriers at Time 1 (T1) and adherence 18 months later at Time 2 (T2) showed that parent-reported barriers at T1 were positively related to nonadherence at T2 at the specific barrier item level. Being forgetful, disorganized, and the lack of parent reminder were significantly associated with parent-reported nonadherence. Parents' belief that the medication has too many side effects was related to more out-of-range serum immunosuppressant levels. Adolescent-reported barriers (e.g., negative side effects, being tired of taking medications, being tired of living with a chronic medical condition) at T1 were also related to adolescent-reported nonadherence at T2, as indicated by both self-report of missed and late doses, as well as out-of-range serum immunosuppressant levels. With regards to medical outcomes, parents and adolescents reporting more ingestion-related barriers (e.g., difficulties swallowing medications) and not wanting others to see the child taking medications at T1 had more instances of organ rejection and death between T1 and T2. Greater numbers of hospitalizations between T1 and T2 were experienced by children who, at T1, reported difficulties with having too many pills to take and experiencing too many medication side effects.

Even though results from this investigation provided answers to important questions about barriers and their relationship to adherence and health, fundamental questions remain about the nature of barriers to medication adherence. First, it is unclear which specific barriers are most frequently endorsed. Data on the relative percentage of patients and parents who endorse specific barriers more frequently than others may help guide clinicians' assessment and intervention efforts. Second, there are no data to indicate whether individual patient's specific barriers to medication adherence are stable or variable over time. If specific barriers, which appear to function as obstacles to adherence, are stable over time, their negative effects on adherence would likely persist. Effective, focused, early intervention to reduce what might otherwise be stable barriers could help patients overcome specific barriers, minimize continual adverse effects, and enhance adherence and health outcomes. These interventions could provide lasting benefits. Conversely, if barriers are unstable over time, interventionists' efforts become more complicated. Unstable barriers would require continual monitoring over time to know how intervention efforts should be directed. Empirical evidence is needed to provide support for the necessity of intervening on specific adherence barriers, which, if stable, would continue to promote nonadherence and poor health outcomes.

To inform intervention design more precisely, a better understanding of which specific barriers are endorsed and whether barriers remain stable or change over time is critical. The current study addresses these fundamental and important gaps in the study of barriers to adherence by examining specific patient- and parent-reported barriers to medication adherence at an initial data collection time point (T1) and 18 months later (T2). The relative percentage of specific adolescent- and parent proxy-reported barriers to medication adherence will be presented to identify highly endorsed barriers. The stability of endorsement for specific barriers will then be determined through varied analytic methods. It is hypothesized that the majority of specific adolescent- and parent proxy-endorsed specific barriers will be stable over an 18-month period.