Inflammatory bowel disease (IBD) is being recognized increasingly in childhood and adolescence. The component disorders ulcerative colitis and Crohn disease, with high morbidity and low mortality, affect a group of patients who eventually must transition to independent self-management of their own disease as they reach adulthood.
Knowledge of medication is an important step in the self-management process and some experts believe that this can occur around ages 10 to 12 years, because it is a concrete and tangible accomplishment (1–3). Such knowledge is necessary but not sufficient for full independent self-management. We chose to examine this aspect of knowledge in our patients with IBD. Furthermore, we know that adult GI practitioners reported that medication knowledge is important but problematic in 54% of young adult patients with IBD visiting their practices (4).
Adolescents spend increasing amounts of time apart from their parents and caregivers. Thus, the capacity to report diagnosis and present medication use becomes increasingly relevant, especially in the event of emergency care. Additionally, patients must begin to self-monitor for signs of a flare, because IBD symptoms such as stool consistency or bleeding are considered private matters. Patients also need to be aware of medication adverse effects, such as increased risk of infection or interaction with alcohol, to report relevant symptoms to their provider and to make less risky behavioral choices.
We sought to determine the ability of patients to independently recall the name and dose of their medication and to report knowledge of adverse effects. We aimed to identify factors that influence patient acquisition of medication knowledge.
PATIENTS AND METHODS
The IBD database at Children's Hospital Boston (CHB) was queried to identify all of the patients with IBD older than 10 years. Exclusion criteria were the absence of a visit in the last 3 years or evidence that patients had moved or transferred to another non-CHB provider. There were 781 eligible patients.
Surveys were distributed at clinic visits or outpatient infusion appointments during a 4-month time period (December 2008–March 2009), with 454 consecutive patients approached. Patients and parents received parallel surveys and were asked to fill them out independently before meeting with the provider. The survey was voluntary, anonymous, and confidential. Surveys were identified only by codes linked to medical record numbers, and data were available to nonclinical research staff.
For the older patients and parents, returning the completed survey implied consent. Parents of children younger than 18 years were required to provide written consent to the child's participation in the study. Although this was usually completed when the parents returned their surveys, consent for using the child's survey could be provided without completing the survey. Surveys were placed into a sealed envelope and returned to the nurse before the entrance of the provider. Clinicians were thus unaware of whether the patient or parent filled out the survey and did not have access to any answers in the survey.
Using a checklist, patients were asked to identify their disease (Crohn, ulcerative colitis, indeterminate colitis, or “I don’t know”) and other demographic characteristics (age, sex, and duration of disease). The term “indeterminate colitis” was used rather than IBD-U (unknown) based on pilot data (11). Questions about medications were open-ended. Patients were asked to list their present IBD medications and doses. They were asked to list the adverse effects for their medications even if they did not experience these effects.
Patients’ answers were linked to data extracted from the medical record, which allowed review for accuracy of diagnosis, medication names, and doses.
Patients also often listed supplements, vitamins, antibiotics, fish oil, and proton pump inhibitors. These were not included in the analysis because the reason underlying the consumption was not clearly related to IBD for all of the patients. Steroid use was not included in the analysis because many patients regarded steroids as temporary and did not always record their use. Also, initiation and tapering reporting were often done outside the documented clinic visits, making the accuracy of steroid dosing at the time of the survey hard to assess.
Medications were divided into 4 classes: calcineurin inhibitors (tacrolimus or cyclosporine), biologics (infliximab and adalimumab), immunomodulators (methotrexate, azothioprine, 6-mercaptopurine), and aminosalicylates (mesalamine, balsalazide, sulfasalazine).
Identification of the correct medication was defined as matching generic or brand name to data in the medical record. Correct dosage was defined as matching either pill counts or milligram amount. For biologics, note was made of whether the patient knew the timing interval or actual dose, or both. Major adverse effects of the medications were considered to be risk of infection, risk of malignancy, low white blood cell counts, liver dysfunction, headache, hair loss, pancreatitis, nausea, and secretory diarrhea. Knowledge of adverse effects was considered positive if a single adverse effect was listed that was appropriate for the medication taken by the patient.
The Committee on Clinical Investigation was consulted and identified the present study to be a quality improvement initiative not requiring formal review. All of the analyses were performed using Stata 10 (StataCorp, College Station, TX). They included descriptive statistics, cross tabulations, and logistic regressions. Significance was determined using Fisher exact test for cross tabulations and Wald test statistics for the coefficients of the logistic regressions. Patient age groups for the cross tabulations were determined before analysis based on developmental considerations. The parent survey analysis is reported in another paper.
A total of 454 surveys were distributed, 313 were returned, but 19 of these were not completed (3 lack of time, 1 Spanish speaking, 3 severe developmental delays, 12 refusals). Thus, 294 completed patient surveys were returned, for a response rate of 65%.
Based on information from the medical record, the mean age of the responding patients was 16.7 years (SD 3.5), 146 (50%) were girls, and 201 (68%) had Crohn disease. Demographics are further detailed in Table 1.
Most respondents were receiving at least 1 medication according to medical record data (92%), whereas 8% did not take any medications at all. Among the 271 patients taking medication, 66% were taking 1, 33% were taking 2, and 1% were taking 3 different medications. The total number of prescribed medications was 365. The distribution of the medications by type and age group is outlined in Table 2, along with the percentage of each age group that accurately identified medication.
Most patients could list their IBD medications, and recall was highest for the aminosalicylates and the calcineurin inhibitors. Nine patients did not list their biologic agent and 8 patients did not list their immunomodulator medication when specifically asked about IBD medications. These patients were evenly distributed across the age groups. Patients forgetting their biologics were evenly split between both sexes, but more boys than girls forgot to list their immunomodulators (results not shown). Knowledge of the medication name was not significantly different across age groups (P = 0.192) or by sex (P = 0.109), and a univariate logistic regression confirmed that this also was the case for duration of disease (P = 0.541).
Recall of dosages was slightly lower than medication name, but still remained in the 50% to 75% range for most medicines except the biologics. Some patients could recall the milligram amount, but did not mention that this dose was taken twice daily. Other patients could recall only that their infliximab was given once every 8 weeks. More than 50% of patients receiving biologics could not provide either the dose or timing of that medicine. Table 3 details the accuracy of medication dosing reporting by age group. Patients ages 10 to 12 often recalled doses at least as well as older adolescents. Dosing knowledge for each medication type did not differ significantly in bivariate analyses with age and sex or in a univariate logistic regression model with increasing duration of disease (results not shown).
Overall knowledge of adverse effects was markedly lower, from 22% to 32% for all but calcineurin inhibitors, which was 0%. Many patients confused symptoms from flares with adverse effects of medication, or listed only the cosmetic effects of steroids such as “swollen face.” Knowledge of medication adverse effects does increase with age (Table 4), almost reaching statistical significance (P = 0.064). Girls are not significantly better or worse in knowing adverse effects (P = 0.207). Duration of disease also is not a predictor of adverse effect knowledge (P = 0.162) using a univariate logistic regression (results not shown).
The multivariate logistic regression analysis with 3 independent variables predicting correct identification of medication among patients receiving immunomodulator therapy shows that neither age, sex, nor duration of disease are predictors of correct medication identification (Table 5). Age, however, has a significant positive effect on medication identification for patients taking biologics when controlling for the effects of sex and duration of disease. The odds of correct identification of medication increase by 38% with an age increase of 1 year.
This positive effect of age also was seen with the logistic regression analysis predicting the knowledge of adverse effects across all of the medication types, controlling for the influences of sex and duration of disease (Table 6). The odds of knowing adverse effects increased by 13% for an age increase of 1 year.
In the present study, the vast majority of adolescents with IBD was able to name their medications; however, 17 (7%) patients forgot to mention a biologic agent or an immunomodulator medication even when specifically asked to list any medication taken for their IBD. It can easily be imagined that this could alter the approach to such a patient in an emergency department or college health center setting, particularly in regard to infection or abdominal pain.
It is difficult to compare adolescent recall with adult recall because studies have not yet addressed parallel settings. The few studies looking specifically at medication recall in ambulatory settings tend to show comparable or lower rates than our population. In 344 general practice patients in New Zealand, 85% could name the medication and 83% could recall the correct dose (5). More concerning, in a US emergency department setting, surveying 200 patients, only 48% could name or produce a list of medications and only 24% knew the doses (6). Similarly, at discharge from the hospital, 43 consecutive patients were surveyed and only 28% were able to list all of their medications (7).
It is interesting to note that patients’ knowledge of their medication or doses did not improve as patients aged. At 10 to 12 years of age, patients were able to recall medication name and dose well. This is unlikely to be the result of parent help during the survey, because we specifically asked patients to check off whether they needed help to answer the question. The small decrease in accuracy over time may be the result of emotional aspects of adolescence, such as denial, or negative response to excessive parental pressure. Alternatively, it may represent a natural decline in knowledge as more time passes from the initial discussion of the medication, if insufficient interval education has taken place.
Wallet-sized “passports” or medical summaries may be useful adjuncts to patient recall; however, in actual use, such as with adult stroke patients, few patients continued to use these tools at 6 months (8). Portable medical records or Internet-based records may be extremely useful, but their use is not yet widespread and benefit has not been proven (9).
A small percentage of our patients were able to report adverse effects of their medications. This is disappointing, because our nurse educators meet with every new patient with IBD at diagnosis and our group has extensive discussions with patient and family before starting immunomodulators or biologics. This poor result may be typical for patients with IBD. In a tertiary clinic that reported considerable efforts to provide medication safety information, a series of 100 adult patients with IBD were surveyed about medication safety; 53% reported “no knowledge” of immunomodulators and 79% reported “no knowledge” of biologics (10). In a previous study of 16- to 18-year-olds with IBD, confidence about adverse effect knowledge was quite low, mirroring their actual knowledge (11). In contrast, physicians often overestimate patient knowledge of adverse effects. In 99 patients at discharge from the hospital, physicians believed that 89% of patients understood adverse effects, whereas only 57% of patients reported understanding (12).
Several studies of adult patients with IBD report that patients do want to know about all of the adverse effects of medication, even if risks are rare (10); however, communicating the risks of medication is difficult. Using visual aids, risk reduction statistics or reframing can improve understanding (13–15). Still, patients often underestimate risk. In 165 patients with IBD or parents of patients, >1 of 3 did not believe that infliximab is associated with risk of lymphoma (16). A hypothetical drug was described that mirrored the risks and benefits of infliximab and 64% indicated they would not take the medication, although 30% of these patients presently or previously were treated with infliximab (16).
Providers and parents may worry that frank discussion of adverse effects may worry their young patients excessively or diminish adherence. Adverse effects of medication ranked as the highest concern of parents in 1 study of 60 patients with IBD (17). There is some evidence for adverse effect aversion, showing that the mere presence of an adverse effect diminishes the patient's willingness to take a medicine regardless of benefit; however, willingness to undergo treatment also takes the perceived benefit into account (18).
This was a single-center study, so results may not be generalizable. Respondents were more likely to be compliant and knowledgeable, and thus may bias the study results toward underestimating the percentage of patients who cannot recall medications or describe the adverse effects of major medications. We did not assess socioeconomic status, ethnicity, or primary home language, so we may have missed other factors that play a known role in medication knowledge and adherence (19).
This is the first study to focus on the specifics of knowledge of patients with IBD regarding medication name, dose, and adverse effects across the adolescent years. It is encouraging that 10- to 12-year-olds knew so much, but disappointing that this rate generally did not rise with increasing age or duration of disease. We confirmed deficiencies in knowledge of adverse effects that could be dangerous in college students and young adults. After developing new formal transition preparation and goals for self-management, we will reassess our patient population to gauge efficacy both during the pediatric years and beyond transfer to adult care.
Dr Fishman acknowledges the support given by the Wolpow and Rubin Family Funds through the IBD Center at CHB.
1. Baldassano R, Ferry G, Griffiths A, et al. Transition of the patient with inflammatory bowel disease from pediatric to adult care: recommendations of the North American Society for Pediatric Gastroenterology, Hepatology and Nutrition. J Pediatr Gastroenterol Nutr
2. Hait E, Arnold JH, Fishman LN. Educate, communicate, anticipate—practical recommendations for transitioning adolescents with IBD to adult health care. Inflamm Bowel Dis
3. Patton SR, Graham JL, Holsclaw D Jr, et al. Survey of professionals’ expectations of developmental task achievement of cystic fibrosis self-care in children. Pediatr Pulmonol
4. Hait EJ, Barendse Re M, Arnold JH, et al. Transition of adolescents with inflammatory bowel disease from pediatric to adult care: a survey of adult gastroenterologists. J Pediatr Gastroenterol Nutr
5. Jaye C, Hope J, Martin IR. What do general practice patients know about their prescription medications? N Z Med J
6. Vilke GM, Marino A, Iskander J, et al. Emergency department patient knowledge of medications. J Emerg Med
7. Makaryus AN, Friedman EA. Patients’ understanding of their treatment plans and diagnosis at discharge. Mayo Clin Proc
8. Ayana M, Pound P, Lampe F, et al. Improving stroke patients’ care: a patient held record is not enough. BMC Health Serv Res
9. Ko H, Turner T, Jones C, et al. Patient-held medical records for patients with chronic disease: a systematic review. Qual Saf Health Care
10. Cullen G, Donnellan F, Long S, et al. Perceptions of medication safety among patients with inflammatory bowel disease. Scand J Gastroenterol
11. Fishman LN, Barendse RM, Hait E, et al. Self-management of older adolescents with inflammatory bowel disease: a pilot study of behavior and knowledge as prelude to transition. Clin Pediatr
12. Calkins DR, Davis RB, Reiley P, et al. Patient-physician communication at hospital discharge and patients’ understanding of the postdischarge treatment plan. Arch Intern Med
13. Paling J. Strategies to help patients understand risks. BMJ
14. Carling CL, Kristoffersen DT, Oxman AD, et al. The effect of how outcomes are framed on decisions about whether to take antihypertensive medication: a randomized trial. PLoS One
15. Carling CL, Kristoffersen DT, Flottorp S, et al. The effect of alternative graphical displays used to present the benefits of antibiotics for sore throat on decisions about whether to seek treatment: a randomized trial. PLoS Med
16. Siegel CA, Levy LC, Mackenzie TA, et al. Patient perceptions of the risks and benefits of infliximab for the treatment of inflammatory bowel disease. Inflamm Bowel Dis
17. Day AS, Whitten KE, Bohane TD. Childhood inflammatory bowel disease: parental concerns and expectations. World J Gastroenterol
18. Waters EA, Weinstein ND, Colditz GA, et al. Explanations for side effect aversion in preventive medical treatment decisions. Health Psychol
19. Traylor AH, Subramanian U, Uratsu CS, et al. Patient race/ethnicity and patient-physician race/ethnicity concordance in the management of cardiovascular disease risk factors for patients with diabetes. Diabetes Care