Medication-Taking Behavior in a Cohort of Patients with Inflammatory Bowel Disease
- First Online:
- Cite this article as:
- Bernal, I., Domènech, E., Garcia-Planella, E. et al. Dig Dis Sci (2006) 51: 2165. doi:10.1007/s10620-006-9444-2
- 270 Views
Recent studies have shown a low adherence rate to maintenance treatment in patients with inflammatory bowel disease (IBD). We sought to assess the medication-taking behavior in a cohort of patients with IBD. We prospectively included IBD patients from the outpatient clinic who agreed to answer a questionnaire about prescribed treatment and adherence. Physicians registered clinical data including prescribed medications. Two hundred fourteen patients (115 Crohn's disease/99 ulcerative colitis) were included. The most prescribed medications were oral mesalazine (56.5%) and immunomodulators (41.1%). Forty-three percent of patients admitted to occasionally forgetting to take their medication but only 7.5% of them did it voluntary. Oral mesalazine and azathioprine were the drugs with the poorest compliance, with nonadherence rates of 45% and 25% of the total prescribed doses, respectively. The only factor associated with a better adherence was a more complicated course of the disease—steroid dependency, steroid refractoriness, need for infliximab treatment, hospitalization, or surgery (P=.02). Twenty percent of patients admitted to self-medicating. An important proportion of patients with IBD admit to forget some doses of the prescribed medication in the setting of a specialized unit of a referral centre.
KeywordsInflammatory bowel diseaseUlcerative colitisCrohn's diseaseAdherenceComplianceSelf-medicationMesalazineAzathioprine
Adherence may be defined as the extent to which patient's behavior (in terms of taking medication, following a diet, modifying habits, or attending clinics) coincides with medical or health advice . As in other chronic diseases, nonadherence is widespread in patients with inflammatory bowel diseases (IBD) [2–4]. These patients usually need long-term treatment, not only to induce remission during flare-ups, but also to prevent relapses and/or further complications. Those studies in IBD patients (mainly ulcerative colitis [UC]) where compliance has been specifically assessed, have shown adherence rates of 20% and 50%, for short- and long-term treatments, respectively [3–6]; these figures are very close to those reported for other chronic diseases . However, data regarding adherence in IBD patients are scarce and most studies have been addressed exclusively to maintenance treatment with aminosalicylates in inactive UC . In a recent prospective trial including UC patients, adherence to mesalazine treatment was only 57% . Nonadherence to maintenance treatment in UC may increase not only the risk of disease relapse , but also possibly the risk of developing colorectal cancer . Many conditions have been associated with a nonadherent behavior in IBD patients, such as sociodemographic factors (male, single, full-time employment), IBD-related factors (short duration or noncomplicated disease), treatment characteristics (3 times daily dosing or multiple concomitant medications), or the presence of psychiatric comorbidity [3, 4, 9].
The primary aim of the present study was to assess the adherence to the prescribed treatment in a large cohort of IBD patients and identify factors associated with higher treatment compliance. Secondary objectives were to evaluate the level of understanding of the prescribed treatments and the self-medication rate in these patients.
Patients and methods
Medication-taking behavior questionnaire
1. Which oral medications are you currently taking for IBD? (List with available drug trade name given.)
2. Which rectal medications are you currently taking for IBD? (List with available drug trade names given.)
3. In a 1-week period, do you forget any doses of your medication? If the answer is affirmative
a. How many days (in a 1-week period) does it happen?
b. When you “forget” taking the medication, is it usually intentional or unintentional?
c. The main reason for you to skip 1 dose is (you can have >1 answer):
• coincidence of medication with working time
• weekends (change in the daily routine)
• going out (work, social events …)
• medication-related side effects
• difficulty in taking the medication (large pills, rectal administration, bad taste)
• I think it is not necessary to take all or “so much” medication
d. Which medication do you tend to “forget”?
4. About the physician in charge of your IBD, how do you qualify the confidence degree that you have with him/her?
5. Have you ever self-medicated (followed any treatment with a medication not ordered by a physician) in any of your disease flare-ups or when you believed you had a flare-up? In the answer is affirmative, which medication do you self-medicate with?
6. What is the reason for you to self-medicate?
• Clinical worsening
• Fear of a new hospital admission/surgery
· Fear of consequences in your professional activity
Results are expressed as mean values ± SD or frequencies. χ2 test for qualitative variables, and Student t-test for quantitative variables were used to compare adherent and nonadherent patients. All statistical analysis were performed using the SPSS 12.0 for Windows package (SPSS Inc., Chicago IL).
Clinical characteristics of the patients (n=214)
Crohn's disease (n=115)
Ulcerative colitis (n=99)
Time from diagnosis (mos)
Ileum/colon/ileocolon/upper GI (%)
Previous hospitalization for IBD attack (%)
Steroid dependence (%)
Steroid refractoriness (%)
Infliximab treatment (%)
Intestinal resection (%)
Mesalazine was the most frequently prescribed oral medication (56.5% of patients), followed by immunomodulators (41.1%). Almost 20% of patients were on topical treatment with mesalazine or steroids. Less than 5% were on systemic steroids or budesonide; most patients had inactive or only mildly active disease, as expected in an outpatient series. Detailed data about prescribed medications is shown in Table 4. Patient–physician relationship was classified by doctors as good or very good in 80% of cases, and patients were confident in their physician in >90% of cases. Surprisingly, almost 1 out of 6 patients (13.6%) did not agree with their physician about which were the currently prescribed drugs.
Ninety-three patients (43.5%) admitted to forget ≥1 dose of the prescribed medication during the week, but only 8% of them did so intentionally. The median number of these “drug oversights” was 2 days per week (interquartile range, 1–2). The main reasons given by patients for a noncompliant behavior were weekend period (16.4%), working days (9.8%), and not being at home (9.8%). Only a minority of cases admitted inconvenience or difficult drug administration, excessive medication, or drug-related side effects. Regarding specific drugs, oral mesalazine had the worse adherence. Forty-five percent of patients under oral mesalazine admitted to “forgetting” some drug doses every week. Considering that mesalazine is usually administered in 2–3 daily doses, and taking into account a median of “drug oversights” of 2 days per week, the estimated adherence rate for these drug would be 71.5–90.5%. Azathioprine had also a poor adherence rate, with 25% of patients with “drug oversights” (median, 2 days per week). In contrast, topical treatments were associated with very good medication-taking behavior, with only 13% of patients admitting “drug oversights.” It is worth mentioning that, in our hospital azathioprine is usually prescribed as a single daily dose, resulting in an estimated adherence rate of 71.5%.
Educational background and employment status
Student (in addition to employment status)
Finally, almost 1 out of 5 patients admitted to practicing self-medication in some instances. Oral (41%) and rectal (21%) mesalazine, and oral (29%) and rectal (17%) steroids were the most often used drugs. The main reasons for self-medication were suspected clinical worsening (80%), fear of hospital admission or surgery (12%), and fear to be on sick leave (10%).
Factors favoring compliance
Time from diagnosis
Education (none or basic/high school–college/university)
Previous hospitalization for IBD attacks
There is no gold standard method to measure adherence, but interview and questionnaire methods are most commonly used because they are easy to obtain and inexpensive. These questionnaires are susceptible to misrepresentation and tend to overestimate adherence, but they detect most nonadherent patients . Furthermore, validated interviews for assessing adherence in IBD do not exist. Although we used a questionnaire to assess the medication-taking behavior, it was conducted according to general standards to increase valid responses (i.e., nonjudgemental questions were used, it was conducted by personnel who do not provide direct medical care, confidentiality was emphasized). In addition, patients with a short duration of disease were excluded to reduce a possible bias because of a hypothetical lower awareness of chronic illness or a short-term relationship with their physician. Patients who have had a colectomy for UC were also excluded, because they become theoretically free of disease, whereas those who had it for CD remain at risk for disease recurrence.
Several studies have prospectively evaluated adherence in IBD [2–4, 9, 10, 13–16], most of them including inactive IBD outpatients (specifically in inactive UC taking mesalazine). In these studies, medication-taking behavior was measured by questionnaires, rates of refilling prescriptions, or urine metabolite determination, and reported nonadherence rates of 17.5–72.0%. Our figures are very close to those obtained in IBD studies using direct inquiries, with nonadherence rates up to 40% [9, 13]. However, a similar study performed in our country with a small series of IBD outpatients showed a nonadherence rate of 72% . When evaluating medication-taking behavior in relation to each specific drug, our data on mesalazine were again very similar to that reported in previous studies with IBD  or UC patients , with nonadherence rates of 43% and 52%, respectively. In our series, patient's compliance was higher for immunomodulators than for aminosalicylates. Similar results were also seen in a recent study performed in a small series of pediatric IBD patients, with self-reported rates of “complete adherence” of 70% and 43%, for immunomodulators and aminosalicylates, respectively . In a observational study, Bloomfeld and Onken  reported the initial azathioprine/6-mercaptopurine metabolite levels of 9187 patients with digestive diseases. In this large series, only 3% of patients had undetectable levels of metabolites, suggesting noncompliance to treatment; however, 46% of patients were underdosed, resulting from an inadequate prescribed regimen by the doctor or an inadequate patient compliance . Topical treatment was not associated with worse adherence in our study, in contrast to larger studies were refill rates for rectal therapies were very low compared with those for oral therapies . However, only 15% of patients were on topical treatment in our series, accounting for a small number of patients.
Major predictors associated with poor adherence include presence of psychological problems or cognitive impairment, treatment of asymptomatic diseases, inadequate follow-up or discharge planning, drug-related side effects, patient's lack of belief in the benefit of treatment or lack of insight into the illness, poor physician–patient relationship, presence of barriers to care or medications, missed appointments, complexity of treatment, and medication cost . Some of these factors have also been reported to be quite specific for IBD patients, specifically those related to physician–patient relationship, depression, disease inactivity, and short disease duration [4, 9, 13, 14]. We found that only a “complex” pattern of disease was associated with a better adherence. However, some pitfalls may have influenced the results of our study. First, patient–doctor relationship was assessed, with excellent results in the awareness by both patients and physicians and, in addition, when patients were asked about the information received from their physician regarding their disease, <10% qualified it as insufficient (data not shown). Second, disease activity was not evaluated as a predictive factor of adherence because there were only a few patients with active IBD. Third, we excluded those patients with disease duration of <2 years. In turn, most of these factors (physician–patient relationship, disease inactivity, and short disease duration) might not be considered in our study. We found a trend toward better adherence in patients with disease-related disability or unemployed (although it did not reach statistical significance), in agreement with other studies showing that full-time employment reduces patient's compliance . As previously stated, we found a higher adherence to azathioprine than to mesalazine. This difference might be related to a less complex disease (thiopurines are usually started after intestinal resection in CD, and as maintenance treatment in steroid dependency, concomitant with infliximab therapy, or after a successful course with cyclosporine in steroid refractory flares) and a cumbersome drug administration (higher number of daily doses, larger tablets or pellets) in the latter group of patients, but the reason for that was not specifically assessed. In this sense, a recent systematic review found that adherence is inversely proportional to frequency of dose .
In summary, in specialized IBD units from referral centers (were adherence might be overestimated) nonadherence is common. This may contribute to a worse disease course, decreased quality of life, and higher economic cost of disease. In addition to general major predictors of poorer adherence, a less complex IBD course is more often associated with worse medication-taking behavior. Despite some favorable conditions found in our series (good patient–doctor relationship, adequate disease information, easily provided nonscheduled visits), these data highlight the importance of developing better programs to help patients understand their disease and their medications, as well as the need for developing advances in drug delivery technology to provide simpler and more convenient dose regimens.
This study was supported in part by a grant of Instituto de Salud Carlos III (C03/02), from the Spanish Ministry of Health.