Adherence to treatment in children and adolescent patients with cystic fibrosis

Article Outline

• Abstract
• Materials and Methods
  • Study design
  • Patients
  • Statistical analysis
• Results
• Discussion
• Acknowledgment
• References
• Copyright

Abstract 

Purpose

The purpose of this study was to monitor medication adherence in cystic fibrosis (CF) patients and its correlation with disease severity and patient age.

Methods

Children less than 12 years of age (group 1) and adolescents 12 years of age and older (group 2) were recruited from the University of Michigan CF Center. The study duration was 3 months. A total of 22 patients per group were enrolled. Adherence to ADEKs, an oral multivitamin, and dornase alfa, a nebulized mucolytic medication, was monitored. Adherence to ADEKs was monitored by using the Medication Event Monitoring System (MEMS) SmartCaps (APREX, AARDx, Inc., Union City, California). Dornase alfa adherence rate was monitored by counting empty medication vials.

Results

Thirty-three patients completed the study, 15 patients in group 1 and 18 patients in group 2. The overall mean adherence rates for ADEKs and dornase alfa were (± SD) 63.6% ± 24.0% and 66.5% ± 31.2%, respectively. The median ADEKs and dornase alfa adherence rate for group 1 was 84.6% and 79.1%, respectively (p = .08); and for group 2 was 56.7% vs. 78.4%, respectively (p = .07). There was a trend toward significance, suggesting that the adherence rate for ADEKs was higher than for dornase alfa (p = .08) in group 1. Group 2 showed a trend toward adherence to dornase alfa than to ADEKs (p = .07). There was a trend for ADEKs adherence between groups 1 and 2 (p = .09), but not for dornase alfa (p = .93).

Conclusion

Parental supervision and disease severity are likely to play a major role in adherence to medical management. Partnership with patients and families about the treatment plan might be important for improving adherence rate. The MEMS SmartCaps is an electronic monitoring technology that should be used to measure drug adherence objectively both in further larger clinical trials and in the outpatient setting.

Keywords:  Cystic fibrosis , Adherence rate , Medication Event Monitoring System Smart Caps

Cystic fibrosis (CF) is a genetic disease that involves the mucus-secreting glands of the lungs, pancreas, and exocrine glands. The incidence is approximately 1 per 2,500 live births in Caucasians. The median age of survival for CF patients is approximately 33 years of age, and patients 18 years of age or older constitute one third of the total patients with CF [1]. Advances in knowledge and medical science have resulted in an increase in life span and improvement in quality of life for patients with CF. The treatment of CF mainly is directed toward correction of organ dysfunction and relieving symptoms that result from this disease [2].

The daily management includes physiotherapy, exercise, oral and inhaled medication for the respiratory system, and oral pancreatic enzyme and multivitamin supplements for treating pancreatic insufficiency and fat malabsorption [3]. The number of medications required to treat the CF patient increases with increasing disease severity. Management issues related to adherence to treatment recommendations have become an increasing focus for physicians and other CF health care providers. Adherence to daily treatment regimens may have short-term and long-term effects on patients with CF [4], [5].

Poor adherence to medication regimens is a common problem with life-threatening disorders. Numerous studies have reviewed and summarized this problem in patients with chronic disease [6]. The mean adherence rate for long-term preventative regimens was 57%, with a range of 33%–94%, and for long-term treatments the mean was 54%, with a range of 41%–61%. More recent work supports these findings [7]. One possible but not well-recognized contributor to nonadherence is that physician treatment recommendations may be poorly understood by parents, children, and/or adolescents [8]. In addition, parents, children, and physicians may have different understandings of severity of illness and expectations for treatment, which may intensify adherence problems [8]. Research on adherence to the medical regimen in CF patients has been relying on subjective measures including the use of interviews and/or questionnaires. Children with CF, their parents, and adult CF patients have been studied using the earlier-described tools [3], [6], [7], [8].

We designed a prospective study to evaluate medication adherence objectively in CF patients. Two patient groups were studied: those less than 12 years of age and those 12 years of age and older, to test for differences in adherence between the 2 groups. We also evaluated adherence to 2 different medications; an oral medication (ADEKs multivitamins), and nebulized medication (recombinant human rhDNase 1, dornase alfa), to assess the effect of route of administration on adherence rate. The nebulized medication adherence rate was measured by counting empty vials. We used the Medication Event Monitoring System ([MEMS] APREX, a division of AARDEX, Inc., Union City, California) to track adherence to the oral medication. MEMS is an electronic monitoring technology that is used to measure drug adherence.

We hypothesized that adherence would be higher in children younger than 12 years of age than in adolescents 12 years of age and older because of parental supervision of the younger age group. This hypothesis was based on our observation of this patient population. In addition, we evaluated adherence rate to 2 medications with 2 different routes of administration: oral versus nebulized treatment, to assess if the route of administration and the time spent taking the medication affected the adherence rate.

Back to Article Outline

Materials and Methods 

Study design 

Children younger than 12 years of age (group 1) and adolescents 12 years of age and older (group 2) with CF, attending the University of Michigan CF Center, were recruited to participate in this study.

The study was conducted over a 3-month period. Patients were recruited during a routine clinic visit to our CF Center. A total of 22 clinically stable patients per group were enrolled. The adherence rate to 2 medications was evaluated. The first medication was ADEKs (multivitamin supplement, plus zinc, specially formulated for CF patients; Scandipharm Inc., Birmingham, Alabama). It is in tablet form and is administered orally. The second medication was recombinant human DNase1 (dornase alfa; Genentech, Inc., South San Francisco, California). It is a mucolytic medication administered through nebulization. The mean time needed for administration is approximately 15 minutes. ADEKs use was monitored by using MEMS SmartCaps. Patients/parents were asked on enrollment to transfer ADEKs to the provided bottle, which was covered with the MEMS SmartCaps. Each MEMS SmartCap contained microelectronics that recorded the date and time the medication bottle was opened. This information was stored until it was retrieved at the end of the study. The dornase alfa adherence rate was calculated by counting the empty medication vials that participants were asked to save and bring at the end of the study.

Patients were evaluated at baseline and at the conclusion of the study as part of their routine follow-up care. Evaluation included history, physical examination, spirometry measurements, and throat/sputum culture.

Patients 

Patients were eligible for this study if they met the following criteria: (1) CF diagnosis as proven by sweat chloride greater than 60 mEq/L or by genetic testing, (2) patients are on dornase alfa and ADEKs as part of their treatment plan. Patient management during the study was decided by the primary pulmonologist during their regularly scheduled visit every 3 months; no modifications of patients’ treatment plans were required for this study. This study was approved by our institutional review board. Written informed consent was obtained from the patients and/or their parent or guardian according to the guidelines of our institutional review board. Early withdrawal from the study was permitted if desired by patient and/or parent or guardian. The total number of patients who met the enrollment criteria was 80 (30% of our total pediatric CF population).

Forty-four patients with CF agreed to participate in the study (55% of eligible patients). The duration of the recruitment was 6 months, and the total study duration was 3 months. Thirty-three patients completed the study. Participants were divided into 2 groups according to their age: group 1 included children younger than 12 years of age, and group 2 included those who were 12 years of age and older. Patients and their parents were informed about the recording capability of the MEMS SmartCaps. The data were analyzed at the completion of the study and return of the MEMS SmartCaps.

Additional patients who met the eligibility criteria were approached during a clinic visit for participation in this study and declined. Reasons for not participating included inconvenience to patient and/or parents of transferring ADEKs to the study bottle, divorced parents with half of the medications at each parent’s residence, and inconvenience of saving the empty dornase alfa vials.

Statistical analysis 

It was determined that approximately 20 patients per group would be required to detect a 40% difference in adherence rate (assuming 80% adherence rate in the younger age group and 40% in the older group), with 80% power at α of .05. To detect a 30% difference (75% and 45% in the younger age group and older group, respectively) at 80% power and α of .05, approximately 40 patients per group would have been required [9].

The primary analysis used adherence as a categorical variable. For this analysis, patients taking 75% or more of their prescribed doses were classified as being adherent to that particular medication. Patients taking less than 75% of prescribed doses were classified as nonadherent to a particular medication. Comparisons of the proportion of patients adherent to each medication between groups were performed using the χ² distribution. The proportion of total prescribed doses taken for each medication also was compared between the 2 groups, and this analysis also used the χ² distribution. To evaluate adherence as a continuous variable, the proportion of prescribed doses taken by each individual patient was determined using either the MEMS data (for ADEKs) or vial count (for dornase alfa), resulting in a subject- and medication-specific score corresponding to the percentage of prescribed doses that were taken. These data then were summarized for each group and compared using the Wilcoxon rank-sum test. Pearson’s correlation coefficient was used to test for linear relationships between continuous variables. Statistical tests were performed using SAS version 8.02 (SAS Institute, Cary, North Carolina), and all tests were 2-sided. Most data are presented as number (%) or mean (SD), except for adherence data, which are presented as median (range) because such data usually are not distributed normally.

Back to Article Outline

Results 

Forty-four patients were enrolled (22 patients for each group) between October 2001 and April 2002. Clinical characteristics of the 2 groups at baseline are summarized in Table 1. Participants’ nutritional status was evaluated by their height for age and weight for age. Their illness severity was assessed by their forced expiratory volume in 1 second (FEV₁) percent predicted. The mean FEV₁ ± SD for group 1 was 89.7 ± 22.6 vs. 67.9 ± 26.0, which indicates that the younger age group has less severe lung disease compared with group 2. Three patients in group 1 did not have spirometry evaluation because they were younger than 6 years of age.

Table 1. Demographic information

	Group 1	Group 2	All
N	22	22	44
Sex	4 M, 18 F	9 M, 13 F	13 M, 31 F
Age (y)	8.5±2.0	16.4±3.1	12.6±4.8
Height (cm)	128.0±12.6	160.2±11.3	144.8±20.1
Weight (kg)	29.6±9.7	51.8±12.1	41.2±15.6
FEV1 (%)⁎	89.8±22.6	67.9±26.0	76.7±26.6

Note: Mean ± SD shown.

M, male; F, female.

Fifteen patients (68%) in group 1 and 18 patients (82%) in group 2 completed the study. For group 1, there was 1 patient who was disqualified because the wrong medication was placed in the MEMS SmartCaps bottle (pancreatic enzymes instead of ADEKs multivitamins). Five patients in group 1 did not complete the study because they forgot or were too busy to complete the study and comply with the instructions. One patient stopped taking dornase alfa and did not complete the study. For group 2, there was 1 patient who was disqualified because of improper use of the monitoring device and 3 patients chose not to complete the study.

Table 2 represents a summary of results. The mean number of days monitored between the 2 groups were not statistically significant with 98.2 ± 26.8 (SD) in group 1 versus 108.3 ± 24.4 in group 2. Overall, there were no statistically significant differences in the percentage of patients adherent to their ADEKs or dornase alfa therapy between the 2 study groups. However, there was a trend toward significance, with a higher proportion of adherent patients in group 1 (age, <12 y) than in group 2 (age, ≥12 y), a trend that was more clear regarding ADEK adherence (53.3% vs. 23.5%, respectively; p = .14) than for dornase alfa adherence (57.1% vs. 50.0%, respectively; p = 1.0). There were statistically significant differences between the groups in the proportion of prescribed doses that were taken (Figure 1). With regard to ADEKs, 67.7% of the prescribed doses for patients in group 1 were taken as compared with only 55.3% of the prescribed doses for patients in group 2 (p < .0001). Conversely, with regard to dornase alfa, only 58.0% of the prescribed doses for patients in group 1 were taken as compared with 67.1% of the prescribed doses for patients in group 2 (p < .0001). Comparisons of mean per drug adherence rates between groups indicated a greater median adherence among patients in group 1 than among those in group 2 for ADEKs (70.5% ± 27.3% vs. 56.7% ± 20.6%, respectively; p = .12). The mean adherence rate for dornase alfa was lower among patients in group 1 than among patients in group 2 (62.9% ± 35.4% vs. 70.0% ± 27.0%, respectively; p = .56) (Table 3).

Table 2. Summary of results

	Group 1	Group 2	p value
N (M/F)	15 (1/14)	18 (7/11)
ADEKs adherence rate	70.5%±27.3%	56.7%±20.6%	.122
Dornase alfa adherence rate⁎	62.9%±35.4%	70.0%±27.0%	.556
Number of days monitored	98.2±26.8	108.3±24.4	.269

Note: Mean ± SD shown.

M, male; F, female.

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Percentage of ADEKs and Dornase alfa doses taken appropriately according to study group.

Table 3. Comparison of median adherence rate for the 2 groups

	Group 1⁎	Group 2⁎	p value†
ADEKs	84.6(97.5–7.7)	56.7(90.7–29.6)	.09
Dornase alfa	79.1(100–17.6)	78.4(100–8.5)	.93
p value†	.08	.07

Comparing median ADEKs adherence to dornase alfa adherence within each group confirmed that adherence was greatest for ADEKs among group 1 patients (84.6% vs. 79.1%, respectively; p = .08), whereas among group 2 patients, the median adherence to dornase alfa was greater than the median adherence to ADEKs (78.4% vs. 56.7%, respectively; p = .07) (Table 3).

The length of time during which patients had their adherence monitored also seemed to be at least somewhat related to adherence rates. The length of monitoring was moderately negatively correlated with both ADEKs adherence (r = −.44, p = .012) and dornase alfa adherence (r = −.54, p = .003), indicating poorer adherence with longer duration of monitoring.

Back to Article Outline

Discussion 

Previous studies mainly have evaluated adherence to medication in CF patients subjectively by relying on patient/parent questionnaires, diaries, and self-reporting [3], [6], [7], [8]. In this study, the adherence rates to 2 medications commonly used in CF patient management were measured objectively. The first medication was ADEKs, an orally administered multivitamin especially formulated for CF patients. ADEKs adherence was evaluated using the MEMS monitors. They are special pharmacy vial caps that are used to record the actual medication vial openings as a measure of dosing, including date and time. The use of MEMS proprietary software allowed us to analyze and produce a compliance report for each patient. The second medication was dornase alfa, which was administered through a nebulized route. The adherence rate to this medication was tested by asking parents to bring in the empty vials used during the study period. We understood that some patients/caregivers could tamper with the vials. However, there was no way for us to rule out this possibility.

The overall mean adherence rates (± SD) of ADEKs and dornase alfa were 63.6% ± 24.0% and 66.5% ± 31.2%, respectively. These were significantly better than reported adherence rates in previous studies [10], [11], [12], [13]. In these studies, the range of adherence rates across all medical prescriptions was between 30% and 70%, with a mean of 50% of patients with chronic disease showing adherence. Because of the chronicity of CF and the necessity of multiple daily medications (the median number of daily prescriptions per patient in one study was 7); patients are likely to have a high degree of nonadherence [10]. The total number of medications taken by each group was not a factor in the adherence rate because the mean number of medications (± SD) was 7.9 ± 2.7 for group 1 and 8.3 ± 3.7 for group 2.

In this study, we evaluated 2 age groups to assess the effect of age, severity of illness, and parental supervision on adherence rate. The first group was younger than 12 years of age with a mean FEV₁ (±SD) of 89.8% ± 22.6%, which indicates mild lung disease. The second group was 12 years of age and older with a mean FEV₁ (±SD) of 67.9% ± 26.0%, which indicates moderate lung disease. The medication intake of the younger age group was expected to be supervised closely by their parents, according to our observation and experience with this group. That is in contrast to the older group, who generally were less supervised by their parents or guardians. In addition, some of the older age group displayed signs of denial or rebelliousness against their disease and could be less adherent to their treatment plan.

The number of patients who did not complete the study was higher in group 1 than in group 2 (7 vs. 4 patients). The adherence rate was higher for ADEKs in group 1 versus group 2, and the opposite for dornase alpha. The highest mean adherence rate (± SD) was for ADEKs in group 1 (70.5 ± 27.3%). The reasons for that could be the ease of administration, fast intake, and the focus of parents on vitamin intake for the younger age group. The dornase alfa adherence rate was 62.9% ± 35.4%, which was worse than ADEKs. The reasons for that could be the amount of time spent administering the medication and the uncertainty of parents of the benefit of this medication in this mild disease group.

The group 2 mean adherence rate was higher for dornase alfa than for ADEKs (70.0% ± 27.0% vs. 56.7% ± 20.6%). One explanation could be that these patients who have moderate lung disease decided that a medication to liquefy their thick secretions is more worthwhile to them than a multivitamin. We analyzed the data by using the median adherence rate for each medication in addition to the analysis using the mean. This analysis was performed because the data were not distributed normally (Table 3). In this analysis group 1 showed a higher rate of adherence to both medications, with adherence being higher for ADEKs with a p value of .08. Group 2 showed a higher adherence rate to dornase alfa. This was not statistically significant; however, it showed a trend toward significance with a p value of .07

In conclusion, this is a preliminary study to evaluate the potential use of the MEMS adherence monitoring system for use in future larger adherence studies. In addition, it seems likely that close parental supervision could be an important factor for improving adherence rate. It seems that disease severity plays a major role in adherence to certain medications that patients view as beneficial and/or leads to a decrease of symptoms. As caregivers, we need to educate our patients and their families more about the value and importance of each medication to the care of these patients and maintaining a good quality of life. Moreover, we need to discuss and work with patients and their families on time management and try to simplify their treatment plan as much as possible.

Larger studies are needed to further objectively evaluate adherence rates to different medications and to investigate ways to improve adherence rates for all patients. The use of MEMS monitors and software can address the medical and economic challenges of objectively evaluating nonadherence with medications.

Back to Article Outline

Acknowledgment 

Supported in part by the Cystic Fibrosis Foundation Traineeship grant, Axcon Scandipharm, Inc., Birmingham, AL and Genentech, Inc., South San Francisco, CA.

Presented in part in a poster session at the North American Cystic Fibrosis Conference, New Orleans, LA, October 3–7, 2002; at the 13th Annual Pediatric Research Symposium, receiving honorable mention, University of Michigan, Ann Arbor, MI, October 18, 2002; and at the American Thoracic Society International Conference, Seattle, WA, May 16–21, 2003.

Back to Article Outline

References 

1. Cystic Fibrosis Foundation . Highlights of 2002 Cystic Fibrosis Patient Registry Annual Data Report . Bethesda, MD: Cystic Fibrosis Foundation; 2001;
   • View In Article
2. Nasr SZ . Cystic fibrosis in adolescents and young adults . J Adolescent Medicine (State of the Art Reviews) . 2000;11:589–603
   • View In Article
3. Abbott J , Dodd M , Bilton D , Webb AK . Treatment compliance in adults with cystic fibrosis . Thorax . 1994;49:115–120
   • View In Article
   • MEDLINE
   • CrossRef
4. Desmond KJ , Schwenk WF , Thomas E , et al.   Immediate and long-term effects of chest physiotherapy in patients with cystic fibrosis . J Pediatr . 1983;103:538–542
   • View In Article
   • Abstract
   • Full-Text PDF (424 KB)
   • CrossRef
5. Warwick WJ , Hansen LG . The long-term effect of high-frequency chest compression therapy on pulmonary complications in cystic fibrosis . Pediatr Pulmonol . 1991;11:265–271
   • View In Article
   • MEDLINE
   • CrossRef
6. Sackett DL , Snow JC . The magnitude of compliance and non-compliance . In:  Haynes RB ,  Taylor DW ,  Sackett DC editor. Compliance in Health Care . Baltimore, MD: The John Hopkins University Press; 1979;p. 11–22
   • View In Article
7. Abbott J , Dodd M , Webb AK . Health perceptions and treatment adherence in adults with cystic fibrosis . Thorax . 1996;51:1233–1238
   • View In Article
   • MEDLINE
   • CrossRef
8. Ievers CE , Brown RT , Drotar D , et al.   Knowledge of physician prescriptions and adherence to treatment among children with cystic fibrosis and their mothers . J Dev Behav Pediatr . 1999;20:335–343
   • View In Article
   • MEDLINE
   • CrossRef
9. Campbell MJ , Julious SA , Altman DG . Estimating sample sizes for binary, ordered categorical and continuous outcomes in two group comparisons . BMJ . 1995;311:1145–1148
   • View In Article
   • MEDLINE
   • CrossRef
10. Haynes RB . Introduction . In:  Haynes RB ,  Taylor DW ,  Sacket DL editor. Compliance in Health Care . Baltimore, MD: The John Hopkins University Press; 1979;p. 1–7
    • View In Article
11. Ley P . Towards better doctor-patient communication (contributions for social and experimental psychology) . In:  Bennett AE editors. Communications Between Doctors and Patients . London: Nuffield Provincial Hospitals Trust; 1976;p. B38
    • View In Article
12. Ley P . Communicating with patients (improving communication, satisfaction and compliance) . London: Croom Helm; 1988;
    • View In Article
13. Barofsky I . The chronic psychiatric patient in the community . New York: Plenum; 1980;
    • View In Article