An educational intervention on drug interactions and contraceptive options for epilepsy patients: a pilot randomized controlled trial

Abstract

Objective

This pilot study investigates whether an educational handout could increase short-term information retention about drug interactions between antiepileptic drugs (AEDs) and hormonal contraceptives among female epilepsy patients of reproductive age.

Study design

This is a pilot randomized controlled trial of an educational intervention among reproductive-age women with epilepsy in an academic neurology clinic. Investigators measured knowledge before and after participants received either usual care or the educational handout. The 10-question test assessed increased knowledge of which AEDs affected efficacy of certain hormonal contraceptives and was assessed by calculating the improvement in score between the pretest and posttest. The educational handout included the names of AEDs that have drug interactions with certain contraceptives and the efficacy of the contraceptives.

Results

A total of 42 epilepsy patients participated in this study. Fourteen participants were taking AEDs that are enzyme p450 inducers and 13 participants were taking Lamotrigine. Twenty women were randomized to receive the educational handout and 22 women were randomized to usual care. We found no statistical difference in the groups with regard to age, ethnicity or level of education. We found a significantly higher improvement in quiz scores in the educational handout group (3.65 point increase) compared to the usual care group (0.68 point increase) as calculated by the Student’s two-sample t test (p<.001).

Conclusions

An educational handout on drug interactions and contraceptives resulted in increased short-term information retention on this topic among reproductive-age female epilepsy patients.

Implications

This pilot study highlights the need for further larger studies to evaluate the impact of educational interventions on improving patient knowledge about the drug interaction of AEDs and hormonal contraceptives.

1. Introduction

There are an estimated 1 million women with epilepsy in the United States with a cumulative incidence of 1.7% in women by age 50 years [1,2]. Hormonal contraceptives can interact with antiepileptic drugs (AEDs) in a variety of ways. Many AEDs such as oxcarbazepine, carbamazepine, topiramate and phenytoin can induce enzymes in the liver that increase the metabolism of hormonal contraceptives that can lead to unintended pregnancy [3]. Unintended pregnancy should especially be avoided in women with epilepsy because of the risk of birth defects with use of several of the AEDs [4]. Lamotrigine is an AED frequently used in reproductive-age women because it is less teratogenic than alternative AEDs. However, combined oral contraceptive can affect the metabolism of Lamotrigine reducing the concentration and potentially causing increased seizure activity [4].

The few small cross-sectional descriptive studies looking at contraceptive use in women with epilepsy found at women on AEDs do not have adequate knowledge of hormonal contraception/AED interactions and feel that they are receiving insufficient counseling [1,5–7]. Looking at what methods of contraception women with epilepsy use, a small study in urban medical center showed that, of 58 women at risk for pregnancy, 10 women (17%) used sterilization and 13 women (22%) used hormonal methods (9 oral contraceptive pills, 3 patch, 1 depot medroxyprogesterone acetate) [1]. We could find no published reports on interventions that address the issues of improving patient knowledge or improving use of highly effective methods of contraception. Specifically, there has not been a study on an educational intervention about the interaction between hormonal contraceptives and AEDs. The goal of this study was to assess the efficacy of an educational handout for participants on the ways AEDs and hormonal contraceptives may interact to alter effectiveness of the medications.

2. Materials and methods

This study was approved by the University of California, San Diego, Human Research Protection Program. Women of reproductive age (18–45 years old) presenting to an outpatient neurology clinic at an academic medical center in San Diego for follow-up epilepsy care were approached to participate in this study. The neurology staff performed eligibility screening and notified the study team member when there was a potential qualifying participant interested in study participation. The study team member screened for inclusion and exclusion criteria prior to consenting eligible participants. Inclusion criteria were sexually activity with a male, taking at least one AED and English speaking. Women were excluded if they were currently pregnant, were attempting pregnancy or had a history of hysterectomy, bilateral oophorectomy or sterilization procedures. Once enrolled in the study, participants were allocated to either the intervention arm or control arm. Randomization was 1:1 and included a block size of 6. Randomization was computer generated and stratified by clinic location. One clinic was in an urban setting and the other clinic was in a suburban setting. The randomization allocation assignments (handout or usual care arm) were prepared by a statistical service. The assignments were sealed in sequentially numbered, opaque envelopes. After participants were consented, the study team member opened the envelope and allocated the participant to the arm of the study indicated in the envelope.

All participants completed a preintervention knowledge examination (pretest) in their patient examination room. With a study team member in the room to ensure no outside information resources were used, participants had approximately 5 min to complete the pretest before it was collected. The 10-question pretest assessed patient knowledge of which AEDs were known to interact with contraceptive methods, patients’ knowledge of rates of unexpected pregnancy and efficacy of different contraceptive methods (Appendix Fig. 1). The maximum score on this test was 10 with each correct answer given one point. The pre-test questions were adapted with permission from studies by Pack et al. and Eisenberg et al. [1,8] participants also received a preintervention contraceptive questionnaire. This questionnaire included contact information, demographics and pregnancy history. The questionnaire was adapted from the Center for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System [9]. After completing the pretest and contraceptive questionnaire, participants randomized to the intervention arm received the educational handout (Appendix Fig. 2). These participants had until the end of their appointment with the neurologist to read and review the handout before it was collected by a study team member. Participants randomized to the control group did not receive the handout and proceeded to their scheduled appointment with the neurologist. All participants completed a posttest after their neurology appointment. The posttest consisted of the same questions as the pretest but with the answer choices scrambled. Finally, all participants also completed a postintervention contraception questionnaire capturing information about counseling and contraceptive plans. Participants were compensated with a gift card after completion of the postintervention contraception questionnaire.

We considered a difference in score of 20% (2 points) to be clinically meaningful. Assuming a standard deviation of 2, the number of participants needed in each treatment arm for a power of 0.80 with an α equal to 0.05 was 17. We needed 42 total participants to account for 20% potential participant discontinuation.

Descriptive statistics were used to report participants’ demographic characteristics and baseline knowledge test scores. Intervention vs. control group differences on demographics and baseline variables were assessed using χ² tests. Separate exploratory two-way (study assignment×subject characteristic) analyses of variance examined whether the effects of the intervention on contraceptive knowledge were moderated by the following subject characteristics: baseline knowledge of hormonal contraceptives/AED interactions, age and education. Point estimates of continuous variables are reported in the form of mean and standard deviation or median and interquartile range (IQR) by specifying the first and third quartiles.

3. Results

A total of 42 epilepsy patients participated in this study. Twenty women were randomized to receive the educational handout and 22 women were randomized to usual care. Fourteen participants were taking AEDs that are enzyme p450 inducers and 13 participants were taking Lamotrigine.

We found no statistically significant difference in demographic background among the women who received the interventional handout and participants that received usual care, as shown in Table 1. The two groups also had similar rates of any contraceptive use, with 10 in the control group and 9 in the intervention group using contraceptives.

Table 1.

Participant characteristics.

Characteristic	Usual care n=22	Handout n=20	Total n=42	p Value
Age: year, mean (S.D.)	32.1 (5.96)	33 (7.27)	32.5 (6.55)	.86
Body mass index:   kg/m2 (S.D.)	26.3 (7.84)	25.7 (5.66)	26.1 (6.81)	.99
Primary language				1.00
English	19 (86%)	18 (90%)	37 (88%)
Spanish	1 (5%)	1 (5%)	2 (5%)
Other	2 (9%)	1 (5%)	3 (7%)
Race				.73
White	11 (50%)	10 (50%)	21 (50%)
Black	2 (9%)	0	2 (5%)
Hispanic	5 (23%)	6 (30%)	11 (26%)
Asian/Pacific Islander	3 (14%)	3 (15%)	6 (14%)
Other/don’t know	1 (5%)	1 (5%)	2 (4.8%)
Degree level				.91
High school diploma	10 (46%)	9 (45%)	19 (45%)
Bachelor’s degree	9 (41%)	7 (35%)	16 (38%)
Master’s degree	2 (9%)	3 (15%)	5 (12%)
Other/not sure	1 (5%)	1 (5%)	2 (4%)

We found no statistically significant difference in baseline level of knowledge between the educational handout group and control group, with median pretest scores of 1 (IQR: 0–2) in both groups (Wilcoxon rank sum test, p=.61). The posttest after the neurology visit, as shown in Table 2, demonstrates a significant increase in knowledge scores for the educational handout group (median score, 5; IQR: 3.75–6.25), which was significantly higher than the increase in the control group (median score, 1; IQR: 0.25–3) (p<.001). Test scores in the educational handout group increased significantly as compared to the control group (mean increase of 3.65 vs. 0.68, respectively, p<.001). Changes in test scores and expected gain in test scores did not vary by recruitment sites (p=.24). We found no statistically significant difference in use of contraception and counseling about planning for a healthy pregnancy between the two arms. Only half of all participants received counseling about ways to prepare for a healthy pregnancy, and 24% of participants never received counseling on the drug interaction between AEDs and hormonal contraceptives. Of the 19 women using contraception, male condoms were the most commonly used method (32%), with combined oral contraceptives following at 21%. A total of 29% of these women were using some form of highly effective contraceptive method [n=1 male sterilization, 1 implant, 2 levonorgestrel intrauterine device (IUD), 2 copper IUD].

Table 2.

Comparison of test scores.

	Usual care (N=22) mean (S.D.)	Handout (N=20) mean (S.D.)	p Value
Pretest score	1.1 (1.32)	1.3 (1.16)	.61a
Posttest score	1.8 (1.65)	4.9 (1.94)	<.001a
Pretest to posttest gain	0.7 (1.73)	3.7 (1.93)	<.001a

^aWilcoxon test.

None of the measures in the postintervention contraception questionnaire were significantly different between the educational and usual care groups, possibly due to the small sample size of this study. Across contraception users, the decision to continue or change their contraceptive method was influenced by information about drug interaction (61%), possible negative effects of certain AEDs on pregnancy (47%) and effectiveness of method (47%). Overall, among all participants, we found a desire to learn more about interactions between hormonal contraceptives and AEDs (62%). We found no statistically significant difference in the desire to learn more the drug interactions of hormonal contraceptives and AEDS among the control arm and intervention arm of the study (p=.90). Half of participants (55%) indicated plans to speak to a doctor (gynecologist, neurologist, primary care or combination of clinicians) about their contraceptive method at a future appointment. Participants were as likely to want to discuss contraceptive options with their neurologist (19%) as with their gynecologist (22%).

4. Discussion

Women with epilepsy and their healthcare providers may benefit from having more educational tools regarding contraception and interactions with AEDs. This pilot study demonstrates that a simple intervention, an educational handout, has the ability to improve short-term patient knowledge about contraceptive options and drug interactions with AEDs. An increase in desire to be educated on these drug interactions was reported by 62% of participants demonstrating the value this information has for women with epilepsy. Furthermore, it is important that tools are developed by collaborative efforts between neurologists and gynecologists and that both fields are well versed in these drug interactions. The results of this study will inform patient education efforts and can potentially increase the number of women following recommended guidelines for effective of contraceptive methods. If patients with epilepsy on AEDs use highly effective contraception, then there may be a decreased risk of unintended pregnancies and exposure to teratogenic effects of AEDs.

A major strength of this study is that it is a collaborative effort between neurologists and family planning specialists. In addition, this study utilizes questions from prior epilepsy and contraception study by Pack et al. [1]. A weakness of this study is that the pretest and the posttest had very low scores indicating that the test may have been too difficult. In addition, the neurologists were not blinded to knowledge of who received the educational handout. It is possible that the discussion with the neurologist became part of the intervention. Another weakness of this study is that we only included English-speaking participants. Therefore, the results are not generalizable to a non-English-speaking population.

Our findings suggest that an educational handout on drug interactions and contraceptive options can increase short-term information retention on this topic among reproductive-age female epilepsy patients. A future study could utilize the educational handout in different study populations such as medical students, pharmacy students, obstetrics and gynecology residents and neurology residents. This study focuses on knowledge and contraceptive intentions only immediately after the intervention. It may be informative to assess knowledge attainment more remotely from the intervention and to determine if gains in knowledge contribute to change in contraceptive behaviors. This study also highlights the feasibility of conducting a randomized controlled trial testing an educational intervention within a neurology clinic to improve epilepsy patient knowledge about drug interactions and contraceptive options.