Self-reported medication adherence to antiepileptic drugs and treatment satisfaction among paediatric patients having epilepsy: A cross sectional study from the Indian subcontinent

Introduction: Non adherence to antiepileptic drugs (AEDs) can vary the course of epilepsy. There is a dearth of data regarding adherence status among paediatric epilepsy patients. Objectives: To determine the level of AED adherence among paediatric patients having epilepsy and to assess the relationship between AED adherence and treatment satisfaction among paediatric epileptics. Method: A cross sectional study was conducted at a tertiary level referral hospital in northern India from July 2015 to June 2016 on children aged 2 to 13 years of age, diagnosed with epilepsy and treatment initiated with at least 1 AED for the past 6 months. Data collection was done using a questionnaire which was divided into three parts. The first part comprised demographical information about patients and their caregivers along with clinical details of epilepsy. The second part gathered information about medication adherence profile with the help of the Morisky Medication Adherence Scale-8 (MMAS-8). The third part gathered data on treatment satisfaction profile about these patients with the help of Treatment Satisfaction Questionnaire for Medication (TSQM 1.4). The estimated sample size was at least 61 patients. Data was analysed using Graph Pad InStat 6.0. Ethical approval was obtained from the institutional ethics committee. Results: A convenient sample of 112 patients with their respective immediate caregivers was enrolled in this study. The mean age of children was 9.82±2.17 years and male to female sex ratio was 1.24. Fifty (44.6%) patients had low adherence, 29 (25.9%) medium adherence and 33 (29.5%) high _________________________________________ Shri Ram Murti Smarak Institute of Medical Sciences, India *Correspondence: drpratikg@rediffmail.com (Received on 02 June 2017: Accepted after revision on 28 July 2017) The authors declare that there are no conflicts of interest Personal funding was used for the project. Open Access Article published under the Creative Commons Attribution CC-BY License adherence to AEDs as per MMAS-8 questionnaire. Altogether 79 (70.5%) children were non adherent when dichotomised MMAS scores were considered. Age of children, duration of epilepsy, mean number of family members, frequency of drug ingestion by patient, employment status of caregiver were significantly associated with nonadherence in epileptic children. Children adhering to medication prescription had significantly higher satisfaction in the effectiveness, convenience and global satisfaction domains of TSQM. Conclusions: Twenty nine percent of epileptic children showed medication adherence to AEDs in the present study. Increasing age of children, increased duration of epilepsy, increasing mean number of family members, patients taking medications more than once per day and employed caregiver status were significantly associated with non-adherence in epileptic children. Satisfaction scores were higher in medication adherent children, particularly in the effectiveness, convenience and global satisfaction domains but not in the side effects domain. DOI: http://dx.doi.org/10.4038/sljch.v47i2.8478 (


Introduction
In India, prevalence of children with epilepsy varies from 2 to 22 per 1,000 1 . However, there is a significant treatment gap, especially in low income countries 2 . Non adherence to antiepileptic drugs (AEDs) has been reported among 55% of paediatric and 70% of adult patients 3 . Improving paediatric adherence to treatment is especially important as there is evidence that complying with treatment improves the health outcomes of children more than that of adults 4 . AED adherence can be measured by self-reporting, drug level monitoring or prescription refill monitoring 5 . Treatment satisfaction can affect a patient's health-related decision making 5 . Measures have been developed for assessing treatment satisfaction 6,7 . A review of the literature did not reveal any studies from the Indian subcontinent which measured the relationship between adherence to drug treatment and treatment satisfaction among paediatric epileptic patients.

Objectives
To determine the level of AED adherence among paediatric patients having epilepsy and to assess the relationship between AED adherence and treatment satisfaction among paediatric epileptics.

Method
A cross sectional descriptive study was conducted from July 2015 to June 2016 in the paediatric outpatient department of a tertiary level care teaching hospital in northern India on children aged 2 to 13 years of age, diagnosed with epilepsy and treatment initiated with at least 1 AED for past 6 months. Approval was obtained from the institutional ethics committee for this study. Data was collected from the primary caregiver of epileptic children. The following inclusion/ exclusion criteria were used:  Absence of developmental disorders or comorbid chronic illnesses in children requiring daily medications.
 Absence of any other sibling or another family member afflicted by chronic diseases or epilepsy which entails daily administration of drugs.
 Caregiver willing to give informed verbal consent for inclusion in the study.
 The respondent had to be literate and able to read and write Hindi, the local language.
 The child should be having normal neurological and cognitive development.
Age selection was based on the expectation that by the age of 13 years self-care responsibilities would be assumed by adolescents and hence the role of the primary care-giver will diminish. Epilepsy was arbitrarily defined as "well controlled" if the patient reported no seizures in the last 3 months and was defined as "poorly controlled" if they reported having had at least one seizure in the last 3 months 5 .  [12][13][14]. The TSQM 1.4 domain scores were calculated as recommended by the instrument's authors and described in detail elsewhere 6,11,12 . TSQM 1.4 domain scores range from 0 to 100, with higher scores representing higher satisfaction in that domain. TSQM provides a unique opportunity to compare various medications used to treat a particular illness on the three primary dimensions of treatment satisfaction (effectiveness, side effects, convenience), as well as patients' overall rating of global satisfaction based on the relative importance of these primary dimensions to patients 7 .
Sample size: This was calculated based on a reported adherence of approximately 50% as explained elsewhere 5 . With z=1.96 for a 95% confidence interval and a total width of confidence interval of 25%, the estimated sample size explained elsewhere was at least 61 patients 5 .
Data analysis: Graph Pad InStat 6.0 was used for data analysis. Qualitative variables were expressed as frequency and percentages and continuous variables were expressed as mean ± standard deviations. Fisher's exact test was used for determining factors associated with adherence. Student t-test was used for comparing two groups. Pearson correlation was used for multivariate analysis. The conventional 5% significance level was used throughout the study.

Results
From a total convenience sample of 150 patients who were clinically examined during the study period, only 132 patients fulfilled the inclusion criteria. Of these only 112 patients agreed to participate in the study, a response rate of 85%. Gender-wise characteristics of the study population are shown in Table 1.  Table 2) significantly affecting adherence adversely were increasing age, higher duration of epilepsy, family history of epilepsy, increasing family size (more than 5 persons in a family), history of at least 1 seizure in the last 3 months, frequency of drug usage more than once a day and employment status of the caregiver.  Table 3. Patients who were adherent had significantly more mean scores in the effectiveness and convenience domains of TSQM.

Discussion
To the best of our knowledge, the relationship between medication adherence and treatment satisfaction among paediatric epileptic patients has not been reported before from the Indian subcontinent. Epidemiological data from India shows a higher prevalence of epilepsy among males as compared to females 13 . Mani et al 14

and
Banerjee et al 15 have reported a peak incidence of epilepsy during early childhood and/or first decade of life. According to Indian community based clinical surveys, most epilepsies are idiopathic or cryptogenic 16 . Udani V also suggested that more than half of children had symptomatic or cryptogenic localization related epilepsies 16 . Asadi-Pooya AA 17 from Iran reported idiopathic epilepsy in 78.5% of paediatric and adolescent epileptics, which is similar to our present findings. Udani V found that generalised epilepsy was twice as common as partial epilepsies 16 . In the present study, too, the majority of children had generalized epilepsy. Similar data was also reported by Asadi-Pooya AA 17 .
The most commonly prescribed drug in the paediatric age group is sodium valproate, according to a single centre study conducted by Bhatt et al in India 18 . Another study from India reported the treatment of focal seizures mainly with carbamazepine in children 19 . Valproate is the most common drug reported for generalized tonic-clonic seizures in few recent studies from India 20 . Majority of paediatric patients were prescribed a single AED in past studies, from India and elsewhere, just like the findings of the present study 17,20 . A positive family history was seen in more than half our patients. A population-based, case control study in India reported family history of the epilepsy as a strong independent predictor of epilepsy 13 .
Less than 1/3 rd (29.45%) of children reported 'high' adherence in the present study and more than 2/3 rd (70.54%) children in the present study did not adhere to their respective antiepileptic regimens. Qoul et al reported high adherence in 29.4% of their paediatric study population 21 . Though Nazziwa et al reported adherence in 79.5% children by self-reporting, this high percentage dropped down to 22.1% when correlated with serum drug levels 22 . Asadi-Pooya AA reported satisfactory drug compliance among 72.4% of study patients which comprised of both adolescents and children 18 . Non adherence may be the most important cause of poorly controlled epilepsy 23 . Seventy percent of patients reported antiepileptic drug omission 5 . Non adherence rates ranging from 12% to as high as 64% have been reported in earlier studies conducted among adolescents and/or adults 5,24,25 . As greater than 95% adherence may be needed to adequately suppress the epileptic seizures, missing a dose or two weekly may suffice to cause failure of therapy and trigger seizures 26 . Thus, epileptic patients should be classified as either adherent or non-adherent 23 . Classifying patients as having high, moderate and low adherence should be discouraged. Hence, in the present study while high adherence score was classified as adherent, moderate and low adherence score on MMAS were clubbed as non-adherent as reported earlier 5 .
Despite India being a male dominant society, our study revealed no statistical significant difference among genders for their number, age, duration of epilepsy, mean family members, seizure frequency and adherence rates. This is agreement with studies by Gabr et al 23  In the present study medication adherence was significantly and adversely affected by increasing age of children and duration of epilepsy. Kyngas et al reported that compliance of adolescents with epilepsy for 1-3 years was greater than the compliance of adolescents with epilepsy for more than three years 28 . However, the same study ruled out age of the patient as an independent factor significantly associated with compliance 28 . Other studies failed to report any similar association 17,22,23 . Our present study and studies by Asadi-Pooya AA 17  In the present study, adherence to AED was adversely affected if one or more seizures occurred in the last 3 months compared to children who had no seizures in the last 3 months. Similar findings were reported by Gabr et al 23 . Kyngas et al reported that adherence was adversely affected if seizures occurred daily compared to less than weekly 28 . In our study, there was no correlation between adherence and seizure frequency. In our study, the number of tablets ingested was not significantly associated with adherence to AEDs. This was similar to findings in studies by Asaadi-Pooya et al 17  The present study showed that children were more likely to be adherent to AEDs if all the AEDs were given at the same time, i.e. once a day, even if more than one AED was used. Asadi-Pooya AA found that patients prescribed a nocturnal dose of phenobarbitone were more compliant than those prescribed carbamazepine or valproic acid twice or thrice daily 17 . Chances of non-adherence increased if the immediate caregiver was employed. Nazziwa et al has reported increased likelihood of noncompliance in a child if caregiver had an occupation 22 . Gender of caregiver was not differentiated in the present study as all the caregivers were females, either mother or grandmother.
Our study demonstrates that satisfaction scores were higher in medication adherent children, particularly in the effectiveness, convenience and global satisfaction domains but not in the side effects domain. In other words, if patients are more satisfied with the treatment then they are more likely to adhere to the medication regimen. Patients on one drug demonstrated significantly more satisfaction in the effectiveness and convenience domains compared with patients on two or more drugs. Patients on monotherapy find it more convenient to take a single medicine. Sweileh et al also reported greater satisfaction in the effectiveness domain among adolescents receiving monotherapy 5 . However, the overall global satisfaction scores were not significantly different and this supports our findings that the adherence status of epileptic children was not affected by the number of drugs.
There are some limitations in this study. This is a single centre study and hence the results cannot be extrapolated to other geographical areas. Our hospital is a tertiary care referral centre for the nearby region of approximately 100 km radius. Hence, patients usually have 'difficult to treat' or 'refractory' diseases. Though the translated version of MMAS used in the present study was checked and found to have an acceptable internal consistency, the Hindi translated and validated version of the same is not available. We relied on interviews with caregivers to measure adherence and this may be inaccurate and biased. Self-report on medication adherence is prone to recall bias and respondents usually overestimate adherence 17 . Serum AED estimation could not be done due to financial constraints. However, single essay for blood drug levels may not reflect variations in drug levels by the day and hence is not acceptable as a full-proof method for medication adherence studies 22 . Further research is warranted to ascertain the specific adherence barriers pertaining to epileptic children from this part of the world.

Conclusions
 Twenty nine percent of epileptic children showed medication adherence to AEDs in the present study.
 Increasing age of children, increased duration of epilepsy, increasing mean number of family members, patients taking medications more than once per day and employed caregiver status were significantly associated with nonadherence in epileptic children.
 Satisfaction scores were higher in medication adherent children, in the effectiveness, convenience and global satisfaction domains but not in the side effects domain.