Our study provides important information on the main causes of non-adherence to antiepileptic drugs in poor communities. Firstly, we found that adherence to treatment is not influenced by gender, age, and duration of treatment. This is consistent with the results from the previous studies [8]. The rate of nonadherence is estimated to be between 30-80% of patients depending on the patient population being analyzed [9].In agreement, we documented AED(s) non-adherence among 58% of the children being treated in 2014, but without any interventional local community childhood epilepsy services [3]. Most of the barriers were effectively removed by provision of free AED(s) and pediatric neurology consultations. We did this by integration and implementation of childhood epilepsy services into primary health care: Rukhsana Shafqat Memorial Primary Health care Center. The integration of CE into this private primary care was done by mutual consent of the owners of the center and visiting TDBUCEP team from Lahore. We documented that the cohort of CWE, being followed at this community center was adherent 80% to AEDs, as compared to the 47.5% of the cohort not being followed at the center. The Epilepsy Centre receives no government funding and is largely funded from private sources, mainly philanthropic donations, and charity. Similarly Tang et al. documented that with community involvement, the number of patients with epilepsy who missed AEDs decreased to 45.0% from 64.3% [10]. Other previous randomized intervention trials demonstrated that there was no consensus about enhancing compliance through various measures [11]. However, others showed that patient reminders such as reminders mailed together with a counseling leaflet have produced promising effects on adherence. In comparison, our simple package of intervention techniques including free monthly camps for CWE, education, reminders, provision of free AEDs and CE consultations proved to be effective in enhancing AED adherence. Similar to other studies, patients may have overestimated their adherence in their self-reported data in our cohort being managed at TDBUCEPC Bhakhar [12].

We found that our current cohort of CWE who were not being followed at TDBUCEPC and were affording all the expenses associated with epilepsy treatment from their own pocket still had AEDs nonadherence in 62.5%, which was documented 58%[3] in the same settings in 2014 by self-reported adherence among CWE: children with epilepsy whose parents were buying AEDs were about 4 times more likely as to be non-adherent as compared with those who were getting their AEDs free of charge from TDBUCEPC Bhakkar. Among the non-adherent CWE in this cohort, 28(56%) were so due to purely/predominantly financial reasons. This is exactly what has been reported by Faris et al. and Snodgrass et al., respectively [13,14]. The cost of drugs limits caregivers from purchasing them in case they are not available free or at a very low cost. Similarly Nazziwa et al. reported that the commonest reason given for missing drugs was lack of drugs due to their high cost as reported by 36(48.7%) of study participants comprising of children and adolescents [15]. Similar findings have been reported from various parts of the world with limited resources, allocated for epilepsy care [16]. Thus, the treatment of childhood epilepsy in outreach districts in developing countries, like Pakistan, remains far from satisfactory, mainly because of the non-availability of medications [17,18].

Although it is widely accepted that nonadherence leads to poor seizure control, our findings also suggest that uncontrolled seizures may, paradoxically, contribute to nonadherence. As 10(31.5%) of the non-adherent patient in our study group discontinue the drugs because there was inadequate seizure control, even in 12% of the cohort not being followed at the community center. In agreement, Jones et al. [19] reported a strong correlation between poor seizure control and non-adherence with 63% of those with poorly-controlled epilepsy being non-adherent as compared to 38% of the well-controlled group.

When most of the barriers are removed, misguidance by local health healers was the predominant cause of non-adherence among 8(25%) of the CWE being followed in TDBUCEPC as compared to 6(12%) of the cohort not being followed at the center. This is similar to the worldwide believe that supernatural powers and spirits influence human being and affect their wellbeing and health especially in epilepsy [20]. Though the no of non-adherent CWE cohort being followed at the center was small, about 08.3% of these prescribed AED(s) were not available at the center pharmacy, where these findings were prevalent among 10% of the cohort not following at this center, so both these cohorts were non-adherent to their prescribed medication. We found that in 08.3% of the patients being followed at the center and 10% not following at the center, the major contributing factor was poor or no counseling by the treating physician.

All this indicates that Epilepsy care should be multidisciplinary and long term, linking primary and secondary care, and empowering patients towards improved management of their condition. Such programs have been successful in improving the diagnosis and management of adults with epilepsy in communities of India, Zimbabwe and Ethiopia [21,22].