Correlates of Seizure Control Among Patients with Epilepsy at Two Referral Hospitals in Zambia

Background and Objective: Epilepsy is a neurological condition that has proven to be a major public health challenge worldwide, including Zambia. Despite the availability of many treatment alternatives, people with epilepsy still experience seizures. There is paucity of information on the control of seizures and associated factors in Zambia. The purpose of this study was to examine the factors associated with seizure control in people with epilepsy in two tertiary hospitals in Zambia. Methods: A cross-sectional study was conducted among epileptic patients at Kitwe and Ndola Teaching Hospitals, two referral hospitals in Northern Zambia, from September 2 to September 23, 2019. A multivariate logistic regression analysis was performed using SPSS to identify independent predictors of seizure control. Adjusted odds ratios (AOR) and their 95% confidence intervals are reported. Results: A total of 220 epileptic patients were enrolled in the study. Of these, 23.6% had controlled seizures, while 76.4% had uncontrolled seizures. Occupation (p value=0.018), level of education (p value<0.001), clinic attendance (p value<0.001), drug adherence (p value<0.001), beliefs about drugs (p value=0.006), and having friends (p value=0.025), were significantly associated with seizure control. A higher level of education (AOR:0.243, CI:0.080-0.738), regular clinic attendance (AOR:8.578, CI:3.327-22.112) and age at first seizure of 3-6 years (AOR: 1.035, CI:0.240-4.454) were predictors of controlled seizures. Conclusion and Implications for Translation: It was shown that there was a low prevalence of controlled seizures among the study population. However, in this study, controlled seizure can be associated with a positive response to medication, regular clinic attendance, employment, having friends, having a higher level of education, and being 3-6 years old when a first seizure occurs. Risk factors should be taken into consideration such as alcohol consumption and drug compliance to better manage seizures.


Introduction
Epilepsy is defined as a condition of recurrent spontaneous seizures due to a primary disturbance of brain function. 1,2It is not regarded as a single condition, but a broad spectrum of disorders that increase an individual's predisposition to seizures.It is characterized by history of at least one seizure, causing a change in the brain that increases the likelihood of future seizures and associated cognitive, psychological or social disturbance. 3pilepsy is known to affect individuals of all ages and across all socioeconomic classes worldwide. 4It is estimated that approximately 50 million people have epilepsy globally, with an estimated proportion with active epilepsy of 6.38 per 1,000 individuals and an annual cumulative incidence of epilepsy of 67.77 per 100,000 individuals.A total of 80% of those affected with epilepsy live in developing countries. 5ople with epilepsy have a high rate of suicide, depression, accidental and unexplained deaths. 6espite being on medication, seizures still persist in an estimated 20-35% of cases, 7 justifying the need to identify factors that can be modified and hence lead to improved seizure control if targeted effectively.The control of seizures can be influenced by several factors such as compliance to medication, adjustments in dosage, and complexity of the drug regimens, which can affect adherence. 8Compliance to Anti-Epileptic Drugs (AEDs) can help control seizures.There are still high rates of noncompliance to AEDs, which is associated with factors such as the patients' belief about the medication, side effects, illiteracy, poverty, ignorance, poorly developed health services, duration of the disease, cost, medical barriers, and the patients' understanding of the disease. 9On the other hand, well-informed patients tend to have seizure control as they are more able to avoid seizure precipitants, have AED compliance, and visit the clinic regularly.However, even with regular clinic attendance and AED compliance, about 20% to 30% of patients continue to have uncontrolled seizures. 10 developing countries, many predisposing factors to epilepsy have been reported. 11-13However, failure of healthcare givers to identify and control these factors has been a major reason for failure of many epilepsy treatment programs to achieve seizure control in patients. 6Many patients believe that their seizures are triggered by external or internal factors such as physical exhaustion, psychological stress, sleep deprivation, prolonged fasting, premature awakening, playing musical instruments, excessive intake of alcohol, and menstruation. 1Evaluation of treatment outcomes and their predictors is important in developing treatment optimization strategies. 14The epilepsy treatment gap is defined as the proportion of people with epilepsy who require treatment but do not receive it. 15Despite the availability of Anti-Epileptic Drugs, a huge portion of people living with epilepsy in many limited resources countries do not receive treatment.The treatment gap is over 75% in low-income countries, with a huge difference in the care of epilepsy patients between high-and low-income countries. 16Epilepsy confers a social and economic burden in Zambia and remains a public health concern. 17r study investigated factors associated with seizure control in people with epilepsy at Kitwe Teaching Hospital (KTH) and Ndola Teaching Hospital (NTH), two tertiary hospitals in Zambia, as to our knowledge, this has not been previously studied.

Study Setting
The study was conducted at KTH, as well as NTH, on the Copperbelt Province in Zambia.NTH has a bed capacity of 851 patients, and is the second largest hospital in Zambia, with a catchment population of 503,649 people.KTH has a bed capacity of 664 patients, and a daily patient population of 1,300.These hospitals have psychiatry sections where psychiatric and epileptic patients attend.The psychiatry department at KTH operates on an outpatient basis, while NTH has provision for admission; each department attends to approximately 50 patients a day.The two tertiary hospitals serve catchment populations that are similar, with some patients being referred between the two institutions.

Study Design and Time Frame
This was a hospital-based cross-sectional study using a structured questionnaire.The study was conducted www.ijtmrph.orgfrom September 2 to September 23, 2019.Seizure control was the dependent variable, and was graded as controlled if a seizure was experienced a year or more ago.Any seizure in less than a year was defined as an uncontrolled seizure.Independent variables investigated included age, gender, level of education, knowledge level, clinic attendance and years at first seizure.

Sample Size and Sampling
The sample size was not calculated as the target population was limited.A non-probability convenience sampling method was used in which participants were selected based on their willingness and availability to participate in the study.The study included both in-patients and out-patients attending the Kitwe and Ndola Teaching Hospital psychiatry departments.All participants were confirmed epileptic patients through a prior diagnosis obtained from patient files.All participants were above 18 years old, had regular follow-ups for at least one year, and were on at least one AED.Primary caregivers were also included in the study.Epileptic patients with less than a year of followup, less than 18 years of age, and those who were not attended to at either KTH and NTH were excluded.Non-epileptic patients were also excluded.

Data Collection
All patients who met the inclusion criteria were included in the study following the signing of the informed consent.Data were collected through the administration of a questionnaire.For participants who could not understand English, the questionnaire was administered in the local language of their choice.
The questionnaire had seven parts which included socio-demographic information, knowledge and attitude towards epilepsy, frequency of seizures and seizure precipitants, availability of anti-epileptic drugs, accessibility to anti-epileptic drugs, compliance to anti-epileptic drugs, and the patient's belief on anti-epileptic drugs.The patient's knowledge level of epilepsy was scored out of 4 questions, with each correct answer being awarded a 1 whereas a wrong answer or no answer was awarded a 0. The scores were graded as follows: 0-2 poor knowledge, and 3-4 proper knowledge of epilepsy.The patient's attitude level towards having epilepsy was scored out of 6 questions, with each correct answer being awarded a 1 whereas a wrong answer or no answer was awarded a 0. The scores were graded as follows: 0-3 poor attitude, and 4-6 proper attitude.Drug adherence and belief was measured by querying compliance, and was defined as irregular if one or more doses were missed, and as regular if no doses were missed.Additional questions were asked on the actual doses missed and possible reasons for the irregular drug compliance in those reporting missing some doses.
Seizure control was defined based on the frequency of a patient's seizures.If a patient experienced either daily, weekly, or monthly seizures, their seizure control was denoted as uncontrolled seizure.A patient's seizure control was denoted as controlled seizure when the patient did not have a seizure for a year or for more than year.Controlled epileptic seizures were regarded as seizures that had stopped completely.Seizures which continue, even after trying medication, are uncontrolled seizures, even if they occur once a year. 18

Data Analysis
Data were processed and analyzed by the Statistical Package for Social Sciences version 21 (IBM Corp, 2012).Frequency distributions were used for demographic characteristics.Descriptive cross tabulation analysis was done on the selected variables.A stepwise forward multivariable binary logistic regression model was used to determine correlates of seizure control.A threshold of p<0.1 in cross tabulations was used to select variables to include in the model.A p-value of <0.05 was considered statistically significant.

Ethical Approval
Ethical approval was obtained from the Tropical Diseases Research Center (TDRC) ethics committee (IRB number 00002911).

Socio-demographic Characteristics of Study Participants
We enrolled 220 epileptic patients wiht a history of epileptic seizures in the study.More than half of www.ijtmrph.orgthe participants were male (58.2%).The participants' age range was 18 to 65 years old, with the mean age being 28.7 years old and standard deviation of 9.9.years.A total of 94 (42.7%) of the participants were aged 18-24 years old while 24 (10.9%) were aged between 30-34 years old.The study showed that 80% of the participants were single, 15.5% were married, 3.6% were divorced, and 0.9% were widowed.A total of 71.4% had no children, and only 12.3% of the participants were employed.It was also found that 13.2% of participants had not attained any education level at all, while less than half (40.5%) had attained primary education.About a quarter of the participants (23.6%) had controlled seizures (Table 1).

Drug Adherence and Patient Belief
More than half (54.5%) of the participants had proper medication adherence, 26.4% missed 2 to 3 doses, and 14.5% missed more than 3 doses; 13.6% of the participants that missed their medication reported that it was due to them feeling well, and the other 13.6% reported that it was due to them not being able to afford their medicine (Table 2).

Factors Associated with Seizure Control
On univariate analysis, employment (p value=0.02),higher level of education (p value<0.00),regular clinic attendance (p value<0.001),drug compliance (p value<0.001),positive patient belief about AEDs (p value=0.01),and having friends (p value=0.03),were significantly associated with controlled seizures.In contrast, drinking alcohol (p value = 0.12), change in drugs (p value=0.21),and the distance to the hospital (p value=0.36),were not significantly associated with seizure control (Table 3).
We performed a multivariate logistic regression analysis to ascertain any association between the dependent variable and independent variables.Those with a higher level of education [adjusted Odds Ratio (aOR):0.24,confidence interval (CI):0.08-0.74)],regular clinic attendance (aOR:8.58,CI:3.33-22.11),and were aged 3-6 years old when they experienced their first seizure (aOR: 1.04, CI: 0.24-4.45),were more likely to have controlled seizures (Table 4).

Discussion
Our findings showed that out of a total of 220 patients, only 23.6% had controlled seizures; this is consistent with literature.Many studies indicate that alcohol causes an increased risk of epileptic seizures and that it is a predictor of uncontrolled seizures. 14,19Alcohol intake may affect a person's adherence to treatment, as they may forget to take their medication, or prefer not to buy their medicine in preference for alcohol. 20Contrary to these prior findings, the present study found that frequent alcohol intake did not have a significant effect on the control of seizures.This variance in findings could be due to the differences in sampling within the study having few participants who consumed alcohol.
Poor drug compliance was significantly associated with uncontrolled seizures.This is reported in other studies which suggest that noncompliance to epilepsy medications can interfere with treatment and may affect the clinical outcome altogether. 10,21,22This study found that the number of drugs that the patient had to take and the number of times drugs were taken in a day, affected adherence, which also agreed with prior studies. 23here were more seizures experienced in patients with negative beliefs towards medication, than those with positive beliefs.This is consistent with findings by other studies. 14,24tients who had proper knowledge about their condition had more controlled seizures than those with poor knowledge.Similarly, among those with a proper attitude were more prone to experiencing controlled seizures than those with poor attitude, although the association was not statistically significant.This can be attributed to the fact that those with proper knowledge would have a better attitude of adherence to their medication and a better understanding of how to take care of themselves.However, there are no studies indicating whether increased knowledge improves self-care. 25ose who received in secondary or tertiary education, and those who were employed, were more likely to have controlled seizures.This finding agrees with prior literature. 10r findings showed that a higher level of education, regular clinic attendance, and were aged 3-6 years old when they experienced their first seizure, were predictors of controlled seizures.Those who had their first seizure between ages 13-  19 years were less likely to have controlled seizures to this is in contrast to findings from a previous study where individuals who were above 20 years of age had better controlled seizures. 10olated patients have been known to experience more seizures than those with friends, colleagues or family. 26This is in agreement with the present study where it was found report that having friends was significantly associated with controlled seizures when compared to those patients who did not have any friends.This could be because peer support may help the patients feel better about themselves, which is important in the treatment process.

Limitations
Even though the study presented important findings in the northern region of Zambia, it precludes other populations in the southern region of the country.This affects the generalizability of the findings to the entire country.The non-probability nature of our sampled population, and the limited period for data collection could have possible limitations on the generalizability of our findings.The study did not stratify participants based on in-patient status or outpatient status.This could affect findings as in-patients are more likely to be supervised during drug administration which may impact treatment outcomes.

Recommendations
Patients should be educated about their condition, the importance of their drugs, compliance and clinic attendance.Close associates, family and friends should be encouraged to be supportive of the patient and their condition altogether.Furthermore, the patients should be encouraged to engage in interactive activities.There is a need for more studies on seizure control in other regions of the country.

Conclusion and Implications for Translation
Our findings show that less than half of the epileptic patients had controlled seizures.Positive beliefs towards epilepsy medication, employment, and having friends, were significantly associated with controlled seizures.Higher level of education, regular