Three hundred subjects (150 PLWE and 150 accompanying healthy NC’s), participated in the study. As shown in Table 1 below, there was no statistically significant (p > 0.05) difference between the two groups in age, religion, area of residence, household size, children had and gender. The PLWE however showed a statistically significantly lower level of education (p < 0.001), annual income (t = −4.552, p <0.001), and MMSE score (t = −5.212, p < 0.001) as compared to the NC’s. They also had unskilled employment (p = 0.041), with the majority of them being unemployed (p < 0.001) and unmarried (p < 0.001), as compared to the NCs.
Among the 150 PLWE who participated in this study, 74.7% had generalized seizures, 23.3% partial, whereas only 2% had unclassified seizures. A total of 37.3% of the PLWE reported using alternative modes of therapy. The commonly used alternative modes of therapy included prayers (21.3%), herbs (12.7%), and witchcraft (2.7%). Majority, (92.7%), of the PLWE in this study reported improvement as a result of AED treatment with 83.3% of them even reporting expectations of being cured. Over half (53.4%) of the unemployed PLWE in this study blamed their illness (epilepsy) as the reason for their unemployment while 50.7% of the PLWE reported not knowing the cause of their illness.
The factors that were found to be statistically significantly associated with a higher seizure burden included use of poly AED’s therapy (x2 = 19.406, p < 0.001), being unmarried (x2 = 8.593, p = 0.035), use of alternative therapy (x2 = 8.585, p = 0.035), low annual income (f = 3.161, p = 0.027), low MMSE scores (f = 4.029, p = 0.009), a longer duration of illness (f = 3.392, p = 0.020), a past history of head injury (f = 3.117, p = 0.026) and an earlier age at onset of epilepsy (f = 5.633, p = 0.001).
Out of the 300 participants in this study, 285 (150 NC’s and 135 PLWE) satisfied the inclusion criteria (MMSE score of ≥ 22) for the self- administered WHOQOL-BREF questionnaire.
The mean QOL among PLWE (49.90%), t = −17.694, p < 0.01, at K.N.H was statistically significantly lower than that of the NC’s (77.60%), t = −18.298, p < 0.01, accompanying them and also statistically significantly impaired, t = −18.298, p < 0.01, as compared to the hypothesized mean of 75±2.5% .
Further analysis of the data was carried out to determine the difference in mean domain and facet QOL scores between PLWE and NC’s using the independent samples test (t-test) at 99% confidence interval. Figures 1, 2, 3, 4, 5 illustrate these findings.
Mean domain scores as determinants of mean QOL. X axis represents the four QOL domains using the WHOQOL-BREF. Y axis represents the mean domain scores (%).
Facets determining the physical health QOL domain. X axis represents the seven facets that comprise the physical health QOL domain. Y axis represents the mean facet score (0 to 5).
Facets determining the psychological QOL domain. X axis represents the six facets that comprise the psychological QOL domain. Y axis represents the mean facet score (0 to 5).
Facets determining the social relationships QOL domain. X axis represents the three facets that comprise the social relationships QOL domain. Y axis represents the mean facet score (0 to 5).
Facets determining the environmental QOL domain. X axis represents the eight facets that comprise the environmental QOL domain. Y axis represents the mean facet score (0 to 5).
As shown in Figure 1, the mean QOL scores for each of the four domains used in the WHOQOL-BREF i.e. physical health QOL (t = − 19.859), psychological QOL (t = −18.698), social relationships QOL (t = −9.934) and environmental QOL (t = −9.934) were all statistically significantly (p < 0.01) lower for PLWE as compared to the NC’s.
As illustrated in Figures 2, 3, 4, 5 above, all the mean facet QOL scores were also statistically significantly (p < 0.01) lower in PLWE as compared to the NC’s, apart from the subjective evaluation on financial resources. Both groups reported not having enough money to meet their needs, t = 0.489, p = 0.625.
As shown in Table 2 below, the factors statistically significantly associated with impairment of QOL among PLWE were a low level of education (p < 0.001), higher seizure burden (p < 0.001), low annual income (p = 0.007), unemployment (p = 0.004), unskilled employment (p < 0.001), and rural residence (p = 0.009). Additionally, those PLWE who reported financial difficulties as the reason for non-compliance to treatment (p = 0.037) and those who blamed their illness (epilepsy) as the cause of their unemployment (p < 0.001) also showed statistically significantly impaired QOL.
Mean QOL and various variables of PLWE
There was no statistically significant (p > 0.05) relationship between mean QOL of PLWE and gender, marital status, age, children had, household size, mode and specific type of drug therapy, seizure type, age at onset of epilepsy, duration of illness or duration of treatment.
In order to determine which variables to include in a regression analysis, all independent variables of PLWE were correlated with the dependent variable, mean QOL, using bivariate (Pearson) correlation. The regression model used then showed that, 11.6% of variation in mean QOL was explained by level of education, 8.1%, average annual seizures, 5.0%, reason for unemployment, 4.5% average annual income, and 2.3%, by type of employment. These variables therefore explained 31.5% of the total variations in mean QOL. The residuals plots showed that data met the assumptions of linearity, homoscedasticity and normality in the regression model used.