Predictors of Quality of Care among People with Epilepsy Receiving Treatment at Selected Hospitals in Nairobi, Kiambu and Machakos counties, Kenya

Abstract

Epilepsy is a common neurological disorder affecting over 70 million people globally and accounting for approximately 1% of the global disease burden, with nearly 90% of cases occurring in low- and middle-income countries. In sub-Saharan Africa, an estimated 10 million people live with epilepsy, while Kenya has a prevalence of about 18 per 1,000 population, corresponding to nearly one million people with epilepsy (PWE), of whom approximately 80% do not receive adequate treatment. Beyond its clinical burden, epilepsy is associated with stigma, discrimination, reduced productivity, and increased healthcare utilization. Despite increasing awareness and improved health-seeking behaviour, evidence on the quality of epilepsy care and its determinants in Kenya remains limited. The main objective of this study was to determine predictors of quality of care among PWE receiving treatment at selected Level Five hospitals in Nairobi, Kiambu, and Machakos counties. A cross-sectional mixed-methods study was conducted between May and September 2021. Quantitative data were collected from 373 PWE using semi-structured questionnaires, while qualitative data were obtained through focus group discussions and key informant interviews. The Donabedian Structure–Process–Outcome Framework and the Aday and Andersen Healthcare Utilization Model guided the assessment of quality domains and determinants. Quality of care was measured using a semantic differential scale, and Exploratory Factor Analysis generated normalized factor weights that were used to construct a composite weighted quality index. A median score of 47 was used to categorize quality of care into high and low levels. Although Fisher’s finite population correction yielded a minimum sample size of 276 from a sampling frame of 969 patients, the study retained the original Fisher sample size of 385 to enhance statistical power, reduce sampling error, and improve precision. Consecutive sampling recruited 373 participants, representing a 96.9% response rate. Quantitative data were analyzed using descriptive statistics, chi-square tests, logistic regression, and multiple linear regression in SPSS version 26, while qualitative data were analyzed thematically in NVivo and triangulated with quantitative findings. Majority participants were 29–49 years (37.5%), and 52.8% reported receiving high-quality care. Respect, communication, and tolerability of medication side effects were the most important contributors to quality ratings. At bivariate analysis, occupation (χ²=19.13, p<0.001), duration before treatment initiation (χ²=6.07, p=0.048), anti-seizure medication use (χ²=4.024, p=0.045), seizure frequency after treatment initiation (χ²=7.337, p=0.026), stigma experience (χ²=5.022, p=0.025), treatment regimen (χ²=10.464, p=0.015), waiting time (χ²=6.49, p=0.031), service availability (χ²=7.137, p=0.029), and affordability of care (χ²=4.034, p=0.043) were significantly associated with quality of care. Multivariate logistic regression identified stigma as the strongest independent predictor of low-quality care (OR=2.123, 95% CI: 1.119–4.026; p=0.021). Qualitative findings showed that respectful provider interactions, effective communication, and treatment effectiveness enhanced patient experiences, whereas inadequate education regarding seizure management and medication side effects reduced perceived quality. The study concludes that quality of epilepsy care is shaped by a complex interplay of sociodemographic, clinical, psychosocial, and health system factors, with stigma emerging as the most pervasive determinant. Strengthening community sensitization, healthcare worker training, patient education, diagnostic capacity, medication supply systems, and structured counselling and peer-support programmes is essential for xxi improving quality of care, narrowing the treatment gap, and informing county and national epilepsy policies.