Clinical spectrum and documentation patterns of cerebral palsy among children at Arthur Davison Children's Hospital and Kitwe Teaching Hospital, Zambia: a retrospective record review.

backgroundCerebral palsy (CP) is the most common cause of childhood physical disability and is frequently accompanied by comorbidities such as epilepsy and feeding difficulties. In many low-resource settings, hospital-based evidence describing CP clinical subtypes, associated impairments, and care patterns is limited, constraining service planning for rehabilitation and pediatric neurology. We aimed to describe the clinical spectrum, associated impairments, interventions, and documentation patterns of CP at two tertiary referral hospitals in Zambia's Copperbelt Province.

methodsWe conducted a multi-centre, hospital-based retrospective record review of children with clinically documented CP at Arthur Davison Children's Hospital (ADCH), Ndola, and Kitwe Teaching Hospital (KTH), Kitwe. CP motor subtype (spastic, dyskinetic, ataxic, mixed) and topographical distribution were extracted where documented. Associated impairments (including epilepsy) and management (rehabilitation and medications) were recorded based on explicit documentation, with undocumented fields treated as missing rather than absent. Categorical variables were summarized as frequencies and percentages. Between-hospital comparisons used chi-square or Fisher's exact tests. Statistical significance was set at α = 0.05.

resultsA total of 195 children with CP were identified (ADCH n = 150; KTH n = 45). This study did not estimate incidence or prevalence. Among records with documented subtype, spastic CP predominated at both sites, while dyskinetic CP was the second most frequently documented subtype and accounted for a higher proportion at ADCH than at KTH. CP motor subtype distributions differed significantly between hospitals (χ²=8.28, p = 0.041). Topographical distribution did not differ significantly (χ²=2.51, p = 0.285), with quadriplegia most commonly recorded at both sites. Epilepsy prevalence differed markedly among records with documented epilepsy status (KTH 40.0% vs. ADCH 84.1%; Fisher's exact p = 0.000025; OR for KTH vs. ADCH = 0.126). Physiotherapy was frequently documented at both hospitals. Documentation completeness varied substantially across sites, particularly for epilepsy status, associated impairments, and functional severity measures (GMFCS) at ADCH.

conclusionsIn this hospital-based cohort from Zambia's Copperbelt Province, CP subtype patterns differed between ADCH and KTH, and epilepsy was highly prevalent (especially at ADCH), underscoring the need for integrated rehabilitation and epilepsy services. Substantial variability in documentation completeness suggests a systems-level opportunity to standardize CP assessment and recording to support clinical management and service planning.