The use of traditional medicine practitioner services for childhood illnesses among childbearing women: a multilevel analysis of demographic and health surveys in 32 sub-Saharan African countries

BACKGROUND: Insights into the use of traditional medicine practitioners (TMP)-for common childhood diseases such as diarrhea and respiratory infections are important to understand the role of Traditional Medicine (TM) in reducing the increasing childhood morbidity and mortality in sub-Saharan Africa (SSA). However, a comprehensive picture of TMP utilisation and its associated factors for childhood illness in SSA is lacking. This study aimed to estimate the prevalence of the use of traditional medicine practitioner services to treat childhood illnesses among women with children under five years old and to identify individual and community-level factors associated with TMP use in SSA. METHODS: The analysis used Demographic and Health Surveys (DHS) dataset collected between 2010 and 2021 among 353,463 under-fives children from 32 SSA countries. Our outcome variable was the use of TMP for childhood illness, defined as having diarrhoea or fever/cough or both. Using STATA v14, we employed the random effect meta-analysis to estimate the pooled prevalence of TMP use for childhood illness and a two-level multivariable multilevel modelling to determine the individual and community-level factors associated with consultation of a TMP. RESULTS: Approximately [2.80% (95%CI: 1.88–3.90)] women who sought healthcare for childhood illnesses utilised the service of a TMP with the highest occurring in Cote d’Ivoire [16.3% (95%CI: 13.87–19.06)] and Guinea (13.80% (95%CI: 10.74–17.57)] but the lowest in Sierra Leone [0.10%(95%CI:0.01–1.61)]. Specifically, approximately [1.95% (95%CI: 1.33–2.68)] and [1.09% (95%CI:0.67–1.60)] of women sought the service of a TMP for childhood diarrhea and fever/cough, respectively. Women with no formal education [AOR = 1.62;95%CI:1.23–2.12], no media access [AOR = 1.19;95%CI:1.02–1.39), who lived in a male-headed household [AOR = 1.64;95%CI:1.27–2.11], without health insurance [AOR = 2.37;95%CI: 1.53–3.66], who considered it a problem getting permission to visit a health facility [AOR = 1.23;95%CI:1.03–1.47] and who perceived the size of their children at birth to be above average[AOR = 1.20;95%CI:1.03–1.41] had higher odds of using TMP for childhood illnesses. CONCLUSIONS: Although the prevalence of TMP for childhood illnesses appeared low, our findings highlight that TMPs continue to play a critical role in managing childhood illnesses in SSA. It is essential that policymakers and service providers should incorporate the potential role of TMPs in the design, review and implementation of child health policies in SSA. Also, the interventions for curtailing childhood illnesses should be focused on the characteristics of women who use TMPs for childhood diseases identified in our study. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12906-023-03972-3.