Child survival remains a critical public health issue in SSA despite the improvements made over the years [1]. Traditional and complementary medicine is known to be used to be widely used to manage communicable and non-communicable diseases in Africa[8]. The role of traditional and complementary medicine, especially TCM practitioners, in managing childhood illness has not been fully explored in the continent. Our study provides the first-ever insight into traditional medicine practitioners’ contributions to managing childhood illness at a regional level. Specifically, our study examined the proportion of women with children less than five years old who sought the service of a traditional medicine practitioner to treat their children illness (diarrhea or fever/cough or both) as well as individual and community factors associated with such a pattern of use in using DHS data from 32 African countries.
Our study indicates that 3% of women who sought healthcare for childhood illnesses used the services of TMPs. Our finding is consistent with a national representative study conducted in China, Mexico, Russia, South Africa, and Ghana[33], which found that the rate of use of TMP was 1.7% for South Africa and 1.5% for Ghana [33]. A similar prevalence has been reported in a similar study conducted in Indonesia in which TMP utilisation rate in childhood illness treatment was 3.4% in the month preceding the survey [34]. Another South African study reported similar findings in which 2.5% of survey respondents reported visiting a TMP when sick, and 3.3% of respondents reported seeking any form of health services consulted a traditional healer[11]. An Ethiopian community-based study reported that 3.3% of mothers visited a Traditional Birth Attendant (TBA) when their child was sick[16].
On the other hand, our finding was lower than the 26.4% average reported in 32 countries majority of which were in Europe and Asia[35]. The only African country included in this multicounty study was South Africa, and the reported prevalence of TM provider use was 24% which is still higher than the pooled prevalence reported in our study[35]. A previous systematic review on TCM use in sub-Saharan Africa reported that traditional medicine practitioner utilisation rate ranges from 1.2–67%, with a lower rate (1.2–44.1%) observed among studies that utilised a larger sample size than those that utilised smaller samples (37.5–67%)[8].
The pooled prevalence of TMP use for childhood diarrhea was 2% in our study. Our finding is lower than what was reported in a systematic review in developing countries in which a 7% median percent of mothers/caregivers used the service of traditional medicine practitioners[36]. Even though Ivory coast and Mali show relatively high prevalence regarding the use of traditional medicine practitioners to treat childhood diarrhea, community studies in these countries and other African countries reported higher use of traditional medicine practitioners. For instance, a community cross-sectional study in Mali found that 57% of parents/guardians sought the service of TMP[15], and 6.4% of caregivers visited a traditional medicine practitioner in the rural areas of Burkina Faso [17]. In comparison, 11.3% of children with diarrhea sought the services of a traditional medicine practitioner[12].
Regarding fever/cough, one in every 100 women who sought care for her child’s fever/cough visited a traditional medicine practitioner, with Ivory Coast and Mali reporting the highest proportions of women who visited a traditional medicine practitioner. Compared with community-based surveys, our overall prevalence is lower, including the TMP utilisation rates of individual countries considered in our study. For example, a community study in Benin and Gabon reported that 7% and 18% of mothers chose a traditional healer as their first healthcare provider, respectively[19]. A similar study in Ethiopia found that 18.5% of parents/guardians first consulted with a TMP for their children's fever[18].
The overall low TMP utilisations rates for diarrhea and fever/cough observed in our study may reflect that TMPs are not the preferred healthcare providers for these conditions; instead, conventional medical care is the mainstay as it has been reported in several multinational studies assessing the healthcare seeking behaviour for childhood illness in Africa[7, 20, 21, 25]. The increased child survival interventions over the years from governments and international organisations, which promote the use of formal healthcare options to manage childhood diseases, may help explain our finding[37, 38]. Also, the current barriers to traditional and complementary medicine use in Africa may explain the low TMP use. These include the lack of belief in the safety and efficacy of TM due to the absence of scientific evidence to support TM practice, absence/limited regulation of TM practice, perceived lack of education and training of TMP in the African region and the association of TM with witchcraft and sorcery[8]. It is important to note that the low use of TMP for childhood illness in our study does not imply that TM plays an insignificant role in managing childhood illness in Africa. Current evidence suggests that most mothers/caregivers self-medicate with home remedies, TM products, or conventional products bought from the market or sourced from friends or relatives before seeking care outside the home[27, 29, 30, 39].Thus, the low TMP utilisation rate in our study may be explained by high self-medication practices with TM products, or combination with conventional products not only among children, but also among adults[8]. Also, we observed a higher prevalence rate in Ivory Coast and Mali and the West African region than in other countries and regions. Such relatively high use may reflect the extent of TM and recognition and its integration into the healthcare system in these countries and region compared to other countries and regions. For instance, as of December 2018, most countries in West Africa, including Ivory Coast and Mali, have met most of the indicators for the integration of TM into national health systems[10].
Results from our multi-level modelling indicate that women with no formal education were more likely to seek care for their children when sick. Similarly, those without media access had higher odds of seeking the service of a TMP. Education has a role in enlightening women about the efficacy and safety of formal healthcare use as opposed to TMP use, which might explain our finding. Similar findings have been reported in a household survey on TMP use for childhood diseases in south-eastern Nigeria[12], the Amhara region of Ethiopia[26], a systematic review on traditional and complementary (TCM) use among the general population in sub-Saharan Africa[8], a survey conducted among the adult population in China, India, Ghana, Mexico, Russia and South Africa[33] and a mixed method study among African migrant women in Sydney Australia[40]. However, our finding contrasted with studies reporting complementary and alternative medicine (CAM) use among children in Germany[41] and Italy[42] and among the adult population in Western countries, including Brazil, in which higher education status was associated with TMP use[43, 44]. The disparity in the targeted population, literacy rates in these regions, and what constitutes TCM may explain the difference observed between educational status and TMP use reported in our study and those reported in Western countries.
In line with community-based African studies on the use of TMP for childhood illness[26], our study revealed that women from poor households had higher odds than their wealthy counterparts of seeking the service of TMPs for their children’s illness. Similar findings were reported among adults in Ghana and India and a systematic review on TCM users’ characteristics in sub-Saharan Africa[8, 33]. However, a systematic review on CAM use among children outside of Africa gives an opposite picture concerning our finding [28]. Since TM is considered affordable due to its relatively low cost compared to conventional medicine in Africa might explain why women from low socio-economic households were likely to seek care from TMP for their children [8]. A South African study on the cost of TM use for non-communicable diseases reported that most participants spent little to nothing to access TM [45].
We also observed that women without health insurance were more likely to seek care from TMP for their children. Our finding is consistent with a nationwide study among older Ghanaians in which TM was primarily used by those who were not insured[46]. However, another Ghanaian study in two districts of the Ashanti region found no significant association between health insurance status and TM [47]. Health insurance is considered an enabling factor to access healthcare services since it provides financial risk protection and decreases healthcare expenditure. However, most health insurance schemes do not cover TM, and the majority of the population in Africa is uninsured[48, 49]. Therefore, uninsured women in our study do not have the financial protection associated with seeking modern healthcare and will seek TM, which is considered less expensive.
Women who reside in households headed by a male had higher odds than a female of seeking treatment from a traditional medicine practitioner for their childhood illness, especially fever/cough. Our finding contrasted with the finding from a multi-country study using DHS data that focussed on the utilisation of conventional healthcare for childhood illness in which living in a household headed by a man was associated with conventional healthcare[7]. However, individual country studies found that the odds of utilising conventional healthcare for childhood illness were higher when the household head was a woman[50, 51]. Our regression analysis indicates that women not involved in decision-making at home and who had issues getting permission to visit the hospital were more likely to seek care for their child’s fever/cough from TMP. Such a finding seems to suggest that the decision to utilise TMP services for fever/cough is often in the hands of the husband/partner or someone else, which further supports our previous finding on the likely use of TMP if the head of the household is male. The reason for such an association is unknown, and this requires further enquiry. The use of the service of a TMP for childhood fever/cough was more likely if the perceived size of the child was above average than if it was average. It is possible that the parent/caregiver considered the above-average size of their child to be linked to supernatural causes rather than biomedical, and as such only TMPs are competent to manage such condition[52, 53].
Policy And Practice Implication
Despite the low patronage of TMP for managing childhood illness, our findings suggest that the TMPs continue to have a role in managing childhood illness in SSA. Thus, policymakers and health service providers need to consider the potential role of TMPs in child health policies and interventions through TMP training and promoting collaboration between TMP and conventional healthcare providers to achieve better child health outcomes.
The identified factors associated with TMP use underscore that women and children with such characteristics are considered risk factors for TMP use for childhood illness in SSA. Thus, women and communities with such characteristics, including low socio-economic backgrounds, must be prioritised when designing and implementing child health interventions in SSA. It is also important that child health policies and interventions are incorporated into women’s empowerment as a tool to enhance women’s active contribution to decision-making regarding the health of their children.
Strengths And Limitation
A key strength of our study is that it uses nationally representative data from the countries involved. Our findings can, therefore, be generalised to all children in all 32 countries included in our study. Also, selection bias was reduced since a multi-stage sampling strategy was employed in the conduct of the DHS in all 32 countries. DHS uses a standardised questionnaire, and data collection is done by trained personnel, which adds rigour to our study. We employed vigorous analysis using multi-level modelling to account for the varied clusters in our data. Notwithstanding these strengths, readers should bear in mind some limitations when interpreting our findings. First, casual relationships cannot be inferred due to the cross-sectional design employed in our study. Second, we could only use variables in the DHS dataset to determine the factors associated with TMP use for childhood disease. As such, our analysis did not include other factors that influence health-seeking behaviour, such as women’s knowledge of childhood disease and health workers’ attitudes, as data on these variables were not captured in the DHS datasets used in our study. We cannot rule out the potential for social desirability and recall biases since some of our study data were collected based on self-report.