Health care seeking behaviour for children with acute childhood illnesses and its relating factors in sub-Saharan Africa: evidence from 24 countries

Background Childhood illnesses and mortality rates have declined over the past years in sub-Saharan African countries; however, under-five mortality is still high in the region. This study investigated the magnitude and factors associated with health care seeking behaviour for children with childhood illnesses in 24 sub-Saharan African countries. Methods We used secondary data from Demographic and Health Surveys (DHSs) conducted between 2013 and 2018 across the 24 sub-Saharan African countries. Binary logistic regression models were applied to identify the factors associated with health care seeking behaviour for children with acute childhood illnesses. The results were presented using adjusted odds ratios (aOR) with 95% confidence intervals (CIs). Results Overall, 45% of children under-5 years with acute childhood illnesses utilized health care facilities. The factors associated with health care seeking behaviour for children with acute illnesses were sex of child, number of living children, education, work status, wealth index, exposure to media and distance to a health facility. Conclusions Over half of mothers did not seek appropriate health care for under-five childhood illnesses. Effective health policy interventions are needed to enhance health care seeking behaviour of mothers for childhood illnesses in sub-Saharan African countries. Supplementary Information The online version contains supplementary material available at 10.1186/s41182-021-00385-1.

Page 2 of 8 Yaya et al. Tropical Medicine and Health (2021) 49:95 countries have a greater burden of childhood illnesses in contemporary time [17][18][19]. In addition, progress to achieve Sustainable Development Goal (SDG) 3.2 (elimination of preventable child death, reduction in neonatal mortality to less than 12 per 1000 live births and those of under-five mortality to less than 25 per 1000 live births for every country by 2030) is slower, and without the necessary policies and interventions, sub-Saharan Africa may not meet the SDG 3.2 target by 2030 [20]. There is evidence that at least one child dies out of every 12 births in sub-Saharan Africa before age five compared to one out of every 147 in developed countries [3]. The majority of under-five mortality in sub-Saharan Africa is due to infectious and parasitic diseases, such as malaria, respiratory infections, meningitis and diarrhea [5,13,14,21]. Diarrhea, for instance, is a major cause of morbidity and mortality globally, as it accounts for 1.3 million deaths among children under-5 years of age each year [21]. Although childhood illnesses can be managed efficiently in sub-Saharan Africa, evidence suggest poor health care seeking behaviour among mothers for childhood illnesses in the region [18,22]. Evidence based information is thus needed on health care seeking behaviour of mothers for effective policies and interventions in sub-Saharan Africa [22]. Meanwhile, prompt and adequate health care seeking behaviour interventions among mothers can substantially reduce childhood mortality due to childhood illnesses in low-income and middle-income countries [10]. Efforts have been made globally and by various countries in sub-Saharan Africa to reduce morbidity and mortality of under-five resulting from childhood illnesses through policies that will promote child health care services among mothers [23][24][25][26]. To achieve the Sustainable Development Goal target of at most 25 deaths per 1000 live births by the year 2030, there are efforts and high-level commitments towards addressing the issues. For instance, at the global level are the Integrated Management of Childhood Illness (IMCI) strategy developed by the World Health Organization (WHO), the Every Woman Every Child Strategy and the Partnership for Maternal Newborn and Child Health [3,10,27,28]. Moreover, many countries in sub-Saharan Africa have policies and programmes that are aimed at reducing under-five morbidity and mortality. For instance, Nigeria introduced the Maternal and Child Health Policy in 1994, National Immunization Policy and Standards Practice in 1996 and Breastfeeding Policy in 1999 [14]. Many African countries have also embraced the International Child Rights Policy, emphasizing and promoting child health care services among [29].
Although prior studies have provided some evidence on factors affecting childhood illnesses in low-income and middle-income countries [30][31][32], there is little knowledge on health care seeking behaviour among mothers for childhood illnesses in sub-Saharan Africa. The factors associated with health care seeking behaviour from studies conducted in different countries include knowledge of danger signs, occupation, residence, education, age, marital status, birth order, mass media exposure [1,13,33] and income [34]. Community level promotion of prompt health care seeking behaviour among mothers for childhood illnesses to enhance the health care of under-five children has been emphasised [5,35]. Meanwhile, underfive mortality is still high in sub-Saharan Africa despite the progress made in the last few decades. Proper health care seeking behaviour of mothers for childhood illnesses can prevent or reduce the magnitude of child mortality resulting from childhood illnesses [10,36,37]. This study, therefore, seeks to examine the factors associated with under-five children illnesses such as diarrhea and fever and health care service utilization in sub-Saharan African countries. Assessing health care services for childhood diseases and associated risk factors may help prevent and reduce under-five morbidity and mortality in sub-Saharan African countries [5].

Methods
The study used secondary data from the Demographic and Health Surveys (DHSs) of 24 countries in sub-Saharan Africa. The countries were selected if the surveys were conducted between 2010 and 2018 [33,38], and outcome and explanatory variables were available. The DHSs use multi-stage stratified sampling method [33,38,39]. These DHSs are nationally representative and comparable surveys conducted worldwide in more than 85 countries [40]. The surveys usually collect a wide range of self-reported and objective data with a strong focus on indicators of reproductive health, fertility, child and maternal health, nutrition, mortality, and self-reported health behaviours among adults [41]. The sample for the final analyses was 98,590. It included children who had diarrhea and/or fever or cough in the 2 weeks preceding the surveys, whether they sought private or public health care or not. The country specific details are presented in Additional file 1.

Measurement of variables Outcome variable
The outcome of interest was under-five children with incidence of diarrhea and/or cough or fever in the past 2 weeks before the surveys. Those who went for consultation in a public or private health care service provider and those who did not go were classified as users and non-users, respectively [14].

Explanatory variables
The explanatory variables considered were maternal education (no formal education, primary and secondary plus), age of mother (15-24, 25-34 and 35-49), occupation of mother (working and not working), wealth index (poorest, poorer, middle, richer and richest), marital status (never married and ever married), residence (urban and rural), distance to health facility (experience no difficulty in getting to a health facility/not a problem in getting to a health facility or experienced difficulty in getting to a health facility/a problem to in getting to a health facility), media access (no access and have access) and sex of the child (male and female) [14].

Data analyses
First, descriptive analyses were performed using frequency and percentage distributions to examine the characteristics of participants and prevalence of health care seeking behaviour among mothers for children with childhood illnesses. Differences in prevalence were examined using chi-square test. Furthermore, to assess multicollinearity, correlation test was performed among independent variables. The findings showed that the assumptions of multicollinearity were not violated. The tolerance value was greater than 0.10 [42]. Non-response and missing data were excluded to arrive at the weighted sample size. Second, a binary logistic regression model was fitted to examine the relationship between explanatory variables and health care seeking behaviour of mothers. To adjust for sampling variability, DHSs incorporate two-stage cluster sampling [33,38,39,43] and sampling weight was applied to account for the complex survey design including weight, cluster, and strata. Stata version 14 (Stata Corp, College Station, Texas, USA) was used to estimate the prevalence of health seeking behaviour of mothers and odds ratios with 95% confidence intervals (95% CI).

Characteristics of the sample population
The descriptive characteristics of respondents are shown in Table 1 of health care seeking behaviour among mothers for children with acute childhood illnesses was found in Sierra Leone (65.1%) and the lowest in Cameroon (22.1%) (see Additional file 1).

Binary logistic regression
The results of the adjusted odds ratios (aOR) and 95% CI for the relationship between socio-demographic variables and health care seeking behaviour of mothers for childhood illnesses are presented in Table 3. The findings showed that the odds of health care service utilization for childhood illnesses were significantly higher among mothers with male children (OR 1.06; 95% CI 1.03-1.08) compared to mothers with female children. The odds of health care service utilization for childhood illnesses were found to be higher among mothers who

Discussion
This study investigated the predictors of health care service seeking behaviour among mothers for childhood illnesses in sub-Saharan African countries. Overall, 98,590 children under-5 years of age in sub-Saharan African countries born within 5 years preceding the surveys, and reported incidence of diarrhea and/or cough or fever in the past 2 weeks before the surveys [14] were included in the study. The findings revealed disparities in health care seeking behaviour among mothers for children with acute childhood illnesses in sub-Saharan African countries. Hence, much is needed to be done to enhance health care seeking behaviour among mothers for children with acute childhood illnesses. For instance, the highest prevalence of health care seeking behaviour of mothers for children with acute childhood illnesses was found in Sierra Leone (65.1%), while the lowest was found in Cameroon (22.1%), (see Additional file 1). These findings reinforce the need for concerted efforts to enhance health care seeking behaviour of mothers for children with acute childhood illnesses in the region [18]. Overall, less than half (45%) of under-five children with acute childhood illnesses utilized health care services, consistent with previous findings [1,34,37,44]. Low health care seeking behaviour among mothers has been shown to be a major determinant of childhood morbidity and mortality in sub-Saharan African countries [22,45], which has been attributed to child, social and maternal factors [1,13,33,34]. For example, there is evidence of an association between mothers' education and health care seeking behaviour for their children, where women with higher levels of education are more likely to seek health care services for their children [33,46,47]. Thus, proper health care seeking behaviour of mothers could prevent and reduce childhood illnesses [10,36,37], and childhood mortality in low-income and middle-income countries [1,34]. In many sub-Saharan African countries, poor and inadequate medical facilities and poor health seeking behaviour are known risk factors for infant and child mortality [1].  49:95 In this study, we found sex of child, number of living children, education, work status, wealth index, exposure to media and distance to a health facility to be predictors of health care seeking behaviour of mothers for children with acute illnesses. We observed higher odds of health care seeking behaviour for male children as compared with female children. This finding corroborates previous findings [48,49], where culture and traditional beliefs have been implicated for this outcome [50]. Male-child preference by some cultures and traditional expectations of men as breadwinners for families are some of the reasons for the extra-care of male children [50]. Furthermore, we observed that education is an important factor in mothers' health care seeking behaviour, especially for childhood illnesses [2,35,51,52]. Thus, enhancing women's educational levels will help advance health care seeking behaviour for childhood illnesses [45,48,53].
The higher odds of health care seeking behaviour found among employed mothers [1] may be linked to empowerment [54][55][56], and women's ability to decide on certain maternal issues [55] and pay for health care service [55,56]. Health care seeking behaviour of mothers had been associated with skilled or semi-skilled employment [36], which can also be linked with women empowerment. Low health care seeking behaviour among the "richer and richest" wealth indices contrast other findings from previous study [2,14,57,58]. This study speculates that geographical location, educational attainment of women and level of awareness may explain the differences in findings. Meanwhile, low socio-economic conditions including poverty has also been shown to be a factor influencing the attitude of mothers towards seeking health care services for their children [36]. For instance, children might not get the required medical attention due to the mothers' inability to pay for health services. Prior studies also noted the importance of access to health care facilities in enhancing mothers health care seeking behaviour, and further reducing childhood illnesses [1,2,48,51].

Strengths and limitations
The study has both limitations and strengths. One of the strengths of the study is that the data sets are from several sub-Saharan African countries and are nationally representative. This permits generalisation of the results in sub-Saharan African countries, despite the time lag in data collection periods. However, comparison of result from different surveys should be done carefully and with caution due to differences in data collection periods [56]. Furthermore, the sample size is sufficiently large as it increases the validity of the findings. Nonetheless, the study is subject to social desirability and recall bias [50,59], because it was based on self-reported data [59].
Finally, the surveys are cross-sectional, and they only permit association and not causal relationship [30,38].

Conclusion
The study analysed secondary data from 24 sub-Saharan African countries on acute childhood illnesses and health care seeking behaviour of mothers in sub-Saharan Africa. The findings indicate a low prevalence of health care seeking behaviour of mothers for childhood illnesses. Socio-economic factors including maternal education were found to be associated with health care seeking behaviour. Public health policies and programmes that target women's empowerment and development are critical to addressing the issue. This may help increase the prevalence of health care seeking behaviour of mothers for childhood illnesses, and consequently improve child health outcomes in sub-Saharan African countries.