Menu
Menu
Menu
Menu
Background Poor healthcare-seeking behaviour is a major contributing factor for increased morbidity and mortality among children in low- and middle-income countries. This study assessed the individual and community level factors associated with healthcare-seeking behaviour for childhood illnesses among mothers of children under five in Chad. Methods The study utilized data from the 2014-2015 Chad Demographic and Health Survey. A total of 5,693 mothers who reported that their children under five had either fever accompanied by cough or diarrhea or both within the two weeks preceding the survey were included in this study. The outcome variable for the study was healthcare-seeking behaviour for childhood illnesses. The data were analyzed using Stata version 14.2. Multilevel binary logistic regression model was employed due to the hierarchical nature of the dataset. Results were presented as adjusted odds ratios (aOR) at 95% confidence interval (CI). Results Out of the 5,693 mothers who reported that their children under five had either fever accompanied by cough, diarrhea or both at any time in the 2 weeks preceding the survey, 79.6% recalled having sought treatment for their children’s illnesses. In terms of the individual level factors, mothers who faced financial barriers to healthcare access were less likely to seek healthcare for childhood illnesses, relative to those who faced no financial barrier (aOR = 0.80, 95% CI = 0.65-0.99). Mothers who reported that distance to the health facility was a barrier were less likely to seek healthcare for childhood illnesses, compared to those who faced no geographical barrier to healthcare access (aOR = 79, 95% CI = 0.65-0.95). Mothers who were cohabiting were less likely to seek healthcare for childhood illnesses compared to married mothers (aOR = 0.62 95% CI = 0.47-0.83). Lower odds of healthcare seeking for childhood illnesses was noted among mothers who did not listen to radio at all, relative to those who listened to radio at least once a week (aOR = 0.71, 95% CI = 0.55- 0.91). Mothers who mentioned that their children were larger than average size at birth had a lesser likelihood of seeking childhood healthcare, compared to those whose children were of average size (aOR = 0.79, 95% CI = 0.66-0.95). We further noted that with the community level factors, mothers who lived in communities with medium literacy level were less likely to seek childhood healthcare than those in communities with high literacy (aOR = 0.73, 95% CI = 0.53-0.99). Conclusion The study revealed that both individual (financial barriers to healthcare access, geographical barriers to healthcare access, marital status, frequency of listening to radio and size of children at birth) and community level factors (community level literacy) are associated with healthcare-seeking behaviour for childhood illnesses in Chad. The government of Chad, through multi-sectoral partnership, should strengthen health systems by removing financial and geographical barriers to healthcare access. Moreover, the government should create favourable conditions to improve the status of mothers and foster their overall socio-economic wellbeing and literacy through employment and education. Other interventions should include community sensitization of cohabiting mothers and mothers with children whose size at birth is large to seek healthcare for their children when they are ill. This can be done using radio as means of information dissemination.
This study was a cross-sectional study that utilized data from the 2014–15 Chad Demographic and Health Survey (CDHS), which is the most recent DHS conducted in the country. The CDHS is conducted by the National Institute of Statistics, Economic and Demographic Studies (INSEED) and the Inner-City Fund (ICF) International [21]. The CDHS utilized a stratified sampling design to recruit eligible participants. The national territory was demarcated into twenty-one study areas with reference to the 22 regions and the city of N’Djaména. Two strata were created in each field (urban and rural). In all, 626 primary survey units (PSUs) or clusters were systematically selected from the list of enumeration areas that were predefined during the 2009 General Population and Housing Census. Households in each cluster constituted the list from which eligible households were selected, with 25 households per cluster in the urban locations and 30 households per cluster in rural locations at random. A total of 17, 965 households from 4,075 urban areas in 163 clusters and 13,890 rural households nested in 463 clusters were selected. All resident mothers 15–49 years or those present the night preceding the survey were eligible to be interviewed. A total of 5,693 mothers reported that their children under five had either fever accompanied by cough or diarrhea or both within the two weeks preceding the survey. This constituted the sample size for our study. The outcome variable for the study was healthcare seeking behaviour for childhood illnesses. This variable was derived as a composite variable from two questions: “Did [NAME] receive treatment for diarrhea?” and “Did [NAME] receive treatment for fever accompanied by cough?” The responses were “Yes” and “No” in the CDHS. All mothers who responded “Yes” to either of the two questions were considered as seeking healthcare for childhood illnesses (coded as 1) whilst those who did not seek healthcare for any of the two childhood illnesses were coded as 0. There were 21 independent variables made up of 18 individual level variables and three community level variables. None of these variables was selected a prior; instead, the selection was based on conclusions drawn by earlier studies on healthcare seeking for childhood illnesses as well as their conceptual and theoretical bearing on healthcare seeking for childhood illnesses [22, 23]. The individual level variables were difficulty with distance to the facility, difficulty in getting money for treatment, difficulty with getting permission to visit a health facility, and difficulty in not wanting to go for medical help alone (each was coded as big problem and not a big problem). These related to geographical, financial, and partner support barriers faced by mothers when accessing healthcare. Big problem means the respondents considered each of these as a barrier to healthcare access whiles not a big problem means that they were not considered as barriers. Other individual level variables were mothers’ age (15–19, 20–24, 25–29, 30–34, 35–39, 40–44, and 45–49), marital status (married and cohabiting), healthcare decision-making capacity (alone and not alone), parity (one birth, two, three, and four or more), employment status (working or not working), religion (Christianity, Islam, and no religion), frequency of exposure to media (reading newspaper, listening to radio, watching television) which were coded as not at all, less than once a week, and at least once a week, sex of household head (male and female), mother’s subjective perception of the size of child at birth (less than average, average, or smaller than average), birth order (one, two to four, and five and above births), twin status (single or multiple births), and sex of child (male and female). Community literacy level (categorized into low, medium, and high), community socio-economic status (captured as low, medium, and high), and residence (rural and urban) were the community level variables. The categorisation of community literacy level and community socio-economic status into low, medium and high was not directly available in the data but generated from maternal education and household wealth quintile through a method of aggregation at the cluster level. We employed both descriptive and inferential analytical approaches. First, we computed the proportion of mothers who sought healthcare for childhood illnesses across the individual and community level variables. Next, a Chi-square test was carried out to assess the level of significance between the independent variables and healthcare seeking for childhood illnesses (see Table 1). At the bivariate analysis stage, due to multiple-comparisons, we introduced a correction method by using the Bonferroni correction method [24]. This was done by dividing the alpha rate (p = 0.05) by the number of analysis performed (21 explanatory variables) [25, 26], that is, 0.05/21 = 0.002. Therefore, at the bivariate analysis, statistical significance was declared at p≤0.002. Following the hierarchical nature of the dataset, the multilevel logistic regression model (MLRM) was employed after the bivariate analysis to examine the predictors of healthcare seeking for childhood illnesses. This comprises fixed effects and random effects [27]. The fixed effects of the model were gauged with binary logistic regression, which resulted in adjusted odds ratios (aORs) (see Table 2). The random effects, on the other hand, were assessed with intra-cluster correlation (ICC) [28] (see Table 2). The sample weight (v005/1,000,000) was applied in all the analyses to control for over- and under-sampling. All the analyses were carried out using Stata version 14.2. Source: 2014–15 Chad Demographic and Health Survey Source: 2014–15 Chad Demographic and Health Survey PSU = Primary sampling unit; ICC = Intra-Class Correlation; LR Test = Likelihood ratio Test; AIC = Akaike’s Information Criterion; N = Sample size Model 0 is the null model, a baseline model without any independent variable Model 1 is adjusted for individual level variables Model 2 is adjusted for community level variables Model 3 is the final model adjusted for individual and community level variables We assessed the fitness of the models with the likelihood ratio (LR) test. The presence of multicollinearity between the independent variables was checked before fitting the models. The variance inflation factor (VIF) test revealed the absence of high multicollinearity between the variables (Mean VIF = 1.21, Max VIF = 1.43, Minimum = 1.05). In order to develop robust models, only variables that showed statistically significant association in the bivariate analysis were included in the models. This study used publicly available data from DHS. Informed consent was obtained from all participants prior to the survey. The DHS Program adheres to ethical standards for protecting the privacy of respondents. The ICF International also ensures that the survey processes conform to the ethical requirements of the U.S. Department of Health and Human Services. No additional ethical approval was required, as the data is secondary and available to the general public. However, to have access and use the raw data, we sought and obtained permission from MEASURE DHS. Details of the ethical standards are available on http://goo.gl/ny8T6X.
N/A
