Data

Child’s age and sex: In terms of the comorbidity, the prevalence defined as experience of both conditions during the two weeks preceding the interview appeared to be impacted by the child’s age but not his/her gender. Using the youngest age group (2-5 months) as a reference, the risk peaked in children aged 12-23 months with nearly 4-fold increase (OR=3.52; 95% CI=1.94 to 6.38). A slightly lower odds ratio was observed in children aged 6-11 months (OR=2.55; 95% CI=1.35 to 4.84).  The risk of comorbid diarrhea and pneumonia decreased constantly for older children.

Weight-for-age Z-score: Co-occurrence of the two conditions was lower in children with weight-for-age Z-scores greater or equal to -2 standard deviation (SD) compared to the malnourished with scores less than -2 SD (OR=0.6; 95% CI: 0.44 to 0.82). Improved sanitation was defined as toilet facility with one of the following characteristics: connection to a public sewer, connection to a septic system, pour-flush latrine, simple pit latrine with a slab, or ventilated, improved pit or composting toilet. This variable was associated to the 14-day comorbidity (improved/unimproved OR= 0.63; 95% CI: 0.45 to 0.90).

Disposal of child feces: Mothers were asked where they disposed of their children’s stools when they do not use any toilet facility. Disposal of child’s feces was considered as safe if the child used the toilet or latrine or if the excreta were put or rinsed into toilet or latrine. Safe fecal disposal showed a statistically significant association with the 14-day comorbidity (OR: 0.73; 95% CI: 0.55 to 0.98).

Parents’ education: Comorbidity was associated with father’s education but not mother’s. Children of fathers with primary education were less exposed to comorbidity compared to those from fathers with no education (OR =0.64; 95%CI: 0.42 to 1:00). Similarly children of fathers educated to secondary level and above experienced less comorbidity   of   the  two  conditions  (OR= 0.68;  95%CI:0.44 to 1.04).

Household wealth Index: Children from the wealthiest household experienced less comorbidity compared to the lowest wealth index quintile (OR=0.45; 95% CI: 0.21 to 0.95). Type of dwelling:  Of the three indices of housing quality examined (main material of wall, floor and roof), only the type of roof  suggested  an  association  with  the comorbidity.

Our results from the community survey add significantly to works reported on co-morbidity and clinical overlap between diarrhea and pneumonia (Fischer-Walker and Black, 2009; Mulholland, 2005; Fischer-Walker et al., 2013; Chisti et al.,  2008,  2011).  Given  that  comorbidity carries more risk on child mortality (Fenn et al., 2005; Fischer-Walker et al., 2013) our findings could be handy in the fight against child mortality. From these analyses comorbid diarrhea and pneumonia is more than would have been expected by chance. While being age-dependent, it peaks at ages between 12 and 23 and is high but slightly lower for the  group of children aged between 6 and 11 months.

Other factors statistically associated to comorbid diarrhea and pneumonia included  malnutrition, parents’ education, unsafe fecal disposal and unimproved sanitation. Reduction of the burden of acute and chronic malnutrition can have a substantial impact on child morbidity and mortality from diarrhea  (Rice  et  al.,  2000; Caulfield et al., 2004) and pneumonia (Fonseca et al., 1996). Finding an association between this factor and the 14-day occurrence of the two conditions is particularly important.

We found increased risks of comorbidity for unsafe disposal of feces and its associated factor namely the quality of the household’s sanitation facility. The safe disposal of children’s feces has been identified as critical for children’s health, and several studies have reported the effects of poor sanitation on childhood diarrhea in sub-Saharan Africa and elsewhere (Clasen et al., 2010; Bhutta et al., 2013; Cairncross et al., 2010; Jinadu et al., 2004; Tumwine et al., 2002; Manun'Ebo et al., 1994; Bauza et al., 2019; Messou et al., 1995; Oyejide et al., 1994; Pokhrel and Viraraghavan, 2004; Traoré et al., 1994; Schmidt et al., 2011; Cronin et al., 2016). Sanitation prevents contamination of the environment by excreta and therefore imposes a barrier to the transmission of pathogens coming from feces of an infected person (Brown et al., 2013). On the other hand, if feces are not properly disposed, they remain a source of contamination in the environment. Due to limited use of potties and diapers in most households in rural settings where young children are often allowed to defecate in the yard or land surrounding the household, a lower risk of diarrheal disease has been associated with safe disposal (Yeager et al., 1999). But finding this practice as related to lower risk of comorbidity could have implications in the targeting of policies designed to reduce child’s mortality.

At the exception of the child’s age, none of the factors associated to the comorbidity was identified as risk factor shared between the two conditions. This echoes findings from elsewhere. In a study of risk factors of diarrhea and Acute Respiratory Infection (ARI), Siziya et al. (2009) found no shared factors except child’s age. Our findings linking malnutrition, unimproved sanitation facility and unsafe excreta disposal to the co-occurrence of childhood diarrhea and pneumonia seem not to support the suggestion that diarrhea and pneumonia occurring in the same individual at the same time could be caused by risk factors the two conditions share (Kahabuka et al., 2012). The above associations to comorbidity known for diarrhea suggest that the co-occurrence of the two conditions is perhaps influenced by the occurrence of diarrhea with a particular level of severity. It is unlikely that there is any shared risk factor playing a leading role. We believe that the co-occurrence of the two conditions is mediated through a weakened immune system or a loss of protective nutrients; we found malnourished children to be at great risk of co-occurring diarrhea and pneumonia.

In addition, the three factors linked strongly to diarrhea as documented in the literature and not as much to pneumonia would suggest rather that comorbidity could be a reflection of diarrhea,  leading to pneumonia. This tends to support findings reported elsewhere among young children in low-income settings and in the case of concomitant  infection   with   diarrhea   and   pneumonia  (Fischer-Walker et al., 2013; Chisti et al., 2008; Schmidt et al., 2009).

A finding of an improved type of roofing associated to higher prevalence of comorbidity is surprising. We  would not speculate on possible explanations because it could be a chance effect. Manun'Ebo et al. (1994) reported similar unexpected results with diarrheal incidence and duration associated to the walling of the house in the wrong direction.

We carried out a second hand analysis of the data previously reported by the national DHS project team; this has some limitations as the data wouldn't fit exactly what we intended to do. The survey data may not be the most appropriate source of data for the study of this question, but this representative dataset covering the entire country allows comparison to other countries and regions of the world. We recognize the inability of a 14-day recall dataset to ascertain the simultaneous occurrence of the two conditions, or the order in which they occur in children. A longitudinal study mapping the days during which the conditions occur would be necessary to answer that question.