Estimated prevalence for diarrhea in the HDSS from the HUAS.
Based on the weighted analysis, the estimated prevalence of any diarrheal episode during the past 2 weeks at baseline was 22.3% (CI = 19.5–25.0) among 20,853 children living in the HDSS area at the time of the HUAS survey.
Diarrhea prevalence in the HDSS from the HUAS-lite.
The prevalence of reported diarrhea in the past 2 weeks among children < 5 years old in the HDSS ranged from the highest rate of 10.8% (May 22 to August 31, 2009) to the lowest rate of 3.9% (September 14 to December 31, 2010) (). Among children with diarrhea in the past 2 weeks in the five rounds of the HUAS-lite, the proportion with MSD ranged from 53% (September 15 to December 3, 2009) to 58% (September 14 to December 31, 2010) ().
Prevalence of diarrhea in the last 2 weeks among children < 5 years old and healthcare-seeking pattern for diarrhea by HUAS-lite round from May 22, 2009 to December 31, 2010 in western Kenya
Prevalence of MSD and LSD in the last 2 weeks among children < 5 years old by the HUAS-lite round from May 22, 2009 to December 31, 2010, in western Kenya.
HUAS study enrollment and background characteristics.
During the baseline HUAS, we successfully interviewed caretakers of 1,043 children 0–59 months of age; 275 (26%) children were reported to have had diarrhea during the 2 weeks preceding the interview (). The children with diarrhea included 112 (41%) infants, 108 (39%) toddlers (12–23 months), and 55 (20%) children (24–59 months).
We used data on household asset ownership to rank household wealth from poorest to wealthiest using five quintiles. Overall, 34.3% of 1,043 respondents surveyed lived in households within the two lowest wealth quintiles; the highest proportion of children with diarrhea (33%) fell within the poorest wealth quintile (). Other characteristics of the households of children with diarrhea in the 2 weeks preceding the study were similar to the characteristics of all households interviewed (). Among caretakers interviewed, 922 (88.4%) caretakers were mothers, 65 (6.2%) caretakers were fathers, 33 (3.2%) caretakers were grandmothers, and 23 (2.2%) caretakers were other relatives.
Wealth quintile ranking of caretakers of children < 5 years old participating in the HUAS in western Kenya in 2007.
Description of the population surveyed in the HUAS study population in western Kenya in 2007 (unweighted analysis; N = 1,043)
Caretakers knowledge, attitudes, and perceptions of illness and health seeking.
We asked caretakers what they would look for to determine if a child is dehydrated; 716 (68.6%) caretakers indicated that they would look for lethargy, 481 (46.1%) caretakers answered sunken eyes, 369 (35.4%) caretakers answered wrinkled skin, 297 (28.5%) caretakers answered dry mouth, and 297 (28.5%) caretakers answered thirst. In total, 206 (19.8%) caretakers said that they would look for both thirst and dry mouth to see if a child is dehydrated.
According to caretakers, the majority (86.2%) of 275 children who had an episode of any diarrhea in the preceding 2 weeks had three to six loose stools per day. Reported accompanying symptoms included lethargy (N = 225, 81.8%), fever (N = 201, 73.1%), being very thirsty (N = 193, 70.7%), mucus or pus in stool (N = 194, 70.6%), dry mouth (N = 188, 68.4%), rice watery stool (N = 163, 59.7%), sunken eyes (N = 162, 58.9%), decreased urination (N = 102, 40.2%), wrinkled skin (N = 92, 33.7%), vomiting (N = 89, 32.7%), coma or loss of consciousness (N = 67, 24.6%), and blood in stool (N = 34, 12.4%). Caretakers reported overall that 6.7% of children with diarrhea were hospitalized and that 6.6% received administered intravenous (IV) fluids for rehydration; 47% of the children hospitalized with diarrhea received IV fluids.
We asked caretakers what they offered their child to drink and eat during the child's diarrheal illness. Of 275 children with diarrhea, 41 (15%) caretakers said they offered the child more drink than usual, 51 (19%) caretakers reported offering the child the same amount, and 183 (67%) caretakers reported that they offered less than usual to drink during the child's diarrheal episode. Of those caretakers who offered less than usual, 96 (52%) children were offered somewhat less, 69 (38%) children were offered much less, and 18 (10%) children were offered nothing to drink during their diarrheal illness. Of 269 caretakers who reported what they offered their child to eat during their diarrheal episode, 3 (1%) caretakers offered more than usual, 43 (16%) caretakers offered the usual amount, and 223 (83%) caretakers stated that they offered less than usual to eat. Of those caretakers who offered less than usual, 74 (33%) caretakers offered somewhat less, 67 (30%) caretakers offered much less, and 82 (37%) caretakers offered nothing during the diarrheal illness. According to their caretakers, 66 (37%) of 180 and 77 (35%) of 220 children who were offered less than usual to drink and eat, respectively, had vomiting accompanying their diarrheal illness.
Healthcare was sought outside the home for 214 (77.8%) of 275 children with diarrhea. For any episode of diarrhea, the places visited as the first source of healthcare outside the home included licensed (62%) and unlicensed (11%) providers and pharmacies (27%). Seeking care outside the home was similar among caretakers of children with bloody compared with non-bloody diarrhea (27 [79%] of 34 versus 187 [78%] of 241, P > 0.05) and more common among caretakers of children who were very thirsty and had a dry mouth compared with children who did not have both conditions (142 [77%] of 184 versus 17 [46%] of 37, P < 0.05). Of 61 caretakers who did not seek care outside their home for their children with diarrhea, the main reasons that they gave were that the child did not seem to need care (44.3%), the cost of treatment was too high (32.8%), the clinic was too far from home (9.8%), and they were unable to find transportation (8.2%).
The most common means of transportation to the nearest health facility of choice was walking (74%) followed by commercial transportation (which included riding on the back of a bicycle; 13%) and personal transport (generally a bicycle; 4%).
We asked caretakers how long it would usually take to reach the health facility of choice; 770 (74%) of 1,035 respondents estimated that it would usually take less than 1 hour. The main circumstances that make it difficult for caretakers to reach their nearest health facility of choice were that it cost too much money (49%) followed by heavy rainfall or flooding (45%) and lack of transportation (24%).
We asked caretakers about who makes the decision to take the child to a health facility when sick; 809 (78%) of 1,041 respondents said that the child's mother makes the decision, whereas 232 (22%) said other relatives, including the child's father, make this decision. We asked all 1,043 caretakers who participated in this survey if they think that vaccines are important to their child's health; 99% said they think that vaccines are important.
Risk factors for diarrheal illness.
The weighted multivariate analysis of risk factors for any diarrheal illness showed that children ages < 12 months (30.5%, aOR = 2.19, CI = 1.50–3.21) and 12–23 months (31.4%, aOR = 2.24, CI = 1.53–3.30) compared with children ages 24–59 months (16.2%) were at increased risk of having an episode of diarrhea (regardless of severity) during the 2 weeks preceding the survey (). Also, caretakers who knew of a child who had died of bloody diarrhea (aOR = 2.30, CI = 1.50–3.54) and who knew that bloody diarrhea is more dangerous than other forms of diarrhea (aOR = 1.68, CI = 1.20–2.35) were associated with the caretaker reporting that their child had diarrhea in the 2 weeks preceding the survey (). The latter findings may indicate that mothers who understand how serious diarrhea can be may be more alert for it in their children and thus, more likely to report it during a survey.
Independent predictors of any diarrheal illness among children < 5 years old in the HUAS in western Kenya in 2007 (weighted analysis; N = 1,043)
Predictors of seeking healthcare outside the home for any diarrheal illness.
On weighted multivariate analysis, seeking healthcare outside the home for diarrheal illness was less common for infants than children ages 24–59 months (aOR = 0.33, CI = 0.12–0.87) (). Caretakers who said lack of transportation was the main factor preventing them from reaching the health facility of their first choice were more likely to seek care outside the home for any diarrhea compared with caretakers who did not answer in this way (90.5% versus 78.9%; aOR = 3.18, CI = 1.13–8.89). This result may indicate that caretakers who have an expressed preference for certain health facilities may be more willing to seek care, perhaps because they are generally more informed. Caretakers of children who had sunken eyes during their diarrheal illness compared with caretakers of children who did not have sunken eyes (92.2% versus 64.9%) sought care more frequently outside the home (aOR = 4.76, CI = 2.12–10.70) ().
Independent predictors of seeking care outside the home for children < 5 years old with any diarrhea in the HUAS in western Kenya in 2007 (weighted analysis; N = 275)
Predictors of seeking healthcare from a health facility among caretakers who sought care outside the home for any diarrheal illness.
On weighted multivariate analysis, seeking care from a licensed health facility (versus a non-licensed health facility) among those caretakers who sought care outside the home for any diarrheal illness was significantly more common for infants versus older children (aOR = 5.06, CI = 1.88–13.61), when the caretaker had some formal education versus none (aOR = 3.32, CI = 1.56–7.07), when caretakers thought that bloody diarrhea could cause harm or death (aOR = 3.25, CI = 1.16–9.09), when caretakers did not report circumstances that make it difficult to reach their preferred health facility (aOR = 3.90, CI = 1.47–10.35), when the child was lethargic during the diarrheal episode (aOR = 5.73, CI = 1.79–18.42), when the child had been offered ORS at home (aOR = 6.99, CI = 3.01–16.22), and when the child was offered no special (i.e., alternative) remedies at home (aOR = 10.17, CI = 2.84–36.37). The latter may possibly be indicative of caretakers' higher education, which was also a predictor of seeking care at a health facility (aOR = 3.32, CI = 1.56–7.07). Caretakers who did not report looking for thirst as a sign of dehydration were less likely to seek care from a health facility for their child's diarrheal illness (aOR = 0.21, CI = 0.09–0.47) ().
Independent predictors of seeking care from a health facility among children < 5 years old with any diarrhea in the HUAS in western Kenya in 2007 (weighted analysis; N = 214)
Factors associated with ORS use among children with any diarrhea.
Most (89.5%) caretakers indicated that ORSs works well to treat diarrhea. However, only 63 (22.9%) of 275 children with any diarrhea, regardless of severity, were offered ORS at home according to their caretakers. A higher proportion of children with MSD (46 of 182, 25.3%) compared with LSD (17 of 93, 18.3%) were offered ORS at home (P > 0.05).
We examined factors associated with the use of ORSs at home for the child's diarrheal illness (). In the multivariate weighted analysis, caretakers were less likely to use ORSs at home for infants versus older children (aOR = 0.35, CI = 0.14–0.89). They were more likely to use ORSs at home if the primary caretaker had some formal education versus none (aOR = 3.01, CI = 1.41–6.42), if the caretaker perceived that dehydration could result in harm or death (aOR = 5.54, CI = 2.23–13.73), if the child had vomiting three or more times per day during the diarrheal episode (aOR = 3.33, CI = 1.56–7.11), if the caretaker knew of a child who died of bloody diarrhea (aOR = 2.73, CI = 1.20–6.20), if the child was offered the usual amount to eat or less than usual during their diarrheal episode (aOR = 8.24, CI = 1.80–37.73), and if the caretaker believed that breastfeeding prevents diarrheal illness (aOR = 16.19, CI = 1.32–199.21).
Independent factors associated with use of ORSs at home among children < 5 years old with any diarrhea in the HUAS in western Kenya in 2007 (weighted analysis; N = 275)
Estimated care seeking for diarrhea in the HDSS from the HUAS and HUAS-lite.
Our weighted analysis estimated that caretakers of 81.5% (CI = 76.5–86.4) of children in the HDSS with any diarrheal episode in the past 2 weeks sought care outside the home. In general, care was sought from licensed providers (57.6%, CI = 49.9–65.5), unlicensed providers (12.5%, CI = 7.2–17.7), and pharmacies (29.9%, CI = 22.5–37.3).
Among children with reported diarrhea specifically in the HUAS-lite, 82.0% of those children with MSD (95% CI = 80.6–83.4) received care outside the home versus 67.3% (95% CI = 65.4–69.2) of children with LSD (when averaged over the five surveys) ().
Care seeking for moderate-to-severe diarrhea in the HDSS from the HUAS-lite.
Among those caretakers seeking care for MSD in the HUAS-lite, 61.9% (95% CI = 59.9–63.9) sought care from a health facility; 35.4% (95% CI = 32.8–37.9) of MSD cases seeking care at a health facility specifically visited one of the GEMS case-control study sentinel healthcare facilities (). Of note, there were no significant differences in the proportions of caretakers who sought care at GEMS case-control study sentinel health facilities for MSD (586/1657, 35.4%) or LSD (293/929, 31.5%) over the course of the five rounds (P = 0.496 for MSD; P = 0.369 for LSD).