Effect of handwashing on the reduction of Escherichia coli on children's hands in an urban slum Indonesia

Diarrhea is a leading cause of mortality in children worldwide, with an estimated 1,438 deaths per day (Levine et al. 2020). Pathogens, including E. coli, are often associated with diarrhea, and hands can serve as a vehicle for pathogen transmission through the fecal–oral route (Abba et al. 2009). Handwashing with soap (HWWS) and water has been shown to eliminate bacteria from the hands (Curtis et al. 2005; Burton et al. 2011) and reduce the risk of diarrhea by 40% (Freeman et al. 2014). However, the impact of hand hygiene on reducing fecal contamination remains unclear, particularly in resource-limited settings (Saboori et al. 2013; Aihara et al. 2014).

Proper handwashing remains a challenge for children in low- and middle-income countries (LMICs), and young children often demonstrate poor handwashing practice (Xuan & Hoat 2011). Several studies have found that a common problem in urban slum communities is the lack of access to soap and clean water, which can affect handwashing technique and bacterial count (Pickering et al. 2011; Paraduth & Biranjia-Hurdoyal 2015). Some experimental trials (laboratory and field studies) confirmed this by manipulating some steps (use of soap, length of time, or hand-drying method) to measure the effect of various methods on bacteria elimination (Burton et al. 2011; Kim et al. 2019; Gizaw et al. 2022). However, these studies did not reflect the natural, daily handwashing behavior of participants in their community. In addition, some of the studies were limited to clinical settings (Seid et al. 2022), and comprehensive assessments of the slum community are scarce.

In urban slum settings, child hygiene is often compromised (Kundu et al. 2018). Our recent study in Indonesia found that young children in urban slums had poor handwashing techniques, which was significantly associated with E. coli detection on their hands. Similarly, awareness and knowledge of HWWS was inadequate (Otsuka et al. 2019). Previous studies in Indonesia have primarily focused on public perception, behavioral intervention, and determinants of handwashing among people (Hirai et al. 2016; Karon et al. 2017; Dwipayanti et al. 2021). However, no study has examined the effect of the handwashing technique on reducing E. coli on children's hands in this setting, particularly during the COVID-19 pandemic.

Therefore, this observational study investigated the effect of the handwashing technique on the reduction of E. coli on children's hands in urban slums. Our study focuses on analyzing which steps are practicable within the community's target population. We also assessed the relationship between E. coli reduction and several key handwashing variables, including duration of soap and water contact, handwashing steps, water used, and hand-drying methods.

This cross-sectional study was conducted among preschool and primary school students in Kiaracondong, an urban slum in Bandung, Indonesia, between August and September 2022. Bandung is the third most populous city, experiencing rapid urbanization and overcrowding, particularly in the slums. Bandung's slum settlements still suffer from poor sanitation and water access (Tarigan et al. 2015). Kiaracondong is one of Bandung's most populous urban slum communities, with a high population density (22,426/km²) and poor water and sanitation infrastructure (BPS 2022). After explaining the purpose and content of the study, 137 students provided written consent and participated. We included students from preschool, grades 2, 4, and 6 with a mean age of 8.9 ± 2.4 years old.

The participants were asked to perform their daily handwashing practice individually in a detached space, while an observer observed the entire handwashing process. They were provided with all the needed materials, such as tap water, soap, tissue paper, and hand towels. Schools and households in developing countries (Kim et al. 2019), including Indonesia, provide hand towels as a common practice. The handwashing process was video recorded using two cameras (front and side views), and the evaluation was based on both recorded video and direct observation. A single observer determined the process and was supported by one research assistant who ensured the recording, use of water, and management of each student's turn. Five main points were assessed in this process: (1) handwashing steps (10 steps) (Otsuka et al. 2019), (2) the use of soap, (3) the volume of water, (4) the length of time (total duration, water contact, and soap contact), and (5) the hand-drying method. The volume of water used for handwashing was measured using a modified jerrycan equipped with a volume scale, while the length of time was measured using stopwatch and video observation (see Supplementary file: Appendix A).

The hand swab test was used to determine the level of E. coli on the children's hands before and after handwashing. The researcher performed the swab test using a swab test kit (Pro Media ST-25 PBS; Series: 64-8070-68, ELMEX, Japan) containing a wiping swab in 10 mL of phosphate buffer saline (PBS). Pre-test with four samples was performed before the school assessment. To minimize bias due to tap water contamination during handwashing, we have already tested the water, and the E. coli results were negative. We also implemented a cleaning protocol by using alcohol-based wet wipes to clean tap handles and soap bottles, thus preventing contamination for each child. To collect ‘hand before washing’ samples, a student's left or right hand was randomly selected. From 137 participants, the right hand was sampled first in 74 students, and the left hand was sampled first in 63 students. Then, students were asked to spread selected hands, and the wiping swab was rolled on the palm, backside, fingers, and in-between fingers. After that, students were requested to wash hands in the usual way. Immediately after handwashing, the opposite hand (hand that had not yet been sampled) was swabbed using the same method. The same hand was not sampled twice because the hand sampling method used could be considered similar to handwashing, which would remove E. coli and introduce bias into our objective of evaluating the effect of handwashing techniques on E. coli count after handwashing (see Supplementary file: Appendix B). After wiping, the swab was kept in PBS of the kit and delivered to the laboratory within 4 h for processing. There are some bacteria that serve as indicators to detect the presence of fecal contamination on hands. However, in this study, we focused on E. coli as a single indicator, as this research is a follow-up to our previous study in the same area (Otsuka et al. 2019). The enzyme substrate method was used to detect E. coli. All bacteria analysis was performed under aseptic conditions in the biosafety cabinet of a microbiology laboratory. Each sample (10 mL) was divided into two volumes (1 and 4 mL), and passed through a 47-mm-diameter 0.45-μm sterilized cellulose filter (Advantec, Tokyo) using a membrane filtration unit (vacuum pump). Approximately 10 mL of sterilized PBS buffer was added to facilitate uniform dispersion over the filter surface. Thereafter, the cellulose filter was placed on XM-G Agar growth media (XM-G; Nissui Pharmaceutical Co, Japan). The sample was incubated on the media for 20 ± 2 h at 37 °C. After incubation, the colony had turned blue and was counted. Blue colony indicates the presence of E. coli in the sample.

The results were analyzed using JMP SAS version 16 (SAS Institute, Japan) for Microsoft Windows 10. For statistical analysis, the number of colonies was measured as a colony-forming unit (CFU) and transformed to log₁₀ per hand before and after handwashing. The matched pair sample t-test (p < 0.05) was used to compare the mean log₁₀ concentration of E. coli before and after handwashing. As the maximum reliable count of E. coli on the filter was 300, the upper detection limit was set at 3,000 CFU/hand. For the categories of handwashing techniques, we used several key variables, such as duration of soap and water contact, handwashing steps, water used, and hand-drying methods (Aihara et al. 2014; Friedrich et al. 2017; Seid et al. 2022). As each parameter performed might influence the effect, multivariate stepwise logistic regression was used to investigate the association and interconnection between handwashing method/steps and its effect on E. coli reduction. The reduction of E. coli was categorized into two groups: reduction and no reduction before and after handwashing. The ‘no reduction’ criterion indicates that the E. coli concentration remained the same or even increased after handwashing. Eligible dependent factors for E. coli reduction were handwashing steps 2, 4, 9, and 10, duration of soap contact and water contact, and hand-drying. These factors were computed using the stepwise forward method to determine those that were significantly associated with E. coli reduction. The p-value threshold for entry and removal from the model was 0.25 and 0.1, respectively.

The 137 preschool and primary school students who participated in the study were aged 4–12 years and were in grades 2, 4, and 6. More than half of the participants (52%) were females. To investigate fecal contamination, we evaluated the E. coli count in 137 paired pre- and post-handwash participants. E. coli was detected in 122 (89%) samples before handwashing, which decreased to 81 (59%) after handwashing (Table 1). Except for preschoolers, there was a significant difference in the mean concentration of E. coli by grade and gender, before and after handwashing (p < 0.05). Second graders had the highest (1.69 ± 1.01 log₁₀ CFU/hand), followed by preschoolers (1.51 ± 1.08 log₁₀ CFU/hand). Overall, handwashing decreased the E. coli count by 34%, with the lowest concentration found in the highest grade (grade 4). There was also a reduction in the number of colonies of E. coli, with most of them having E. coli concentrations below 100 CFU/hand. Interestingly, in some students, we also observed that E. coli levels remained the same or even increased after handwashing. These findings support our hypothesis that children's handwashing technique influences the presence and concentration of E. coli after handwashing.

We measured the mean E. coli concentration before and after handwashing by the method practiced (Table 2). Table 2 displays the log₁₀ CFU of E. coli before and after handwashing, categorized according to the techniques used by the participants. The average time spent by participants for the entire handwashing process was 18.5 ± 9.3 s. A greater reduction in E. coli was found when using water and soap (handwashing duration ≥ 20 s), soap contact (soap contact during handwashing duration ≥ 10 s), 4–5 steps, ≥500 mL amount of water, and single-use tissue for the hand-drying method. Handwashing with ≤3 steps did not significantly influence E. coli reduction. Although we provided tissue paper, some students used hand towels to dry their hands, resulting in no significant reduction after handwashing. Table 2 also shows the correlation between handwashing technique and E. coli reduction. The reduction of E. coli on children's hands was significantly associated with soap use, duration of soap contact, number of steps, and hand-drying method (p < 0.05).

Table 3 shows the association between handwashing step and E. coli reduction. More than 90% of the participants wet their hands and rubbed them palm to palm and about two-thirds used soap during handwashing. Nevertheless, less than 25% completed scrubbing of the back of their fingers, thumbs, and fingertips, indicating poor knowledge of this aspect of handwashing. Soap use during handwashing significantly reduced E. coli concentration on children's hands, compared to water only (p < 0.05). Among those who scrubbed some parts of the hand, cleaning in-between fingers had a significant correlation with E. coli after handwashing. Although some students skipped the final two steps in handwashing, we found a significant correlation between rinsing with water and drying hands with a towel or tissue paper and E. coli reduction.

The results of the regression analysis model predicting the handwashing technique and E. coli reduction are presented in Table 4. This regression model represents the function of the handwashing technique performed and reduction of E. coli, which might be influenced by other techniques or steps. The odds of E. coli reduction were three times and four times higher when handwashing included the use of soap and water (AOR: 2.81, 95% CI: 1.62–11.8) and fingers interlaced cleaning (AOR: 4.45, 95% CI: 1.54–12.8), respectively. Additionally, scrubbing the hand with soap for more than 10 s was five times (AOR: 5.07, 95% CI: 1.34–19.1) more likely to reduce E. coli. Using multi-use towels or clothes for hand-drying had a protective effect in reducing E. coli (AOR: 0.03, 95% CI: 0.01–0.47).

A high level of fecal contamination on hands has been observed in an urban slum, which has been linked to poor handwashing technique (Otsuka et al. 2019). Contamination may come from the household and environment, such as physical contact with contaminated water or materials (Daneshmand et al. 2018). Although several studies have investigated E. coli in children's hands (Aihara et al. 2014; Paraduth & Biranjia-Hurdoyal 2015; Ogba et al. 2018), limited research has documented this during the COVID-19 pandemic. The pandemic may have influenced handwashing behavior, as people increased their frequency of handwashing to prevent the spread of the infection (Dwipayanti et al. 2021). Frequent handwashing during the pandemic was associated with a reduction in infections (Seid et al. 2022). Our study found that 89% of the children had E. coli on their hands. This was lower than the 98.7% previously documented in the same population (Otsuka et al. 2019). However, the level of contamination in this study was higher than in previous studies. For example, a systematic review revealed that the mean E. coli prevalence on hands in LMICs was 69% (Cantrell et al. 2023). Although different measurements and counting methods were used in a previous study (Cantrell et al. 2023), the findings from our present study indicate that there is still a high risk of contamination among young children in urban slums in Indonesia.

The concentration of E. coli on hands has been measured as 0.6–3.5 log₁₀ CFU/hand (Pickering et al. 2011; Friedrich et al. 2017; Kundu et al. 2018). This is consistent with our study which found that the average E. coli concentration by grade was 1.5–2.3 log₁₀ CFU/hand, with higher contamination observed among participants in the lower grades. However, a previous study conducted in an urban slum setting in India found a lower concentration of 0.64 log₁₀ CFU/per two hands in children under 5 years (Kundu et al. 2018). This higher concentration of E. coli contamination observed in our study could be attributed to some children spending more time engaging in outdoor activities, which exposes them to more germs and dirt. Some children in this community were involved in garbage sorting to help their parents who worked as garbage collectors (Sai et al. 2020).

Except for preschoolers, we found a statistically significant reduction in bacteria after handwashing. Lower graders performed fewer handwashing steps, which could explain the low effectiveness in bacteria reduction. Furthermore, performing three or fewer steps did not result in significant E. coli reduction when compared to four steps or more. We found that in some students, the E. coli concentration remained the same or even increased after handwashing. Similar observations have been made by other authors in which young children had limited knowledge of HWWS (Xuan & Hoat 2011) and performed handwashing with a low volume of water and without soap, which could have limited the reduction in E. coli on their hands after handwashing (Aihara et al. 2014; Agestika et al. 2019). This underscores the importance of the age of children as an essential factor in bacteria reduction of handwashing techniques.

Despite the availability of soap, 39% of the students did not use soap during handwashing. A similar trend was found among children in LMICs, with estimates showing that only 27–39% of children wash their hands with soap (Saboori et al. 2013; Otsuka et al. 2019). A possible explanation for this is the lack of effectiveness of handwashing promotion or prevalent social practices at home. In urban slum communities, some families cannot provide handwashing soap because it is an additional cost. Moreover, some respondents describe using soap as challenging because it is a new practice, and they believe washing with water is enough to make hands physically clean. Experimental trials have shown that HWWS is more effective in removing bacteria than using water alone (Burton et al. 2011; Amin et al. 2014). Similarly, our findings showed that the use of soap was twice as effective as using water alone in eliminating bacteria. HWWS for 20 s has been highlighted as an effective method for removing bacteria from the hands (WHO 2009). Wetting the hands with water and scrubbing with soap creates a lather that traps and eliminates these bacteria (Burton et al. 2011). Although some students performed handwashing for more than 20 s, only 6% of them scrubbed their hands with soap for that entire duration, while others did so briefly. It is important to note that scrubbing the hands with soap for more than 10 s significantly reduces E. coli concentration. This finding adds to our understanding that using soap for a short duration is insufficient to remove bacteria. Therefore, future handwashing promotion campaigns that target children should emphasize the use of soap and the appropriate duration of soap contact with the hands.

Previous studies have demonstrated the presence of diverse bacteria on different parts of the hands, such as fingertips (Julian et al. 2015), palms (Fierer et al. 2008), and in-between fingers (Rosenthal et al. 2014). Hence, scrubbing specific parts of the hands is essential for effective bacteria removal during handwashing (Julian et al. 2015). Similarly, we found a significant correlation between E. coli reduction and scrubbing in-between fingers during handwashing. Hand scrubbing during handwashing is an important action that physically destroys germs. However, we observed that more than half (58%) of the students did not wash in-between their fingers during handwashing (step 4). This confirms a previous finding before the pandemic that only 55% performed step 4 (Otsuka et al. 2019), indicating that handwashing promotion during the pandemic did not have a significant effect on accomplishing this step. Students often consider the palm as the dirtiest part of the hand (Ray et al. 2011), and they may overlook other parts, including in-between fingers, during handwashing. Our findings highlight the importance of focusing on specific parts during handwashing promotion campaigns for children.

Our study also revealed that the method of hand-drying has contrasting effects on E. coli reduction after handwashing. The largest reduction was found in the hands of students who used single-use tissue. This parallels a previous study that showed that single-use or paper towels could dry hands while causing less contamination in the washroom environment (Huang et al. 2012). Drying hands with single-use paper towels resulted in a larger reduction than air-drying (Jensen et al. 2015). By contrast, using reusable hand-drying methods such as cloth towels or clothes had a much smaller reduction effect and was considered a limiting factor in reducing E. coli. This may corroborate our finding that there was no reduction in E. coli levels for some students after handwashing. Clothes may not dry the hand completely, leaving it slightly moist. Residual water on the other hand may increase the risk of contamination because bacterial transmission is more effective in wet conditions (Patrick et al. 1997), so undried hands increase the possibility of recovery of E. coli (Ansari et al. 1991). Cloth towels have the lowest reduction in the number of fecal bacteria when drying hands compared to warm air and single-use paper (Ansari et al. 1991). A possible reason for this is that sharing hand towels with users who practice poor handwashing techniques increases the chance of the transfer of bacteria from poorly washed hands to the towel. Many studies have focused on hand-drying methods by comparing paper towels and air-drying. However, most urban slums lack access to these methods due to financial constraints. Therefore, it is crucial to emphasize the hand-drying method when promoting handwashing. Private reusable cloth towels may be an effective alternative when a school or household cannot provide single-use paper.

On average, we found that handwashing had a minimal effect on bacteria count (34%). This finding is consistent with other studies which reported that a single handwashing is insufficient; sustained and consistent hand hygiene behavior is required to reduce bacteria significantly (Aihara et al. 2014). Furthermore, the complexity of E. coli through environmental contamination in urban slum areas cannot be underestimated. Microbial contamination of water used to wash hands in schools or households poses a higher risk of hand contamination (Daneshmand et al. 2018; Berhanu et al. 2021). Some households and schools in Bandung's urban slum use groundwater for activities, such as handwashing, cooking, or bathing (Otsuka et al. 2018). Groundwater in some areas of Bandung is moderately to heavily polluted, which can lead to cross-contamination with humans (Hasanawi et al. 2022). A previous study indicated that high levels of E. coli have been detected in various types of water in peri-urban areas (Asada et al. 2022). If the water is potentially unsafe, it will potentially become a contaminant for hands. This necessitates further research to investigate the pathway of E. coli contamination in school and household environments in order to track and identify the sources.

This study has some limitations. This study employed a purposive sampling of two schools with relatively small sample size and the same level of sanitation in the study area. Hence, the findings may not be generalizable to all children in LMIC urban slums. In addition, there was no observation at the household level, which would represent students' daily practice in a natural setting. The presence of observers and the facilities provided during the handwashing observation may have biased the participants to modify their handwashing technique. To minimize this bias, we conducted the evaluation in a detached space and consistently reminded participants to wash their hands as they normally would. Nonetheless, this study enabled us to assess the importance of handwashing steps in reducing bacteria. Regarding the method of E. coli detection, we used an enzymatic method that cannot specifically recognize the type of E. coli. E. coli has many types of strains, some of which can cause severe health hazard, such as diarrhea. Lastly, E. coli is not the sole indicator for fecal contamination on hands; there are other bacteria that may serve as indicators, which should be considered for future studies.

This study highlights the correlation between poor handwashing practices among young children living in an urban slum and fecal contamination on their hands. There may be a misconception that the primary purpose of handwashing is simply to wet hands and clean palms, while other parts of the hands are overlooked. The study recommends that the focus on handwashing techniques for children should be extended to include scrubbing specific parts of the hands, such as between fingers and under fingernails, where fecal contamination can be found. Additionally, this study found that the low percentage of students who use soap and single-use tissue is likely due to the limited availability of facilities or prevalent social practice at home. To improve the removal of E. coli contamination from hands and prevent cross-contamination, it is recommended that schools provide soap and single-use tissue paper. To improve fecal contamination removal on the hands, it is recommended that schools provide soaps and single-use tissue paper. Proper handwashing is critical in preventing the spread of pathogens, and education on proper handwashing should start early for young children. Handwashing intervention should be tailored to the needs of LMICs, where resources may be limited. Further research that incorporates a comprehensive assessment of handwashing behavior and environmental contamination is needed to understand why handwashing does not effectively reduce fecal contamination in some students.

This research was supported by the JSPS KAKENHI Grant in Aid for Scientific Research B (Grant Number: JP18H00992). The authors are grateful to the respondents who participated in this study. We also wish to thank the research team and members of the Research Unit for Clean Technology, National Research and Innovation Agency (BRIN) of Bandung, for their support and teamwork in the field.

All relevant data are included in the paper or its Supplementary Information.

The authors declare there is no conflict.