WASH, vulnerability, severity, and the response of urban slum dwellers to the COVID-19 pandemic

Water, sanitation, and hygiene (WASH) behaviours have become critical contributors to coronavirus illness prevention and response during the greatest health challenge of the current global coronavirus crisis. However, one-third of the world's population, mainly marginalized and vulnerable individuals living in urban slums in many developing nations, still lack access to WASH. Most individuals in vulnerable areas suffer from water-borne diseases such as diarrhoea, acute respiratory infections, and skin infections as a result of contaminated water, inadequate sanitation, and poor hygiene. Due to poverty, limited access to basic WASH services, and inadequate housing, slum residents are particularly vulnerable and exposed to illness during the crisis (Konteh 2009). In addition, slum dwellers have a low level of awareness and information about illness and WASH practices, which can lead to poor health and dangerous protective behaviours (Hsan et al. 2019).

Understanding slum dwellers’ WASH conditions, their vulnerability, severity, health-protection knowledge, perceptions, and practices can aid in identifying elements that influence their protective behaviours when it comes to adopting response practices (Podder et al. 2019). In dealing with pandemics, knowledge enhancement and education initiatives are likely methods for improving the self-protection of marginalized people (Al-Hanawi et al. 2020). The COVID-19 infectious agent spreads fast within the entire society and communities; therefore, protecting marginalized groups helps reduce the risk faced by individuals and society.

In public health research and practice, the knowledge, attitude, and practice (KAP) theory is used to examine the role of information, attitudes, and behaviours in minimizing the risk of communicable diseases caused by poor hygiene and unsanitary settings (Mubarak et al. 2016). Among developing countries, Ethiopia (Berhe et al. 2020) and Bangladesh (Islam et al. 2021) showed low hygiene KAP. Knowing, learning, and practising WASH can help people reduce their risks of contracting communicable diseases (Berhe et al. 2020). Some researchers have confirmed the link between knowledge and practice, as well as attitudes and behaviours (Nguyen et al. 2019). Previous research in the field has shown that people with high WASH and health-protection knowledge and attitudes are more likely to adopt safe hygiene and self-protection practices (Ozdemir et al. 2020).

In addition, Rogers (1975) established the Protection Motivation Theory (PMT), which is frequently used to predict health intention and practice in a variety of research, including water management, disaster, waste management, and infection illnesses (e.g., Janmaimool 2017; Tang & Feng 2018).

In the PMT, people's perceptions of the severity and vulnerability are assessed using the threat appraisal component. The coping appraisal, on the other hand, is used to assess people's ability to cope with and respond to dangers, as well as to remove barriers to certain behaviours, such as response efficacy and self-efficacy. Prior studies in the health field have found that response efficacy and self-efficacy are linked to health-protective behaviours (Hernández-Padilla et al. 2020). Threat appraisal and coping appraisal both have a stronger potential to predict protective intents, according to several studies (e.g., Bashirian et al. 2020). In addition, there is an external factor such as knowledge which is linked to threat perception and intention, thus influencing the behaviour change (Chamroonsawasdi et al. 2017). The intention was found to be a mediator variable between PMT elements and actual behaviours by Tang & Feng (2018). Cognitive factors such as information, knowledge, and actual behaviours are linked to current practice and influence people's perceptions of risk, susceptibility, and severity (Eppright et al. 1994), This study used a combined theoretical framework that incorporated knowledge–practice and PMT to examine (i) the relationship between WASH and urban slum dweller's vulnerability and severity to the COVID-19 pandemic, (ii) how their perceived severity, self-efficacy, and response efficacy are driven by their knowledge and practice and vulnerability, and (iii) how their protective behaviours are shaped by their current practices, perceived severity, self-efficacy, and response efficacy through a case study of urban slum dwellers in Bangkok, Thailand.

By comparing the characteristics of two sub-communities: environment, living conditions, location, and sensitive concerns within the community, Lock 1-2-3 and Ban Guay were chosen as case study places. The Lock 1-2-3 sub-community is located near Art Narong Road and has a dense population, whereas the Ban Guay sub-community is located along the canal and under the toll-way road. In the chosen areas, low-quality drinking water with a high quantity of bacteria was found inside water tanks in vending machines (Office of Water Quality Management 2019).

The study applied the phenomenological approach, focusing on slum residents’ personal experiences with the COVID-19 pandemic (Creswell & Creswell 2018). Data were gathered using a combination of techniques, including semi-structured interviews and questionnaire surveys.

Semi-structured interviews were conducted with 20 community members and nine staff members from seven organizations and local authorities that work with marginalized people. Each interview was conducted for 30 min. Most questions are open-ended to explore their experiences with the COVID-19 pandemic and challenges regarding inadequate WASH facilities, knowledge, attitude, and protective behaviours towards COVID-19 protection.

A total of 453 persons responded to the questionnaire survey, with 385 from the Lock 1-2-3 sub-communities and 68 from the Ban Guay sub-communities. Using Slovin's Formula at 5% significance, the calculated sample size required was 383, but we oversampled cases to 453 to increase the reliability of the data analyses.

The snowball sampling method was used as it is appropriate for hidden and unknown populations (Baltar & Brunet 2012), particularly in slums with limited information or census data that are unknown and anonymous. Other sample criteria, such as ethnicity, age range, gender, health condition, and disability, have been created to encompass varied socio-economic, ethnic, and cultural features of the entire slum population. The population list for the Khlong Toei slum was compiled using information from the National Statistical Office of Thailand (2019) and a recommendation from the Duang Prateep Foundation.

The questionnaire aimed to examine knowledge, experience, perceived vulnerability, severity, self-efficacy, response efficacy, and protective behavioural intention towards COVID-19 in a larger population of slum communities. The questions were developed from the themes/issues that emerged from the semi-structured interview transcript coding.

The questionnaire consisted of three parts:

• Demographic information included age, gender, nationality, occupancy, education, family member, marital status, state of residential occupancy, access, and non-access to WASH.

• Perception of vulnerability and severity, knowledge and threat appraisal of COVID-19.

• Coping appraisal, practices, and protective behavioural intention towards COVID-19.

Eleven trained field assistants and three staff members from the Duang Prateep Foundation performed the questionnaire survey in Thai. They have a lot of expertise working with people who live in slums. All field assistants were instructed to take the social survey's ethical considerations, privacy, and integrity.

The data analysis methods employed in this study were a mix of qualitative and quantitative. The information gathered during the interview and questionnaire survey was kept anonymous.

In data analysis, two groups of ‘access WASH’ and ‘non-access WASH’ were categorized by using the criteria of having clean water at home, basic sanitation, and hygiene facilities (e.g., toilets, washrooms), and accessing basic public healthcare, including sanitation and hygiene education towards COVID-19 protection based on the respondents’ answers to the relevant Yes/No question in the questionnaire. As a result, 349 respondents who had access to WASH and 104 respondents who did not had access to WASH were grouped.

The phenomenology research methodology was used to qualitatively describe urban slum dwellers’ experiences and behaviours in relation to the COVID-19 phenomena, as well as the main factors influencing their perceptions, attitudes, and responses. Based on PMT components, the authors constructed themes for assessing noteworthy utterances and developed meaning clusters through interviewing (Creswell & David Creswell 2018).

Descriptive statistics was applied to describe the level of respondents’ knowledge, practices, attitudes, and behaviours. The difference in socio-demographic and cultural variables between the two respondent groups was tested using Fisher's exact test and the T-test. With a significant level of less than 5%, all statistical analyses were performed in IBM SPSS Statistics 26.

The measurement/outer and structural/inner models were the two stages in the performance of the Partial Least Squares Structural Equation Modelling (PLS-SEM) (Hair 2015). Formative measures were used to model seven constructs in order to analyse the causal links between the latent constructs estimated by 22 items using the route model analysis. The inner model is evaluated using two criteria: collinearity and the importance and relevance of the formative indicators (Hair 2015). The variance inflation factor (VIF) was calculated to determine the extent of collinearity. The tolerance (VIF) of the Sll indication was less than the threshold value of 5. To interpret the formative data, the values of the outer weights and the loading of the associated items were estimated.

Table 1 depicts the demographic features of two slum dweller groups: those who had access to WASH and those who did not. In terms of age, gender, ethnicity, and dwelling occupancy, there were no variations in numbers between the two groups.

In general, there are no differences in age, gender, and nationality between the two examined groups of access WASH and non-access WASH. Squatters can access WASH services in 49% of cases, whereas 43% of them do not have access to WASH (p < 0.001). Respondents who had access to WASH had a higher education level (p < 0.001) and were more likely to be single (p = 0.004). Daily wage earners had access to WASH more than other groups, whereas 32.7% of the jobless do not (p < 0.001).

While the majority of respondents have access to tap water, approximately 100 homes do not. Residents who can afford the tap water payment have water filters in their houses for drinking water. Those who cannot afford must use canal water for washing and other domestic use, and have to purchase water from a vending machine for drinking. Some residents who cannot pay a monthly water payment must buy tap water from their neighbours, as stated by a community volunteer ‘Here, some people do not have water access because they cannot afford the water meter and cannot pay monthly. They rent water from their neighboring with the higher prices’. Another young male migrant expressed ‘I buy water from water vending machines for drinking and use water from the canals for personal wash and domestic use’. Furthermore, due to the lack of work, residents of the Khlong Toei slum are unable to pay their water bills during the crisis, worsening their living conditions. They survive on donated food and clean drinking water. Foreign workers, the homeless, and families living in non-standard housing, such as under the toll-way and train, make up the majority of those without access to clean water (Pattanasri et al. 2022). However, the water quality of community water vending machines has been found unsafe for consumption due to high levels of pathogens. Furthermore, the water quality of the Khlong Toei canal, where these people gather and store water for domestic use, is greater than the average permissible value for biochemical oxygen demand (BOD) (around 52.6%) according to a report by the Office of Water Quality Management, Thai Ministry of the Interior (2019). The canal water is contaminated by solid and organic trash dumped straight into the canal from marketplaces and private residences.

The majority of houses had their own private toilet, although it was undersized and lacked cleanliness standards. People who resided beneath the toll-way bridge have no access to a private restroom, as a disabled man confided ‘I live under the toll-way bridge. I do not have my own toilet, thus I use the canal or toilet of my friends’ houses’. Regarding access to health care services, Thai citizens can freely access basic health treatment under the Thai government's 30-baht Universal Health Coverage Programme. ‘We have the 30bth golden card for all diseases, thus, we can access the public medical clinic nearby’, a Thai woman explained. However, migrants face difficulties in accessing healthcare services when they need them; as a migrant stated, ‘I cannot access healthcare service and other COVID-19 services because of my social status. I don't know where to go when I am sick so mostly take care of myself at home’. Or ‘health care services are expensive and inaccessible. We really cannot afford it’, a male migrant expressed.

Table 2 displays statements made by the interviewed slum dwellers about their views of dangers and hazards, coping benefits and barriers, and protective intentions. The interviewees regarded themselves and their family members to be susceptible to the COVID-19 pandemics because of their current health and hygiene situations, such as chronic sickness, inadequate housing, population density, and lack of WASH facilities. They confided that their seasonal work does not allow them to adopt COVID-19 protective techniques, and they perceived the situation as serious. They were more concerned about starvation and poverty than the COVID-19 infection. At the same time, they were concerned about the coronavirus's potential for death and illness. They did, however, believe they had self-efficacy in dealing with the pandemic. They had faith in illness and pandemic awareness and practice, as well as community collaboration and government information in the fight against the pandemic. Wearing masks, social distancing, and handwashing were among the COVID-19 preventative activities that interviewees said would help them lower risks. They also saw coping barriers in the high cost of medical masks, which they could not afford, their nature of work in crowded markets and ports, which prevents them from practising social distancing, and their social status, which prevents them from accessing WASH, public healthcare, and online information.

T-test was performed to determine whether there was a significant difference between the mean values of two groups, access WASH and non-access WASH, on the COVID-19 knowledge, prevention practices, perceived vulnerability, perceived severity, self-efficacy, response efficacy as well as protective intention. The results showed that there are no differences between the two groups in terms of knowledge, practice, vulnerability, perceived self-efficacy, response efficacy, or response cost. Respondents with non-access to WASH, on the other hand, rated the COVID-19 as being more severe (t = −3.807, p < 0.001; df = 451), whereas respondents who has access to WASH had higher protective intention (t = −3.947, p < 0.001; df = 431) than those without. Squatters can access WASH services in 49% of cases, whereas 43% of them do not have access to WASH (p < 0.001). Respondents who had access to WASH had a higher education level (p < 0.001) and were more likely to be single (p = 0.004). Daily wage earners had access to WASH more than other groups, whereas 32.7% of the jobless do not (p < 0.001).

The model results show current practice (β = 0.456, p < 0.001) and perceived vulnerability (β = 0.180, p < 0.001) are both influenced by knowledge, whereas the current practice is driven by vulnerability (β = 0.163, p = 0.002). At the same time, current practice influences perception of self-efficacy (β = 0.314, p < 0.001), response efficacy (β = 0.557, p < 0.001), and severity (β = 0.198, p = 0.003). The model's findings also revealed that perceived severity has an impact on the intention to engage in a protective activity (β = 0.102, p = 0.043). People will be more motivated to undertake protective actions if they are more aware of the health and illness repercussions. Individuals with strong self-efficacy believe they can cope with the sickness, which leads to high preventive motivation since perceived self-efficacy promotes protective motivation actions (β = 0.308, p < 0.001).

It has also been proven that perceived response efficacy influences protective motive behaviours (β = 0.409, p < 0.001). Response efficacy was a factor that boosted self-efficacy by increasing self-protection. Self-efficacy and response efficacy were both found to be useful predictors of people's protective behaviours.

The findings of this study revealed that urban slum dwellers face inequalities in access to clean water, poor sanitation, personal protective equipment shortages, and public healthcare services. The condition of urban slum dwellers in Thailand is comparable to that of poor people and low-skilled worker migrants in many developing nations such as Kenya, Africa, Pakistan, and India, where they struggle to get healthcare and pricey safety equipment (Zulu et al. 2011). Many scholars across the world (e.g., Iribarnegaray et al. 2015; Hashemi 2020) have tried to develop frameworks and indicators for the evaluation of the sustainability of sanitation systems; however, there is often a lack considering knowledge, practices, perceived self-efficacy, and response efficacy of the community, especially vulnerable, the marginalized population as the social indicators of a sustainable sanitation system. The findings showed that many squatters do not have access to WASH and people who had access to WASH had a higher education level. Daily wage earners had access to WASH more than the jobless group. The results also depicted that people with non-access to WASH rated COVID-19 as being more severe, but their counterparts had higher protective intention. The barriers to WASH and healthcare services are due to unemployment, hidden social status, expensive health insurance, low health literacy, language barrier, and social stigma decrease the protective motivation awareness of marginalized people (Nguyen & Pattanasri 2022) which may constraint their protection intentions towards COVID-19. This implies that the urban sustainable sanitation systems do not only require the improvement of technical and physical components (i.e., sanitation and hygiene facilities) but also enhancing equity of WASH access among socially marginalized groups considering social status, ethnicity knowledge, practices, and attitudes toward self-protection against the pandemics such as the COVID-19.

The findings demonstrated that urban slum dwellers had a basic understanding of COVID-19 and WASH practices and a high sense of threat, coping assessment, and protective purpose. People who have access to WASH have a stronger protective intention against COVID-19.

In this study, perceived severity, self-efficacy, and reaction efficacy in the PMT model plays a role in the ‘attitude’ component of the KAP model. Knowledge linked to people's awareness, motivation, and competence to understand, appraise, and assess health risk or vulnerability situations leads people to judge and form accurate decisions by maintaining or improving their health practices regarding disease prevention and health protection (He et al. 2016). Through threat appraisal and coping appraisal, the model revealed that knowledge was a critical element that had a higher impact on practice and behaviour intention (Renner et al. 2008). It can also influence people's risk perceptions, lowering their perceived severity and vulnerability beliefs (Eppright et al. 1994). Vulnerability can help predict deliberate acts because it is intimately tied to people's knowledge and emotions. Fear and anxiety are essential for people's survival, and they can be used to anticipate health-preventive behaviours (Stangier et al. 2021).

The findings also revealed that existing behaviours may have an impact on how people perceive coping and threat evaluation. People with good current practices can reduce or prevent the intensity of adverse occurrences and raise their impression of self-efficacy and response efficacy when it comes to disease prevention. Poor health habits or a poor attitude toward health habits might sabotage protective behavioural intentions (Park et al. 2020). People's health behavioural intentions can be influenced by current practices (Chamroonsawasdi et al. 2017).

According to the model results, perceived severity, self-efficacy, and coping efficacy are all characteristics that influence protective motivated actions. People who are well-informed about their health and the severity of their diseases are more likely to adopt protective practices. Perceived self-efficacy is the strongest component among PMT constructs, as a result, numerous studies (Hernández-Padilla et al. 2020) have demonstrated that people with high efficacy are more likely to manage the disease, leading to increased protective motivation. Furthermore, perceived response efficacy is discovered to be a factor that supports self-efficacy (Leigh et al. 2020). Self-protection self-efficacy can be boosted by response efficacy. Threat assessment and coping evaluation have resulted in protective motivated actions (Tang & Feng 2018).

This research provided a detailed picture of the lives of underprivileged people living in slums. Many disadvantaged urban populations are being denied access to water and sanitation, which has serious consequences for epidemic and pandemic prevention and survival.

The study found that urban slum dwellers’ perceived susceptibility and current practices are strongly influenced by current practice, whereas severity, self-efficacy, and response efficacy are strongly influenced by current practice. Furthermore, severity, self-efficacy, and reaction efficacy are all linked to protective motivation behaviours. Although the urban slum dwellers in this study had a high degree of protective behaviour intention and protective practices, they were unable to carry out the basic COVID-19 prevention activities due to lack of clean water, insufficient sanitary infrastructure, and pricey protective equipment. Protecting a neighbourhood against a contagious disease entails protecting the entire civilization against a pandemic. This research calls for immediate action and specific help from development organizations, the government, and society to meet marginalized groups’ requirements by assuring their access to safe drinking water, sanitation, and health care. In the future, all WASH and health initiatives responding to epidemics or pandemics must include a socio-cognitive behavioural intervention aimed at enhancing people's understanding, practice, and perception of danger assessment and coping techniques.

Finally, the study used a social cognitive conceptual model to predict health behaviours by merging KAP and PMT to develop intervention programmes and policies for a deliberate behaviour change. Even though KAP and PMT have been widely utilized in psychology research to predict health behaviours, this is the first study to look at the relevance of the combined KAP and PMT in understanding COVID-19 protective behaviours in Thailand's underprivileged communities.

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

The authors declare there is no conflict.