Hygiene knowledge and practices and determinants of occupational safety among waste and sanitation workers in Bangladesh during the COVID-19 pandemic

Waste and sanitation workers provide essential services to society. In most low-and middle-income countries, they are often mistreated and lack access to necessary personal protective equipment (PPE) and hygiene facilities that ensure occupational safety in workplaces. COVID-19 has also imposed serious health risks upon these worker groups. This study explores factors associated with poor occupational health and safety based on a conceptual framework. We conducted 499 surveys with five categories of waste and sanitation workers across ten cities in Bangladesh. We performed descriptive analysis and used Firth's logistic regression model following the conceptual framework. The analysis revealed consistent distinctions between workers considered to be in “safe” versus “unsafe” working conditions. The result showed that workers had not been adequately trained, not provided with proper equipment, and many had an informal status that prevented access to hygiene facilities. The workers who received occupational training, knew how to prevent COVID-19 by wearing a face mask, hand washing, and maintaining social distance, maintained protective measures, and practiced proper disposing of PPEs were more likely to be in safe condition. Initiatives to improve the situation of the waste workers who work in unsafe work conditions are still inadequate. Therefore, we recommend supplying proper protective equipment, ensuring a regular supply of gender-specific PPEs, and providing functional facilities necessary to practice personal hygiene and occupational safety, such as handwashing stations, changing rooms, and disposal facilities of used PPEs at the workplace. We also urge increased institutional management procedures, infrastructure that facilitates hygiene practices, and social policies to reduce occupational hazards for the waste workers in Bangladesh during and beyond the COVID-19 pandemic.


Introduction
Waste and sanitation workers -those involved in cleaning and managing human and solid waste -are some of the most vulnerable workers in low and middle-income countries like Bangladesh. They are often ignored, ostracised and their work undervalued ( World Bank, 2019 ). These worker groups are likely to be extremely poor and marginalised members of society who take on risky work. In some high-income coun-workers. However, efforts are fragmented, and the underlying factors responsible for weak occupational safety need to be identified and addressed more systematically to mitigate the existing hazards.
Sanitation workers mainly engage in cleaning latrines and emptying septic tanks or may take on multiple similar jobs ( Kabir et al., 2018 ). In Bangladesh, waste and sanitation workers commonly have limited access to various essential services, and the evidence of social and economic marginalisation is perceptible in their living and working conditions ( Kabir et al., 2018. They are also exposed to various occupational hazards ( CWAS & University, 2020 ). A study in Khulna, Kushtia, and Jhenaidah cities in Bangladesh showed that over 90% of septic tanks and pits were being emptied manually without any protective measures taken ( Ganguly, 2018, Karim, 2017 which is also prevalent among street or drain-cleaners ( Kabir et al., 2015 ). They are often exposed to various risks such as exhaust gases, extreme noise, poisonous/toxic substances, sharp objects and dust particles, increasing the risk of infections and injuries ( Kabir et al., 2015, Khurana et al., 2009. Waste and sanitation workers are also more prone to chronic hypertension, heart disease, respiratory disorders, kidney disease, and liver issues ( Practical Action, 2020 ). Moreover, pit-emptiers and other waste workers have neither health insurance nor compensation for the days lost due to illness or any occupational accidents ( World Bank, 2019 ).
During the COVID-19 pandemic in Bangladesh, most waste workers and their families experienced COVID-19-related symptoms ( Hoque, 2020 ). The outbreak has imposed serious health risks upon the waste workers and increased the risk of losing alternate income opportunities due to movement restrictions ( Practical Action, 2020 , WaterAid ). Since waste workers collect solid waste from residential areas, roads, hospitals, and other institutions, they are highly vulnerable to economic shocks from COVID-19 infection and spread ( WaterAid 2020 ). Most waste and sanitation workers are deprived of any handwashing facilities or soaps and sanitisers at their workplace, essential for COVID-19 prevention ( Practical Action, 2020, Root, 2020. Spending money on handwashing agents is considered a luxury since the workers earn a low salary ( WaterAid 2020 ). If they get infected with COVID-19, they usually seek treatment from nearby drug stores or administer self-treatment ( Habib & Sujan, 2020 ).
In Bangladesh, medical waste handlers are currently the most vulnerable ( Ali et al., 2017, Behnam et al., 2020 since they are highly exposed to the increased infectious wastes generated in healthcare facilities during the pandemic ( Tsukiji et al., 2020 ). Approximately 23% of medical waste cleaners in Bangladesh do not wash their hands after collecting medical waste, which is often dumped openly, directly exposing the workers ( Syed et al., 2012 ). Most waste workers lack sufficient knowledge of maintaining personal hygiene ( Zaqout et al., 2021 ) and opportunities to practice hygiene at their workplace. Most waste workers in Bangladesh appointed for sweeping the streets are female and often do not have enough facilities to maintain personal hygiene at their workplace ( WaterAid 2020 ), including menstrual hygiene. On the other hand, personal protective equipment (PPE) is inadequate ( WaterAid 2020 , Dubey et al., 2020 ), and ill-fitted for several body types ( Sakamoto et al., 2020 ). Furthermore, most workers have never received any training on occupational health and safety issues ( WaterAid 2020 , Zaqout et al., 2021 ) or how to wear and dispose of PPE, which is essential for controlling the re-emergence of viral infection .
Currently, more research is being conducted towards understanding the challenges of these waste-related jobs ( Gutberlet, 2018 ). However, little work has focused on the relationship between training and access to PPE materials, and occupational safety. Understanding infectious disease exposure, work-related injuries, and the vulnerability of waste workers is critical, especially in light of the pandemic. Underlying factors that contribute to poor occupational safety for the waste workers in Bangladesh must be explored. A well-developed conceptual framework can assist our understanding of the pattern of interconnections across knowledge and experiences. However, we found only a few conceptual frameworks available to understand occupational safety at the workplace ( Israel et al., 1996, Makin & Winder, 2008, and none were applicable to the present context of Bangladeshi waste and sanitation workers. Limited knowledge exists about waste and sanitation work, especially in the context of COVID-19, and there are few studies that quantify the factors associated with occupational safety practices.

Study objectives
The objectives of the study were to: a) assess the working conditions of different waste worker groups in ten cities of Bangladesh during COVID-19; b) explore the level of waste worker's knowledge and practice for occupational safety, infectious disease prevention, and hygiene practice during work; and c) identify the factors associated with waste and sanitation workers' occupational safety.

Study sites
We conducted this study in ten cities of Bangladesh with two different local government structures: City Corporations and Municipalities. While both Municipalities and City Corporations are self-governing bodies, their autonomy is bounded by the fact that the government serves as the Municipality's prescribed authority and holds the authority to intervene in the activities of the City Corporations. Both governing bodies were included to capture the different socio-demographic backgrounds of the waste and sanitation workers. We included three City Corporations (Mymensingh City Corporation, Cumilla City Corporation, Rangpur City Corporation) and seven Municipalities (Satkhira Municipality, Sirajganj Municipality, Cox's Bazar Municipality, Moulvibazar Municipality, Bhola Municipality, Manikgonj Municipality, Laksam Municipality). Municipalities (locally known as "Paurashava ") are the local governing bodies that deal with the state government directly ,' while the City Corporations deal with the state government through the district administration and often work with the national ministry.

Study design, sampling and respondent selection
This cross-sectional study includes five categories of waste and sanitation workers: septic tank/pit-emptiers, solid waste collectors, drain cleaners, road sweepers, and medical waste handlers.
We estimated sample size by assuming 50% prevalence of the outcome of interest among the waste and sanitation workers, a 95% confidence level, and a 5% margin of error. After adjusting by 20% nonresponse rate, the estimated total sample size was 480. For the equal allocation of the workers among these ten selected cities, we conducted 499 surveys, with 50 from each city targeted. Participants were selected randomly from the list of workers available in the City Corporations and Municipalities offices. We used a snowball sampling procedure to select unlisted (some waste and sanitation workers were not on the list as the list was incomplete) workers of the City Corporations and Municipalities.

Data collection
Before the field-level data collection, we informed each Mayor about the study via official letters and sought permission for data collection in their administrative areas. After obtaining permission from each Mayor, we collected a list of the waste workers from each City Corporation/Municipality office. Afterwards, we communicated with those workers and chose study participants among them to collect primary data. Participation was completely voluntary, and participants had the right to refuse participation at any stage. Our field staff maintained social distancing and took mandatory protective measures during the data collection period, such as maintaining safety protocols, carrying handwashing agents, wearing face masks and gloves, and cleaning devices periodically. We developed the questionnaire through an iterative process so that questions were culturally sensitive, context-oriented, and easy to understand. The research team was experienced in this research topic and we adopted a few questions from the team's previous studies. We went through an intensive literature review to determine the appropriate variables needed to examine and arranged several brainstorming sessions to finalise the questions. We made multiple modifications to the questionnaire after pilot testing, and during enumerator training sessions. To ensure that the data collectors understood the questionnaire, we arranged an intensive 5-days long training session for the enumerators before primary data collection. We recruited data collectors from the respective study locations so that they could communicate in the local language, accent, and culturally approved gestures.
The duration of this study was three months, from September to November 2020. Throughout this assessment, we conducted face-toface surveys with the waste workers to explore their present conditions, knowledge, practices, and attitude towards occupational safety and infectious diseases. Face-to-face interviews were preferred to help assess the emotions, behaviour, and non-verbal cues of the respondents ( Lavrakas, 2008 ). We designed a close-ended questionnaire that was easy to respond to, code, and analyse ( Tashakkori & Teddlie, 2010 ).

Conceptual framework
We generated a conceptual framework ( Fig. 1 ) to identify the factors associated with the occupational safety of waste and sanitation workers at their workplace considering our study objectives, current evidence and literature ( Ali et al., 2017 ;World Bank, 2019 ;Coroiu et al., 2020 ;Hosen et al., 2021 ;Kabir et al., 2015 ;Kabir et al., 2018 ;Patwary et al., 2021 ;Sorensen et al., 2021 ). The framework was used to specify a priori potential relationships between variables of interest to inform our analytical approach. The framework was beneficial to understand and explore the gaps in occupational health and safety of waste and sanitation workers.

Outcome variables
We constructed the outcome 'occupational safety' considering five indicators: have necessary safety equipment according to job type 1 (i. e for septic tank/pit emptier, required safety equipment is a face mask, apron, gloves, gumboot), received occupational safety guidelines from recruiters, received training on PPE use, cleaned working equipment with cleaning products, and have any form of PPE. Workers who responded positively to all of the above indicators were categorised as "safe, " and others were "unsafe " as negative responses to any of the indicators represent a risk.

Independent variables
We explored the socio-demographic characteristics of the waste and sanitation workers to assess the relationship with their occupational safety at the workplace. We assessed waste and sanitation workers' 'knowledge' by six binary variables and 'practices' by six variables. The items under each of these variables were merged together to create two new binary variables by considering only items with higher frequencies.
We considered waste and sanitation workers' 'attitudes' towards the use of safety equipment by a binary variable -'feeling problem with PPE', and a Likert scale variable -'the level of satisfaction with the quality of PPE'.
International Labour Organizations (ILO) defines "decent work involves opportunities for work that are productive and deliver a fair income, security in the workplace and social protection for families, better prospects for personal development and social integration, freedom for people to express their concerns, organise and participate in the decisions that affect their lives and equality of opportunity and treatment for all women and men " ( International Labour Organization, 2015 ). As such, we considered 'job facilities' here by three binary variables (i.e having incentives, receiving payment regularly, and having treatment facilities from the recruitment agencies) and one categorical variable (i.e type of accommodation facilities).
We considered the socio-economic conditions of sanitation and waste workers as a factor in the conceptual framework. This dimension focuses on the social and economic status depending on their caste, ethnicity, accommodation types, and individual income as exposure to their occupational safety measures.

Statistical analysis
We collected our quantitative data by tablets using the KoBoCollect App. We conducted an exploratory analysis to present the frequency distribution of the socio-demographic variables of the waste and sanitation workers by their working station. Afterwards, we performed bivariate analysis to measure the association between workers' different sociodemographic factors, and each of the variables within the knowledge, practice, attitudes, and job facilities with their occupational safety. We conducted a Chi-square test to estimate the prevalence difference for workers' occupational safety for the interested independent variables considered in our conceptual model. We used this association to identify the relevant variables with a p < 0.25 in our final model. We then fitted a binary logistic regression model separately among the variables within each block 'knowledge', 'practice,' 'attitude' and 'socio-economic condition' considering all the significant variables retrained from the bivariate analysis and considering other essential variables in our conceptual model to estimate the crude odds ratio (OR) and 95% confidence interval. Then we built an adjusted multivariate logistic regression model for each of the blocks separately considering those variables from the unadjusted model which were significant at the p < 0.5 level. Further, we adjusted the internal causal relationship captured among the knowledge, practice, attitude and socio-economic condition factors in our conceptual framework. We considered socio-demographic characteristics as confounders in our final adjusted model. Finally, we only considered those variables which have a statistically significant association with the outcome of interest at a p-value of < 0.05 threshold. We performed all analyses using the statistical software Stata version 14.

Ethical approval
The Ethics Review Board of the International Training Network Centre (ITN-BUET) approved this study. Enumerators read out an information sheet to the respondents in the local language and answered any questions/confusions if raised, and obtained written informed consent from the respondents for participation in the study. Respondents were given a copy of the information sheet with the contact numbers of the study lead. We also took consent from the respondents before capturing any field photos with them.

Respondents and the demographics
We found that more than half of the waste workers were appointed as "Muster Roll " employees, whereas 31% were listed as "Non-Muster Roll. " The Muster roll engagement is a semi-formal contractual service where the workers are hired for temporary services usually under a project for a certain period. Locally it is also known as 'no work no pay' job type. Though these employees often have a chance of being hired as permanent staff or extension of their present contract; except for overtime, they usually do not get any other regular job facilities (e.g., medical, insurance, accommodation, provident fund) that the permanent ( "Non-Muster Roll ") employees can. We also found that 16% of workers were unlisted (informal daily laborers). The waste workers usually received salaries based on the total number of days they worked in a month. Only the septic tank cleaners reported receiving assignments via direct calls from city authorities and individual citizens.
Around 81% of the respondents were male, and waste workers were split equally between Muslims and Hindus. Females were most likely to be road sweepers, comprising about half of the total, followed by approximately 17% being medical waste handlers. On the other hand, septic tank/pit emptiers were over 98% male, followed by drain cleaners (97%) and solid waste collectors (92%). In Mymensingh and Rangpur City Corporation and Laksham Municipality, more than 70% of the waste workers belonged to the Hindu religion, most of whom were from Harijan (low-caste Hindu) caste. Most respondents were Bengali (60%) and aged between 25-40 years. Around 45% of the workers had attained education levels within grades 1-8, whereas 48% had received no institutional education or could only sign their names.
More than half of the workers lived in either their own homes or rented accommodations. Waste workers working in City Corporations received more government-provided accommodation facilities (60%) than those of municipalities (34%). Salaries of these different types of waste workers vary with their work type and working hours. Nearly half of the total waste workers earned a salary within the range of 5000-10,000 BDT (US$ 59 -118). However, NGO-appointed workers received a higher payment than those appointed by the City Corporations or municipalities. A septic tank cleaner earned 1000-1500 BDT (US$ 12 -18) from private clients per contact, whereas the street cleaners earned 2500-3500 BDT (US$ 30 -42) per month. Medical waste handlers/solid waste collectors appointed by NGOs earned 6000-16,000 BDT (US$ 71 -189). We recorded the average monthly income of all respondents at 8362 BDT (US$ 99) ( Table 1 ).

Relationship between socio-demographic characteristics with occupational safety at the workplace
We explored the association between characteristics of the waste workers' occupational safety at their workplace ( Table 2 ). We found no statistically significant relationship between socio-demographic characteristics such as gender, age, education or caste of the waste workers with occupational safety.
There was a significant relationship between the ethnicity and occupational safety of the workers. Among the total waste workers who worked in an unsafe workplace, around 42% of the waste workers belong to the Harijan ethnicity. Accommodation conditions were also observed to have a statistically significant relationship with occupational safety-those who lived in government facilities were less likely to be in a safe category (33% vs 44%; P-value < 0.05). Furthermore, those who earned over 10,000 BDT (US$ 118) were around twice as likely to be in the safe workplace category compared to those who earned less than that.

Knowledge about occupational safety and COVID-19
Waste workers who received occupational training were more likely to be considered safe (40% vs 13%; Prevalence Difference (PD): 33) ( Table 3 ). Similarly, workers who knew about maintaining their protective gears were in safe working conditions (89% vs 49%; PD: 31). Knowledge about the spread of COVID-19 varied widely across waste workers, though it did not appear to differ substantially between the safe and unsafe group. Among these, coughing (86% safe; 83% unsafe) and sneezing (73% safe; 71% unsafe) were regarded as the prime reasons for the spread of COVID-19, followed by a social gathering (56% safe; 51% unsafe) which is more commonly known among the workers. Regarding knowledge about the prevention of COVID-19, the participants under the safe category reported wearing face masks (97%), maintaining social distance (73%) and handwashing (70%) as the prime reasons. Knowledge about the meaning of social distance also varied among the workers. Over 81% of the safe category of workers knew the meaning of social distancing as maintaining a 3-6 feet distance vs 58% in the unsafe category. Also, around 70% of workers from both the safe and unsafe workplace categories did not know how long to quarantine. On average, 85% of workers from both categories did not know how to isolate.

Current practices and attitudes regarding PPE, COVID-19 and maintaining occupational safety
Almost all the workers expressed that they had government guidelines to maintain personal hygiene in the working area. Supervisors instructed the workers to follow the guidelines. However, all the workers did not follow proper hygiene protocols. They mostly argued that the nature of their work did not provide the opportunity to maintain hygiene. Another reason is that they were not provided with proper protective equipment, which often led to various occupational injuries and diseases among them. The practices include maintaining occupational safety guidelines while working, currently using PPE at their workplace, cleaning vehicles while wearing PPEs and disposing of PPEs by burning or throwing them in respected bins show a significant association with their occupational safety. Most workers reported using PPEs (92% from the safe category; 90% from the unsafe category). After returning home, over 93% from both the safe and unsafe category of workers reported taking a bath with soap. Other standard practices such as washing hands before meals (54% and 33%) and washing clothes (64% and 52%) happened less often. Over half (54%) of the workers from the safe workplace category reported problems with PPE use, such as a feeling of discomfort which included breathing problems, sweltering problems, and improper fitting. However, over 50% of the total workers reported they were moderately satisfied with the quality of the PPE. None of the workers from the city corporation expressed strong dissatisfaction with PPE quality. However, approximately 99% of the workers from the unsafe category and 90% from the safe category reported that they did not have decent job facilities ( Table 3 ).

Worker's occupational safety
All variables within the knowledge block in the adjusted multivariate model were significantly associated with the workers' occupational safety ( Table 4 ). Workers' practices of maintaining safety guidelines at their workplace, current practices of using PPE while working, and practices of disposing of PPE after use were associated with having an occupationally safe working environment. Workers' attitudes towards the PPE and their types of ethnicity were found to be strongly associated with their occupational safety.
After adjusting the models in the conceptual framework, the estimates revealed that workers who received the occupational training had nearly four times higher odds (AOR = 3.77, 95% CI = 2.18-6.52) of being occupationally safe. Those who maintained protective gear were nearly nine times more likely (AOR = 8.70, 95% CI = 4.55-16.65) to have an occupationally safe environment than those who did not. Moreover, workers' occupational safety was less common among the workers who knew about communicable diseases (AOR = 0.49, 95% CI = 0.29-0.94), as well as those who wear face masks while handling waste (AOR = 0.57, 95% CI = 0.35-0.94). Workers who followed safety guidelines (AOR = 13.11, 95% CI = 6.31-27.24), wore PPE at their workplace (AOR = 10.93, 95% CI = 4.38-27.23), and disposing of PPE safely (AOR = 3.20, 95% CI = 1.59-6.44) were more likely to have the occupational safety as compared to the workers without such practices. Workers who felt some problems with their PPE had 3.5 times higher odds (AOR = 3.51, 95% CI = 2.08-5.90) of being occupationally safe at their workplace than workers who did not perceive such attitudes. Most importantly, occupational safety was more commonly associated with the workers having decent job facilities (AOR = 7.09, 95% CI = 2.17-23.09). Workers with decent job facilities were 7.09 times more likely to be exposed to a safe occupational workplace than those without. Workers from the Bengali ethnic group had nearly three times higher odds (AOR = 2.71, 95% CI = 1.06-6.94) of being in a safe occupational environment than workers from the Harijan ethnicity.

Discussion
The study found a strong association between waste workers' limited safety knowledge, their attitudes, hygiene practices in the workplace, socio-demographic characteristics, and lack of job facilities with poor occupational safety. Waste workers commonly face various occupational injuries, diseases, and even deaths because they are not provided with the necessary safety measures and knowledge. In Bangladesh, from March 2020, the COVID-19 pandemic has added additional health risks to their work, which was presented in this study.
Waste workers' knowledge was identified as one of the key factors responsible for occupational safety status. We observed that only a few workers participated in any occupational training arranged by their recruiters. Similarly, a severe lack of occupational safety compliances and training for different waste workers was identified by several previous studies ( Patwary et al., 2021, Patwary et al., 2012, Akram, 2016, Sarker et al., 2014, Akter & Reza, 2019. Studies also found that only 3-5% of the waste workers received institutional training in Bangladesh ( WaterAid 2020 , Zaqout et al., 2021 ). A recent study in Table 2 Bi-variate relationship between socio-demographic and socio-economic variables with the occupational safety of workers at workplace

Government facilities
Own facilities Rent 7 (35) 7 (35) 6 (30) 84 (63) 23 (17) 26 (20) 35 (33) 18 (17) 53 (50) 81 (34) 62 (26) 97 (40) 42 (33) 25 (20) 59 (47) 165 (44)  Bangladesh showed that 95% of the medical waste workers had experienced occupational accidents and 73% never participated in any occupational training ( Patwary et al., 2012 ). They require a high standard of occupational training or PPE demonstration to achieve a good level of knowledge and skills towards occupational safety measures. Workers who knew how to wear and maintain protective gear and had a positive attitude toward the quality of PPE, had a better occupational safety status. However, only a few workers chose to wear PPE regularly, which indicated that the waste workers were not provided with adequate knowledge about the importance of using PPE in preventing infections. Workers' perception of various problems (fitting issues, suffocation, heat & sweat) with their PPE were associated with their occupational safety. Therefore, distributing PPEs only is not sufficient. Instead, improving knowledge and attitudes towards PPEs through training is crucial for enhancing occupational safety. Workers who lacked the basic understanding of COVID-19 preventive measures had less occupational safety. We noted that the workers who had knowledge regarding basic hygiene practices and necessary PPE use during work increased their occupational safety at their workplace. A recent nationwide study in Bangladesh also supports this finding by concluding that regions that have people with better knowledge about COVID-19 exhibited higher rates of COVID-19 preventive behaviours ( Hosen et al., 2021 ). This further implies that providing basic knowledge to the waste workers may increase safety in their occupation as they are directly exposed and are susceptible to infection. Providing knowledge to individuals is essential because it can motivate others to prevent spreading the virus in a community, including their family members ( Coroiu et al., 2020 ). We also found that the nature of the jobs and accommodation facilities of the waste workers often render them unable to maintain adequate distancing since they did not know whether anyone infected with COVID-19 was near them. Similar results were reported by recent assessments ( Paul & Bhattacharjee, 2022 ;An et al., 1999 ).
Another significant finding from the study was that waste workers' workplaces in Bangladesh were indeed not decent, which was one of the key factors contributing to the occupational safety status. This study affirmed that all categories of waste workers manually handled various raw and harmful wastes, including biomedical wastes that severely threatened their health. A recent study in South Asian regions also indicated such health threats ( Islam et al., 2021 ). Study recommended the disposal of used PPEs and the separation of domestic waste from plasticbased hospital waste, suggesting that mixing these wastes raises the risk of disease transmission to waste workers . Even we did not find any sterilisation system or separation of biomedical wastes exists for these worker groups. In this context, Sharma and colleagues previously indicated the urgency of these practical measures for the workers who manage solid wastes to reduce the risk of spreading the infection among them ( Sharma et al., 2020 ). Moreover, a recent study across different municipalities of Bangladesh found that more than 80% of these worker groups were suffering from psychological distress ( Haque et al., 2022 ).
Our study identified that workers' attitudes and scope of hygiene practices were connected with poor occupational safety status. Such connections accelerate many negative consequences like the spread of infection among them through contaminated PPEs. An exploratory study in Bangladesh supports our findings, showing that only 19% of the study participants burned PPE-related waste, while most reported using less protective disposal measures ( Islam et al., 2020 ). On the other hand, it is estimated that nearly 40,000 workers in Bangladesh who informally collect wastes from households, clinics or hospitals are at a high risk of being infected by the COVID-19 virus if proper protection and disposal methods are not followed. But, it was evident that we lack proper management of the increased medical waste generated daily due to COVID-19, despite adopting the Medical Waste Regulation (Management and Processing) in 2008 ( Rahman et al., 2020 ). A recent study also reported improper application of even newer policies and guidelines  (50) 10 (50) 64 (48) 69 (52) 57 (54) 49 (46) 124 (52) 116 (48) 67 (53) 59 (47) 188 (50)  2) 28 (7.5) Using elbow while coughing or sneezing (yes) 6 (30) 25 (19) 4 (3.8) 16 (6.7) 10 (7.9) 41 (11) Cleaning toilets (yes) 1 (5) 9 (6.8) 2 (1.9) 10 (4.2) 3 (2.4) 19 (5.1) Washing shoes or other equipment (yes) 3 (15) 9 (6.8) 2 (1.9) 11 (4.6) 5 (3.9) 20 (5.4) Medicine (yes) 0 (0) 1 (0.8) 1 (0.9) 6 (2.5) 1 (0.8) 7 (1.9) Vaccination (yes) 0 (0) 1 (0.8) 0 (0) 1 (0.4) 0 (0) 2 (0.5) Cleanliness (yes) 0 (0) 2 (91) 2 (1.9) 16 (6.7) 2 (1.6) 18 (4.9) Don't know (yes) 1 (5) 7 (5.3) 0 (0) 10 (4.2) 1 (0.8) 17 (4.6) Other 1 (5) 1 (0.8) 1 (0.9) 10 (4.   ( Barua & Hossain, 2021 ). While in other countries like China, local waste management agencies arranged mobile incinerators for disposing of the atypical quantities of contaminated PPEs during the peak of the outbreak ( Singh et al., 2020 ). Therefore, our analysis marked a lack of necessary policies responsible for this condition. So, through this study, we want to emphasise the urgency to modify these policies that will initiate similar actions. Correspondingly, a few recent studies in Bangladesh  and other countries also suggested such actions could be, more resource allocation to tackle these infectious diseases ( Hosen et al., 2021 ), health promotion -mainly targeted toward COVID-19 prevention measures, and proper use and disposal of PPEs ( Zhong et al., 2020, Byanaku & Ibrahim, 2020. Finally, we suggest key approaches to improve the occupational health and safety of waste and sanitation workers. A tailored and specif- ically designed intervention is needed to increase the workers' knowledge and attitude and create a culture of maintaining occupational health safety. Employers could also provide practical training that would be applicable and sustainable beyond the COVID-19 period. Training is crucial to create awareness about job-related health risks and how to mitigate them. In addition to using social and behavioural change communication (SBCC) messages designed for waste workers in the training, knowledge on the proper use of PPEs should be provided, and the importance of wearing PPEs should be emphasised. Several countries have followed these strategies to improve the health and safety of their waste workers ( Kawakami & Khai, 2010 ). Moreover, to ensure their access to a safe work environment, appropriate policy measures and their implications should be taken at the government level. Institutional guidelines are also required to report and monitor health behaviours at the workplace. Along with these, work benefits like payment, job security, healthcare insurance, overtime and compensation should be improved. Lastly, it is crucial to prioritise sanitation workers in receiving vaccination to protect the safety of these frontline warriors against COVID-19 and other infectious diseases.

Limitations
Through this study, we could not assess the rates of occupational deaths and underlying causes as Bangladesh lacks a systematic record of deaths and occupational injuries among sanitation and other waste workers. A national survey would be helpful to identify waste worker groups from various regions that are most vulnerable and the reasons for that vulnerability. In addition, despite focusing on the waste workers' physical injuries and health risks, we could not conceptualise and assess the factors impacting their mental health within the scope of this study. To ensure a decent workplace for these worker groups, the impact on their mental health is also a vital issue.
We have selected field sites purposively, so our sample might not be representative to the entire waste and sanitation worker population.
However, we have selected field sites from across the country, so we have sufficiently captured variation among waste and sanitation workers in terms of socio-economic conditions, job types and facilities in our sample population.
As this was a cross-sectional study, it was not possible to directly establish causal relationships between framework variables, but clear associations noted here should provide strong hypotheses for subsequent experimental testing.

Conclusion
Sanitation and waste workers provide essential services for cities, and so protective measures must be ensured while they work in hazardous environments that cause various injuries and diseases in addition to increased likelihood contracting COVID-19. Most sanitation and waste worker do not have a clear understanding of proper preventive measures, and necessary protective equipment is not adequately provided to them. Infrastructural facilities are also inadequate to support preventive measures.

Recommendations
We recommend the following measures to improve the work conditions of sanitation and waste workers: • Provide proper protective equipment to waste workers and encouraging them to use them during work • Ensure a regular supply of gender-specific PPEs (in terms of suitability, fitness, work type, and weather-specific comfortable materials) from the recruiting authorities • Provide functional facilities necessary to practice personal hygiene and occupational safety such as handwashing stations, changing rooms, and disposal facilities of used PPEs at the workplace • Introduce national institutional programs to increase awareness, training on proper use and disposal procedures for PPE, health and safety monitoring committees, committees to oversee PPE supply, quality, and disposal to ensure the sustainability of the initiatives mentioned above • Introduce national policies to increase the salaries and benefit packages of waste workers and enable their transition to more secure and dignified jobs and work conditions.

Author contributions
MA, FS, together with TA developed the first draft of the manuscript. MZ, JBT, MR, MF, MAR, AA, DMS, SF, MH, KFT, TA critically reviewed the research design, data collection tools, and the manuscript. DMS, SF, FS, MA, MH supervised the data collection activity. DMS, FS, MH, KFT cleaned and analysed data. MH, SF, MR was involved in data analysis and provided intellectual input. All authors provided scientific inputs and critically reviewed the manuscript. The entire study was conducted under the supervision of TA.

Funding
This study was funded by the Bill and Melinda Gates Foundation and administered by the International Training Network of Bangladesh University of Engineering and Technology (ITN-BUET) -Centre for Water Supply and Waste Management.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.