Health care waste management among health workers and associated factors in primary health care facilities in Kampala City, Uganda: a cross-sectional study

Background Healthcare wastes (HCWs) are one of the most hazardous wastes globally; second to only radiation waste. HCW management needs to be prioritized because of the devastating effects on human health and environment if not well managed. Health workers play a crucial role in management of HCWs. This study investigated the management of HCWs among health workers and associated factors in primary health care facilities in Kampala City, Uganda. Methods We conducted a cross-sectional health facility survey in 8 primary health care facilities in Kampala City from March to April 2017. We interviewed health workers who provided data on socio-demographic characteristics, knowledge, attitudes and practices on HCW management. Prevalence ratios (PRs) and their corresponding 95% confidence intervals were used as a measure of association between HCW management and associated factors. The PRs were obtained using a multivariable modified Poisson regression using a generalized linear model of Poisson family with a logarithm as the canonical link function, with robust standard errors while applying a forward elimination method. Results A total of 200 health workers responded to the survey; Knowledge of HCW management was high 143 (71.5%, 95% CI (65.2–77.8)). About 160 (80.0%) wore appropriate personal protective wear when handling HCWs. Overall, 148 (74.0%, 95% CI (67.8–80.1) had satisfactory HCW management practices. Health workers with diploma education (Adjusted PR = 1.49, 95% CI (1.13–1.96), working in the teenage corner (Adjusted PR = 1.10, 95% CI (1.01–1.29), previous training on HCW management (Adjusted PR = 1.19, 95% CI (1.01–1.42) and those who thought HCW management was important (Adjusted PR = 2.81, 95% CI (1.22–6.47) were more likely to have satisfactory HCW management practices. Conclusion The practices of health workers on HCW management were largely satisfactory. Higher odds of HCW management were determined among health workers with diploma education, previous HCW management trainings and among those who perceived HCW management as important. There is need to organize HCW management trainings in order to improve their HCW management practices among health workers.


Background
Health care wastes are considered the second most hazardous wastes globally after radiation waste [1]. Healthcare wastes encompasses various forms of waste such as sharps, human body parts, blood, chemical wastes, pharmaceutical wastes, and medical devices [2]. These wastes are majorly generated from hospitals and primary care facilities, laboratories, mortuaries, autopsy centres, laboratories, blood banks, nursing homes among others [3]. Healthcare wastes can have devastating effects on human health if not properly handled [4]. Mismanagement of HCWs can result into various hospital-acquired infections, occupational health hazards and food contamination [5]. In addition, mismanaged wastes such as sharps contaminated with blood facilitates transmission of infections such as hepatitis B, hepatitis C, HIV/AIDs and other viral infections [2,6,7]. Health care workers, patients, workers in support services, visitors to health facilities, waste handlers, scavengers, fetuses in the wombs, and the general public are susceptible to these infections [2,8].
In developing countries especially in Africa, healthcare waste has not received the much needed attention that it deserves [9][10][11]. This is because of the inadequate resources in these countries resulting into low priority for HCW management [12]. In many countries, there is limited segregation of hazardous and medical wastes and usually mixed with non-infectious waste [13,14]. Inadequate knowledge and unsatisfactory management practices among the health care workers are major challenges in the management of HCWs [15]. Previous research indicate that HCW management may be affected by lack of formal training, lack of knowledge on HCW management, limited interest from hospital administration [16][17][18].
In Uganda, the HCW generated is averaged at 92 Kilograms (Kg) per day in hospital while primary health care facilities (Level IV health centers, level III health Centre and level II Health Centre) generate about 42Kg, 25 kg and 20 kg respectively on daily basis [19]. Moreover, most of these primary health care facilities lack proper HCW management facilities [20,21]. Despite the policy guidelines on injection and HCW management developed by the Ministry of Health, Uganda, there is sufficient evidence that HCWs including in Kampala health facilities is not properly handled [22,23]. Although health care workers are routinely involved in managing HCW at their workstations, there is little published data on HCW management among health workers and more especially in primary health care facilities. Previous studies [16,23,24] on HCW management in East Africa were mainly conducted in high level health facilities (hospitals) and not primary health care facilities. These studies also delved more on HCW at facility management level and did not explore HCW management among health workers. There is paucity of studies conducted to assess the management of HCWs among health workers in primary health care facilities and factors that may influence it.
We therefore set out to understand HCW management practices among health workers in primary health care facilities in Kampala and the factors that influence those practices.

Study design, setting and study population
The study was cross sectional in design and employed quantitative techniques of data collection. It was conducted in Kampala, Uganda's Capital City between March to April 2017. According to 2014 census results, Kampala has an estimated resident population of 1,507,080 people with an annual growth rate of 2.02% [25].
This study was conducted among health workers in 8 primary health care facilities in Kampala. In Uganda, the health care system is organised into a four-tier system (i.e., hospitals, health centres of levels IV, III and II). Hospitals are considered high level facilities while health centres; II, III and IV are lower level and are classified as primary health care facilities. These primary health care facilities offer basic services for example Health centre IIs offer out-patients consultations, health centre IIIs offer inpatient care and some laboratory services. In addition to services offered at lower levels, Health centre of level IV offer some caesarean operations and blood transfusion services. The main waste streams generated in these facilities included sharps, pathological wastes, infectious wastes and general wastes.
The study units were the primary health care facilities and the study participants were health workers. In this study, we defined a health worker as a person who works at health care centre and is engaged in all actions aimed primarily on enhancing health of patients. These health workers included nurses, midwives, medical officers, laboratory personnel, counsellors and social workers. All selected health workers who consented were interviewed. Those who were absent on leave or medically not well were excluded.

Sample size and sampling procedure
The Kish Leslie formula for cross sectional studies was used to calculate sample size of 231 health workers [26].
We considered, the following assumptions: Z score, Zα = 1.96; prevalence, p = 16.3% (adopted from a study on adherence to waste disposal guidelines among health workers in Kenya) [27], a precision, σ = 5% and non-response rate of 10%. It is important to note that actual population of health workers in these primary health care facilities in Kampala was small (< 500).
Proportionate sampling was done to select a specific number of health workers from each of the 8 primary health care facilities based on the number of health workers in each health facility (Table 1). This was based on the numbers of health workers obtained from Kampala Capital City Authority (KCCA).
From each of the health facilities, stratified proportionate sampling was again employed to select the different cadres of health workers (nurses, midwives, medical officers, laboratory personnel, counsellors and social workers).

Data collection
We used a semi-structured questionnaire to collect quantitative data from health workers on socio-demographic characteristics, knowledge of HCW and perceptions of risk due to HCW and to capture health workers' HCW management practices such as segregation, availability of necessary accessories, occupational safety and health in their work stations. The tools were developed based on findings from published studies [1,9,23]. We pretested data collection tools with 15 health workers in a health centre in the neighbouring Wakiso district. Pretesting was done to mitigate any errors and avoid ambiguity in preparation for actual data collection. This facility was chosen because it is close to Kampala and has similar HCW challenges as the primary health care facilities in Kampala.

Data management and analysis
Data were checked daily for completeness, entered using EpiData version 3.02 (EpiData association, Denmark). We used Stata 14.0 (Statacorp Texas; USA) for cleaning and analysis. To determine the status of HCW management (Outcome variable), which was classified as either satisfactory or unsatisfactory, seven questions were asked on practices on healthcare waste management with the responses "Yes and No". "No" was assigned 0 and "Yes" was assigned 1. We used mean scores to divide the HCW management status of health workers into two groups as suggested in the SAGE encyclopaedia of communication research methods [28]. As such, respondents who scored above 4.995 (average score) on 7 questions were considered as having satisfactory healthcare waste maintenance practices and those that scored less than 4.995 were deemed not to have satisfactory healthcare waste management practices. Prevalence rate ratios (PRs) were computed using a generalized linear model of the Poisson family with the logarithm as the canonical link function, with robust standard errors while applying a forward elimination method. This was done to measure the association between the outcome and independent variables. Prevalence rate ratios (PRs) were used instead of odds ratios since the prevalence of the outcome variable was > 10%, yet logistic regression's odds ratios tend to overestimate the relative risk in such instances [29,30]. Simple models consisting of the outcome and one independent variable were run to obtain the crude PRs. In the multivariable model, variables that were significant at simple models (p < 0.05) and those that had plausibility with the outcome were included. The crude and the adjusted PRs and their corresponding 95% confidence intervals were presented. Level of knowledge which was a predictor variable was also classified as either good knowledge or poor knowledge, five questions were asked on knowledge on healthcare waste management with the responses "Yes and No". "No" was assigned 0 and "Yes" was assigned 1. Respondents who scored above 4.6 (average score) were considered as having a good knowledge and those that scored less than 4.6 were deemed to have poor knowledge on HCW management. Level of knowledge was however excluded in the final regression model due to collinearity. Strengthening the reporting of observational studies in epidemiology (STROBE) guidelines informed the reporting of our study [31].

Discussion
This study aimed at understanding management of HCWs and associated factors among health workers in primary health care facilities in Kampala, Uganda. Our findings show that majority of the health workers had high knowledge on HCW management; knew how wastes are segregated and the risk to health. Most of them, used waste bins and wore personal protective wear when handling HCWs. A high proportion had satisfactory HCW management practices. Health workers in teenage corner, those who had attended training on HCW management and those who thought that HCW  [32,33]. The prior trainings on HCW management that most health workers had, partly explains the high knowledge. In our study, 60% of participants had prior trainings consistent with what has been reported in other studies which indicated   adding that this has to be complemented with continuous trainings [37,38]. Wearing personal protective equipment (PPE) such as gloves, masks, clinical coats, shoes helps to minimise exposure to infections and injuries [39]. In this study, most health workers wore appropriate (PPE) which is a good practice since it minimises risk of contact with the waste. Our findings did not corroborate with findings of a cross sectional study conducted in a Tanzanian Municipality in which most health workers did not wear appropriate personal protective gear [40]. The low usage in the aforementioned study was attributed to the fact that health workers were not provided with protective gear by their employees. It is appropriate to ensure adequate provision of PPE and then supervision for proper and consistent use.
Generally, the practices of health workers on HCW management were satisfactory which relates to appreciable knowledge as there was significant association between practice and knowledge scores (p < 0001). The satisfactory practices suggest that health workers might be less likely to experience adverse effects associated with poor handling. Our finding corroborates with findings of a similar study conducted in Egypt in which most health workers had satisfactory practice scores [41]. It was found that health workers with diploma education were more likely to have satisfactory HCW management practices as compared to those with higher secondary education, a finding which is understandable given that a diploma generally offers better opportunity to acquire extensive knowledge on HCW management and yet such opportunity is almost non-existent in secondary school. Continued professional development may help improve the practices of health workers on management of HCWs.
Health workers who had received training on HCW management were more likely to have satisfactory practices. A possible explanation may be because they are able to put into practice what they have trained in. Our findings also support the findings of a similar study in Ethiopia which revealed that health workers trained on healthcare waste management were more likely to exhibit satisfactory practices on HCW management [42]. Trainings should therefore be intensified as they have shown to improve practices of health workers regarding how they handle HCW.
Health workers who thought that HCW management was important were more likely to have satisfactory HCW management practices as compared to those who thought otherwise. This finding is understandable because it is expected that such health workers based on their attitude would take practical measures to properly handle HCWs. It has been shown that health workers are more likely to be cautious and take necessary measures when they realise that HCW poses risk [42]. Health workers in the teenage corner were more likely to have satisfactory HCW management practices as compared to those in the outpatient clinic. This is understandable because further analysis found that all health workers in the teenage corner had diploma education unlike in the Outpatient department. The fact that higher education levels were associated with satisfactory practices may have modified the association between department and HCW management.

Limitations
This study had some limitations; firstly, the sample size was relatively small which limits the generalizability of