Simulation for hand hygiene compliance and prevention of healthcare-associated infections among healthcare providers in maternity wards in Bukavu, Democratic Republic of Congo

¹Evangelical University in Africa, Medical Simulation Centre, Bukavu, Democratic Republic of Congo, ²Department of Gynaecology and Obstetrics, Panzi General Referral Hospital, Bukavu, Democratic Republic of Congo, ³All'Sims Healthcare Simulation Center, Angers University Hospital, Angers, France, ⁴"Médecins d’Afrique", Coordination Europe, Savigny-sur-Orge, France

^&Corresponding author
Jean Paul Buhendwa Cikwanine, Evangelical University in Africa, Medical Simulation Centre, Bukavu, Democratic Republic of Congo

Introduction    

Hand hygiene is a major global challenge for patient safety [1]. The term "hand hygiene" includes both washing hands with water and ordinary soap or soap containing an antiseptic, and the use of alcohol-based products (gels, solutions, foams) that do not require the use of water [2].

Hand hygiene is a simple practice that is considered to be one of the most effective methods of reducing healthcare-associated infections (HCAIs) and improving patient safety [3]. These infections occur in developed countries, countries in transition, and developing countries. They are among the main causes of mortality and morbidity among hospitalised patients [2]. At the global level, the World Health Organization (WHO) estimated in 2022 that for every 100 patients hospitalised in acute care facilities, seven patients in high-income countries and 15 patients in low- and middle-income countries contracted at least one healthcare-associated infection during their hospital stay [4]. "On average, one in ten patients dies as a result of this infection", points out the WHO [5]. In England, around 300,000 patients contract HCAIs every year [6]. Despite the paucity of reports, it is clear that the impact of HCAIs in developing countries is more pronounced. There are few data and publications on this subject in sub-Saharan Africa, where barriers to compliance with hand hygiene rules persist [7]. It is estimated that the probability of contracting an HCAI is approximately 2 to 20 times higher in developing countries, and the proportion of infected patients is over 25% [8,9].

Sepsis is a key factor in maternal and neonatal mortality, accounting for 15% of all neonatal deaths [10] and 1 in 10 maternal deaths [11]. Despite the implementation of various strategies to promote hand hygiene in healthcare establishments [12,13], compliance in maternal [14] and neonatal care [15-17] remains low.

Evaluations of quality of care have shown that compliance with hand hygiene protocols is often much lower than other evidence-based quality of care interventions [12]. There is considerable heterogeneity in hand hygiene training among infection prevention and control professionals worldwide [13]. Countries face many challenges that prevent healthcare workers from participating in educational programmes, including a lack of trained professionals and financial constraints [7].

Current approaches to improving hand hygiene practices among healthcare professionals focus largely on a limited set of known behavioural determinants (knowledge, skills, and physical opportunities such as the availability of water and alcohol-based hand rub (ABHR)) [1]. The contribution of simulation to the retention of knowledge and skills has already been the subject of studies [14-18], as has its contribution to improving compliance with hand hygiene and reducing the number of HAIs. Simulation can be used as a teaching tool to improve and reach healthcare professionals efficiently, thereby reducing healthcare-associated infections and, hopefully, patient morbidity and mortality.

This study sought to determine whether an educational intervention on hand hygiene supported by simulation sessions was effective in improving knowledge, reinforcing behavioural intention, and hand hygiene practices among healthcare workers performing deliveries in maternity units in the city of Bukavu.

Achieving this objective would provide answers to three fundamental questions on which this study was based: Does simulation-based training improve knowledge of hand hygiene more effectively than theoretical training alone? Does simulation lead to better adherence to hand hygiene practices in the maternity wards of Bukavu? Is the knowledge acquired during the training retained at 3 and 6 months?

Methods     

Study design: a randomized controlled trial was conducted in line with WHO hand hygiene (HH) guidelines in healthcare [3] and the study was reported in line with the STROBE statement for reporting observational studies [19].

The study took place in 3 phases: the first consisted of preparing training modules, scenarios (Annex 1), multiple-choice questionnaires (MCQs) (Annex 1), objective structured clinical examinations (OSCEs) (Annex 1), and acquiring the equipment used for hospital-based HH. Secondly, we trained the trainers and tested the OSCEs. Finally, the intervention and follow-up evaluation took place.

Setting: this study was carried out at the simulation centre of the Faculty of Medicine of the Evangelical University of Africa in South Kivu in the Democratic Republic of Congo, between December 2023 and May 2024. The training was carried out by a team consisting of an obstetric gynaecologist, a midwife, and a resuscitation anaesthetist, all three of whom had received training as simulation trainers. Fifty-four healthcare providers were included (11 physicians, 13 midwives, 19 nurses, 11 trainees), representing half of the available staff in four maternity wards performing ≥50 deliveries per month. A post-intervention evaluation was organised 3 and 6 months after training.

Ethical issues: authorization to conduct the study was obtained from the directors of the selected facilities. A consent form signed by each participant authorising the use of the results obtained was required. We respected anonymity and the rules governing the ethics of scientific research. Two simulation sessions were then organised for group B, after our data had been collected, for reasons of pedagogical equity.

Participants, sampling and randomization: the study population consisted of 54 care providers practising in the selected maternity units. They included doctors, midwives, nurses, and trainee doctors. In each health facility, we selected half of the total number of maternity care providers at the time of data collection. Only half of the workforce was retained so as not to disrupt the normal course of activities in the departments. In each facility, half of the maternity staff were selected by simple random sampling. Participants were then randomly assigned to two groups using R-generated sequences: group a (theory only) and group b (theory plus simulation). This sampling method helped to prevent service disruption and ensure continuity [20].

Variables, Intervention, and data collection: all observations took place at the simulation centre of the Evangelical University in Africa. At the beginning, an identification sheet was distributed, followed by a pre-test on knowledge and skills on HH and healthcare-associated infections (HCAIs), which was organized in two stages, firstly with answers to MCQs and then with OSCEs. The MCQs consisted of 15 items, 10 of which were single-answer questions and the other 5 were multiple-answer questions. The OSCEs were organised into 2 stations of 8 minutes per learner. The number of stations and the duration were determined by the limited number of trainers and the planned duration of the training. Subsequently, a nurse with expertise in hospital hygiene, assisted by members of the research team for this study, organised theoretical training. The training module was designed using the WHO guidelines on HH. The hand hygiene training focused on the definition of HCAIs, the impact of hand hygiene on patient outcomes, modes of transmission with a focus on hand hygiene, and WHO recommendations on why, when, and how hand hygiene should be implemented in healthcare settings. There were 2 theoretical training sessions. Each training session lasted 2 to 3 hours. At the end of the training, the module was distributed to all participants. For group B, two simulation sessions focusing on HCAIs and the simple hand-washing procedure (washing with tap water and soap) and hand disinfection with alcohol-based hand rub (ABHR) were carried out.

The MCQs and OSCEs were carried out as a pre-test, immediately after theory training, after simulation training, and at 3 and 6 months after theory and simulation training, depending on the group. The same MCQs and OSCEs were used for the different evaluation periods. During the evaluation, for the MCQs, a score out of 15 was given to each learner, and for the OSCEs, the practices of the agents surveyed were scored as follows: 1 = complete completion of the step, and 0 = not done or incomplete. The total score obtained from the OSCEs was added together and translated into the median percentage of correct answers.

Data analysis: all data collected from observations during the OSCEs and responses to the training questionnaires were recorded on Microsoft Excel 2019. SPSS 27 was used for statistical analysis. The results are presented in terms of the distribution of numbers for the qualitative variables and the median and the 25^th and 75^th percentiles for the quantitative variables, irrespective of the distribution. The Shapiro-Wilk test was used to evaluate the distribution of the data, which was normal for a p-value greater than 0.05. The statistical analysis aimed to answer the study’s three key questions, whilst additional analyses investigated the role of qualification in variations in practices. The Mann-Whitney U test was used to compare compliance with hygiene practices between groups A and B immediately following the training. The chi-square test was used to assess the association between provider qualifications and compliance with hygiene practices. Fisher’s correction was applied for variables with fewer than five observations in either category of the independent variable or in either group of the dependent variable. No regression models were applied to the results, as the statistical power to fit a multivariate model was limited, particularly by the sample size, which was restricted to 54. We opted for non-parametric tests suitable for small samples. The results of all these different tests were considered at a p significance level of less than 0.05 (CI=95%).

Results     

In this study, we recruited 54 healthcare providers: 11 doctors, 13 midwives, 19 nurses, and 11 trainee doctors. There were more nurses because they were the majority of maternity workers. They were divided into two randomised groups of 27 people per group.

Distribution of healthcare providers according to simple hand washing and use of ABHR: the results of this work showed that the healthcare providers who took part in the study were not familiar with HH before the training. Only 27.8% of participants had used HH, either by simple hand washing (18.5%) or by using ABHR (9.3%), prior to care. After care, 64.8% of participants used one of the HH methods. There was no significant difference between groups A and B. After training, there was a clear improvement in compliance with HH before and after care, with 75.9% and 85.2% of participants, respectively (Table 1, Table 2) (table gives a description of the population studied and their habits before and after care, and before and after training. The median experience is estimated at 5 years, with the 25^th and 75^th percentiles at 2 and 10 years. There was no statistical difference in median age between the two groups. Qualification 3 was more represented in this study with 35.2% of participants. Trainee doctors (1), midwives (2), nurses (3) and doctors (4). This table shows that there was no statistically significant). The majority used the ABHR (61.1% of participants). The evaluation of compliance with HH showed that after 3 and 6 months, participants continued to perform HH before and after care in a similar manner in both groups (p>0.05) and with almost similar proportions compared with the immediate post-training evaluation.

Knowledge outcomes (MCQs): changes in scores before and after the intervention: the median pre-test score was 40.0% (IQR 33.3-40.0) in both groups. The median score immediately after the training rose to 86.7% in both groups, with no significant difference between the two groups (p>0.05). At 3 and 6 months after the intervention, scores remained slightly higher than at the pre-test, with no statistically significant difference between the two groups (p>0.05). Both interventions improved knowledge, with no statistical superiority of simulation over theoretical training.

Clinical outcomes (OSCEs): comparison of clinical performance following simulation-based training versus theoretical training: the median pre-test score was 42.1% (IQR 36.8-47.4). After the intervention, this score rose to 73.7% (IQR 68.4-78.9) in the simulation group and to 57.9% (IQR 52.6-63.2) in the theoretical training group, with a statistically significant difference between the two groups (p<0.001). At 3 and 6 months post-intervention, the scores remained higher than in the pre-test, with a decrease of approximately 10% (Table 3 (this table presents analyses of the distributions of MCQ and OSCE scores as median percentages between the two groups and their evolution over time. No statistically significant difference between groups A and B for MCQs (p>0.05) but statistically significant for OSCEs (p<0.001))). This indicates an immediate superiority of simulation over theoretical training, without maintenance of the differential effect in the long term.

Retention over time (3 and 6 months): retention of learning in the medium and long term: scores for theoretical knowledge, as assessed by multiple-choice questions, remained above the baseline level, with no statistically significant difference between the two groups. Practical scores, as assessed by OSCEs, remained higher than in the pre-test, but showed a gradual decline of around 10%, with no significant difference between the two groups. Taking into account the qualification of the participants in the study (Table 4 (this table shows that there was no statistically significant difference (p> 0.05) for the MCQs in relation to the qualification of participants in groups A and B), Table 5), there was no statistically significant difference (p>0.05) for the MCQs in groups A and B. As for OSCEs, the study showed that before training, qualification did not have a statistically significant influence on the median OSCE percentage (p>0.05). However, in the immediate post-training period, there was a statistical difference for the other provider qualifications, except for doctors.

Discussion     

This study was carried out in the DRC, in a context of precariousness and gaps in compliance with HH in healthcare settings. Non-compliance with HH is recognised as a factor contributing to HCAI, which is one of the three major causes of maternal mortality in sub-Saharan Africa. This study enabled us to verify the contribution of simulation in the progression of knowledge and compliance of maternity care providers in the city of Bukavu on the subject of HH. The methodology used was to compare two randomised groups, one representing theoretical training and the other theoretical training plus a simulation session. It is important to point out that, despite the importance of HH in the prevention of HCAI, there are few studies comparing simulation training and theoretical training, and the impact on improving the knowledge and practices of healthcare workers in obstetric care settings.

The results of this study showed a low median HH compliance value in the 2 groups before training. However, this value improved and more than doubled after training. Regardless of the training method chosen, there was no statistically significant difference in the percentage obtained in the MCQs. In Kisangani in the DRC, Longembe and Kitronza found an overall hand hygiene compliance rate of 39% (CI95 0.37; 0.41) and that friction with hydroalcoholic solution was very infrequent (5%). They justified this low rate by the lack of trained providers, the non-existence of institutional policies to promote and monitor the application of HH, the lack of resources to support the operation of healthcare establishments, and the lack of adequate infrastructure [21]. The OSCEs in our study showed a significant difference between the two groups. Learners who received simulation training in addition to theoretical training had higher median percentages on the post-training assessment.

The WHO, in its article "a strategy for promoting hand hygiene in 2013" [22] in 6 sites in Saudi Arabia, Costa Rica, Italy, Mali, and Pakistan, showed that compliance with good MH practices had risen from 51% before the study to 67% and, at the same time, staff knowledge had also increased significantly at all sites. The study also showed that this change in safety practices and attitudes lasted for at least two years after the end of the test phase. Our study found the same increase in knowledge, raising the median value before care from 27.8 to 75.9%, and in practice as assessed by OSCEs from 36.8 to 73.7 for the group trained by simulation versus 42.1 to 57.9 for the group who had theoretical training alone.

Looking at the period during which learners observed HH more, the level of HH adherence was particularly low before care because there were several factors influencing good HH adherence. There are sometimes shortages of tap water and the absence of ABHR in the health facilities in our study environment, which could largely explain why healthcare workers are unable to practise rigorous adherence to the HH. The study showed that study participants practised hand hygiene more after than before care. This may be explained by the feeling of having to protect oneself more than the feeling of having to protect patients. In their study, Wilson et al. [23] found the same trend, which they interpreted as ritualised behaviour, performed mainly for self-protection rather than to protect patients from cross-infection. Moreover, in our opinion, this behaviour could also reflect the inadequacy of a hospital hygiene culture in which healthcare workers who consider their hands to be clean believe that they cannot pass on germs to their patients.

In studies comparing simulation and other learning methods, the results are commented on in different ways. Most have found a clear superiority of simulation over other learning methods for MH in different areas of health [14,24,25]. Bounou et al. [25] state on the one hand that educational methods based on theoretical knowledge are often inadequate for changing behaviour. On the other hand, in their study, Jansson et al. [15] concluded that a one-off educational intervention, with no audiovisual feedback or opportunity for retraining, had only minor and temporary effects on HH compliance.

In delivery rooms, training in hand hygiene should be frequent, as HAIs are the third leading cause of maternal mortality. Even with correct HH compliance, the risk of recontamination persists in the maternity ward. Gon et al. [26] reported in their study that birth attendants risked recontamination of their hands or gloves in 45.3% of cases where they rubbed, washed, or put on gloves.

Limitations: finally, there are limitations to this study, primarily a potential source of bias related to the limited sample size; this resulted in the median percentages being similar in the two groups. The second likely factor of bias is the fact that the study included providers who came from different maternity units, and their qualifications were not the same. The sample size of our study remains insufficient to meet the statistical power required for this type of study. This is due to the fact that developing countries, notably the Democratic Republic of the Congo and, in particular, South Kivu, which is the setting for our study, suffer from a significant chronic shortage of healthcare staff [20].

Conclusion     

This study showed the simulation has an immediate improvement in practice, without demonstrating superiority to theoretical training in the long term. The gradual decline of theoretical learning outcome and practical simulation-based learning outcome knowledge over the long term suggests the need for continuous refresher training. It is important to multiply efforts to limit barriers to HH compliance, whatever the working conditions.

Competing interests     

The authors declare no competing interests.

Authors' contributions     

Jean Paul Buhendwa Cikwanine, Denis Mukengere Mukwege, Denis Verron, Fabien Ganywamulume Balagizi, John Kivukuto Mutendela, and Ludovic Martin contributed to the drafting of the article and the revision of its intellectual content; Jean Paul Buhendwa Cikwanine participated in the design, acquisition, analysis, and interpretation of the data, and in the development and delivery of the theoretical training and simulation; Denis Mukengere Mukwege and John Kivukuto Mutendela participated in the design and interpretation; Fabien Ganywamulume Balagizi participated in the analysis and interpretation; Ludovic Martin participated in the design, acquisition, analysis and interpretation of data, and in the development and delivery of theoretical and simulation training. All the authors read and approved the final version of this manuscript.

Tables     

Table 1: demographic characteristics and professional qualifications of healthcare givers recruited from four maternity wards in Bukavu, Democratic Republic of Congo, between December 2023 and May 2024 (N = 54)

Table 2: hand hygiene practices (simple handwashing and/or use of alcohol-based hand rub) among healthcare givers before and after patient care, in four maternity wards in Bukavu, Democratic Republic of Congo, between December 2023 and May 2024 (N = 54)

Table 3: median scores on multiple-choice questionnaires and objective structured clinical examinations among healthcare givers, assessed at baseline, immediately post-training, and at 3- and 6-month follow-up, in four maternity wards in Bukavu, Democratic Republic of Congo, between December 2023 and May 2024 (N = 54)

Table 4: multiple-choice questionnaire results stratified by professional qualification (physicians, midwives, nurses, trainees) and training group (theory only vs theory plus simulation), among healthcare givers in four maternity wards in Bukavu, Democratic Republic of Congo, between December 2023 and May 2024 (N = 54)

Table 5: objective structured clinical examinations results stratified by professional qualification (physicians, midwives, nurses, trainees) and training group (theory only vs theory plus simulation), among healthcare givers in four maternity wards in Bukavu, Democratic Republic of Congo, between December 2023 and May 2024 (N = 54)

Annex     

Annex 1: supplementary materials (PDF - 316KB)

References