What is the impact of water sanitation and hygiene in healthcare facilities on care seeking behaviour and patient satisfaction? A systematic review of the evidence from low-income and middle-income countries

Patient satisfaction with healthcare has clear implications on service use and health outcomes. Barriers to care seeking are complex and multiple and delays in seeking care are associated with significant morbidity and mortality. We sought to assess the relationship between water, sanitation and hygiene (WASH) provision in healthcare facilities (HCF) and patient satisfaction/care seeking behaviour in low-income and middle-income countries. Pubmed and Medline Ovid were searched using a combination of search terms. 984 papers were retrieved and only 21 had a WASH component warranting inclusion. WASH was not identified as a driver of patient satisfaction but poor WASH provision was associated with significant patient dissatisfaction with infrastructure and quality of care. However, this dissatisfaction was not sufficient to stop patients from seeking care in these poorly served facilities. With specific regard to maternal health services, poor WASH provision was the reason for women choosing home delivery, although providers’ attitudes and interpersonal behaviours were the main drivers of patient dissatisfaction with maternal health services. Patient satisfaction was mainly assessed via questionnaires and studies reported a high risk of courtesy bias, potentially leading to an overestimation of patient satisfaction. Patient satisfaction was also found to be significantly affected by expectation, which was strongly influenced by patients’ socioeconomic status and education. This systematic review also highlighted a paucity of research to describe and evaluate interventions to improve WASH conditions in HCF in low-income setting with a high burden of healthcare-associated infections. Our review suggests that improving WASH conditions will decrease patience dissatisfaction, which may increase care seeking behaviour and improve health outcomes but that more rigorous research is needed.

Patient satisfaction with healthcare has clear implications on service use and health outcomes.Barriers to care seeking are complex and multiple and delays in seeking care are associated with significant morbidity and mortality.We sought to assess the relationship between water, sanitation and hygiene (WASH) provision in healthcare facilities (HCF) and patient satisfaction/care seeking behaviour in low-income and middle-income countries.Pubmed and Medline Ovid were searched using a combination of search terms.984 papers were retrieved and only 21 had a WASH component warranting inclusion.WASH was not identified as a driver of patient satisfaction but poor WASH provision was associated with significant patient dissatisfaction with infrastructure and quality of care.However, this dissatisfaction was not sufficient to stop patients from seeking care in these poorly served facilities.With specific regard to maternal health services, poor WASH provision was the reason for women choosing home delivery, although providers' attitudes and interpersonal behaviours were the main drivers of patient dissatisfaction with maternal health services.Patient satisfaction was mainly assessed via questionnaires and studies reported a high risk of courtesy bias, potentially leading to an overestimation of patient satisfaction.Patient satisfaction was also found to be significantly affected by expectation, which was strongly influenced by patients' socioeconomic status and education.This systematic review also highlighted a paucity of research to describe and evaluate interventions to improve WASH conditions in HCF in low-income setting with a high burden of healthcareassociated infections.Our review suggests that improving WASH conditions will decrease patience dissatisfaction, which may increase care seeking behaviour and improve health outcomes but that more rigorous research is needed.

InTroduCTIon
The water, sanitation and hygiene (WASH) attributable burden of disease is large and concentrated within low-income and middle-income countries (LMIC).A total of 842 000 diarrhoeal disease deaths (of which, 361 000 occurred in children under 5 years old) were attributed to inadequate WASH in 145 countries. 1Despite considerable progress in improving access to WASH services under the Millennium Development Goals (MDGs), a significant proportion of the world's poor still lack access to safe WASH. 2 However, reporting for the MDGs focused on WASH access in the community.By contrast, there has been little exploration of the impact of inadequate WASH provision in healthcare facilities (HCF) in LMIC.In 2015, WHO and

What is already known?
► A WHO/Unicef report (2015) highlighted the lack of adequate water, sanitation and hygiene (WASH) provision in many healthcare facilities (HCF) in low-income and middle-income countries (LMIC).► Patient satisfaction and care seeking behaviour have been extensively used to monitor and improve the quality of care.► The evidence on the contribution of poor WASH to patient dissatisfaction and care seeking behaviour is unclear.
What are the new findings?
► This systematic review sought to assess the relationship between WASH in HCF and patient satisfaction/care seeking behaviour in LMIC.► Our findings showed that WASH status was not the main driver of patient satisfaction as other factors were more significant to users.► Nevertheless, poor WASH provision was associated with significant patient dissatisfaction and stopped women from seeking care at maternity services.► This is the first systematic review to be published on this topic.on September 15, 2023 by guest.Protected by copyright.

BMJ Global Health
Unicef assessed WASH status in 66 101 HCF in 54 LMIC. 3 This assessment showed that 38% of facilities lacked access to water, 19% had no improved sanitation and 35% had no soap and water facilities.The issue of lack of WASH in HCF is of paramount importance because vulnerable populations are over-represented in these settings and the risk of infection and death is heightened.There is a growing awareness about this issue at a national and international level and an intergovernmental commitment to address this inequity.][6] Healthcare-associated infections (HCAI) are a major challenge in LMIC, where it has been estimated that the risk is 2-20 times higher than in developed countries. 7he highest rates of HCAI have been reported from the Eastern Mediterranean and South East Asia regions (11.8% and 10%, respectively) but this is an underestimation due to poor recording and lack of patient follow-up. 7As most HCAI are transmitted via the hands of healthcare workers through direct contact or environmental contamination, hand washing remains the single most important preventive strategy. 7 8][13][14] In contrast to high-income countries, there is relatively little evidence on the burden of HCAI in LMIC.A recent systematic review estimated that HCAI prevalence in LMIC was 15.5 per 100 patients, compared with 7.1 and 4.5 per 100 patients, in Europe and USA, respectively. 15It is plausible that much of this excess is due to inadequate WASH.However, the disease burden associated with inadequate WASH provision is likely greater than the HCAI burden alone.Indeed, inadequate WASH could have large impacts on health outcomes through its influence on patient satisfaction, care seeking behaviour and staff morale.
The barriers to care seeking are characterised using the three delays model developed by Thaddeus and Maine 16 comprising: delays in deciding to seek care (primary delay), delays in reaching the health facility (secondary delay) and delays in receiving quality care once at the health facility (tertiary delay). 17Delays in receiving care have been estimated to be responsible for 30% newborn deaths in Uganda, 17 45% of child deaths from diarrhoea and acute respiratory infections in Mexico 18 and an increased odds of intrauterine fetal death of 6.6 (95% CI 1.6 to 26.3) for over an hour delays among Women in Afghanistan. 19are seeking barriers are multiple and include caretakers' failure to identify early danger signs that should trigger appropriate care seeking behaviour, cost (especially for medication), distance to the facility, impediments related to weather or social unrest, lack of supervision for other children at home, lack of transport and, particularly relevant to this review, dissatisfaction with the quality of care. 20Afsana and colleagues 21 consider that barriers to using hospital care are mainly related to care quality, especially for maternity services (often inadequate, unaffordable, insufficiently staffed and lacking medically trained professionals).Patient satisfaction is a commonly used indicator of healthcare quality and was shown to affect service use, clinical outcomes and patient retention. 22It is considered a reliable measure to understand patients' needs and to make strategic decisions to improve care quality. 23However, no standardised system exists and a wide range of patient satisfaction indicators have been used as highlighted in a recent systematic review. 23The aim of this systematic review was to assess the impact of poor WASH provision in HCF in LMIC on two relevant indicators of healthcare quality: patient satisfaction and care seeking behaviour.

MeTHods
The review methods are reported in accordance with the 'Preferred Reporting Items for Systematic Reviews and Meta-Analyses' (PRISMA) 24 (checklist: online supplementary file 1).

search strategy and inclusion criteria
Pubmed and Medline Ovid were searched in March 2016 for articles published in English after the year 2000 using the search terms outlined in table 1.A combination of specific and broad search terms was used in order to retrieve all relevant papers.'Developing countries' was included as a search term in two out of five searches so as not to exclude relevant studies.LMIC were classified based on income level as defined by the World Bank data.No restrictions on study design and duration were applied.Reference lists were manually scanned for additional relevant papers, which were included if eligible.Papers that had no WASH component were excluded.
data extraction and analysis Relevant data were extracted from all included papers using a standardised form.These data were: geographic

Key questions
What do the new findings imply?► Inadequate WASH provision in HCF in LMIC may increase the risk of healthcare-associated infections (HCAI).► Beyond the HCAI burden, poor WASH provision may increase patient dissatisfaction and limit care seeking behaviour, leading to adverse health outcomes.► Improving WASH provision in HCF should be prioritised as a means of addressing HCAI but also to address patient satisfaction and encourage timely care seeking.

resulTs
This systematic review assessed the effect of WASH in HCF on two quality of care outcomes: patient satisfaction and care seeking behaviour.Although WASH was rarely the primary focus of the included studies, all included some assessment of WASH conditions in HCF and their impact on patient satisfaction and/or care seeking behaviour.
The search strategy retrieved 984 articles (table 1).After removal of duplicates and screening of abstracts, 54 papers were considered eligible (figure 1).Following full The level of satisfaction with WASH provision was reported in most studies.However, some studies reported on composite indicators of patient satisfaction and these were also noted.The papers were categorised according to the type of healthcare system, in particular, findings for maternity services were presented separately.Additionally, three papers investigated improvement interventions.

WAsH in HCF other than maternity services
Several papers reported patient dissatisfaction rates with WASH in non-maternal health service.Woldeyohanes and colleagues assessed patient satisfaction with in-patient services in Ethiopia and reported 81.5% were dissatisfied with toilet cleanliness (table 2). 25 A study in antiretroviral treatment clinic in Ethiopia showed a lower, but significant, dissatisfaction with toilet cleanliness (35.3%). 26The authors highlighted the importance of maintaining good hygiene levels, especially for patients with HIV/AIDS.Ezegwui and colleagues investigated patients' satisfaction with eye care hospital in Nigeria and found that 71.7% of patients were dissatisfied with toilet facilities (only one toilet for patients and no running tap water). 27A study of rural healthcare system in India highlighted the link between poor WASH provision and patient dissatisfaction, with 50% respondents reporting that in surveyed government hospitals toilets are either 'not at all usable' or 'dirty needed cleaning'. 28In addition, 3% of health facilities did not have toilets and drinking water was available in only 55% of hospitals.The authors concluded that provision of clean toilets with privacy and safe drinking water would improve client satisfaction. 28While all these studies reported low patient satisfaction with WASH provision, a study in an eye care hospital in India reported high patient satisfaction with toilets (83.2%), water facilities (99.4%) and cleanliness (99.4%). 29ndeed, no respondent judged these as poor.However, 16.9% did not answer the toilet question.It is unclear if WASH provision was adequate in the HCF investigated as the paper was not focused on WASH, thus this information was not provided.
Khamis and colleagues investigated patient satisfaction with quality of care in an outpatient department in Tanzania using perception and expectation questions and calculating mean gap score between the two components. 30The study reported high overall dissatisfaction with quality of care, with a mean gap score of −2.88. 30The mean gap score was −0.5 and −0.67 for general cleanliness and sufficient chairs and toilets, respectively (table 2), showing a moderate level of dissatisfaction. 30ohammed and colleagues assessed the responsiveness of healthcare services for insured patients in Nigeria. 31One of the responsiveness domains was quality of facilities, which included provision of clean toilets in the hospital.Only 42.8% of users were satisfied with the quality of facilities and low-income users reported better quality of services than high-income users. 31Westaway and colleagues investigated interpersonal and organisational dimensions of patient satisfaction in a diabetic clinic for black patients in South Africa and found that the most important items for satisfaction were availability of a seat and a toilet in the waiting area and cleanliness. 32n a study investigating quality of care and contraceptive use in Kenya, 78.5% of facilities had running water; however, facility infrastructure and patient satisfaction indicators were not associated with contraceptive use. 33The cost of service and toilet facilities were the main areas of dissatisfaction.
Glick investigated the reliability of exit surveys frequently used to assess patient satisfaction. 34The respondents' opinions were collected and answers were compared between exit and household surveys.Courtesy bias was found to influence respondents' answers resulting in overestimates of patient satisfaction from exit surveys.This bias was stronger for subjective questions such as treatment by staff and consultation quality compared with objective questions such as facility conditions. 34sH in maternity services Nine out of 21 studies focused on WASH conditions specifically around maternal health services, covering antenatal, delivery and postnatal care.Srivastava and colleagues conducted a systematic review investigating determinants of women's satisfaction with maternal healthcare in developing countries and covered all three dimensions: structure, process and outcome. 35A good physical environment was found to be associated with a positive assessment of the health facility.In Bangladesh, when availability of services (a composite of waiting area and time, drinking water and clean toilet) was rated good, mothers were more satisfied with care quality. 35Cleanliness and maintenance of hygiene were also significant determinants of satisfaction in Bangladesh, Gambia, Thailand, India and Iran.Interpersonal behaviour, specifically provider courtesy and non-abuse, were the most widely reported determinants of women satisfaction. 35However, other factors influenced perceived maternal satisfaction including access, cost, socioeconomic status and reproductive history. 35teinmann and colleagues assessed women's satisfaction with latrines and hand washing stations in rural India and their impact on care seeking behaviour. 36They reported significant discrepancies between public and private health facilities.The average number of latrines per HCF was 2.4 (1.3 in public and 3.5 in private facilities).One healthcare centre had no latrine and dedicated latrines for woman were rarely available. 36The     Socioeconomic status was not found to be a barrier to service use when women perceived the benefits of the service to outweigh the cost, and when the service was within reasonable distance.
The table summarises the characteristics of included studies and their main findings.WASH components are presented in bold.WASH, water, sanitation and hygiene.

BMJ Global Health
mean number of hand washing stations was 2.3 (0.8 for public and 3.7 for private facilities), with two public centres lacking any hand washing facilities.WASH provision is generally acceptable in private healthcare centres but needs improvement in government facilities. 36ood reputation, competent and respected doctors and ability to deal with complications were the main factors influencing the choice of HCF.For ambulatory care, including child birth, WASH provision was considered less important compared with prolonged hospitalisation settings. 36bwele and colleagues investigated the quality of neonatal healthcare in Tanzania. 37Two per cent of mothers commented on hygiene issues and one mother suggested that improvements in hygiene were needed.Most respondents reported that the condition of toilets was as expected, while a few found them worse than expected (table 2). 37The main reason for primary delay was quality of treatment followed by cost of medical care, while secondary delay was due to distance from home, transport and complaints about unfriendliness of care workers.Tetui and colleagues investigated the quality of antenatal care in Uganda and reported that 74.6% of respondents were satisfied with care quality, while 70% were satisfied with cleanliness. 38Although data on piped water and hand washing were collected as part of the assessment, no report on WASH and patient satisfaction was provided.Infection control was a major focus and 73.4% HCF were deemed to have good infection control measures. 38MacKeith and colleagues assessed women's experience of urban maternity care in Zambia and reported that 74% would like to see general improvements; however, only 18.23% clearly expressed the need for better hygiene in toilets and bathrooms. 39abrysch and colleagues reported that women criticise dirty toilets, lack of water and aseptic practices, highlighting combined shortcomings in personal interaction, medical care and hygiene. 40They concluded that the perceived quality of care had a major influence on care seeking behaviour. 40Griffiths and colleagues investigated users' perspectives of barriers to maternal healthcare use in India through identification of key social, economic and cultural factors influencing women's decision to seek maternal care. 41Quality of care and safety issues as well as lack of WASH provision were motivating women to give birth at home.A respondent stated, 'It was safe in the house and the nurse was present to do the delivery.In the government hospital, the delivery room is not there.Toilet and water facilities are not there.So I felt safer to give birth in the house'. 41Socioeconomic status was not a barrier to service use when women considered the benefit to outweigh the cost, providing it was within reasonable distance. 41Philibert and colleagues reported that, in Burkina Faso, socioeconomic status influences patients' expectation and satisfaction, with the poorest women more satisfied with delivery environment than the wealthiest ones. 42Courtesy bias leads women to respond more positively to care quality questions, which does not reflect their true opinion. 42Courtesy bias was more pronounced for interpersonal relationships between patients and care providers, 42 which is in accordance with the findings of Glick (in a non-maternity setting). 34

Improvement interventions and accreditation in HCF
Developing accreditation standards in Ugandan hospitals was investigated by Galukande and colleagues. 43Accreditation items included physical infrastructure, infection control and waste management.While the majority of hospitals reported having infection control protocol in place, only half were recording needle stick injuries and vermin control. 43Perhaps more surprisingly, 27.5% hospitals were not tracking infection rates even for caesarean sections.In addition, the authors reported inadequate capacity to sterilise equipment in all hospitals, which would contribute to HCAI. 43The study reported good provision of running water but no mention of sanitation.Okwaro and colleagues investigated community perception of healthcare improvement intervention in rural Uganda. 44The formative research showed that many HCF (in this case malaria treatment centres) lacked running water.Following the intervention, antimalarial drug availability has improved; however, other requirements including more health workers, provision of clean water and clean toilets have not been addressed.Therefore, this intervention was not sufficient to elicit major changes or influence patients' decision about healthcare use. 44Indeed, several patients continued to seek care at inadequate heath centres.The authors reported that the main limitation of such an intervention is the focus on a particular disease and therefore failing to address multiple inadequacies observed in HCF in LMIC.
One paper investigated a criteria-based audit to improve a maternity unit in Malawi, where an initial audit resulted in the formulation of recommendations and a second audit 3 months later would report on any observed improvements. 45Significant improvements in cleanliness were achieved post audit; however, no significant changes in provision of clean toilets and bathrooms were noted. 45he authors reported that one health facility requested and obtained a new toilet, which should contribute to address the issue of inadequate WASH provision in healthcare setting. 45sCussIon Patient satisfaction is a good indicator of quality of care provided and impacts on care seeking behaviour.In the reviewed studies, inadequate WASH in HCF was associated with increased patient dissatisfaction and was even a barrier to service use in some settings (most notably maternity services).This systematic review of current evidence has informed a conceptual model of patient dissatisfaction, detailing relevant factors and repercussions of low patient satisfaction (figure 2).In this model, patient dissatisfaction results in delayed care seeking, poor health outcomes and reduced staff morale.Good

BMJ Global Health
infrastructure including adequate WASH provision is an integral part to high quality of healthcare.Inadequate WASH provision is one of the elements influencing patient dissatisfaction, though it was not found to be a major driver.Other factors relevant in resource-poor settings were significantly influencing patient satisfaction and care seeking behaviours in LMIC.The relative importance of WASH on patient satisfaction is context-specific and depends on the type of healthcare service and the length of stay.Indeed, the lack of safe WASH facilities in delivery rooms was frequently cited as the reason for women to prefer home delivery.Women expect HCF to have adequate WASH, and rightly so, as this is pivotal for their human right, dignity and infection prevention.Achieving this, however, remains a distant prospect in many healthcare settings in LMIC.
The limitations of this study include a relatively small publication window (2000-2016), which was chosen to exclude historic (or outdated) WASH provision and a search strategy that could have been further optimised to retrieve all relevant papers.Potential further limitations are the difficulty of retrieving eligible LMIC research, likely to be published in national journals not indexed in the databases searched and studies are not necessarily indexed properly (particularly regarding LMIC status/country affiliation).Finally, the studies included were mostly cross-sectional with potentially biased outcome measures and perhaps more importantly no study designed to specifically assess the causal effect of WASH provision on patient satisfaction and/or care seeking behaviour was found.The limitations of some of the included studies are related to study design, such as small sample size, lack of randomisation and patient recruitment procedures, as well as outcome measures such as heterogeneous indicators of patient satisfaction and potentially biased findings.
This review focused on WASH and patient satisfaction/care seeking because of the large disease burden associated with delayed care seeking.The link between perceived quality of care and attendance at HCF (patients who received quality care tend to return and recommend the facility to relatives) was supported by several studies and the WHO recommends the evaluation of patients' satisfaction for the improvement of HCF. 46However, perceived quality of care is highly subjective.It includes satisfaction with the outcome, the interventions and the service received (staff friendliness, availability of supplies and waiting times) as well as objective measures of care quality such as facility infrastructure, equipment and staffing. 40However, even these measures are subjective because they depend on the discrepancy between expectation and reality, strongly influenced by socioeconomic traits and subpopulation groups.Indeed, it was reported that wealthier women and patients with higher education were consistently less satisfied with delivery environment and quality of care, respectively. 31 42It was noted, however, that factors other than WASH actually drive the selection and use of health facility. 36Therefore, it is perhaps not surprising that patients continue to use HCF with inadequate WASH provision (table 2). 44he evaluation of patient satisfaction is usually performed using patient questionnaires, administered at either the HCF or households.It has been shown that exit questionnaires tend to overestimate patient BMJ Global Health satisfaction level due to courtesy bias (although this was mainly for treatment by staff and consultation quality and not facility condition). 34Intimidation bias was also reported when female interviewees felt intimidated by a male interviewer. 42Therefore, household surveys may provide more reliable estimates of patient satisfaction. 34owever, household surveys could also yield biased results as they are associated with substantial under-reporting of healthcare use, especially when the recall period was over 1 month. 47he availability of skilled birth attendants is crucial to provide emergency obstetric care and reduce maternal and newborn mortality. 48This is part of the official guidance and improving WASH provision should increase use of maternal health services in LMICs.Concernedly, a study reported higher mortality rates after obstetric care. 49The reasons were: seeking help very late and in critical condition and lack of timely and adequate care once at the health facility.Birth attendants may not provide socioculturally appropriate and respectful care leading to poor uptake. 48Previous delivery by a male provider made women choose home delivery during the subsequent pregnancy (OR 3.90; 95% CI 2.30 to 6.65). 46It was stated that 'efforts aimed at improving maternal and child health in developing countries should take cognisance of the sociodemographic and cultural underpinnings of maternal health seeking behaviour'. 50Complaints of abuse, neglect and poor treatment are common in maternity services. 51Therefore, in addition to improving facilities' infrastructure, care quality and cost-effectiveness, improvements in maternity services should also address providers' attitudes and interpersonal behaviours. 48This highlights the scale and complexity of the issues investigated and the high number of shortcomings that need to be addressed.
The importance of WASH in HCF extends beyond patient satisfaction and care seeking behaviour because inadequate WASH may also be associated with a significant burden of HCAI.Poor sanitary conditions and hand hygiene in hospital settings would result in several gastrointestinal and opportunistic infections.Unfortunately, poor hand washing practices around birth are still prevalent and continue to pose risks to mother and baby.In an observational study, the proportion of birth attendants who washed their hands prior to assisting with delivery was 24% in India, 69% in Bangladesh and 32% in Nepal. 52Hand washing of birth attendants was associated with 49% reduction in maternal mortality (OR 0.51, 95% CI 0.28 to 0.93) 52 and 19% (range 1%-34%) reduction in all cause neonatal mortality. 53Effective hand washing in HCF has benefits for a wide range of other HCAI, 54 although adherence to good hand hygiene practices is a persistent challenge.Addressing this issue requires changes in both behaviour and infrastructure; hand hygiene practices will only improve if healthcare workers are motivated to change their behaviour and when adequate facilities (taps with running water and soap) are available.

ConClusIon
The provision of adequate WASH in HCF is important to protect vulnerable populations and reduce HCAI.However, WASH provision is still inadequate in many HCF in LMIC.This systematic review assessed the impact of WASH provision on care seeking behaviour and patient satisfaction.Our review suggests that improving WASH conditions will decrease patience dissatisfaction, which may increase care seeking behaviour and improve health outcomes but that more rigorous research is needed.

Figure 2
Figure 2 Conceptual model of implications of patient dissatisfaction with care quality.The model details the interactions between patient dissatisfaction, inadequate WASH provision, care seeking behaviour and health outcomes.WASH, water, sanitation and hygiene.

Table 1
Combined search strategy and number of papers retrieved

Table 2
Impact of WASH in healthcare facilities on patient satisfaction and care seeking behaviour

Table 2
Continued on September 15, 2023 by guest.Protected by copyright.