Does WASH FIT improve water, sanitation, and hygiene and related health impacts in healthcare facilities? A systematic review

2.1. Study overview

We systematically reviewed peer-reviewed and grey literature on WASH FIT implementation and evaluation. We extracted data on WASH FIT study types and locations, length of WASH FIT implementation, and WASH FIT domains measured in each study to assess study designs and WASH FIT domains. To determine the inputs and activities required to deliver WASH FIT, we extracted data on WASH FIT inputs and activities. We mapped activities to the WASH FIT’s intended five-step continuous quality improvement cycle (WHO/UNICEF, 2022a) and inductively developed categories for inputs based on observed similarities in the data. We used a similar approach to compile and evaluate WASH FIT outputs, outcomes, and impacts.

2.2. Conceptual framework

We used a logic model to guide our data extraction and narrative analysis. Logic models visually depict the relationships between an intervention’s inputs, activities, outputs, outcomes, and impacts (Figure 1). They help identify and communicate causal pathways related to a program or policy and are frequently used for program monitoring and evaluation (W.K. Kellogg Foundation, 2004.) We used a logic model to document the inputs and activities required to deliver WASH FIT. We assessed outputs, outcomes, and impacts within the logic model framework to evaluate whether WASH FIT improved health impacts or service-level WASH outcomes.

• Download figure
• Open in new tab

Figure 1.

Defining Components of the logic model. Adapted from W.K. Kellogg Foundation (2004)

2.3. Search strategy

We conducted systematic searches of peer-reviewed and grey literature. We performed database searches in PubMed and Scopus for peer-reviewed literature on February 13, 2024. We searched by title and abstract in PubMed and title, abstract, and keyword in Scopus for four terms: “WASH FIT,” “WASH FAST,” “Water and sanitation for health facility improvement tool,” and “Water and sanitation for facility improvement tool.” We restricted our search to studies published in 2017 or later (the publication year for the first edition of WASH FIT).

For grey literature, we searched washinhcf.org, the largest online repository of resources related to environmental health services in healthcare facilities, curated by the WHO and UNICEF. Resources are tagged with keywords describing their content when uploaded. We searched washinhcf.org on February 6, 2024, for all resources tagged as WASH FIT. We identified additional references through a 2022 report on WASH FIT case studies (WHO/UNICEF, 2022b) and consultation with experts from the WHO and UNICEF.

2.4. Inclusion and exclusion criteria

We included studies that reported WASH FIT implementation in a healthcare facility in a low- or middle-income country and described at least one domain related to WASH FIT in the logic model (i.e., inputs, activities, outputs, outcomes, or impacts). Studies of all designs that used either the first or second edition of WASH FIT were eligible for inclusion. We restricted studies to English, French, and Spanish. We included only studies that reported implementing at least one step of the WASH FIT cycle. For example, we included some studies that partially implemented the WASH FIT cycle but did not report all five steps. However, we excluded studies that reported solely on pre-implementation processes (e.g., national-level training of trainers) without describing WASH FIT implementation in a healthcare facility. A single author screened all articles.

2.5. Data extraction and synthesis

Before data extraction, we developed, piloted, and revised the extraction form. The authors met to discuss the study goals and objectives and developed a preliminary form in Microsoft Excel. Two authors then piloted the form on five studies, met to discuss the challenges and shortcomings of the extraction form, and revised it accordingly. The final extraction included the following categories: study design and context (e.g., location, design, methods, healthcare facility characteristics), WASH FIT domains used in the assessment step, WASH FIT inputs and activities, and WASH FIT outputs, outcomes, and impacts. A single author extracted data from each study. The complete data extraction form is provided in Supplemental Information File 1.

We used narrative synthesis to address our three objectives. To describe the study designs and WASH FIT domains, we described similarities and differences in how each study measured and reported WASH FIT domains. To document inputs and activities, we organized extracted data in the logic model framework and inductively developed categories for inputs based on observed similarities in the extracted data. We mapped activities to the WASH FIT cycle (WHO/UNICEF, 2022a). There were insufficient data for a meta-analysis, so we organized extracted data on WASH FIT outputs, outcomes, and impacts in the logic model framework and inductively developed categories for outputs and outcomes based on observed similarities in the extracted data. We summarized the most common effects hypothesized in our logic model.

3.1. Included studies

Our search yielded 156 unique articles. We identified 17 studies that were not identical but reported data on the same WASH FIT program (typically a grey literature report subsequently published in peer-reviewed literature). We included multiple studies on the same WASH FIT program only when they reported different data across inputs, activities, outputs, outcomes, and/or impacts. Where articles presented the same data, we excluded the older or less comprehensive version. Details of unique articles excluded based on duplicate data are provided in Supplemental Information File 1.

In total, 24 articles met our inclusion criteria. One article was a WHO report containing eight eligible case studies, each summarizing WASH FIT implementation in a different country (WHO/UNICEF, 2022b). We extracted and reported these case studies separately in our results for 31 studies included in our synthesis.

• Download figure
• Open in new tab

Figure 2.

Flow chart of studies included in a review of WASH FIT effectiveness.

We found published reports on WASH FIT implementation and evaluation in 20 of the 53 countries where WASH FIT has been implemented as of 2024 (Figure 3). Studies from the African Region and South-East Asian Region represented approximately 71% (n=22) of the studies.

• Download figure
• Open in new tab

Figure 3.

Geographic distribution of WASH FIT studies. Cropped regions of the map did not have any countries with WASH FIT studies.

More than half of the included studies (n=17, 55%) reported implementing WASH FIT in ten or more healthcare facilities, but sample sizes varied from one to 256. Of studies that reported healthcare facilities characteristics, WASH FIT implementation spanned all healthcare facility types (i.e., primary, secondary, tertiary), both publicly and privately ownership, and diverse types of health service provision, including inpatient, outpatient, maternity, neonatal, pediatric, medical, surgical, family planning, and laboratory services. The reporting of healthcare facility size varied. Some studies reported number of beds (n=4, 13%), some reported number of patient consultations (n=3, 10%), and others reported both (n=2, 6%). Fourteen studies (45%) reported no healthcare facility characteristics besides facility type or location. Of studies that reported location (n=14, 45%), 11 included HCFs in rural areas.

We identified ten studies of WASH FIT implementation in special settings or situations: three in refugee camps (Kabir et al., 2023; Morshed, 2019; WHO/UNICEF, 2022b) and seven as part of pandemic response or recovery for COVID-19 (Ashinyo et al., 2021; Hirai et al., 2021; Saadeh et al., 2022; WHO/UNICEF, 2022b), Ebola (Kanagasabai et al., 2021), or cholera (Ndumbi et al., 2020).

3.2. Study designs and follow-up time

Study designs comprised case study (n=17), cross-sectional (n=8), and quasi-experimental (n=6). Not all studies reported the length of WASH FIT implementation. Of those that did (n=19, 61%), WASH FIT implementation ranged from two days to three years, and most of these (n=11) reported implementing WASH FIT for one year or less. Studies of shorter durations typically used WASH FIT as a one-time assessment tool to identify gaps in environmental health service provision, set priorities, and inform the development of improvement plans. Nine studies (29%) recorded multiple rounds of WASH FIT assessment and systematically compared scores over time. Twelve studies (39%) completed one WASH FIT assessment to assess pre-improvement scores. The remaining ten studies (32%) did not specify how many assessments were conducted.

3.3. Evaluation of WASH FIT domains and indicators

In all studies, WASH FIT implementation and data collection was carried out before 2022 (the release date for the second edition), so we determined that all studies used the first edition. The aggregate “hygiene” and “sanitation” domains complicated our efforts to extract information on the specific WASH FIT domains measured and reported by each study. Fourteen studies reported disaggregated indicators for “hand hygiene” versus “facility environment, cleaning, and disinfection.” In contrast, two studies generically reported that they measured “hygiene” but did not specifically describe whether they measured “hand hygiene” indicators, “facility environment” indicators, or both. Similarly, 15 studies reported disaggregated indicators for “sanitation.” In contrast, three studies generically reported that they measured “sanitation” but did not specify whether they measured only “sanitation” or both “sanitation” and “healthcare waste” (Figure 4).

• Download figure
• Open in new tab

• Download figure
• Open in new tab

Figure 4.

WASH FIT domains assessed in WASH FIT step 1 (assess the facility).

Nine articles reported on all WASH FIT first edition domains and sub-domains. Studies that did not report all domains were most likely to exclude the management or hand hygiene sub-domain. Not all studies explicitly reported the total number of WASH FIT indicators measured. Among those that did (n=16, 52%), the number of indicators ranged from seven to 67. Eight studies (26%) adapted WASH FIT indicators or supplemented them with indicators from other sources. For example, Ndumbi et al. (2020) reported using WASH FIT, but the indicators did not directly match any of the first edition WASH FIT indicators. Saadeh et al. (2022) measured 150 indicators, including WASH FIT and other indicators for infection prevention and control, COVID-19 safety, and pharmacy safety.

3.4. Inputs and activities required to deliver WASH FIT

Approximately 84% (n=26) of studies reported inputs and 100% (n=31) reported activities. Among these, most reported inputs and activities qualitatively. Case studies were most likely to report inputs, while cross-sectional studies were least likely to report them (Figure 5).

• Download figure
• Open in new tab

• Download figure
• Open in new tab

Figure 5:

Reporting of inputs, activities, outputs, outcomes, and impacts among WASH FIT studies.

After reviewing similarities in our extracted data, we inductively developed the following input categories: financial support, government support, human resources, and technical assistance (Table 1). Financial support and human resources were reported both qualitatively and quantitatively, with five studies reporting specific funding amounts or numbers of staff, while others reported them qualitatively. Studies described government support and technical assistance qualitatively, though the level of detail varied. Some studies named specific government entities or international organizations, while others described receiving support from government entities or international organizations (Table 1).

Like inputs, most studies reported activities qualitatively, describing the actions taken to implement WASH FIT. Studies did not directly tie activities to the WASH FIT cycle. Still, most reported activities could be mapped to the five steps in the cycle (i.e., establishing and training the WASH FIT team, assessing healthcare facilities, identifying and prioritizing areas for improvement, developing an improvement plan, and acting, monitoring, reviewing, adapting, and improving).

We also identified a commonly reported activity that was not included in the WASH FIT cycle but was suggested as an optional preparation in the WASH FIT guide. Ten studies reported adapting WASH FIT to local contexts or current needs. For example, the WASH FIT team in the Philippines consulted with stakeholders to harmonize WASH FIT indicators with national policies and guidelines (WHO/UNICEF, 2022b). Ogando dos Santos et al. (2020) translated and adapted WASH FIT material in Venezuela, and Teme (Teme, 2023) integrated 12 indicators from a COVID-19 scorecard into WASH FIT.

3.5. Outputs, outcomes, and impacts of WASH FIT

Most studies reported quantitative outputs related to WASH FIT implementation (n=23, 74%) (Figure 5). Outcomes were reported less frequently among all three study types. Eighty-three percent (n=5) of quasi-experimental studies, 76% (n=13) of case studies, and no cross-sectional studies reported outcomes. One case study reported qualitatively that staff perceived a reduction in “communicable diseases” after WASH FIT, but no study directly measured impacts. Based on observed similarities in our extracted data, we identified four output categories and seven outcome categories (Table 2).

We classified WASH FIT assessment indicators as output indicators because these indicators are designed to be assessed in Step 2 to inform improvement plans and then re-assessed in Step 5 to evaluate outputs and performance of the improvement plan. Most studies in our review reported conducting a facility assessment as Step 2 of the WASH FIT cycle. However, only six quasi-experimental studies and three case studies assessed and reported changes in WASH FIT indicator scores over time.

We identified an additional output indicator not included in the WASH FIT assessment tool: personnel trained on WASH FIT. Studies presented data about personnel trained on WASH FIT in varied ways, with some reporting the number of personnel trained while others only shared that an unspecified number received training.

Outcomes varied more widely than outputs and were less likely to be reported quantitatively. Ten studies reported that WASH FIT contributed to policy changes or informed governments’ decisions to scale up WASH FIT. Four studies reported that WASH FIT implementation helped improve collaboration with government entities. Seven studies reported that WASH FIT helped justify allocated budgets for WASH services. Twelve studies also noted improvements in occupational safety or knowledge, attitudes, or practices among healthcare staff, patients, or community members. Most of these changes were reported qualitatively through interviews or observations. Finally, four studies indicated that WASH FIT helped improve community engagement by creating new avenues and incentives for healthcare workers to engage with community members.

No studies rigorously reported impacts. One case study reported interview debriefing with program staff, stating that “two-thirds of the interviewees cited that there is a lower incidence of communicable diseases among the staff.” However, the methods did not specify how, if at all, communicable disease incidence was directly measured. Most studies referred to a connection between improved WASH conditions and various long-term impacts to justify the importance of WASH FIT. These included reduced healthcare-associated infections (Saadeh et al., 2022; Weber et al., 2018), antimicrobial resistance (Hirai et al., 2021; Sehar et al., 2021), infectious disease transmission (Ashinyo et al., 2021), and healthcare costs (Ashinyo et al., 2021; Morshed, 2019) as well as improved quality of care (Hirai et al., 2021; Sehar et al., 2022), health-seeking behaviors (Ashinyo et al., 2021; Hirai et al., 2021), and maternal and child health (Kanagasabai et al., 2021).

4.1. Disaggregate and report results of WASH FIT assessments

By design, implementing WASH FIT will yield quantitative data regarding output indicators for environmental health services (i.e., the results of the WASH FIT assessment tool). The WASH FIT guide recommends that healthcare facilities iterate an assessment cycle every 6-12 months. WASH FIT teams should review these data to guide improvement cycles and track the overall progress and effectiveness of the WASH FIT program (WHO/UNICEF, 2022a). As these data should be collected and managed in such a way as to allow routine review as part of the recommended WASH FIT process, including them in published studies should be straightforward. Yet, we identified several gaps in reporting.

Eighteen studies (58%) measured and reported quantitative data (i.e., WASH FIT indicator scores) for three or more WASH FIT domains or sub-domains. “Hand hygiene” and “management” were the domains least likely to be reported. Domains were not reported for several reasons. First, studies reported data from the WASH FIT first edition assessment tool for “hygiene” or “sanitation” without specifying which specific subdomains were measured for hand hygiene, environmental cleaning, sanitation, or waste management. This aggregation and ambiguity impeded our efforts to evaluate changes at the output level. However, the second edition of WASH FIT separates domains for hygiene, sanitation, cleaning, and waste management in the assessment tool, which should correct this issue. Second, studies indicated that they measured a domain but did not report data—or reported qualitatively that an output had changed but did not quantify specific metrics. In these cases, reporting checklists can help ensure that all essential information is reported, aiding the interpretation of individual studies and eventual meta-analyses (von Elm et al., 2007; Zwarenstein et al., 2008). Third, studies did not indicate whether they measured a specific domain and reported neither quantitative nor qualitative data. This may reflect a reporting deficiency or an adaptation to WASH FIT in which a particular domain was intentionally omitted from the assessment and improvement process.

We recommend that future studies disaggregate and report all WASH FIT assessment data. Documenting output changes is an essential preliminary step to understand whether and how downstream effects on health outcomes and impacts may occur. Furthermore, the effort required to track and report output-level changes should be minimal, as data collection and ongoing review are essential components of the WASH FIT cycle. Where multiple assessment rounds are conducted, studies should report the results of all assessments. Where programs intentionally modify or omit specific domains or indicators from their assessment and improvement plans, studies should explicitly describe the adaptation and rationale.

4.2. Identify inputs, activities, outputs, outcomes, and impacts

We used a logic model framework to identify inputs and activities required to implement WASH FIT and its resulting outputs, outcomes, and impacts. A logic model illustrates the core components of a program and outlines expected outputs, outcomes, and impacts that will be achieved when the specified inputs and activities are delivered (W.K. Kellogg Foundation, 2004). Logic models are an important tool for program planning and evaluation. They identify required inputs and activities to support process evaluations to understand if programs are being delivered as intended. They identify expected outputs, outcomes, and impacts, which can support impact evaluations to understand if programs are working as intended. Furthermore, logic models help understand a program’s theory of change (i.e., theory of how and why a program creates change, identifying specific pathways by which inputs and activities yield short-term outputs, intermediate outcomes, and long-term impacts) (Breuer et al., 2016; Savaya & Waysman, 2005; W.K. Kellogg Foundation, 2004).

Our review suggests that the underlying logic model and theory of change for WASH FIT are poorly documented or understood. We found selected examples of inputs, activities, outputs, outcomes, and impacts that were far from comprehensive. Furthermore, we did not identify studies that proposed links between specific inputs or activities and expected outcomes and impacts.

Other studies have hypothesized logic models. Within the WASH FIT guide, the WHO and UNICEF propose five input categories (political; financial/material; human; civil society engagement; intersectoral collaboration: energy and climate, health), activities for the five-step WASH FIT cycle, five outcome categories (improved staff morale and performance; increased care seeking; improved infection prevention and control and reduced antimicrobial resistance; less environmental pollution and more sustainable waste management; more efficient use of resources and lower healthcare costs), and three long-term impacts (dignified, safe pregnancy and reduced maternal and newborn mortality; healthier, more productive families and communities; improved outbreak response and resilience) (WHO/UNICEF, 2022a). Weber et al. (2019) proposed a conceptual evaluation framework for WASH FIT that closely follows the framework developed by WHO and UNICEF and additional outcomes for changes in finance and infrastructure operations and management. Our review identified similar categories, but overall, we found that the quality of evidence was too weak to corroborate the logic model. Furthermore, we identified many studies that did not complete all five steps of the WASH FIT cycle yet still attributed changes in outputs, outcomes, and impact to WASH FIT. This suggests that the underlying logic model and theory of change may differ from what has been proposed by the WHO, UNICEF, and others.

Understanding the logic model and underlying theory of change for WASH FIT is important for implementing and evaluating WASH FIT programs. It is also important to inform adaptations so that inputs and activities may be tailored to the context without undermining the core functions of the theory of change (Anderson et al., 2022; Kirk et al., 2020; Movsisyan et al., 2021). A well-defined logic model and theory of change can also enhance WASH FIT by guiding improvement plans based on local needs. For example, we hypothesize that hand hygiene and environmental cleaning are more likely to affect impacts related to surgical site infections and other healthcare-acquired infections compared to sanitation (Carboneau et al., 2010; Lederer et al., 2009; Pittet et al., 2000). Practitioners can develop improvement plans to target salient health issues if these pathways are understood.

We recommend research to define a logic model and theory of change for WASH FIT. Prior research on costing may help identify inputs and activities. Bottom-up costing studies on environmental health services in healthcare facilities have developed frameworks to identify essential inputs and activities (Anderson, Cronk, et al., 2021; Anderson, Wren Tracy, et al., 2021). However, these frameworks are not tailored to WASH FIT programs and will require verification and likely adaptation. Process evaluation or similar techniques (Saunders et al., 2016) to examine the implementation of WASH FIT programs will likely help develop and refine a logic model.

Research will also be needed to identify outputs, outcomes, and impacts. Output indicators are the best established, with 95 indicators already available in the WASH FIT second edition assessment tool (WHO/UNICEF, 2022a). Outcome and impact indicators will prove more challenging to evaluate. Prior systematic reviews have found insufficient evidence to assess the impacts of environmental conditions on healthcare-acquired infections (Watson et al., 2019) and patient satisfaction and healthcare-seeking (Bouzid et al., 2018). We expect systematic reviews for other outcomes and impacts to yield similar results, if not less conclusive.

We suggest that initial research should emphasize qualitative, mixed methods, and other formative work that identifies broad categories and specific types of outcomes and impacts rather than quantifying individual relationships. Health impact studies are needed but will be slow and costly. Qualitative and formative research can more rapidly and cost-effectively identify and map plausible benefits (e.g., Anderson et al., 2023). Once these broader frameworks have been established, systematic reviews can be used to compile evidence for specific impacts, generate meta-analyses, identify evidence gaps, and develop a more precise research agenda.

4.3. Conduct experimental studies and exploit natural experiments

Experimental studies will be necessary to rigorously evaluate WASH FIT’s effectiveness in improving health outcomes and impacts. These studies will be costly. However, given the influence and investment in WASH FIT by countries—both countries adding line items for WASH FIT into their national health systems budgets and bilateral donors—we argue that they are important to ensure these investments are effective and identify improvement opportunities.

In parallel to experimental studies, there are opportunities to exploit natural experiments among countries’ WASH FIT programs using national health management information systems. Reducing healthcare-acquired infections is routinely cited as a benefit of WASH FIT (Ashinyo et al., 2021; Hirai et al., 2021; Saadeh et al., 2022; Sehar et al., 2021; Weber et al., 2018).

Measuring healthcare-acquired infections is expensive and requires large sample sizes for experimental trials (Blanco et al., 2019). However, many countries monitor healthcare-acquired infections as part of their national health management information systems. In countries implementing WASH FIT in a subset of regions or healthcare facilities within a region, matched controls could be identified among non-program areas to evaluate impact using existing data more rigorously. Other health outcomes and impacts could be assessed through proxies such as changes in patient volume or facility revenue to approximate changes in care seeking (Lopez et al., 2020).

We recommend that country governments, implementing partners, and any external funders incorporate more rigorous evaluation plans into WASH FIT programs. One possible solution to support countries and implementing partners is to incorporate evaluation guidelines into the WASH FIT implementation guide or create a companion guide as part of the suite of available WASH FIT materials. WHO and UNICEF have created supplemental materials, including a training manual and technical fact sheets (WHO/UNICEF, 2022c). An evaluation guide could supplement these materials. We recommend that an evaluation guide encourage WASH FIT teams to document inputs and activities needed to implement WASH FIT and report on output, outcome, and impact-level indicators to help strengthen causal pathways and evaluate the WASH FIT’s long-term effectiveness.

4.4. Limitations

We limited our search to two peer-reviewed databases and one grey literature database. Washinhcf.org is the most comprehensive grey literature database for environmental health services in healthcare facilities, but it does not actively solicit studies. Our review captured available published studies, but there are likely many unpublished reports. Dedicated efforts to make those reports publicly available would strengthen the evidence base.

We recognize that definitions to classify outputs, outcomes, and impacts vary. Our review adapted definitions from the Kellogg Foundation (2004). We classified extracted data accordingly, but arguments could be made to classify some long-term outcomes (e.g., policy changes) as impacts. Vague descriptions hindered our efforts to extract data into the logic model framework in the literature. For example, studies that reported that a specific organization or government agency gave “support” were classified as inputs. However, this could be classified as an output if the “support” indicates knowledge, attitudes, or policy-level action changes after WASH FIT implementation.