What the WASH Score Actually Counts
This chapter examines the measurement systems behind WASH scores in health care facilities: what the WHO/UNICEF service ladders define as “basic,” what a WASH-FIT-related proxy score records, and where those measurements stop. The practical judgment is straightforward: a score becomes more useful when its reader can name its unit, threshold, and omissions. The goal is not to distrust measurement, but to know which part of the safety problem it has actually observed.
“Basic” is a defined service level, not a verdict on care
The WHO/UNICEF Joint Monitoring Programme, or JMP, uses service ladders to make WASH conditions comparable across health care facilities. A ladder has three broad levels: basic service, limited service, and no service. These labels do not rank the quality of every clinical activity in a facility. They classify whether specified conditions are present.
For water, basic service means that an improved water source is available on the premises. “Improved” refers to the design and construction of the source—for example, a piped supply, borehole, protected well, rainwater, or delivered water source with the potential to protect water from contamination. The definition is useful because water on the premises is a concrete condition for care. It is not a complete test of the volume available during a busy shift, the reliability of supply, or the microbiological quality at the point of use.
Basic sanitation sets a more demanding composite condition. Facilities must be improved and usable, with at least one toilet dedicated for staff, at least one sex-separated toilet with menstrual-hygiene facilities, and at least one toilet accessible to people with limited mobility. The details matter because a toilet that exists but cannot be used safely by staff, patients, or people with disabilities does not provide the same service. Yet the ladder still focuses on the facility and its users' access. It does not by itself establish that wastewater or faecal sludge will be safely managed after leaving the toilet.
Basic hygiene requires functional hand-hygiene facilities with water and soap and/or alcohol-based hand rub at points of care and within five metres of toilets. This definition ties hand hygiene to two locations where contamination risk and opportunity for prevention meet. It is stronger than asking whether a hospital has a sink somewhere. But it does not prescribe how many points of care must be covered, whether supplies remain present all day, or whether workers perform hand hygiene at the moments when it is needed.
Basic environmental cleaning means that cleaning protocols are available and that all staff with cleaning responsibilities have received training. The protocol-and-training pair recognizes that cleaning is not simply the visible condition of a floor. It requires defined tasks, responsibilities, frequency, and safe methods. Even so, a protocol and trained staff do not prove that the right equipment was available, that the schedule was followed, or that contamination was controlled.
Basic waste management requires safe segregation into at least three bins and safe treatment and disposal of sharps and infectious waste. This is a critical distinction from merely having a waste container. Separation must happen where waste is generated, and the hazardous streams must reach an acceptable treatment or disposal route. The service ladder therefore captures an important risk boundary, while still leaving questions about operational functionality—such as whether treatment equipment works or has fuel—to additional measures.
The value of these definitions is their clarity. A reader can ask whether the condition was present instead of relying on an impression such as “the hospital looked clean.” Their limit is equally important: basic service is a minimum monitoring level. It is not a certificate that the facility has reliable utilities, correct behavior, clean surfaces, safe excreta management, or lower infection incidence.
Why the weakest component can govern the result
A composite indicator combines several underlying conditions into one service-level judgment. The JMP method generally uses the minimum value across available primary indicators. In plain language, the weakest required component can determine whether the facility meets the composite threshold. That rule can feel harsh when most components are strong, but it prevents a critical failure from disappearing inside an average.
Imagine a facility with usable improved toilets and toilets dedicated for staff, but little accessibility for people with limited mobility and almost no menstrual-hygiene provision. An average could make the overall sanitation picture appear comfortable. A limiting-factor approach instead asks whether the service is basic for the groups and purposes that the definition explicitly includes. Similarly, waste segregation without safe treatment and disposal is not the same as basic waste management. A set of bins cannot compensate for a hazardous waste stream that is later burned or discarded unsafely.
The rule also changes how a score should be used. If a composite result is below the basic level, the next management question is not simply “How do we raise the average?” It is “Which required component is limiting the service, and is that component the one that creates the most consequential exposure?” A low composite level may reveal a specific repairable bottleneck. Conversely, a composite result above a threshold tells a manager that defined conditions were met; it does not prove that every operational weakness has disappeared.
There is a second interpretive issue: the underlying indicators do not always come from the same observation or the same data source. The JMP report explains that primary indicators can originate in different datasets, and data may not always be combinable at the level of an individual facility. That makes the published estimate valuable for comparable monitoring while reminding readers that it is not necessarily a single visit in which every component was observed together. The closer a claim gets to a particular ward, shift, or patient pathway, the more direct the evidence needs to be.
A proxy can be useful without being complete
The Indonesian hospital study used the WASH-FIT approach as its conceptual frame, but it did not have every WASH-FIT indicator. WASH-FIT is a risk-based management approach that includes water, sanitation, hand hygiene, waste management, environmental cleaning and disinfection, energy and power, and management. The Indonesian analysis used six service areas represented in the Rifaskes data and treated their available variables as proxy indicators.
That distinction matters because a proxy stands in for a broader concept without containing everything in the concept. A proxy for hand hygiene might record the presence of clean running water or alcohol hand rub and promotional materials. It may not record whether stations are functional at every relevant point of care, whether soap is replenished, or whether staff use the stations before and after the moments specified in IPC practice. A proxy for environmental cleaning may record procedures or the availability of laundry services, while leaving the condition of rooms, floors, equipment, lighting, and cleaning supplies unmeasured.
The study itself notes missing or incomplete criteria. Its sanitation measures could not fully capture accessible facilities, amenities, or safely managed sanitation. Its data could assess facilities but not compliance with handwashing and cleaning protocols. The result can still be useful: it identifies how much of the measured readiness is present and where resources may be directed. What it cannot do is silently expand from “the available proxy items scored well” to “the entire WASH-FIT risk has been controlled.”
This is why similar words should not be merged. “Hand hygiene service,” “hand hygiene practice,” and “infection prevention” are related but different evidence objects. “Environmental cleaning protocol,” “clean environment,” and “reduced infection” occupy different points in the causal chain. Keeping them separate may make a result sound less decisive, but it makes the decision more honest.
The diagram separates the measurement objects. The JMP basic level and the Indonesian proxy score can tell us something about facility conditions. They do not directly observe all of operation, practice, environmental control, or clinical outcomes. The arrows are therefore not interchangeable: “may influence” is a mechanism claim, while “does not prove” is an evidence boundary.
The conditions that a minimum threshold can miss
Several omissions are especially important when a reader turns a service label into a safety judgment.
Reliability and sufficiency. The JMP's basic water level records an improved source on the premises, not whether water is continuously available or sufficient for a facility's workload. The report discusses WASH-FIT assessments of storage capacity for interruptions because a source can exist while the service fails during an interruption.
Quality and downstream management. Basic sanitation describes usable facilities and the needs they meet. It does not include every step of wastewater treatment or faecal-sludge management. An on-site system can contain waste at the toilet while still creating environmental risk if it is not emptied, treated, or covered safely.
Coverage and supplies. A hand-hygiene station at one point of care may satisfy a minimum definition even if other wards have gaps. A cleaning protocol may exist while disinfectant, protective equipment, or working tools are missing. A waste-treatment device may be installed while non-functionality or lack of fuel prevents safe treatment.
Behavior and experience. Users can have different experiences of the same facility. The JMP report describes patient and staff reports of missing soap, dirty toilets, poor waste disposal, and dirty points of care in a Mozambique pilot. These experiences show why infrastructure presence and user experience are related but not identical measures. They also show why a facility-level average can conceal differences between staff, patients, wards, and social groups.
None of these boundaries invalidates the basic service ladders. Minimum indicators are valuable precisely because countries need a common starting point. The mistake is to ask them to answer a question they were not designed to answer. A service ladder can establish that a specified condition is present; it cannot, without additional evidence, establish that it was reliable, used correctly, effective in controlling exposure, and followed by fewer infections.
Before comparing scores, name the thing being counted
The Indonesian study classified composite proxy scores as poor, adequate, or good, with “good” above 75 percent. The JMP basic service level is a standardized threshold built from defined primary indicators and, in composite domains, can be constrained by the weakest component. These are not two names for the same scale.
The difference is not a technical footnote. A score above 75 percent might mean that a set of available items was present in enough proportion to enter the good category. A basic service estimate might require a particular combination of minimum conditions, such as a usable sanitation facility plus user-specific criteria. One measure may reward broad availability across proxy items; another may be limited by a missing essential element. Neither is automatically more truthful. They answer different questions.
The same discipline applies to the denominator. Is the result a proportion of hospitals, all health-care facilities, a population-weighted estimate, or a subgroup such as government hospitals? Is the reference year the year of data collection, the year an estimate is assigned to, or the latest point produced by a model? Until those questions are answered, two percentages can look comparable while describing different populations and conditions.
The safest reading is therefore layered: first identify the definition, then the component conditions, then the operational gaps, and only then ask what health outcome the measure might plausibly influence. A WASH score is a map of selected facility conditions. It becomes a better map when its legend includes what it leaves outside the frame.