In 2012, the United Nations (UN) Secretary General declared that the water component of MDG target 7c, to reduce by half the proportion of people without access to safe drinking water, had been met, five years ahead of the 2015 deadline [1]. By contrast, the sanitation target, to reduce by half the proportion of people without access to safe sanitation, was declared seriously off-track and unlikely to be achieved [2]. Whereas 780 million people are estimated to lack access to an ‘improved’ source of drinking water, an estimated 2.5 billion lack access to an ‘improved’ sanitation facility [2]. This apparent deficit, with global progress on extending access to safe sanitation lagging behind that of water, has been coined the ‘sanitation crisis’ and has contributed to various calls to action being issued [3].

The current MDG target for water and sanitation and the methods for monitoring its progress are part of a longer-term evolution of international goal setting and monitoring for water and sanitation [4]. Although there is one joint target for water and sanitation – “to halve, by the year 2015, the proportion of people without sustainable access to safe drinking water and basic sanitation” [2] – the respective benchmarks for water and sanitation differ significantly with regard to the level of access deemed adequate. For water, the benchmark for ‘improved’ is community-level access and, for sanitation, it is household-level access [2]. This decision to establish different benchmarks was seemingly pragmatic; a combination of evidence and experience but also realism in terms of what could be achieved at that time in light of existing levels of progress, resource constraints, and historic levels of ambition. In addition, keeping the water target at the level of the community may have been considered a progressive measure to incentivize reaching those with distant and unsafe water, rather than improving the level of access of those already served at the community-level through subsidized household connections.

The inclusion of water and sanitation within the MDG framework reflects the important contribution of these basic services to human health and wellbeing as well as to the realization of human rights. It has been estimated that as much as 6.6% of the global burden of disease is attributable to poor water, sanitation and hygiene, and this problem is heavily concentrated in low income settings [5]. In particular, diarrhoeal diseases, largely preventable with safe water, sanitation and hygiene, persist as a leading cause of child deaths globally [6]. Water and sanitation also contribute to economic development [7], education [8] and improving the nutritional status of children [9]. Beyond these specific benefits, access to “sufficient, safe, accessible and affordable” water and sanitation is recognized as a human right [10].

The extent to which these benefits differ between community- and household-level access to water and sanitation is unclear but that they do differ significantly is well established and generally accepted [11] and this is discussed further below. The human rights considerations of sufficiency, safety, accessibility and affordability are also inextricably linked to where services are located in relation to the household. The benchmarks for water and sanitation established under the MDG target however were not consistent in this regard; the minimum requirement for water was a protected community-level source, such as a tubewell, but for sanitation it was a household-level sanitation facility, such as a household pit latrine. It is striking that within one MDG target, such different benchmarks should have been established in relation to both human rights obligations and the potential health benefits.

In the coming year, a new set of development goals will be agreed and an ‘illustrative’ proposal for these has been put forward which includes a dedicated universal water and sanitation access goal [12]. However the meaning of this goal is contingent on which benchmarks will be used in monitoring changes in access to water and sanitation. In this analysis, we assess progress on water and sanitation access from 1990 through 2015 under two scenarios: the first with a community-level benchmark for both water and sanitation; and the second with a household-level benchmark for both. We refashion the current MDG target definitions of ‘improved’ in order to critically assess progress in bringing water and sanitation closer to the household where health and other benefits are greatest. A better understanding of trends in global progress for community- and household-level access to these services can support current debates as to the appropriate level of ambition and focus for new international goals on water and sanitation.

This study uses the same data sources used by the World Health Organization (WHO)/United Nations Children's Fund (Unicef) Joint Monitoring Programme (JMP) to assess national, regional and global progress toward the water and sanitation MDG target. This analysis models global progress on water and sanitation under the two scenarios described in Table 1; the first with a community-level benchmark for access and the second with a household-level benchmark. Under the ‘community-level benchmark’ scenario, both community-level access (i.e. a public water point or sanitation facilities that are shared by more than one household) and household-level access are classified as improved. Under the ‘household-level benchmark’ scenario only household-level access for both water and sanitation (i.e. only individual household water sources and sanitation facilities) are classified as improved.

It is necessary to ‘gap fill’ for certain countries and years where data are unavailable and for this we use an alternative approach to that currently used by the JMP. The methods are described in three parts: (1) the approach to gap-filling for missing data; (2) estimation of progress under Scenario 1 with a community-level access benchmark; (3) estimation of progress under Scenario 2 with a household-level access benchmark.

In total, 170 countries were included for the drinking water analysis and 169 countries for the sanitation analysis, equivalent to 99.5% and 98.4% of the global population respectively (Table 2). Countries with populations under 100,000 and/or under 100 km² total area [13] were excluded from the analysis and all countries without any data for water and/or sanitation coverage were excluded.

We estimated global progress between 1990–2015 on community and household-level access to water and sanitation, further disaggregated by rural and urban areas. Using these alternative benchmarks, we were also able to estimate whether the MDG target, refashioned in accordance with these alternative benchmarks, would be met under these two scenarios. Finally, we estimated when universal access to water and sanitation would be achieved using these two benchmarks.

Global progress on community-level drinking water and sanitation (see Table 1 for definitions) is depicted in Fig. 2. Under this scenario, with equivalent community-level benchmarks for both water and sanitation, the nominal MDG target of halving the proportion without access is met by 2010 for water and in 2014 for sanitation; both ahead of the 2015 target date. While the water target is met earlier, the rate of progress for sanitation exceeds that of water such that the difference in the proportion with access to water versus sanitation is almost halved between 1990 and 2015. Under this scenario, we estimate that in 2015 the global population without access to water and sanitation at the community-level will be approximately 517 million and 1.58 billion, respectively. We estimate that at current rates of progress, universal community-level access will not be achieved until 2025 and 2037 for water and sanitation respectively.

An equivalent household-level access benchmark for both water and sanitation was considered under Scenario 2 (see Table 1 for definitions). Global progress against this benchmark between 1990 and 2015 is shown in Fig. 3. Under this scenario the estimated levels of access in 2015 are almost equal such that approximately one third of the world's population will lack access to safe water and sanitation at a household level, equivalent to 2.35 billion and 2.46 billion for water and sanitation respectively. The rate of progress for household-level access is greater for sanitation compared to water such that, on current trends, access to sanitation at a household-level will overtake that of water in 2022. Based on the same linear regression, we estimate that universal household-level access would not be achieved until 2075 and 2061 for water and sanitation, respectively, should progress continue at the current rate.

Globally, levels of access to water and sanitation and rates of progress, whether at a community- or household-level, vary between rural and urban settings. Fig. 4 shows global progress against a community-level benchmark disaggregated by rural and urban settings. In rural areas, between 1990 and 2015 substantial progress has been made on both water and sanitation against a community-level benchmark and the gap between rural and urban levels has reduced substantially. In urban areas, there has been little change in the proportion without community-level access to water whilst the proportion without community-level access sanitation has reduced by half.

Between 1990 and 2015, levels of household-level access to water and sanitation improve at an almost equal rate in rural areas (4). Access to rural household-level coverage for water and sanitation rose from approximately 949 million for water and 855 million for sanitation in 1990, to 1.69 billion to 1.80 billion respectively in 2015. Although the proportion of the population in urban areas gaining household-level access to both water and sanitation has changed little for sanitation and remained constant for water between 1990 and 2015 (Fig. 5), in absolute numbers, we estimate a large change. Between 1990 and 2015, we estimate that an additional 1.26 billion people will have achieved household-level access to water and 1.29 billion to household-level sanitation in urban areas. These dramatic increases in absolute numbers have barely kept pace with increases in the global urban population. In 2015, we estimate that for water and sanitation respectively 32% and 34% of the global urban population will be without household-level access.

Applying equivalent benchmarks to both water and sanitation reveals remarkable similarity in progress on water and sanitation since 1990 (Figs. 2 & 3). Although the proportion of those with community-level access to water will exceed that of sanitation in 2015, the proportions with household-level access to water and sanitation will be almost equal (Fig. 3). Disparities in levels of access between rural and urban areas are also diminishing (Figs. 4 & 5).

The purpose of this analysis was to assess progress on water and sanitation between the MDG baseline and end date (1990–2015) against different benchmarks of access. The two scenarios constitute alternative benchmarks for monitoring MDG progress, with equivalent levels of access for water and sanitation in each scenario; Scenario 1 to halve the proportion without community-level access by 2015, and Scenario 2 to halve the proportion without household-level access by 2015. If the sanitation benchmark for the MDG target had, like water, considered community-level access, the target would have been met before 2015. If, alternatively, the water benchmark for the MDG target had been household level access, like that for sanitation, the target would be seriously off-track.

The often reported ‘sanitation deficit’ is apparent only when household-level sanitation access is contrasted with community-level water access, as is done under the current MDG target monitoring framework. Our analysis suggests substantial and comparable progress has been made since 1990 under the MDG target on both water and sanitation and at both levels of community and household access; and, that progress towards MDG attainment is broadly similar under each scenario. Indeed, since 1990, the rate of progress has been greater for sanitation than for water for both community- and household-level access and in both rural and urban areas.

These benchmarks correspond to important differences in the likely benefits associated with water and sanitation. The importance of distance to water source in determining the level of benefits enjoyed by the individual or household has long been recognized. In their seminal study of domestic water use in East Africa, White and colleagues observed that, “diarrhoeal diseases also seem to diminish when water supplies are made more accessible” [11]. This observation was confirmed by the findings of a later review on the effect of water supply that found the only water supply interventions to improve health were those where water was made available at or near the home [20]. More recent systematic reviews have found that increased distance to water source was significantly associated with an increased risk of diarrhoeal disease although the reasons for this association could not be elucidated [21], [22]. These findings are supported by an analysis of household survey data from 26 African countries which found that time spent walking to a household's water source was a significant determinant of under-five child health after adjustment for various potentially confounding variables [23]. One of the more persuasive arguments for this apparent jump in health benefits is that it is driven by a dramatic increase in consumption when water is available at the household-level which enables improved domestic hygiene [24].

For sanitation too there is evidence that the benefits of household level access outweigh those offered by community-level access. In a recent systematic review, Heijnen and colleagues found that that individuals using shared sanitation facilities had a higher risk of diarrhoeal disease infection in comparison to those using household facilities [25]. This is supported by common sense about relative ease of access, especially at night and for certain sub-populations (including women, children, people with disabilities and those with chronic diseases). In general, shared facilities are assumed to be less acceptable to populations and therefore less likely to be used, particularly by women [26]. Joshi and colleagues cite two examples of different types of community-level sanitation facility, one in Kenya and one in Bangladesh; the prohibitive costs of the former driving households to unsafe alternatives, and the latter, with inadequate provision for maintenance, presenting an environmental risk to the community [27]. A study of shared toilets in Bhopal, India, including facilities managed by the community, the municipal authority and private providers, found that the ratio of male to female users was 2∶1 and this was consistent for both adults and children [28]. One reason why women, in particular, may opt to not use shared facilities is the associated risk of violence they may experience, whether this be psychological, physical or sexual [29].

Current evidence suggests that household-level access is required to maximize the benefits associated with the use of water and sanitation. Indeed, many benefits are limited or negated by community-level access such that the human rights obligations of governments to progressively realize sufficient, safe, and accessible water and sanitation services will be difficult to ensure without universal household access. The rates of progress for community- and household-level access have been remarkably consistent for water and sanitation, likewise across rural and urban areas. It seems though that underlying assumptions regarding the attainability of household-level access to water have led to a substantially lower level of ambition with regard to water access as compared to sanitation.

Current proposals for the ‘post-2015’ development agenda appear to perpetuate differentiated benchmarks for water and sanitation with regard to community and household-level access [2]. Whilst universal household-level access to sanitation is deemed attainable within the horizons of a post-2015 goal – albeit with a modification to include facilities shared between a small number (<5) of households -, universal household-level water access is not. The linear progress trends estimated here do not confirm the implicit assumption that household-level access to safe drinking water is significantly less attainable than sanitation. Our analysis instead suggests that the same target year for achieving household-level access to water and sanitation is both appropriate and attainable.

The claim that sanitation lags behind water is largely an artifact of the benchmarks adopted for monitoring progress towards the MDG target. Global progress towards universal access to both safe water and sanitation at the household level, where the health and other benefits are maximized, is inadequate and the deficit is as great for water as it is for sanitation. Aligning future goals, targets and monitoring efforts with this challenge is critical to secure the full health and other benefits offered by water and sanitation. Future benchmarks should be clear and concise; equally meaningful for those households currently without these services and those with responsibility to progressively realize these goals. Expressing the targets for a future global water and sanitation target in equivalent terms of community or household-level access for water and sanitation, rather than the existing terminologies of ‘basic’ or ‘intermediate’, would focus attention on these two critical service thresholds. Rendering future targets equivalent will also reinforce the interdependency of these services, facilitating greater coordination in planning, resourcing and delivery. One path to achieving this may be a global target with a single benchmark set at the critical level of household access and for both water and sanitation together.