Framing Water, Sanitation, and Hygiene Needs Among Female-Headed Households in Periurban Maputo, Mozambique

Abstract

Objectives. Water, sanitation, and hygiene challenges in the global south require analyses that capture more than urban–rural differences. A new taxonomy is required to help systematize and respond to basic sanitary needs. My aim was to test a new framework for understanding these concerns in periurban spaces.

Methods. I conducted semistructured qualitative interviews with a randomized sample, stratified by settlement density, of mostly female-headed households in KaTembe, the largest municipal district of Maputo, Mozambique. The survey included questions on the adequacy, accessibility, and affordability of water, sanitation facilities, and waste management as well as awareness of illnesses and safe hygiene practices.

Results. Despite being part of a capital city, KaTembe residents face a diverse mixture of sanitary challenges, as revealed through an analysis of adequacy, accessibility, affordability, and awareness issues. The interaction of these 4 lenses provides insight into residents’ behaviors and the obstacles they face in securing adequate provisions.

Conclusions. International water, sanitation, and hygiene studies continue to depend on urban–rural distinctions. However, an adequacy, accessibility, affordability, and awareness framework can improve the utility of their data.

Water, sanitation, and hygiene concerns in the global south require analyses that capture more than urban–rural differences. The numerous surveys conducted to date that focus only on urban–rural categorizations fail to highlight the important mixture of vulnerabilities in different types of settlements. Water, sanitation, and hygiene concerns in settlements with mixed urban and rural characteristics—especially typical in periurban areas—demand improved survey methodologies that can effectively capture the rich texture of challenges in such environments. The analytical lens of accessibility, adequacy, affordability, and awareness of basic needs provides a strong framework through which to better systematize data on water-, sanitation-, and hygiene-related needs in diverse settlements.

Cities are often depicted as locales where, as a result of population density, the accessibility of basic needs (i.e., distance from basic services) is not as problematic as in rural areas.^1–3 Likewise, the adequacy of such provisions (e.g., the volume and quality of water available or of sanitation facilities used) in urban areas is generally considered better than in rural districts because of the quality control mechanisms or coverage incentives of provider institutions. Affordability concerns in urban areas have been allayed by the presumption that higher or more stable wages and income opportunities allow urban residents access to basic services. Urban residents are also often described as having more diverse exposures to and knowledge of various health risks.

These generalizations about urban vulnerabilities regarding the accessibility, adequacy, affordability, and awareness of basic needs have been effectively revealed as inaccurate and misleading in the urban development literature. Over the past decade, intraurban inequalities have been increasingly well documented in the global south, particularly in case studies from countries including India, South Africa, and Brazil, where a significant number of residents live in precarious urban settlements.^4–9 In the present study, in an attempt to add to this compendium, I examined the situation in Mozambique, focusing on one of the districts of the capital city of Maputo. My analyses call into question the utility of continued analytical dependence on categorizations such as urban and rural in internationally comparative health survey reports.

There is little international agreement on what constitutes urban or rural locales, save that these definitions or demarcations are subject to diverse national interpretations and should correlate research objectives with the appropriateness of the defining variables used (e.g., density, economic base, lifestyles; for additional discussion of urban–rural definitions, see United Nations Statistics Division¹⁰). Yet, health statistics—particularly water, sanitation, and hygiene data—continue to be aggregated into urban versus rural categories without reference to typologies of settlements and often without clear indication of how these categories have been defined.

Data on Mozambique represent a case in point. The Global Health Observatory of the World Health Organization (WHO) indicates overall deficiencies in the use of improved water sources and sanitation facilities but also strong divergence in urban and rural settings. For example, from 1990 to 2008, the percentage of urban Mozambicans with improved drinking water sources increased slightly from 73% to 77%, and the percentage with improved sanitation facilities increased from 36% to 38%. During the same period, the percentage of rural Mozambicans with better drinking water increased from 26% to 28%, but there was no increase in the percentage with improved sanitation facilities, which remained at an unacceptably steady 4% between 1990 and 2008.¹¹

Today roughly 7 million of Mozambique's 22 million residents live in urban areas composed of cities and towns almost entirely with populations of less than 1 million. This urban population is estimated to reach 50% of the country's total in the next 10 years.¹² The capital city of Maputo is already home to roughly 1.2 million residents. However, the residents of Maputo's 7 municipal districts do not live in urban areas alone, and thus an urban categorization belies the diversity of the city's settlements.

I focused on water-, sanitation-, and hygiene-related needs in the largest periurban district, KaTembe, which comprises 45% of Maputo's total land area. It is significant for its lack of density (with only 119 people/km², compared with other districts nearing 20 000 people/km²), as well as for its poverty incidence of more than 70%, the highest in the city.¹³ Despite the geography of the area, boasting a long coastline and inland ponds, KaTembe's population also complains of water inadequacy. In other words, from a perspective of density, income, and geography, KaTembe is a municipal district with a decidedly inland, rural feel. Given that periurban populations such as that of KaTembe are not atypical in the global south, its residents’ stories can help provide an understanding of how an accessibility, adequacy, affordability, and awareness lens on basic needs can improve on the urban–rural distinction used in many international surveys addressing the health situation in such areas.

METHODS

I drafted a survey instrument to assess residents’ water, sanitation, and hygiene concerns before I arrived in KaTembe in 2010 and later revised the instrument on-site in consultation with 2 local professionals. The first was the secretary (chief of staff) of the Municipal District of KaTembe, Jorge Aissa, who has had extensive experience in the water and sanitation sectors in Mozambique. The second was Odete Muximpua, a professional staff member of the Water and Sanitation Program in Mozambique, who has had significant experience conducting household surveys in vulnerable neighborhoods in Maputo.

My final questionnaire had 5 parts. The first section gathered data on respondents’ sociodemographic characteristics, including gender, age, employment, household makeup, and educational attainment. The next 3 sections were dedicated to exploring adequacy, accessibility, and affordability concerns in the areas of water, sanitation, and waste management. The final section, included at the bequest of the local government district office, asked about hygiene, in particular water-, sanitation-, and hygiene-related illnesses and personal hygiene practices, topics about which the local government wished to gather additional data.

I used a semistructured survey format to gain a perspective on basic services and hygiene practices in KaTembe, initially asking some specific and some open-ended questions about water, sanitation, and waste management services. For example, to obtain information on accessibility issues, I asked interviewees what types of water resources they used, whether they used different water sources to gather water for human consumption and water for cleaning, the time required for them to gather water, how they evaluated the distance to their most used water sources, how many standard (20 or 25 L) buckets their respondent households used in 1 day, whether they had their own latrine or other sanitation system, whether open defecation was a problem in their neighborhoods, and whether their neighborhoods had waste removal services.

Adequacy questions centered around the quality of the water consumed (assessed according to smell, taste, and color), the frequency of disruptions in water availability, the type of latrine used (traditional or improved [i.e., covered]), the latrine's quality vis-à-vis past years, how many people used the household latrine, whether the volume of waste and waste removal services had increased in recent years, and what households did with waste not removed. Affordability concerns were addressed in questions about how much respondents paid for water per bucket or per access to a maintained pump, how often they rebuilt latrines and whether they hired anyone to do so for them, whether they paid for removal of waste, and the ordering (from most costly to least costly) of basic living expenses (e.g., expenses related to food, water, electricity, housing, health care, and transportation).

Questions in the final section on hygiene awareness were also both specific and open-ended, including questions about how and where respondents kept water; the frequency and manner of cleaning buckets and cups; practices used to maintain latrines; knowledge of water-, sanitation-, and hygiene-related illnesses and how to treat them; and illnesses among household members in the preceding month. All interviews were between 45 minutes and 1.5 hours in duration depending on interviewee responses and whether they required translation between Portuguese (which I speak) and Changana, the local Bantu language largely spoken in southern Mozambique (which I do not speak fluently).

To identify household respondents, I used Google Maps to print out aerial-view maps of KaTembe, and, in conjunction with updated views of KaTembe from Google Earth, I stratified settlement categories within the district by density levels. On these printed maps, I marked residential settlement areas as having high, medium, or low relative densities according to the clustering of shelters. I used a density stratification approach because settlement densities are important in the design of basic services such as water and sanitation.¹⁴

Together with my partner in this research, nongovernmental association Associação IMAGINE, I then randomly chose households within each type of density group, targeting 12 respondents in high-density areas, 9 respondents in medium-density areas, and 6 respondents in low-density areas. In total, I conducted 27 qualitative, semistructured interviews in July 2010. Administratively, the 27 random households were located in 4 of KaTembe's 5 formally recognized subdistrict neighborhoods (“bairros”): Guachene, Chalí, Chamissava, and Incassane. Imagine provided me with transportation to the farther reaches of KaTembe, and the organization's locally based expert staff helped with translations to and from Changana into Portuguese.

The typical resident with whom I spoke was a middle-aged woman with little or no education who was engaged in subsistence agriculture and cared for a household with anywhere from 1 to 6 members younger than 18 years. Although, according to the most recent census (2007), KaTembe's population of roughly 21 000 is only 52% female, almost all of my interviewees were female. Women and children traditionally gather water and manage sanitation and hygiene,^15,16 and accordingly 26 of my 27 interviewees were conducted with adult women, almost all of whom reported themselves as the head of the household.

The age range of respondents was wide, with the youngest interviewee aged 18 years and the oldest aged 80 years. However, the majority (74%) of the interviewees fell between the ages of 25 and 65 years, with the median being aged 52 years. Thirteen interviewees had no education, another 9 attended some primary school, 1 had completed primary school, and the remaining 4 had completed some secondary school. In terms of livelihood, 14 interviewees were engaged in subsistence farming on small plots of land around their home. Four respondents earned their living either in housekeeping or by cleaning streets for the municipal government, another 4 worked in local commercial enterprises, and the remaining 5 were elderly retirees living on social security benefits. An average of 4 to 5 residents lived in the households I visited, and 24 of the 27 interviewees’ households included minors.

RESULTS

Information typifying spaces as urban or rural (or even periurban and semiurban) is helpful in gauging how density and landscapes influence water, sanitation, and hygiene issues and concerns among residents of different areas. However, this taxonomy is often used in survey reports (as described in the Discussion section) to aggregate more than density and landscape characteristics, expanding to work typologies, social networks, cultural norms, and general lifestyles. As a result, use of the urban–rural designation can obscure the mixture of vulnerabilities found in varied geographies. The observations stemming from my interviews with residents of KaTembe (again, a rural district in a capital city) show how an accessibility, adequacy, affordability, and awareness lens on water-, sanitation-, and hygiene-related concerns can reveal unique challenges faced by these individuals that would be lost in urban–rural aggregations.

Sanitation

Although their levels of adequacy and affordability vary, latrines are the standard sanitation option for most residents of KaTembe. The typical latrine in KaTembe is situated outdoors and most often surrounded by a small, open-roof room for privacy. Among the residents I interviewed, traditional pit latrines (i.e., those without covers) were almost as common as improved pit latrines, or latrines that have covers and are built with cement blocks to keep the latrine's walls from falling in, particularly during the rainy season. Latrines tend to last anywhere from 6 months to 3 years, depending on their quality, how many individuals use them, and to what extent the rainy seasons affect their walls. Salt or cooking ashes are most often added for upkeep, although respondents reported using salt only when they could afford to purchase an adequate amount (for example, 5 or 6 kg of salt every 3 to 4 months).

Several interviewees complained that their facilities had worsened over the preceding 3 years. They explained that even though they have had improved pit latrines made with cement blocks and covers, they must dig another hole when these latrines are full. The “replacement” latrines of residents I spoke with were always traditional pit latrines.

The explanation for the relatively common reports of latrine deterioration is of interest here. Residents of KaTembe benefited from recent sanitation campaigns and donations of improved pit latrines; one initiative was organized by the municipal government in 2007, and another was led by the local nongovernmental association, Imagine, between 2007 and 2009. However, despite benefits of such one-time sanitation improvement projects, affordability concerns limited residents’ ability to maintain improved facilities by purchasing latrine covers or cement blocks to line new latrines. Residents indicated that the cheaper option was simply to dig another hole. Indeed, almost all of the residents I interviewed did just that.

The digging of latrines is also an income-generating opportunity for some in KaTembe. Eleven interviewees reported paying someone to dig their latrine hole for them. However, this finding raises further affordability concerns. Three residents I interviewed—all female household heads—reported that they could not dig an adequate traditional pit latrine themselves, nor could they afford to pay someone else to do so. Instead, they indicated that they simply dug small holes in the ground to relieve themselves, and they emphasized in their conversations with me that they covered the hole afterward with dirt.

Indeed, one of the more difficult sanitation topics to broach and discuss—for the interviewee as well as the interviewer—was that of open-air defecation. Much effort has been made in rural parts of Mozambique to conduct educational campaigns on the subject and to include questions on open-air defecation in household surveys on water and sanitation (e.g., Grupo de Água e Saneamento, a national network of associations and professionals dedicated to water and sanitation, meets monthly to share implementation strategies and research on public health interventions related to water, sanitation, and hygiene¹⁷).

Rather than asking whether respondents had ever needed to openly defecate themselves, I asked them whether they believed or witnessed that this was a problem in their neighborhoods. Eight residents I spoke with reported that open-air defecation was indeed sometimes a problem in their neighborhoods. Interestingly, these respondents were evenly distributed between administrative bairros, with each of the 4 subdistrict bairros representing 2 such replies. However, open-air defecation was apparently most problematic in high-density areas, representing half of the 8 complaint locations.

There are 3 possible explanations of why open-air defecation was still considered a problem in these neighborhoods. One interviewee located in a high-density area of KaTembe noted that it was a problem because of a close-by transit stop where travelers await pickup. Another high-density settlement respondent noted that the culprits were often children. Finally, a respondent in a low-density area also related the problem to transit, noting that people sometimes defecate in open at night because they are walking long distances to get home.

Hygiene

The hygiene questions included in the survey allowed exploration of how awareness of health risks related to water, sanitation, and hygiene influenced residents’ behaviors and hygiene practices. It is important here to highlight that awareness is different from educational attainment. For example, of the 12 residents I interviewed who were aware of water-, sanitation-, and hygiene-related illnesses (namely cholera, diarrhea, and stomach viruses), 7 had no formal education.

Indeed, there was little correlation between education and awareness of such illnesses, methods of treating them, and ways to avoid them. Respondents who knew how to treat these illnesses represented a mixture of educational attainment and age. Furthermore, almost all respondents also knew that they should cover their water containers or cans when not in use and knew to keep water cups on clean surfaces to avoid contamination. However, affordability issues influenced whether residents actually did cover their water. The few residents who said that they did not cover standing water at their homes explained that they did not own covers. Interestingly, this was an issue for those who could afford to conserve water at their homes in containers larger than the 20- or 25-liter buckets used to gather water. Residents I spoke with who could not afford the larger water containers simply maintained water in jerry cans or in bottles that had tops or were easier to cover (with a plate, for example) than larger containers.

Affordability issues also influenced residents’ cleaning habits. Although all but one of the respondents with whom I spoke reported regularly cleaning the containers used to gather water, variations existed with respect to frequency (ranging from every day to once a month) as well as method of cleaning. Respondents who gathered water from standpipes, water pumps, or open wells reported 3 methods of cleaning the inside of containers: water alone, a combination of water and pebbles or sand found around the water point (to grate the inside of jerry cans where visible residue builds up), and a combination of water, soap, and sometimes pebbles. The primary factor determining which method was used was whether residents could afford the addition of soap. All residents using soap used Omo, a detergent that has risen in price over the past 3 years¹⁸ and is actually marketed as a clothing detergent rather than an all-purpose cleaner.

Waste Management

Waste management did not appear to be a major concern of residents with whom I spoke. Only 3 respondents, all in a high-density settlement in the high-density bairro of Guachene, had municipal roadside waste pickup services. None of the other respondents had any formal trash removal service. Instead, they reported burning and burying trash in their own yards. Respondents explained that they often used this method to enrich dirt for farming, and the fact that a number of the interviewees farmed for subsistence helps explain why residents often said that they did not require any trash removal.

Interestingly, 2 respondents did indicate that there was a question of fairness in the provision of municipal waste removal services. These respondents indicated that although electricity bills include a tax that is channeled to provide waste removal services, their neighborhoods did not receive such services despite paying this tax. It is worth noting that electricity provision in KaTembe is itself limited to more dense settlements and that almost half of the residents with whom I spoke reported having no electricity options.

Water

Water sources were largely deemed accessible by the residents interviewed, with most respondents indicating that the distances they had to travel were reasonable. However, concerns with adequacy related to the time burden of gathering water emerged from the interviews. These concerns stemmed not from the distance to water points but from the wait time for one's turn at a source. Both water pumps and fountain standpipes—the water sources most used by respondents—have appointed access times, with one daily access time in the morning and another in the afternoon. As a result, some residents use unattended jerry cans to hold their place in line. One respondent complained that at times it takes 2 hours just to get water, although the water source is only a 10- to 15-minute walk from home.

Residents using such water pumps pay a monthly maintenance fee alone rather than paying by water volume, as those using fountain standpipes do. Interestingly, however, residents did not voice any explicit complaints regarding the affordability of the water itself. Instead, they related how they used different water sources for different purposes and during different times and seasons. If standpipes and water pumps are not accessible owing to affordability issues or breakdowns in equipment, for example, residents reported that they also have access to free water from wells or to rainwater. However, wells are most often uncovered, and as often as not the water is salty or dirty and thus inappropriate for drinking.

These findings highlight how drinking water options within our periurban study area have characteristics more typically associated with rural areas. This reality is especially problematic for respondents who entirely or partially depend on open wells and have unstable incomes that preclude dependence on filtered water from fountains or a standpipe. Given that only a single resident with whom I spoke reported having a fixed income, this represents a serious concern, one more commonly attributed to rural settlements.

DISCUSSION

Although physical geographies are helpful aggregate characterizations for shaping public sector interventions in some basic service areas, it is important to recall that environmental health systems must address both physical and nonphysical challenges. Subtle and tangible vulnerabilities exist in diverse but shared physical spaces (see, for example, the proximity of the pastoral and concrete landscapes in Figure 1). Although these vulnerabilities find expression in the accessibility, adequacy, affordability, and awareness framework, this framework is still not widely integrated into the reports of important institutions in the international public health sphere.

FIGURE 1—

Urban or rural? The Guachene bairro in the municipal district of KaTembe, Maputo, Mozambique.

Source. Photograph by Gabriella Y. Carolini.

The challenges described by the residents of KaTembe with whom I spoke question the dualistic urban–rural divide as a classification meant to provide texture to aggregated health statistics. Yet, many influential survey reports continue to aggregate vulnerabilities by this very dichotomy. For example, the WHO/UNICEF Joint Monitoring Programme for Water Supply and Sanitation uses data from 9 surveys in its reports. All of the program's estimations of improved water and sanitation usage are aggregated according to urban–rural categories.^19,20

Furthermore, in the latest Mozambique Demographic and Health Survey (DHS),²¹ Maputo city in particular often emerges as a relatively strong outlier relative to other aggregated urban and rural areas of the country. Whereas only 0.3% of the rural population and 17.3% of the urban population interviewed for the Mozambique DHS had access to clean drinking water in their homes, the figure for Maputo was 39.8%. Similarly, whereas 63% of rural DHS interviewees and 21.6% of urban interviewees reported having no sanitation infrastructure, only 0.5% of Maputo-based DHS respondents indicated having no latrines or toilets.

Although possibly helpful in some public health interventions, such survey data do not account for important intraurban spatial variations and vulnerabilities. Indeed, in Maputo in particular, they often paint a picture of improvement. Nonetheless, cholera still threatens the city's health,²² and riots broke out first in February 2008 and more recently in September 2010 to protest the increase in living costs, including proposed increases in water, electricity, bread, and oil prices. These events are not only political, economic, or security concerns. They are a reflection of the environmental health challenges faced by residents of periurban areas such as KaTembe, and they bring to light the importance of designing health systems that combine more diverse geographic sensibilities with an accessibility, adequacy, affordability, and awareness lens that captures how vulnerabilities intermingle.

For example, this study shows that adequate water-, sanitation-, and hygiene-related services compete with other basic necessities such as food. In KaTembe, and Maputo more widely, it appears that expenditures for such services are not being as highly prioritized as expenditures for other basic needs. A recent multiyear report on poverty among tracked households in Maputo indicates that whereas 92.5% of households had monthly water expenditures in 2007, only 81.2% of the same households had such expenditures in 2010.¹⁸

All but one of the residents I interviewed listed food as their biggest cost of living expense, followed by electricity (in cases in which households had such utilities available). None listed water or sanitation costs as their top concern. Although this might indicate that these provisions are affordable, my findings show—via the added dimensions of adequacy and accessibility—that more is at play in household decisions about expenditure priorities. My respondents reported that that when money is tight, they can obtain free water from wells (uncovered or not), and sanitation options can revert back to basic holes in the ground when necessary. However, these other options have costs. They take more time away from other productive or income-generating activities than having a household standpipe or latrine. Also, although they might be temporary solutions (and thus fall outside the scope of health surveys), they are not adequate.

In short, KaTembe residents’ demand elasticity for potable water and sanitation was greater than their demand elasticity for other basic needs such as staple foods. Future studies could help determine whether this helps explain, for example, some of the volatility in reported cases of cholera in Maputo, with almost 12 000 cases reported in 2004, no cases in 2005 or 2006, approximately 900 cases in 2007, and none again until April 2011, when there were 20 new cases.^22,23

Vulnerabilities exist almost everywhere, but vulnerable residents of periurban areas such as KaTembe run the risk of paying the combined price of higher costs of living than residents in more central or dense settlements and being less targeted for public health interventions than their rural counterparts given the starker overall circumstances of those individuals. In Mozambique, both governments and donors still largely associate poverty and environmental degradation with rural areas.¹⁸ The addition of an accessibility, adequacy, affordability, and awareness framework would be a tremendous contribution to improving the quality of survey data reports and the public health interventions they inform. Such changes would help lead to a focus on which residents of any given area need further assistance, as opposed to which areas need more or less aid.

Because of the physical nature of water and sanitation systems, water-, sanitation-, and hygiene-related interventions have often focused on aggregated physical geographies in targeting needs. However, as this study shows, water, sanitation, and hygiene vulnerabilities include more than location-specific concerns, and public health interventions—especially those supported by international organizations—must do more to incorporate an accessibility, adequacy, affordability, and awareness approach into data gathering and systems planning.