Spatial inequality in safely managed water access in Ghana

Sustainable Development Goal (SDG) 6 calls for the availability and sustainable management of water and sanitation for all by 2030 (United Nations 2015). Target 6.1 specifically focuses on drinking water. It seeks to achieve universal and equitable access to safe and affordable drinking water for all by 2030 (United Nations 2015), with the indicator being the proportion of the population using safely managed drinking water services (United Nations Children's Fund (UNICEF) and World Health Organization (WHO) 2018). In the words of the WHO/UNICEF Joint Monitoring Programme (JMP), safely managed water is ‘drinking water from an improved water source that is located on premises, available when needed and free from fecal and priority chemical contamination’ (UNICEF and WHO 2018, 7).

The JMP in its 2020 WASH report at the household level estimates global access to safely managed water to be 74%, an increase of 4% compared with the baseline figure of 70% in 2015 (WHO and UNICEF 2021). The report revealed significant disparities in safely managed water coverage by SDG regions – 96% in Europe and Northern America, 79% in Northern Africa and Western Asia, 75% in Latin America and the Caribbean, 62% in Southern and Central Asia, and 30% in sub-Saharan Africa (WHO and UNICEF 2021). According to the JMP, the world is currently off track in achieving SDG target 6.1 unless the current rate of progress is quadrupled. In most sub-Saharan African countries, achieving universal access to safely managed water service by 2030 will require an increase in current rates of progress by about 10-20 times. Estimates by the JMP show that all sub-Saharan African countries, including Ghana, will miss SDG target 6.1 at the current rate of progress (WHO and UNICEF 2021).

In 2020, access to safely managed water service in Ghana was estimated to be 41% and projected to be 85% by 2030 (WHO and UNICEF 2021). Like most sub-Saharan African countries, disaggregated data on safely managed water access by relevant socio-economic and spatial variables are currently lacking in Ghana. This data gap has the potential to undermine interventions aimed at achieving universal access to safely managed water service in Ghana by 2030 as targeted by SDG 6.1 as well as SDG 10 which seeks to reduce inequalities within and among countries. The United Nations in recognition of the need to bridge inequalities in development calls on surveillance agencies of the SDGs to disaggregate national data by income, gender, age, race, ethnicity, migratory status, disability, geographic location, and other characteristics relevant in national contexts to measure inequality (United Nations 2015). Seven years post the adoption of the SDGs, knowledge of spatial inequality in water access in Ghana can inform interventions on universal access to safely managed water as targeted by SDG 6.1 (Antwi-Agyei & Monney 2018).

It is against this background that the study draws on the Gini index to examine the level of inequality in drinking water access among regions, and between urban and rural areas in Ghana with a focus on access to safely managed water service, access to safe water, access to water on premises, access to sufficient quantities of water, and access to improved water, using household level microdata of the 2017/18 Ghana Multiple Indicator Cluster Survey (MICS). Although there are over 50 inequality measures (Eliazar 2018; McGregor et al. 2019), the Gini index was employed in this study because it is the most widely used inequality measure (Cole et al. 2018). It is also easy to compute and interpret (Sitthiyot & Holasut 2020). Moreover, it accounts for the deviations of all the values in the population and is very robust to inequalities in the tails (Kovacevic & Binder 1997). The Gini index was traditionally developed to measure the distribution of income inequality, but it has been applied to measure inequality in the distribution of land and water resources (for example, Cullis & van Koppen 2007; Wang et al. 2011; Cole et al. 2018).

The study contributes to Water, Sanitation, and Hygiene (WASH) scholarship by providing empirical evidence on spatial inequalities in drinking water access in Ghana, reinforcing the call by the United Nations on the need to disaggregate the SDGs data at a subnational level. The study also provides a case study on SDGs data disaggregation, inequality assessment, and reporting at a subnational level. Lastly, the findings of the study will inform the design and implementation of inclusive and equitable water policies in Ghana.

The population of Ghana as recorded in the 2021 Population and Housing Census (PHC) was 30,792,608, with an average annual growth rate of 2.1% (GSS 2021). The proportion of females recorded in the 2021 PHC is slightly higher (50.7%) than males (49.3%) (GSS 2021). Educational attainment in Ghana is generally low. From the 2010 PHC, 5.7% of persons aged 6 years and above had tertiary education, 3.5% had secondary education, 64.8% had basic education, 2.5% completed only nursery/kindergarten, and 23.5% never attended school (GSS 2013b). As with most developing economies, the main occupation of most households in Ghana is agriculture. From the 2010 PHC, 45.8% (2,503,006) of households in Ghana were engaged in the agricultural sector (GSS 2013b).

With respect to the objective of the study, the data required for the study include households’ source of drinking water, location of households’ drinking water source, safety of household drinking water source, households’ access to sufficient quantities of water when needed, and household size across urban and rural areas, and administrative regions. The sixth round of the Ghana MICS was found suitable for the study because it is the latest national representative household survey in Ghana with data on the above-mentioned variables across urban and rural areas, and the 10 old administrative regions.

MICS is a national representative household survey initiated by the UNICEF to assist countries in collecting and analyzing data for monitoring the well-being of children and women every five years (GSS 2013a). It has enabled many countries to produce statistically sound and internationally comparable estimates on a range of indicators in the areas of water, sanitation, health, education, child protection, HIV/AIDS, among others (GSS 2013a). The MICS was originally developed around the mid-1990s in response to the need to monitor the progress of the World Summit for Children (WSC) Goals and its Mid-Decade Goals. Since the mid-1990s, six different rounds of the MICS have been carried out in Ghana. The sixth round of the MICS which was used for this study was conducted in 2017/18 by the GSS in collaboration with the Ministry of Health, Ministry of Education, Ministry of Sanitation and Water Resources, Ministry of Gender, Children and Social Protection, Ghana Health Service and the Ghana Education Service as part of the Global MICS Programme (GSS 2018). A detailed methodology of the 2017/18 Ghana MICS is attached as supplementary material.

Household level microdata of the 2017/18 Ghana MICS was downloaded from the GSS website in SPSS format. The file contained records of all 13,202 households that were sampled for the study. Of these, data were successfully collected from 12,886 households, yielding a response rate of 99.4%. With the exception of data on household main drinking water source safety (with the indicator being E. coli), all other data required for the study were collected in the 12,886 households’, that is, household main source of drinking water, location of household main drinking water source, household access to sufficient quantities of water when needed, household size as well as the area and region of the household. Perhaps, due to the high cost involved in water quality testing, testing of E. coli in household drinking water sources was carried out for 3161 households, representing approximately 25% of sampled households. The 3161 households with data on the amount of E. coli in household's main drinking water source were extracted for the study because they equally have data on all other relevant variables required for the study. A summary of the variables and their associated codes as extracted from the SPSS file is presented in Table 1.

As part of data cleaning, 30 households were deleted, bringing the number of households used for the analysis to 3,131. The deleted households comprised of:

• seven households with main source of water captured as ‘others’;

• one household with no information on the location of the main water source;

• seven households without information on household access to sufficient quantities of drinking water when needed; and

• 15 households without results on the amount of E. coli in drinking water source.

Before analysis, the amount of E. coli in household's main drinking water source which was captured as continuous data (Coliform Forming Unit –CFU/100 mL of water) was reclassified into four classes based on the WHO E. coli-risk-to-health classification (WHO 2011) as follows: <1 CFU/100 mL (low risk), 1–10 CFU/100 mL (intermediate risk), 1–100 CFU/100 mL (high risk), and 100 CFU/100 mL (very high risk). The authors also created a new variable in the SPSS dataset called ‘safely managed water status’ with two categories – 1 (household has safely managed water service) and 0 (household has no safely managed water service). The data were analyzed in SPSS using the custom tables tool as it allows for cross-tabulation of two or more variables. The custom table was used to cross-tabulate the amount of E. coli in household's main drinking water source, main source of drinking water, location of drinking water source, and access to sufficient quantities of drinking water when needed and safely managed water status with the locality of the household and the region of the household to generate frequencies.

To understand the level of spatial inequality in water access, the Gini index was computed for safely managed water coverage, access to water on premises, access to sufficient quantities of water when needed, improved water coverage, and safe water access at both regional and locality levels using the frequencies obtained from the cross-tabulations.

To obtain the Gini indices for access to improved water, access to water on premises, access to safe water, access to sufficient quantities of water, and access to safely managed water at the regional level, we plotted the cumulative percentages of water access for each of the above-mentioned indicators on the y-axis of the Lorenz curve and the cumulative percentage of regions on the x-axis. The Gini index was then computed as a ratio of the area between the Lorenz curve and the equality line, and the total area under the line. This process was repeated to compute the Gini indices for improved water coverage, access to water on premises, access to safe water, access to sufficient quantities of water, and access to safely managed water at the locality level. The Gini index usually ranges from 0 to 1, with 0 being total equality while 1 means complete inequality (Wang et al. 2011). No detailed classification of the Gini index exists for water access. However, based on Cole et al. (2018) interpretation of the Gini index, a Gini index of less than 0.2 was classified as low inequality, 0.2–0.3 as moderate inequality, and greater than 0.3 as a high level of inequality. The data were presented in the form of tables, graphs, simple percentages, and indices.

One of the elements of safely managed water service is that people must have access to drinking water sources on their premises (UNICEF & WHO 2018). From the 2017/18 Ghana MICS, approximately 14.8% of the population in Ghana has access to drinking water sources on premises. Low access to drinking water sources on premises is largely due to financial constraints on the part of water utility companies to extend water to households, inaccessibility of housing units to physically connect to the water networks, high water tariffs, and ill-suited housing arrangements (Ablo & Yekple 2018; Kumasi 2018).

The third and last element of safely managed water service is that drinking water sources must be free from fecal and priority chemical contamination (UNICEF and WHO 2018). The 2017/18 Ghana MICS, which was used for the study only monitored fecal contamination in drinking water at source with the indicator being E. coli. Table 3 shows population distribution based on the number of E. coli per 100 mL of source water as recorded in the 2017/18 Ghana MICS. Overall, 51.3% of the population drinking water was free from fecal contamination at source. The Gini index (0.10) revealed a low level of inequality between rural and urban population exposure to E. coli in drinking water (Figure 4). From Table 3, the proportion of the population without E. coli in source water in the urban areas was 61.7% as against 43.1% in the rural areas. High access to safe drinking water at source in the urban areas compared with rural areas can also be explained by higher levels of open defecation in the rural areas than in the urban areas (GSS 2018). Also, sachet water and piped water which most urban population are dependent on are often treated.

At the regional level, the Gini index (0.17) also showed low inequality in the proportion of the population without E. coli in drinking water sources (Figure 3). It was highest in the Ashanti Region (71.2%), followed by the Upper West Region (62.6%), Greater Accra Region (60.4%), Upper East Region (58.4%), Western Region (58.4%), Central Region (56.7%), Brong-Ahafo Region (52.0%), Eastern Region (48.8%), Northern Region (22.1%), and lastly Volta Region (21.1%) (Table 3). Access to safe drinking water at source in Northern and Volta regions was lower than the national average of 51.7%, and this may be explained by high levels of open defecation and unimproved water use in those regions as compared with other regions (GSS 2018).

Safely managed water coverage in the urban areas is higher than in the rural areas because urban areas have slightly better access to improved water, safe water at source, and drinking water source on premises than rural areas (Tables 2 and 3, and Figure 5). Similarly, regional differences in access to safely managed water service are largely determined by the level of access to safe water and drinking water source on premises. For instance, the Ashanti Region tops the safely managed water coverage chart (13.8%) because it recorded the highest proportion of the population with access to safe water (71.2%) and drinking water sources on premises (26.1%).

This study applied the Gini index to examine the level of inequality in drinking water access among regions, and between urban and rural areas in Ghana with a focus on access to safely managed water service, access to improved water, access to water on premises, and access to sufficient quantities of water when needed, using microdata of the 2017/18 Ghana MICS. Access to safely managed water service in Ghana is low with moderate disparity among regions, and a high disparity between rural and urban areas. At the regional level, moderate inequality was recorded for access to water on premises and access to safe water while low inequality was recorded for improved water coverage. Between urban and rural areas, moderate inequality was recorded for access to water on premises while low inequality was recorded for access to sufficient quantities of water when needed, access to safe water, and improved water coverage.

The results of the study showed that all regions and localities deserve attention with regard to efforts aimed at achieving universal access to safely managed water as no region or locality has attained 100% coverage. However, in the face of limited resources, priority should be given to the Volta Region, followed by Upper West, Upper East, Northern, Brong-Ahafo, Western, Central, Eastern, Ashanti, and lastly Greater Accra. Between urban and rural areas, the results point to the need to prioritize the latter over the former with regard to interventions aimed at achieving universal access to safely managed water services. In terms of intervention focus by the elements of safely managed water service, priority should be given to access to water on premises where all regions performed poorly, followed by access to safe water, and lastly access to sufficient quantities of water. To this end, the authors strongly support the call by the United Nations for monitoring and disaggregation of national data of the Sustainable Development Goals by relevant socio-economic and spatial variables to help in the design and implementation of inclusive and equitable policies. In addition to data disaggregation at a subnational level, the inequality measure (Gini index) presented in this study provides insight on the extent of inequality in water access at regional and locality levels and thus could be used as a tool for monitoring spatial inequality in water access.

The authors would like to thank the Government of Ghana, UNICEF, and all other institutions that funded the sixth round of the Multiple Indicator Cluster Survey and made the data available for public use.

This research received no external funding.

All relevant data are available from an online repository or repositories (https://www.statsghana.gov.gh/gssdatadownloadspage.php).

The authors declare there is no conflict.