Household Water Insecurity in the Western Amazon, Amazonas, Brazil: A Preliminary Approach

Abstract

The objective was to evaluate the quality of an instrument to measure the experience of household water insecurity (WI) and the factors associated with the prevalence of WI in an urban area in a municipality in the Western Brazilian Amazon. A cross-sectional, population-based study was conducted to investigate 983 urban households. The Household Water Insecurity Experiences (HWISE) scale was used to measure the psychometric properties of reliability and validity. An exploratory factor analysis was conducted, and the prevalence ratio (PR, 95% CI) was calculated, considering WI as the dependent variable and the other household variables as independent variables. WI affected 46.2% (95% CI: 43.0–49.4%) of the households, independently associated with: head of the family as parent/other and presence of a child in the household. The instrument exhibited unidimensionality in the factor analyses and was considered to be both reliable and valid, as indicated by a Cronbach’s α coefficient of 0.958. Household WI is a serious public health problem in the Amazon in correlation with both social vulnerability and a lack of public services. As a preliminary approach, the scale proved to be valid and reliable. However, considering the Amazonian context, misunderstandings about some issues by respondents were identified, and further validation studies are needed to improve the intelligibility of these questions.

1. Introduction

Water insecurity (WI) affects both developed and developing countries, especially those that are located in the Global South [1,2], which face the most diverse socioeconomic and political vulnerabilities [3]. One such country is Brazil, which, despite its possession of one of the largest watersheds on the planet (equivalent to 12% of the global total), has insufficient and unequal access to safe water in geographical and social terms [4,5].

Paradoxically, the North Region of Brazil, which features abundant water resources, accounting for approximately 70% of the country’s fresh water, is characterized by the lowest rates of access to and distribution of water for human consumption, as well as the lowest rates of access to the public sewage system [6].

In this context, the State of Amazonas is located in the Western Amazon, which is the most well-preserved part of the Brazilian Amazon. This region, which features a continental scope, consists of a political-administrative structure that is divided into 62 municipalities, most of which are small. These municipalities feature extensive territorial areas but low levels of population density [6].

In contrast, the capital, Manaus, exhibits the largest population, including members of the native population, including 71,700 indigenous inhabitants, and is interconnected with numerous tributaries that are spread across the extensive Amazon watershed. The lack of basic sanitation and access to public services is latent [7,8], thus highlighting the so-called Amazonian factor [7].

Populations who are located far from the capital suffer more severe effects from such disparities, which result in inadequate conditions in terms of basic sanitation and low levels of access to health, education, food, and nutrition services [9,10,11,12,13,14]. With respect to public health policies, access to water for human consumption in Amazonas is a dramatic problem, and has resulted in serious mishaps in terms of quality, supply, storage, and monitoring [15]. This reality is even more concerning, given that WI can exacerbate the issue of food insecurity (FI) at home, especially in its severe form, that is, hunger [9].

At the same time as the hunger epidemic that occurred in Brazil in 2022, approximately 12% of the Brazilian population had no access to water. Additionally, severe FI and WI were observed to be interrelated. Approximately 65% of households facing WI also experienced quantitative food restriction, and a combination of both issues affected the North Region more strongly (48.3%) than the other regions of the country, whose national average was 42.0% [16].

Both types of damage have powerful consequences with respect to populations’ ways of life and survival, especially for the most vulnerable populations [17,18,19]. Thus, the use of instruments that are capable of measuring the experiences of individuals/households makes it possible to quantify and understand the associated factors as well as to support political action targeted at protecting people’s rights to adequate and healthy food and health, which represent a major challenge for public policies [20].

The present study aimed to evaluate the quality of an instrument that can be used to measure experiences of household WI in the Amazon context alongside various factors associated with the prevalence of WI in an urban area in a municipality in the Western Brazilian Amazon.

2. Materials and Methods

2.1. Study Design and Location

This cross-sectional, population-based study was conducted between August and November 2021; it focused on households in an urban area in the municipality of Itapiranga, Amazonas, which is located in the Amazon River basin, Western Amazon. The municipality of Itapiranga features an area of 4335.075 km² and a population density of 1.94 inhabitants/km; furthermore, it is located 226 km from Manaus, the state capital, and has both land (via road, i.e., AM-010 and Várzea) and fluvial access [7].

2.2. Sampling

The study population was selected by stratified probability sampling. The population of interest was composed of adults, concentrating on the age group between 20 and 59 years old. The registry of the Primary Care Information System (SIAB) of the Municipal Health Department (SEMSA) from 2021 was used, which is the largest data source available and updated at the time, corresponding to the three health areas among which the local Unified Health System (SUS) is organized, namely: area 04 (1210 inhabitants), area 05 (2255 inhabitants), and area 06 (1682 inhabitants).

To calculate the sample size, the prevalence of 38.4% of households living in water insecurity in the North Region was considered [21]. An absolute error of 5% with 95% confidence was established to estimate the prevalence of HI. Additionally, 10% of individuals were added to the final calculation of the sample to minimize possible losses and/or refusals for each of the three health areas of the municipality, respectively, 308, 345, and 330 individuals, which resulted in a total of 983 individuals.

2.3. Eligibility Criteria and Data Collection

All households that featured at least one adult resident between the ages of 18 and 59 years who agreed to participate in this research by freely signing the informed consent form (ICF) were considered eligible for inclusion in this research, except people with disabilities (PCD). The interviews were conducted face-to-face by trained interviewers, who employed sanitary biosafety protocols.

2.4. Household Water Insecurity Scale (HWISE): Psychometric Properties

The Household Water Insecurity Experiences (HWISE) scale [1] consists of 12 questions (Box 1), and cut-off points are established on the basis of the scores assigned to the items, which are evaluated as follows: water security: 0–11 points; water insecurity: 12–36 points. To produce the final classification of households, the stratification of the levels of water security/insecurity was considered in light of the sum of the scores assigned to the 12 items included in this measure, which resulted in a set of scores ranging from 0 to 36. To assess the quality of the instrument, psychometric properties such as reliability and validity were used [22].

Box 1. Household Water Insecurity Experiences (HWISE).

Label	Survey Item
Worry	In the last 4 weeks, how frequently did you or anyone in your household worry you would not have enough water for all of your household needs?
Interrupt	In the last 4 weeks, how frequently has your main water source been interrupted or limited?
Clothes	In the last 4 weeks, how frequently have problems with water meant that clothes could not be washed?
Plans	In the last 4 weeks, how frequently have you or anyone in your household had to change schedules or plans due to problems with your water situation?
Food	In the last 4 weeks, how frequently have you or anyone in your household had to change what was being eaten because there were problems with water?
Hands	In the last 4 weeks, how frequently have you or anyone in your household had to go without washing hands after dirty activities because of problems with water?
Body	In the last 4 weeks, how frequently have you or anyone in your household had to go without washing their body because of problems with water?
Drink	In the last 4 weeks, how frequently has there not been as much water to drink as you would like for you or anyone in your household?
Angry	In the last 4 weeks, how frequently did you or anyone in your household feel angry about your water situation?
Sleep	In the last 4 weeks, how frequently have you or anyone in your household gone to sleep thirsty because there was not any water to drink?
None	In the last 4 weeks, how frequently has there been no usable or drinkable water whatsoever in your household?
Shame	In the last 4 weeks, how frequently have problems with water caused you or anyone in your household to feel ashamed/excluded/stigmatized?
Note: Adapted from Young et al. [1]. The responses to items are: never (0 times), rarely (1–2 times), sometimes (3–10 times), often (11–20 times), always (more than 20 times), do not know, and not applicable/I do not have this. Never is scored as 0, rarely is scored as 1, sometimes is scored as 2, and often/always are scored as 3.

2.5. Characterization of Households

To characterize the households, the following variables were considered: number of residents (0 to 4 and 5 or more residents); presence of children (yes or no); socioeconomic variables; family income (expressed in terms of the minimum wage); social class; and access to the Bolsa Família Programme (BFP) public policy.

Social class was assessed on the basis of the standardized index provided by the Brazilian Association of Research Companies (ABEP) [23]. The classification is based on the number of consumer goods, the number of domestic workers, the level of education attained by the head of the family, and access to public services. The individuals included in this research were subsequently divided into subgroups: A, B1, B2, C1, and C2 (least frequent) and D and E (most frequent).

2.6. Statistical Procedures

The 12 items included in the HWISE scale focused on questions that corresponded to a recall period of four weeks and featured five possible response categories: never, rarely, sometimes, often, and always. Initially, for both the descriptive analysis and the exploratory factor analysis, four response categories were used [i.e., ‘never’ (score 0), ‘rarely’ (score 1), ‘sometimes’ (score 2), and ‘often/always’ (scored as 3)]. The categories ‘often’ and ‘always’ were unified because ‘always’ was rarely chosen as an answer. Thus, the scores obtained in this context ranged from 0 to 36.

The multidimensionality of the HWISE scale was assessed using an exploratory factor analysis (EFA) to explore the latent structure of the items. We applied oblique rotation (Promax), as we assumed that the underlying dimensions of household water insecurity are conceptually and statistically correlated. The number of factors to retain was determined through examination of the scree plot, the Kaiser criterion (eigenvalues > 1), and the theoretical interpretability of the factor structure [24]. The internal reliability of the scale was also tested by reference to Cronbach’s α coefficient (>0.80).

Scores on the HWISE scale were calculated as the sum of each of the participants’ responses to the 12 questions. A 12-point cut-off was used to classify each response as indicating either ‘water security’ (when the score in question was between 0 and 11 points) or ‘water insecurity’ (when the score was between 12 and 36 points).

Per capita income was calculated as the ratio between the average income and the number of residents; subsequently, the value of per capita income was categorized on the basis of quartiles and the results of the ABEP classification.

The prevalence of WI is described in terms of numbers and percentages, alongside the corresponding 95% confidence interval (95% CI). To estimate the associations between WI, FI, and the socioeconomic indicators selected for this research, the prevalence ratio (PR) with the corresponding 95% CI was calculated, specifically when WI was included as the dependent variable and the other household variables and the head of household were included as independent variables. The variables that exhibited p values of up to 20% in the univariate analysis were selected for the multivariate model. We chose to calculate the prevalence ratio rather than the odds ratio because the event evaluated in this research (i.e., WI) was frequent (≥20%). Prevalence ratios and the corresponding 95% confidence intervals were estimated on the basis of the generalized linear model (GLM), with a binomial distribution and log link function [24].

All tests were bilateral, and p values lower than 5% (p < 0.05) were considered to indicate statistical significance. The analyses were performed with the assistance of STATA software, version 18.0, for Windows (STATA, College Station, TX, USA).

2.7. Ethical Aspects

The study was approved by the Ethics Committee of the Federal University of São Paulo under protocol number 55669522.5.0000.5505. Respondents who agreed to participate in this research freely signed the ICF.

3. Results and Discussion

In exploratory factor analysis [25,26], models that featured two and one factors were evaluated sequentially to determine their interpretability and theoretical significance. Following the relevant adjustments, the scale was identified as unidimensional, as the first factor explained approximately 69% of the variability in the data, and the second factor explained approximately 10% of this variability. In other words, the first factor was predominant, thus justifying the use of only one factor as an optimal balance between scope and significance. The Cronbach’s α coefficient was equal to 0.958, thus suggesting a high level of consistency among the items.

Figure 1 indicates that these experiences change as the WI becomes more severe. The first relevant factor, which has commonly been identified in studies conducted in Latin America [1], was worry, followed by interruption, and a lack of sufficient water to perform domestic activities, such as washing clothes and preparing food. The experience of a lack of water to support hand and body hygiene occurred later, and finally, families ran out of water, thus leading them to feel ashamed and to go to sleep thirsty.

A visual scan of the correlation matrix of the data revealed that all the coefficients were greater than or equal to 0.30; however, none of these coefficients exceeded 0.90 [25]. In line with Bartlett’s sphericity test (1950), the hypothesis that the correlation matrix was configured as an identity matrix was rejected (p < 0.001). The Kaiser-Meyer-Olkin (KMO) sampling adequacy measure was acceptable, as indicated by a value of 0.953 for the total model and values ranging from 0.888 to 0.977 for each of the variables under evaluation [26]. These measures jointly indicated that the correlation matrix was appropriate for the exploratory factor analysis [26,27].

Velicer et al. [28] recommended that a combination of parallel analytical methods and the minimum partial mean should be used to determine the number of factors that should be retained for rotation, in which context a scree plot [29] is a potentially useful complement. If these three criteria are adopted, one or two factors can be sufficient to ensure an optimal balance between comprehensiveness and parsimony; thus, we chose to view the scale as one-dimensional.

Notably, the high level of consistency observed between the items [23] is similar to the figure reported by Young et al. [1] in an analysis of the level of WI in middle- and low-income countries, as indicated by Cronbach’s α values ranging between 0.84 and 0.93, thus suggesting that the HWISE is a reliable instrument that can be used to measure WI in different contexts, even in those that exhibit heterogeneous geographical characteristics.

Among the 983 households whose residents were interviewed, 46.2% experienced water insecurity (95% CI: 43.0–49.4%). According to

Table 1. Distribution of sociodemographic data concerning the head of the family and the household unit in terms of water insecurity in an urban area in the municipality of Itapiranga, Amazonas, Brazil, 2025 (univariate analysis).

		Water Insecurity
	Total	N	%	Prevalence Ratio (p < 0.20)	IC-95%	p
Family Head						0.003
Grandparent	39	6	15.4	Ref.
Parent/Other	944	448	47.5	3.08	1.47; 6.46
Sex						0.377
Female	326	144	44.2	Ref.
Male	657	310	47.2	1.07	0.92; 1.24
Age						<0.001
<55 years	726	363	50.0	1.41	1.18; 1.69
≥55 years	257	91	35.4	Ref.
Education						0.416
Illiterate/Fund. 1 Incomplete	266	119	44.7	1.09	0.83; 1.42	0.544
Complete Fund. 1 or Fund. 2 Incomplete	113	60	53.1	1.29	0.97; 1.72	0.085
Fund. 2 Complete or Incomplete Middle School	179	86	48.0	1.17	0.88; 1.54	0.276
Complete High School or Incomplete Higher Education	323	147	45.5	1.11	0.85; 1.43	0.452
Completed Higher Education or Higher	102	42	41.2	Ref.
Occupation
Registered	269	116	43.1	Ref.
Self-Employed/Other	714	338	47.3	1.10	0.94; 1.29
Per capita Income						0.001
<200.00	203	113	55.7	1.47	1.20–1.78	<0.001
200 to <300.00	275	135	49.1	1.29	1.06–1.57	0.010
300 to <600.00	239	105	43.9	1.16	0.94–1.43	0.173
≥600.00	266	101	38.0	Ref.
Job Loss during the Pandemic						<0.001
No	668	276	41.3	Ref.
Yes	307	174	56.7	1.37	1.20; 1.57
Bolsa Família Programme						0.030
No	438	185	42.2	Ref.
Yes	544	268	49.3	1.17	1.02; 1.34
Presence of Child						<0.001
No	539	220	40.8	Ref.
Yes	444	234	52.7	1.29	1.13; 1.48
Water at Home						<0.001
Public Network	804	314	39.1	Ref.
Other	179	140	78.2	2.00	1.78; 2.25
Housing						0.007
Own	811	359	44.3	Ref.
Assigned/Leased	104	54	51.9	1.17	0.96; 1.43	0.119
Other	68	41	60.3	1.36	1.11; 1.68	0.004
Electricity						0.001
Yes	964	440	45.6	Ref.
No	19	14	73.7	1.61	1.22; 2.13
Type of Street						<0.001
Asphalt	876	386	44.06	Ref.
Earth/Gravel	107	68	63.6	1.44	1.23; 1.69
Type of Household						0.192
Masonry	713	342	48.0	Ref.
Masonry/Timber	152	65	42.8	0.89	0.73; 1.09	0.259
Wood	118	47	39.8	0.83	0.66; 1.05	0.121

, with the exception of gender, education, occupation, and type of household, the other variables were statistically associated with HI, considering a p value < 0.20. The prevalence of HI was higher among: families headed by parents or others, under 55 years of age, per capita income less than BRL 300.00, job loss during the pandemic, access to the PBF, presence of a child in the household, access to water by means other than the public network, living in a house occupation regime classified as ‘other’, and not having electricity in the household. Those variables that exhibit p values lower than 0.20 were included in a multivariate model.

Following the initial adjustment, the analysis was continued by removing the non-significant variables one by one, which resulted in the final adjusted model (

Table 2. Association between water insecurity and sociodemographic data concerning the head of the family and the household unit, Itapiranga, AM, Brazil. 2025 (multivariate analysis).

	Initial Model			Final Model
	Adjusted Prevalence Ratio	IC-95%	p	Adjusted Prevalence Ratio	IC-95%	p
Family Head
Parents/Other x Grandparents	2.40	1.17; 4.93	0.017	2.59	1.27; 5.26	0.009
Age in years
<55 x ≥55	1.12	0.93; 1.34	0.235
Per capita Income
<200.00 x ≥600.00	1.16	0.94; 1.44	0.173	1.08	0.89; 1.33	0.430
200 to <300.00 x ≥600.00	1.06	0.87; 1.30	0.539	1.04	0.86; 1.27	0.661
300 to <600.00 x ≥600.00	1.09	0.89; 1.34	0.402	1.14	0.93; 1.40	0.206
Job Loss during the Pandemic
Yes x No	1.25	1.10; 1.42	0.001	1.25	1.10; 1.42	0.001
Bolsa Família Programme
Yes x No	1.02	0.88; 1.17	0.815
Presence of Child
Yes x No	1.14	0.99; 1.31	0.073	1.16	1.01; 1.33	0.036
Water at Home
Other x Public Network	1.81	1.60; 2.06	<0.001	1.87	1.66; 2.12	<0.001
Housing
Assigned/Leased vs. Owned	1.14	0.93; 1.39	0.212	1.17	0.96; 1.43	0.128
Other x Own	1.22	1.00; 1.48	0.046	1.24	1.02; 1.51	0.028
Electricity
No x Yes	1.15	0.80; 1.65	0.453
Type of Street
Earth/Gravel and Asphalt	1.31	1.11; 1.54	0.001	1.31	1.11; 1.53	0.001
Type of Household
Masonry/Wood x Masonry	0.97	0.80; 1.17	0.735
Wood x Masonry	0.84	0.67; 1.07	0.153

). In the multivariate analysis, the following variables remained independently associated with WI: head of household, job loss during the pandemic, presence of a child in the household, type of water supply, and type of street of the household.

In the final model, all the prevalence ratios thus obtained were adjusted to account for per capita income. The probability of WI was greater in households whose family heads were parents or others (PR = 2.59; 95% CI: 1.27–5.26) than in those whose family heads were grandparents. Families who experienced job loss during the pandemic were more likely to experience WI (PR = 1.25; 95% CI: 1.10–1.42) than were those who did not experience such job loss. The presence of children in households increased the probability of WI (PR = 1.16; 95% CI: 1.01–1.33) in comparison with households without children. Similarly, WI was more common in households whose occupancy regime was ‘Other’ (PR = 1.24; 95% CI: 1.02–1.51) than in those in which residents lived in their own house.

With respect to access to water, WI was more common (PR = 1.87; 95% CI: 1.66–2.12) in households that were supplied through means other than the public services than in those that were supplied with water by the public network. This result was replicated with respect to the housing conditions, as families who lived in houses that were located on dirt or gravel streets were more likely to experience WI (PR = 1.31; 95% CI: 1.12–1.53) than were families who lived in houses that were located on paved streets.

Among the 983 households evaluated as part of this research, 46.2% were identified as water insecure (95% CI: 43.0–49.4%). The findings of this investigation indicate a figure that is approximately twice as high as the value reported in a study that was conducted in the semiarid region of Ceará (in the Northeast Region), where the level of WI in the urban area was 24.89% and the corresponding level in the rural area was 11.88%. These differences are related to availability, access, and socioeconomic inequalities that can shape populations’ access to quality water [30].

According to data at the national level, between 2021 and 2022, approximately 12% of Brazilian households did not have adequate access to water. In addition, severe FI and WI were interconnected in this context. Approximately 65% of the households that faced WI also experienced quantitative food restriction, namely, hunger. The combination of these two conditions affected the North Region (48.3%) more severely than the Southeast (43.0%), Midwest (41.8%), and Northeast Regions (41.2%). Therefore, both WI and FI occur predominantly among vulnerable populations, thus composing the national scenario of regional inequalities, especially in the Amazon.

Internationally, in vulnerable regions such as the US-Mexico border, WI has contributed to increased gastroenteritis, exposure to carcinogenic compounds as a result of contaminated water, and psychosocial suffering [31]. These health problems are related to the historic social and environmental injustices that characterize these countries [1]. In Brazil’s Legal Amazon, the contamination of water, soil, and the atmosphere with mercury as a result of illegal gold extraction impacts both the ecosystem and human lives in these territories [32].

A multivariate analysis revealed that the following variables were independently associated with WI: head of household, job loss during the pandemic, presence of a child in the household, water availability, and type of street of the household. WI is an indicator of health disparities, given the structural dynamics underlying poverty and economic development [1].

In Brazil, the lack of access to water is an important indicator of inequities and is related to the social and environmental conditions as well as to the water infrastructure that is available at each location [33]. Therefore, it is necessary to reflect on the governance of water resources and territorial specificities, considering that water insecurity is expressed in different ways across the country. Understanding factors related to individual or family water insecurity helps to direct public programs and policies, especially for the most vulnerable groups and/or regions.

Although the Northern Region of Brazil has one of the largest watersheds in the world, its water supply system is deficient, with impacts on the quality of distributed water, supply, storage, and monitoring of water distributed for human consumption [7,17,34]. This reality, consequently, influences the FI of families and contributes to increasing the severe FI that already exists in this state.

Moreover, climatic emergencies resulting from deforestation efforts aimed at addressing the needs of capital, such as the production of commodities, in addition to illegal mining in the Amazon, have strongly impacted the biodiversity of this region and the livelihoods of its population [35]. The resulting increase in temperature and reduction in rainfall and water regime have been identified as important effects of these changes [36]. It is highlighted that deforestation results in significant losses of biodiversity in the Amazon region. Concerning illegal mining, mercury contamination causes environmental damage to water quality, in addition to contaminating fish and human lives [35,36]. Specifically, regarding water quality, the precariousness of the public water supply and monitoring services in the region is impressive [15].

Notably, the deleterious effects of WI and FI, alongside the impacts of climate change, are remarkable among vulnerable population groups. This hard scenario challenges the purposes of the 2030 Agenda in relation to the Sustainable Development Goals (SDG), especially SDG 2—Zero Hunger and Sustainable Agriculture, SDG 6—Clean Water and Sanitation, and SDG 13—Action Against Global Climate Change [37].

Although this study was carried out during a drought season in the Amazon, it is worth noting that the investigation took place in households located in the urban area of the municipality of Itapiranga, whose water supply is provided by the public service, through collection from artesian wells. Therefore, the high prevalence of water insecurity indicated is related to problems in the supply service.

In this study, households with greater social vulnerability, such as those with children or those who lacked public services (no access to water and basic sanitation), were more susceptible to FI, which can contribute significantly to the occurrence of FI, especially in its severe form, i.e., hunger, as both conditions are interrelated and FI is directly associated with family income [9]. Furthermore, in the municipality of Itapiranga, most residents of the households included in the survey had family incomes of as much as twice the minimum wage (minimum wage = BRL 1212.00) in 2021 and were strongly impacted by the COVID-19 pandemic.

Contributions and Limitations of This Study

The scale used to measure the experience of household WI (HWISE) has been developed only recently, and this study is the first one to use it in the interior of the Western Amazon.

The diversity of peoples and the continental scope of the Amazon require reflection on the research instruments used in epidemiological surveys conducted in the Amazon. Although the present study employed a valid instrument and relied on a previous pilot study, it exhibited certain limitations, ranging from the fact that the participants misunderstood some of the questions included in the scale to the period during which data were collected, which took place during a drought in the Amazon.

These misunderstood questions were related to participants’ experiences with a lack of water, as well as to the pattern responses of the instrument, which were: ‘never’, ‘rarely’, ‘sometimes’, ‘often’, and ‘almost every day’. Furthermore, such misunderstandings were related to difficulties for participants to comprehend some statements in the questions themselves.

On the other hand, the innovative nature of this study is evidenced by the fact that it represents the first research to investigate the experience of household WI in Brazil’s Legal Amazon, a region that is renowned for its vast waters and high level of biodiversity, although it is impaired by the current context.

Notably, forests and rivers represent means of production and channels of circulation that host all forms of life [38,39]. Thus, the lack of potable water for human consumption and access to adequate and healthy food can trigger new health–disease processes. Therefore, issues pertaining to water quality and access should be viewed as priorities for One Health given the ongoing climate emergency.

4. Conclusions

Though it is necessary to delve into the particularities of the Amazonian context, the use of an international water insecurity scale allowed for the evaluation of the dimension of water insecurity in a territory abundant in water resources, but scarce in terms of the quality of water available for human consumption.

As a preliminary approach, the scale proved to be valid and reliable. Nevertheless, in the face of misunderstandings about some issues, further validation studies, such as cross-cultural validation, are necessary to improve the intelligibility of these questions. Thus, it would be required to carry out a qualitative study in order to understand the sociocultural and linguistic specificities of the Amazonian population.

A high prevalence of household water insecurity in an urban area of a municipality in the State of Amazonas, associated with social vulnerability and lack of public services, was identified. This finding is unprecedented and brings to light a problem hitherto not addressed in households in the Amazon, providing evidence for decision-making by the government. In this direction, understanding the magnitude of household HI and its possible impacts on households and individuals supports the confrontation of food insecu rity and, consequently, may contribute to the reduction of hunger in this Region.

As recommendations based on our findings, we highlight the urgency of strengthening intersectoral actions, which converge with social protection policies at the local–regional and federal levels, such as the National Sanitation Policy (PNSB), the Water Quality Surveillance Information System for Human Consumption (Sisagua), the National Water Resources Policy (PNRH), and the National Food and Nutrition Security Policy (PNSAN).