Diarrhea in children and sanitation and housing conditions in periurban areas in the city of Guarulhos, SP*



Mariana Gutierres Arteiro da PazI; Márcia Furquim de AlmeidaII; Wanda Maria Risso GüntherIII

IPrograma de Pós-Graduação em Ciência Ambiental do Instituto de Eletrotécnica e Energia da Universidade de São Paulo
IIDepartamento de Epidemiologia da Faculdade de Saúde Pública da Universidade de São Paulo
IIIDepartamento de Saúde Ambiental da Faculdade de Saúde Pública da Universidade de São Paulo

Mail responsible author




A cross-sectional study was carried out to identify the association between diarrhea in 0-2 year-old children and children´s characteristics, access to sanitation and housing conditions in a periurban area served by the Family Health Program, in the city of Guarulhos, SP. Data were obtained from FHP registration forms. Multiple logistic regression showed interaction for Housing*Sewage (other materials and non collected wastewater, p < 0.001), age group (4 - 9 months old, p = 0.054; 10 months and older, p = 0.008) as risk factors for diarrhea. Information collected by the Family Health Program could be an excellent tool to identify populations with poor housing and sanitation conditions at locations where sanitation indicators are not efficient to identify populations living at risk.

Keywords: Diarrhea. Sanitation. Family Health Program. Periurban area. Epidemiologic study. Environmental health.




Diarrhea is a relevant public health issue related to hygiene conditions and water quality (Özkan et al1, 2007). WHO reports diarrhea as the second leading cause of death in children, representing about 1.5 million annual deaths in children under five years old (UNICEF/WHO2, 2009). In global data, diseases related to diarrhea are among the greatest causes of mortality in underdeveloped countries (Fewtrell et al 3, 2005), where they are frequent and can be fatal (Bozkurt et al4, 2003), especially in young children (Moe e Rheingans5, 2006). In these countries, it is estimated that 1.5% of newborn deaths in 1993 (Bozkurt et al4, 2003) and 7.1% of deaths between 1998 and 2002 in Latin America and the Caribbean (Teixeira e Pungirum6, 2005) were caused by diseases related to diarrhea. Transmission of infectious diseases such as diarrhea is a complex process, with many determinants (Trevett el al7, 2005). Nevertheless, 88% of the deaths caused by diarrhea are a consequence of unsafe water, unsuitable sanitation and poor hygiene conditions (UNICEF/WHO2, 2009). This is why water provision in quality and amount, treatment and removal of domestic waste and promotion of sanitation in the community are actions that can be taken to prevent diarrhea in children (UNICEF/WHO2, 2009). These services must be promoted by the urban infrastructure sector, with participation from the public health sector, jurisdiction of the Sistema Único de Saúde (SUS - Brazilian Unified Health System) and community, as stated in the Federal Constitution of 1988 (BRASIL8, 1988).

In many Brazilian localities, populations without urban infrastructure such as sanitation services and public equipment are served by the Programa de Saúde da Família (PSF - Family Health Program). This program was created in 1994 as an alternative health promotion program in the country. Teams of Agentes Comunitários de Saúde (ACS - Communitarian Health Agents) visit families periodically to collect data about habitation and family and health conditions for records and to compose the Sistema de Informação da Atenção Básica (SIAB - Information System for Primary Care). ACS follows the families monthly and records the information regarding housing, sanitation and health of children 0 to 2 years old.

In periurban areas, sanitation services are usually not universal, ergo in many localities there is no access or it occurs irregularly. Precarious access is often responsible for a large part of occurrences of diarrhea, and it is associated with poverty (Blakely et al9, 2005).

The objective of this study is to verify if there is an association between the dependent variable prevalence of diarrhea and the variables: i) child characteristics (gender, age, nutritional state, breastfeeding and birth weight); ii) sanitation conditions (water supply, home water treatment and sewage); and iii) housing type.



Area and studied population

The study examined a county in the metropolitan region of São Paulo (RMSP), a low-income periurban settlement with the PSF implemented since 2005. Considering that the PSF can be an instrument in identifying priority areas of sanitation investment, a transversal study was done in order to verify the possibility of using information raised by ACS in the health-environment interface. The study field is represented by the Recreio de São Jorge neighborhood, Guarulhos, SP. The locality is attended by health unit Recreio de São Jorge, from PSF and is located in the supply reservoir catchment protection area of the Cabuçu reservoir. In 2006 (research interest period) the Recreio de São Jorge neighborhood had an estimated population of 17,502 inhabitants and unsatisfying sanitation indices with 88.7% of the houses served by the public water supply system and only 16.9% served by sewer collection. Considering the number of records in the PSF register for the period of May 2005 to April 2006, all children between 0 and 2 years old were studied during the reference period (817 children) as well as registered families responsible for these children (771 families).

Variables used and data collection

The dependent variable is the diarrhea occurrence in children between 0 and 2 years old. This indicator is used in most studies to assess sanitation impacts on public health Andreazzi et al10 (2007), for response capacity, viability in its use and easy determination of disease (Heller11, 1997). The exposure variables refer to child characteristics, housing and sanitation conditions. As information sources were used Form A - family characteristics and Form C - health conditions form and medical monitoring for children under 2 years old.

From Form A the variables used were: i) Material used in habitation construction: wood, brick, reused material, others; ii) number of rooms in the house; iii) number of residents; iv) adult residents per home, by age; v) water supply conditions: by public network, by well or spring or others; vi) kind of water treatment in house: by filtration, boiling, chlorination or without treatment; vii) waste disposal: by public collection, burned or buried, waste dump; and viii) sewer destination: sewer, sink or open sewers. The waste disposal variable is not part of the study because according SIAB information, 99% of local families were served by city public collection. Information found on Form A was collected by the ACS between March and April 2005, when the program was implemented, being that there were no sanitary or educational interventions in the referenced period (May 2005 to April 2006).

The variables used on Form C were: i) age (age group); ii) birth weight; iii) nutritional state (presented malnutrition during period, yes or no); and iv) breastfeeding (exclusive or mixed). The age variable was stratified across three age groups: 0-3 months, 4-9 months and 10 months or more. Information about child characteristics was collected monthly by the ACS during home visits.

Data analysis

Univariate analysis was initially performed between the closure diarrhea occurrence and environment and child health variables to the variables pre-selection, using the value of p<0.2 (Table 1) as a cutoff. The association measure was the Prevalence Ratio (PR) for a Confidence Interval (CI) of 95%. To calculate the age PR and the diarrhea occurrence the Mantel and Haenszel PR to CI 95%, were used (Kelsey et al12, 1996). This analysis was used to identify the variables that should be used in the logistical regression analysis. In the multivariate analysis, the multivariate logistic regression model was used, in STATA 9 statistical software. Three adjustments were made until the final model was complete, selecting the variables that were associated with the closure of statistical level of significance <0.05.

Ethical aspects

The Comitê de Ética em Pesquisa de Saúde Pública of Universidade de São Paulo (COEP - Research Ethical Committee) reviewed and approved the Research Protocol N° 1435, according to requirements of Resolução CNS/196/96 (Resolution of Brazilian National Health Council).



During the analyzed period 4,048 cases of diarrhea in children between 0-2 years old were recorded. Regarding the characteristics of children and families in the univariate analysis, greater diarrhea prevalence in children between 4-9 months old was identified. Taking the age group 4-9 months old as a reference, age group 0-3 months old and more than 10 months old showed protection effect to diarrhea occurrence. Table 1 shows that age group was associated to diarrhea occurrence. The variable elderly also was associated to malnutrition: all children with malnutrition (22) lived with adults older than 55 (chi2=784.00; p<0.001). Due to collinearity between malnutrition and elderly, the variable elderly was not contemplated in the regression model, while malnutrition remained in the analysis. Weight when born variable was not a part of the model because of high loss of record (24.7%).

Table 2 shows that material other than brick used in construction of habitation was considered a risk factor to diarrhea occurrence in children (p<0.001). This variable portrays family socioeconomic conditions and is highly correlated to sewer destination, when absence of sewage (p<0.001). Because of this high correlation between housing and sewer destination, an interaction variable was created: housing*sewer to integrate the regression model.

In the first adjustment of the regression model, three variables show significant statistical association with diarrhea occurrence: malnutrition (p=0.020), age group (10 months and more) (p=0.056) and housing*sewer (p<0.001) as risk factors. In the second adjustment of the regression model, with significant statistical variables to diarrhea occurrence, malnutrition was excluded.

Table 3 shows final logistic regression model: housing*sewer (material other than brick used in construction of habitation and absence of sewage); and age group (10 months and more) as risk factors to diarrhea occurrence. The final model showed the variables: housing*sewer (materials other than bricks used in construction of habitation and absence of sewage); and age (10 months or more) as risk factors to diarrhea occurrence (Table 3), being that the other variables were no longer significant to the closure.




The material other than brick used in the construction of habitation was identified as a risk factor to diarrhea occurrence in the univariate analysis, suggesting that the precarious housing conditions can result in home hygiene difficulty, a factor that can increase the occurrence of diarrheal diseases. On the other hand, this variable is also an indicator of poverty and low family income (IBGE13 2000).

The absence of sewage was not associated with diarrhea occurrence in then univariate analysis; however, an interaction of this variable with the use of precarious material in the construction of habitation has been identified, raising the risk of diarrhea occurrence by almost 15 times for children that live in residences where this condition was found. This result confirms the healthy housing concept of Azeredo et al14 (2007), which considers that housing is a health agent and relates to the geographic and social territory, the materials used in its construction, health education of the inhabitants and other characteristics in the surrounding context.

Absence of sewage interferes with the health of children by polluting the environment and enabling the spread of excreta related diseases, especially those of the parasitic variety that have diarrhea as the main symptom. Absence of proper sewer disposal systems in urban settlements, is not only an important cause of surface and underground water pollution, but is also a risk to the population's health, especially when there is no knowledge of waterborne diseases. Giatti15 (2004). Sanitation research in Iporanga (SP), verified that the researched watercourses presented microbiological indices that indicated the presence of pollution caused by domestic sewage due to local sanitation failure, considering that 91% of the households had feces in trenches, mostly rudimentary.

Scenarios found in Recreio de São Jorge reflect the situation of RMSP´s periurban areas, which are characterized by indices of population increase in protected areas (Porto16, 2003) or legally restricted areas. This situation generates precarious housing, water and sanitation conditions and contributes to environmental degradation mainly in reservoir catchment protected areas, besides the verified public health problems. Studies verified an association between housing in invaded areas or slums and infant mortality in Campinas (Almeida17, 2004) and the southern zone of the city of São Paulo (Shoeps18, 2007); suggest that the locality of this housing is indicative of social exclusion. In this study interaction between precarious home construction materials and non-existence of sewage indicates social exclusion and increment of conditions to pathogen exposure explains the high risk of diarrhea in children living in precarious housing with poor sanitation.

Water supply access and diarrhea occurrence do not show significant association with the research results, possibly because almost the whole population has this service. However, according to ACS, many families use water from the public supply together with well and river water because of the high level of intermittence of this service. Access to and consumption of water from public services decrease the probability of diarrheal disease occurrence because of the required potability standard in the system that guarantees the water as a safe for human consumption.

Quality and quantity of water for basic needs are not guaranteed when using alternatives sources of water (Razzolini e Günther19, 2008). However problems with intermittence in water supply enable satisfactory conditions for the infiltration of pathogens in the water supply network because of a negative pressure on it (Lee e Schwab20, 2005). Another consequence of interrupted provision is the possibility of storing water in precarious recipients without sanitation, representing a vulnerability factor to water quality.

Checkley et al21 (2004), in periurban area studies of Lima, Peru, show that young children exposed to precarious sanitation (water source, reservoir localization and sewer destination) present 54% more cases of diarrhea than non exposed children. A study in Asian periurban areas (Briscoe22, 1987) found a 33% reduction in diarrhea in children served by public water supply and sewage.

Vulnerability factors of the public network such as frequent intermittence of water supply and illegal connections expose water to contamination or housing vulnerability factors like: non existence of housing reservoir sanitation, transport and storage in inadequate recipients, introduction of objects into water reservoir and inadequate practices of users. These vulnerability factors pose the risk of water contamination.

Regarding child characteristics, it was observed that the age group of 10 months and beyond was more associated with occurrences of diarrhea. In this age group, other kinds of food were introduced into the children's diets, and without adequate health care in preparation, could present risks of pathogen transmission. Furthermore, children from this age group possess greater mobility, thus increasing the chance of environmental contamination. It was observed that age group 4-9 months is four times more likely to experience occurrences of diarrhea, but this is statistically insignificant, suggesting that the passage of exclusive breastfeeding to mixing with other foods could have some effect on occurrence of diarrhea in this age group. This fact confirms WHO´s recommendation (WHO23 2003) about the necessity of exclusive breastfeeding up to 6 months of age for children's survival and health, especially related to the prevention of diarrheal diseases.

Child malnutrition was associated with occurrences of diarrhea in the univariate analysis; but in the last adjustment of the regression model, it was excluded. Many studies show an association between child malnutrition and occurrences of diarrhea (WHO/UNICEF/USAID/SIDA24, 1990). Elderly presence in families and malnutrition were observed collinearly, suggesting that children can live in nontraditional families and such families can present unfavorable characteristics (Camarano25, 2004).

Programa Saúde da Família´s information use in environmental health studies

Results obtained show that SIAB´s information could constitute an important data source to identify populations living in situations of environmental risk.

Often environmental and health studies have been interrupted by the difficulties in obtaining local and specific data. In many cases available information is aggregated on a municipal level, masking heterogeneous living conditions and making it hard to understand the risk factors at local levels. In that sense, SIAB could represent a data source for studies about the interrelation between environment and health, contributing to the obtainment of housing, health and sanitation data that respect the heterogeneity of Brazilian urban areas.

However the use of SIAB data in environmental studies has some limitations that could be easily attenuated. Data records must be improved; especially the environmental information, which should be periodically updated since the environmental conditions are dynamic and can be altered with time. Using families' samples to inspect the homes and update the recorded information or even to add information to the existing data collection record to obtain indices according to the study's needs can be a way to handle this difficulty.

SIAB´s information could be utilized in environmental-health studies; incorporating variables that better characterize environmental and housing conditions. It could also be included as suggestions for improvement, data regarding: water origin from alternatives sources; water quality; water intermittence frequency; existence, type and quantity of domestic animals; housing surrounding conditions; housing at irregular areas or environmental vulnerability; waste reaprovement; and sanitation practices. Regarding the form on children, it is important to contain the age and mother or occupation.

A study about environmental health indicators Calijuri et al26 (2009) proposed some sanitation variables: lack of water in housing to verify the service intermittence and kind of water supply in cases of alternative sources present. There is also an interesting use of variables that identifies the presence of animals, flies and plants inside the house in the environmental health group.



Large conurbation water and sanitation indicators cannot be efficient in identifying communities that live in at-risk conditions of diarrheal occurrence. Information collected by the Programa de Saúde da Família is an excellent tool to identify settlements in which portions of the population live in poor housing conditions and have no access to water and sanitation services. In this scenario this study verified that the risk of diarrheal occurrence in children is almost 15 times greater than that of children living in good water and sanitation conditions.



1. Özkan S, Tüzün H, Görer N, Ceyhan M, Aycan S, Albayrak S. Water usage habits and the incidence of diarrhea in rural Ankara, Turkey. Trans Soc Trop Med Hyg 2007; 101(11): 1131-5.         

2. UNICEF-United Nations International Children's Emergency Fund; WHO-World Health Organization. Diarrhea: why children are still dying and what can be done. UNICEF; WHO: Geneva; New York; 2009.         

3. Fewtrell L, Kaufmann RB, Kay D, Enanoria W, Haller L, Collford JM Jr. Water, sanitation and hygiene interventions to reduce diarrhea in less developed countries: a systematic review and meta-analysis. Lancet Infec Dis 2005; 5(1): 42-52.         

4. Bozkurt AI, Ozgür S, Ozçirpici B. Association between household conditions and diarrheal diseases among children in Turkey: a cohort study. Pediatr Int 2003; 45(4): 443-51.         

5. Moe CL, Rheingans RD. Global challenges in water, sanitation and health. J Water Health 2006; 4: 41-57.         

6. Teixeira JC, Pungirum MEM de C. Análise da associação entre saneamento e saúde nos países em desenvolvimento da América Latina e do Caribe, empregando dados secundários do banco de dados da Organização Pan-Americana de Saúde - OPAS. Rev Bras Epidemiol 2005; 8(4): 356-76.         

7. Trevett AF, Carter RC, Tyrrel SF. The importance of domestic water quality management in the context of fecal-oral disease transmission. J Water Health 2005; 3(3): 259-70.         

8. Brasil. Constituição da República Federativa do Brasil (1988). Brasília: Senado; 1988.         

9. Blakely T, Hales S, Kieft C, Wilson N, Woodward A. The global distribution of risk factors by poverty level. Bull World Health Organ 2005; 83(2): 118-26,          .

10. Andreazzi MAR, Barcellos C, Hacon S. Velhos indicadores para novos problemas: a relação entre saneamento e saúde. Rev Panam Salud Publica 2007; 22(3): 211-7.         

11. Heller L. Saneamento e saúde. Brasília: OPAS; 1997.         

12. Kelsey JL, Whittemore, AS, Evans, AS, Thompson, WD. Methods in observational epidemiology. New York: Oxford University Press; 1996.           

13. IBGE-Instituto Brasileiro de Geografia e Estatística. Censo 2000. Rio de Janeiro: IBGE; 2001.         

14. Azeredo CM, Cotta RMM, Schott M, Maia T de M, Marques ES. Avaliação das condições de habitação e saneamento: a importância da visita domiciliar no contexto do Programa Saúde da Família. Ciênc Saúde Coletiva 2007; 12(3): 743-53.         

15. Giatti LL, Rocha, AA, Santos FA dos, Bitencourt SC, Pieroni SR de M. Condições de saneamento básico em Iporanga, Estado de São Paulo. Rev Saúde Pública 2004; 38(4): 571-7.         

16. Porto M. Recursos hídricos e saneamento na Região Metropolitana de São Paulo: um desafio do tamanho da cidade. Brasília: Banco Mundial; 2003.         

17. Almeida, SDM, Barros, MBA. Atenção à saúde e mortalidade neonatal. Rev Bras Epidemiol 2004; 7(1): 22-35.         

18. Shoeps D, Almeida MF de, Alencar GP, França Jr I, Novaes HMD, Siqueira AAF et al. Fatores de risco para a mortalidade neonatal precoce. Rev Saúde Pública 2007; 41(6): 1013-22.         

19. Razzolini MTP, Günther WMR. Impacto na saúde das deficiências de acesso à água. Saúde Soc 2008; 17(1): 21-32.         

20. Lee EJ, Schwab KJ. Deficiencies in drinking water distribution systems in developing countries. J Water Health 2005; 3(2): 109-27.         

21. Checkley W, Gilman R, Black RE, Epstein LD, Sterling CR, Moulton LH. Effect of water and sanitation on childhood health in a poor Peruvian peri-urban community. Lancet 2004; 363 (9403): 112-8.         

22. Briscoe J. Abastecimento de agua y servicios de saneamiento; su función em la revolución de la supervivencia infantil. Bol Oficina Sanit Panam 1987; 103(4): 325-39.         

23. WHO-World Health Organization. Infant and young child feeding: a tool for assessing national practices, policies and programmers. 2003. Disponível em http://www.who.int/nutrition/publications/infantfeeding/ [Acessado em 30 de janeiro de 2007]         

24. WHO-World Health Organization; UNICEF-United Nations International Children's Emergency Fund; USAID-United States of America Agency International Development; SIDA-Sweden International Development Agency. Innocent declaration on the protection, promotion and support of breastfeeding. Florence: WHO/UNICEF; 1990.         

25. Camarano, AA. O idoso brasileiro no mercado de trabalho. Instituto Brasileiro de Pesquisa Econômica Aplicada - IPEA, Rio de Janeiro: 2001. Disponível http://www.ipea.gov.br/pub/td/td_2001/td_0830.pdf [Acessado em 18 de abril de 2008]         

26. Calijuri ML, Santiago A da F, Camargo R de A, Moreira Neto RF. Estudo de indicadores de saúde ambiental e de saneamento em cidade do Norte do Brasil. Eng Sanit Amb 2009; 14(1): 19-28.         



Mail responsible author:
Mariana G. A. da Paz
Address: Rua Antonio Carlos, 196 ap. 56
A ZIP: 01309-010 - São Paulo/SP Brasil
E-mail: mariana_gutierres@yahoo.com.br

Received: 26/08/10
Final version: 16/03/11
Approved: 22/03/11



* Master´s research realized between 2005-2007 in the Departamento de Saúde Ambiental, Faculdade de Saúde Pública, Universidade de São Paulo (Environmental Health Department, School of Public Health, University of Sao Paulo): "Epidemiologic study in periurban locality in Guarulhos city, SP: Sanitation access and child health conditions" was defended in 2007. We state that there is no conflict of interest. - Research funding: ICO-DEV - International Cooperation with Developing Countries mediante contrato ICA4-CT-2002-10061.

Associação Brasileira de Pós -Graduação em Saúde Coletiva São Paulo - SP - Brazil
E-mail: revbrepi@usp.br