Slum upgrading strategies involving physical environment and infrastructure interventions and their effects on health and socio‐economic outcomes

Background Slums are densely populated, neglected parts of cities where housing and living conditions are exceptionally poor. In situ slum upgrading, at its basic level, involves improving the physical environment of the existing area, such as improving and installing basic infrastructure like water, sanitation, solid waste collection, electricity, storm water drainage, access roads and footpaths, and street lighting, as well as home improvements and securing land tenure. Objectives To explore the effects of slum upgrading strategies involving physical environment and infrastructure interventions on the health, quality of life and socio-economic wellbeing of urban slum dwellers in low and middle income countries (LMIC). Where reported, data were collected on the perspectives of slum dwellers regarding their needs, preferences for and satisfaction with interventions received. Search methods We searched for published and unpublished studies in 28 bibliographic databases including multidisciplinary (for example Scopus) and specialist databases covering health, social science, urban planning, environment and LMIC topics. Snowballing techniques included searching websites, journal handsearching, contacting authors and reference list checking. Searches were not restricted by language or publication date. Selection criteria We included studies examining the impact of slum upgrading strategies involving physical environment or infrastructure improvements (with or without additional co-interventions) on the health, quality of life and socio-economic wellbeing of LMIC urban slum dwellers. Randomised controlled trials (RCTs), controlled before and after studies (CBAs) and interrupted time series (ITS) were eligible for the main analysis. Controlled studies with only post-intervention data (CPI) and uncontrolled before and after (UBA) studies were included in a separate narrative to examine consistency of results and to supplement evidence gaps in the main analysis. Data collection and analysis Two authors independently extracted data and assessed risk of bias for each study. Differences between the included study interventions and outcomes precluded meta-analysis so the results were presented in a narrative summary with illustrative harvest plots. The body of evidence for outcomes within the main analysis was assessed according to GRADE as very low, low, moderate or high quality. Main results We identified 10,488 unique records, with 323 screened as full text. Five studies were included for the main analysis: one RCT with a low risk, two CBAs with a moderate risk and two CBAs with a high risk of bias. Three CBAs evaluated multicomponent slum upgrading strategies. Road paving only was evaluated in one RCT and water supply in one CBA. A total of 3453 households or observations were included within the four studies reporting sample sizes. Most health outcomes in the main studies related to communicable diseases, for which the body of evidence was judged to be low quality. One CBA with a moderate risk of bias found that diarrhoeal incidence was reduced in households which received water connections from a private water company (risk ratio (RR) 0.53; 95% confidence interval (CI) 0.27 to 1.04) and the severity of diarrhoeal episodes (RR 0.48; 95% CI 0.19  to 1.22). There was no effect for duration of diarrhoea. Road paving did not result in changes in parasitic infections or sickness in one RCT. After multicomponent slum upgrading, claims for a waterborne disease as opposed to a non-waterborne disease reduced (RR 0.64; 95% CI 0.27 to 0.98) in one CBA with a high risk of bias but there was no change in sanitation-related mortality in a CBA with a moderate risk of bias. The majority of socio-economic outcomes reported within the main studies related to financial poverty, for which the body of evidence was of very low quality. Results were mixed amongst the main studies; one RCT and two CBAs reported no effect on the income of slum dwellers following slum upgrading. One further CBA found significant reduction in monthly water expenditure (mean difference (MD) -17.11 pesos; 95% CI -32.6 to -1.62). One RCT also showed mixed results for employment variables, finding no effect on unemployment levels but increased weekly worked hours (MD 4.68; 95% CI -0.46 to 9.82) and lower risk of residents intending to migrate for work (RR 0.78; 95% CI 0.60 to 1.01). There was no evidence available to assess the impact of slum upgrading on non-communicable diseases or social capital. Maternal and perinatal conditions, infant mortality, nutritional deficiencies, injuries, self-reported quality of life, education and crime were evaluated in one study each. Nine supporting studies were included that measured varying outcomes (6794 households or observations within eight studies reporting sample sizes). One CPI evaluated cement flooring only while three UBAs and five CPIs evaluated multicomponent slum upgrading strategies. All studies but one had a high risk of bias. The studies reinforced main study findings for diarrhoea incidence and water-related expenditure. Findings for parasitic infections and financial poverty were inconsistent with the main studies. In addition, supporting studies reported a number of disparate outcomes that were not evaluated in the main studies. Five supporting studies included some limited information on slum dweller perspectives. They indicated the importance of appropriate siting of facilities, preference for private facilities, delivering synergistic interventions together, and ensuring that infrastructure was fit for purpose and systems were provided for cleaning, maintenance and repair. Authors' conclusions A high risk of bias within the included studies, heterogeneity and evidence gaps prevent firm conclusions on the effect of slum upgrading strategies on health and socio-economic wellbeing. The most common health and socio-economic outcomes reported were communicable diseases and indicators of financial poverty. There was a limited but consistent body of evidence to suggest that slum upgrading may reduce the incidence of diarrhoeal diseases and water-related expenditure. The information available on slum dwellers’ perspectives provided some insight to barriers and facilitators for successful implementation and maintenance of interventions. The availability and use of reliable, comparable outcome measures to determine the effect of slum upgrading on health, quality of life and socio-economic wellbeing would make a useful contribution to new research in this important area. Given the complexity in delivering slum upgrading, evaluations should look to incorporate process and qualitative information alongside quantitative effectiveness data to determine which particular interventions work (or don’t work) and for whom.