PHIRN: Population Health Improvement Research Network


smaller text tool iconmedium text tool iconlarger text tool icon

The online registry is a free, searchable database of quality-appraised systematic reviews evaluating the effectiveness of public health interventions. The registry is populated through an extensive ongoing search (1985-present) of seven electronic databases (MEDLINE, EMBASE, CINAHL, PsycINFO, Sociological Abstracts, BIOSIS, SportDiscus), handsearching of 46 journals, and screening the reference lists of all relevant reviews [10]. Reviews are assessed for relevance, and then relevant reviews are indexed by commonly-used public health terms and quality assessed by two independent reviewers who come to agreement on the final rating of each review (strong, moderate, weak). More detail on has previously been published [10].

The registry was used to search for reviews on the built environment in May 2011. The term “built environment” was used to search the Main topic lists and Intervention strategy sections of the registry. All search results were limited to reviews rated as being of strong methodological quality.

Two reviewers used a standardized quality assessment tool to assess the methodological quality of each identified review. Using a ten-point quality assessment tool (available at:, all reviews were assessed by two reviewers independently and disagreements resolved through discussion. The ten criteria used to assess methodological quality were: (1) a clearly focused question; (2) inclusion criteria explicitly stated; (3) comprehensive search strategy; (4) adequate number of years covered in the search; (5) description of level of evidence; (6) assessment of the methodological rigor of primary studies; (7) methodological quality of primary studies assessed by two reviewers and results given; (8) tests of homogeneity or assessment of similarity of results conducted and reported; (9) appropriate weighting of primary studies; and (10) author’s interpretation of results supported by the data. Each criterion, worth one point each, was given equal weight in the overall assessment score. Reviews were given an overall score out of 10 and were classified into three categories: Strong, Moderate, and Weak. Reviews receiving an overall rating of eight or more were considered strong, those with a score of five to seven, moderate, and those with four or less, weak. Quality ratings for reviews included in this synthesis project ranged from 8-10.

All outcome data was extracted from all the reviews included in this project and organized into a matrix table. Additional tables were then created according to the most prominent outcomes to further summarize and present the data for each of the topic areas. Data extraction was conducted on strong reviews using a previously developed tool. Extracted data included author and year of publication, methodological quality rating, author’s country, number of included studies, research design, population examined, interventions evaluated, details describing which outcomes where evaluated, the effectiveness of the intervention, the results, and additional comments. The data are presented in Tables 1 to 3.