Presenting Author: Ana Ortigoza, MD, MS, Drexel University Urban Health Collaborative

ABSTRACT

Background: Country-level population health surveys are necessary for health surveillance. Health surveys may also be useful for cross-country studies examining population health trends, but require harmonization to make meaningful comparisons.

Objectives: To describe the collection and harmonization of population health surveys from SALURBAL. An additional aim was to assess the availability of georeferenced data for city and subcity-level analyses across and within countries.

Methods: Survey availability was systematically determined for each country from official sources, the Pan-American Health Organization (PAHO), and local collaborators. Harmonization focused on demographics, anthropometry, diet, physical activity, smoking, alcohol consumption, hypertension, and diabetes. Questions were collated thematically to compare wording and responses to determine composite variable feasibility. We also determined the smallest available geographic identifier for participants.

Results: Seven priority surveys were identified in ten countries (Argentina, Brazil, Chile, Colombia, Costa Rica, El Salvador, Guatemala, Mexico, Nicaragua, and Peru), spanning 2007 to 2016. All were national surveys with multi-stage sampling of males and females, 17 years old. We found completely consistent variables for: age, sex, marital status, occupation, housing conditions, current smoking, and diabetes and hypertension diagnosis and treatment. We found consistency in 75% for: education, health insurance, income, weight, height, dietary habits, type of physical activity, gestational diabetes, blood pressure, and glucose levels. Available geographic levels ranged from census- block to county. Other variables were consistent in 75% of surveys.

Implications: This harmonization process for city-level analyses may be useful for future multi-country studies that require collection of heterogeneous secondary data.

Authors: SALURBAL Investigators