PURE-AIR: Home | Cohort | Outdoor air | Household air | Epidemiological analyses

The PURE cohort is run out of the Population Health Research Institute.

The PURE cohort is an established prospective cohort study that recruited over 155,000 adults (35–70 years of age) as of December 2013 from 17 low, middle and high- income countries and has conducted 4 years of follow-up. An additional six years of follow up is scheduled. PURE was a multi-stage convenience-sample, with countries and communities chosen purposively at the first and second stage, and individuals selected by random sampling at the third stage. At the community level 123,114 households were adjudicated. Within these household 197,332 individuals were between 35-70 years of age, of which 153,662 (77.9%) agreed to participate in the PURE study. This includes 85,461 women and 68,117 men. The objective of PURE is to examine a broad range of factors that influence disease in different social, cultural, physical and economic environments. The study is coordinated by the Population Health Research Institute (PHRI), Hamilton, Canada.


PURE CommunitiesThe 155,327 individuals currently recruited in PURE live in 628 communities (Figure 1) that are located in 4 low-income countries: Zimbabwe, Bangladesh, Pakistan, India; 10 middle-income countries: South Africa, Brazil, Chile Argentina, Columbia, Poland, China, Malaysia, Iran, Turkey; and 3 high income countries: Canada, Sweden, UAE. While each country aimed to include a diverse range of communities, most countries identified communities to be included purposefully or conveniently. Both urban and rural communities were included in the PURE study since in many countries urban and rural environments exhibit distinct social and physical determinants of health. For some countries (e.g. India, China, Canada, Turkey, Columbia, Malaysia) communities from urban and rural areas in different states/provinces have been included to capture the diversity in environmental factors (which incorporates further variation in outdoor air pollution levels).

In smaller countries between one and three pairs of urban/rural communities have been selected. PURE study communities represent small areas (i.e. neighborhoods) within cities, while in rural areas represent a small town or village. Previous large air pollution epidemiological studies have used entire cities as the geographic unit for exposure assessment and the smaller size of the PURE communities is a substantial improvement that will reduce exposure misclassification.

PURE-AIR communities

Figure 1. Location of the 628 PURE study communities located in 17 countries. The satellite photographs (left) show the size of PURE study communities in Vancouver, Canada which is an urban site; and Pulamhue, Chile which is a rural site.

Baseline Data Collection: Within each PURE community, recruitment was designed to achieve a representative sample of adults aged 35- 70 years. The age, sex, urban/rural status, educational profiles and mortality statistics of the PURE sample have been validated against national statistics and show good agreement. In addition, information was collected on non-responders to further examine potential selection bias. Data was collected using standardized protocols at the community, household and individual levels. Data are entered online by each center directly to the central PHRI database and evaluated for incomplete and inconsistent data. Briefly, community level data include information on the social and environmental factors that influence health (e.g. tobacco, nutrition, and built environment); household data includes information on house structure, solid fuel use, and ventilation (which will allow this study to control for household air pollution exposure as well as examine joint exposures); family data includes the number of individuals in a household and their tobacco use, education and major morbidities; and individual data includes information on the nine risk factors for CVD identified in the INTERHEART study: lipids, smoking, hypertension, diabetes, abdominal obesity, psychosocial factors (stress and symptoms of depression), diet and alcohol and regular physical activity. Detailed information on physical activity and diet was collected using the International Physical Activity Questionnaire (IPAQ) (or regional questionnaires with comparable variables) and semi-quantitative Food Frequency Questionnaires (FFQs). All surveys were validated in each country.

Upon entry into the PURE cohort, participants provided a self-reported medical history. This included information on MI, heart failure, stoke, COPD, asthma and health symptoms experienced over the last six months (e.g. cough for at least 2 weeks, wheezing or whistling in chest, etc.). Verification of self-report diagnoses obtained through the interview-based questionnaires was performed by examining a sub-sample of 455 reported cardiac events using medical and hospital records, which showed a confirmation rate of 89% during central adjudication. In addition, a physical examination was conducted that included spirometry (FEV1, FVC), blood pressure and an electrocardiogram (ECG). A 10 ml fasting blood sample (serum, plasma and genetic materials) was also collected from all consenting participants and transferred to centralized storage.

The availability of spirometry in the PURE cohort offers an unprecedented opportunity to examine the effect of air pollution on adult respiratory health.  Lung function was measured following ATS protocol with a portable Spirometer (MicroGP, MicroMedical, Chatham, IL, USA). Each subject attempted up to six forced manoeuvres, while standing and wearing a nose clip. Each maneuver was observed to ensure maximal effort and forced exhalation for 6 seconds. The three highest forced expiratory volume in 1 second (FEV1) and forced vital capacity (FVC) were recorded.

Cohort Follow-Up: Currently, the mean follow-up duration of the cohort is 4 years with an overall retention rate of ~93%. The goal of the PURE cohort is to have loss to follow-up <10% after 10 years of follow up. Follow-up visits involve simple procedures and active follow-up of missed visits is conducted with verified primary and secondary contact information. Repeat surveys every three years will establish changes in common risk factors for cardiovascular and respiratory disease, including dietary measures, obesity, and smoking information. Biannual contacts are made to document major morbidity and mortality (coded for cause using ICD-10 codes). Health events captured in the PURE follow-up are ascertained with standardized case report forms, which are entered online by each Center directly to the central PHRI database and evaluated for incomplete and inconsistent data.