Enhancing COPD Management in Primary Care Settings

Jill A. Foster, MD, MPH; Barbara P. Yawn, MD, MS, MSc, FAAFP; Maziar Abdolrasulnia, PhD; Todd Jenkins, MPH; Stephen I. Rennard, MD, FCCP; Linda Casebeer, PhDs

In This Article

Materials and Methods

Survey Development

To investigate the needs and practice patterns of primary care physicians in COPD management, a 30-item case-based survey was developed and validated for face validity with Family Medicine and Pulmonary Medicine experts. The first step in developing the survey items was a review of the scientific literature that examined gaps between actual clinical practice and guideline recommendations, using Global Strategy for the Diagnosis, Management, and Prevention of Chronic Obstructive Pulmonary Disease, Updated 2005[9] as a benchmark. Practice issues that appeared important but were not well studied were also identified. A second step used a Nominal Group Technique to understand significant barriers to optimal COPD diagnosis and management from the perspective of primary care physicians. Finally, case writers used this information to create COPD case-vignettes with multiple-choice questions that examined diagnostic and management choices. This approach was chosen because clinical vignettes have shown good correlation with physicians' actual practice choices[20,21] and is cost-efficient. Additional questions examining perceptions about COPD, guideline awareness, and learning preferences were also included.

Survey Distribution

Surveys were distributed by e-mail and fax to a random sample of US primary care physicians during November and December 2006. Individuals who did not respond after 3 contacts were replaced by others drawn randomly until a usable sample of at least 700 respondents had been achieved. A small monetary incentive was offered for completing the survey.


To visually assess factors relevant to COPD care, maps were created using ArcView v9.1 (Environmental Research Systems Institute [ESRI], Redlands, Calif.) to display COPD mortality rates, smoking prevalence, and pulmonologist practice locations. Crude county-level COPD mortality rates [ICD10: J40-J44] from the years 1999-2002 combined[22] were categorized and plotted as quartiles. Counties with fewer than 20 COPD deaths were considered to have unreliable rates, and were classified as 'insufficient data.' States with the highest smoking prevalence rates in the 2005 Behavioral Risk Factor Surveillance System (BRFSS)[23] were identified and indicated using cross-hatching. Pulmonologist office locations were acquired from the Physician Masterfile of the American Medical Association (AMA).[24] These practice locations were then geocoded using ArcView v9.1 and Streetmap USA (ESRI, Redlands, Calif.), a reference street network database derived from Census 2000.

Statistical Analysis

For categorical data, chi-square analyses were performed. T-tests were used to evaluate normally distributed continuous data. All analyses were performed using SAS System v9.1 (SAS Institute, Inc.; Cary, NC). The level of statistical significance was set at P < .05.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: