Representativeness of Patient Visits to a National Practice-Based Research Network: A Comparison of Pediatric Research in Office Settings (PROS) and National Ambulatory Medical Care Survey (NAMCS) Data

EJ Slora ¹, K A Thoma ¹, RC Wasserman ^1,2, SE Pedlow ³ and AB Bocian ¹. ¹Dept of Practice & Research; Ctr for Child Health Research, AAP, Elk Grove Village, IL, United States; ²Dept of Peds, University of Vermont, Burlington, Vermont, United States and ³Stats and Mthd, NORC, Chicago, IL, United States.

Background: Practice-based research network (PBRN) studies have become important in primary-care research; however, it is unclear to what extent PBRN samples are generalizable to those from typical primary-care settings.

Objective: To assess the extent to which patient visits to Pediatric Research in Office Settings (PROS), the national pediatric PBRN, are representative of those seen in US pediatric primary care as described by the National Ambulatory Medical Care Survey (NAMCS).

Design/Methods: PROS data were collected in the offices of 57 active network practitioners identified by stratified random sampling, and included 1,706 randomly selected pediatric visits between March and June 2002. National comparison data were the 948 pediatric visits seen March-June 2000 in the offices of 33 primary-care pediatric practitioners participating in NAMCS. The groups were compared on patient demographics and visit characteristics.

Results: Comparisons revealed many similarities on PROS and NAMCS patient demographics, including gender (53.9% males in the PROS sample vs 53.4% for NAMCS, respectively), Hispanic/Latino ethnicity (10.6% vs 13.1%), and Medicaid status (22.0% for PROS vs 22.3% for NAMCS). PROS children were one year older than their NAMCS counterparts (6.45 years vs 5.47, p<.0001). PROS had a smaller proportion of visits from African-Americans (7.5% vs 17.6 %, p<.0001) and a higher proportion of visits from multiracial children (6.8% vs 0.5%, p<.0001). With respect to visit characteristics, similarities were observed for acute (vs chronic) visits (57.2% acute visits for PROS vs 55.6% for NAMCS, respectively), top 5 physician diagnoses (well child, OM, URI, pharyngitis, and virus), and proportions of patients referred (4.3% vs 3.8%). Similarities also were noted for the top 6 visit chief complaints (well child, cough, fever, sore throat, earache, rash), with a small difference in the proportions of the remaining complaints (40.7% vs 45.4%).

Conclusions: With the exception of modest differences in patient age and racial distribution, visits in the PROS sample were remarkably similar to those in NAMCS. These data suggest that visits to PROS practitioners are largely representative of and generalizable to visits to pediatricians nationwide.