Clinical approach to non-resolving pneumonia: a survey of Wisconsin primary care clinicians


Department of Family Medicine, Center for Urban Population Health, Aurora UW Medical Group

Presentation Notes

Presented at 2013 Aurora Scientific Day, Milwaukee, WI


Background/significance: Little has been reported regarding the approach of primary care clinicians to ambulatory, non-responsive pneumonia (ANRP), variously defined as pneumonia which has not improved with 3-10 days of antibiotic therapy. Anecdotal experience suggests that changes in antibiotic prescription may occur without diagnostic testing for uncovered etiologic agents such as fungi (estimated 1-7% of pneumonia in Wisconsin).

Purpose: To determine the stated approach to this clinical problem by primary care clinicians. Methods: A survey containing an algorhythmic, scenario based clinical case of ANRP was sent electronically to the 103 eligible members of the Wisconsin Research and Education Network Survey Group of Wisconsin area primary care clinicians. Respondents were presented with potential diagnostic and therapeutic responses to the case scenario which was constructed from recent consensus guidelines.

Results: Surveys were returned from 53/103 members of which 44 were completed (61% male, 30% rural, mean 20 years practice, 13 counties). X-rays were “ordered” by 39 (89%) at initial presentation, and all 39 ordered antibiotics (70% azithromycin). After 4 days of no response, 25/39 (64%) did additional diagnostic testing, 26/39 (67%) changed antibiotics (5 without further testing); after 11 days of no response 3/39 (59%) ordered CT, 16 (41%) bronchoscopy and 28 (72%) either CT or bronchoscopy. Five clinicians (11%) did not order initial X-rays (3/5 started antibiotics), nor did they order further testing after 4 days of no response. Overall, 29/44 (66%) had done no sputum or antigen testing for bacteria by 4 days, and 24/44 (55%) had done no specific testing for fungi by 11 days of ANRP.

Conclusion: When presented with an ANRP scenario, a majority of regional primary care clinicians both change antibiotics and do further testing after 4 days of no response, but do not test for fungi by 11 days. These findings highlight the need for practice guidelines based on research outcomes and expert experience to establish pathways for optimal treatment of ANRP. Studies are underway to determine what is done in actual practice.

Document Type


Link to Full Text