AIM: To describe trends in the prescription of medication for acne vulgaris in the USA from 1990 to 2002 with particular reference to the shift from antibiotic to non-antibiotic treatment. METHODS: Retrospective cross-sectional observational study which analysed the data from the 4922 patient visits for acne vulgaris from the 1990-2002 National Ambulatory Medical Care Survey (NAMCS) to assess medications mentioned or prescribed at office visits for acne vulgaris. We specifically analysed visits for benzoyl peroxide or combination benzoyl peroxide products, topical retinoids, clindamycin, erythromycin, tetracycline group antibiotics and isotretinoin. A priori predictions were made for changes in physician behaviour over time, and were tested by regression. Multivariate regression also assessed the impact of demographic variables such as patient age, gender, race and type of physician seen on prescribing behaviour. RESULTS: From 1990 to 2002 there were significant declines (p<0.01) in the likelihood of use of several drug classes that are reliant on antimicrobial mechanisms for acne including the following: benzoyl peroxide, topical clindamycin, oral erythromycin and tetracycline group antibiotics. There were significant increases in likelihood of utilization for agents not reliant on antimicrobial mechanisms including topical retinoids and oral isotretinoin during this period. These changes in utilization were not affected by controlling for demographic factors in multivariate analysis. CONCLUSION: There has been a shift towards non-antibiotic treatment in acne management. Growing awareness of antibiotic-resistant Propionibacterium species may contribute to increased use of topical and systemic retinoid agents.
AIM: To describe trends in the prescription of medication for acne vulgaris in the USA from 1990 to 2002 with particular reference to the shift from antibiotic to non-antibiotic treatment. METHODS: Retrospective cross-sectional observational study which analysed the data from the 4922 patient visits for acne vulgaris from the 1990-2002 National Ambulatory Medical Care Survey (NAMCS) to assess medications mentioned or prescribed at office visits for acne vulgaris. We specifically analysed visits for benzoyl peroxide or combination benzoyl peroxide products, topical retinoids, clindamycin, erythromycin, tetracycline group antibiotics and isotretinoin. A priori predictions were made for changes in physician behaviour over time, and were tested by regression. Multivariate regression also assessed the impact of demographic variables such as patient age, gender, race and type of physician seen on prescribing behaviour. RESULTS: From 1990 to 2002 there were significant declines (p<0.01) in the likelihood of use of several drug classes that are reliant on antimicrobial mechanisms for acne including the following: benzoyl peroxide, topical clindamycin, oral erythromycin and tetracycline group antibiotics. There were significant increases in likelihood of utilization for agents not reliant on antimicrobial mechanisms including topical retinoids and oral isotretinoin during this period. These changes in utilization were not affected by controlling for demographic factors in multivariate analysis. CONCLUSION: There has been a shift towards non-antibiotic treatment in acne management. Growing awareness of antibiotic-resistant Propionibacterium species may contribute to increased use of topical and systemic retinoid agents.