Yoshimi Anzai MD, Presenter: Nothing to Disclose

William Hollingworth PhD, Abstract Co-Author: Nothing to Disclose

Sean D Sullivan PHD, Abstract Co-Author: Nothing to Disclose

Jeffrey Gil Jarvik MD, MPH, Abstract Co-Author: Nothing to Disclose

Acute bacterial sinusitis is clinically over-diagnosed and antibiotics (Abx) are over-prescribed. This results in Abx resistant infection, which is a major public health problem. Purpose of this study is to perform cost utility analysis of diagnosis and management of acute sinusitis from societal perspective using Markov model.

Four strategies are compared; 1) empirical Abx treatment, 2) CT based Abx treatment, 3) clinical guideline based Abx treatment, and 4) no Abx. The analysis was targeted to urgent care setting for adult patients. Time horizon was 1 month. Quality adjusted life day (QALD) was measured based on the Health Utility Index. Costs estimate included cost of diagnostic test, antibiotics, lost days from work, and patient time for CT scan. The basic assumptions included a) no recurrent infection, b) no complication from not treating acute sinusitis, and c) no adverse events from Abx treatment. Markov model included two states of disease (sick or well) and utility of 0.82 and 1.0 were assigned from Health Utility Index, respectively. Cost of antibiotics was calculated from a weighted average of the frequency of prescription and cost of each drug. Medicare reimbursement was used to estimate cost of screening sinus CT.

Clinical guideline based treatment was most inexpensive strategy and empirical Abx treatment was cost effective ($23/QALD: $8450/QALY) compared with clinical guideline based treatment with the baseline parameter estimates (37% prevalence of acute sinusitis, $48.40 for antibiotics, 83% sensitivity of CT, 70% specificity of CT). Empirical Abx treatment was most cost effective in a setting of high prevalence of acute sinusitis. When prevalence of acute sinusitis is low and cost of Abx is high, clinical guideline based treatment was most cost effective strategy.

Clinical guideline based treatment or empirical Abx treatment are cost effective strategies, depending on the prevalence of acute bacterial sinusitis and cost of Abx. The 'No Abx' strategy is not cost effective, as patients remain sick for a longer period costing more to the society. CT guided treatment was not cost-effective across a wide range of baseline assumptions.

Y.A.: This study was supported in part by AHRQ K-award, K08 HS 013613-02

Anzai, Y, Hollingworth, W, Sullivan, S, Jarvik, J, Cost Effectiveness Analysis of Diagnosis and Treatment of Acute Bacterial Sinusitis.  Radiological Society of North America 2005 Scientific Assembly and Annual Meeting, November 27 - December 2, 2005 ,Chicago IL. http://archive.rsna.org/2005/4415598.html