Novel application of published risk factors for methicillin-resistant S. aureus in acute bacterial skin and skin structure infections.

ABSTRACT

Methicillin-resistant Staphylococcus aureus acute bacterial skin and skin structure infections (MRSA ABSSSIs) are associated with a significant clinical and economic burden; however, rapid identification of MRSA remains a clinical challenge. This study aimed to use a novel method of predictive modeling to determine those at highest risk of MRSA ABSSSIs. Risk factors for MRSA ABSSSI were derived from a combination of previously published literature and multivariable logistic regression of individual patient data (IPD) using the 'adaptation method.' A risk-scoring tool was derived from weight-proportional integer-adjusted coefficients of the predictive model. Likelihood ratios were used to adjust posterior probability of MRSA. Risk factors were identified from 12 previously published studies and adapted based on IPD (n = 231). Risk factors were history of diabetes with obesity (adapted odds ratio [aOR] = 1.1), prior antibiotics (90 days) (aOR = 2.6), chronic kidney disease/hemodialysis (aOR = 1.4), intravenous drug use (aOR = 2.8), previous MRSA exposure/infection (12 months) (aOR = 2.8), previous hospitalization (12 months) (aOR = 7.5), and HIV/AIDS (aOR = 4.0). Baseline prevalence of MRSA was 42.7%. Scores ranged from 0 - 8 points. Post-test probability of MRSA score 0 = 35.0%; score 1 - 2 = 45.0%; score 3 = 63.0%. The newly derived risk-scoring tool is proof-of-concept of the adaptation method. This study is hypothesis generating and such a tool remains to be validated for clinical use.