ASCO 2014. Abstract 627. An electronic prompt prior to myelosuppressive therapy to improve hepatitis B virus screening.
Jordan J. Feld, Lisa K. Hicks, Joshua Juan, Truong Judy, Urszula Zurawska, Angie Giotis, Kelvin K. Chan; University Health Network, University of Toronto, Toronto, ON, Canada; St. Michael's Hospital, University of Toronto, Toronto, ON, Canada; University Health Network, Toronto, ON, Canada; Odette Cancer Centre, Sunnybrook Health Sciences Centre, Toronto, ON, Canada; St. Joseph's Health Centre, Toronto, ON, Canada; Sunnybrook Health Sciences Centre, Toronto, ON, Canada
Background: Hepatitis B virus (HBV) reactivation is a potentially fatal and yet preventable complication of myelosuppressive therapy. However, HBV screening rates have been low despite the recommendation from the Centre of Disease Control (CDC) to screen all patients prior to myelosuppressive therapy. We evaluated the effectiveness of an electronic prompt on HBV screening rates.
Methods: An electronic prompt was established in Nov 2010 at a large academic oncology centre in Toronto (study centre). The electronic prompt reminded ordering physicians to order HBV surface antigen (HBsAg) at the time of booking a new patient’s first chemotherapy electronically. For physicians who agreed, HBsAg was automatically ordered. The prompt was not implemented at another large academic oncology centre in Toronto (control centre). Both centres received the same educational rounds in Nov 2010. The primary endpoint was the rate of HBV screening. Actual HBV screening rates were determined in both centres both prior to the intervention (Nov 2009 to Oct 2010) and during the intervention (Nov 2010 to Oct 2011). Multivariable logistic regression with random effects was conducted to assess the effect of the electronic prompt adjusting for potential confounders and clustering of patients by physicians.
Results: 6,116 patients received first chemotherapy during the study period (2,095 study centre, 4,021 control center). In the pre-prompt period, the screening rate was 17% in the study centre and 25% in the control centre. In the prompt period, the screening rate increased to 61% in the study centre and was unchanged at 25% in the control centre. The overall screening rate for patients with hematological (heme) malignancies was higher (62.3% vs. 20.4% in non-heme patients, p<0.0001). After adjusting for study period, centres and heme malignancies, the electronic prompt improved the HBV screening rate significantly (odds ratio 11.0, 95% CI 8.3-14.7, p<0.0001), which remained significant adjusting for clustering of patients by physicians (p<0.0001).
Conclusions: An electronic prompt increased the rate of HBV screening, however screening rates remained relatively low. Educational rounds did not appear to improve the HBV screening rate.