A Risk Prediction Model for Sporadic CRC Based on Routine Lab Results

http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = Dig Dis Sci (2016) 61:20762086 DOI 10.1007/s10620-016-4081-x

http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = ORIGINAL ARTICLE

http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = http://crossmark.crossref.org/dialog/?doi=10.1007/s10620-016-4081-x&domain=pdf

Web End = A Risk Prediction Model for Sporadic CRC Based on Routine Lab Results

Ben Boursi1,2,3,5 Ronac Mamtani2,3,4 Wei-Ting Hwang2,3 Kevin Haynes2,3

Yu-Xiao Yang1,2,3

Received: 15 September 2015 / Accepted: 5 February 2016 / Published online: 19 February 2016 Springer Science+Business Media New York 2016

AbstractBackground Current risk scores for colorectal cancer (CRC) are based on demographic and behavioral factors and have limited predictive values.

Aim To develop a novel risk prediction model for sporadic CRC using clinical and laboratory data in electronic medical records.

Methods We conducted a nested casecontrol study in a UK primary care database. Cases included those with a diagnostic code of CRC, aged 5085. Each case was matched with four controls using incidence density sampling. CRC predictors were examined using univariate conditional logistic regression. Variables with p value \0.25 in the univariate analysis were further evaluated in

multivariate models using backward elimination. Discrimination was assessed using receiver operating curve. Calibration was evaluated using the McFaddens R2. Net reclassication index (NRI) associated with incorporation of laboratory results was calculated. Results were internally validated.

Results A model similar to existing CRC prediction models which included age, sex, height, obesity, ever smoking, alcohol dependence, and previous screening colonoscopy had an AUC of 0.58 (0.570.59) with poor goodness of t. A laboratory-based model including hematocrit, MCV, lymphocytes, and neutrophillymphocyte ratio (NLR) had an AUC of 0.76 (0.760.77) and a McFaddens R2 of 0.21 with a NRI of 47.6 %. A combined model including sex, hemoglobin, MCV, white blood cells, platelets, NLR, and oral hypoglycemic use had an AUC of0.80 (0.790.81) with a McFaddens R2 of 0.27 and a NRI of 60.7 %. Similar results were shown in an internal validation set.

This work was performed in partial fulllment of the requirements for a Ph.D. degree of Ben Boursi, Sackler Faculty of Medicine, Tel-Aviv University, Israel.

Electronic supplementary material The online version of this article (doi:http://dx.doi.org/10.1007/s10620-016-4081-x

Web End =10.1007/s10620-016-4081-x ) contains supplementary material, which is available to...