Colorectal cancer screening comparing no screening, immunochemical and guaiac fecal occult blood tests: A cost‐effectiveness analysis

Abstract

Comparability of cost-effectiveness of colorectal cancer (CRC) screening strategies is limited if heterogeneous study data are combined. We analyzed prospective empirical data from a randomized-controlled trial to compare cost-effectiveness of screening with either one round of immunochemical fecal occult blood testing (I-FOBT; OC-Sensor V R ), one round of guaiac FOBT (G-FOBT; Hemoccult-II V R ) or no screening in Dutch aged 50 to 75 years, completed with cancer registry and literature data, from a third-party payer perspective in a Markov model with first-and second-order Monte Carlo simulation. Costs were measured in Euros (€), effects in life-years gained, and both were discounted with 3%. Uncertainty surrounding important parameters was analyzed. I-FOBT dominated the alternatives: after one round of I-FOBT screening, a hypothetical person would on average gain 0.003 life-years and save the health care system €27 compared with G-FOBT and 0.003 life years and €72 compared with no screening. Overall, in 4,460,265 Dutch aged 50-75 years, after one round I-FOBT screening, 13,400 life-years and €320 million would have been saved compared with no screening. I-FOBT also dominated in sensitivity analyses, varying uncertainty surrounding important effect and cost parameters. CRC screening with I-FOBT dominated G-FOBT and no screening with or without accounting for uncertainty.Cost-effectiveness studies on colorectal cancer (CRC) screening suffer from heterogeneous data based on partially ambiguous evidence.1 Usually, the effect of CRC screening is performed by exploring a single type of test or strategy. A literature search showed that, if data from studies with more than one type of test were available, they suffered from no random allocation of tests and limited comparability due to differences in inclusion criteria, follow-up periods and geographical regions. [1][2][3][4][5][6] Most studies are accompanied by extensive sensitivity analyses covering a broad range of outcomes, but authors seldom report on decreased precision and possible lack of generalizability. [2][3][4][5][6][7] Besides these forms of potential biases, several cost-effectiveness studies suffered from unrealistic assumptions. For example, participation in screening can vary between populations and screening tests. Primary endoscopy screening hardly ever exceeds 30% and is consistently lower than participation in guaiac fecal occult blood test (FOBT) screening, which hardly ever exceeds 50% participation. [8][9][10][11][12] However, in cost-effectiveness analysis, participation is often assumed to be 100%, regardless of the population, test or strategy. When imperfect participation is taken into account, it is often assumed to be only dependent on the population and not on the test or strategy. 3,5,7 In reality, participation levels related to different types of tests can vary considerably within a population, because of test-specific aspects such as the design of the test or the number of days a test has to be performed. Also, adherence to colonos...

S3-Leitlinie – Kolorektales Karzinom

“…FlexSure ® ) verglichen wurde, fand sich hingegen kein signifikanter Unterschied [139]. Insbesondere in den beiden größten randomisierten Studien aus den Niederlanden [103,140] fand sich jedoch ein signifikanter Unterschied, der für eine Überlegenheit des eingesetzten iFOBT (OC-Sensor) gegenüber dem Hämoccult ® Test spricht.…”

Alternatives to colonoscopy for population-wide colorectal cancer screening

“…13,15,16,20,33 The iFOBT is more costly than its guaiacbased counterpart, 20 but modelling studies showed that it is more cost-effective. [34][35][36][37] This is largely explained by the higher participation rate, detection rate, sensitivity and PPV, and with lower NNScope and NNScreen. There is a general consensus that it should replace gFOBT.…”

Laboratory screening markers in gastroenterology - state of the art

“…The collection of two stool samples improves the qiFOBT´s diagnostic yield, but it also doubles the screening costs 23,24 . Recent analyses of the cost-effectiveness of qiFOBT were published by Dutch working groups 25,26 ; the optimal cut-off value was found to be 50 ng Hb/mL. It is notable that the identical value was selected for screening in the Czech Republic based on a pilot study with OC-Sensor 27 .…”