Skip to main content
. Author manuscript; available in PMC: 2013 May 6.
Published in final edited form as: Ann Intern Med. 2012 Nov 6;157(9):611–620. doi: 10.7326/0003-4819-157-9-201211060-00005

Table 1.

Screening test characteristics and costs used in the analyses.

Sensitivity* for adenomas by size and for CRC (%)
Analysis/screening test Small (1–5mm) Medium (6–9mm) Large (≥10mm) CRC Specificity (%) Source Test cost, $ Cost description
Base-case analysis
 HSFOBT 7 12 24 70 93 (40) 23 2007 national average Medicare reimbursement for fecal occult blood assay (HCPCS G0394) adjusted to 2010 dollars, plus cost of one hour of patient time
 FIT 5 10 22 70 95 (40) 46 2007 national average Medicare reimbursement and beneficiary copayment for immunoassay-based fecal occult blood test (HCPCS G0328) adjusted to 2010 dollars, plus cost of one hour of patient time
 Colonoscopy 75 85 95 95 84 (39,41) 1,153 without polypectomy Without polypectomy: weighted average of 2007 national average Medicare payments and beneficiary copayments for diagnostic colonoscopy (CPT 45378), colon screen in high-risk individual (CPT G0105), and colon cancer screening for non-high-risk individual (CPT G0121) (40) adjusted to 2010 dollars, plus the costs of colonic preparation** and 24 hours of patient/escort time
1,347 with polypectomy With polypectomy: weighted average of 2007 national average Medicare payments and beneficiary copayments for colonoscopy and biopsy (CPT 45380), colonoscopy with submucosal injection (CPT 45381), colonoscopy/control bleeding (CPT 45382), lesion removal colonoscopy — fulguration (CPT 45383), lesion removal colonoscopy-hot biopsy (CPT 45384), and lesion removal colonoscopy-snare polypectomy (CPT 45385) (40) adjusted to 2010 dollars, plus the costs of colonic preparation** and 24 hours of patient/escort time
 CTC -- 57 84 84§ 88 (42) 530 2010 national average Medicare reimbursement†† and beneficiary copayment for a diagnostic CTC (CPT 74261), plus the costs of colonic preparation** and eleven hours of patient time
Sensitivity analysis ‡‡
 Colonoscopy§§ 60 68 76 76 84 assumption 577/770 to 5,765/5,959 0.5–5 times the base-case estimate of colonoscopy without polypectomy, plus the incremental cost of polypectomy
 CTC -- 84 92 95 80 (4344) 530 Cost estimate was not varied from the base-case value

CPT = Current Procedural Terminology code; CRC = colorectal cancer; CTC = computed tomographic colonography with ≥6mm threshold for colonoscopy referral; HSFOBT = highly-sensitive guaiac fecal occult blood test; HCPCS = Healthcare Common Procedure Coding System code; FIT = fecal immunochemical test, -- indicates sensitivity is not provided because size is smaller than the referral threshold for a colonoscopy.

*

Sensitivity is provided per individual for HSFOBT and FIT and per lesion for colonoscopy and CTC.

The lack of specificity with colonoscopy reflects the detection of non-adenomatous lesions. Non-adenomatous lesions induce polypectomy and biopsy costs.

Sensitivity for CTC for medium adenomas was calculated from published tables (42).

§

Sensitivity for CRC was assumed to be the same as for large adenomas.

The lack of specificity with CTC reflects detection of non-adenomatous lesions, artifacts, and adenomas smaller than the lesion size threshold for referral to colonoscopy of 6mm

The value of an hour of patient and/or caregiver time was assumed to equal the 2010 US median hourly wage rate for the civilian population, $18(45). The amounts of patient and escort time assumed for each test are detailed in Appendix Table 2.

**

Estimated at $23 (2010 average wholesale price of GoLYTELY® (46)).

††

With implementation of the Out-patient Prospective Payment cap on the technical component of imaging procedures (47).

‡‡

The sensitivity analysis on colonoscopy test characteristics was performed separately from the sensitivity analysis on colonoscopy cost.

§§

Assuming a 20% reduction from the base-case values. To account for the possibility that colonoscopy may not be as protective against right-sided disease, in the sensitivity analysis on colonoscopy test characteristics, we assumed that only 80% of colonoscopies are compete to the cecum (vs. 95% in the base-case analysis)