The triple effect of colorectal cancer screening: reducing deaths, government spending, and mortality disparities

Abstract

Colorectal cancer (CRC) screening accounts for over 60% of cancer screening costs in the United States, prompting recurrent debates about its value. Yet CRC screening remains the main tool to curb overall CRC incidence, mortality, and disparities that affect Black Americans. Using the race-specific CRC-SPIN microsimulation model, we show that CRC screening in the United States simultaneously achieves three goals: it saves lives by preventing 24 deaths per 1000 Black Americans screened with the fecal immunochemical test (FIT) and 26 screened with colonoscopy; saves tax dollars by shifting costs from Medicare to private payers; and reduces racial incidence and mortality disparities, helping offset disparities in CRC survival. Both FIT and colonoscopy screening are cost-effective relative to no screening, with annual FIT remaining the most cost-effective option. Changes to policy requiring coverage of preventive care services must avoid compromising the effectiveness of CRC screening—arguably the greatest equalizer of cancer disparities.

Colorectal cancer (CRC) screening requires a major healthcare expenditure in the United States (US), accounting for 64% of the estimated $43 billion spent annually on cancer screening, with colonoscopy procedures alone constituting more than half of these costs.¹ As healthcare systems face increasing pressure to improve spending efficiency, CRC screening programs may become targets for cost-minimization efforts.

Yet healthcare efficiency goes beyond cost minimization. Efficiency in healthcare means saving the maximum amount of life with the least resources—selecting interventions with good value for money and reducing preventable mortality disparities. The burden of CRC is not evenly distributed, with Black Americans still bearing the second-highest CRC incidence and mortality among race and ethnicities recognized in the US.² Whereas prior studies have shown that CRC screening can save lives and is cost-effective from a societal perspective,^3-5 even when adherence to screening is low,⁶ there has been less emphasis on the distributional effects of CRC screening (i.e. who pays for it and who benefits the most). Modeling studies have estimated the impacts of improvements in screening quality and adherence on CRC incidence and mortality among Black Americans,^7,8 yet there is little recognition that mitigating racial/ethnic disparities in CRC mortality does not have to come at the expense of the taxpayer.

This paper employs the Colorectal Cancer Simulated Population model for Incidence and Natural History (CRC-SPIN) model⁹ to assess whether CRC screening can simultaneously achieve three seemingly incompatible outcomes: saving lives, saving taxpayer dollars, and narrowing CRC mortality disparities. CRC-SPIN is a validated microsimulation model^10,11 that simulates the natural history of CRC, including adenoma development, cancer progression, and mortality.

We simulate lifetime outcomes for multiple cohorts of 45-year-old, average-risk non-Hispanic White (hereafter, White) and 45-year-old average-risk non-Hispanic Black (hereafter, Black) Americans to illustrate how screening can mitigate existing disparities in CRC survival. We use the race-specific version of CRC-SPIN calibrated in Rutter et al.⁷ The model captures differences in cancer location distribution, with proximal colon cancer being more common in Black Americans and rectal cancer more prevalent in White Americans. Our economic analysis incorporates updated costs for screening tests estimated by Halpern et al.¹ Costs and utility weights were informed by the literature (Table S1). We estimate quality-adjusted life-years (QALYs) gained and CRC care costs, subdivided into screening (including follow-up colonoscopy after abnormal screening), surveillance colonoscopy after polypectomy,¹² and treatment costs. Following prior studies,^3,4,13 QALYs gained and costs were discounted at 3% per year. Our analysis adopts a societal perspective. We calculated the net monetary benefit ($\mathit{NMB}=\lambda \times \mathit{QALYs} \mathit{gained}-\mathit{Net} \mathit{Costs}$) for each screening strategy compared to no screening, allowing for direct comparison of their absolute value and their Incremental Cost Effectiveness Ratio (ICER =$\mathrm{ }\mathrm{\Delta }\mathit{Costs}/\mathrm{\Delta }\mathit{QALYs} \mathit{gained})$for assessing their incremental value. The net monetary benefit calculations assume a willingness to pay of$\mathrm{ }\lambda =\mathrm{\$}\mathrm{100\hspace{0.17em}} 000\mathrm{ }$per$\mathrm{ }\mathrm{QALY} \mathit{gained}$, which is a threshold in the middle of the $50 000-$150 000 range widely used in cost-effectiveness studies.¹⁴

We simulated three screening strategies: no screening, annual fecal immunochemical test (FIT), or decennial colonoscopy from age 45 to 75 (the two most used tests in the US as of 2021).¹⁵ Our target estimands are the benefits (QALYs gained) and costs of screening for individuals adherent to US screening¹⁶ and surveillance¹² guidelines. These screening modalities were selected as they represent the most common screening approaches in the US and are recommended in current clinical guidelines.¹

Following Rutter et al.,⁷ we examined the sensitivity of screening benefits and cost to colonoscopy quality and CRC survival. For colonoscopy quality, we contrasted two scenarios: the high-quality colonoscopy scenario considered high adenoma sensitivity (0.75 for 1-5 mm adenomas, 0.85 for 6-9 mm adenomas, and 0.95 for adenomas ≥10 mm and preclinical cancers) and better reach representing better preparation (95% reaching the cecum) as assumed in analyses that informed USPSTF CRC screening guidelines; ¹⁷ and lower-quality colonoscopy with reduced detection rates (0.55 for 1-5 mm adenomas and 0.70 for 6-9 mm adenomas) and lower cecal intubation rates (85%).⁷ To assess the impact of treatment disparities, the Equal CRC Survival scenario eliminates the estimated Black-White gap in stage-specific CRC survival⁷ by applying the same survival rates to both populations, whereas our Base-Case scenario maintains those disparities.

In our base-case scenario, the unscreened Black cohort is estimated to sustain 31.38 deaths per 1000 population (Table 1). Annual FIT screening could avert 24.18 of those deaths, yielding 0.13 QALYs gained per person, with a net cost of 1170 dollars per person relative to no screening, providing good value for money (Incremental Cost Effectiveness Ratio [ICER] = $8838 additional dollars spent per QALY gained). Decennial colonoscopy could avert 26.08 CRC deaths per 1000 population, yielding 0.15 QALYs gained per person at a net cost of 4633 per person relative to no screening. When compared against commonly used cost-effectiveness thresholds of $50 000-$150 000 per QALY gained, both screening strategies would be cost-effective relative to no screening, with FIT demonstrating particularly favorable cost-effectiveness. The incremental value of decennial colonoscopy from age 45 to 75 relative to annual FIT was small (0.02 QALYs gained requiring an additional cost of $3463 per person, implying an ICER of $213 552 per QALY gained). Results are similar for the White population (Table S2).

Table 1.

Lifetime CRC outcomes for non-Hispanic Black individuals.

Scenario	Modality	CRC deaths per 1000a	QALYsb	Treatment costsc	Screening costsd	Surveillance costse	Total costsf	Net monetary benefitg	ICERi
Base-case	No screening	31.38	20.41	5214	0	0	5214	Reference	Reference
	FIT	7.20 (−24.18)	20.54 (+0.13)	2018 (−3196)	2915	1451	6383 (+1170)	12 065h	8838
	Colonoscopy	5.30 (−26.08)	20.55 (+0.15)	1322 (−3892)	6375	2149	9847 (+4633)	10 224h	213 552
High-quality COL	No screening	31.38 (+0.00)	20.41 (+0.00)	5214 (+0)	0	0	5214 (+0)	0h	Reference
	FIT	4.56 (−26.81)	20.55 (+0.15)	1492 (−3722)	2898	1583	5972 (+758)	13 938h	5159
	Colonoscopy	2.62 (−28.76)	20.57 (+0.16)	789 (−4425)	6292	2377	9458 (+4244)	12 121h	208 897
Equal CRC survival	No screening	27.90 (−3.48)	20.42 (+0.01)	5370 (+157)	0	0	5370 (+157)	985h	Reference
	FIT	6.13 (−25.25)	20.54 (+0.13)	2043 (−3171)	2916	1456	6415 (+1201)	12 293h	8454
	Colonoscopy	4.52 (−26.85)	20.56 (+0.15)	1342 (−3872)	6374	2152	9868 (+4655)	10301h	236 295
High-quality COL & equal CRC survival	No screening	27.90 (−3.48)	20.42 (+0.01)	5370 (+157)	0	0	5370 (+157)	985h	Reference
	FIT	3.83 (−27.55)	20.55 (+0.15)	1507 (−3707)	2900	1587	5994 (+780)	14 102h	4535
	Colonoscopy	2.20 (−29.18)	20.57 (+0.16)	797 (−4417)	6291	2380	9469 (+4255)	12 080h	239 209

All outcomes except CRC deaths are presented per person and discounted at 3% per year.

^aCRC deaths per 1000 people, with difference from baseline (no screening) in parentheses.

^bQuality-adjusted life years per person, with difference from baseline in parentheses.

^cTreatment-related costs per person, with difference from baseline in parentheses.

^dScreening-related costs per person. Includes follow-up colonoscopy costs for screening with FIT.

^eSurveillance costs per person.

^fTotal costs per person, with difference from baseline in parentheses.

^gValue of screening compared to baseline using $100 000/QALY threshold.

^hValue relative to base-case no screening scenario.

ⁱIncremental costs per QALYs gained relative to preceding scenario.

In the high-quality colonoscopy scenario, which corresponds to higher adenoma detection rates and lower interval CRC rates,¹⁸ screening outcomes improve for both screening modalities. FIT yielded 0.15 QALYs gained instead of 0.13, and colonoscopy yielded 0.16 QALYs gained instead of 0.15. If disparities in CRC survival were eliminated and Black population CRC survival improved to match that estimated for the White population (equal CRC survival scenario, Table 1), outcomes for the unscreened population would improve, reducing 3.48 CRC deaths per 1000 population. Screened cohorts would achieve smaller improvements because screening’s preventive effect already mitigates mortality.

Finally, when high-quality colonoscopy is combined with equal CRC survival, health outcomes improve the most—but both screening modalities with a high-quality colonoscopy already provided most of the benefits. For example, FIT with a high-quality colonoscopy averted 25.25 CRC deaths per 1000 individuals, whereas FIT with a high-quality colonoscopy and equal CRC survival averted 27.55 CRC deaths per 1000 individuals. This finding highlights CRC screening’s capability to mitigate CRC treatment disparities through prevention even when tests only partially act at the “root” causes of disparities by preventing cancer onset and detecting cancer at earlier stages. Hence, screening among populations that face poor CRC survival outcomes helps mitigate incidence and survival disparities.

CRC screening also saves tax dollars. While our previous results (Table 1) present aggregate lifetime cost estimates, examining when costs accrue shows that CRC screening effectively generates cost savings to Medicare, as costs are primarily borne by private insurance and ultimately individuals (Figure 1). This contrasts markedly with unscreened cohorts, where treatment costs accumulate predominantly after age 65, when Medicare is primarily responsible for costs. Healthcare costs increase with screening over the lifetime of individuals (Figure 1), but the shift to screening is not a burden to Medicare, and the projected benefits of screening outweigh the projected costs that individuals pay through insurance (evidenced by positive net monetary benefit estimates in Table 1). Those costs include the costs of overdiagnosis due to removal of lesions that would never have progressed to cancer, over-referral to colonoscopy surveillance, and diagnosis of cancers that would not have become symptomatic within an individual’s lifetime. However, these analyses do not account for potential increases in the ability of gastroenterologists to detect and act on diminutive (<5 mm) lesions, which could increase overdiagnosis and overuse of surveillance colonoscopy. Still, a shift away from screening would pose a great burden to Medicare. Hence, an effort to reduce government spending should suggest increasing CRC screening participation—not the opposite.

Figure 1.

CRC costs per person by screening modality and age. This figure displays discounted lifetime costs per person across three screening strategies (no screening, annual FIT, and decennial colonoscopy), separated by when costs are incurred (before age 65 vs. at/after age 65). All costs are discounted at 3% annually from age 45, representing present value. Costs are categorized into three components: screening, surveillance, and treatment. For the FIT modality, the screening component includes the costs of follow-up colonoscopies. The left panel shows costs typically covered by private insurance, while the right panel shows costs typically covered by Medicare. Both screening strategies shift costs from later years (Medicare) to earlier years (private insurance), with colonoscopy requiring a larger upfront investment but achieving a greater reduction in later treatment costs compared to annual FIT. Although both FIT and colonoscopy result in higher costs before age 65, both are cost-effective relative to no screening because of the value of QALYs gained, which is not shown in this graph but is shown in Table 1.

These results demonstrate that CRC screening can achieve three goals simultaneously. First, CRC screening saves lives through two mechanisms: finding precursor lesions and removing them before they become cancer, and detecting cancer at earlier, more treatable stages. Second, CRC screening saves tax dollars by shifting costs from Medicare to individual and private insurance payers. Finally, CRC screening helps reduce long-standing disparities by preventing deaths among populations that may otherwise have a worse chance of survival, such as Black Americans. Although this analysis investigates the effects on Black Americans, who bear the second-highest CRC mortality in the US,¹⁹ this point holds broadly to other groups that may face the poorest CRC survival outcomes.

The ability of CRC screening to deliver this triple benefit—reducing mortality, generating taxpayer savings, and narrowing health disparities—is a remarkable public health achievement that few interventions can match. While we do not suggest that only interventions achieving all three outcomes deserve public funding, the exceptional value proposition of CRC screening demands sustained policy support. As healthcare systems continue to face cost pressures, our findings strongly caution against viewing CRC screening programs as targets for cost reduction. Instead, policymakers should recognize CRC screening as an exemplary case of healthcare efficiency.

Given these benefits, healthcare policy should prioritize not only maintaining but also expanding CRC screening adherence, with particular emphasis on reaching populations that stand to benefit most (i.e. uninsured individuals). Policies currently requiring insurers to cover preventive care services (42 U.S.C. § 300gg-13, 2010),^20,21 with no out-of-pocket patient costs, safeguard the quality and accessibility of CRC screening—arguably the most powerful equalizer in cancer care—while delivering value to individuals and society.

Author contributions

Pedro Nascimento de Lima (Conceptualization, Formal analysis, Investigation, Writing—original draft, Writing—review & editing), Lilian Bartholomew (Data curation, Writing—review & editing), Folasade P. May (Writing—review & editing), Gloria Coronado (Writing—review & editing), and Carolyn M. Rutter (Funding acquisition, Supervision, Writing—review & editing)

Supplementary material

Supplementary material is available at JNCI: Journal of the National Cancer Institute online.

Funding

This research was made possible with support from grant R01-MD017599 from the National Institute on Minority Health and Health Disparities. This project was also supported in part by grant U01-CA253913 from the National Cancer Institute as part of the Cancer Intervention and Surveillance Modeling Network (CISNET). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the National Institute on Minority Health and Health Disparities or the National Cancer Institute.

Conflicts of interest

Dr May reported serving on scientific advisory boards for Exact Sciences, Geneoscopy, Natera and Medtronic. No other disclosures were reported.

Data availability

Detailed documentation of the CISNET colorectal cancer models used in this study is publicly available at https://cisnet.cancer.gov/colorectal/. The code of CISNET models is not open source.