Higher Quality Colonoscopy: Worth the Wait?

ABSTRACT: To meet the high demand for colonoscopy, the Veterans Health Administration often refers veterans to community practices. Petros et al. compared colonoscopy quality at a Veterans Affairs Medical Center with that of local community practices. Although community providers performed more timely colonoscopy than the Veterans Affairs medical center (mean 25 days earlier), colonoscopy quality was significantly lower for other important quality metrics, including adenoma detection, advanced adenoma detection, adenomas per colonoscopy, and appropriate surveillance recommendations. This study highlights the need for continued efforts to assure high-quality colonoscopy in all settings.

Despite being the largest integrated healthcare system in the United States, the Veterans Health Administration (VA) relies on community partnerships to meet the needs of veterans, especially for those who do not live near VA facilities and for services that may not be readily available because of high demand, such as colonoscopy. To provide veterans with more healthcare options, the Maintaining Internal Systems and Strengthening Integrated Outside Networks Act (1) made non-VA community care available to veterans largely based on wait-time and/or travel-time eligibility criteria. While timely access to care is an important aspect of quality, the Institute of Medicine also includes 5 other dimensions of quality: safety, effectiveness, patient-centeredness, efficiency, and equity (2). Over the past 2 decades, there has been increasing emphasis on colonoscopy quality metrics because of the demonstrated association between these measures and important patient outcomes, especially related to colorectal cancer (CRC). For example, a provider's adenoma detection rate (ADR) has been strongly linked to colonoscopy's effectiveness at reducing future CRC incidence and mortality (3–5).

In a recent issue of CTG, Petros et al. (6) compared the quality of colonoscopy that veterans receive in the VA vs in the community on several quality domains. The authors identified 235 veterans receiving community care colonoscopy (CCC) who were then age-matched to 235 control subjects with a VA colonoscopy (VAC) at the Oklahoma City VA Medical Center. Screening indications were comparable for CCC and VAC; surveillance indications were 9% higher for VAC while abnormal FIT indications were 10% higher for CCC. Although CCC was performed earlier than VAC (58 vs 84 days, P < 0.0001), other aspects of quality were significantly worse for CCC. Specifically, CCC was associated with a lower ADR than VAC (37% vs 63%, P < 0.0001), lower advanced adenoma detection (8.9% vs 18.3%, P = 0.003), fewer overall adenomas detected per colonoscopy (0.77 vs 1.83, P < 0.0001), and worse adherence with surveillance guidelines (74.9% vs 93.3%, P < 0.0001). When adjusting for potential confounders, compared with VAC, CCC had an odds ratio (OR) for adenoma detection of 0.39 (95% confidence interval 0.20–0.63) and the OR for appropriate surveillance recommendations was 0.21 (95% confidence interval 0.09–0.45).

Although the CCC ADR exceeds the 30% benchmark for men (7), a landmark study from Kaiser Permanente found that each 1% increase in ADR was associated with an absolute 3% decrease in CRC incidence and 5% decrease in CRC death, without a ceiling effect for benefit (3). Thus, veterans receiving CCC may experience higher rates of postcolonoscopy CRC incidence and mortality than those receiving VAC, although further study is needed. Appropriate colonoscopy surveillance recommendations fall into the quality dimensions of efficiency, safety, and patient-centeredness because overuse of surveillance is inefficient (for both patients and the healthcare system) and exposes patients to the risk, inconvenience, and cost of low-value procedures while underuse (i.e., inappropriately long surveillance intervals) poses an increased risk of interval cancer. When evaluated against contemporaneous guidelines (8), CCC was more frequently associated with surveillance recommendations that would be classified as both overuse (15% vs 6%, P = 0.001) and underuse (10% vs 1%, P = 0.0001) than VAC. Although these findings are from one center and may not be generalizable to the entire VA, they do build on previous studies from other VA facilities that also raise concerns about the quality of colonoscopy in the community (9–11).

Given the impact of CRC on the population and the importance of high-quality colonoscopy in improving patient outcomes, colonoscopy quality assurance is a key priority of the VA National Gastroenterology and Hepatology Program (NGHP). Over the past 3 years, over 645,000 veterans underwent VAC and another 219,000 veterans had CCC. With approximately 1000 employed colonoscopy providers, the VA is not immune to quality issues (12,13). Therefore, the NGHP has taken several steps to assure high-quality colonoscopy with coordinated efforts for transparent measurement, accountability, and improvement. The VA established requirements for colonoscopy quality monitoring at all VA facilities in 2014, with stronger requirements published in 2020 (14). In 2021, the VA updated requirements for colonoscopy quality monitoring as part of the ongoing professional practice evaluation required by the Joint Commission (15). While this monitoring is currently performed locally, the NGHP is working to build a national quality assurance system, including interventions at the facility and provider level for low performance (16). However, these efforts do not apply to CCC and obtaining non-VA provider-level quality data is difficult, if not impossible, because each provider only performs a relatively small number of colonoscopies on veterans.

The results of this paper raise the following question: Is the VA disproportionately purchasing lower quality colonoscopies or is there something about the non-VA setting that promotes lower quality colonoscopy? There is currently no easy way for the VA to assure that they are contracting for care only from providers who perform high-quality colonoscopy and reimbursement is based on a fee-for-service model rather than a value-based care model. This is a significant dilemma because the VA currently lacks capacity to provide colonoscopy to all veterans in need, and VAC is not available in all regions where veterans live. Although CCC was associated with an average 25-day decrease in the time to colonoscopy and offering CCC is required when the VA lacks access within 28 days of the date of request, adverse outcomes after abnormal fecal occult blood testing have been demonstrated to be associated with wait times exceeding 9 months, far longer than VA policy expectations (17). Moreover, it has been found that community care records are often not available in the VA (10), thereby limiting care coordination and planning for future surveillance. Unfortunately, the MISSION Act does not address other dimensions of quality beyond access. Ideally, the VA would improve access to colonoscopy (e.g., through improved efficiency and/or building capacity) such that wait times and reliance on community care could be reduced because this would both allow for quality assurance and facilitate care coordination.

The results of this single-center study have implications far beyond the VA because this highlights the question that all patients and purchasers of care face today: How does one know whether they can expect to receive high-quality care? Although we acknowledge the complexities inherent in colonoscopy quality measurement and reporting, we are also cognizant of its tremendous value. Much progress has been made, but there is still much work to be performed by all stakeholders (e.g., physicians, payers, electronic health record vendors, and professional societies) to enhance colonoscopy quality assurance efforts, such as audits (including the use of artificial intelligence), data reporting requirements, and reimbursement models that reward high-quality care. Ultimately, all patients deserve access to high-quality colonoscopy for optimal CRC prevention.

CONFLICTS OF INTEREST

Guarantor of the article: Jason A. Dominitz, MD, MHS.

Specific author contributions: A.J.G.: drafting of the manuscript. J.A.D.: drafting of the manuscript.

Financial support: This manuscript is the result of work supported in part by resources from the Veterans Health Administration.

Potential competing interests: A.J.G. and J.A.D. are employees of the Veterans Health Administration. The content is solely the responsibility of the authors and does not represent the views of the Department of Veterans Affairs or the United States Government.