Colorectal cancer is the third most common cancer and the third leading cause of cancer death in the United States. In 2010, there were approximately 140,000 new cases and more than 50,000 deaths due to colorectal cancer.1 Although screening colonoscopy, recommended for adults age 50 or older at average risk, can reduce the incidence of cancer through removal of precancerous polyps and improve outcomes through earlier tumor detection,2, 3 this disease continues to pose a substantial public health burden.
In 2003, the Commissioner of the New York City Department of Health and Mental Hygiene (NYC DOHMH) launched a colorectal cancer screening initiative, which included a colonoscopy patient navigator program, within selected public hospitals. Patient navigators are trained to guide individuals through complex clinical settings in order to assist with the scheduling, preparation, and completion of the procedure. Commonly cited barriers addressed by navigators include confusion and fear about the procedure and bowel preparation, and logistical challenges such as lack of health insurance, difficulty scheduling, and finding an escort home.4 For members of ethnic minority groups, patient navigators also assist in overcoming cultural and linguistic obstacles.5
Initial evaluations of the NYC DOHMH colonoscopy patient navigator program and a similar program implemented at 2 tertiary academic medical centers in New York City suggest that these programs can be effective in increasing the number of screening colonoscopies performed, increasing the probability of colonoscopy completion, increasing the number of precancerous colorectal polyps removed, and reducing the number of missed appointments for colonoscopy.6-8 Our objective was to evaluate the economic impact of the patient navigator program, specifically the cost, cost-effectiveness, and net financial benefit associated with the program implemented at 3 public hospitals.
- Top of page
- MATERIALS AND METHODS
- FUNDING SOURCES
In the urban public hospital settings evaluated in this study, a colonoscopy patient navigator program was associated with increases in colonoscopy volume and in the likelihood of colonoscopy completion. The cost-effectiveness of the program varied from about $200 to $700 per additional colonoscopy, with variation between sites primarily attributable to differences in the time and salaries of program personnel. The program was associated with a net monetary benefit of about $45 per additional colonoscopy at the lowest cost site and a net cost of $464 per additional colonoscopy at the highest cost site.
Of the 3 program sites included in this evaluation, the site with the highest costs (Hospital B) was the first to implement the patient navigator program. This site's substantially greater personnel costs were likely related to its pioneering role in the program, with more staff and higher paid staff initially involved in program planning and administration. All of these personnel might not have been necessary for ongoing program operation, but the hospital chose to maintain program staffing at its original level. The staffing arrangements at the 2 other sites suggest that a similar impact on colonoscopy volume and appointment completion can be achieved at a lower cost. Program costs were lower at Hospital A and Hospital C, and the program was sufficiently effective to yield a net financial benefit.
In the cost-benefit analysis presented here, conducted from the perspective of the hospital, net financial outcomes were a function of 3 factors: navigator program cost, reimbursement associated with increased colonoscopy volume, and the cost of performing these additional procedures. We did not explicitly consider the economic value of reducing missed colonoscopy appointments. Instead, the analysis implicitly assumed that endoscopy suites and personnel did not sit idle when a scheduled patient failed to show. If resources cannot be immediately reassigned when appointments are missed, then our analysis may underestimate the economic benefit of the program, because an increase in the colonoscopy completion rate would be associated with financial savings from a reduction in missed appointments.
Despite enthusiasm for patient navigation programs, particularly those that facilitate cancer screening, diagnosis, and treatment, evidence of their efficacy and cost-effectiveness is relatively limited. A recent review identified 16 studies that reported outcomes of cancer-related patient navigation programs.10 Six of the studies involved programs designed to improve cancer screening rates, and 3 studies specifically addressed colorectal cancer screening.8, 11, 12 Although patient navigation was associated with an increase in colonoscopy completion rates, the review authors noted that methodologic limitations precluded a definitive determination of program efficacy. More importantly, of all 16 studies reviewed, none included an analysis of program cost-effectiveness.
A more recent study found an increase in colorectal cancer screening rates, including colonoscopy specifically, among low-income adults who were randomized to a patient navigator program in an urban primary care setting in Massachusetts.13 Although that program was not identical to the patient navigator program evaluated here, based on the authors' projections of program cost ($75,000 per year) and the difference in colonoscopy use associated with the program in a 9-month period (20.8% of 409 adults randomized to the intervention versus 9.6% of 814 controls), the cost-effectiveness of the Massachusetts patient navigator program would be more than $8000 per additional patient screened with colonoscopy.
The scope of economic evaluation in our study was limited to the proximal endpoint of cost per additional colonoscopy. Ultimately, the benefit of increased screening derives from its impact on the more distant outcomes of morbidity and mortality due to colorectal cancer. Several investigators have used decision-analytic methods to estimate the long-term outcomes of colon cancer screening.14 Five studies simulating the experience of adults aged 50 to 85 years who were at average risk of colorectal cancer found that screening colonoscopy every 10 years was associated with a cost of $10,000 to $20,000 per life-year gained (given in 2000 US dollars).15-19 In these models, adherence with screening varied from 50% to 100%. By increasing colonoscopy volume and completion rates, the patient navigator program evaluated here can be viewed as increasing adherence with screening recommendations in the population served by the program. One decision-analytic simulation showed that enhancing the rate of colonoscopy screening adherence from 50% to 80% increased the average number of life-years gained by more than 30%, and increasing the screening adherence rate from 50% to 100% increased life-years gained by more than 60%.20 Thus, by increasing adherence, colonoscopy patient navigation likely has a beneficial impact on long-term health outcomes.
Following standard methods of cost-effectiveness analysis, we did not analyze the financial cost of lost productivity associated with colorectal cancer.21 However, the economic value of reducing productivity losses due to morbidity and premature mortality may be quite substantial. One recent estimate suggests that every death due to colorectal cancer is associated with more than $210,000 in lost productivity.22 Of the $142 billion projected value of lost productivity due to all cancer deaths in 2010, colorectal cancer deaths account for $12 billion or 9% of these economic losses. Colorectal cancer mortality is second only to lung cancer mortality in contributing to the value of lost productivity associated with all cancer deaths.
Limitations of our study should be noted. The nonrandomized design of the program evaluation yielded measures of program effectiveness, rather than efficacy. Although program hospitals and comparison hospitals served similar populations, and we used multivariable statistical methods to control for selection bias, there may be some residual confounding in our estimates, although the likely direction of such bias is not obvious. The racial and ethnic composition of colonoscopy patients at program and comparison hospitals was not identical, and colonoscopy patients at program hospitals came from neighborhoods with more high school graduates. Both groups of hospitals were located in New York City, limiting potential bias due to local secular trends in cancer screening. However, the generalizability of our findings to nonpublic hospitals and hospitals in other cities and regions is uncertain. It is notable that the program was both effective and cost-effective in an urban public hospital system serving a population composed predominantly of low-income racial and ethnic minorities, among whom the burdens of colorectal cancer morbidity and mortality are greater compared with their white peers.23 Finally, navigator program and colonoscopy costs and reimbursement amounts are from 2006. Although adjustment for inflation would increase the absolute values of these items if reported in more recent year dollars, the relative values of these items would likely be similar. The relative input prices and technologies associated with patient navigation and screening colonoscopy have not changed substantially in the past 6 years.
The colonoscopy patient navigator program studied here was effective in increasing colonoscopy volume and rates of colonoscopy completion, at a cost of $200 to $700 per additional colonoscopy. From the perspective of a hospital or other provider, the return on investment in a colonoscopy patient navigator program depends on several factors, most importantly the program cost, procedure cost, and reimbursement. If colonoscopy reimbursement exceeds the procedure cost by an amount greater than the cost of the navigator program per additional colonoscopy (ie, the cost-effectiveness ratio), then the program generates a net monetary benefit. In this case, net revenue associated with each additional colonoscopy exceeds the cost to achieve that additional colonoscopy. Alternatively, if colonoscopy reimbursement exceeds procedure cost, but by an amount less than program cost per additional colonoscopy, then the navigator program generates a net monetary cost. Even if the navigator program is effective and cost-effective relative to other similar interventions, if procedure cost exceeds reimbursement, then the provider loses money on every colonoscopy.
Colorectal cancer screening rates have increased in the past decade, both nationwide and in New York City. By 2007, 62% of eligible New York City residents had received a colonoscopy, compared with 42% in 2003,24 and similar trends have been observed across the country.25 Despite these advances, nationally at least two-thirds of Americans aged 50 years and older have not had a screening colonoscopy.26 As the population ages, the number of adults eligible for colonoscopy will grow substantially. Patient navigation programs, in conjunction with other efforts, may improve colonoscopy rates. Our results suggest that such programs are economically feasible and can yield a financial benefit to providers.