Cost-Effectiveness of a Patient Navigation Program for Colorectal Cancer Screening

05/12/17
Issue
Citation

Journal of Clinical Pathways. 2017;3(4):31-35.

Affiliation

1University of Colorado School of Medicine, Department of Family Medicine, Aurora, CO

2Colorado School of Public Health, Aurora, CO

3Colorado School of Public Health and University of Colorado Cancer Center, Aurora, CO

4University of Colorado School of Medicine, Division of Health Care Policy and Research, Aurora, CO.

Correspondence

Dr Whitney C Rhodes

University of Colorado School of Medicine

12631 East 17th Avenue, Room 350

Aurora, CO 80045

Phone: (303) 724-8975

Email: whitney.rhodes@ucdenver.edu

Disclosures

The authors report no relevant financial relationships.

Key Words

Acknowledgment: This work was supported by a contract (no. 13 FLA 45130) from the Colorado Department of Public Health and Environment. The authors would like to thank Holly Wolf, PhD and Shannon Pray, MSPH for their contributions to the manuscripts.

Abstract: Patient navigators (PNs) support colorectal screening through increased attendance and improved bowel preparation and are now frequently embedded in primary care settings. In order to determine whether colonoscopy providers might benefit economically from paying for PN services, we compared navigation costs with gains in appointment attendance and bowel preparedness. Using data from five community clinics receiving flat-rate state reimbursement for colonoscopy PNs, we conducted a break-even analysis using a microcosting framework. Payment offset direct costs of PN services in the highest-volume clinic. Variance was observed in cost per patient, with cost related to screening volume. PNs may benefit providers by decreasing missed screening opportunities.

Received March 20, 2017; accepted May 3, 2017

-----

Patient navigation has been used to address disparities in colorectal screening and has been shown to successfully increase screening rates and adequacy of preparation for the procedure.1-6 Patient navigators (PNs) may deliver patient education before and after the procedure, answer questions and concerns, and provide reminder calls and support for patients to attend scheduled appointments.1

Colorectal cancer is the second-leading cause of cancer-related mortality in the United States,7 although it is preventable through early detection and removal of adenomas.8 Colorectal screening is widely accepted as effective in reducing colorectal cancer incidence and mortality9,10 and is cost-effective when compared with no screening.8,10,11 However, disparities in colorectal cancer screening, incidence, and mortality persist in underserved populations, including among racial and ethnic minorities, individuals in low-income households, individuals with lower education levels, and the uninsured.12,13

Primary care settings increasingly use PNs in colorectal screening,1-4 but PN services are currently ineligible for insurance reimbursement, and thus must be funded by grants or become part of a facility’s or system’s general operating expenses. As a result, most studies of PN costs have focused on financial returns to hospitals or health systems2,14 and net costs of health care.15 In contrast, the present study examines the economic impact to the screening provider as a potential means for sustaining PN services. When PNs decrease no-show rates,2,3 the economic benefits accrue not only to the facility, but also to screening providers by reducing idle, unbillable time attributable to missed appointments.

The current evaluation focused on the Colorado Colorectal Screening Program (CCSP), a grant-funded program that reimbursed 52 community health clinics for providing PN services to low-income patients, the specifics of which have previously been described.16 Under the CCSP model, PNs were employed by the community primary care clinics, while screening procedures were delivered in unaffiliated facilities. The program’s grant funds ended in 2015, and the current evaluation explored a potential approach to sustainability, comparing PN service costs with projected provider revenue gains attributable to PNs through increased appointment attendance and improved bowel prep quality to ensure full visualization of the colon to reach the cecum. The evaluation question concerned whether the total projected gain in provider revenues would be comparable to total PN service costs, which might give screening providers an economic incentive to pay for PN costs on a fee-per-procedure basis. Sustainability of PN services for colonoscopy has previously received only limited attention.8,17

Methods

Study Setting and Recruitment

CCSP program staff identified a purposive sample of 10 community clinics based on likely availability of necessary cost data from each clinic. The sample included urban and rural clinics, with a wide range of patient volumes.
To study economies of scale, the sample was stratified by volume using a natural breakpoint in mean number of completed screenings reported to CCSP in fiscal year 2013 and fiscal year 2014. Low-volume clinics had fewer than 45 patients, while high-volume clinics had more than 120 patients.

Data Collection

Cost data included five components: (1) PN staff costs; (2) supervisor and other program support staff costs; (3) PN hiring and training costs; (4) other direct program costs; and (5) indirect costs, including facility and administrative costs. Participating clinics provided these data in Microsoft Excel spreadsheets, using an instruction guide provided by the evaluation team. Clinics were instructed to list PN and staff time attributable to colonoscopy-related duties that took place between initial scheduling and procedure completion for a typical patient. PN staff costs included hourly wages/salaries, hours worked per week or percentage of a full-time equivalent for the specified activities, and fringe benefit rates. Similar information was requested for supervisors and other support staff. Hiring and training costs focused on PNs, because supervision and other support staff devoted only a small percentage of time to the program. PNs received standardized training from the CCSP program to ensure service uniformity. Other direct program costs included travel expenses for program delivery, materials, telephone charges, and office equipment used in delivering the program. Facility, administrative, and other indirect costs were provided as a percentage of direct program costs, or as direct costs of office space and other facilities dedicated to colonoscopy screening navigation.

Analyses

Two types of analyses were conducted: PN-related costs per completed screening colonoscopy, and break-even analysis for colonoscopy providers. Analysis of cost per screening used a microcosting framework that calculated programmatic cost by aggregating specific cost-estimates associated with distinct program elements. For each clinic, total reported costs were summed and divided by the number of completed colonoscopies for which the clinic received CCSP payment, resulting in an average cost per screen. An aggregated estimate across clinics was calculated by summing all participating clinics’ programmatic costs and dividing by the sum of the average number of screenings per clinic during the 2-year study period. The cost-per-screening analysis included a comparison of the study-estimated programmatic costs with the CCSP reimbursement rate for PN services.

The break-even analysis—a type of return-on-investment method—adopted the financial perspective of colonoscopy providers and focused on the provider’s lost revenue or opportunity cost attributable to patient no-shows or inadequate preparation for the procedure. Endoscopy and colonoscopy reports were used to verify exam completion. In 99% of cases when patients presented for their appointment, endoscopy reports indicated good or excellent bowel prep, and full visualization and ability to reach the cecum. As such, we assumed all attended appointments represented completed colonoscopies. The analysis thus focused on no-show rates, and compared rates with vs without PN services. For simplicity, we assumed colonoscopy costs (facility charge, anesthesiologist, nursing support, etc) were invariant and always paid by the provider, so a no-show appointment costs the provider the entire amount that would have been reimbursed for a completed colonoscopy. During the grant period, providers were paid regardless of whether a patient attended the procedure. The current analysis assumed providers would be paid only when a patient attends the appointment, which is a more realistic assumption for a sustainability analysis. The no-show rate with PN services was based on fiscal year 2013 CCSP data; comparison rates were taken from published reports for colonoscopy attendance among similar populations without PN services.18,19

Results

High-volume vs low-volume clinics reported annual screening means of 178 patients (range, 122-221 patients) and 43 patients (range, 13-101 patients), respectively.
Half of the sampled clinics agreed to participate and provided cost data, including two high-volume clinics (range, 122-181 patients) and three low-volume clinics (range, 25-42 patients).

The observed CCSP no-show rate was 9%, compared with published no-show rates without PN services of 42%16 and 38%.17

Program Costs

The average PN-related cost per CCSP-sponsored screening was $470 ($415 excluding indirect costs; Table 1). PN-related personnel costs (PNs, supervisors, and other support staff) accounted for 79% of direct programmatic costs, including 49% attributable to PNs and 30% attributable to other staff. Personnel costs included salary and fringe benefits. Weighted hourly rate for PNs was $20.37 including benefits, with PNs working an average of 14.8 hours per week on CCSP (range, 8-26 hours per week). Weighted hourly rate for supervisors was $45.36 including benefits, with supervisors working an average of 4.2 hours per week on CCSP (range, 2-6 hours per week). The weighted average hourly rate for support staff was $31.43 including benefits, with support staff working on average 1.25 hours per week. Other direct costs included training and travel, office equipment purchases for PNs, and program materials such as printed information and other supplies. Indirect costs included organizational overhead; each participating organization provided its overhead rate.

PN service costs varied nearly four-fold among study clinics and were inversely related to clinic volume. The lowest direct cost at the highest volume clinic was $280 per screen, while the highest direct cost at a low-volume clinic exceeded $1000 per screen. Average direct costs among low-volume clinics was $662 per screen, more than double the $316 average cost per screen among high-volume clinics.

The CCSP paid PN clinics $280 per completed screen, an amount equal to the lowest direct cost of PN services at the highest volume clinic. The program paid providers $966 per screen to cover professional services, facility fees, pathology services, anesthesia services, bowel prep kit, and population-averaged costs to treat adverse events and remove large polyps. 

Break-Even Analysis

In the hypothetical scenario where providers would pay for PN services, necessary reimbursement rates ranged from $779 to $1471 per completed colonoscopy to break-even; the reimbursement payment range depends on clinic-specific costs of PN services, and is based on missed appointment rates before PN services were introduced (Table 2). The payment per colonoscopy that was paid by CCSP ($966) would be more than adequate where the PN service fee matched the lowest observed direct cost of PN services ($280, excluding indirect costs). 

Providers paid at the CCSP rate ($966 per colonoscopy) would break-even if they paid between $309 and $347 per colonoscopy for PN services (Table 3). These PN service costs would exceed the CCSP payment for PN services ($280 per colonoscopy), and would cover the average observed cost per colonoscopy of PN services at the highest volume clinic.  

Discussion

An evaluation of PN support for screening colonoscopy found higher rates of completed procedures than published completion rates in the absence of navigator support. PN service costs per colonoscopy varied nearly four-fold among the five study clinics; the difference was inversely related to clinic colonoscopy volume. 

The break-even analysis suggests that it would be in the economic interest of colonoscopy providers to pay for direct PN costs at the rate that is paid by CCSP or at the actual cost of the highest-volume clinic. The higher the decrease in no-show rates and the lower the cost of the PN services, the more likely colonoscopy providers would be to break-even on paying for PN services. 

The cost of delivering PN services in primary care settings is driven primarily by personnel costs, including PN staff and supervisory and other support staff, with personnel costs accounting for 79% of direct cost on average for the clinics in this evaluation. The predominance of personnel costs is expected since the PN intervention is interpersonal, involving delivery of health messages and tailored reduction of barriers, and requires only office equipment and limited materials. 

We also observed considerable variance in cost per patient screened for the clinics in this evaluation, with highest costs among the low-volume clinics. A substantial driver of this variation is differences in personnel cost per completed colonoscopy, with estimated PN hours per completed colonoscopy ranging from 6 hours in the highest-volume clinic to almost 17 hours in the lowest-volume clinic. Similarly, supervisory and other support staff effort is more than three times higher per completed colonoscopy in low-volume clinics compared with high-volume clinics. This suggests that for the clinics in this evaluation, there may be substantial economies of scale in providing PN services to increase colorectal screening rates.

Although wages and hours per week for PNs were the lowest at the highest-cost, low-volume clinic, this lower cost was more than offset by the small patient volume, resulting in the higher completed colonoscopy cost for navigation services. Moreover, the number of hours of supervision and other support staff were relatively equal across all of the clinics, resulting in higher per completed colonoscopy cost for this component even though average wages for supervisors and other support staff were considerably lower in the rural area where this low-volume clinic is located.

Our finding that supervisor salaries and time can have a large impact on per-patient cost is consistent with other PN programs that have examined the cost of providing PN services, such as Elkin and colleagues, which examined cost from a health system perspective.4 One cost study that examined PN services for colorectal screening at three hospitals attributed substantial variance in programmatic cost to physician administrators and other supervisory personnel who contributed to the program in the highest-cost hospital, and only had minimal involvement in the other two hospitals.4 Although future studies could seek to determine optimal staffing models for clinics based on patient volume, the economies of scale appear to be sizable. This creates a significant challenge to sustaining this service in primary care settings under current fee-for-service payment models, particularly in the low-volume clinics. 

The evaluation results suggest that PN services are most likely to be sustainable in high-volume primary care practices. The results in the literature suggest that PN services provided by the colonoscopy provider are potentially sustainable since the entity incurring the PN costs directly benefits from the decreased no-show rates. What this evaluation did reveal, however, is that the use of PNs for CCSP did result in a lower percentage of missed appointments than what is being seen nationally.  This suggests there are cost benefits to using PNs to prepare clients for colonoscopies but that the provider lens may not be the most effective to really evaluate the true cost benefit.  Examining the cost benefits of PN from the societal perspective may be beneficial to understand the fuller picture. However, we are not aware of any comparative effectiveness studies examining difference between primary care-based PNs and colonoscopy providers in terms of decreases in no-show rates.

This project has several important limitations, including the fact that it is an evaluation of a specific program implemented in safety net community clinics, and is therefore not generalizable. However, the lessons may be informative to other organizations. Additionally, five out of 10 purposely selected clinics provided microcosting data, and it is possible that the costs in these five clinics are not comparable to costs in other clinics, particularly given the variation in costs across sizes of the clinic. Additionally, clinics that monitor their costs may be more likely to have lower costs, since they are mindful of them. It is also important to recognize that the idea of colonoscopy providers making payments to referring clinics for providing PN services is a conceptual analysis, as there are a number of potential regulatory issues related to payments from one health care provider to another that would need investigation before recommending this approach to sustaining the CCSP or similar programs. Also, the study did not examine noneconomic factors that could affect provider willingness to pay for PN services. Finally, the break-even analysis assumes that colonoscopy providers pay for unused procedure-support services when a screening appointment is missed; break-even figures for PN services would be lower if some or all procedure-support services are not charged to providers when not used.

Conclusion

For CCSP, using a PN led to an increased percentage of patients presenting for their colonoscopy exam fully prepared. As a result, the contributions made by PNs—including increasing screening attendance and preparation—may result in an economic benefit for screening providers by decreasing missed opportunities for screening. However, this economic benefit may be insufficient to create long-term sustainability, especially in the low-volume clinics that demonstrated substantially higher costs than the high-volume clinics. More research is needed to develop and test models to sustain PN services in primary care clinics to increase attendance and preparation for screening colonoscopy. 

References

1.    Riley, S, Riley C. The role of patient navigation in improving the value of oncology care. J Clin Pathways. 2016;2(1):41–47.

2.     Lairson DR, Dicarlo M, Deshmuk AA, et al. Cost-effectiveness of a standard intervention versus a navigated intervention on colorectal cancer screening use in primary care. Cancer. 2014;120(7):1042-1049.

3.    Jandorf L, Stossel LM, Cooperman JL, et al. Cost analysis of a patient navigation system to increase screening colonoscopy adherence among urban minorities. Cancer. 2013;119(3):612-620.

4.    Elkin EB, Shapiro E, Snow JG, Zauber AG, Krauskopf MS. The economic impact of a patient navigator program to increase screening colonoscopy. Cancer. 2012;118(23):5982-5988.

5.    Nash D, Azeez S, Vlahov D, Schori M. Evaluation of an intervention to increase screening colonoscopy in an urban public hospital setting. J Urban Health. 2006;83(2):231-243.

6.    Lebwohl B, Neugut AI, Stavsky E, et al. Effect of a patient navigator program on the volume and quality of colonoscopy. J Clin Gastroenterol. 2011;45(5):e47-53.

7.    Centers for Disease Control and Prevention. Colorectal Cancer Statistic 2015. Centers for Disease Control and Prevention website. http://www.cdc.gov/cancer/colorectal/statistics/. Accessed May 9, 2017.

8.    Pignone M, Saha S, Hoerger T, Mandelblatt J. Cost-effectiveness analyses of colorectal cancer screening: a systematic review for the U.S. Preventive Services Task Force. Ann Intern Med. 2002;137(2):96-104.

9.    Joseph DA, King JB, Miller JW, Richardson LC, Centers for Disease Control and Prevention. Prevalence of colorectal cancer screening among adults--Behavioral Risk Factor Surveillance System, United States, 2010. MMWR. 2012;61(2):51-56.

10.    Maciosek MV, Solberg LI, Coffield AB, Edwards NM, Goodman MJ. Colorectal cancer screening: health impact and cost effectiveness. Am J Prev Med. 2006;31(1):80-89.

11.    Lansdorp-Vogelaar I, Kuntz KM, Knudsen AB, van Ballegooijen M, Zauber AG, Jemal A. Contribution of screening and survival differences to racial disparities in colorectal cancer rates. Cancer Epidemiol Biomarkers Prev. 2012;21(5):728-736.

12.    Rim SH, Joseph DA, Steele CB, Thompson TD, Seef LC. Centers for Disease Control and Prevention colorectal cancer screening - United States, 2002, 2004, 2006, and 2008. MMWR Surveill Summ Suppl. 2011;60(1):42-46.

13.    Siegel R, Desantis C, Jemal A. Colorectal cancer statistics, 2014. CA Cancer J Clin. 2014;64(2):104-117.

14.    Ladabaum U, Mannalithara A, Jandorf L, Itzkowitz SH. Cost-effectiveness of patient navigation to increase adherence with screening colonoscopy among minority individuals. Cancer. 2015;121(7):1088-1097.

15.    Wilson FA, Villarreal R, Stimpson JP, Pagan JA. Cost-effectiveness analysis of a colonoscopy screening navigator program designed for Hispanic men. J Cancer Educ. 2015;30(2):260-267.

16.    Wolf HJ, Dwyer A, Ahnen DJ, et al. Colon cancer screening for Colorado’s underserved: a community clinic/academic partnership. Am J Prev Med. 2015;48(3):264-270.

17.    Holt CL, Schroy PC 3rd. A new paradigm for increasing use of open-access screening colonoscopy. Clin Gastroenterol Hepatol. 2008;6(4):377-378.

18.    Kazarian ES, Carreira FS, Toribara NW, Denberg TD. Colonoscopy completion in a large safety net health care system. Clin Gastroenterol Hepatol. 2008;6(4):438-442.

19.    Turner BJ, Weiner M, Yang C, TenHave T. Predicting adherence to colonoscopy or flexible sigmoidoscopy on the basis of physician appointment-keeping behavior. Ann Intern Med. 2004;140(7):528-532.