The articles in this supplement were commissioned based on participation in evaluating the Centers for Disease Control and Prevention–funded Colorectal Cancer Screening Demonstration Program.
The opinions or views expressed in this supplement are those of the authors and do not necessarily reflect the opinions or recommendations of the journal editors, the American Cancer Society, John Wiley & Sons, Inc., or the Centers for Disease Control and Prevention.
This program would not have been possible without the cooperation and participation of the following individuals from the DHMH CCSC: Barbara Andrews, Marsha Bienia, Carmela Groves, Annette Hopkins, Kitty Musk, Eileen Steinberger, Vicki Varsalone; the following individuals from the participating hospitals: Mary Austin, Jennie Boyer, Kira Eyring, Michael Farrier, Nancy Goldstein, Christine Hendrix, Alva Hutchison, Myra James, Jin Lee, Marylu Manning, Eileen Marks, Candace Shaffer, Eden Stotsky, Toby Tighe, Linda Wieczynski; and the participating colonoscopists: Richard Baum, Sudhir Dutta, Anthony Kalloo, Gopal Kowdley, Robert Mathieson, Patrick Okolo, Justin Somerville, and Rakesh Vinayek.
This article is dedicated to the memory of Eileen Marks. Her wisdom, compassion, and commitment were instrumental to the success of the program.
Maryland, excluding Baltimore City, began public health screening for colorectal cancer in 2000. Initiating colorectal cancer screening in Baltimore City was an objective in the Maryland Comprehensive Cancer Control Plan. The Centers for Disease Control and Prevention's (CDC's) funding announcement for the “Colorectal Cancer Screening Demonstration Program” (CRCSDP) was seen as a potential opportunity for Maryland to begin screening in Baltimore City.
The Maryland Department of Health and Mental Hygiene (DHMH), the American Cancer Society, and five Baltimore City Hospitals collaborated to develop the funding application and model for the Baltimore City CRCSDP. After receipt of funding, between 2005 and September 2009, screening sites collaborated with the DHMH to implement the multi-site colorectal cancer screening program in Baltimore City.
Close collaboration across organizational boundaries enabled the funding, formation, and implementation of the CRCSDP in Baltimore City. The Baltimore City CRCSDP illustrates the complexity of establishing a functional public health screening program. The program overcame expected and unexpected fiscal, programmatic, and clinical challenges to successfully perform 709 colonoscopies screening cycles among 696 people and detect three cancers during the 38 months of screening.
In 1998, as part of Maryland's portion of the multistate Master Settlement Agreement with the tobacco industry, the Cigarette Restitution Fund (CRF) program was created by the Maryland General Assembly and signed into law by the governor. The major goals of the program were to reduce cancer mortality and to lessen cancer disparities among ethnic and racial minorities in Maryland. The statute called for each of Maryland's 24 jurisdictions (23 counties and Baltimore city) to form Community Health Coalitions and solicit stakeholder input into the direction of the CRF programs at the local level. Colorectal cancer was the second-leading cause of cancer death among men and women nationally and in Maryland in 2000.[2, 3] African Americans in Maryland had an age-adjusted colorectal cancer mortality rate 1.4 times that of whites (31.1 versus 22.1 per 100,000, respectively).
With their CRF funding in 2000, 23 of Maryland's 24 jurisdictions prioritized colorectal cancer, developed education and/or screening programs for their populations, and focused their enrollment for colorectal screening on underinsured and uninsured persons. In contrast, Baltimore city, in collaboration with its Community Health Coalition, selected prostate, oral, breast, and cervical cancer as priority areas for screening and outreach rather than colorectal cancer. As a result, underserved residents of Baltimore city continued to lack access to no-cost colorectal cancer screening services.
In 2003, a “Colorectal Cancer Collaborative” committee of the Baltimore City Community Health Coalition was formed under the leadership of a representative of the American Cancer Society (ACS). The Colorectal Cancer Collaborative included representatives from the Maryland Department of Health and Mental Hygiene (DHMH), the Baltimore City Health Department, major city hospitals, Johns Hopkins Medical Institutions, and the University of Maryland, Baltimore. With limited resources, the Colorectal Cancer Collaborative initiated colorectal cancer education and awareness activities in Baltimore city.
At the same time, the Maryland Comprehensive Cancer Control Plan was developed, and members of the Colorectal Cancer Collaborative who served on this group ensured that the plan included a specific objective, stating, “Increase funding for colorectal cancer screening among uninsured, low-income Maryland residents, especially in Baltimore City.”
An opportunity to realize this objective emerged in April 2005, when the Centers for Disease Control and Prevention (CDC) offered a funding opportunity supporting an organized colorectal cancer screening program. The goal of the program, the Colorectal Cancer Screening Demonstration Program (CRCSDP), was to assess the practicality of varied screening delivery models in differing geographic settings, including costs of the models.
The objectives of this article are to describe: 1) the planning and development of a multisite, multiorganizational collaborative screening program in Baltimore, Maryland, an urban setting; and 2) the implementation and operation of the program, including fiscal, organizational, programmatic, and clinical challenges and successes, as well as screening outcomes.
Program Planning and Development
Developing the funding application and the program model
In response to the CDC funding announcement, the Maryland DHMH approached the Baltimore City Colorectal Cancer Collaborative. Given mutual interest, the chair of the collaborative, a strong “champion” of colorectal cancer screening from the American Cancer Society, convened a series of planning meetings that included several Baltimore city hospitals, local nonprofits, academic medical institutions, and other groups all interested in the initiative.
Several decision points were addressed during the initial stakeholders' meeting, including those related to fiscal leadership, colorectal cancer screening test selection, level of organizational centralization, and screening site locations.
1. Identifying the primary applicant for the funding
The primary applicant would have ultimate responsibility for the project—serving as the fiscal agent for the grant, coordinating the screening program, and communicating and reporting to the CDC. Potential agencies to fulfill this role included the DHMH, the Baltimore City Health Department, an academic medical center, a hospital, and a nonprofit agency. Factors affecting the choice of the applicant organization included costs and requirements of the funding announcements and related agency capacity.
The fiscal implications of identifying an academic medical center as the grant recipient were considered, including the amount of the award funds that would be committed to indirect costs (up to 60% of the award). DHHM staff contacted nonprofit groups and thought that the Maryland chapter of 1 large, national nonprofit would be the perfect applicant; however, the fiscal, clinical, and data reporting requirements of the award were an obstacle to that agency's ability to serve as the primary applicant.
The stakeholders' group determined collaboratively that the DHMH would be the strategic and appropriate grant recipient given the DHMH's experience with CDC grant applications and management.
2. Colorectal cancer screening test selection
In selecting a primary screening test to be used for the program, the collaborative gave primacy to screening the largest number of people in the most efficient manner. In making this decision, the collaborative first sought guidance from the larger, Maryland CRF colorectal cancer screening program, which had been conducting colorectal cancer screening using various screening tests for almost 5 years. For the CRF program, colonoscopy was preferred because it provided screening, diagnosis, and primary prevention through polypectomy; there was sufficient capacity for colonoscopy; and a resultant screening interval of 10 years for average-risk clients simplified patient support services such as case management. Moreover, CRF programs in Maryland that first selected fecal occult blood tests (FOBTs) with sigmoidoscopy switched to colonoscopy during the initial 3 years because of the above reasons.[4, 5] A second factor influencing test selection was the influence of hospital gastroenterologists involved in the planning who recommended colonoscopy. Consequently, the collaborative chose colonoscopy as the primary screening test for the program while also allowing individual screening sites to use FOBTs (either guaiac-based or immunochemical) with or without flexible sigmoidoscopy for asymptomatic average-risk patients who refused colonoscopy.
3. Level of organizational centralization
Two potential screening models, one decentralized and the other centralized, were considered.
In the decentralized model, the primary applicant (DHMH) would serve as the fiscal agent and would have a central coordinator who would oversee the program and be the liaison with the CDC. The fiscal agent would then contract with screening sites and distribute grant funds for each site to perform the colonoscopies and case management activities including contracting with endoscopists, facilities, and laboratories for services; recruiting patients; determining patient eligibility; navigating/managing patients through screening and beyond; paying providers; and entering data.
In the centralized model, the primary applicant (DHMH) would serve as the fiscal agent, hire central staff to coordinate all program activities, and perform the screening case management activities of contracting with providers, recruitment, and case management of screening, and the screening sites would only perform endoscopy under contract to the DHMH.
The members of the collaborative selected a decentralized model (Fig. 1) based on both administrative issues and problems of mission conflict with a centralized model. Specifically, a decentralized model was selected given the following: 1) the difficulty of the DHMH recruiting and quickly hiring nursing staff for case management, 2) the difficulty of the DHMH establishing contracts with providers for clinical services, 3) the fact that it was not the mission of the DHMH to be a primary case manager, and 4) the desire of the screening sites to have their own staff recruit and manage their patients and interact with their providers.
4. Identifying screening sites
Criteria for screening site selection included the ability to accept a grant from the DHMH, hire staff, establish contracts for clinical services, and accomplish all the clinical and technical requirements set forth by the CDC CRCSDP and the DHMH. Several Baltimore city hospitals and academic medical centers expressed interest. High indirect costs precluded 1 academic medical center from participating.
Five sites were selected for participation: Harbor Hospital and Union Memorial Hospital (both of MedStar Health), Johns Hopkins Medical Institutions, St. Agnes Hospital, and Sinai Hospital. To prepare the budget proposal for the CDC application, each site was given a preliminary dollar figure and asked to develop a budget that would support staff and fund as many colonoscopies as possible.
The chair of the collaborative asked each site to determine its cost for colonoscopy. This created a sense of friendly competition among the sites to reduce staffing and clinical costs and to make them roughly comparable by changing some items (eg, fringe benefits, pathology laboratory costs, precolonoscopy physical exams) into in-kind donations to the program. The DHMH prepared its budget, which included the costs for a program coordinator and for some central office operations. The DHMH contributed its existing CRF program client database so that screening sites could directly enter client data required by the CDC.
After collecting final budgets from each of the sites, the DHMH submitted its application to the CDC, which contained 90% of funding for the sites and 10% for the DHMH.
In June 2005, the CDC notified the DHMH that Baltimore city was selected as 1 of 5 sites in the country for the 3-year grant award that would begin on September 1, 2005. However, based on available funding at CDC, the funds awarded were less than what the DHMH had requested in their proposal.
Given the actual funding award, adjustments in program design and staffing were required. Leveraging other program funds, the DHMH identified 1 of its federally funded coordinators for the Maryland Comprehensive Cancer Control Plan as a half-time coordinator for the CRCSDP. The DHMH also provided in-kind staff for program and grant management, database design and implementation, clinical input, quality assurance, and data collection of screening costs. These contributions on the part of the DHMH allowed for the majority of CRCSDP funds to support direct screening services and staff at the clinical sites.
The Baltimore City CRCSDP began operations on September 1, 2005, with 5 screening sites projected to perform 260 colonoscopies each year.
Collecting and analyzing data
The CDC proposed and modified the required clinical data elements that were to be reported to the CDC from each site. In Maryland, the DHMH modified its existing client database to add the additional CDC-required variables, to obtain a data extract, and to recode data to meet the CDC requirements. Each hospital screening site entered data on patients screened through the CDCSDP into the database, a centralized, SQL, intranet-based client database housed at the DHMH and accessed through a virtual private network. The number of patients screened and their clinical outcomes were available to the DHMH through direct views of the input data, database reports, and downloaded data.
Program Implementation and Operation
Contracting with sites and providers
The 5 screening sites (Fig. 2) were located throughout Baltimore city. In the decentralized model, the DHMH was spared the responsibilities of hiring, contracting with colonoscopy providers, and direct patient management; however, the DHMH had to establish contracts with each of the 5 screening sites (in a process that took up to 5 months) and then to monitor the contracts. To receive funding, each screening site had to agree to the programmatic, billing and data reporting requirements and to agree to give the DHMH a copy of every clinical invoice received for each client screened so that the DHMH could meet the CDC requirement for cost assessment data.
Each screening site then had to establish contracts with its providers who would perform colonoscopies, anesthesiology, and pathology testing within the facilities where colonoscopies would be performed. In these clinical contracts, providers had to agree to follow guidelines developed for the program by the CDC and by the DHMH Maryland Colorectal Cancer Medical Advisory Committee. The Medical Advisory Committee for the CDCSDP was composed of gastroenterologists, surgeons, a medical oncologist, and a pathologist who were on the existing statewide committee, to which was added an endoscopist from each Baltimore city screening site. The committee reviewed and approved the Baltimore city program policies and procedures and clinical eligibility. Each hospital screening site then had to establish policies and procedures that reflected its own staffing and operations within these guidelines.
Identifying and enrolling eligible clients
The ACS offered to have their 1-800-ACS-2345 cancer hotline number advertised as a central number for client intake and eligibility determination for the 5 sites. Maryland houses the ACS South Atlantic Division's fully staffed cancer information “call center,” located 10 miles north of Baltimore city. The call center is staffed by cancer information specialists who answer calls, E-mails, and other inquiries from cancer patients, families of cancer patients, or those seeking information on cancer topics. The service is free to all who call.
Normally when individuals contact the call center, their names are entered into a central database used for fundraising. Because of privacy and other concerns, the DHMH had to ensure that clients who called the call center and were interested in the CRCSDP would have their identifying information kept separately and not be solicited for ACS donations in the future. The ACS worked with the DHMH to establish an internal logging system to keep these callers out of the ACS database. The log also allowed the DHMH to track call data. DHMH staff trained call center staff about SDP eligibility by residence, age, income, and insurance. Call center staff did not determine risk factors, prior screening, or clinical eligibility, which was done by the screening sites. Call center staff knew which of the screening sites were currently enrolling new clients, asked the potential clients which available site they preferred, and faxed intake information to the site that the client selected.
Potential clients also came from existing programs that had established patient populations:
Two of the sites were screening sites for the CDC-funded breast and cervical cancer screening program that served women of the same income and insurance eligibility.
One site managed the prostate cancer screening program for the Cigarette Restitution Fund in Baltimore city.
Other sites partnered with Federally Qualified Health Centers with whom they were affiliated.
Nursing staff hired as case managers at each site (0.5 full-time equivalents at each site) completed standard forms to determine eligibility by obtaining family and personal risk history, symptoms, and prior colorectal cancer screening history to assure that these met program requirements for eligibility.
Establishing procedures and communicating with and monitoring sites
The DHMH provided template materials for each site for education and outreach, policies and procedures, guidelines, result letters, data collection forms, and training materials that had been modified from the CRF colorectal cancer screening program. Training was held to familiarize the staff with colorectal cancer, the client database and how fields should be completed, billing and the requirement to reimburse at Medicare rates for colonoscopy and pathology, and the fiscal requirements of the grant with the DHMH. DHMH staff reviewed the final 5 “site manuals” and made them available to CDC staff before any site was approved to begin screening. Screening began within 9 months of the CDC award, 3 months longer than initially anticipated.
DHMH staff performed site visits to each screening site after the first few colonoscopies were completed and periodically thereafter. The DHMH provided feedback on case management, data entry, and billing accuracy and completeness. Monthly teleconferences were held among staff of the DHMH and the 5 sites to review progress and to communicate new information or answer questions. Data quality review was ongoing and proceeded as follows: sites entered data and DHMH staff compared text with coded data to assure accuracy and completeness; the DHMH received data submission feedback reports on the clinical data elements submitted quarterly to IMS, Inc., the data contractor retained by CDC to help manage the flow of clinical data; and the DHMH worked with site staff to correct problems identified such as missing required information, duplicates, errors in coding of findings, and recall recommendations that deviated from the CRCSDP-required national standards.
Managing fiscal, programmatic, and clinical challenges
Over the course of the development and implementation of the Baltimore City CRCSDP, several challenges emerged with the screening model. None of these proved insurmountable, but they did cause DHMH, the screening sites, and partners to adapt to the challenges of the start-up of a new program.
Because of initial budget constraints and fear that interest in the program would overwhelm the resources, the Baltimore CRCSDP reduced originally planned outreach and education activities. Having limited funding allocated for outreach and education activities meant that clinical case managers assumed these responsibilities, including recruitment of individuals to colorectal cancer screening.
The ACS provided its literature, which included its centralized call center number, and the CDC provided in-kind Screen for Life materials. Except for the part-time coordinator funded under the CDC Comprehensive Cancer Plan grant, DHMH staff for central program management was in-kind.
Recognizing and managing intersite differences
Rather than 1 program with 5 screening sites, the Baltimore city CRCSDP functioned as 5 separate miniprograms, which the DHMH treated as individual entities and managed for CDC reporting. This posed challenges for program implementation such as differing contract approval processes by each hospital, differing contracting methods between hospitals and their providers, and varying methods to track invoices and bill the DHMH. The DHMH considered the demonstration project public health implementation and not research and did not seek institutional review board (IRB) approval; however, we required that the hospital staff present the program to their respective institutional review boards (IRBs). Four hospital IRBs decided that the program was not research and exempted the program, whereas 1 hospital IRB required that all participants sign an additional consent form for participation in the program before screening.
Meeting CDC program requirements
During the program start-up and early implementation, evolving policy and requirements required flexibility and resulted in some unexpected costs.
The CDC finalized the required Colorectal Cancer Data Elements 3 months into implementation, and the DHMH modified its existing DHMH database thereafter. The delay in Office of Management and Budget clearance for sites to submit line-listed patient data (without identifiers) to the CDC led the CRCSDP to require a supplemental aggregate data report to monitor site screening activities in the interim. This monthly report required additional effort and continued for 2 years into the SDP.
Other adjustments were made based on CRCSDP reimbursement policies. Some hospital sites absorbed the cost for standard precolonoscopy tests that were not reimbursable through the program. Individuals presenting with colorectal symptoms were not eligible for CRCSDP screening. Maryland was initially interested in accepting clients into screening who had minor gastrointestinal symptoms, but the CRCSDP required a provider to certify that the symptoms were not suggestive of colorectal cancer and did not reimburse for this provider visit.
During the program start-up and early implementation, the Baltimore City CRCSDP adapted to changes in CDC program requirements.
Covering the cost of colonoscopy in Maryland
Hospital rates in Maryland are regulated by the Healthcare Services Cost Review Commission (HSCRC), and rates change frequently. All payers, including Medicare and the Baltimore City CRCSDP, must pay the rates that the HSCRC sets. For the CRCSDP, this included HSCRC rates for colonoscopy facility fees, which were higher than the Medicare rates published for nonregulated facilities. This unavoidable cost meant that for the amount of funding awarded, fewer clients could be screened than might have been expected if paying published Medicare rates for Maryland.
Billing DHMH for reimbursement
Three of the 5 sites had not worked under the provisions of a DHMH contract before and were not familiar with the level of reporting required under this CDC award. When sites invoiced the DHMH for costs associated with the program, some sites submitted invoices with an inaccurate or disallowed billing procedure code charge, requiring the DHMH to reject payment.
To address this, DHMH staff reiterated that we could only reimburse for specified procedures listed in the program manual. If a site or 1 of its providers wished to perform procedures not on the list, the site needed prior approval from the DHMH, which then had to seek approval from the CDC. Also, the DHMH visited the sites to meet with clinical and hospital billing staff to review billing procedures and approvals.
Using the CDC Cost Assessment Tool
Obtaining and reporting cost data were burdensome. The CDC required completion of a cost assessment tool that summarized the costs at the site for personnel, administrative costs, and in-kind contributions. In addition, the CDC required reporting of line-listed billing data for every invoice on each client. To comply, the DHMH collected each invoice from the screening sites and entered them into a database. Screening sites found that copying and submitting to the DHMH each bill on hundreds of clients was burdensome, yet they recognized that omitting bills would lead to underestimates of cost. The DHMH assumed the responsibility of entering each invoice into a database, a complex task that was undertaken in-kind.
Collecting clinical data and adhering to guidelines
The DHMH reviewed the quality of the data that were entered by the screening sites into the central database according to its usual procedures that assure accuracy and completeness of the fields. The DHMH performed site visits to the 5 screening sites to assure that the clinical and billing procedures and records met program standards of accuracy, completeness, and clinical acceptability. In addition, the CDC contractor, IMS, Inc., provided quality assurance feedback on the screening data that the DHMH submitted quarterly. These feedback reports enabled the DHMH to review data entry and further assure the quality and consistency of the data by national standards. The DHMH and its sites turned to the colonoscopists and pathologists to assure that the colonoscopy and pathology reports had all the required data and that the screening and surveillance recall intervals agreed with the CRCSDP requirements. One major challenge for the program was to code in the database the adequacy of bowel preparation. Often the preparation was reported as “excellent,” “good,” or “fair,” rather than “adequate to identify lesions ≥5 mm” or “inadequate,” as recommended. This raised several questions. What should the recall interval be on an average-risk client with no findings on colonoscopy in which a section of the colon was not “adequately” prepared? What did the colonoscopist mean by “fair prep”? Should an average-risk patient with a “fair prep” and no findings return immediately to have a repeat colonoscopy, or wait 5 years? If the colonoscopist said “fair prep” with a 5-year recall interval, should the prep be entered in the database as “inadequate”? These provider descriptions of bowel preparation and recall recommendations were not covered by existing national guidelines.
Monitoring outcomes of screening
Table 1 shows the screenings performed in the Baltimore City SDP by hospitals with the outcomes of screening. A total of 696 people received 1 or more screenings in the Baltimore City SDP. Of these, 597 (86%) were of minority race or ethnicity, and 190 (27%) were men. Success in reaching minority clients was because of the recruitment from existing cancer screening programs that serve mostly minorities, the patient populations seen at these hospitals, and the outreach and education efforts to communities in Baltimore city, which has a predominantly minority population.
Table 1. Colonoscopy Screenings Performed in the Baltimore City Screening Demonstration Program by Hospital and Outcome
Seven hundred nine colonoscopy “cycles” in which 744 colonoscopies were performed (eg, repeat colonoscopy because of inadequate bowel preparation or not reaching the cecum).
There were a total of 709 colonoscopy cycles with 744 colonoscopies performed. Of these, 139 colonoscopies detected 1 or more adenomas (19%), and 2 found adenomas with high-grade dysplasia. As a result of screening, cancer was detected in 3 people, 1 with adenocarcinoma and 2 with rectal carcinoid tumors. Unplanned events during or following colonoscopy included 5 colonoscopies with postpolypectomy bleeding (1 requiring emergency room visit with repeat colonoscopy), 3 cardiopulmonary events (2 with arrhythmia, 1 with hypertension), 3 drug-related reactions, and 12 gastrointestinal complaints.
Per CDC program policy and guidelines, payment for cancer treatment was not covered by the CDC SDP. The stage III adenocarcinoma was treated by the hospital as charity care. The 2 carcinoids did not require treatment beyond colonoscopy, which was covered in the program. Costs of the Baltimore SDP are summarized in other articles in this monograph.[9, 10]
Through its participation in the CRCSDP, Baltimore city achieved unprecedented collaboration across multiple competing health care systems throughout the city. The lessons learned through this demonstration helped position Maryland to successfully compete for future federal colorectal cancer funds.
The CRCSDP award brought additional funds to the Maryland DHMH, and this provided an opportunity to contract with 5 Baltimore city hospitals for colorectal cancer screening services, case management, and data entry in the city. The hospitals used template contracts, case management procedures, a screening database, and quality assurance procedures modified from the existing statewide CRF colorectal cancer screening program to provide infrastructure to manage and implement the CRCSDP at the local level.
There were fiscal, programmatic, and organizational challenges that the Baltimore city program encountered and surmounted, and overall the CRCSDP was an unqualified success. Clinically, from 2005 to 2009, the CRCSDP provided access to screening to individuals who would not have had access without this program, namely, the program performed 744 colonoscopies, detected 3 cancers, and found and removed adenomas in 139 individuals.
From a collaborative partnership perspective, staffs of 5 hospital sites that had previously not interacted with each other on a regular basis formed a cohesive screening program across Baltimore city. The sites demonstrated their commitment to the partnership and their belief in the public health mission in Baltimore city through this collaboration. The successes of the Maryland CRCSDP and the ongoing CRF colorectal screening program led to Maryland's successful new funding under the federal Colorectal Cancer Control Program in 2009.
The Colorectal Cancer Screening Demonstration Program evaluated in this supplement was funded by the Centers for Disease Control and Prevention Funding Opportunity Number: RFA AA030. This study was also supported by the Centers for Disease Control and Prevention CDC) Colorectal Cancer Screening Demonstration Program Cooperative Agreement U55/DP325046, CDC National Comprehensive Cancer Control Program Cooperative Agreement U58DP000827, the State of Maryland, and the Maryland Cigarette Restitution Fund.