Abstract
Background
Cancer patients treated in community hospitals receive less guideline-recommended care and experience poorer outcomes compared to academic medical centers or National Cancer Institute (NCI)-Designated Cancer Centers. The Markey Cancer Center Affiliate Network (MCCAN) was designed to address this issue in Kentucky, the state with the highest cancer mortality.
Study Design
Using data obtained from the Kentucky Cancer Registry, the impact of patients treated in MCCAN hospitals was evaluated on four evidence-based, Commission on Cancer (CoC) quality measures using a pre/post, matched-cohort study design. Each group included 13 hospitals, matched on bed size, cancer patient volume, community population and region (Appalachian vs. non-Appalachian). Compliance with quality measures was assessed for the three years prior to joining MCCAN (T1) and three years after (T2).
Results
In T1, control hospitals demonstrated higher compliance with two quality measures compared to MCCAN hospitals. In T2, MCCAN hospitals achieved higher compliance in three measures, compared to controls. From T1 to T2, MCCAN hospitals significantly increased compliance on three measures (vs. 1 measure for controls). While most hospitals were not accredited by the CoC in T1, 92% of MCCAN hospitals had achieved accreditation by the end of Time 2 (vs. 23% of controls).
Conclusion
After joining the network, compliance with quality measures and achieving CoC accreditation increased significantly among MCCAN hospitals compared to control hospitals. MCCAN provides a unique academic/community-collaboration model able to make a significant impact on improving cancer care. Future research is needed to adapt and evaluate similar interventions in other states and regions.
Keywords: community cancer care, quality measures, breast cancer, lung cancer, accreditation
Precis
This study evaluated the Markey Cancer Center Affiliate Network, an academic/community-collaboration model designed to improve the quality of cancer care in community hospitals. The study found participating hospitals demonstrated significant improvement in three of four evidence-based, Commission on Cancer quality measures, and achieved greater compliance than non-participants in three measures.
Introduction
Community hospitals treat approximately 80–85% of all cancer patients in the U.S.1,2 There are significant advantages for cancer patients and their families when they can receive high quality cancer treatment close to home.3–5 This is particularly important for rural and underserved patients.6 However, there is evidence that cancer patients treated in community hospitals receive less guideline-recommended care and experience poorer outcomes compared to academic medical centers or National Cancer Institute (NCI)-Designated Cancer Centers.2,7–10 In addition to providing excellent patient care and conducting cutting-edge research, NCI-Designated Cancer Centers are expected to serve their catchment areas with programs and services tailored to their unique needs and populations.11
According to the most recent CDC data, the Common wealth of Kentucky leads the nation in cancer incidence and mortality.12 Disparities in access to and use of high-quality care have also been noted for Appalachia in general and Kentucky in particular.13–15 The NCI-designated University of Kentucky (UK) Markey Cancer Center developed the Markey Cancer Center Affiliate Network (MCCAN)for the purpose of enhancing access to high-quality cancer services and programs for cancer patients treated in community hospitals throughout the state. MCCAN assists Affiliate Network hospitals with achieving and maintaining American College of Surgeons Commission on Cancer (CoC) accreditation for their cancer programs, as part of affiliation requirements.
CoC accreditation recognizes hospitals for achieving high-quality, comprehensive and multidisciplinary cancer care. In order to earn and maintain CoC accreditation, a hospital cancer program must meet CoC standards, which include establishment of a cancer committee, multidisciplinary cancer conferences, and oncology nursing services.16 CoC accredited hospitals must also provide a set of specific services, either on-site or by referral to other hospitals. These services include diagnostic imaging, radiation oncology, systemic therapy, psychosocial support, rehabilitation, nutrition, and access to clinical research.
In addition, CoC accreditation requires that the hospital establishes a cancer registry that collects patient, tumor, treatment and outcome information for each cancer case and submits these data to the National Cancer Database.16 This data is compiled from all CoC-accredited hospitals and used to provide feedback to individual programs regarding their performance on quality of care measures. These quality measures, evidence-based and endorsed by the National Quality Forum, are collectively called Cancer Program Practice Profile Reports (CP3R).18,19 CoC requires that the annual performance rate for these measures be equal to or greater than the expected rate specified by the CoC, or that the upper confidence interval for the performance rate encompass the expected rate. The confidence interval is influenced by the number of eligible cases for any given hospital and would, therefore, be narrower for very large hospitals and wider for very small hospitals.
MCCAN uses the CoC standards as a frame work to provide services to support data collection, quality improvement initiatives, assessment of clinical performance measures and clinical consultation to help improve the quality of care in community hospitals. Peer-to-peer support is provided to administrative leadership and physician champions to share best practices across the network and provide mentoring on how best to establish and enhance key structures including tumor boards and cancer committees. MCCAN staff provide support and guidance for cancer committee responsibilities including goal setting, implementing and evaluating cancer-related activities, reviewing CP3R measures, and initiating quality improvement activities. Peer-to-peer support is also provided for data collection/reporting staff, which involves training and quality assurance monitoring to ensure the capture of high-quality data. MCCAN provides ongoing educational programs for clinicians/staff, including live events, webinars and instructional videos that address the clinical practice criteria necessary to treat cancer patients according to each of the CoC required CP3R quality of care measures. Direct services are also provided to support implementation of CoC standards that are particularly challenging in smaller community hospitals, such as genetic counseling delivered via telehealth. MCCAN facilitates access to clinical trials/studies and provides referral navigation support to coordinate care for patients who receive some of their cancer care at UK. This multi-faceted approach promotes a focus on quality care and quality improvement among affiliate hospitals.
In this article, we evaluated the impact of participation in MCCAN on four CP3R evidence-based cancer care quality measures for Affiliate Network hospitals and a matched set of control hospitals using a pre/post, matched-cohort study design.
Methods
Study Design
There are currently 21 MCCAN hospitals (Figure 1).17 These hospitals joined MCCAN over a number of years beginning in 2006 and additional hospitals have joined the network in subsequent years. For this study, a hospital had to be part of MCCAN for at least three years prior to 2018, so that compliance with the quality measures could be assessed for the three years before the hospital joined MCCAN and the three years following joining MCCAN. Thirteen MCCAN hospitals met the criteria for inclusion in this study.
Figure 1.
Location of the MCCAN Hospitals
A control hospital was selected for each MCCAN hospital, matched using a pre-specified, objective algorithm. This algorithm included dividing MCCAN hospitals into two groups by whether they were located in an Appalachian or non-Appalachian area of the state. The Appalachian area of Kentucky is primarily rural, with high rates of poverty, low educational attainment and limited access to health care.13,20,21 These two groups were further divided by whether the MCCAN hospital was in a county with a population less than 50,000 or in a county with a population of 50,000 or greater. Each MCCAN hospital was then matched to a control hospital in the same subgroup that also had the closest number of 1) licensed, acute-care beds and 2) incident analytic cases of cancer per year.
We selected all CP3R quality measures that were established at least three years before the first hospital joined MCCAN and remained in place through 2017. Four CP3R measures met these criteria:
Radiation therapy administered within 1 year (365 days) of diagnosis for women under age 70 receiving breast conserving surgery for breast cancer (BCSRT).
Radiation therapy recommended or administered following any mastectomy within 1 year (365 days) of diagnosis of breast cancer for women with ≥ 4 positive regional lymph nodes (MASTRT).
Adjuvant chemotherapy recommended or administered within 4 months (120 days) of diagnosis for patients under the age of 80 with AJCC Stage III (lymph node positive) colon cancer (ACT).
At least 12 regional lymph nodes removed and pathologically examined for resected colon cancer (12RLN).
In each control hospital, the proportion of cancer patients treated according to the CP3R quality measures were evaluated over the same three-year period before its matched MCCAN hospital joined the network and the same three-year period after its matched MCCAN hospital joined the network. The study design is shown in Figure 2.
Figure 2.
Research Design
Data Source
Data to calculate the proportion of patients treated according to the four quality of cancer care measures were obtained from the Kentucky Cancer Registry (KCR). KCR is part of both the NCI Surveillance, Epidemiology, and End Results (SEER) Program and the CDC National Program of Cancer Registries (NPCR). Since the inception of the CP3R measures, the KCR has created and implemented computer algorithms that carefully assesses 1) the eligibility of each cancer patient for quality measures that can be evaluated using data from KCR and 2) the proportion of eligible patients for a given hospital compliant with these CP3R criteria.
Statistical Analyses
The proportion of patients treated according to each CP3R measure in the MCCAN hospitals at time 1 (T1) were compared to the proportion of patients treated according to each CP3R measure in the matched control hospitals at T1. The same was done for time 2 (T2). Changes in the compliance rate between T1 and T2 were also calculated for MCCAN and control hospitals. Contingency tables that contained cells with five or fewer subjects were evaluated using a Fishers Exact test. Contingency tables in which all cells contained more than five subjects were evaluated using a Chi Square test and the differences between means were evaluated using a T test. All analyses were performed in SAS version 9.4 and statistical significance was set at the 5% level.
Results
Prior to joining the Network, the average number of acute-care beds and analytic cancer patients were slightly higher among the MCCAN hospitals compared to the matched control hospitals (Table 1). However, neither of these differences were statistically significant. In contrast, the average county-level population for the MCCAN hospitals was slightly less than that for the matched control hospital; again, not significantly different. Prior to each MCCAN hospital joining the network there was no significant difference in the number of MCCAN hospitals (3) and the number of matched control hospitals (2) that were CoC accredited.
Table 1.
Comparison of baseline (T1) characteristics between MCCAN and control hospitals
| MCCAN Hospitals | Control Hospitals | Significance α=.05 | |
|---|---|---|---|
| N=13 | N=13 | p-value | |
| Average number of acute care beds | 134 | 124 | 0.35 |
| Average number of analytic cancer cases | 352 | 261 | 0.21 |
| Average county population | 40,984 | 63,747 | 0.19 |
| Number located in Appalachian region | 7 (59%) | 7 (59%) | 1.00 |
| Accredited by Commission on Cancer (CoC) | 3 (23%) | 2 (15%) | 1.00 |
| Not accredited by CoC | 10 (77%) | 11 (85%) | |
The proportion of patients receiving radiation therapy following breast conserving surgery (BCSRT) and the proportion having 12 nodes examined following a colon resection (12RLN) were significantly lower among the MCCAN hospitals prior to joining the network (T1) compared to the matched control hospitals (Table 2a). During T1, five MCCAN hospitals met the BCSRT criteria and four met the 12RLN criteria. In contrast, six matched control hospitals met the BCSRT criteria and five met the 12RLN criteria during T1. There were no significant differences in the proportion of patients receiving radiation therapy following mastectomy (MASTRT) and the number of node positive stage III colon cancer patients receiving chemotherapy (ACT) between the MCCAN hospitals and the matched control hospitals prior to joining the network. Three MCCAN hospitals met the MASTRT criteria and six met the ACT criteria during T1. Among the matched control hospitals, three hospitals met the MASTRT criteria and five met the ACT criteria during T1.
Table 2:
Compliance with quality measures: MCCAN vs. Control hospitals by time period
| MCCAN Hospitals | Control Hospitals | Significance α = .05 | |||||
|---|---|---|---|---|---|---|---|
| Eligible Patients | Compliant | Eligible Patients | Compliant | ||||
| N | N | % | N | N | % | p-value | |
| 2a. TIME 1 | |||||||
| BCSRT | 296 | 202 | 68% | 347 | 277 | 80% | 0.0008 |
| MASTRT | 56 | 33 | 59% | 64 | 35 | 55% | 0.64 |
| 12RLN | 383 | 249 | 65% | 407 | 295 | 72% | 0.02 |
| ACT | 84 | 67 | 80% | 99 | 68 | 69% | 0.09 |
| 2b. TIME 2 | |||||||
| BCSRT | 266 | 228 | 86% | 353 | 273 | 77% | 0.009 |
| MASTRT | 61 | 51 | 84% | 50 | 33 | 66% | 0.03 |
| 12RLN | 327 | 262 | 80% | 347 | 292 | 84% | 0.17 |
| ACT | 83 | 72 | 87% | 96 | 71 | 74% | 0.03 |
During the 3 years after the MCCAN hospitals joined the network (T2), more of the eligible patients treated in MCCAN hospitals received radiation therapy after breast conserving surgery (85.7%), and mastectomy (83.6%), and stage III colon cancer patients were more likely to receive chemotherapy (86.8%) compared to the matched control hospitals (77.3%, 66.0%, and 74.0%, respectively). All of these differences were statistically significant (Table 2b). During T2, all 13 MCCAN hospitals met BCSRT and ACT criteria while 12 of the MCCAN hospitals met the MASTRT criteria. In contrast, six matched control hospitals met the BCSRT criteria, three met the MASTRT criteria, and five met the ACT criteria. There was no significant difference between the eligible patients in the MCCAN Hospitals having 12 nodes examined following a colon resection (80.1%) and control hospitals (84.1%). During T2, 12 MCCAN hospitals met the 12RLN criteria and eight matched control hospitals met the 12RLN criteria.
Compliance with the CP3R measures increased during the three years after the MCCAN hospitals joined the network compared to the three years prior to joining the network for all four quality measures. The difference was statistically significant for three of the four CP3R measures (BCSRT increased by 17.5%, MASTRT increased by 24.7%, and 12RLN increased by 15.1%) (Table 3a). The proportion of Stage III colon cancer patients having adjuvant chemotherapy within 4 months of diagnoses also increased (7.0%), but this was not statistically significant.
Table 3:
Compliance with quality measures: Time 1 vs. Time 2 by hospital group
| Time 1 | Time 2 | % Change | Significance α = .05 | |||||
|---|---|---|---|---|---|---|---|---|
| Eligible Patients | Compliant | Eligible Patients | Compliant | |||||
| N | N | % | N | N | % | p-value | ||
| 3a. MCCAN Hospitals | ||||||||
| BCSRT | 296 | 202 | 68% | 266 | 228 | 86% | 17% | <0.0001 |
| MASTRT | 56 | 33 | 59% | 61 | 51 | 84% | 25% | 0.003 |
| 12RLN | 383 | 249 | 65% | 327 | 262 | 80% | 15% | <0.0001 |
| ACT | 84 | 67 | 80% | 83 | 72 | 87% | 7% | 0.27 |
| 3b. CONTROL Hospitals | ||||||||
| BCSRT | 347 | 277 | 80% | 353 | 273 | 77% | −2% | 0.42 |
| MASTRT | 64 | 35 | 55% | 50 | 33 | 66% | 11% | 0.5 |
| 12RLN | 407 | 295 | 72% | 347 | 292 | 84% | 12% | 0.0001 |
| ACT | 99 | 68 | 69% | 96 | 71 | 74% | 5% | 0.42 |
In contrast, there were no statistically significant changes in the CP3R compliance rate for three of the quality measures in the control hospitals (Table 3b). The only significant change was an increase in the proportion of eligible patients having 12 nodes examined following a colon resection (11.7%). A graphic summary of all results is provided in Figure 3.
Figure 3.
Percent compliance with quality measures by time period and hospital group.
Discussion
Substantial improvements in cancer care, according to CP3R measures, occurred in the MCCAN hospitals after joining the Network. Several of these increases were exceptional. A significantly higher percent of the eligible patients with breast conserving surgery received radiation therapy in the matched control hospitals (79.8%) compared to the MCCAN hospitals (68.2%) before the MCCAN hospitals joined the Network. However, after joining the Network, there was a 17.5% absolute percent increase in the compliance rate for the MCCAN hospitals (85.7%), while there was a 2.5% reduction in the compliance rate in the matched control hospitals (77.3%). There was a nearly 25% absolute percent increase in the proportion of patients receiving radiation therapy following mastectomy in the MCCAN hospitals after joining the Network and only an 11.3% increase in the matched control hospitals. Prior to joining the Network, compliance with the four CP3R measures among the MCCAN hospitals ranged from 58.9% to 79.8%. In contrast, after joining the Network, compliance with each of the four measures was over 80%. Only one of the measures among the matched control hospitals exceeded 80%.
One of the likely drivers of these improvements is related to the MCCAN requirement that its affiliate member hospitals pursue CoC accreditation. Only three (23%) of the MCCAN hospitals had CoC accreditation prior to joining the Network and a similar rate of two (15%) of the matched control hospitals. However, in the 3 years after joining the network, all but one of the MCCAN hospitals (92%) had achieved CoC accreditation whereas only one additional matched control hospital received accreditation (23%). As previously mentioned, one of the requirements for CoC accreditation is to achieve the expected performance rate on the CP3R measures. Accredited hospitals must therefore track their performance on these measures and take action when compliance falls below expected levels. MCCAN provides specific education and quality-improvement support to help affiliate hospitals comply with all of the CP3R measures. The requirements of the CoC and the support provided by MCCAN appear to strongly incentivize enhanced performance on the quality measures examined in this study.
There are some limitations that are important to consider when interpreting these results. This study was conducted in a single state and with a relatively small number of hospitals. Even so, results were significant. A second limitation is that we do not have information regarding patient preferences or other circumstances that could influence their treatment, and thus it is not possible to determine why CP3R quality measures were not met. In addition, of the four CP3R measures we were able to evaluate, two addressed breast cancer and two addressed colorectal cancer. These are common cancer sites and very important to assess given that they can theoretically be treated by general surgeons in hospitals of just about any size. There are, however, other CP3R measures that address the quality of care for patients with other types of cancer. As additional community hospitals join the Network and are part of the Network for longer periods of time, we will be able to include both more hospitals in this type of analysis and evaluate CP3R measures that address the quality of care for additional cancer sites.
Finally, MCCAN requires and supports CoC accreditation as a primary component of its services to affiliate hospitals. However, our study was not able to identify which of the specific MCCAN educational and quality improvement services resulted in the greatest improvement in cancer care.
Conclusions
Improvements in the quality of care for breast and colon cancer patients as measured by the four CP3R criteria were associated with joining MCCAN. Delivering high-quality cancer care remains a challenge for many community hospitals in the U.S. MCCAN provides a unique model with the potential to make a significant positive impact on the care of cancer patients. Future research and evaluation are needed to determine which network activities are most impactful so they can be replicated in other states and regions.
Acknowledgments
Support: This research was supported by a National Cancer Institute, Cancer Center Support Grant (P30 CA177558) awarded to the University of Kentucky Markey Cancer Center. The Cancer Research Informatics Shared Resource Facility of the University of Kentucky Markey Cancer Center developed computer algorithms used in this study. This research was also supported by a National Cancer Institute, Cancer Center Support Grant (P30 CA086862) awarded to the University of Iowa Holden Comprehensive Cancer Center.
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
References
- 1.Petrelli NJ. A community cancer center program: getting to the next level. J Am Coll Surg. 2010;210(3):261–270. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Pfister DG, Rubin DM, Elkin EB. Risk Adjusting Survival Outcomes in Hospitals That Treat Patients With Cancer Without Information on Cancer Stage (vol 1, pg 1303, 2015). Jama Oncol. 2015;1(9):1323–1323. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ambroggi M, Biasini C, Del Giovane C, Fornari F, Cavanna L. Distance as a Barrier to Cancer Diagnosis and Treatment: Review of the Literature. The oncologist. 2015;20(12):1378–1385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Farrow DC, Hunt WC, Samet JM. Geographic variation in the treatment of localized breast cancer. The New England journal of medicine. 1992;326(17):1097–1101. [DOI] [PubMed] [Google Scholar]
- 5.Bakker DA, DesRochers C, McChesney C, Fitch M, Bennett J. Community cancer clinics: patients' perspectives. Supportive care in cancer: official journal of the Multinational Association of Supportive Care in Cancer. 2001;9(4):234–240. [DOI] [PubMed] [Google Scholar]
- 6.Joynt KE, Harris Y, Orav EJ, Jha AK. Quality of care and patient out comes in critical access rural hospitals. Jama. 2011;306(1):45–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Friese CR, Earle CC, Silber JH, Aiken LH. Hospital characteristics, clinical severity, and outcomes for surgical oncology patients. Surgery. 2010;147(5):602–609. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Paulson EC, Mitra N, Sonnad S, et al. National Cancer Institute designation predicts improved outcomes in colorectal cancer surgery. Ann Surg. 2008;248(4):675–686. [DOI] [PubMed] [Google Scholar]
- 9.Birkmeyer NJO, Goodney PP, Stukel TA, Hillner BE, Birkmeyer JD. Do cancer Centers designated by the National Cancer Institute have better surgical outcomes? Cancer-Am Cancer Soc. 2005;103(3):435–441. [DOI] [PubMed] [Google Scholar]
- 10.Shulman LN, Palis BE, McCabe R, et al. Survival As a Quality Metric of Cancer Care: Use of the National Cancer Data Base to Assess Hospital Performance. Journal of Oncology Practice. 2018;14(1):41–+. [DOI] [PubMed] [Google Scholar]
- 11.National Cancer Institute. NCI-Designated Cancer Centers. 2019.
- 12.Kochanek KD, Murphy SL, Xu J, Tejada-Vera B. Deaths: Final Data for 2014. In. Hyattsville MD: National Center for Health Statistics; 2016. [PubMed] [Google Scholar]
- 13.Lengerich EJ, Tucker TC, Powell RK, et al. Cancer incidence in Kentucky, Pennsylvania, and West Virginia: disparities in Appalachia. The Journal of rural health: official journal of the American Rural Health Association and the National Rural Health Care Association. 2005;21(1):39–47. [DOI] [PubMed] [Google Scholar]
- 14.Behringer B, Friedell GH, Dorgan KA, et al. Understanding the Challenges of Reducing Cancer in Appalachia. Calif J Health Promot. 2007;5(SI):040–049. [Google Scholar]
- 15.Meilleur A, Subramanian SV, Plascak JJ, Fisher JL, Paskett ED, Lamont EB. Rural residence and cancer outcomes in the United States: issues and challenges. Cancer epidemiology, biomarkers & prevention: a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2013;22(10):1657–1667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Commission on Cancer. Cancer Program Standards: Ensuring Patient-Centered Care. 2016. ed2015.
- 17.Markey Cancer Center Affiliate Network.https://ukhealthcare.uky.edu/markey-cancer-center/refer-patient/affiliate-network. Accessed November 5, 2019.
- 18.Commission on Cancer. Value and Benefits of Accreditation. https://www.facs.org/quality-programs/cancer/coc/apply/benefitscoc. Accessed September 7, 2018.
- 19.Shulman LN, McCabe R, Gay G, Palis B, McKellar D. Building Data Infrastructure to Evaluate and Improve Quality: The National Cancer Data Base and the Commission on Cancer's Quality Improvement Programs. J Oncol Pract. 2015;11(3):209–212. [DOI] [PubMed] [Google Scholar]
- 20.Wilson RJ, Ryerson AB, Singh SD, King JB. Cancer Incidence in Appalachia, 20042011. Cancer epidemiology, biomarkers & prevention: a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2016;25(2):250–258. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Yao N, Alcala HE, Anderson R, Balkrishnan R. Cancer Disparities in Rural Appalachia: Incidence, Early Detection, and Survivorship. The Journal of rural health: official journal of the American Rural Health Association and the National Rural Health Care Association. 2017;33(4):375–381. [DOI] [PubMed] [Google Scholar]