Background

In the past decade, hospital and health system consolidation has increased significantly.^1–3 There are multiple drivers of this phenomenon, including economies of scale, shifts to population health and value-based reimbursement models, and a need for clinical standardization.^1,2,4 These factors coupled with increased patient consumerism have given rise to a range of partnerships, from mergers and acquisitions to joint ventures and clinical affiliations.⁵ Nowhere is this more evident than in oncology, where community health systems have begun to partner with academic medical centers, which are attempting to extend their reach into the community setting, where 85% of cancer care is delivered.^6,7 Several NCI-designated Comprehensive Cancer Centers have established partnership and affiliation models, including the Duke Cancer Network,⁸ MD Anderson Cancer Network,⁹ and Fox Chase Cancer Center Partners Program.¹⁰ Although each network’s relationship model is unique, these programs typically provide community hospital partners with access to consultative services, treatment protocols, research, and clinical trials.^8,9,11 A potential benefit for the academic centers is enhanced accrual to and more timely completion of clinical trials and the inclusion of more diverse populations in these studies.

One such partnership model is the Memorial Sloan Kettering Cancer Center (MSK) Cancer Alliance.¹² MSK created the Alliance with the goal of rapidly bringing the newest standards of care (SoC) into the community setting and increasing patient access to clinical trials in local communities. The Alliance is predicated on the notion that bidirectional engagement between MSK and community providers will lead to more timely adoption of cancer care advances, increased clinical trial activity from greater trial availability in the community, improved diversity of participants enrolled in clinical trials, and more willingness of cancer center investigators to open trials in the community.¹³ Since 2014, three members have joined the MSK Cancer Alliance: Hartford HealthCare Cancer Institute (HHC; 2014), Lehigh Valley Cancer Institute (LVCI; 2016), and Miami Cancer Institute at Baptist Health South Florida (MCI; (2017).¹⁴ Through the Alliance, MSK provides clinicians across disciplines opportunities to observe and present cases to MSK tumor boards, connect with MSK physicians for clinical input, attend clinical lectures, participate in MSK disease-specific retreats, provide local access to clinical trials for MSK patients, and have MSK physicians participate in local meetings.

A particularly distinguishing feature of the MSK Cancer Alliance is the development of SoCs by MSK disease management teams, which are multidisciplinary groups of clinicians treating patients with specific cancers. The SoCs are disease-specific process measures for 8 cancer sites (breast, colon, endometrial, kidney, lung, and prostate cancers; melanoma; and non-Hodgkin’s lymphoma), intended to cover at least 50% of all cancer cases at each Alliance facility. These SoCs aim to codify MSK practice from the time of a patient’s initial presentation through diagnosis, treatment planning, treatment administration, follow-up, and survivorship.

MSK uses internal data to benchmark practice for each SoC in each disease. Using data from each institution’s tumor registry and billing and medical records, MSK assesses each Alliance member to identify practice gaps and practice variation within the system and uses this information to inform strategies to address these gaps. The SoCs are assessed in conjunction with MSK resources and capabilities (R&C) that explore staffing, operational, and infrastructure requirements across multiple departments, including radiology, nursing, pharmacy, medical physics, pathology, and research administration. MSK has continually refined the R&C and SoCs to reflect changes in practice and evidence, enhance measurability, and incorporate feedback and improvements from Alliance physicians. MSK and the Alliance members reassess alignment with the SoCs and R&C on a periodic basis.

Given the multifaceted nature of the Alliance relationship, Alliance leadership was interested in obtaining insight from the members’ physicians into the specific components they found most valuable to their cancer-related clinical practice. Alliance leadership administered a quality improvement survey to all member sites with the intent of identifying opportunities to expand engagement across the Alliance. We report the findings from this survey with the objective of contributing insights into relationships involving academic medical centers and community health systems.

Methods

Results

A total of 103 clinicians across the 3 Alliance member sites completed the survey: 58 from HHC, 27 from LVCI, and 18 from MCI. There was representation from clinicians practicing in each of the 8 disease sites for which MSK SoC and other activities had been developed (Table 1). Analysis was limited to physicians reporting at least 1 disease management team affiliation in order to focus on clinicians for whom the activities in the survey were likely relevant. The final analysis included 87 respondents. Results for individual questions are included in supplemental eAppendix 2.

Table 1.

Respondents’ Institution and Disease Management Team Affiliations

View Table

Most respondents were aware of opportunities to participate in Alliance activities; awareness ranged from 90% for observing and presenting to MSK tumor boards to 62% for MSK disease-specific retreats (Figure 1). Most respondents considered the following valuable to their practice: observing MSK tumor boards (94%), presenting prescheduled and urgent cases to MSK tumor boards (91% and 91%, respectively), and MSK lecture series (96%), among those who reported them applicable. Accessibility remained an area for improvement: 79% of respondents considered MSK tumor boards accessible, and 70% felt the same about the lecture series. Across all respondents, most reported satisfaction with engagement opportunities regarding observing MSK tumor boards (79%), MSK participation in their institution’s meetings (76%), meetings about the MSK SoCs (75%), and presenting to MSK tumor boards (72%) (Figure 1). Smaller proportions of respondents reported satisfaction with meetings about clinical research (57%) and MSK service retreats (41%), although the reasons for dissatisfaction are unclear. When asked where they would like to see increased engagement, the most commonly reported response was for more lecture series (45%). Write-in suggestions included focusing on expanding the scope of Alliance activities to include other tumor sites and increasing opportunities for one-to-one physician discussions.

View Full Size

Reported awareness and satisfaction with MSK Cancer Alliance activities.

Abbreviation: MSK, Memorial Sloan Kettering Cancer Center.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 17, 9; 10.6004/jnccn.2019.7295

Most respondents (n=77; 92% of those who responded to the question) reported that either their practice or the practice of other providers in their institution had changed because of the MSK Alliance. When asked a follow-up question of which interactions with the Alliance most contributed to practice change, of those who responded, 63% attributed practice change to the SoC assessment and alignment process, followed by individual physician-to-physician interactions (30%) (Table 2). Several responses referred to specific changes in practice related to discrete SoCs, such as, “We started biopsying [sic] metastatic site in patients presenting with de novo metastatic breast cancer and routinely started doing [fine-needle aspiration] of axillary lymph node in [patients] with breast cancer and clinically positive lymph nodes receiving neoadjuvant chemotherapy,” and, “More rapid and global adoption of certain treatment regimens such as hypofractionated breast [radiotherapy].”

Table 2.

Practice Change Attributed to Alliance Membership

View Table

Discussion

Survey findings revealed lessons relevant to practitioners participating in models that establish a connection between community health systems and academic medical centers. Importantly, the activities most valued by community physicians were heavily physician relationship–based. The value of physician-based relationships cannot be underestimated; they have enabled ad hoc collaborations, ease of obtaining second opinions, and bidirectional information sharing. Such relationships require buy-in and time from physicians across institutions. These findings support the ability of this relationship-centric model to be used to improve care without direct financial incentives or penalties.

There were several encouraging results from the survey. The respondents were largely satisfied with Alliance activities, such as MSK physicians attending Alliance member tumor boards. Most respondents reported practice changes, either for themselves or for their institution, and many attributed this practice change to the SoCs. Development of the SoCs and the direct physician-to-physician activities clearly required a substantial time commitment, but the survey results suggest that this commitment is valued by the intended audience. It will be important for future research to assess whether the reported practice change translates into measurable and objective improvements in care quality, outcomes, and costs. Given that retrospective analyses have reported variation in practice and outcomes depending on setting of care and type of hospital, evaluating whether outcomes improve for hospitals after the establishment of these relationships will be instructive for patients, community and academic hospitals, and other stakeholders.^15,16

Survey results led both MSK and Alliance members to reevaluate opportunities to enhance engagement across the Alliance and improve communication of existing opportunities to Alliance members. For example, MSK leadership added several new lecture series and improved the lecture streaming software to increase virtual accessibility. A total of 23 MSK clinicians participated in virtual Alliance lecture series in 2018, with topics ranging from the operationalization of cold caps to the use of bone marrow transplant in patients with follicular lymphoma. To expand engagement opportunities beyond physicians, MSK also made nursing grand rounds available remotely for Alliance clinicians and offered access to on-site nursing education events both at MSK and locally. To further facilitate in-person interaction between physicians at MSK and Alliance members, in 2018 a total of 39 MSK physicians participated in 26 speaking engagements at Alliance sites, including informal meetings with local physicians, research discussions, tumor boards and case conferences, grand rounds, and patient education events. Finally, SoCs and R&C metrics are being re-collected, with a focus on areas in which an Alliance member was not in alignment with MSK practice at initial collection, the measurement method changed (eg, from interview to quantitative measurement), or the metric reflected new practice.

With regard to research, as of December 2018, MSK Cancer Alliance members have opened a total of 40 trials on site in multiple disease areas. A total of 243 patients have accrued locally to MSK therapeutic trials, with an additional 518 accruing to MSK-IMPACT (Integrated Mutation Profiling of Actionable Cancer Targets) for genomic profiling of their tumors and 301 to nontherapeutic studies. To further enhance these efforts, the Alliance’s Clinical Research Subcommittee created disease-specific working groups for direct discussion between investigators in 11 areas: breast, central nervous system, gastrointestinal, genitourinary, gynecologic, head and neck, and thoracic cancers; melanoma; hematology; radiation oncology; and phase I research. Between April and December 2018, these working groups reviewed 157 MSK trials to evaluate interest, patient population, and feasibility of opening MSK trials locally, fostering and expediting the identification of suitable protocols for collaboration.

Although our survey findings were valuable to Alliance leadership, they may not reflect the perceptions of physicians who did not respond. Respondents may have been disproportionately likely to have awareness of or active involvement in Alliance activities or to have strong opinions about the Alliance, potentially overestimating physician satisfaction. Regardless of the representativeness of the respondents, the survey achieved the Alliance leadership’s goal of obtaining practical insight into the relative value of the different activities and opportunities to increase engagement. Alliance leadership plans to assess the frontline clinician experience on an ongoing basis to ensure responsiveness to clinician needs and requests.

Conclusions

Given the proliferation of relationships between community health systems and academic medical centers across the United States, it is important to understand the types of models that will support improved outcomes in cancer care.^17,18 The experience of one such relationship, established through the MSK Cancer Alliance, suggests that activities involving substantial physician investment may be effective for promoting practice change in the context of cross-institutional relationships. Future research is needed to identify key components of these models that might lead to actual improved patient outcomes, especially as these models evolve.