Partnering With Payers for Success: Quality Oncology Practice Initiative, Blue Cross Blue Shield of Michigan, and the Michigan Oncology Quality Consortium

Short abstract

More than 16% of the total sites participating nationally in the QOPI survey are in Michigan. A significant component of the growth in QOPI participation in Michigan can be attributed to the involvement and quality improvement efforts of Blue Cross Blue Shield of Michigan.

Introduction

In 1999, the Institute of Medicine published Ensuring Quality Cancer Care, which defined elements of quality cancer care, described how quality could be measured, and documented gaps in the quality of care for patients with cancer.¹ The report recommended the creation of a quality monitoring system to report regularly on the quality of care for these patients. The National Initiative on Cancer Care Quality, initiated by ASCO in 2000, was the first systematic investigation of the quality of cancer care in the United States. The investigators found that initial management of patients with breast and colorectal cancers in the United States seemed consistent with evidence-based practice; however, substantial variation in adherence to some quality measures was observed. The methods used to study quality of care—identification of eligible patients, medical record abstraction—are similar to measures used in conducting clinical trials, and examining processes of care is analogous to clinical trial adherence measures.² Several studies outside the realm of oncology have linked process measures to health care outcomes.^3–6

The ASCO QOPI Program

In 2002, a group of ASCO volunteers developed the Quality Oncology Practice Initiative (QOPI), as described elsewhere in this issue. QOPI is a voluntary, practice-based program designed to assist medical oncology practices in assessing quality. QOPI was developed to provide a framework for oncology quality improvement on par with efforts already developed in diabetes care and cardiovascular surgery. In 2006, QOPI became available for use by all physicians who were members of ASCO.

QOPI assesses practice performance on a series of process of care measures that are indicative of quality. Each process measure is based on a quality-of-care process or published practice guideline, which, if followed, should translate into positive patient outcomes.⁷ Early results showed that practices that participated in QOPI demonstrated improved performance in process measures, with the greatest improvement in initially low-performing practices.⁸ In 2008, a feature was added to QOPI enabling participating practices to request that ASCO send verification of QOPI participation to applicable health plans that participate in the QOPI Health Plan Program. This voluntary program is intended to demonstrate to third-party payers a practice's commitment to self-assessment and clinical quality improvement.

Although ASCO created a quality improvement platform via QOPI that is relatively simple, objective, and evidence based, there are significant barriers to participation. Busy oncology practices must divert staff away from patient care for medical record review, data abstraction, and data entry. Oncologists may not be familiar with quality improvement principles or know how to use QOPI data to enhance quality. Practices incur nonreimbursable costs relating to data abstraction.

Even given these barriers, since its inception, the number of practices participating in QOPI has grown. In the spring of 2009, 248 practices from across the country participated in the QOPI survey, voluntarily reporting on 81 measures encompassing a total of 18,177 records. In the state of Michigan, the number of participating practices has grown more than 400%, from fewer than 10 in the fall of 2007 to 40 in spring of 2009. Michigan now constitutes more than 16% of the total sites participating nationally in the QOPI survey. A significant component of the growth in QOPI participation in Michigan can be attributed to the involvement of Blue Cross Blue Shield of Michigan (BCBSM).

BCBSM's Quality Improvement Efforts

BCBSM, a nonprofit health plan and a member of the Blue Cross Blue Shield Association, has an estimated 60% of the third-party insurance market for Michigan. As part of its business strategy and its social mission, BCBSM developed a program to partner with and reward physician organizations (POs) for improved health care delivery. This program, known as PGIP—Physician Group Incentive Program—was first established in 2005 and initially focused on primary care. Since that time, PGIP has begun to add initiatives for specialists—including oncologists. PGIP currently includes 7,618 physicians from 36 POs from across the state of Michigan and encourages information sharing and development of quality improvement initiatives that improve the state's health care system, without regard for the patient's third-party insurance. PGIP has supported and facilitated practice transformation using a wide variety of initiatives, rewarding the POs for improved performance in health care delivery. PGIP currently has 25 initiatives in which POs can choose to participate.

The establishment of physician-led collaborative quality initiatives (CQIs) has been another important feature of the BCBSM's Value Partnerships program that also includes PGIP. CQIs are multi-institutional quality improvement projects that are data driven, relying on comparative performance reports derived from clinical registries. CQIs focus on common and costly areas of surgical and medical care. Currently, there are nine CQIs financially supported by BCBSM.⁹ Six of these CQIs are hospital based and address a variety of clinical topics, including bariatric surgery, general and vascular surgery, cardiac catheterization, cardiac surgery, peripheral vascular interventions, and breast oncology. Another CQI addresses cardiac imaging and includes both hospitals and physician practices. The remaining two CQIs are deemed PGIP professional CQIs and involve statewide coalitions of physician practices and ambulatory clinics—one professional CQI addresses oncology and the other addresses therapeutic anticoagulation.

These Michigan CQIs have achieved dramatic improvements, using comparative performance reporting and structured quality improvement strategies in the context of a statewide consortium. Examples of their accomplishments include reducing postangioplasty complications and radiation exposure^10–11 and reducing complications after coronary artery bypass grafting.¹²

BCBSM and QOPI

The PGIP oncology initiative was formed in collaboration with ASCO. BCBSM chose to use ASCO's QOPI as the foundation for the PGIP oncology initiative because it was an established and nationally validated program. BCBSM recognized that the QOPI platform, developed by practicing clinicians, was an ideal foundation for quality improvement for oncology practices participating in PGIP. It is the expectation of BCBSM that with greater adherence to clinical practice guidelines and self-assessment using QOPI measures, improvements will result that positively affect quality of life, lead to a reduction in off-label drug use, and reduce health care costs—results worthy enough to warrant subsidizing participation. Although there are 14 health plans that have established a way of recognizing QOPI participation, BCBSM is the only health plan to provide financial support to help defray the cost of data management for oncology practices that participate in QOPI.¹³

As an initial step, Michigan oncologists who participate in the PGIP oncology initiative register their practice with QOPI and participate in the semi-annual (spring and fall) data collection and reporting that is required. Practices are required to opt into the QOPI Health Plan Program for disclosure purposes and select BCBSM as a payer that can be notified of the practice's participation in QOPI. To allow review of performance at the level of each PO, each practice shares QOPI data collection reports with the larger PO to which it belongs. POs are also required, as an expectation of participation, to attend quarterly PGIP meetings, which provide a forum for the exchange of ideas and sharing of best practices. The QOPI coordinating center has begun providing Michigan aggregate data in addition to national aggregate data.

BCBSM has developed a member attribution model to determine the amount of the payment to each PO. Current payments are approximately $3,000 or more per eligible physician in single or small practices, and adjusted so that large practices are not disproportionately rewarded for a body of work that is often similar to those of the smaller practices. The payments are intended to offset the cost of data abstraction associated with the QOPI survey. At this point, the support provided by BCBSM can be considered “pay for participation,” versus “pay for performance,” as payment is based solely on participation in the PGIP initiative along with active participation in QOPI, and not tied to outcomes associated with the data.¹³ BCBSM understands that improving health care quality is a continuum, starting with oncologists' willingness to participate in data collection, followed by sharing of best practices and collaboration, ultimately leading to process improvements with the goal of improving clinical outcomes.

QOPI in Michigan Oncology Practices

In Michigan, small and large private community practices and large academic practices alike demonstrated the feasibility of participation and utility of the QOPI survey tool. The University of Michigan Comprehensive Cancer Center, a large hospital-based academic medical center, reported its experience with five rounds of QOPI data collection with the conclusion that QOPI can be adapted for use in practice improvement at an academic medical center. This center demonstrated that sharing performance data with physicians was sufficient to change some aspects of physician behavior, whereas improvement in other measures requires structural practice changes.¹⁴ QOPI data abstraction for the University of Michigan was completed by tumor registry staff with the assistance of a clinical pharmacist. Time involved in data collection averaged just more than 60 minutes per medical record abstracted.

In smaller private practices, carving out the time for QOPI data collection and submission can be daunting. One PO, Oncology Physician Resource (OPR), successfully implemented a model to encourage QOPI participation and assist private practices with data abstraction. OPR practices agreed to allow the financial support payment provided by BCBSM to be pooled and managed by OPR to assure the data could be abstracted without affecting the daily operations of the oncology practices, the majority of which had fewer than three physicians. OPR's goal was to allow the practices to participate in a manner most appropriate for their particular circumstances. They were given the option to have their own office staff contract with OPR to complete the data management and abstraction training during off hours or allow a central data abstracting team from OPR to come to their offices to complete medical record abstraction. These teams were centrally managed, and often included personnel who managed the data collection tasks associated with clinical trial conduct. Data abstraction averaged 54 minutes per medical record in fall of 2008.

OPR oncology practices are using the data to drive changes within their practices. Several community practices demonstrated improvements of 20% to 30% in areas related to documentation of assessment and management of pain, discussion of purpose of chemotherapy treatment, and inclusion of cancer stage in physician notes. In addition, one practice, noting a low administration of antiemetics, implemented a standard order set, leading to a 40% improvement in compliance with this guideline. Nurses who participated in data collection reported that once care providers became aware of the areas being measured, the health care teams started discussing methods to ensure they could improve the quality of documentation and care. Members of the OPR PO demonstrated that community practices can obtain important quality data in a cost-effective manner and use the data to advance the quality agenda.

Because of its interest in quality improvement and its financial support for QOPI participation, BCBSM has been described as a “visionary health insurer” by Joseph Simone, MD, who first conceptualized QOPI.¹⁵ BCBSM is building on that visionary accolade with the establishment in July 2009 of a new professional CQI, the Michigan Oncology Quality Consortium (MOQC).

MOQC

The objective of MOQC is to develop a multipractice regional collaborative that seeks to improve the quality of cancer care. MOQC will help Michigan practices that have participated in QOPI move the process to the next level—using data to apply process improvements to enhance patient care quality. MOQC will also offer support and education to other oncologists in Michigan who have not yet been involved in QOPI, encouraging their future participation. MOQC will use QOPI survey data to inform practices, analyze variation between practices, identify best practices and opportunities for improvement, and develop specific quality-improvement initiatives and interventions that will demonstrate improved clinical quality and outcomes for patients receiving cancer care. The scope of work will integrate MOQC, BCBSM, ASCO's QOPI, and individual oncology practices and their associated POs. MOQC will serve as the data coordinating center and Michigan clinical champion for QOPI.

MOQC will collaborate with ASCO regarding use, analysis, and dissemination of the QOPI survey data, with the goal of developing Michigan-specific data reports and analysis to assess individual participating practice concordance with quality measures. MOQC will use the data, along with practice investigation, to identify best practices and opportunities for improvement. MOQC plans to develop forums for communication between the participating practices and convene regular meetings of clinical representatives from the practices to formulate quality improvement ideas and act as a quality improvement information clearinghouse. MOQC will use lean health care principles and continuous quality improvement techniques to guide quality improvement efforts and develop specific actions and process improvement steps that should help improve adherence to standards, thereby improving quality of care. The consortium will oversee and guide quality improvement efforts while providing practical support to POs to jumpstart quality initiatives.

In the most recent QOPI Measures Summary Report (spring 2009), each Michigan practice was able to compare itself with the Michigan aggregate and QOPI national aggregate data, including mean, minimum, and maximum scores. There are several core measures that demonstrate more than five percentage points in variation between the Michigan and national aggregate means, indicative of poorer concordance with these measures in Michigan. These include documentation of plan of care for pain; assessment of constipation at the time of narcotic prescription or after visit; documentation of plan for chemotherapy, including doses, route, and time intervals; appropriate completion of chemotherapy plan; and assessment of patient emotional well-being within 1 month of the patient's first office visit. Each of these areas of variation denotes a departure from the underlying standard corresponding to the measure, represents real issues in the quality of care provided to the affected patients, and provides MOQC with areas of focus for its initial investigation and ongoing collaborative activities.

Using the QOPI program developed by ASCO and engaging practices in voluntary quality measurement and improvement activities under the encouragement and with the financial support of BCBSM has led to preliminary evidence of process and quality improvement with minimal financial incentives. As this statewide consortium moves forward, we plan to optimize our measurement techniques, continue to minimize the financial and time burden on busy oncology practices, and work collaboratively to improve oncology care for all residents in the state of Michigan.

Authors' Disclosures of Potential Conflicts of Interest

Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.

Employment or Leadership Position: Philip J. Stella, Oncology Physician Resource (C); Thomas Ruane, Blue Cross Blue Shield (C) Consultant or Advisory Role: Douglas W. Blayney, Amgen (U), Bristol-Myers Squibb (U); Beth LaVasseur, Oncology Physician Resource (C) Stock Ownership: None Honoraria: None Research Funding: Douglas W. Blayney, Blue Cross Blue Shield, Amgen Expert Testimony: None Other Remuneration: None