Measuring Value and Outcomes for Continuous Quality Improvement. Noelle Flaherty MS, MBA, RN, CCM, CPHQ 1. Jodi Cichetti, MS, RN, BS, CCM, CPHQ

Transcription

1 Noelle Flaherty MS, MBA, RN, CCM, CPHQ 1 Jodi Cichetti, MS, RN, BS, CCM, CPHQ Leslie Beck, MS 1 Amanda Abraham MS 1 Maria Uriyo, PhD, MHSA, PMP 1 1. Johns Hopkins Healthcare LLC, Baltimore Maryland Corresponding Author: Noelle Flaherty, MS, MBA, RN, CCM, CPHQ Biographical Sketches: Noelle Flaherty, MS, MBA, RN, CCM, CPHQ is Director of Quality Improvement at Johns Hopkins HealthCare LLC. (JHHC). Ms. Flaherty earned her BA from Bryn Mawr College, BS from Johns Hopkins School of Nursing, MS in Health Services Leadership and Management from the University of Maryland School of Nursing, and MBA from the University of Baltimore. Jodi Cichetti, MS, RN, BS, CCM, CPHQ is the Senior Director of Quality and Clinical Improvement, WellSpan Health System, and past Senior Director Medical Management, Johns Hopkins Health System. Leslie Beck, MS, is the Health Services Project Manager at JHHC. Ms. Beck served as Co-Chair on the JHHC Diversity Management Leadership Committee and is a member on the Johns Hopkins Berman Institute of Bioethics Patient Advisory Group. Ms. Beck holds a Bachelor of

2 Science degree in Business Management and a Master of Science degree in Healthcare Management from the Johns Hopkins University Carey Business School in Baltimore, Maryland. Amanda Abraham, MS, is the Data Analyst for Quality Improvement at JHHC. She received her Bachelor s degree from Penn State University in Health Policy & Administration and a Master s degree from George Mason University in Health Systems Administration with a concentration in Executive Management Her primary role as an analyst is to provide actionable data on the Value-Based Purchasing project and Medicaid line of business (Priority Partners MCO) to the administrative team. Maria Uriyo, PhD, PMP, is the Project Manager for NCQA Accreditation at JHHC. She received her PhD in Food Science, Master s in Food Chemistry from Virginia Polytechnic Institute & State University and a Masters in Health Systems Administration from Georgetown University. In her current role Dr. Uriyo manages the NCQA accreditation process for JHHC. Acknowledgement: The Quality Improvement Team developed the framework under direction of Dr. Chester Schmidt, Chief Medical Officer, JHHC. Disclosures: No grants or external support funded this work. 2

3 Abstract Quality of care is critically important to health plans, systems and care providers striving for a healthier population at a decreased cost. Prioritizing action related to population health and quality of care is a challenge for leaders in the healthcare industry. Poor performance on quality outcome measures can cost health plans and providers millions of dollars in missed revenue, fines, penalties and other related health care costs. Challenges for the health care industry include hundreds of National Quality Forum (NQF) endorsed measures, variations between government, payer and provider quality indicators and related definitions, as well as current or future financial penalties linked to different measures by regulatory and accreditation bodies. In response to the need for meaningful and actionable health care initiatives, the Johns Hopkins HealthCare LLC Quality Improvement (QI) Department developed a framework for quality project planning based on a mathematical model in order to easily identify, forecast and target population health measures. This effort is in alignment with Johns Hopkins Medicine s mission to continuously reduce preventable harm, improve patient outcomes and enhance the value and equity of care around the world by advancing the science of patient safety and quality through discovery, implementation, education, evaluation, and collaborative learning. Keywords: Value Based Purchasing; Performance Improvement; Pay for Performance (P4P) Introduction Johns Hopkins HealthCare LLC (JHHC) administers health care benefits and services for over 408,000 lives enrolled in managed care organizations, government and employer sponsored programs. JHHC supports plan members through added benefits, including outreach, disease 3

4 management, complex case management and health education. Quality Improvement (QI) Program activities support and promote the JHHC mission to improve the lives of our plan members by providing access to high quality, cost effective, member-centered healthcare. Additionally, the JHHC QI program supports the Johns Hopkins Medicine (JHM) mission to improve the health of the community and the world by setting the standard of excellence in medical education, research, and integration. The QI Department works collaboratively with Johns Hopkins entities to ensure that members are receiving exceptional health care. Annually, health outcomes for our covered members are measured through the Healthcare Effectiveness Data and Information Set (HEDIS ). Based on the HEDIS results, the JHHC Quality Improvement (QI) Department will propose projects to improve patient outcomes. A standard work method was needed to efficiently and effectively focus and allocate resources to measure directed projects that will improve care quality, and also provide optimal impact to quality ranking results for the health plan. In response, the QI Department developed a framework for quality project planning based on a mathematical model to easily identify, forecast and target population health quality measures. Healthcare outcomes and patient/member satisfaction data are important indicators in measuring care quality, and when monitored with due diligence, support good financial stewardship. Quality measurement and improved outcomes are a high priority due in part to the importance of care quality for the patient population, and also to the potential financial impact of pay for performance programs including value based purchasing (VBP), the Centers for Medicare and Medicaid Services (CMS) Five-Star Quality Rating System (Stars), and other pay for performance (P4P) programs. From a financial perspective, the return on investment for 4

5 improving health outcomes and patient/member satisfaction can be estimated in the millions of dollars for some organizations. Health care systems and leadership are challenged with developing strategies for the management of quality improvement projects due to the sheer number of quality measures and related data metrics, and the competing priorities for resources in the healthcare environment. The National Quality Forum (NQF) has endorsed over 700 measures 1 that have undergone peer review, and are considered clinically relevant to population health and quality of care. Measure types include process, outcome, intermediate clinical outcome, efficiency and cost/resource utilization. 1 Quality measures that are used by health plans and federal and state agencies are based on NQF endorsed measures. For example, the Maryland Department of Health (MDH) has a VBP program for HealthChoice, which is the Medicaid managed care program 2, ten (10) of the thirteen (13) VBP measures are NQF endorsed measures, while the other three (3) were developed by the MDH. In addition to the NQF measures, the Agency for Healthcare Research and Quality (AHRQ) National Quality Measure Clearinghouse includes over 2,000 measures, including some measures that overlap with NQF. 3 In order to manage the volume and variety of measures, a standard framework is needed to support a focus on clinically relevant opportunities with the greatest positive impact on population health. 4 Meltzer & Chung evaluated thirteen (13) AHRQ quality measures and proposed a framework for prioritization called net health benefit. This clinical framework focuses on the evaluation of cost and population health outcomes. Porter s model is another method for prioritization that focuses on the value for the patient 5. Value is quantified through an analysis of health outcomes relative to costs. Similar to the model proposed by Meltzer & Chung, Porter s model is population health based and clinically driven 5. The framework 5

6 developed by our health plan includes the clinical and cost elements, but also includes administrative and effort scoring with the goal of the best impact on patient health, patient experience and cost of care. Methods In 2012, the QI Department analyzed patient outcomes data as defined by HEDIS and identified many opportunities for improvement. The QI Department had limited resources to impact all of the measures, and identified a need to develop a targeted approach to quality improvement initiative planning. The measures that were targeted by the QI team were specific to the health plan accreditation requirements from the National Committee for Quality Assurance (NCQA ), which is a key indicator of health plan excellence. The QI team created an initiative planning framework to focus on decision making and appropriately allocate resources. Institutional Review Board Approval (IRB) was not required for the framework development or associated data analysis. The framework developed by the QI Department has three (3) phases, and includes consideration of multiple factors, including national and regional benchmarks, administrative effort, individual circumstances of the business, and variations in population demographics. This framework includes tools for quality improvement reporting, planning and intervention. The framework is cost effective and does not require technology beyond standard data collection of quality measures. In the first phase of the framework, previous year results are used to prioritize measures with the greatest opportunity for improvement when compared to national or regional benchmarks. The evaluation of previous year results includes the following: 6

7 1. Basic population evaluation components: age, gender, geography, and ethnicity. 2. Determine multifaceted population evaluation components identified in previous quality program evaluations, targeting best practices and opportunities for improvement. 3. A barrier analysis. 4. Identify population targets by geography and population size. 5. Literature review of best practices. Figure 1: Example of an evaluation of a behavioral health measure After completing the evaluation and barrier analysis, measures are color coded to highlight opportunity for improvement. Green indicates maximum improvement opportunity. Yellow is the next best point, and red indicates minimal movement or need to maintain. Minimal or need to maintain usually indicates measures that are in the goal range (threshold or target when compared to benchmarks). The use of color coding and benchmarks in this first phase are shown in the example in Figure 2. Although NCQA benchmarks are used in the example; other national or regional benchmarks can be applied using the same process. The color coding can be modified to meet the business or population health requirements of the organization. 7

8 Figure 2 HEDIS Measures 2017 Final HEDIS Rate 2017 NCQA National Percentile Benchmark Threshold Additional 2018 HEDIS Points (if next benchmark attained) Breast Cancer Screening th Chlamydia Screening in Women th Follow-up After Hospitalization for Mental Illness - within 7 days th After the measures are color coded, the methodology focuses on measures that are identified as having the best or next best opportunities for improvement (color coded green or yellow). The JHHC QI department may also decide to focus on high performing measures (color coded red) if there is a business reason (P4P) or population health rationale to focus on those measures. During phase two of the methodology, a percentage to goal value is calculated for each measure using current year (prospective) data to identify likelihood of success based on current data. The percent to goal (PG) of a measure moving to the next threshold percentile level is calculated using the following formula: PG= CCR/NBPTR. In this formula, PG is the percent to goal; CCR is current compliance rate and NBPTR is the national benchmark percentile target rate. This calculation helps focus attention to the amount of effort that will be needed to meet the next percentile benchmark. Some measures might be identified as potentially reaching the target based on little to no effort. Figure 3 is an example of the second phase of the framework for CY 2018 planning. The measures in this example have demonstrated improvement, so a control plan can be put in place for these measures and other measures can be targeted in C

9 Figure 3 During phase three, the QI Department meets internally to review the priority measures identified in the first two phases. The QI Director then meets with business leaders, physicians and other stakeholders to learn about any business needs or strategies already underway that could positively (or negatively) impact the quality outcomes for the current year. Factors to consider prior to scoring measures include specifics related to the financial history of project planning, patient/member population, company workforce, and historical quality approach. After gathering input from the QI team, business leaders and other stakeholders, the QI Director scores each measure using four pre-defined categories: 1) administrative effort; 2) population impact / relevance; 3) reporting requirements; and 4) anticipated expense. Definitions of the categories are as follows: 1) Administrative effort is defined as the level or intensity of the total business work effort related to quality or care management programs; 2) Population impact or relevance is defined as the number of opportunities within the measure or the relevance assigned. The lower the point score, the higher the anticipated impact on the population or higher relevance; 3) Reporting effort is defined as the level of complexity for ongoing monitoring and evaluation of effectiveness of quality projects; AND 9

10 4) Anticipated expense includes projected costs for the quality program and potential financial risks such as VBP penalties, potential loss of contracted members, or other regulatory fines. The amount of expense is compared with the potential financial gain or loss. Each category is scored on a scale of 0-5. Effort scoring is a total cumulative score (maximum of 20 points) where the higher point value represents the higher/heavier work effort comparatively for corporate consideration and project planning. Measures identified as having a lower point value may be selected for rapid cycle quality improvement projects. Measures with a higher point value that are selected for quality initiatives may need more time for planning and budgeting to successfully manage the work effort. The cumulative score can be considered in addition to the other factors in determining the most valuable quality measure for corporate focus and quality planning for improved clinical and reported patient outcomes and enhanced value. Figure 4 is an example of the third phase of the framework that was used for initiative planning in CY Figure 4 Results In 2012, the QI team used the framework and identified breast cancer screening as a measure for improvement for the commercial line of business. A women s health screening 10

11 brochure was sent to members with opportunities in 2011 and A significance validation was calculated for women who received health reminders. The evaluation of the members who received a mailing had a significantly better rate of compliance (p value <0.0001) than those who did not receive the mailing in Based on evaluation, the brochure mailings likely served as a benefit to remind the member to schedule the appointment for breast cancer screening. HEDIS scores increased for this measure from 70.82% (50 th percentile) in Calendar Year (CY) 2012 to 76% (75 th percentile) in CY The QI Initiative Project expanded in 2014 and 2015 and the plan has maintained 75 th percentile since CY 21013, indicating that a control plan is in place to maintain performance for this measure. Another measure that was identified for improvement using this framework was the quality measure for follow up care after a mental health hospitalization. A project was initiated in 2013, and was refined and updated in As a result of the ongoing focus on this measure, the results in 2016 demonstrated a statistically significant improvement (9.93%). This improvement also resulted in an increase in overall accreditation score for the plan, contributing to the US Family Health Plan ongoing Excellent accreditation. Figure 5 demonstrates the improvement in this measure over time and with a sustained effort to improve the coordination of care project for mental health hospitalization within seven (7) days. 11

12 Figure 5 In addition to the measure specific successes referenced above, overall quality results for JHHC health plans have been above national standards. Results for 2017 (CY 2016) that demonstrates the overall success of the framework for one of our programs includes: NCQA Excellent accreditation for US Family Health Plan, the highest level of health plan accreditation NCQA Rating 5 out of 5, indicating that the US Family Health Plan is a highly rated national plan. The highest possible percentile ranking (NCQA 90th percentile) for the following HEDIS measures identified through the framework: Cervical Cancer Screening, Comprehensive Diabetes Care, Follow-Up After Hospitalization for Mental Health 7 days 12

13 Ongoing compliance for the Breast Cancer Screening measure at the 75 th percentile level. Significant improvement (6.90 %) of compliance with chlamydia screening for women. Limitations The framework and associated tools are not automated and do not incorporate statistical analysis using large data sets. Access to quality data, including administrative, pharmacy, laboratory, claims, encounters and electronic data is needed to maximize the use of the framework. Changes within health care, including changes to clinical practice guidelines, quality measure specifications, payer plan benefits or decreased access to care can result in lower than expected results. Although leadership input is engaged for the effort scoring, strategic decisions may be made by leadership that could result in selection of quality measures that are not identified as priority through the framework. Discussion For most quality measures, the framework is applied to the current year using past results and a minimum of three months of quality data. When there is a new quality measure, or when there are significant revisions to the specification for an established measure, projected trended rate and percentage to goal are not applicable because there is no historical data available to calculate the projected trend rate. In these circumstances, the goal percentile ranking is listed as requiring the greatest effort. There are measures that require chart audit/abstraction of a sample population, so prospective administrative data is not available. An example is the HEDIS measure for controlling blood pressure 6, which cannot be easily measured for a health plan throughout the 13

14 year because the measure is based on a representative sample of the entire population, and measures the last blood pressure of the calendar year. For this type of measure, the projected rate is derived from the two basic assumptions that past patterns will persist into the future, and measurable fluctuations in past trends will recur regularly and can be projected into the future. Previous year s eligible reporting population and information on total plan membership is referenced. Conclusions The framework provides a standard approach to prioritizing quality improvement projects with the goal of improving patient outcomes. Executive decisions are made easier by clear recommendations from subject matter experts that are supported by data. The final product for executive decision making is an objective report inclusive of a pre-determined and consistent methodology, that once applied, supports quality improvement prioritization and focus for intervention. The model developed by the JHHC QI Department has been effective over the past five years, and quality measures identified through the framework have maintained or improved each year when the proposed projects were approved and funded. In addition to proven maintenance or improvement of quality measure performance (by measured year) identified by the framework, there is a correlated improvement in overall quality ranking. Implications The NQF has recognized the importance of developing meaningful measures as well as improving and prioritizing existing measures, as evidenced by the NQF Strategic Plan to answer an unmet need for NQF to lead, prioritize, and collaborate to drive measurement that can result in better, safer and more affordable healthcare for patients, providers, and 14

15 payers. 1 CMS has also recognized the need for a core measure set to focus on patient outcomes. CMS, in collaboration with America s Health Insurance Plans (AHIP), released seven set of quality measures in The work between CMS and AHIP is ongoing, but does not address the need for health plans, health systems and providers to prioritize quality improvement activities and related interventions to improve patient health care. The framework developed and implemented by the JHHC QI department supports both the NQF and CMS goals related to identifying key measures for prioritization to improve health outcomes. This framework can be easily modified for adoption for use in a variety of settings, including health plan/ payor, Accountable Care Organization (ACO), hospital and provider. References 1 National Quality Forum (NQF). Healthcare Measurement. Retrieved from on June 7th, Maryland Department of Health and Mental Hygiene. HealthChoice. Retrieved from on July 22, Agency for Healthcare Research and Quality. National Quality Measures Clearinghouse. Retrieved from on June 7th, Meltzer, D. O., & Chung, J. W. (2014). The population value of quality indicator reporting: a framework for prioritizing health care performance measures. Health Affairs, 33(1), Porter, M. What is Value in Health Care? The New England Journal of Medicine 363:26 (2015): National Committee for Quality Assurance (NCQA). HEDIS & Performance Measurement. Retrieved from on June 7th,

16 7 Centers for Medicare & Medicaid Services. (2016). CMS and major commercial health plans, in concert with physician groups and other stakeholders, announce alignment and simplification of quality measures, Retrieved from CMS.gov on July 22,