New York State Department of Health SPARCS Training Improving and Expanding Race and Ethnicity Data Collection Albany, NY October 10, 2013

Agenda Topic Presenter Welcome and Introductions Eric Niehaus Vice President Healthcare Association of New York State Health and Health Care Disparities Race and Ethnicity Barbara A. Dennison, MD Director, Policy and Research Translation Unit Division of Chronic Disease Prevention Office of Public Health New York State Department of Health SPARCS Data Collection John M. Skerritt Technical Specialist SPARCS Operations Office of Quality and Patient Safety Questions & Answers Dr. Dennison and John Skerritt

Objectives Describe why improved race and ethnicity data will help in identifying disparities in health care quality. Identify national legislative/regulatory attention to race and ethnicity data. Describe steps to improve quality of data collection and expand race and ethnicity categories. Describe how to code the expanded race and ethnicity data categories. Thank you, Vice President Niehaus Good Afternoon. Thank you for joining the first in a series of five Webinars and a Webcast focused on increasing health equity. Improving the quality and granularity of disparities data, such as race and ethnicity, is the first step. In this Webinar, we will briefly discuss the rationale for paying increased attention to health care disparities and why NYS DOH, Am. Hospital Assn. IOM, AHRQ and other national organizations and experts are focusing on improving race/ethnicity data, using the stratified data to identify unequal health care, and then developing targeted interventions to reduce these disparities. We will also describe the technical and procedural steps needed to collect and code the enhanced race and ethnicity categorical data, and to update your IT system to capture and report these data.

Definitions Health Disparities: Differences in the incidence, prevalence, mortality, burden of disease and other adverse health conditions that exist among specific population groups.  Source: National Institute of Health Health Care Disparities: Includes differences in treatment provided to members of different racial or ethnic groups that is not justified by the underlying health conditions or treatment preferences of patients. Source: Institute of Medicine Although disparities in health and health care can be inextricably tied to one another, distinguishing between them is important to understand the complexity of the problem. Health Disparities are “Differences in the incidence, prevalence, mortality, burden of disease and other adverse health conditions that exist among specific population groups.” They are related to multiple factors such as education, geography, income distribution, social capital, etc…. In an ideal world, even if we were to eliminate disparities in health care, we would still see disparities in health. Whereas, Health Care Disparities include differences in treatment provided to members of different racial or ethnic groups that is not justified by the underlying health conditions or treatment preferences of patients. We’re focusing on disparities that exist within the health care delivery system because hospitals and facilities have control over what happens in their environments.

What are Disparities in Health Care Quality? Racial and ethnic minorities tend to receive a lower quality of health care than non-minorities Less likely to receive: Cancer screening Cardiovascular therapy Kidney dialysis Transplants Curative surgery for lung cancer Hip and knee replacement  Pain medicines in the ER The IOM published 2 sentinel works in the last decade: Crossing the Quality Chasm and Unequal Treatment. These reports demonstrate that although the U.S. spends billions on healthcare, there is wide variation in the care that Americans actually receive and the care they SHOULD receive. Not only are patients not necessarily receiving the right care, but racial and ethnic minorities are less likely to receive medical procedures, and experience a lower quality of health care and services, even when insurance status, income, age, and severity of conditions are comparable, and one adjusts for access-related factors, and patient preference. These reports find that minorities, compared with non-minorities, tend to receive lower quality of health care in a number of areas, including: cancer screening, cardiovascular therapy, and pain medicines in the ER. These disparities probably exist in your hospital, but you won’t know unless you look. 5

Unequal Health Care The health care system contributes to disparities in care: Increased medical errors Prolonged length of stays Avoidable admissions and readmissions Over and under-utilization of procedures Studies also show disparities in other areas – such as length of stay, hospital readmissions, and procedures – ---these are factors that can affect the bottom line. ---which supports the case that equitable, high quality care is not just the right thing to do, but it’s also good business practice. According to a recent Agency for Healthcare Research and Quality (AHRQ) report, 60% of disparities in the quality of have care have stayed the same or worsened for Blacks, Asians, Hispanics and for low-income populations. Source: Institute of Medicine. (2002). Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care. Washington, DC: National Academy Press.

Growing U.S. Minority Population Population Projections, 2010 to 2050 This graph provides information about our changing national demographics. The U.S. Census Bureau projects that the number of minorities living in the U.S. (shown in blue) will increase and exceed the number of non-Hispanic whites (shown in red) as early as 2040.  Minorities will become the majority in America. This is one of the reasons why addressing disparities in health care is becoming even more pressing. Changes in the U.S. population already impact health care providers and the patients they care for, and it will continue. Have you noticed changes in your hospital/facility’s patient demographics? Is the change anecdotal or based on accurate patient data that you trust? Do you know how your hospital’s community demographics have changed in recent years? Source: U.S. Census Bureau, 2009 National Population Projections (Supplemental). Projections of the Population by Sex, Race, and Hispanic Origin for the United States: 2010 to 2050

Minority Groups Will Be Majority This shows expected changes in the U.S. population between 2000 and 2050. There is expected to be a doubling in the percentage of Hispanics, shown in green (13% -> 24%) Asians, shown in purple (3.8% -> 8.0%) With a concomitant reduction in the percentage of Americans who are NH, white. Source: Eliminating Disparities: Why It’s Essential and How to Get It Done, American Hospital Association.

New York Population, 2012 This slide shows the population distribution in NYS, separately by race and ethnicity, so one can't strictly compare to the previous slide. Compared to the U.S. population, NYS has: More Hispanics ( 18% vs. 15%) More Asians (8% vs. 5%) Fewer Native Hawaiian & Other Pacific Islander (0.1% vs. 2%)

Increasing Legislative and Regulatory Attention to Race and Ethnicity Data American Recovery and Reinvestment Act of 2009 To be eligible for “meaningful use” incentive payments Patient Protection and Affordable Care Act of 2010 If receiving federal money Revised Joint Commission Standards - 2012 New requirement NYS SPARCS – All discharges effective January 1, 2014 Expanded race and ethnicity categories (CDC Race and Ethnicity Code Set - Version 1.0) Increased attention to data quality Besides shifting demographics, there are also legislative and regulatory forces that will impact your organization and provide additional reasons to take action. There have been 3 major national movements over the past few years related to health disparities and race, ethnicity and language data collection. 1) The American Recovery and Reinvestment Act of 2009—better known as the stimulus bill—created financial incentives for physicians and hospitals to use electronic health records (EHR); the collection of race, ethnicity and language information were included in “meaningful use” criteria. 2) The health reform bill —the Patient Protection and Affordable Care Act of 2010—(also known as Obama Care) requires health care organizations that receive federal funds to collect self-reported patient race, ethnicity and language data. The Joint Commission revised its accreditation standards. Previously, organizations were required to provide written information appropriate to the language of the patient, and to provide interpretation services as necessary. The new standards, which began in 2012, require hospitals to collect and record in the patient’s record, information about the patient’s race, ethnicity, language and communication needs.  NYS SPARCS – requires Hospitals and facilities Expand the race and ethnicity categories, effective January 1, 2014 In addition, there is increased attention to the quality and accuracy of the data collected and reported

“Although the collection of race, ethnicity and language “Although the collection of race, ethnicity and language* data does not necessarily result in actions that will reduce disparities and improve care, the absence of the data guarantees that none of that will occur.” While these national acts require collecting a patient’s preferred language, in addition to their self-reported race and ethnicity, it should be pointed out, that NY legislation requires race and ethnicity data be reported as part of SPARCS. But, at present, there is no such requirement for reporting preferred language. Of note, several other states have recently passed legislation/regs that require language data also be submitted. However, just collecting the data, won’t improve health care. One needs to use the data to identify populations at risk of receiving lower quality care within and across health care organizations. Source: IOM (Institute of Medicine). 2009. Race, Ethnicity, and Language Data: Standardization for Health Care Quality Improvement. Washington, DC. *Note: The language reference is part of a direct quotation. SPARCS collects race and ethnicity data, but not language data.

Three Steps to Address Health Disparities Standardize collection of self-reported race and ethnicity data Stratify and analyze performance measures by race and ethnicity Identify and develop quality improvement interventions targeted to specific patient populations The IOM identified 3 steps to address issues of equity and health disparities within a quality improvement framework. First, standardize the collection of data by using the same categories throughout the organization and allowing the patients to self-report.  No more just looking at the patient and guessing. The second step is to stratify performance measures to determine if and where disparities exist. Third, is to use the stratified data to drive quality improvement efforts. In this webinar series we will discuss the standardized collection process, but this is just the beginning of addressing disparities. In the Public Health Live Webcast on Oct. 17th, Commissioner Shah, national experts, and NYS hospital leaders will outline the charge to hospitals and make the business case for improving data, quality and equity of healthcare to improve health outcomes. Subsequent Webinars will address staff training, and provide examples of how hospitals have improved collection of patient's race and ethnicity, analyzed their data, and improved care for all their patients.

U.S. Hospital Survey 82% of hospitals collect race and ethnicity data, but… Categories vary within and across hospitals Staff collect data mostly by observation Staff at some hospitals trained to “not ask” Most hospitals do not use data for quality improvement Only 17% use data to assess and compare health outcomes among different patients A recent U.S. Hospital Survey found that most hospitals collect race and ethnicity data, but only a small percentage use the data to assess differences in care, and then to make improvements. In part, because they may doubt the quality or validity of their data Many Staff report entering information based on appearance, educated guesses or secondary sources such as identification documents. Staff may be reluctant to ask patients their race and ethnicity 47% Patient self-reports only 41% Registrar observation AND patient self-reports 8% Registrar observation only Source: Hospitals, Language, and Culture: a Snapshot of the Nation, 2010 N = 60 U.S. Hospitals

Some Anticipate Obstacles to Modifying Registration IT System Information technology Training/educating staff Time Costs 45% do not anticipate any obstacles This survey found that half of hospitals anticipated obstacles to modifying their registration IT system. The 4 most common concerns were: 1) information technology, 2) training/educating staff, 3) time and 4) costs. Of all the barriers listed, training/educating staff can be the most challenging – but not insurmountable. And, as many of you know, training is an ongoing process. One of the upcoming Webinars will focus on staff training and provide resources, toolkits and materials.

Registration Staff Face Challenges/Barriers to Race and Ethnicity Data Collection Patient reluctance to provide the data Staff reluctance to ask the questions Inability of staff to communicate in patient’s preferred language Lack of staff training on data collection Potential barriers faced by registration staff in collecting patient race and ethnicity data include: Presumed patient and/or staff reluctance, cited by 55% of hospitals. Lack of staff training in collecting R/E data And, difficulty communicating to patients in their preferred language .

Barriers/Challenges to Using Race and Ethnicity Data Accuracy of the data Lack of consistent data collection process Lack of standardized data categories Data systems and integration with QI practices 41% reported no barriers or challenges Potential barriers to using patient race and ethnicity data include concerns with: the accuracy of the data currently being collected the lack of a consistent data collection process not having/using standardized data categories how to integrate the race and ethnicity data with current data systems and quality improvement practices

NYS Assessment of Hospitals and Ambulatory Surgery Facilities Policy or Procedure for collecting patient race and ethnicity information (74%) Provide staff training on R/E data collection (78%) Registration/Admissions Supervisors (13%) Outpatient Registration Staff (27%) Ambulatory Surgery Admissions Staff (24%) ER Registration Staff (21%) Admissions Clerks (32%) Registration Clerks (35%) Training offered only once to new employees (65%) In August 2013, a survey was sent to 20% of hospitals facilities and ambulatory surgery center in NYS (N=63). The response rate was 75%, thanks to you who completed our survey, and to the persistence of our colleagues at the SPH. The survey found that: Most facilities have a policy/procedure for collecting race. 40% collected Race and ethnicity data at every visit and 40% collected R/E data only during the patients’ initial visit. Most hospitals provided training, though 65% only provided training to new employees. The type of staff trained varied across hospitals and facilities, though training of admission and registration staff were the most frequently reported.

NYS Assessment of Hospitals and Ambulatory Surgery Facilities Frequency of collecting patient race and ethnicity information: Initial Visit (34-40%) Every Visit (42-53%) Don’t Know (5-25%) Method of collecting patient race and ethnicity information: Verbally asking the patient questions (87%) Getting patient information from existing records (53%) Having the patient fill out a form (40%) Observing the patient’s physical characteristics (34%) 34-40% facilities indicated that race and ethnicity data is collected at every visit , 42-53% during the patients’ initial visit. Most hospitals/facilities reported they ask the patient questions, but 34% reported observing the patient. 40% of facilities have patients fill out a form and In 53%, the information comes from existing records

NYS Assessment of Hospitals and Ambulatory Surgery Facilities Reported barriers to collecting patient race and ethnicity information: Patient declines to respond (64%) Staff have not been trained (15%) Question too sensitive (53%) Language or communication barriers (39%) There is not a good opportunity to collect (16%) Not enough time to collect (17%) No method to collect this information (2%) The NYS Assessment of Hospitals and Facilities found similar barriers to collecting R/E data as did the U.S. hospital survey. Lack of staff training was less commonly reported. Patient refusal, language barriers, and questions perceived as too sensitive were commonly reported. These issues will be briefly addressed in next week’s Webcast and then later in more detail in a Webinar specifically designed for Registration/Admission Supervisors and staff.

Components of Standardized Race and Ethnicity Data Collection Use standardized categories across the organization Ask patient to self-report ethnicity, then race No more “eyeballing” the patient Data are collected from all patients Tell the patient why we are collecting his/her race and ethnicity and how the information will be used What are the basic components of standardized data collection? The same categories for patient race and ethnicity should be used across the hospital, system or medical group.  The patient needs to be asked to self-report his/her ethnicity, and then to self-report his race. And every patient needs to be asked – at least once, not necessarily at every visit. The patient also needs to be told why the information is being collected, and how the information will be used.

Article 28 SPARCS Timeline May 31, 2013: Letter to Article 28 facilities from Office of Quality and Patient Safety. July 1, 2013: Health care facilities may begin submitting files in the expanded format. All discharges effective January 1, 2014 Facilities must be fully transitioned to collect and report the expanded race and ethnicity categories. The Timeline: NYS Article 28 Facilities, that are required to report SPARCS data, were notified of the changes May 31, 2013 -- 6 months prior to the effective date – All discharges on or after Jan. 1, 2014, need to report the expanded race and ethnicity data. On July 1, 2013, changes were made such that facilities can begin submitting and testing race and ethnicity files using the new format.

Key Decision Points Who needs to be engaged? What system modifications need to be made? How will the registration process change? How will staff be trained on the new collection procedures? How will you monitor the data to ensure completeness and accuracy? As hospitals and other facilities set out to change your data collection system, there are many things that need to be considered. Some of the key decisions include: Deciding who needs to be involved, and then including all the players Deciding what systems need to be modified If your outpatient and inpatient health records are linked, and data will populate one system from the other, it may be more efficient in the long run, to change your entire system. Will there be changes to your registration or admission process? How and which staff need to be trained? With what frequency? How will you monitor race and ethnicity data and ensure that it is being collected correctly – i.e., patients are being asked? Data are being entered properly? That you don’t have high rates of unknown or other race or ethnicity?

Quality Improvement Requires high-quality data. First Step: Helping hospitals gather data on patient race and ethnicity to obtain a more accurate and complete picture of their patients. Second Step: Use data to critically examine care delivered to learn whether they are providing equitable care. Third Step: Design Quality Improvement efforts to improve quality of care and reduce disparities. Source: Robert Wood Foundation, Expecting Success: Excellence in Cardiac Care Program Health Care organizations must have good quality data on the race and ethnicity of those they serve in order to identify health care disparities, and then to make changes to ensure that they are providing high quality care to all. As mentioned, the upcoming Webcast and Webinars will be discussing Steps hospitals can take, and the advantages of doing so. Now, I’d like to turn this over to John Skerritt, who is a Technical Specialist, SPARCS Operations, NYSDOH, who will explain some of the more technical details of submitting the expanded datasets to SPARCS

SPARCS Ethnicity Standards Are you Hispanic, Latino/a, or Spanish origin? (One or more categories may be selected) Current Data Standard Expanded Data Standard X12 Value Ethnicity E1 Spanish/Hispanic Origin   E1.02 Mexican, Mexican American, Chicano/a E1.06 Puerto Rican E1.07 Cuban See: SPARCS Appendix RR for list of codes (CDC Code Set) Additional Hispanic, Latino/a, or Spanish Origin categories E2 Not of Hispanic, Latino/a, or Spanish origin E9 Unknown

Expanded Data Standard SPARCS Race Standards Race Standards What is your race? (One or more categories may be selected) Current Data Standard Expanded Data Standard X12 Value Race R1 American Indian or Alaska Native R2 Asian   R2.01 Asian Indian R2.06 Chinese R2.08 Filipino R2.11 Japanese R2.12 Korean R2.19 Vietnamese See: SPARCS Appendix RR for list of codes (CDC Code Set) Additional Asian categories

Expanded Data Standard SPARCS Race Standards Race Standards What is your race? (One or more categories may be selected) Current Data Standard Expanded Data Standard X12 Value Race R3 Black or African American R4 Native Hawaiian or Pacific Islander   R4.01.001 Native Hawaiian R4.02.001 Guamanian or Chamorro R4.01.002 Samoan See: SPARCS Appendix RR for list of codes (CDC Code Set) Additional Pacific Islander categories R5 White R9 Other Race

How Will SPARCS Collect Data? X12-837, Version 5010 format Repetition separator in ISA DMG segment format Edit reports Data dictionary SPARCS Appendix RR (CDC Race and Ethnicity Code Set - Version 1.0) http://www.health.ny.gov/statistics/sparcs/sysdoc/apprr.htm

SPARCS Data Collection The X12-837 file: Using Version 5010R: this is the only format supported. The race and ethnicity data elements are collected in the DMG segment and make use of the repetition separator. ISA 11 segment must contain the same character as DMG 05 separating multiple race values. ISA*00* *00*….. *^* <- ISA 11

SPARCS Data Collection The DMG segment contains the race and ethnicity information. The race and ethnicity can repeat; up to 10 total in the segment. The repetition separator is used to identify each unique value. DMG*D8*20130115*F**RET:R2.02^:RET:E5* <- DMG 05

SPARCS Data Collection SPARCS edit reports have an error code for race and ethnicity: 2010DMG5000 RACE and ETHNICITY CODE Most errors to date have been missing the repetition separator or missing the values completely

SPARCS Data Dictionary

SPARCS Appendix RR

New York State Department of Health Resources Data Dictionary Race and Ethnicity Addendum Pages: http://www.health.ny.gov/statistics/sparcs/sysdoc/race_ethnicity_072013.pdf Appendix RR: http://www.health.ny.gov/statistics/sparcs/sysdoc/apprr.htm Frequently-Asked Questions: http://www.health.ny.gov/statistics/sparcs/faqs/#ERE

Staff at the Facility Get everyone at the facility on board from the top down. Standardize the collection process: Patient should self-identify/report Report ethnicity(ies) first, then race(s) Data are collected on all patients Review in-house security to protect data. Train all staff to collect data and answer the patient with the same response.

Patients at the Facility Tell patients you are collecting the information before you collect it and explain why. Create forms, so they can self-identify. Assure them the data will be protected. Engage the community.

Resources NYS Toolkit to Reduce Health Disparities: Improve Race and Ethnicity Data Health Research and Educational Trust (HRET) Toolkit: On-line resource to help hospitals and facilities systemically collect race and ethnicity data from patients: http://www.hretdisparities.org

Next Steps Collecting the Data: First Steps in Achieving Health Equity October 17, 2013, 9-10:30 a.m. http://www.phlive.org Several Webinars for: Physicians, Hospital Executives, Quality Improvement Advisors, and Medical Staff Registration and Admission Supervisors and Staff Community-Based Organizations and Community Leaders NYS Toolkit to Reduce Disparities: Improving Race and Ethnicity Data Collection

Our Goal… Improve the quality of race and ethnicity data collected. Expand the granularity (number of categories) of race and ethnicity data. Questions? To truly eliminate disparities in health care, there must be fundamental changes to our health delivery system.

SPARCS Operations John Skerritt, Trainer Website: http://www.health.ny.gov/statistics/sparcs/ E-mail: sparcs@health.state.ny.us Phone: (518) 473-8144 Fax: (518) 486-3518