Appendix G. Kaiser Permanente: Evolution of Data Collection on Race, Ethnicity, and Language Preference Information Race, Ethnicity, and Language Data: Standardization for Health Care Quality Improvement IntroductionKaiser Permanente is nationally recognized for its innovation and groundbreaking initiatives in health care. Over the years, we have continuously assessed the needs of our membership and public community, and proactively designed programs and solutions to meet the new challenges. Race, Ethnicity, and Language Data Collection underscores our commitment toward the elimination of health disparities and inequities for racial, ethnic, and linguistic groups.Kaiser Permanente PolicyWith the advocacy and leadership of our National Diversity Council, the Member Demographic Data Collection Initiative on Race, Ethnicity, and Language preference information became an organizational policy in 2004.The objectives of the policy extend beyond collecting information for compliance reasons; it is intended to inform patient-centered care, culturally and linguistically competent care and services, and the elimination of health and health care disparities/inequities. Timely, valid, and representative data collection is essential to the provision of quality and equitable care.Data Collection Improvement ProcessAs our commitment to continuous quality improvement, Kaiser Permanente National Diversity recommended several updates and modifications for KP HealthConnect, our electronic medical/health record system. These accepted recommendations are based on research and collaborative input from various stakeholders and constituents to improve Race, Ethnicity, and Language data collection.During the initial phase of data collection, we utilized an approach that was common throughout the industry, which was to ask for the Office of Management and Budget (OMB) federal 'Race' and 'Ethnicity'* categories. We addressed the need to identify diverse subpopulations by also including a limited number of detailed groups.At that juncture, our technical systems were not designed to house granular information separately on subpopulation groups, therefore, granular descriptions were coupled to the OMB federal standards to accommodate compliance and reporting needs.During initial implementation, important concerns were raised, which included the following:The need for more granular categories for ethnicity.The feedback that Hispanic/Latino patients felt uncomfortable selecting 'Race' categories that they may not identify with.The issue of coupling of federal standards to granular information did not allow for self-identification.To address these concerns, in 2005, our National Diversity office undertook a comprehensive assessment, review, and analysis of industry collection standards, practices, and lessons learned as well as integrating notable work and contributions from the fields of public health, sociology, law, race relations, journalism, demography, and statistics.During this assessment for quality improvement, collective discovery concluded that a new collection framework was needed. We explored and were able to identify technological innovations in our electronic medical/health record system that offered viable options in how the data fields and code sets could be built to accommodate the proposed changes.We seized the opportunity to improve data collection by 1) further understanding the constructs of Race, Ethnicity, and Language as well as their intersections in-depth and 2) developing organizational structure, decision making processes, and guiding principles that would advance uniform, standardized, and systematic data collection.Code Selection ProcessThe Race, Ethnicity, and Language Code Sets that were developed represent the first phase of Kaiser Permanente's Member Demographic Data Collection Initiative on Race, Ethnicity, and Language. Our Code Sets are open and responsive to accommodate preferred selfidentification as well as changes in national/international developments and demographic trends.Granular EthnicityKaiser Permanente examined the significance of ethnicity in the context of health, and concluded that the creation of a more encompassing definition of Ethnicity was needed. Granular Ethnicity would be able to capture patient data at a level that would inform beyond the limited federal standards to ensure timely, valid, and representative data collection.The Code Sets are reviewed on an on-going basis primarily to ensure alignment with changing demographic trends in the United States as well as relevance in culturally and linguistically competent care and patient-centered care.Ongoing Analysis and Review for Comparability (Selected Examples Provided)National Data Sets and ResourcesU.S. Decennial Census/American Community SurveyYearbook of Immigration StatisticsHealth Surveys and Research (Population or Region Specific)California Health Interview SurveyNational Latino and Asian American StudyTransnational and International Data Sets and ResourcesUnited Nations Statistics DivisionWorld Health OrganizationData Aggregation (Roll-Up)Kaiser Permanente's collection framework is asking the two part question separately (Combined Format and Granular Ethnicity), thereby eliminating the need/use of automatic linkage and data aggregation (commonly known as "roll-up"). Data aggregation should be specific to the research and/or reporting needs for the population groups under study/report.Automatic linkage or aggregation would produce deteriorated or inaccurate data if patients are linked/aggregated in a way that contradicts or misrepresents their personal selfidentification, especially for multiracial persons who identify with more than one federal 'Race'/'Ethnicity' standard.The effects of globalization, population displacement, and social movements are undermining the capacity of one nation to fix a person's identity. Simply, national or geographic origin does not automatically determine one's race or ethnicity, nor does ethnicity determine race, nor does race determine ethnicity.Data aggregation without the consideration of the intersecting layers that construct identity would be similar to suggesting that 'American' or being born in the United States could be automatically linked/aggregated into 'White'.The Combined Format and Granular Ethnicity fields were developed as two separate questions to achieve health care, research, compliance, reporting, and resource allocation needs. As importantly, it ensures the patient's right to self-identify as well as ensures the highest data quality and integrity.Language (Spoken and Written)Kaiser Permanente focused on developing timely, valid, and representative language codes to capture our nation's increasing diversity. We are collecting both spoken and written language to ensure the provision of quality and equitable health care.The Spoken/Written Language Code Sets are reviewed on an on-going basis primarily to ensure alignment with linguistic trends in the United States, current organizational language utilization, as well as relevance in culturally and linguistically competent care and patientcentered care.Ongoing Analysis and Review for Comparability (Selected Examples Provided)National Data Sets and ResourcesU.S. Decennial Census/American Community SurveyOrganizational Language Utility (Regional and National)Vendor Utilization DataKaiser Permanente Utilization DataTransnational and International Data Sets and ResourcesInternational Organization for StandardizationRace, Ethnicity, and Language Code SetsNOTES (Applies to all Code Sets):'Decline to State', 'Other', and 'Unknown' are collection codes utilized by Kaiser Permanente.Categories that represent one code in the collection system but have multiple names/designations are listed together and separated by a (/) slash mark.Combined FormatIssued by the Office of Management and BudgetAmerican Indian/Alaska NativeAsianBlack/African AmericanHispanic/LatinoNative Hawaiian/Other Pacific IslanderWhiteDecline to StateOtherUnknownThe Combined Format is a multiple response field.Granular EthnicityAcadian/CajunAfghan/AfghanistaniAgikuyu/KikuyuAkanAlaska AthabascanAlbanianAleutAlgerianAlsatianAmara/AmharaAmazigh/Imazighen/BerberAmerican/United StatesAmerindian/Indigena/IndioAntiguan/BarbudanApacheArab/ArabicArgentine/ArgentineanArmenianAsian Indian/Indian (Asia)Assyrian/Chaldean/SyriacAustralianAustrianAzerbaijaniAzeriBahamianBajan/BarbadianBamar/BurmanBangladeshiBantuBasqueBelarusian/BelorussianBelgianBelizeanBengaliBlackfeetBolivianBosniakBosnian/HerzegovinianBrazilianBritish Isles/British Isles originBritish West Indian/IndiesBriton/BritishBulgarianBurmeseCambodianCameroonian/CameroonCanadianCanadian American IndianCape VerdeanCarpatho RusynCelticCentral American IndianCherokeeCheyenneChickasawChileanChineseChippewaChoctawColombianColvilleComancheCongolese/CongoCosta RicanCreekCreoleCroat/CroatianCrowCubanCypriotCzechCzechoslovakianDane/DanishDelawareDominicanDutchDutch West Indian/IndiesEast IndianEastern CherokeeEcuadorianEgyptianEmirati/United Arab EmiratesEnglishEritreanEskimoEstonianEthiopianFijianFilipino/PhilippineFinn/FinnishFleming/FlemishFrenchFrench CanadianFulani/HausaGeorgianGermanGerman RussianGhanaian/GhanianGreekGrenadianGuamanian/ChamorroGuatemalanGuyaneseHaitianHawaiian/Native HawaiianHmongHonduranHopiHungarianIbo/IgboIcelanderIndonesianInupiat EskimoIranianIraqiIrishIroquoisIsraeliItalianIvoirian/Ivory CoastJamaicanJapaneseJavanese/Java/JawaJewish/JewJordanianKazakh/QazaqKazakhstaniKenyanKeresKhmerKinh/VietKiowaKittitian/NevisianKoreanKurd/KurdishKuwaitiLao Loum/Lowland LaoLao/LaotianLatin American IndianLatvianLebaneseLiberianLibyanLithuanianLumbeeLuxemburgerMacedonianMalayMalaysianMalteseMaoriMendeMestizoMexicanMexican American IndianMohawkMoldovan/MoldavianMontenegrinMoroccanMuscogee (Creek) NationNavajoNepalese/NepaliNew Zealander/New ZealandNicaraguanNigerianNorwegianOglala SiouxOkinawanOklahoma ChoctawOneida Nation of New York/OneidaOromoOsageOttawaPaiutePakistaniPalestinianPanamanianParaguayanPart HawaiianPashtun/PathanPennsylvania GermanPersianPeruvianPimaPole/PolishPortuguesePotawatomiPuebloPuerto RicanPuget Sound SalishQuechuaRomanianRosebud SiouxRussianSaint LucianSalvadoranSamoanSan Carlos ApacheSaudi/Saudi ArabianSault Ste. Marie ChippewaScandinavianScotch-IrishScottishSeminoleSenecaSenegaleseSerb/SerbianShawneeShonaShoshoneSierra LeoneanSingaporean/SingaporeSinhalese/SinghaleseSiouxSlavicSlovakSlovene/SlovenianSomali/SomalianSouth AfricanSouth American IndianSoviet/Soviet UnionSpaniard/SpanishSpanish American IndianSri LankanSudaneseSwede/SwedishSwissSyrianTagalogTahitianTaiwaneseTanzanianTemne/Temme/ThemneTeton SiouxThaiTigrinya/Tigray/TigrawayTlingitTlingit-HaidaTogolese/TogoTohono O'OdhamTonganTrinidadian/TobagonianTunisianTurk/TurkishTurtle Mountain Band/Turtle MountainU.S. Virgin IslanderUgandanUkrainianUruguayanUteUzbek/UzbegUzbekistaniVenezuelanVietnameseWelshWest IndianWhite Mountain ApacheYakamaYaquiYemeniYorubaYugoslavianYup'ik/Yupik EskimoZimbabweanZuniDecline to StateOtherUnknownCurrently, Granular Ethnicity is not available as a multiple response field for multiethnic persons. Request for IT change is underway.Spoken LanguageAbkhazAfarAfrikaansAkanAlbanianAlgonquinAmerican Sign LanguageAmharicApachean/ApacheArabicAramaicArmenianAssyrian Neo-Aramaic/AssyrianAymaran/AymaraAztec/NahuatlBantuBashkirBasqueBengaliBhojpuriBislamaBosnianBretonBulgarianBurmeseCantonese/Chinese, YueCatalan-Valencian-BalearCebuanoChamorroChaochow/Chinese, Min NanChaozhou/Chinese, Min NanCherokeeChuukese/TrukeseCorsicanCroatianCushitic/CushiteCzechDakotaDanishDari, Zoroastrian/DariDinkaDutchEnglishEsperantoFarsi/PersianFijianFilipinoFinnishFormosanFrenchFrench CreoleFrench, Cajun/CajunFukienese/Chinese, Min NanFula/FulaniFuzhou/Chinese, Min DongGaelic, Irish/Irish GaelicGeorgianGermanGerman, Pennsylvania/Pennsylvania DutchGreekGujarati/GujarathiGuranaiHaitian Creole French/Haitian CreoleHakka/Chinese, HakkaHawaiianHebrewHindiHmongic/HmongHmong-Mien/Miao-YaoHuizhou/Chinese, HuizhouHungarianIgbo/IboIlocanoIndonesianInuktitut, GreenlandicItalianIu Mien/MienJamaican Creole English/Jamaican CreoleJapaneseKannadaKashmiriKeresKoreanKruKurdishKusaiean/KosraeanLadinoLanguedocien/OccitanLao/LaotianLatvian/LettishLingalaLithuanianMacedonianMalayMalayalamMandarin/Chinese, MandarinMandeMarathiMarshalleseMaya, Yucatan/YucatecoMicronesianMixtecoMongolian Proper/MongolianMon-Khmer/KhmerNavajoNepaliNorwegianOriyaOromoPalauan/PalauPampangan/PampangoPanjabiPashtoPatoisPohnpeianPolishPortugueseProvencalPunjabiQuechuan/QuechuaRomanian/MoldavianRundi/KirundiRussianRwanda/KinyarwandaSamoanSango/SanghoSerbianShanghainese/Chinese, WuShonaSign Language (Other)Sinhala/SinhaleseSlovakSomaliSonghai/SonghaySoninkeSorbian, Lower/LusatianSorbian, UpperSotho, Southern/SesothoSpanishSwahili/KiswahiliSwedishTagalogTaiwanese/Chinese, Min NanTamilTatarTeluguThaiTibetanTigrigna/TigrinyaToishanese/Chinese, YueTonga/Western NyasaTonganTsimshianTsongaTswana/SetswanaTupi-Guarani/GuaraniTurkishTurkmenTwi (Akan)UkrainianUrduVietnameseVlaams/FlemishWolofXhosaYapeseYi, Sichuan/SzechuanYiddishYorubaZuluDo Not SpeakDecline to StateOtherUnknownSpoken Language is a multiple response field.Written LanguageAbkhazAfarAfrikaansAkanAlbanianAlgonquinAmharicApachean/ApacheArabicAramaicArmenianAssyrian Neo-Aramaic/AssyrianAymaran/AymaraAztec/NahuatlBantuBashkirBasqueBengaliBhojpuriBislamaBosnianBrailleBretonBulgarianBurmeseCatalan-Valencian-BalearCebuanoChamorroCherokeeChinese, SimplifiedChinese, TraditionalChuukese/TrukeseCorsicanCroatianCushitic/CushiteCzechDakotaDanishDari, Zoroastrian/DariDinkaDutchEnglishEsperantoFarsi/PersianFijianFilipinoFinnishFormosanFrenchFrench CreoleFrench, Cajun/CajunFula/FulaniGaelic, Irish/Irish GaelicGeorgianGermanGerman, Pennsylvania/Pennsylvania DutchGreekGujarati/GujarathiGuranaiHaitian Creole French/Haitian CreoleHawaiianHebrewHindiHmongic/HmongHmong-Mien/Miao-YaoHungarianIgbo/IboIlocanoIndonesianInuktitut, GreenlandicItalianIu Mien/MienJamaican Creole English/Jamaican CreoleJapaneseKannadaKashmiriKeresKoreanKruKurdishKusaiean/KosraeanLanguedocien/OccitanLao/LaotianLatvian/LettishLingalaLithuanianMacedonianMalayMalayalamMandeMarathiMarshalleseMaya, Yucatan/YucatecoMicronesianMixtecoMongolian Proper/MongolianMon-Khmer/KhmerNavajoNepaliNorwegian, Bokmal/BokmalNorwegian, Nynorsk/NynorskOriyaOromoPalauan/PalauPampangan/PampangoPanjabiPashtoPatoisPohnpeianPolishPortugueseProvencalPunjabiQuechuan/QuechuaRomanian/MoldavianRundi/KirundiRussianRwanda/KinyarwandaSamoanSango/SanghoSerbianShonaSinhala/SinhaleseSlovakSomaliSonghai/SonghaySoninkeSorbian, Lower/LusatianSorbian, UpperSotho, Southern/SesothoSpanishSwahili/KiswahiliSwedishTagalogTamilTatarTeluguThaiTibetanTigrigna/TigrinyaTonga/Western NyasaTonganTsimshianTsongaTswana/SetswanaTupi-Guarani/GuaraniTurkishTurkmenTwi (Akan)UkrainianUrduVietnameseVlaams/FlemishWolofXhosaYapeseYiddishYorubaZuluDo Not Read/WriteDecline to StateOtherUnknownWritten Language is a multiple response field.For more information, please contact:Gayle Tang, RN, MSNDirector, National Linguistic & Cultural ProgramsNational Diversity, Kaiser PermanenteOne Kaiser Plaza, 17 LakesideOakland, CA 94612E-mail: Gayle.Tang@kp.orgPhone: 510-271-6828 * 'Race' in quotes refers to the federal OMB defined race standards. 'Ethnicity' in quotes refers to the federally defined OMB ethnicity standards. Current as of March 2010 Internet Citation: Appendix G. Kaiser Permanente: Evolution of Data Collection on Race, Ethnicity, and Language Preference Information: Race, Ethnicity, and Language Data: Standardization for Health Care Quality Improvement. March 2010. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/research/findings/final-reports/iomracereport/reldataapg.html
Back
