San Francisco, California
The Subcommittee on Populations of the NCVHS was convened on November 13, 2003 at The Palace Hotel in San Francisco, California. The meeting was open to the public. Present:
OVERVIEW OF SUBCOMMITTEE ON POPULATIONS, PURPOSE OF HEARING AND REVIEW OF AGENDA - Vicki Mays
Dr. Mays outlined the hearing questions. Noting the geographical diversity inherent in covering this population, she encouraged people in other areas to participate.
OVERVIEW OF PACIFIC ISLAND HEALTH DATA ISSUES - Christina Perez, Regional Minority Health Coordinator, Region 9
Ms. Perez discussed characteristics of Region 9 and its data projects. Goals are increased participation of Asian Americans, Native Hawaiian and other Pacific Islander (ANHOPI) populations in federal policy development; participation in regionally funded activities; and closing the health data gap through research and collection of data for ANHOPIs.
THE NEED FOR DETAILED ANHOPI FOR HEALTH POLICY - Ho Tran, M.D., Asian Pacific Islander American Health Forum
Dr. Tran said the paucity of data on Asian Americans and Pacific Islanders (AAPI) in federal data sets makes it difficult to measure programs and eliminate health disparities. Only three of Healthy People 2010’s 22 objectives comply with OMB. Under representation of AAPIs and inadequate sample sizes in national, state and local surveys render many databases useless for examining demographic- and policy-related trends affecting AAPIs. She conveyed the Forum’s recommendations.
ASSESSING ANHOPI DATA - Gem Daus, Asian Pacific Islander American Health Forum Census Information Center
Mr. Daus cited two issues: whether the health of multi-racial categories could be characterized and changes in health status measured. He noted methods for bridging to gain a standard denominator and cautioned about comparisons with data collected before 2003. Noting the American Community Survey (ACS), which replaces the long form in Census 2010, is still being tested and that outreach and translation will not be as extensive as for Census 2000, he advocated retaining funding.
ANHOPI MEASUREMENT AND CLASSIFICATION ISSUES - Elena Yu, Ph.D., The John Hopkins University
Noting ANHOPIs are the nation’s second fastest growing minority, Dr. Yu discussed the demographic growth rate and population characteristics, identified classification and identification issues and their relevance in terms of health for classifying and measuring ANHOPIs, detailed data and data needs, and made recommendations calling for paradigm shifts.
HEALTHCARE QUALITY INDICATORS FOR ANHOPI POPULATIONS - Ellen Wu, Executive Director, California Pan Ethnic Health Network
Ms. Wu described the California Pan Ethnic Health Network’s (CPEHN) policy advocacy work on multi-cultural health issues related to Proposition 54 and signing of SB 853. She stressed dispelling stigma and myth about collection of race/ethnicity data, not letting a sample’s population size limit research, and partnering with community-based policy advocates to ensure research efforts are relevant. She emphasized cultural and linguistic services and working across agencies/departments to standardize and mandate collection of race/ethnicity data.
CANCER RESEARCH AMONG ASIANS - Scarlett Lin Gomez, Ph.D., Northern California Cancer Center
Ms. Gomez expressed concern that population data from the Census and Department of Finance is inconsistent with how race/ethnicity in numerator data is collected. She encouraged medical reporting facilities to uniformly collect data on race/ethnicity. She suggested a national task force could formulate collection policies and guidelines and advocated desegregation of subgroups. Ms. Gomez urged researchers to engage community and advocacy groups in the design and conduct of studies, and make data and results available in appropriate languages.
PALAU TELEPHONE CONFERENCE
Ms. Tellie discussed Palau’s spiritual life, health, and how they integrated in dealing with community health.
Dr. Dever recommended that U.S. federal agencies harmonize data requirements and processes and develop standardized, user-friendly software to promote data collection, analysis and reporting.
HAWAII - Catherine Sorenson, Ph.D., Department of Health, Hawaii
Dr. Sorenson noted seven national data sets do not use 1997 OMB recommended race categories and Hawaii Department of Health (HDH) programs do not employ a standardized method to define, collect or report race/ethnicity data. The Hawaii Health Data Warehouse is being developed to coordinate resources across administrations and divisions, ensure consistency in public-health data with national recommendations, and increase public access to health data.
QUALITATIVE DATA ISSUES FOR NATIVE HAWAIIANS AND AMERICAN SAMOANS
Dr. Murry discussed the University of Hawaii’s research with a HERSA state planning grant studying Hawaii’s uninsured. She emphasized that the numbers were not necessarily “a good way” to determine collections. She advocated collecting data on mixed populations, cautioned that understanding issues and addressing health disparities was limited without qualitative data, encouraged multiple translations, and discussed guidelines for community-partnership research.
Mr. Ohta stressed harmonizing federal-agency approaches and requirements. Mr. Spoehr noted the political relationship to data-collection in the Hawaiian community.
SUBCOMMITTEE DELIBERATIONS OF TESTIMONY
Members reviewed the recommendations heard and identified those the Committee should focus on in updating their work plan, preparing to present those with resource implications at the November 21 Executive Subcommittee Retreat. Privacy and confidentiality issues will also be introduced along with the question of where language and a matrix to enable core data sets to follow consistent definitions and collections in aggregable categories might best be worked on.
Health U.S. will be asked if focusing and targeting surveys on well defined subgroups, combining results and creating information about larger groups and the whole population would be useful. A meeting will be arranged with Dr. Sondik, director of NCHS, and the Census, OMB, NHII and the Subcommittee on Privacy and Confidentiality to explore whether sharing with the Census Bureau is “doable.” Members will draw upon input from the IOM panel’s hearings and meetings with Social Security. Members noted a discussion with the Data Council will be helpful. Members discussed hearing from HHS about display of geo-reference data and privacy/confidentiality, standards and security, and threshold issues. A conference call will be scheduled to draft a letter of recommendations for the January 28 full Committee meeting; Ms. Dearing will be included. ACS will be asked to present at the March full Committee meeting.
The staff will prepare a list of evolving issues for the January 28 Subcommittee meeting; preparations will be made for a discussion about harmonization of data reporting at another meeting. NLM will be invited to the meeting to discuss their experiences posting accessible data sets with large bodies of questions on a Web site. Census will be asked to present their recommendations about data issues. Members also will request to hear from the Social Determinates, Sampling, Health Disparities and Environmental Justice Work Groups.
DETAILED SUMMARY
The details of all presentations and letters drafted can be found in the meeting transcript posted on the NCVHS Web site, www.ncvhs.hhs.gov
-DAY ONE-
OVERVIEW OF SUBCOMMITTEE ON POPULATIONS, PURPOSE OF HEARING AND REVIEW OF AGENDA - Vicki Mays
Dr. Mays outlined the hearing questions. Noting the geographical diversity inherent in covering this population, she said there would be questions and answers after the presentations and a call-in number and encouraged people in other areas to participate.
OVERVIEW OF PACIFIC ISLAND HEALTH DATA ISSUES - Christina Perez, Regional Minority Health Coordinator, Region 9
Ms. Perez discussed characteristics of Region 9 and their data projects. The Asian population in California is close to 11 percent; Native Hawaiian and other Pacific Islanders’, 03 percent. Well over 100 languages are spoken by Asian populations throughout this region. Travel extends 8,627 air miles across Oceana and the Pacific Islands. The infant mortality rate in the Federated States of Micronesia for 2003 is 32.9 deaths per 1,000 live births. Vietnamese women have a cervical cancer incidence rate five times that of white women; Vietnamese men have the highest rates of liver cancer for all racial and ethnic groups. Problems on the Pacific Islands include inadequate supplies of potable water, insufficient facilities for disposal of waste, threats to marine ecosystem, over- and illegal fishing practices, and a high unemployment rate.
Ms. Perez said their data project’s overall goals were increased participation of ANHOPI populations in federal policy development; participation in regionally funded activities; and closing the health data gap through research and collection of data for ANHOPIs.
She noted a collaboration of the regional Office of Population Affairs and health administrator: a research study with the University of Hawaii School of Social Work assesses perceived barriers to prenatal care access among AAPI immigrant women in Hawaii. Another project, the Office of Minority Health (OMH) state partnership initiative was established in federal fiscal 1999 to help state minority health entities develop or expand existing infrastructures, address public health needs of racial and ethnic minorities, and undertake special projects addressing emerging health-related issues impacting minority communities. Another project funded by the divisional OMH office is contracting with the Association of Asian Pacific Community Health Organizations (APCHO) to develop a strategic plan and curriculum to implement a geographic information systems technical assistance and training module tailored to data needs of ANHOPIs.
Ms. Perez reported that OMH Regions Eight, Nine and Ten held a tri-regional data workshop in March 2002 to assist minority community-based organizations, researchers, and state and local health officials understand issues related to the collection, use and dissemination of health data on minority populations, and develop strategies addressing these issues. In October 2002 the Office of Women's Health announced development of a database to provide state and local public health agencies, faith- and community-based organizations, and health advocates with reliable comparative data.
She commented on the defeat of Proposition 54, noting the initiative would have restricted collection of data and classification of individuals by race/ethnicity, broadly affecting many aspects of state and local government.
Questions and Discussion
Ms. Perez said getting accurate information was always a challenge for many smaller islands and territories. Much information came from community partners. The region hoped to collectively develop a way to capture more data. She suggested the best way to begin to understand “the degree of depravity of information” was to travel to the islands.
Ms. Perez explained that she would have to rely on private researchers’ data to get information to answer many questions (e.g., whether the high rate of liver cancer was related to a genetic predisposition among Vietnamese or environmental) and that she would not necessarily learn anything about health issues in the subpopulations. Getting a “snapshot” of health or health status at the county level would require a month, full time, “trying to piece together” as much information as possible. She emphasized the difficulty of getting a better picture of health disparities for Pacific Islanders and their causes without a standard way of collecting their data. She noted they had to go to a number of resources for information and that there were few opportunities for sharing and updating information usable for the region’s populations and subpopulations. Experts, time, commitment and funding were key resources.
Asked about the region's role in the collection, analysis and utilization of data, Ms. Perez said it was building the communities’ capacity so efforts to improve the health of the population were supported. She emphasized enabling community advocates to utilize and develop data so they could look at and, hopefully, improve health outcomes. Developing partnerships with community-based organizations aimed at building their technical capacity around data was another goal. She noted that limited resources to act in larger ways were a barrier.
Ms. Perez said the Pacific Islands’ linguistic diversity added richness to the population but was a barrier to receiving services. She noted that private providers of health care did not always understand their needs and could incorrectly use data, making wrong assumptions.
Asked if the region could provide interviewers and others to translate survey instruments if the federal government had resources, Ms. Perez said California had thousands of health professionals with advanced degrees who could not practice because of licensing issues but were a source of information, expertise and translation.
She reported there were Internet stations on the islands but few cell phones; one had to walk “a block or two” to use a phone. The region worked closely with the Pacific Islander Health Officers Association on data training. Collaboration in research was welcome.
Mr. Localio noted that, if resources were available for a NCHS targeted survey of a subpopulation along with Geographical Information Systems (GIS) input to enhance the data’s usefulness, under current statutes and regulations those data could only be used in NCHS’s main office. He questioned if that could be responsive to their needs. Ms. Perez replied that having five satellite offices around the country “would be great”.
THE NEED FOR DETAILED ANHOPI FOR HEALTH POLICY - Ho Tran, M.D., Asian Pacific Islander American Health Forum
Dr. Tran reported growth between 1990 and 2000 of 48 percent for the Asian population alone, 72 percent for Asian in combination. Pacific Islander alone increased 9.5 percent; in combination, 140 percent.
Remarking that the Census might be the best example of how data collection could work for AAPIs, Dr. Tran noted that implementation of the 1997 OMB standards mandated use of the minimum five-race categories, including Asian and a separate category for ANHOPI, when collecting race data. She commended HHS for implementing revised standards for 149 (85 percent) of the 175 forms. Some 128 forms use multiple race reporting.
Dr. Tran noted that Healthy People 2010 served as a plan of action and resource allocation for HHS. She pointed out, however, that the paucity of data on AAPIs in federal government data sets made it difficult to measure programs and eliminate health disparities. She commended HHS for adopting Forum’s recommendation to specify whether Asian or Pacific Islander data had not been collected (DNC), collected but not analyzed (DNA), and analyzed but not reported due to small sample size (DSU). She said this clarification was useful for prioritizing future efforts.
Dr. Tran reported that only three of Healthy People 2010’s 22 objectives comply with OMB and under representation of AAPIs and inadequate sample sizes in national, state and local surveys render many databases useless for examining demographic- and policy-related trends affecting AAPIs. Six objectives only have aggregate AAPI data. Five have aggregated data for Asians but not ANHOPIs. Eight do not have statistically reliable aggregated data even for AAPIs.
Dr. Tran conveyed the Forum’s recommendations to: collect enough disaggregated data for subgroups as well as the minimum five classifications, convene experts to determine solutions to methodological problems posed by small populations, conduct more small targeted surveys assessing populations underrepresented in larger national surveys, conduct culturally competent surveys in other languages on a regular basis, increase the capacity for ANHOPI researchers and organizations to conduct health disparities research, require the Social Security Administration to collect race/ethnicity data, and fund ACS.
ASSESSING ANHOPI DATA - Gem Daus, Asian Pacific Islander American Health Forum Census Information Center
Mr. Daus said over 50 census information centers (CIC), ranging from direct-service community-based organizations, universities, national advocacy organizations, and tribal nations, disseminate information to under served communities. The Bureau of the Census Customer Liaison Office provides CIC with free data, training, special products, access to databases and other support including an advanced query system. As a CIC, Health Forum responds to data requests, conducts needs assessments, prepares fact sheets, and does advocacy. The Forum provides technical assistance and training for using the Census query system and coordinates with other AAPI CICs.
Mr. Daus mentioned two issues: (1) whether, given the mass of data, the health of multi-racial categories could be characterized and (2) changes in health status measured. He noted methods for bridging in order to gain a standard denominator and cautioned about comparisons with data collected before 2003, which is presented differently and can be difficult to match.
Noting ACS, which will replace the long form in Census 2010, was still being tested and that Census funding dips between the big censuses, he advocated retaining funding. He expressed concerned that outreach and translation would not be as extensive as for Census 2000.
Questions and Discussion
Asked for estimates of the percentage of the population and subgroups missed in collecting data mainly in English or Spanish, Mr. Daus noted both the Asian and Pacific Islander populations were pocketed (e.g., Hmong, Vietnamese). Dr. Breen pointed out that, according to CHIS, 60 percent of Asians in California are immigrants.
Mr. Daus clarified that one can access data through a CIC down to the census-tract level, though there is a Title 13 census restriction, confidentiality filters, criteria and a level beyond which one might not get data.
Ms. Tran noted a difference in confidentiality between census data, which reflects few health-related issues, and health data, which involves more personal and confidential issues but affects health issues. Mr. Daus suggested that HHS could help people understand how census data and health surveys are connected and how a change in the census affects health data.
Asked if a definitive listing of languages was recognized, Mr. Daus said the Census collected language data and had an incomplete list of hundreds (i.e., Native Hawaiian was absent).
Noting the Committee scheduled an ACS presentation for the December Data Council meeting, Mr. Hitchcock asked about methodological issues users wanted pursued and people with expertise to address them. Mr. Daus suggested Dr. Yu. He said Health Forum was convening some HIV/AIDS researchers and would pose the questions to them.
Mr. Daus advocated assessing Southeast Asian and Pacific Islander populations with more small targeted surveys. Noting Healthy People 2010 only used national data sets, he suggested utilizing local or other data that had limitations but might shed light on a population’s progress.
He asked the Committee to evaluate the extent that technical assistance, personnel, and training had been received to implement previous recommendations.
ANHOPI MEASUREMENT AND CLASSIFICATION ISSUES - Elena Yu, Ph.D., The John Hopkins University
Noting ANHOPIs are the nation’s second fastest growing minority, Dr. Yu reported the population increased by 76 percent between 1990 and 2000. By self reports, 14 percent of Asian Americans and 54 percent of ANHOPIs are multiracial; Immigration contributed 86 percent of that growth.
She said Census terminology negated the fact that many ANHOPIs are U.S. born and some are U.S. citizens. The concept of territorial preemption is implied in the identification of Native Hawaiians and other Pacific Islanders. ANHOPI population has a large household size. Two-thirds speak a non-English language. Twenty-five-year-olds and above are twice as likely as non-Hispanic whites to have either a college or ninth-grade-or-less education. Income and employment distributions also are bimodal: many are professionals; many others are service or unskilled workers. ANHOPIs are twice as likely as non-Hispanic whites to be renters
Dr. Yu cautioned that classification methods significantly impact health estimates for Native Hawaiian and other Pacific Islanders. Tiny numerical change result in big percentage shifts. She wondered whether everyone lost sight of what mattered by focusing so much on the classification and measurement of ANHOPIs that more substantive issues were neglected.
Dr. Yu said many non-genetic confounders must be collected when studying ANHOPIs: nativity, generation in the U.S. culture, English proficiency, health literacy, socioeconomic status, dietary preferences, region of origin, whether urban or rural, and use of traditional medicine.
Noting the need for transparency in classification and measurement, she urged publications studying AAPIs and ANHOPIs to specify how they classify and identify populations. She said appropriate method of classification depended on the question. She stressed that data obtained using different classification methods should not be compared or only compared with caution.
Dr. Yu cited a need to control for SES when the playing field is not level. Noting minorities had less purchasing power, she said controlling for income did not make the groups comparable.
Noting ANHOPIs are not distributed throughout the U.S., she said national surveys and sampling population in proportion to a group’s representation in the population would always show unreliable estimates.
Dr. Yu cited a study published in 2000 showing five measures of secondary access were worst for AAPIs: preventive services, time-related access for routine care or general need, specialty services, telephone and general access. Asians receive less counseling or medical encounters requiring verbal communication and fewer longer physician-patient contacts
Leading causes of death for AAPIs are cancer (27 percent), heart disease (26 percent), stroke, accidents, chronic lower respiratory disease, pneumonia, influenza, and diabetes.
Dr. Yu noted and statistically debunked myths about Asian Americans (e.g., Asian Americans had a low infant mortality rate and thus were a healthy population without health disparities; AAPIs had low death rates for cancer).
She recommended special large-scale population-focused surveys on ANHOPIs; legislation authorizing Congressional funding to NCHS, CDC and the Census to conduct surveys on ANHOPIs; self-identification of ANHOPIs; and including multiple access of classification and health risk identifiers in health surveys
Dr. Yu noted that nine of the ten places with the highest percentage of Asians are in California. The other is in Hawaii. The ten places with the largest percentage of Native Hawaiians and other Pacific Islanders are in Hawaii, Utah, and California. Eighty percent of the ANHOPIs live in ten states. Over half the Pacific Islanders live in two states.
She said who framed questions and why they were framed as they were affected answers. She stressed the need for health surveys in ANHOPI languages, coordinated multi-site population-focused survey, addressing the issue of inclusion, case control and alternative study designs.
Dr. Yu advocated having at least two super-funded multi-site surveys in AAPI languages through Congressional appropriations: one focused on chronic diseases with common-risk factors (cancer, heart disease, stroke, diabetes, substance abuse and depression), another aimed at vaccine-preventable infectious diseases. She noted government surveys were the only source of large-scale national data on ANHOPIs. NCHS was “a major player;” California a potential. Some data exists from routine government sponsored surveys and investigator-funded studies, but numbers are small and confidence intervals wide. Dr. Yu said most epidemiologists and all schools of public health lacked training to analyze data not normally distributed. Trained ANHOPI epidemiologists might be spoken to before a proposal was submitted and sometimes invited as consultants, but not included from “conception to write-up” or in the authorship.
Noting she was a full, tenured professor, she said universities were archaic for ANHOPI minority students who could not write dissertations on minority health research issues because many intellectual resources were unfamiliar with the topic. Universities pigeonholed minority health researchers as computer data analysts rather than train and educate them to become leaders and decision makers. She contended that universities seldom considered community involvement and public health practices productive work and lacked infrastructure for that work.
She proposed the creation of an ANHOPI policy research institute that would “transcend university walls” but work within them and with community organizations and ANHOPI clinics. It would provide health professionals with fellowships and internships for students at top universities wanting to do ANHOPI research but lacking mentors. Students and researchers’ collaborations (Ph.D and doctorate as well as master's level) would address cutting edge research and policy issues. ANHOPI data banks would be developed and maintained jointly with government agencies (e.g., NCHS, Census and CDC).
Questions and Discussion
Asked how urban-rural origins were assessed, Dr. Yu noted capitals in developing countries were “a far cry” from the rest of those lands in terms of exposure to Western values, science and medicine. These different experiences affected one’s ability to be aggressive and assertive, seek a personalized relationship, or deal with bureaucracy and public health.
Remarking that she heard sampling was used because it was relatively cheap and adequate, Dr. Breen said it sounded expensive. Dr. Yu said cost was a problem but relative to what one thought data was worth. Asked about a way to get adequate sample sizes without using list samples, Dr. Yu said the best strategy for using census data to sample Asian Americans was to identify blocks in census tracts with 20 percent or more of the subgroup and boundaries defining them, then identify and with “hard legwork” canvas households in those blocks utilizing a reverse directory.
Dr. Yu contended that good data on ANHOPIs could help answer critical, substantive questions about majority Americans: i.e., enough descriptive data collected about Asian Americans in earlier eras could have supplemented research on menopausal symptoms that only emphasized drugs and looked at physical exercise.
In order to measure and classify minority groups to meet objectives of Healthy People 2010, Dr. Yu said surveys had to be supplemented with questions about where responders and their parents were born, nativity, rural/urban origin, generation in the U.S., and level of education.
Asked about cultural equivalency, Dr. Yu told how a pilot project drew upon State-side members of China’s main language groups and professional translators in Hong Kong, comparing and evaluating differences and underlying meanings. She emphasized that working with multiple languages was “doable” and a matter of being willing to invest resources and go through the rigorous, labor-intensive process of back translation and “ironing out” nuances.
Noting NCHS applied cognitive psychology strategies to serve questions, Dr. Yu observed that saying this cost too much to do for ANHOPIs reflected a value judgment and prejudice that had to be changed.
Asked if differences between subgroups were more important than those between cities within a subgroup, and how to deal with those problems, Dr. Yu said it depended on the research question. Sampling and sampling strategy had to be targeted.
Noting one overlooked or “hit” a sizeable segment of the AAPI population with each three-generational household, Dr. Yu emphasized utilizing available technology and knowledge in primary enumeration. Asked about other statistical needs, she suggested looking at how degrees of cultural integration affect health care and access to health in a three-generation family. Noting the confidentiality issue and problem of proficiency, she proposed using the household as the unit of analysis and “cutting” the data differently: looking separately at each generation and comparing them. She said methodological research should be included in every survey in order to measure the sensitivity and specificity of the classification method in locating AAPIs.
Dr. Yu said some AAPI data was collected on earlier computer technology but there was no way to convert them. NCHS surveys (e.g., the vital statistics record) provided a minimum amount of data; the difficulty was getting them written up and analyzed. Asked if someone could find the investigators that had collected early data and assess if it was feasible and worth the resources to archive, Dr. Yu explained that most investigators “wouldn't mind sharing the data” but were concerned about authorship, misinterpretations and contradictions. Noting there were data use policies, Dr. Mays said the Committee should consider this issue.
Asked about focusing public health investigation on the adolescents with multiple racial and ethnic identities identified at increased risk for adverse health, Dr. Yu urged prioritizing resources and prevention efforts for them.
Participants noted religion was an important variable for certain AAPI groups, but the federal government could not ask about specific beliefs or practices.
SUBCOMMITTEE DELIBERATION OF ISSUES
Dr. Cain noted hearing the need for a willingness to “tackle and deal with” their incredible diversity rather than categorize and “gloss it over.” Dr. Steinwachs recommended looking at why there might be biases and misperceptions about “who was healthy and worse off,” considering ways to talk about the issue of size and public health significance, and an agenda that recognized “we can learn something for all of us.”
Recalling hearing a need for funding and legislative language requiring improvements in data, Mr. Hitchcock suggested advocates might focus on getting changes written into law. Dr. Steinwachs noted examples of what could be done through Congress. He noted that members of the Committee were in positions to discuss what had to be done regarding future priorities for investing in and developing information, both in research and public health statistics.
Mr. Localio proposed a paradigm with surveys focused and targeted on well defined subgroups and combining those results and creating information about larger groups and the whole population. Dr. Mays noted to ask Health U.S. if offering data in this way might be useful.
Dr. Steinwachs suggested asking about information needed to define an intervention and action agenda, then identifying clusters lacking it. Ms. Burwell pointed out that the disadvantaged minority health improvement act documented in 1990 substantial need for special studies of racial and ethnic populations, new methodological approaches, mentoring, training, and ensuring these populations could do research, but resources interfered. She said the 21st Century report is one way NCHS and others hope to build momentum and address paradigm shifts
Members discussed supplementing the surveys by having a common core but asking different questions to people in different strata. Dr. Mays suggested the National Center for Minority Health and Health Disparities consider funding to answer population-specific groups questions.
Members discussed methodological issues, studies and sampling frames noting: NHANES was broad, designed to represent the country; HIS accounted for geographical considerations and made regional estimates; SLATES was a state and local survey. Mr. Localio noted that restrictions currently placed on access to data were a difficulty in methodological work and for individual subpopulations.
Mr. Localio reported that Dr. Sondik, director of NCHS, was open to the request made at the full Committee meeting to explore whether sharing with the Census Bureau was “doable” and had suggested a small meeting early in 2004. Mr. Localio noted that nearly everything they had heard about peoples' needs fell within the e-Gov statute and sharing for statistical purposes as defined in recent legislation. The Subcommittee will arrange the meeting and ask for representatives from the Census, OMB, NHII and the Privacy and Confidentiality Group.
Dr. Hungate questioned that the National Children's Study could be responsive to inputs heard that day. Noting there had to be a stratification decision, based on group size, he proposed looking at Census data and sizes for populations of interest. He emphasized not only thinking about the survey’s outcome, but also about who would use it and ensuring that smaller populations partner in the work. Dr. Mays put this on the agenda for the next day.
Dr. Hungate suggested considering how to define population sizes and approaches related to size. Mr. Hitchcock suggested designing surveys so outside groups could link up and build unto the survey with smaller parallel studies comparable on some items. Dr. Mays said they should aim to have a letter of recommendations ready to go to the January full Committee meeting.
Dr. Mays noted they heard recommendations about translation in terms of the language and survey questions, particularly with NCHS surveys. Dr. Heurtin-Roberts pointed out the tension between comparability across surveys and realizing something truly appropriate and useful for a specific population. Members discussed the advantage of having questions and whole instruments posted on a Web site so others could draw from them. Noting NLM did this on a large scale, they noted to ask them to come to the January meeting.
Dr. Yu recommended encouraging investigators who had translated instruments and published papers to share them on a Web site. She suggested that concern about equivalence versus standardized translations could be resolved by having core-versus-supplement questions.
Dr. Mays asked the staff to differentiate between what had to go to the Secretary or directly to NLM, NCHS, or the Board of Scientific Counselors, noting it might be useful in the January meeting to receive this as a set of evolving issues. She recalled that they heard the need for training about how to use the Census data at the community level, ensuring there was technical assistance, and developing investigators from these populations.
Noting “the incredible diversity” among Pacific Islanders, Dr. Heurtin-Roberts mentioned the need to train local populations to access, use and ask questions of the data. Dr. Hungate suggested that a contract within the islands for development of the information was a way to develop that strength
Dr. Yu recommended thinking about ways to utilize the National Research Service Award that did not always have enough minority applicants. She also emphasized thinking about local settings where they could get a critical mass of ANHOPI researchers to work together and train people from these populations to utilize the data. Summer programs were identified.
Ms. Burwell said, short term, there were models for bringing together students and ways to fund them at existing public health institutes. She noted OMH has a center for linguistic and cultural competency that could by linked to the OMHRC Web site.
Dr. Cain recommended emphasizing social and behavioral factors involved in health disparities. Dr. Mays said that they might want to hold hearings to come up with recommendations about what core measures should be and adapt them to fit the surveys. A body of variables, rationale, and encouragement in the RFAs would generate use.
Dr. Cain said many biomedical researchers recognized the need to do more and, given a reasonable approach, would be amenable to collecting some data. Dr. Heurtin-Roberts advocated shifting the emphasis to invest resources in a coherent, systematic and studied fashion.
Dr. Steinwachs cautioned that the surveys were not designed to reflect the social environmental community context. He noted that the Center for Setting Health System Change had a design for making national estimates with high sampling rates. He said health utilization data could be put in a framework like NHIS that sat in communities for a decade. A mini-longitudinal series might help them understand why different things were seen for ANHOPIs in various contexts.
Dr. Breen reflected that collaborating required significant resources as well as a research agenda and plan. She suggested that the Committee make recommendations to “beef up” the activity of OBSSR and others with more social emphasis.
Dr. Mays said the Subcommittee should introduce any major project that would require commissioned work or other activities at the November 21 Executive Subcommittee Retreat so it was considered in the planning phase. Recalling that they talked about mental health statistics and asking core questions around health without any mental health data, she suggested: recommending broadening the portfolio of health data to include social determinants and mental health, NIH make funding available for research, and NCHS rethink the core and commit to ongoing collaborative funding. The agenda would be specific, as in a targeted survey, addressing population health and racial and ethnic groups.
Reflecting that genetics and environment had to be viewed together and that this called for a large survey, Mr. Localio pointed out that there were practical issues to consider. Dr. Hungate proposed identifying groups with unique interests who wanted to self-report data on the Internet. Dr. Mays suggested that data integration was a possibility.
Dr. Lengerich urged the Subcommittee to keep individual social and behavioral determinant issues separate from environment and social determinant ones. Dr. Breen added that it was important to ground individual data collected in geography. Diverse data collection efforts with different methodological concerns collected in dissimilar ways had to be linked to unify broader social context. Citing a general consensus, she urged the Committee to “push the thinking in that direction.” Dr. Lengerich concurred, emphasizing three areas: individual-level information about social and behavioral issues, the environment, and methods to put these together.
Dr. Heurtin-Roberts revisited the question of how long the survey should be. Concurring with the possibility of self-reported data, she said the survey core also needed a social component.
Noting not everything the social security system received was in the database or summarized, Dr. Breen encouraged attending to that data. She noted a combination of IRS and Social Security data might provide a profile of employment, income and wealth. Dr. Mays said they would hear from them. Reporting that the IOM panel responsible for Adequacy of the Racial and Ethnic Data Collected by HHS for Administering its Programs held hearings and met with Social Security, she suggested tapping into any workshop data that might be shared.
Members discussed the need to study persons of mixed ethnicities as a group in their own right. Dr. Breen proposed focusing on the increasingly high rate of adolescents with mixed ethnicity as “the canary in the mine,” contending that examining the microcosm of problems they face in struggling to find their identity and transition in this society could bring a broader understanding.
Noting a need for translation of ACS and a recommendation that it collect health data, Dr. Mays said they will be asked to present at the March full Committee meeting.
HEALTHCARE QUALITY INDICATORS FOR ANHOPI POPULATIONS - Ellen Wu,
Executive Director, California Pan Ethnic Health Network
Ms. Wu described CPEHN’s policy advocacy work in California on multi-cultural health issues related to Proposition 54 and the signing of SB 853, which gives the state regulatory body for health plans, the Department of Managed Health Care (DMHC) authority to regulate cultural and linguistic services in commercial plans. Other victories included the health plans collection of race, ethnicity and language preference of enrollees; availability of translated materials and interpreter services; regulations to ensure limited-English proficient members access to plans’ grievance processes and that plans collect the race/ethnicity of members with grievances and identify if these are due to cultural or linguistic barriers; and inroads in ensuring that DHMC’s 800 HMO help line also collects race/ethnicity data on callers and is accessible to limited-English proficient.
Ms. Wu said CPEHN worked closely with Medi-Cal regarding analyze of their encounter data utilization by race, ethnicity and language preferences. She cited the Healthy Families Quality Work Group as a model, noting they match the heath plans reports on HEDIS requirements with membership data and issue a report including HEDIS measurements by race/ethnicity. She noted that CPEHN’s polling data identified that the message that resonated most with voters about “Prop 54” was that this information was needed for everyone’s health. She emphasized the importance of having mobilized the health community and convening around policy implications of multi-race categories on health disparities research and resource allocations.
Ms. Wu stressed dispelling the stigma and myth about the collection of race/ethnicity data, noting the reason that CPEHN couldn’t talk about data per se around “Prop 54” was that they could not identify when discrimination happened or disparities. She advised not letting the sample population size limit the research conducted, noting the data might identify where research was needed or outreach. She said researchers must partner with policy advocates in community-based organizations to ensure their efforts are relevant to the community.
Stating quality health care could not be provided without knowing the enrollee population and speaking their language, Ms. Wu emphasized cultural and linguistic services and working across agencies and departments to standardize and mandate collection of race/ethnicity data.
Questions and Discussion
Mr. Localio noted the need to distinguish between collecting data upon enrollment and after a person is enrolled and seeks care. Noting the defeat of “Prop 54”, Ms. Wu contended that, if a credible source and community “came out” and there was a public education campaign, people would care. Dr. Mays recalled that community advocates cited the need for ombudsman or litigation resources for individuals. Noting help and consumer hot lines could serve as models, Ms. Breen suggested cultural-competency training for physicians and the front desk so that data is collected and used in a respectful way. Noting appropriate language and cultural sensitivity for HMO’s complaint lines was a well-identified problem, Mr. Hungate questioned whether another organization could better provide a culturally-sensitive complaint service for the plans.
Raising the question of guarantees that data will be used appropriately, Ms. Heurtin-Roberts suggested requiring HMOs and others to draft statements limiting how data would be used.
Asked if purchasers of health-plan services put up barriers to collection of this data, Ms. Wu said they leaned toward “the whole report card” and, so far, collection of race/ethnicity data wasn’t on employers’ radar. She suggested that advocates offer to collect race/ethnicity data on the enrollment form and tell health plans to look at quality measurements that way. Discussing language preference versus language proficiency, Ms. Wu said assessment of the bilingual proficiency of providers, staff and interpreters should have both cultural and medical-terminology components.
Ms. Wu said CPEHN was discussing with NCQA their work on identifying quality indicators around language access and looking at integrating the race/ethnicity data in HEDAS
CANCER RESEARCH AMONG ASIANS - Scarlett Lin Gomez, Ph.D., Northern California Cancer Center
Ms. Gomez said the Northern California Cancer Center’s (NCCC) cancer registry covers over half the Asians in the U.S. and nearly 70 percent of Native Hawaiians and other Pacific Islanders. Estimates of the population at risk come from the decennial census and are delineated by the geography at various levels. In California, population counts come from the Department of Finance, which corrects Census estimates for under and over count, other racial/ethnic groups, and provides extra- and interpolated intercensal estimates for certain populations, but none for specific Asian subgroups.
NCCC works with a demographer using an algorithm based on external sources of data (e.g., school-enrollment and voter-registration records, birth and death certification) to derive intercensal population estimates for five large Asian subgroups. NCCC uses cancer registry data to explicate patterns and incidents, treatment and survival for specific disaggregated subgroups.
Ms. Gomez explained that cancer incidence data (e.g., comparing racial/ethnic groups in the U.S. or migrants to those in their homeland) provide clues to cancer etiology as well as support public health prevention efforts targeted to specific groups. Studies show Asians vary regarding their treatment for cancer. NCCC plans follow-up studies to explicate cultural demographic clinical and economic reasons for women’s choices of treatment. An ongoing study of survival gauging trends by hospital characteristics indicated private and teaching hospitals are likely to collect patient information on race.
Ms. Gomez noted that while the percentage of the population reporting multiple races in the 2000 Census was relatively small, it is increasing. She noted strategies to analyze the data and collapse the number of groups to a manageable number. She said the Census and NCHS collaborated on bridging methods that will assign multiple race individuals to one group based on socio-demographic indicators. Minimum/maximum methods will be used by the Department of Finance in California to assign a given multiple-race individual to the race under-represented in that individual’s county. Ms. Gomez expressed the registry’s concern that population data from the Census and the Department of Finance is inconsistent with how race/ethnicity in the numerator data or number of cancer cases is collected.
Ms. Gomez expressed concern that the number of hospitals that do not collect birthplace data (32 percent) and a bias in the completeness of that data limit use of this information in analyses. She encouraged medical reporting facilities to uniformly, systematically and deliberately collect data on race/ethnicity for all patients and emphasized recording these data on birth and death certificates. She emphasized self-reporting, noting identification with multiple categories should be allowed and categories should be specific as possible.
She concurred with educating the public about the importance of collecting these data and ensuring data are of high quality. Noting one strategy could be systemizing data collection across hospitals, she suggested a national task force comprised of national health organizations, hospital representatives and researchers could formulate collection policies and guidelines. She called for continuing to advocate for the desegregation of subgroups.
Ms. Gomez urged researchers to engage community and advocacy groups in the design and conduct of their studies, and make data and results available in a language appropriate and applicable for a broad audience. Questions should be comparable to the Census or NCHS or at least have the same elements. Studies should incorporate measures of socio-economics, immigration and acculturation. Noting the biggest limitation in surveillance of cancer mortality and incidence data trends in Asians was a lack of intercensal population estimates for subgroups, Ms. Gomez emphasized advocating for the disaggregation of subgroups in analyses of this data.
Questions and Discussion
Asked whether men and women in these communities access the health-care system differently, Ms. Gomez said cultural reasons might be important in a woman’s choices of treatment for breast cancer. She noted that geography also made a difference.
PALAU TELEPHONE CONFERENCE
Ms. Tellie discussed Palau’s spiritual life, health, and how these integrated into the environment in dealing with community health. She noted a clash between Western and traditional medicine.
Dr. Dever said data collected must reflect the ethnic group’s health indicators; AAPI’s needs had to be supported by funding research in small populations. Data must allow for mapping health indicators and human-resource development, facility development, program design, service provision and evaluation of effectiveness. Data on medical health, environment, socio-economic and spiritual indicators must be collated to allow ease in application to newer technology.
Noting the Ministry of Health participated in 27 U.S. federal programs and each required different data sets in different formats, Dr. Dever said Pacific jurisdictions were challenged with collecting quality data appropriate for their needs. He recommended that U.S. federal agencies harmonize data requirements and processes and develop standardized, user-friendly software to promote data collection, analysis and reporting.
Palau AHEC, a joint John A. Burns School of Medicine, University of Auckland Faculty of Medicine and Fiji School of Medicine Program, will provide formal post-graduate education and training for regional physicians in general practice and community health. A community-development program (a full diploma program in community health with the Fiji School of Medicine) builds off the Ministry of Health’s work plan and will provide both appropriate data and physicians from Palau and the region trained in community-health processes.
Questions and Discussion
Dr. Dever said Palau was fortunate compared to its neighbors and other Pacific Islands “down south” in terms of health indicators like infant mortality, per-capita income, and the amount of health-care dollars spent per capita,.
Dr. Dever described a dilemma with an issue of intellectual property rights and elders who are the repository of information developed over more than 3,000 years. Elders are wary in a time outsiders scrutinized potions of plants for commercial purposes. He noted discussions about registering practitioners and encouraging the transfer of their knowledge to appropriate successors. Ms. Tellie said relations between practitioners of the traditional and Western medical system were beginning to be more positive (Western-trained physicians had begun to advocate that out-patients coordinate both treatments) but that even the concept of documentation could be offensive in Palau where knowledge is a private property.
He said travel was expensive, tedious and effected off-island medical referral. Many islands had a bandwidth problem; Internet and e-mail supported consultation and distance learning. He noted WHO, CDC and the Secretariat to the Pacific Community collaborated to set up an e-mail-based program to alert the islands about infectious disease outbreaks, and he encouraged including government and non-governmental organizations in any harmonization process.
Ms. Tellie said an increasing percentage (25-28 percent) of the population were laborers and professionals from Asia, Japan and the U.S. who helped build the community but introduced health problems. Dr. Dever said Palauans enjoyed excellent access to primary health care at reduced cost, but some foreign workers paid a month’s wages to go to the emergency room. The Ministry of Health tried to address these disparities. Data on race/ethnicity for indigenous people was collected; faced with SARS, materials were translated in several languages.
Dr. Dever said Palau was not only over-represented in non-communicable diseases, but bore the lingering problem of significant infectious as well as emerging diseases. Westernization and departure from the local diet, morbidity and mortality were similar to the U.S. Palau had been so successful in family planning that there was concern whether Palau could adequately support its aging population. The infant-mortality rate jumped dramatically, primarily due to issue of numerators over denominator: one extra death doubled the rate.
Reflecting that this was an unprecedented opportunity for members to learn more so they could try and be more helpful, Dr. Mays acknowledged the work of Dr. Stinson and Ms. Burwell and thanked the testifiers for their effort and “being up so early in the Pacific.”
HAWAII - Catherine Sorenson, Ph.D., Department of Health, Hawaii
Dr. Sorenson noted seven national data sets, including both government and private agencies, did not use 1997 OMB recommended race categories. Hawaii data revealed disparities in health by ethnicity and other population variables. Although ethnicity is one of the most widely collected data variables among HDH programs, a standardized method to define, collect or report race/ethnicity data is not employed.
Review of 100 race/ethnic categories from 11 HDH data sets indicated inconsistencies in spelling, combining of groups and formatting differences, as well as use of nondescript or general categories. Disparity across descriptors was largely based on past practices and linked to funding-agency guidelines, including federal reporting requirements. Ethnic-specific data were frequently only reported at an aggregate level. Use, collecting and self-describing of multi-race and multi-ethnic were problematic. HDH has no standard business rules for aggregating and handling this information. The only standard is in the Office of Health Status Monitoring, which has statutory authority and responsibility to collect and report births and deaths.
HDH’s decision rules were established in the 1940s for assigning infant race/ethnicity as reported by a parent. Infant ethnicity is derived from the father. If unknown or not reported, the mother’s ethnicity is used. Parents may report multiple ethnicities on the birth certificate, but only one is captured electronically. When Hawaiian is among multiple ethnicities listed, “Part Hawaiian” is coded. If non-Caucasian ethnicity is listed with Caucasian, the first non-Caucasian ethnicity listed is coded. If more than one Caucasian ethnicity are listed, the first is coded. Other programs have different rules. Death certificates, often based on observation, are problematic. Dr. Sorenson noted Hawaii has no major ethnic group.
In 1995, 58 percent of 15,000 births reported mixed ethnicity. In 2000, 61 percent were reported. Technical and programmatic issues preclude many divisions in HDH from reporting multiple ethnicities and those that attempt to report use an array of methodologies. Language preference and comprehension or first language is seldom requested. All program-level data collection instruments are solely written in English. Dr. Sorenson emphasized that the lack of denominators for discrete groups in AAPI populations is a barrier.
She noted the Hawaii Health Data Warehouse was being developed as a way to coordinate resources across administrations and divisions, ensure consistency in public-health data with national recommendations, and increase public access to health data. Design focused on integrating data from disparate data sets. HDH created a model that allows programs to continue utilizing current data collection processes for program-level data, while aggregating towards meaningful, consistent data. The model identifies a process whereby ethnicity data can be “clustered up” using a set of standards. OMB race categories disaggregate down to other ethnic groups. Adoption and application of the model will be implemented through consensus.
QUALITATIVE DATA ISSUES FOR NATIVE HAWAIIANS AND AMERICAN SAMOANS
Dr. Murry discussed research the University of Hawaii did with a HERSA state planning grant looking at Hawaii’s uninsured. Qualitative research drew upon economists, political scientists, anthropologists, sociologists, public health experts, and community groups (including policymakers, state workers, labor, Chambers of Commerce and business owners). Over 200 uninsured and over 50 providers were interviewed, including 10 Native Hawaiian healers. Provider locations included caravan sites on the beach, youth outreach program in Waikiki, hospitals, community health centers and social-service organizations.
Dr. Murry said Native Hawaiian health organizations did an excellent job of helping people get through the system in ways other organizations couldn’t do. Providers emphasized that: a lack of access to land and water impacted the health of Native Hawaiians, spirituality was important, and loss of cultural identity was a risk factor. Issues were especially severe among Pacific Island immigrants who lacked information about eligibilities and health system. Some spoke no English, had different cultural values, poverty was an issue, and they feared that, if they said anything, they might be “sent away.” Their expectations of a health-care system were based on an assumption that they would receive care.
Dr. Murry emphasized that until they collected data and identified small population subsets, disparities could never be discerned or subset populations identified. Emphasizing that the numbers were not necessarily a good way to determine whether something should be collected, she noted that there were 1,100 Kipingameranges in Micronesia and they “found all of them.” She emphasized that without this kind of data collection, the fact that Kipingameranges had the world’s highest incidence of leprosy (a virgin-soil epidemic) would not have been discerned.
She advocated collecting data on mixed populations, noting risk factors were linked to heritage. She cautioned that without collecting qualitative data (that tended to be marginalized) it was hard to understand issues, know what to change, and develop programs to address health disparities. She stressed the need for multiple translations, noting in Hawaii they had used 15 languages. Infrastructure was a barrier and an issue requiring a long-term solution from the bottom up, good research training, and the involvement of the communities experiencing health disparities.
Dr. Murry discussed guidelines for community-partnership research: (1) having the targeted community at the table, helping shape the research and values that go into the questions and training representatives as researchers, and (2) ensuring the data is in a language that can be understood and accessible. Noting it was good to build on existing partnership and that federal funding often required breaking them off, she suggested advocated long-term funding.
Qualitative Data Issues for Native Hawaiians and American Samoans
Mr. Ohta stressed harmonizing federal-agency approaches and requirements. Noting the collection categories still in use were created in 1972 when “only six data sets fit in computer systems,” he emphasized the inertia of setting up a data-processing system and the effort it took to create a more rational framework. He said additional sampling might always be needed in order to get data for small areas.
Qualitative Data Issues for Native Hawaiians and American Samoans
Mr. Spoehr said Papa Ola Lokahi, the Native Hawaiian Health Initiative, looked at over-sampling of smaller populations to identify need areas. He noted the political relationship to data-collection in the Hawaiian community and that the identifiers conveyed entitlements for the Native Hawaiian population.
Questions and Discussion
Mr. Spoehr explained there were state and federal definitions of Hawaiian and part Hawaiian. The federal definition required tracing one’s genealogy back to the Hawaiian Islands prior to 1778. The state definition denoted a 50-percent Native Hawaiian blood-quantum.
Ms. Sorenson said Hawaii’s model of consistent and detailed categories for race/ethnicity, aggregated and disaggregated ethnic groups was introduced last December at a HDH forum for division chiefs and data specialists who were appreciative that “something was being done.”
Dr. Murry said the University of Hawaii’s partners valued quotes and stories useful in gaining support for the advocacy to make these changes and funding. She emphasized that one could not understand change without qualitative questions. She noted that not wanting to accept charity, especially in using health care, was a big issue, especially with Native Hawaiians. Colleagues asked Dr. Murry to convey the issue of “carving out” funding for the infrastructure for data collection in all endeavors. Asked whether it was preferable not to collect data in the detail needed or risk breeching confidentiality, Mr. Spoehr said to ask the community.
SUBCOMMITTEE DELIBERATIONS OF TESTIMONY
Dr. Steinwachs said the model of the comprehensive community survey in Palau was attractive, but he pointed out that they did not discuss how those data might become a national resource. Members mentioned the utility of qualitative methods. Remarking that stories were powerful tools, Dr. Steinwachs wondered if national health statistics should collect them at an analytic level. Dr. Mays noted that qualitative methods described and analyzed processes by which outcomes occurred. Dr. Steinwachs revisited their earlier discussion about a paradigm shift that viewed health statistics as supporting the designing of interventions.
Recalling harmonization of data reporting was another recommendation they had to consider, Dr. Mays said they would prepare for a discussion of the different points at another meeting. Members noted a discussion with the Data Council also would be helpful.
Remarking that they heard disparities “get repaired” locally and that they were trying to build an information system that strengthened communities’ ability to reduce disparities, Mr. Hungate said certain core data sets had to follow consistent definitions and collections had to be in aggregable categories. He recommended establishing a voluntary standards organization to generate language and matrix, noting it had to be developed by the participants. Dr. Mays suggested asking at the retreat where this would be best worked on. She noted that Standards knew the methodology, issues and content. Dr. Steinwachs suggested that the Public Health Standards Data Consortium, with NCVHS’s oversight, could be a framework for voluntary activity. Recalling they heard that Census were often denominators, Ms. Paisano said the Bureau was a key resource in how they did classifications. Noting Census had made recommendations about data issues that the Subcommittee had to receive, Dr. Mays said would ask if they could attend the next Subcommittee meeting.
Members noted hearing a classification system had to: be hierarchical, maintain room for newly identified ethnic groups, capable of handling mixed or multiple ethnic groups used in combination, able to handle fluidity, flexible, with enough detail for local use, contain a temporal element so changes in race/ethnicity were not processed as a mistake, convey contextual effect as well as uncertainty, be useable in the field and relatively simple to execute at the local level.
Ms. Heurtin-Roberts said immigration patterns kept coming up and were collected in terms of where individuals were born, when they came into the U.S., and language preferred
Noting they heard that many places discussed had “next to nothing” in terms of data collection capability or capacity, members recalled testifiers emphasized that, if the action was to be at the local level, infrastructure and support had to be built for data collection.
Dr. Mays noted they had to understand the concept of nations, sovereignty and territories so they knew the relationships and issues relative to HHS. Expressing concern that they did not know enough about the infrastructure in terms of NHII, she suggested they consider a visit.
Mr. Lengerich concurred on the need to hear from the Census, adding that they also had to learn about issues that caused states to seek other denominators to produce estimates and suggested a general discussion about denominators. He also recalled that groups were generally supportive of data collection and research on small identifiable populations as long as there was community involvement in conceiving the research and collecting and using the data.
Mr. Hungate wanted to hear more about OMH’s sampling for the National Children’s Study and suggested asking which populations the study would provide statistically valid data on and the strategy for dealing with those missed. Dr. Mays said they would request again to hear from the Social Determinates, Sampling, Health Disparities and Environmental Justice Work Groups. She noted there was considerable planning in the background and that Mr. Sheip said he would prefer any recommendations “sooner rather than later.” Members will schedule a conference call to draft a letter for the January 28 full Committee meeting
Dr. Mays noted the issue of using geo-coding and GIS in terms of displaying data. She emphasized communicating the importance of these national surveys, why data on race/ethnicity was collected, and how it would be used. Recalling some said health informatic’s components were not widely enough disbursed, she suggested thinking broadly about substitutions.
Noting an updating function was recommended, Dr. Steinwachs suggested that NLM might provide resources or an appropriate government home for classification systems useful to people and the practice of medicine and public health. Ms. Heurtin-Roberts stressed the diversity of languages and cultures that had to be represented and the inadequacy of just OMB categories. Dr. Mays suggested bringing Ms. Dearing into the conference call.
Mr. Localio said the tradeoff and choice between data specificity and identifiability should be determined by desires of the sub-population of interest rather than current federal regulation’s one-size-fits-all approach. Ms. Burwell suggested it was even more basic: with informed consent, almost anything could be disclosed.
Mr. Lengerich proposed hearing at the full Committee from those at HHS looking at display of geo-reference data and privacy/confidentiality, standards and security, and threshold issues. Noting Mr. Localio will follow up on the privacy and confidentiality issues, Dr. Mays said they would introduce this at the retreat and see how it could be done. Citing Dr. Cooper’s absence, Dr. Mays noted members would schedule a conference call to work on the letter. Thanking the participants in this and the previous hearing held in Los Angeles, she adjourned the meeting.
I hereby certify that, to the best of my knowledge, the foregoing summary of minutes is accurate and complete.
Vickie Mays___________ _/S/_________________________9/23/04________
Chair Date