Click for next page ( 255


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 254
H NIH Committee on Minority Health and Health Disparities Research Definitions and Application Methodology FINAL REPORT (REVISED JANUARY 2004) The NIH Committee on Minority Health and Health Disparities Research Definitions and Application Methodology (see Attachment A, Committee) was established in the Fall of 2002 at the request of Dr. Elias Zerhouni (Director, NIH) and Dr. John Ruffin (Director, NCMHD) to determine operational defini- tions and a methodology to form the basis for future reporting on all NIH activi- ties related to minority health and health disparities. Given the inconsistencies in the manner in which Institutes/Centers (ICs) have previously reported on their minority health-related activities (see Attachment B, Minority Direct), the Com- mittee focused on developing a methodology that could be applied uniformly by all ICs and in as automated a way as possible. Guidance for two reports--the Minority Health Report and the Health Dis- parities Report--was developed (see Attachment C, Guidelines). In both reports, activities are reported in each of five categories: basic research; clinical research; infrastructure; research training and career development; and outreach. The Health Disparities Report builds upon the Minority Health Report, adding to it those activities directed at low socioeconomic status (low-SES) and rural populations that are not included in the Minority Health Report. The process followed by the Committee to develop the Guidelines is outlined in Attachment D (Activities). The Committee believes that limiting reporting according to the recommen- dation made by the Institute of Medicine (IOM) in its 1999 report (Recommenda- tion 3-3, page 140, The Unequal Burden of Cancer) would severely under-report and misrepresent the NIH commitment to minority health and health disparities 254

OCR for page 254
APPENDIX H 255 research. The IOM report asserts that percent relevancy accounting methods, such as the one used at the NIH, that calculate minority health research funding based on the percentage of minority individuals in the study population, "over- state the relevance of research for addressing ethnic minority health issues." The Committee believes, however, that restricting NIH reporting to "studies whose purpose is to address a priori research questions uniquely affecting ethnic minor- ity and medically underserved groups" (as the IOM recommends) would do a disservice to NIH efforts to foster attention to minority health issues. In fact, most of the major NIH-supported clinical studies are now designed with the statistical power to detect racial/ethnic differences in incidence, prevalence, or efficacy of treatment even when the study is not focused primarily on minority health issues. They provide valuable results applicable to minority groups, as well as to the majority population. With this in mind, the Committee established the following criteria for inclu- sion in the minority health and health disparities reports: For basic research, infrastructure, and outreach activities, include only projects targeted to minority health and health disparity issues. For clinical research, include non-targeted as well as targeted studies. For clinical research related to minority health, report a percentage of the award based on the ratio of minority participants to total participants (subject to a 25 percent minority participation minimum); for clinical research related to low- SES and rural populations, report a percentage of the award based on the empha- sis level assigned by CRISP. For research training and career development, include non-targeted as well as targeted activities. For non-targeted activities, include only dollars re- ceived by under-represented minority individuals. Minority Health Report For the Minority Health Report, the following operational definitions were developed: Targeted research (report at 100 percent) includes: basic research on a disease, condition, or biological process that affects exclusively or almost exclusively one or more minority populations basic research on a behavior that is found exclusively or almost exclu- sively in one or more minority populations basic research on whether and/or how the mechanisms of disease or basic biological processes differ in minority populations; or how behaviors differ in minority populations clinical research conducted exclusively or almost exclusively in one or more minority populations

OCR for page 254
256 EXAMINING THE HEALTH DISPARITIES RESEARCH PLAN OF THE NIH clinical research investigating whether and/or how the manifestations, consequences, or responses to treatment of diseases or other conditions differ in minority populations clinical research investigating how behaviors differ in minority populations. Non-targeted research (report at the percent minority participation, as long as it exceeds the threshold of 25 percent) is clinical research focused on prevention, diagnosis, or treatment of diseases or other conditions that affect minority populations. Targeted infrastructure (report at 100 percent) includes solicited programs that are focused exclusively on developing the research capabilities of his- torically black colleges and universities (HBCUs) and minority medical schools, as well as solicited programs designed to enhance the research resources available specifically to under-represented minorities at other in- stitutions educating or training high numbers of one or more minority popu- lations that are under-represented in biomedical or behavioral research. Targeted research training and career development (report at 100 percent) includes programs that focus exclusively on supporting research training or career development of under-represented minorities. Non-targeted research training and career development (report only dollars received by under-represented minority individuals) includes activities that support research training or career development of under-represented minor- ity investigators, but that are not focused exclusively on these population groups. Targeted outreach (report at 100 percent) includes: programs that focus exclusively on providing health-related information to minority populations programs that focus exclusively on providing information to health care providers about preventing, diagnosing, or treating diseases or other con- ditions in minority populations. Health Disparities Report Minority health issues are considered to be a subset of health disparities issues. The Health Disparities Report includes all activities in the Minority Health Report plus activities related to low-SES and rural populations. An activity that addresses minority health as well as low-SES or rural health is included in the Health Disparities Report based on either its percent relevance to minority health or its percent relevance to low-SES and rural health--whichever is higher.

OCR for page 254
APPENDIX H 257 For the low-SES and rural component of the Health Disparities Report, the following operational definitions were developed: Targeted research (report at 100 percent) includes: basic research on a disease, condition, or biological process that affects exclusively or almost exclusively low-SES or rural populations basic research on a behavior that is found exclusively or almost exclu- sively in low-SES or rural populations basic research on whether and/or how the mechanisms of disease or basic biological processes differ in low-SES or rural populations; or how be- haviors differ in minority populations clinical research on whether and/or how the manifestations, consequences, or responses to treatment of diseases or other conditions differ in low-SES or rural populations clinical research conducted exclusively or almost exclusively in low- SES or rural populations. Non-targeted research (report at 50/25 percent if coded in CRISP as Secondary/ Tertiary for low-SES or rural health--or at the minority participation percent, if higher) includes: clinical research related to the manifestations, consequences, or responses to treatment of diseases or other conditions in low-SES or rural populations clinical research conducted in populations that include low-SES or rural populations. Infrastructure is equal to the minority health figure for this category (due to difficulties in identifying institutions that educate large numbers of people from low-SES or rural populations). Research training and career development is equal to the minority health figure for this category (due to difficulties in identifying trainees of low-SES or rural backgrounds and institutions that educate large numbers of people from low-SES or rural populations). Targeted outreach (report at 100 percent) includes: programs that focus exclusively on providing health-related information to low-SES or rural populations programs that focus exclusively on providing information to health care providers about preventing, diagnosing, or treating diseases or other con- ditions in low-SES or rural populations. The Committee recommends that the attached Guidelines (Attachment C) be used to report on NIH activities in FY 2003 related to minority health and health

OCR for page 254
258 EXAMINING THE HEALTH DISPARITIES RESEARCH PLAN OF THE NIH disparities, but that a system be developed and implemented by FY 2006 that will accommodate more automated (perhaps even fully automated) and consistent reporting of these activities. ATTACHMENT A COMMITTEE ON MINORITY HEALTH AND HEALTH DISPARITIES DEFINITIONS AND APPLICATION METHODOLOGY Membership Andrew von Eschenbach, NIH/NCI (co-chair) Claude Lenfant/Barbara Alving, NIH/NHLBI (co-chair) Tommy Broadwater, NIH/NCMHD Harold Freeman, NIH/NCI Ann Hagan, NIH/NIGMS Milton Hernandez, NIH/NIAID Raynard Kington, NIH/OD Donald Poppke, NIH/OD Carl Roth, NIH/NHLBI Nathan Stinson, DHHS/OMH

OCR for page 254
APPENDIX H 259 ATTACHMENT B MINORITY DIRECT, TOTAL IC BUDGETS, AND MINORITY DIRECT AS PERCENT OF TOTAL IC BUDGETS Based on Old Definition Minority Health and Assistance FY 2004 President's Budget Minority Direct/Specifically Targeted Request from 2004 Direct as Programs Congressional Justification % of (Dollars in millions) (Dollars in millions) Budget FY 2003 FY 2002 FY 2003 FY 2004 FY 2002 Amend FY 2004 FY IC Actual Estimate Estimate Actual PB Estimate 2002 NCI $121.9 $128.0 $131.9 $4,177.8 $4,609.0 $4,770.5 2.9 NHLBI 241.7 256.0 262.5 2,569.8 2,762.4 2,868.0 9.4 NIDCR 19.1 20.8 21.3 342.3 369.3 382.4 5.6 NIDDK 133.2 144.0 150.0 1,560.0 1,703.1 1,820.0 8.5 NINDS 49.2 53.9 57.0 1,325.2 1,416.4 1,468.9 3.7 NIAID 331.3 375.4 392.6 2,339.8 3,981.1 4,335.3 14.2 NIGMS 141.7 154.0 154.0 1,722.9 1,849.0 1,923.1 8.2 NICHD 277.2 298.6 310.8 1,110.5 1,194.9 1,245.4 25.0 NEI 23.2 25.0 25.9 580.0 625.1 648.3 4.0 NIEHS 23.4 25.3 26.1 644.7 608.9 630.8 3.6 NIA 75.9 81.6 84.7 891.3 957.6 994.4 8.5 NIAMS 49.1 53.3 55.2 447.7 485.5 502.8 11.0 NIDCD 5.8 6.3 6.5 341.3 365.7 380.4 1.7 NIMH 26.7 28.8 29.9 1,245.3 1,332.5 1,382.1 2.1 NIDA 74.7 81.0 84.0 892.6 960.0 995.6 8.4 NIAAA 35.9 37.7 39.3 383.2 414.9 430.1 9.4 NINR 28.1 29.8 30.7 120.2 129.7 134.6 23.4 NHGRI 10.8 13.1 13.8 428.2 457.8 478.1 2.5 NIBIB 0.3 1.0 1.0 111.7 270.5 282.1 0.3 NCRR 48.7 52.6 55.3 1,010.2 1,065.1 1,053.9 4.8 NCCAM 7.4 8.0 8.2 104.3 112.4 116.2 7.1 NCMHD 147.2 173.9 180.3 157.4 185.8 192.7 93.5 FIC 1.0 1.1 1.1 56.8 61.8 64.3 1.8 NLM 8.8 9.6 9.6 275.4 305.9 316.0 3.2 OD 5.9 5.9 5.3 234.8 274.0 318.0 2.5 Total* 1,888.4 2,064.4 2,137.1 23,073.4 26,498.4 27,734.0 *May not add due to rounding. SOURCE: Budget Office, OD, NIH; Budget Office, NHLBI, NIH.

OCR for page 254
260 EXAMINING THE HEALTH DISPARITIES RESEARCH PLAN OF THE NIH ATTACHMENT C GUIDELINES FOR MINORITY HEALTH AND HEALTH DISPARITIES REPORTING (REVISED DECEMBER 2003) The Guidelines for Minority Health and Health Disparities Reporting de- scribe reporting requirements for two reports--the Minority Health Report and the Health Disparities Report. Minority Health Report: This report should include all funding for minority health research as defined by PL 106-525 (i.e., basic, clinical, and behavioral research on minority health conditions, including research to prevent, diagnose, and treat them.) Minorities are defined by statute as American Indians/Alaskan Natives (including Eskimos and Aleuts); Asian Americans; Native Hawaiians and other Pacific Islanders; Blacks; and Hispanics (i.e., individuals whose origin is Mexican, Puerto Rican, Cuban, Central or South American, or any other Spanish-speaking country). Health Disparities Report: For the purposes of this report, health disparities populations are minority populations, low socioeconomic status (low-SES) popu- lations, and rural populations. Therefore, this report will include all funding included in the Minority Health Report plus funding for activities that address health issues affecting low-SES and rural populations. General Reporting Principles: Any distinction between "targeted research" and "non-targeted research" is for internal methodological and computational purposes only. Only the aggre- gate (i.e., sum of targeted and non-targeted research) will be reported outside the NIH. For co-funded grants, each IC should report its own dollar contribution only. All reporting should be done at the subproject level (consistent with the decision of the Director, NIH, in his memo dated October 18, 2002, to report clinical research at the subproject level starting with FY 2003). Starting with FY 2003 reporting, ICs are required to report costs attribut- able to clinical research at either 0, 25, 50, 75, or 100% of the total award amount for all projects and subprojects (see memo from Director, NIH, dated October 18, 2002, referenced above). Consistent with this decision, when calculating clinical research dollars for minority health and health disparities, each IC should first multiply each award amount by the percent relevance assigned for clinical research. For FY 2003, ICs should report intramural minority health and health disparities research consistent with their procedures in past years. It is the inten-

OCR for page 254
APPENDIX H 261 tion of the NIH to develop a standardized trans-NIH reporting procedure for this area in the future. MINORITY HEALTH REPORT Each IC will develop its own list of activities and corresponding dollars based on the guidelines below. Only the total is to be reported, but separate guidance is provided for each of five categories: basic research, clinical research, infrastructure, research training and career development, and outreach. For extramural research, dollar estimates should be generated for each of the areas. For intramural research, dollar estimates should only be developed only for basic research and clinical research. Basic Research Report only targeted research projects. This would include research on: A disease, condition, or biological process that affects exclusively or almost exclusively one or more minority populations A behavior that is found exclusively or almost exclusively in one or more minority populations Whether and/or how the mechanisms of disease or basic biological processes differ in minority populations; or how behaviors differ in minority populations. All targeted projects should be included in the report at 100% of the funded amount. For multi-project grants, ICs may include a percentage of the funding for their core(s), based on the number of subprojects in the grant that perform tar- geted basic minority health research. Note that for this report, career development awards (K series awards) should be considered "research training and career development," rather than "research." Non-targeted basic research (i.e., basic research focused on diseases, condi- tions, or processes that affect minority populations but that does not satisfy the requirements of the definition of targeted research) should be excluded from the report. Clinical Research All targeted and non-targeted minority clinical research should be reported. However, funding for clinical research studies conducted on foreign populations

OCR for page 254
262 EXAMINING THE HEALTH DISPARITIES RESEARCH PLAN OF THE NIH should be omitted. Note that for this report, career development awards (K series awards) should be considered "research training and career development", rather than "research." Targeted Clinical Research Targeted clinical research should be reported at 100% of the funded amount. (Remember to adjust the award amount to reflect the clinical percent relevance-- 25, 50, 75, or 100%--assigned by your IC for each clinical research project. See "General Reporting Principles" on page 260 of this book.) This category would include clinical research: Conducted exclusively or almost exclusively in one or more minority populations Investigating whether and/or how the manifestations, consequences, or responses to treatment of diseases or other conditions differ in minority popula- tions. (Studying one minority group exclusively is sufficient to qualify as tar- geted minority research; i.e., no comparison group is necessary.) Investigating how behaviors differ in minority populations. Non-Targeted Clinical Research Non-targeted clinical research is clinical research focused on prevention, diag- nosis, or treatment of diseases or other conditions that affect minority populations. Reporting non-targeted clinical research The dollar amount for each non- targeted clinical project should be multiplied by the percent minority participa- tion in that study. (Remember to adjust the award amount to reflect the clinical percent relevance--25, 50, 75, or 100%--assigned by your IC for each clinical research project. See "General Reporting Principles" on page 260 of this book.) Note however that only non-targeted clinical projects with aggregate minor- ity participation of 25 percent or more should be included. Determining Percent Minority Participation ICs have several options for determining percent minority participation to use for this report. The IMPAC II Population Tracking database (POP Tracking) should be the "starting point." Minority participation represents the aggregate of ethnic and racial minority participation. If recruitment is not complete, projected numbers for minor- ity recruitment should be used. If projected numbers do not exist, the actual recruitment should be used. ICs may: use the percent minority participation contained in POP Tracking.

OCR for page 254
APPENDIX H 263 override the percent minority participation in POP Tracking if they deter- mine that the study should be considered "targeted" clinical research (by the definition provided above). In this case, ICs should report the project at 100% of its award amount. try to obtain participation data from other sources when data are missing from POP Tracking. For example, recruitment data (projected or actual) might be contained in the original applications or progress reports, even when they have not been entered into POP Tracking. In addition, not all projects in POP Tracking are required to have participation data; examples of such "exceptions" are studies with fewer than 10 participants and studies performing secondary analysis of data. In these cases, ICs should obtain participation data from other sources when possible. For example, project officers might have data on the original study population for studies performing secondary analysis of data. Infrastructure Report only targeted infrastructure activities. This would include solicited programs that are focused exclusively on developing the research capabilities of historically black colleges and universities (HBCUs) and minority medical schools, as well as solicited programs designed to enhance the research resources available specifically to under-represented minorities at other institutions educat- ing or training high numbers of one or more minority populations that are under- represented in biomedical or behavioral research. Examples of targeted infra- structure activities include the Extramural Associates Research Development Award (EARDA) and the HBCU Research Scientist Award. All targeted infra- structure activities should be reported at 100% of the funded amount. Research Training and Career Development All targeted and non-targeted activities should be reported. Training and career development activities include career development awards (K series awards), fellowships (F series awards), institutional training awards (T series awards), and minority supplements. ICs may support other training and career development activities, as well. Note that for this report, a research grant (e.g., R01) awarded to a minority investigator, or to an investigator at a minority serving institution, should not be included in the Training and Career Develop- ment category. Targeted Activities Report all targeted activities at 100%. Targeted activities are programs that focus exclusively on supporting research training or career development of under-represented minorities (e.g., Mentored Research Scientist Development

OCR for page 254
264 EXAMINING THE HEALTH DISPARITIES RESEARCH PLAN OF THE NIH Award for Minority Faculty-K01; Minority Predoctoral Fellowship-F31; Mi- nority Research Supplements). Note regarding Minority Research Supplements: IMPAC II does not in- clude separate records (type 3 records) for the minority supplements after their first year of funding. (i.e., After the first year, minority supplement dollars are included in the total award amount of the parent grant.) To avoid double count- ing, only new minority research supplements should be reported separately. Non-Targeted Activities Non-targeted activities are ones that, although not focused exclusively on minority groups, support research training or career development of under- represented minority investigators. For non-targeted activities, report only the dollars know to be received by under-represented minority individuals. For example: A fellowship (F32) awarded to an under-represented minority individual would be included at 100%. A T32 that supports 6 graduate students, only 2 of whom are under- represented minorities, would be reported at 33%. Outreach Report only targeted outreach activities. Targeted outreach activities are those that focus exclusively on providing health-related information to minority populations or information to health care providers about preventing, diagnosing, or treating diseases or other conditions in minority populations. All targeted activities should be included in the report at 100% of the funded amount. In general, non-targeted activities that provide education appli- cable to both minorities and non-minorities should be excluded. However, if a non-targeted outreach project has identifiable costs associated with specific activities targeted toward minorities (e.g., translating brochures into Spanish), those costs should be included in this report. On the other hand, simply taking a percentage of an outreach project for "general" minority-related activities is not appropriate. Note: For this report, conference grants (R13s) should be reported under the "outreach" category. HEALTH DISPARITIES REPORT The report will include all activities that address issues with low socioeco- nomic status (low-SES) and rural populations plus those that are included in the minority health report. The basis for the low-SES and rural component of the

OCR for page 254
APPENDIX H 265 report is to be a list of activities identified by a CRISP search. (See page 9 for the list of CRISP search terms used to generate the list.) For each grant, the percent relevance should be determined as follows. If any of the CRISP search terms is designated as "primary" for a particular grant, then its relevance is 100%; if no term is "primary," but at least one of the terms is "secondary," then its relevance is 50%; and if no terms are "primary" or "second- ary," but at least one term is "tertiary," then its relevance is 25%. For example, if an activity is coded as "primary" for low income and rural area and "secondary" for Medicaid, then its percent relevance should be 100%. Similarly, if a grant does not have any of the health disparity terms coded as "primary", but medically under-served population is coded as "secondary" and homeless is coded as "tertiary", its percent relevance should be 50%. Once the list of grants is generated: Sort it into the same five categories used in the Minority Report--basic research, clinical research, infrastructure, research training and career develop- ment, and outreach--and process according to the guidance below (presented under each category heading). For each category, add the activities compiled for the Minority Health Report, being careful not to double-count. For any grant that included in both the Minority Health Report and the health disparities CRISP search, count the grant only once and at the higher of the two percent relevances. (For example, if the CRISP list shows a grant at 50% relevance due to a low-SES search term, but the percent minority participation in the study is 63%, 63% relevance should be used for the health disparities report. Likewise, if a grant is considered 100% relevant for rural health but is reported at only 50% for minority research, it should be included at 100% in the health disparities report. (However, a higher percent relevance for an activity in the health disparities list should not be transferred back to the minority health report.) Add activities funded by any mechanisms not coded in CRISP (e.g., contracts). Assign each a percent relevance (100%, 50%, or 25%) based on whether the CRISP terms that would be appropriate would be of "primary," "secondary," or "tertiary" importance to that activity. (Use only those terms listed on page 9. See page 10 for a short description of how CRISP coders determine relevance for a particular term.) Delete activities from the CRISP-generated list that do not address low- SES or rural health issues. However, if a grant should be included that is not on the list (or should be included at a relevance different from that assigned by CRISP), the CRISP office (CRISP contact person: Ms. Dorrette Finch, 435-0656) should be contacted to work together to resolve any concerns about how the study is coded in CRISP. Multiply the corresponding dollar amount for each activity by the associ- ated (disparities) percent relevance to calculate the amount to include in this

OCR for page 254
266 EXAMINING THE HEALTH DISPARITIES RESEARCH PLAN OF THE NIH report for each activity. (Remember to adjust award amount to reflect the clini- cal percent relevance--25, 50, 75, or 100%--assigned by your IC for each clinical research project. See "General Reporting Principles" on page 260 of this book.) Basic Research Report only targeted research projects. For the low-SES and rural compo- nent, this would include research on: A disease, condition, or biological process that affects exclusively or almost exclusively low-SES or rural populations A behavior that is found exclusively or almost exclusively in low-SES or rural populations Whether and/or how the mechanisms of disease or basic biological pro- cesses differ in low-SES or rural populations; or how behaviors differ in minority populations. All targeted projects should be included in the report at 100% of the funded amount. Leave in all non-duplicate basic research activities from the Minority Report. Clinical Research Report the sum of all targeted and non-targeted research activities. For the low-SES and rural component: targeted research is: research on whether and/or how the manifestations, consequences, or responses to treatment of diseases or other conditions differ in low- SES or rural populations research conducted exclusively or almost exclusively in low-SES or rural populations. non-targeted research is: research related to the manifestations, consequences, or responses to treatment of diseases or other conditions in low-SES or rural populations research conducted in populations that include low-SES or rural populations. For calculating the low-SES and rural component, use the CRISP relevance (100%, 50%, 25%) assigned to each activity. However, for those activities that

OCR for page 254
APPENDIX H 267 appear in the CRISP low-SES and rural list and in the Minority Report, use the larger of: (1) the CRISP relevance or (2) the percent minority population used for the Minority Report. Be careful not to double-count. Remember to add any research contracts with an IC-specified relevance of 100%, 50%, or 25% to low-SES or rural health. Leave in all non-duplicate research activities (targeted and non-targeted) from the Minority Report. They should be reported just as they were in the Minority Report--at 100% for the targeted activities; at the percent of minority participation for the non-targeted activities. Infrastructure Report the same total here as that reported for Infrastructure in the Minority Health Report (due to difficulties in identifying institutions that educate large numbers of people from low-SES and rural populations). Research Training and Career Development Report the same total here as that reported for Research Training and Career Development in the Minority Health Report (due to difficulties in identifying trainees of low-SES or rural backgrounds and institutions that educate large numbers of people from low-SES and rural populations). Outreach Report only targeted outreach activities. For the low-SES and rural health component, include: programs that focus exclusively on providing health-related information to low-SES or rural populations programs that focus exclusively on providing information to health care providers about preventing, diagnosing, or treating diseases or other conditions in low-SES or rural populations. These would include outreach activities in the CRISP-generated list with a relevance of 100%. In addition, any targeted outreach activities addressing low- SES or rural health disparities issues that are not in the CRISP list (e.g., contracts) must be added. Report these activities at 100% of their funded amount. Leave in all non-duplicate Outreach activities from the Minority Report (reported at 100%). Activities providing education that is applicable to health disparity popula- tions, but that is also applicable to all populations, should be excluded. However, if a non-targeted outreach project has identifiable costs associated with specific

OCR for page 254
268 EXAMINING THE HEALTH DISPARITIES RESEARCH PLAN OF THE NIH activities targeted toward low-SES or rural populations (e.g., partnering with community-based organizations in low-income communities), those costs should be included in this report. On the other hand, simply taking a percentage of an outreach project for "general" health disparity activities is not appropriate. Note: For this report, conference grants (R13s) should be reported under the "outreach" category. CRISP THESAURUS TERMS TO IDENTIFY LOW-SES AND RURAL HEALTH DISPARITIES ACTIVITIES health disparities homeless income income maintenance low income Medicaid medical indigency medically underserved population migrant public assistance rural area social class social mobility social status socioeconomics urban poverty area HOW CRISP CODERS DETERMINE RELEVANCE FOR A TERM CRISP EMPHASIS CODES: P, S, T An emphasis code: 1) is assigned to every indexing term selected for a research project; 2) reflects the relevancy of the indexing term to the aims and objectives of the project; 3) may have one of three values - P = primary S = secondary T = tertiary

OCR for page 254
APPENDIX H 269 Primary Emphasis The primary emphasis code, "P", should be used to indicate the main aspects of a single project or subproject. Assigning "P" emphasis should be reserved for those terms with major significance to the project or subproject. A good common sense rule to follow would be to ask the question "would this project be wanted as primary on a subject search by the user or requestor?" It will always require judgment on the part of the indexer to determine how many "P" terms as well as which ones are assigned "P" emphasis. There will usually be 1-5 primary concepts per project or subproject. Many times it will require more than one P term to "cover" one of the major areas of a project. EXAMPLE: A project may be studying obesity as a risk factor for stroke. The following terms should all be assigned a P: disease /disorder proneness /risk stroke obesity If the example above were modified to emphasize, instead, the many differ- ent risk factors for stroke, the following terms should be assigned a P: disease /disorder proneness /risk stroke The specific risk factors, which are of less importance in this example, should also be indexed but with an "S" emphasis. obesity exercise diet Secondary emphasis "S" is the emphasis code for indexing terms lacking primary significance but of greater importance to the research than routine tools, subjects, methods, etc. Some specific terms are required to be indexed only at this level (i.e., cooperative study). Tertiary emphasis "T" is the emphasis code for indexing terms representing materials and methods such as instruments and routine methodology, research subjects, includ- ing animals, and tag terms, etc.

OCR for page 254
270 EXAMINING THE HEALTH DISPARITIES RESEARCH PLAN OF THE NIH ATTACHMENT D ACTIVITIES OF THE NIH COMMITTEE ON MINORITY HEALTH AND HEALTH DISPARITIES RESEARCH DEFINITIONS AND APPLICATION METHODOLOGY NOVEMBER 2002OCTOBER 2003 As a first step, the Committee surveyed the Institutes/Centers (ICs) to deter- mine how each currently reports on activities related to minority health and health disparities. ICs were asked whether they report at the project or sub-project level and how they determine the percent relevance for activities in each of four categories: basic research, clinical research, infrastructure and training, and out- reach. They were also asked which populations they include in health disparities reporting: minorities, low socioeconomic, rural, and other. Based on the data reported, the Committee discussed the pros and cons of the different methods used and decided upon some basic standards, such as requiring all reporting at the sub-project level. Members of the Committee reviewed definitions and reporting requirements in PL 106-525 (the Minority Health and Health Disparities Research and Educa- tion Act of 2000), and drafted a range of suggestions (including operational definitions) on how to report each of the four categories--basic research, clinical research, infrastructure and training, and outreach--for minority health and health disparities. The suggestions were used for discussion among the full Committee to determine the rules upon which each of the two new reports--the minority health report and the health disparities report--should be based. The rules for reporting on minority health include, for example, for basic research, reporting only targeted activities, and for clinical studies reporting a percentage of the total dollars based on the ratio of minority to non-minority participants in the study. The rules also specify that only racial/ethnic minorities, low socioeconomic sta- tus (low-SES) populations, and rural populations should be included in the health disparities report at this time. Draft guidelines were developed to aid the ICs in preparing their reports. Two different methodologies were proposed for producing each report. One meth- odology for the minority health report allowed the ICs to develop their own lists of minority activities, while incorporating the definitions and rules agreed upon by the Committee; the other methodology relied upon the NIH CRISP database for the lists of minority research activities. Both methodologies for the health disparities report relied upon the same CRISP-generated list for the low-SES and rural health component; the only difference was the methodology used to develop the minority health component of the report. The Committee then decided to conduct a pilot using the draft guidelines to determine whether they enabled easy and consistent reporting. Six Institutes (two

OCR for page 254
APPENDIX H 271 large, two mid-size, two small) were asked to participate in the trial: NCI, NHLBI, NINDS, NICHD, NIAAA, NIDCR. The ICs were provided the requisite instruc- tions and data files, and were invited to a training session that was held early in the trial for representatives from each IC. NHLBI staff were available for ques- tions and provided technical help throughout the trial. Upon completion of the pilot, a meeting was held at the request of the participating ICs to provide them an opportunity to share their experiences. Several collaborations were created to produce basic data required by all ICs in the pilot: CRISP coders worked closely with a sub-committee of the larger group to devise a way for the reporting to be based on the existing NIH-wide system. Lists of grants, created using a standard set of search terms, were provided to each of the ICs as their starting point for one of the minority health reports and for the low-SES and rural health component of the health disparities reports. The Information Resources and Technology Program at the NHLBI devel- oped a program for all the ICs to use to pull clinical data from the Population Tracking database, a subset of the NIH IMPAC II system. Again, this will allow all ICs to generate a list of qualifying activities that will be based on the same criteria. The Office of Extramural Research, Office of the Director, NIH worked out a way for participating institutes to report funding awarded to minority indi- viduals who have received NIH training grants and fellowships that were not specifically targeted to minority individuals or minority institutions. Results of the pilot were presented to the Committee in early September 2003. Most of the ICs participating in the pilot expressed reservations about the adequacy of CRISP as the basis for an automated reporting system for minority health and health disparities, noting that in their view it would seriously underes- timate the NIH commitment to health disparities. Therefore, after reviewing the pilot results and hearing the opinions of the pilot ICs, the Committee selected the methodology that allows the ICs to develop their own lists of minority activities (within the rules agreed upon by the Committee and described in the revised Guidelines) for the Minority Health Report in FY 2003. Consequently, the meth- odology selected for the Health Disparities Report is the one that incorporates this selected methodology for the minority health component and that uses CRISP only to generate lists of activities for the low-SES and rural health component. A sub-group of the Committee then conducted a final review and refinement of the definitions and review of the CRISP terms for health disparities. The Guidelines were revised to reflect the Committee decisions and final definitions. Looking ahead, the Committee recommended that a group be selected to develop and implement a more automated and perhaps fully automated system by FY 2006, possibly involving the CRISP database, for reporting minority health and health disparities activities.