An Issue Paper Prepared by the National Rural Health Association— April 2001


In rural minority populations, unique patterns and occurrences of disease produce health needs that sometimes differ greatly from those of the general population. In order to determine the true picture of need among these at-risk groups, data are essential. Yet, currently, there exist sparse and inconsistent health related data concerning rural minority populations. While there are a multitude of data being collected and analyzed nationwide, little pertains to rural minority populations. Thus, the results are not useful in terms of designing programs and initiatives for these populations. These limitations of data result in barriers that affect programs, services and efforts focused on health and quality of life issues. These barriers must be addressed collectively and individually as part of the initiative for data accuracy and completeness.


In order to deal more effectively with data problems and offer solutions, the commonly shared limitations of data nationwide must be considered. In a review of the major data bases, five common areas of need emerged concerning the limitations. The five areas of identification were:
In the review of materials concerning coding and minority populations, the single category of largest error was racial misclassification. Several published articles directly mentioned the use of the category "other" in data analysis methods and its adverse effect on developing an accurate picture of the true state of affairs in rural minority populations. The problem of race directly affects data that have been collected in the past, data analysis, and future data needs. Researchers have grappled with problems associated with racial misclassification and all its implications for decades. They express concern, however, that among minority populations, there may be inaccurate reporting of race. All national databases use death certificates and birth records for racial categorization if race is previously unknown. Yet if the racial classification is wrong, there will forever be false data, resulting in erroneous conclusions and analysis.

Mortality and incidence rates are affected by inaccurate coding, due to inaccurate information being present in state records and passed on to national databases. The researcher, program planner or service organizations must be aware of these gaps and limitations when designing, implementing, funding and evaluating any health program or data issue relative to rural minority populations. The categories of race for the purpose of data are needed to assist in defining differences involving factors such as ancestry, racial, ethnic, tribal or cultural groups. Variances in lifestyle, diet and health behaviors contribute to the health differential experienced by racial and ethnic populations. Data to be usable must be reflective of the minority populations and their needs.

Another area of concern is the collapsing of data into primary categories. This combining of data from several small sets into one larger category, to facilitate analysis and management needs, frequently results in lost or hidden data in regard to minority populations. Errors such as these often result in poor coding, or worse yet, coding errors themselves. In turn, the findings may not be representative of the population in question, leading to errors in inference. Racial misclassification on vital records, in disease registries, and census data, which are usually used for denominators in calculating rates, relies on self-identification of race. Vital records are often entered by a third party (funeral director/health care practitioner) based on appearance or surname. This is a major barrier to understanding and developing a true picture of the morbidity/mortality of underserved rural minority populations. The appropriate linkage of numerators and denominators in calculating health statistics is essential. Because denominator data for health statistics come from the Census Bureau, which relies on racial self-reporting, unrealistically low estimates of morbidity and mortality among rural minority populations in particular has occurred. At the same time, numerator data from vital records and disease registries is used and includes the same erroneous data counts based on racial misclassification.

Rural minority populations are at a disadvantage in today’s databases, due to coding errors, among other factors. Steps to correct this, such as implementation of commonly shared data linkages, will help overcome lack of statistical insignificance and undercounting. Coding using a universal system is essential today as technology moves more into computer based study procedures to facilitate data access, management and analysis.

The lack of standardized definitions upon which to base coding decisions is another area of concern. Lack of common definitions for use in broader data base linkage and merging of data bases are seen as a major handicap and are significant barriers to research. Comparison and combination of data from databases using differing definitions or coding is prevented, limiting sample size. The sample sizes for racial and ethnic populations in most data bases are not sufficient to reveal statistically significant differences among populations when those differences exist and do not allow analysis and inclusion due to low discriminatory power of the studies. Studies that focus on particular subgroups within racial and ethnic categories (e.g. tribal affiliations) have an even greater problem with insufficient statistical power. In the past, a common solution to this problem was to collapse all racial categories into three groupings: Caucasian, African American and "other". The result was and remains hidden important information about specific racial and ethnic populations. This not only obscured findings at the time, but prevents secondary analysis of the data in new studies or retroactive studies today.

"Undercount" refers to the difference between the number of persons counted in the census (or other programs) and the true populations size. Most researchers are concerned about undercount and whether the undercount is concentrated in certain groups or areas, leading to non-random biases in the data and to misinterpretation of results. Bias also is introduced when the methods of the study lead to an undercount. A telephone study or survey is an example in which the homeless, precariously housed individuals, those too poor to afford telephones or areas which do not have telephone service available (true in several isolated reservations) are all automatically undercounted. The undercounting problem is so widespread and of such duration that the U.S. Census Bureau does not question if it occurs, but rather measures the size of the undercount. Clearly, the undercount in national level data sets must be addressed before any type of significant data analysis or even data collection can be undertaken. To develop data about rural minority populations, a statistical significant number to allow formulation of conclusions must be a prerequisite of all research efforts.

Areas such as these highlight the critical need for inclusion of rural minority populations not only in national data bases, but in the process of data collection itself. National data that is accurate and represents the true state of affairs is essential. The diversity of health needs and health status at the community level or even at the geographic area level is masked because of these recorded errors in data bases as a matter of routine. Considering how culturally and geographically diverse persons are in the United States, this information is essential for the service organizations and programs as well as the community itself, to use in health care planning and needs assessment. The status of the community and the communities needs cannot be determined without it. The system of Federal funding is set up from a resource allocation standpoint, which uses people served and priority areas of needs, which are based on patterns and trends of utilization, morbidity and mortality, all derived from national data bases. Lack of rural minority population specific data confounds decisions concerning these areas and prevents the necessary targeting of services to address specific health problems among high-risk groups within these populations. Inconsistencies in data can have important implications for health planning and resource allocation, which affects generations of people, not just the present group members.


The National Rural Health Association recognizes that these problems have a significant impact on the quality of U.S. data concerning rural minority populations. Measurement error strongly affects all phases of the data process and its usefulness. There are reliability problems with the assessment of race that suggest there is an acute problem of undercounting racial and ethnic status for American Indians/Alaskan Natives, Asian Americans and Pacific Islanders and Hispanics, among others. Research and needs assessment of minority communities must be conducted using more accurate methods. In an effort to address these needs and issues, steps must be taken to effectively ensure adequate data concerning the health status, health care utilization patterns, health care financing and health outcomes for all rural minority populations incorporating the "umbrella" of the recognized five common areas of need previously listed.


In order to begin addressing these barriers the following recommendations are made.


Data and information are not available for all health issues of concern to all rural minority communities. Currently, systems are not set up to address the issues presented in this statement. To do so, there must be an understanding of what systems exist, what they do, and how they are set up to work. To do this, there should be a focus on two axes:

The definitions, even the studies themselves, should be determined by who is being studied, the purpose of the study, and why the research is being conducted. The data that already exist needs review and additional needs and gaps within these data needs to be identified. Data fitting the individual needs to be developed by asking the right questions and finding the best way to accomplish what needs to be done.

Common Terminology

Definitions used in data collection should use common terms which truly reflect the rural minority populations and their needs. In order to facilitate this, we recognize data collection of race and ethnicity issues are hindered by problems of definitions and measurements. Developments which would facilitate the linkage of existing data bases include:

Local Data

The need for more local level data is recognized. The development and review of mechanisms to allow this to occur and to facilitate the process of information and data collection and utilization are necessary. To allow data to be disaggregated from the national level down through all data collection levels would require the following steps:

Community "Ownership"

Community "ownership" of all data is a cornerstone of the data process itself. Cultural ways and the reality of rural community life must be concerns of researchers and be factored into all research efforts and initiatives. Cultural competency, cultural appropriateness and cultural sensitivity are needs which are recognized and are most fully met when the "community" is a full partner in each step of the data process. In order to encourage community partnerships, the following steps are urged:


The National Rural Health Association favors adoption of the strategies listed above in data gathering on rural minority populations. Five areas of common needs and four areas of focused efforts are urged in this issue paper. Their purpose is to achieve and maintain a quality of life within the multigenerational rural minority populations from the individual, community and local level upwards to the national level, and to eliminate all health disparities within these at-risk persons and groups and to advocate a healthy lifestyle with prevention as well as treatment as an accepted norm.

Return to home page