repayments of Federal and State chartered credit unions. This study should address the lack of direct monthly data for State credit unions and the adequacy of the sample of Federal credit unions used in obtaining these data.

2. The interagency committee on financial statistics should explore alternative methods for obtaining direct information on the use of mortgage loans on existing housing.

Mortgage and Construction Loan Interest Rates.Current information on actual interest rates charged on Federal Housing Administration and Veterans Administration home mortgages is not available. These data on interest rates at the time of commitment and closing for new and existing homes were collected from June 1972 to September 1977, initially on the recommendation of the Interagency Committee on Interest and Dividends which functioned during the price and wage control period of the early 1970's. The series was discontinued in 1977 on the basis that other information on mortgage interest rates was adequate for economic analysis.

The Interagency Committee on Financial Statistics should study the uses and differential movements of the discontinued series on Federal Housing Administration and Veterans Administration interest rates on home mortgages to assess the need for reinstituting the collection of these data.

Current information on interest rates for mortgages for apartments and commercial buildings is available from diverse sources-for example, monthly survey of 15 life insurance companies, monthly figures from the Advance Mortgage corporation (subsidiary of Citicorp), and weekly data for

apartments only from a survey of 25 savings and loan associations conducted by the Federal Home Loan Mortgage Corporation. There is no comprehensive survey of mortgage interest rates charged on these buildings.

The Interagency Committee on Financial Statistics should review the need, feasibility and costs for developing an "official" series on mortgage interest rates for apartment and commercial buildings.

Construction loans are an early source of mortgage credit. For commercial banks, current data on the volume and interest rates of construction loans are available. For savings and loan associations, information on construction loans is limited to the volume of closings.

Because of the importance of construction loans by savings and loan associations, the Federal Home Loan Bank Board should collect current information on interest rates for these loans.

Small Business Data Needs.-In the previous sections on Statistics of Income and the Quarterly Financial Report, note was made of the use of financial statistics for administering the small business program. However, at the present time there is no overall framework for assessing data needs for fostering small business, including those owned by minorities. The current review by the Small Business Administration of its data needs is an important step in remedying this situation.

When the current study by the Small Business Administration of its data base needs is completed, these should be included in the overall planning of economic statistics programs.

Introduction and Overview

Over the past two decades, health statistics at the Federal level have evolved from the measurement of natality, mortality, and morbidity and their determinants, to an expanded array of information designed to assist the health administrators, practitioners, planners and scientists, as well as others concerned with health issues. To paraphrase the World Health Organization sponsored Second International Conference of National Committees on Vital and Health Statistics, health statistics are no longer simply being utilized as background and descriptive information, they have been brought directly into the arena of policymaking.

Health statistics are defined to include statistics concerning the health of people; the health services. they receive; the manpower and facilities resources that provide the services; and certain basic demographic data which are closely linked to the epidemiology of health problems and the characteristics of the population being served. Also covered under health statistics are statistics on health attitudes and practices and on payment for health services.

The need for the Federal Government to be involved in the collection of health statistics stems from the Federal role in the training of health professionals, direct delivery of health care, financing of a significant portion of care delivered, protection of the public, prevention of disease and insuring reasonable public access to health care resources. Many of these roles and concerns overlap with those of States, localities and the private sector. Consequently, it is quite essential that Federal data activities be, to the extent practicable, integrated with, and built upon program activities, and that they be guided by appropriate statistical principles.

Guiding principles and goals for health statistics are basically no different from those for any other Federal statistical activity. They include costefficiency, minimizing respondent burden, reliability and validity, comparability, standardization, elimination of data gaps, and so forth. In attempting to comply with such principles, recent progress

should be noted in the development of minimum basic data sets, and the standardization of definitions for health statistics. These principles will be discussed further in other sections of this chapter.

Current needs for health statistics in the United States can be broadly placed under several categories: To describe the nature of the country's health problems and needs;

To assist in assessing how well and at what costs these problems and needs are being and can be met;

To assist in the allocation of the Nation's health resources;

To serve in a focal capacity in basic health research; and

To assist in governmental regulatory and planning activities.

The focus of this chapter will be on the broad range of activities undertaken to meet the needs identified above. A discussion of future data needs in the health area presumes a degree of clairvoyance beyond the scope of this report. This is not to say that attempts should not be made to predict or anticipate future data needs. On the contrary, such activities should be ongoing in organizations with statistical policy responsibilities. This chapter should not be viewed as a plan or program for health data activities for the next decade, but as a framework for more detailed examination and activity relating to the improvement of existing statistical programs.

The major Federal data collector in the health area is the Department of Health, Education, and Welfare (DHEW). Within the Department's components there exist both distinctive responsibilites for data gathering as well as areas of considerable duplication and overlap.

The Alcohol, Drug Abuse and Mental Health Administration (ADAMHA) surveys all alcoholism treatment programs which it funds, monitors all drug abuse treatment facilities in the United States, obtains detailed information from federally

sponsored community drug treatment projects, and collects comprehensive information on the Nation's mental health facilities and services.

The Center for Disease Control (CDC) assesses programs and problems in disease control, laboratory resources and proficiency, health education and occupational safety and health. CDC's disease surveillance program, its major statistical activity, includes collection of data on the incidence and prevalence of notifiable diseases such as tuberculosis, venereal disease, botulism, measles, birth defects, and lead poisoning.

The Food and Drug Administration (FDA) collects data designed to monitor FDA regulated products. In addition to collecting information on adverse reactions to the use of foods, drugs, and devices, FDA measures X-ray trends, conducts studies related to the health effects of radiologic exposure, and supports surveys of consumer perceptions of products regulated by FDA.

The Health Services Administration (HSA) operates a wide variety of statistical reporting systems related to its many and varied programs. These programs include Emergency Medical Systems, Health Maintenance Organizations, Community Health Service programs, Maternal and Child Health programs, and Indian Health Service programs.

The major statistical programs in the National Institute of Health (NIH) relate to the collection of biomedical and epidemiologic data in such areas as cancer, heart disease, stroke, hypertension, kidney disease and respiratory disease. These studies include major longitudinal efforts (e.g. the Framingham Heart Study), national registries (e.g., kidney dialysis) and numerous single-time studies targeted to individual Federal program interests.

In 1977, the Office of Health Policy, Research and Statistics (OHPRS) was created in the Office of the Assistant Secretary for Health to provide better access to timely research data and health statistics, and an analytic capability for policy analysis and evaluation support.

The National Center for Health Statistics (NCHS) and the National Center for Health Services Research (NCHSR) were transferred from the Health Resources Administration (HRA) to OHPRS in order to align health policy, statistics and services research functions under a single official reporting to the Assistant Secretary for Health. The Office of Policy Development and Planning and the Office of Health Information and Health Promotion, were also transferred to OHPRS to provide a close relationship

between health strategy and health policy functions, and health information and health promotion functions.

In addition to a clearer decisionmaking process, elevating NCHS to the office of the Assistant Secretary for Health (OASH) was intended to produce more effective use of NCHS as the agent responsible for the collection and analysis of general purpose health statistics and for coordinating statistical cooperation with State and local agencies. Health data collection efforts should become more systematic and uniform and the Public Health Service's (PHS) ability to rapidly obtain data needed for policymaking and to resolve data policy issues should be enhanced.

Placement of the National Center for Health Services Research (NCHSR) within the principal OASH policy analysis staff emphasizes the Center's crosscutting responsibilities as the primary source of technical and professional assistance in the area of health policy research and in the development of a national health strategy. This action also promotes NCHSR as the PHS focus for health services research.

The National Center for Health Statistics (NCHS) is the focal agency for the production of generalpurpose health statistics. The mission of the Center is "...to develop and maintain systems capable of producing reliable general purpose national descriptive health statistics on a continuous basis...." This is accomplished through the design and administration of data collection systems which provide a broad range of data for the United States through national sample surveys and the Cooperative Health Statistics System (CHSS). To fulfill its mission, the Center collects data on health status (the Health Interview Survey and the Health and Nutrition Examination Survey), health facilities (the Master Facility Inventory), vital statistics (births, deaths, marriages and divorces, etc.), health care utilization (Hospital Discharge Survey and National Ambulatory Medical Care Survey), and health professions statistics.

The reorganization of HEW in 1977 had other major impacts on the statistical organizations of the Department's health agencies. The creation of the Health Care Financing Administration (HCFA) led to the transfer of the health care statistical and operating components from the Social Security Administration (SSA) and the Social and Rehabilitation Services (SRS) to the new agency. Statistical and research functions were combined into the Office for Policy, Planning and Research (OPPR) in HCFA. HCFA now has responsibility for the Medicaid, Medicare and Professional Standards

Review Organization (PSRO) programs. These include the statistical systems designed to review and evaluate the Medicare and Medicaid programs.

Medicare program data, a byproduct of the claims review and payment process, are used to answer programmatic questions, such as how much money is being spent, how many people are being served, and what kinds of services are being used. In addition, there is a major research program designed to investigate alternatives to existing health care delivery and reimbursement procedures. Data on the Medicaid program are obtained from the participating State agencies. The Medicaid statistical systems are designed to assist in the administration and evaluation of health delivery and health care financing.

Although the major Federal data collector in the health area is the Department of Health, Education, and Welfare, a number of major health statistics efforts are conducted under the auspices of other major agencies. Nutrition data are also collected by the Department of Agriculture, drug abuse data are also collected by the Department of Justice, environmental effects on health are also measured by the Environmental Protection Agency and the Energy Resources Development Administration, data on health professionals are also collected by the National Science Foundation and the Bureau of Labor Statistics. Both the Veterans Administration and the Department of Defense collect data on the health status of their client populations as well as information on the characteristics of the institutions under their jurisdiction which deliver health care.

Adequacy of Health Statistics

Criteria employed to assess the adequacy of the Federal role in health statistics include timeliness, quality, comparability, and utility of the data collected, as well as the degree to which the collection efforts impose a minimum burden on the public. An additional criterion is comprehensiveness, or the existence of gaps in the data. Utility of data collected is perhaps the most crucial of these, encompassing to some extent elements of each of the other criteria.

Since Federal data collectors usually gather data for a multitude of users, utility must be examined from a number of perspectives including those of decisionmakers in the collection agency, Department and Executive Office levels, Congress, State and local governments, the research community, as well as other public and private institutions, and the general public.

Moreover, utility should be examined not only in

terms of utilization but also in terms of effect. The effectiveness of a data gathering program may be assessed on the basis of answers to questions about the extent to which data are being used by the various potential users, which (and why) potential users are not utilizing available data, the effect of utilization on policies, programs and the public, and the degree to which the activity is cost-effective and beneficial. Given this set of general criteria, to what degree and how well are Federal health statistics being used?

Because of the myriad activities in the health statistics area it is not practicable to assess or discuss all activities, or even all major activities. Consequently, the following discussion is designed to present some general viewpoints, and to relate these to certain specific statistical operations.

Basic Health Statistics and Health Status

Timely assessments of the health status of the Nation are essential for set priorities and for efficient management of Federal efforts in the health area. A major program for assessing health status is the vital statistics program of NCHS. Birth, death and infant death records have traditionally provided some of the most widely used data for geopolitical comparisons of health status. An additional major program in this area is CDC's National Disease Surveillance Program, a joint Federal-State information/data system. The Public Health Service began collecting morbidity data in 1878. Gradually, a program has evolved through which all States now provide reports of "notifiable" diseases to a central Federal collector/analyzer, presently the Center for Disease Control. Although the present system does provide essential data designed to monitor the incidence of communicable diseases and their spread, the system heavily relies on voluntary physician reporting which has been demonstrated to be variable and inconsistent. States differ in their authority to require physician reporting. Consequently, many incidence measures cannot be directly obtained through reported data, but must be estimated through statistical manipulations. The possibility of developing greater standardization in reporting from State to State, and of obtaining improved physician cooperation, are areas which need further exploration.

Two of the other major data gathering devices for assessments of health status are the Health Interview Survey which collects household health status data and the Health and Nutrition Examination Survey, which combines household interviews with actual medical examinations. These two collection efforts, sponsored by the National Center for Health. Statistics, gather data on a continuous basis; the