|
||||||||||||||||
|
||||||||||||||||
The following HTML text is provided to enhance online readability. Many aspects of typography translate only awkwardly to HTML. Please use the page image as the authoritative form to ensure accuracy. –1–
|
||||||||||||||||
 |
 |
Page 1 |
 |
 |
 |
|
| ||||||||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||||
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 1
Vital Statistics: Summary of a Workshop
–1–
Introduction
LITERALLY THE FIGURES OF LIFE AND DEATH, vital statistics hold an undeniable position of importance among a nation’s data resources. In their basic content, the measurement of births and deaths is one of the longest-standing data collection priorities of the U.S. government, dating to at least 1850. Over the past few decades, the specific program that gathers the data has evolved into a complex cooperative program between the federal and state governments for social measurement. The vital statistics themselves are a critical national information resource for understanding public health and examining such key indicators as fertility, mortality, and causes of death, and the factors associated with them.
Vital statistics begin as individual, geographically focused vital events that are registered or certified after their occurrence. Figure 1-1 provides a basic illustration of the process by which the records of these vital events make their way into the tabulations of the vital statistics of the United States. Today, the Vital Statistics Cooperative Program (VSCP) is maintained by the National Center for Health Statistics (NCHS). Registrars in 57 vital event registration areas—the 50 states, the District of Columbia, New York City (separate from the rest of New York state), and four U.S. commonwealths and territories—collect vital event data from local officials and transmit them to NCHS. NCHS compiles those data and issues public-use data files and other products, as well as analytical reports. In the past, the VSCP also compiled records of the vital events of marriage and divorce so that it was also a critical resource for documenting changes in American family and household structures. However, budget constraints on the program in the mid-1990s, combined with declines in reporting by the registration areas, led
OCR for page 2
Vital Statistics: Summary of a Workshop
Figure 1- 1 Flow of vital records and statistics in the United States
SOURCES: Adapted from Hetzel (1997: 62; reproduction of 1950 original) and National Center for Health Statistics (Appendix B, this volume: note 1). Fetal deaths are not explicitly indicated in this simplified diagram; Records of Fetal Death (bearing much of the same demographic information on characteristics of the mother as the Certificate of Live Birth) are completed and filed separately.
OCR for page 3
Vital Statistics: Summary of a Workshop
to the abandonment of the marriage and divorce series. Though marriage and divorce records are no longer compiled at the national level, the natality and mortality components of the vital statistics program have endured and continue to be essential to a wide variety of governmental and research uses.
At the workshop summarized in this report, then-Census Bureau director and former Texas state demographer Steve Murdock marveled at the degree to which the vital statistics on birth and death are taken for granted. He observed that these statistics have grown to be sufficiently critical to so many processes and assessments of the nation’s well-being that it is assumed that they always have and always will exist. Yet that is the paradox of vital statistics: data on births and deaths seem so fundamental and—at first glance—so simple a metric of social conditions that their existence is assumed to be automatic and their collection is assumed to be easy. A history of vital records collection commissioned in 1950 (Hetzel, 1997:43) captured this theme well:
Most people take vital statistics for granted, assuming that any statistics they need should be freely available as part of today’s culture…. Death rates are among the typical vital statistics that most people assume we have always had available and, without much effort, will continue to have. The real story is quite different: national statistics of deaths and births were achieved only within the present generation, after two centuries of intermittent struggle and building.
The U.S. vital statistics system relies on the original information reported (and the consistency of that reporting) by myriad physicians, new parents, and funeral directors; channeled through state and local information systems of widely varying levels of sophistication and automation; and coordinated and processed by a federal statistical agency that has experienced relatively flat funding for many years. The challenges facing the vital statistics system and the continuing importance of the resulting data make it an important topic for periodic examination, assessing both current and emerging uses of the data and considering the methodological and organizational features of compiling vital data.
1–A
THE WORKSHOP ON VITAL DATA FOR NATIONAL NEEDS
Pursuant to its charge to improve the statistical information and methods on which public policy decisions are made, the Committee on National Statistics (CNSTAT) of the National Academies convened a Workshop on Vital Data for National Needs on April 30, 2008. The workshop was designed to consider “the critical importance of adequate vital statistics for the statistical, research, and policy communities” and “improvements that are needed at NCHS’s vital statistics program.”
OCR for page 4
Vital Statistics: Summary of a Workshop
CNSTAT organized the workshop with support from its core sponsors, as well as additional support from the Census Bureau, the Office of Research, Evaluation, and Statistics of the Social Security Administration, and NCHS. The workshop drew nearly 100 participants, including the invited speakers and discussants. Prior to the workshop, the workshop’s planning committee asked that two background papers on two basic perspectives on the vital statistics be prepared in order to inform the discussion. Steven Schwartz (New York City Department of Health and Mental Hygiene) prepared a paper on the role of the states and vital registration jurisdictions and NCHS staff prepared one on the national-level, administrative perspective; these background papers are presented in Appendixes A and B of this summary. Appendix C adds information on the workshop agenda and lists the workshop participants.
1–B
SUCCESSES AND CHALLENGES OF THE VITAL STATISTICS PROGRAM
Workshop remarks by Harry Rosenberg (NCHS, retired) identified 22 major successes in the current vital statistics program; the full set of these successes is described in Box 1-1 and his comments provide a useful orientation to the range of topics touched on by the workshop.
Rosenberg elaborated on four particularly notable successes, each of which he said represents highly effective collaboration between the states and the federal government.
Production of the annual vital statistics files: Rosenberg argued that the greatest success of the VSCP is its basic regular product—statistical files covering 6–7 million records, including about 4 million live births and 2.5 million deaths. He said that the complexity of producing these statistical files cannot be overstated; production requires constant interaction between the states and NCHS in terms of receipt and control, intensive processing and quality evaluation, and the preparation of tables and public-use data files for analysis.
Revisions of the U.S. standard certificates: Rosenberg observed that it is important to acknowledge the periodic revisions of the standard birth and death certificates as an achievement of the program (see Chapter 4 for a detailed discussion and Appendix D for reproductions of the current standard instruments). Through these periodic updates, both administrative and research ends are achieved: they permit national compilation to keep abreast of the changing legal and administrative environment in the reporting areas while improving the data content of the final files to be of greater use to public health officials and researchers.
OCR for page 5
Vital Statistics: Summary of a Workshop
Box 1-1
Successes of the U.S. Vital Statistics System
Vital Statistics of the United States Data Files—production of annual statistical data files based on about 6–7 million records of births and deaths
Model Legislation—1907 template for state legislation on birth and death registration
Model Vital Statistics Act—fuller version of model legislation that expanded coverage of system to include marriage and divorce records; first suggested to states in 1942 and revised in 1959 and 1977
Completing the National Vital Registration Areas for Births and Deaths—as of 1933, all 48 states and the District of Columbia had adopted laws consistent with the model legislation, adopted the suggested birth and death certificates, and reported 90 percent (or greater) total registration of events
Tests of Birth Registration Completeness—series of three experiments (1940, 1950, 1964–1968) conducted by the Census Bureau to verify the completeness of reporting; 1964–1968 study suggested over 99 percent registration of births
Query Programs—development of manuals and training materials to assist source reporters (e.g., physicians coding causes of death) in consistently completing vital record data items
Current Mortality Sample (discontinued)—beginning in the 1940s, state registration areas directly forwarded a 10-percent sample of incoming death records to the national vital statistics office, thus enabling publication of national estimates of causes of death with only a 4-month lag after month of occurrence; discontinued around 1995 because of resource constraints
Classifying Causes of Death According to International Standards—World Health Organization’s International Classification of Diseases (ICD) (and subsequent revisions) adopted as coding standard for U.S. national vital statistics since 1900
Comparability Studies—tests of consistency of classification and coding of cause-of-death data after implementing new ICD revisions, last done after current ICD revision (ICD-10) implemented in 1999
Ranking Leading Causes of Death—1951 standard developed by national and state vital statistics offices for producing ranked lists of causes of death, separately for infants and noninfants
Mortality Medical Data System—set of software programs, originally developed in the late 1960s, to simplify cause-of-death coding and consistently resolve multiple cause-of-death codes; the programs have been adopted by other countries as well as the states, and recent revisions have worked toward a goal of permitting natural language entry of death causes rather than numeric codes
Race and Ethnicity Data (see Chapter 4)
Fetal Death Reporting—data on stillbirths have been collected since 1939, and World Health Organization standards for defining fetal death were adopted in 1950
Abortion Reporting (discontinued)—From the 1970s through 1993, NCHS and the states worked on a reporting system for induced terminations of pregnancies; the system was ended in 1993 because of resource constraints
OCR for page 6
Vital Statistics: Summary of a Workshop
Follow-back Surveys—periodic surveys to collect additional information on samples of birth and death records, conducted as early as the mid-1950s; effectively discontinued because of resource constraints (none conducted since 1993)
Training—training sessions on medical coding and specific methodological techniques (e.g., using the Mortality Medical Data System) were conducted by NCHS beginning in 1983, but have been reduced greatly in number in recent years
Mortality Workshops—wide-ranging practitioner workshops on improving cause-of-death data convened by NCHS in 1989 and 1991
Electronic Registration of Vital Events (see Appendixes A and B)
Electronic Microdata Sets—public-use data files of vital statistics on birth and death, with measures to protect the confidentiality of individuals, have been available in various formats since 1968: data tapes, CD-ROM, and most recently via the Internet
Revisions of the U.S. Standard Certificates (see text and Chapter 4)
The Linked File of Infant Deaths and Live Births (see text)
The National Death Index (see text)
SOURCE: Adapted from workshop presentation by Rosenberg and his follow-up paper (Rosenberg, 2008).
Creation and development of the linked file of infant deaths and live births: Rosenberg noted that infant mortality is one of the most widely used measures of overall health of a community. He said that, for this reason, the creation of a national linked file of live births and infant deaths has been a remarkably useful tool for understanding the medical circumstances and causes of death of infants and informing possible interventions to curb specific infant mortality types.
Creation and development of the National Death Index: Prior to 1979, researchers who wanted to conduct epidemiological studies on cause of death (particularly following up on previous studies using human subjects) had to contact every registration area separately and make arrangements for a death records check, purchasing death certificates that matched their study subjects, and coding the relevant information. Given the cumbersome nature of this process, the 1982 creation and continued updating of the National Death Index—a compilation of over 62 million death records for 1979–2005 (as of September 2007)—has been an invaluable resource for research. Rosenberg estimated that the National Death Index has assisted 1,500 research projects by performing about 4,300 searches. Among the noteworthy projects to make use of the index data are drug surveillance studies by pharmaceutical companies, evaluations of the cancer registries of
OCR for page 7
Vital Statistics: Summary of a Workshop
the National Cancer Institute and the Centers for Disease Control and Prevention, and studies of post-employment death due to exposure to hazardous substances conducted by large oil and chemical companies.
Yet the current vital statistics cooperative program faces significant challenges; indeed, Rosenberg’s tally of some of the key successes was accompanied by his notation of several shortcomings:
Struggle for timeliness in data production: As a cooperative partnership, the timeliness and quality of vital statistics rise and fall with the input of contributing registration areas. Physically, compiled files for the nation as a whole cannot be put together and released until the last state or registration area submits its data. Rosenberg also noted that the complexity of the data adds to the lag time between the end of a calendar year and when birth and death data for that year become available. At the time of the 2008 workshop, the most recent available vital statistics data covered births and deaths in 2005; the lag time between close of the data year and publication of the final data is 24–25 months. Rosenberg cited a survey of vital records participation by Friedman (2007), observing that this lag time has varied widely between 1985 and 2004, from less than 2 years to as many as 4 years.
Difficulty in achieving adoption of 2003 certificate revision: The most recent revision of the standard certificates of live birth and of death have been slow to win acceptance by the registration areas (see Chapter 4). This slow adoption has been particularly problematic because the revision implemented new standards for permitting reporting of multiple-race categories; until full compliance is achieved, the “national” vital statistics data are a patchwork of different reporting formats and styles for a critical data item.
Discontinuation of national collection of some vital records and downgrading of some quality assurance methods: There have been major casualties of data streams within the national vital statistics collections, which Rosenberg attributed principally to the inability to secure adequate and sustained funding for the system. The most prominent of these casualties is data on marriage and divorce; marriage and divorce records do continue to be developed at the state and local level but national-level collection and compilation was discontinued almost 20 years ago because of budget concerns. A portion of the vital statistics program also briefly collected national-level data on terminated pregnancies and the circumstances—data that, objectively compiled, would inform the ongoing national debate on abortion—but that system was also discontinued: if it were still in operation, Rosenberg said that it would add over 1 million additional records to annual vital statistics. Other reductions have been more subtle but are still very
OCR for page 8
Vital Statistics: Summary of a Workshop
consequential. Rosenberg said that NCHS has stopped coding occupation and industry of decedents, which can be important markers of both socioeconomic status and possible deaths due to workplace characteristics. The current mortality sample (meant as a quicker system for surveillance of death types) and the natality and mortality followback surveys (used for quality assurance) have been dropped (or effectively discontinued). Budget constraints have also led to reductions in NCHS-provided training courses for vital record collectors at the state and local level.
1–C
REPORT OVERVIEW
This workshop summary largely follows the topic blocks that were used in scheduling the workshop, though some rearrangement has been made when that seemed logical. Following this introduction, Chapter 2 briefly describes the current uses of vital statistics as presented at the workshop, particularly their use in deriving population estimates and various projections. Chapter 2 also discusses the emerging field of public health surveillance and the possible roles for vital statistics in that framework. In Chapter 3 we turn to the structure of the existing VSCP, from both the state or registration area perspective and NCHS’s perspective as the national-level coordinator and primary funder of the system. The workshop featured selected case studies of analogous partnership systems in the federal statistical system, and those are briefly recounted in the chapter. Chapter 4 considers methodological issues and, in particular, those raised by the 2003 revision of the standard birth and death certificates, which includes a new format for race and Hispanic origin data and preliminary findings from new public health data items included on the certificates. Finally, Chapter 5 summarizes the concluding session of the workshop, which identified different possible visions for the vital statistics program and featured a roundtable set of reactions from a discussant panel.