The Death Certificate: Civil Registration, Medical Certification, and Social Issues

Dennis L Peck. Handbook of Death and Dying. Editor: Clifton D Bryant. Volume 2. Thousand Oaks, CA: Sage Reference, 2003.

Thanatology is both an art and a science. As an art it emphasizes humanistic approaches to death, dying, and bereavement. As a science, it stresses the need for education, inquiry, systematic investigation, and research in approaching these once taboo subjects.

~ Kutscher, Carr, and Kutscher (1987:316)

Although death is an individual event, it is also of great social significance. Like birth, marriage, and divorce, death is a socially significant event that is celebrated by rituals and is institutionalized through documentation (Hanzlick 1997). As a social ritual, death is, in the words of John W. Riley, Jr. (1968), “the final rite de passage” (p. 19). As an institutionalized entity, death entails a legal process of death registration that in turn generates an important by-product—namely, vital mortality statistics.

The death certificate represents one of the officially reported documents surrounding an individuals’ life, or, more significantly, death. This document includes a variety of information, the most important being the cause(s) and circumstances of death (Hanzlick 1997). Indeed, the cause-of-death category is the most important item of information for public health analysts, government officials, and private business interests. According to Moriyama (1968), mortality statistics serve many purposes, such as “estimating a component of population growth and preparing population projections; delineating health problems, planning public health programs, and assessing health progress, and studying the natural history of disease” (p. 498). This kind of orientation receives strong endorsement from epidemiologists such as D’Amico et al. (1999), who argue, “Mortality statistics are important in epidemiological studies and enable us to generate hypotheses about disease aetiology, to identify trends in diseases and estimate their prevalence in population groups” (p. 141) (for an extensive discussion of such issues, see Willcox 1934; Rosen 1968).

Mortality statistics represent an important institutionalized by-product of death registration for other reasons as well. Information gleaned from the death certificate provides the data essential to the construction of the types of death rates, such as the total or general crude death rate, age-specific death rates, and the age-adjusted death rate. The creation of life tables useful to determine the expectation of life among a stationary living population also is possible using the accumulated death registry data. Such information allows for the study of sociodemographic correlates of multiple causes of death as well as more basic comparisons of death rates by age, sex, race, marital status, occupation, and level of education (see, e.g., Mackenbach et al. 1997; Johnson and Christenson 1998; D’Amico et al. 1999). Although governments, private business sectors, and professionals of various orientations are the biggest consumers of government statistics, statistics are no less critical to keep the citizenry informed, thereby satisfying an essential component of democracy (Keyfitz 1968; Straf 2001).

In the following sections, the discussion will focus on an historical overview of death registration and the evolution of the use of vital statistics and issues surrounding the recording of data considered central to the mission of epidemiological analysts. These sections are followed by a discussion of the importance of the death certificate and problems inherent in establishing the underlying cause of death. Finally, comments pertaining to the limitations of the recording process as social issues affect these conclude the discussion of the death certificate.

The Evolution of Death Registration/Vital Statistics

The evolution of the registration process and the use of vital statistics have a notable history. The historical essay “Statistics: History” crafted by Walter F. Willcox (1934) offers an enlightening view of a term that is undoubtedly taken for granted by most whose endeavors depend on the use of data. Coined in the late Middle Ages, the term statistics is derived from the Latin phrase ratio status and the Italian equivalent ragione di stato. Initially employed in the study of practical politics, the term has undergone a dramatic transformation from its original 17th- and 18th-century meaning and application, intended to enhance understanding of the political science of European nation states. Later, influenced by the statistician Quetelet during the early 1800s, the numerical element of a two-part conceptualization of statistics began to take root, and this numerical component dominated definitions of the term as we currently think of it. By the time Willcox (1934) wrote his essay for the Encyclopedia of the Social Sciences, statistics had evolved in application to mean “the numerical study of groups, and a narrower one, the numerical study of social groups” (p. 357).

Vital statistics are data recorded as important events in the life of an individual. Such statistics are recorded at the time of an event such as birth, marriage, divorce, and death. Parochial registration of baptisms and burials were recorded as early as 1538, and the first numerical study of death records, dating to 1662 and authored by John Graunt, is based on these data (Vold 1958:163; Grebenik 1968:341). Titled Natural and Political Observations upon the Bills of Mortality, Graunt’s work, according to Willcox (1934), represents “an honest and intelligent effort to interpret by inductive methods the figures showing the number of deaths in London during the preceding third of a century…. The study was informed by the spirit of modern science” (p. 357). John Graunt’s recording categories included causes of death; this kind of early scientific inquiry underscored recognition of the importance attributed to the recording of vital information encouraged by the evolving system of registration. In this area, Graunt’s work was to have, in the words of Shneidman (1976), “great social and medical significance” (p. 246).

The investigation of death by coroners originated with the Articles of Eyre in 1194 England. These articles called for the election of three knights and a clerk for each English county whose assigned task was to protect the financial interests of the crown when the death of other than an Englishman occurred. By the late 1400s, English justices of the peace had assumed the coroner’s fiscal duties, and by 1538, parish clergy were required to keep record of christenings, marriages, and burials. This form of graves registration documented by English clergy marked the origin of death and burial records (Cordner and Loff 1994; Hanzlick 1997), although Henry VIII may actually have initiated the registration of deaths in 1532, when widespread panic and fear existed over the plague (Keyfitz 1968).

The oldest systems of registration of vital statistics are found in Scandinavian countries beginning with Finland in 1628, Denmark in 1646, Norway in 1685, and Sweden in 1686 (Dublin 1933; Grebenik 1968). Throughout the course of the 17th and 18th centuries, improved documentation of public records and registries made quantitative analysis of health-related problems possible, especially when civil registration of births and deaths was first introduced in England and Wales in 1836 (Grebenik 1968) or in 1837 (Maudsley and Williams (1996).

By 1836, concern about the concealment of homicide and grave robbing led to passage of the Birth and Deaths Registration Act, resulting in the requirement that a death certificate be issued prior to burial in England and throughout Europe (Hanzlick 1997). In the 18th and early 19th centuries, the practice of body snatching to supply cadavers to English physicians for instruction and research prompted the British Parliament to require a registrar’s certificate or a coroner’s order prior to burial (Williams 1987; Cordner and Loff 1994). It was not until 1911, however, that the gathering of data on the cause of death, undertaken in Sweden, established the process of officially recording each death (Kastenbaum 1989). Death Registration in the United States

Colonists who subscribed to the European custom brought the graves registration system to the North American continent. In 1632, the Virginia colony passed a law requiring a representative from every church and parish to present at court a register of christening, marriage, and burials for the year. This practice continued until 1639 when a registration law passed for the colony of Massachusetts mandated that responsibility for registration be transferred from the church to government officials, who began the task of recording vital events—namely, for every marriage, birth, and death. This law was strengthened in 1644 when registration was made obligatory and, later, in 1692, when penalties for failure to register each birth and death were reinforced.

The separation of church and state, as required by the Constitution of the United States, had implications for death registration and the resultant vital statistics. This early interest in vital statistics was based on fear of epidemics and a belief that population size is the most important element of state power, especially its resources and military strength (Willcox 1934; Keyfitz 1968). As local and state governments developed an infrastructure to deal with 19th-century epidemics such as cholera, responsibility for death registration increasingly shifted to urban health departments and boards of health, finally leading to a more comprehensive system of registration (Grebenik 1968; Hanzlick 1997). These agencies also developed epidemiological strategies that led to the collection of information pertaining to regions where infectious diseases were localized and the number of deaths caused by infectious disease could be enumerated. As noted by Hanzlick (1997:265; Willcox 1934:358), Maryland and Massachusetts adopted a statewide registration law in 1842, and by 1851, seven states had enacted similar death registration laws. It was not until 1900, however, that under law a national centralized system of death registration was established. Finally, in 1933, all states had been accepted as a registration area.

Centralization of death registration was obviously a slow process. The earliest U. S. mortality statistics were based not on death certificates but on the information collected during the decennial census. Beginning in 1850 until 1900, the decennial remained the single source of official mortality statistics in the United States. Then, in 1900, mortality data were gathered and issued annually by the federal Bureau of the Census on the basis of death registration (Dublin 1933).

Until 1946, the U.S. Bureau of the Census functioned to collect and report national registration information. But in 1946, this function was assigned to the Public Health Service, an organization that created a National Office of Vital Statistics. In 1960, this office was merged with the National Health Survey program to form the National Center for Health Statistics, which in 1987, became a part of the National Centers for Disease Control. This kind of centralization continues as the responsibility for the collection and management of national mortality data remains with the National Center for Health Statistics.

The Death Certificate

Although the concept of death certificate is not defined by statute in all jurisdictions (Maudsley and Williams 1996:60), this document has three basic purposes: (a) It registers the fact that a person has died (the registration process); (b) it certifies an opinion regarding the cause, circumstances, and manner of death (the certification process); and (c) it provides information that may be used for postmortem statistical and epidemiological research purposes (Hanzlick 1997; D’Amico et al. 1999). Such information may be used by local-, state-, and national-level government agencies and private sector organizations to monitor morbidity and mortality and for scientific study, planning, changing public health programming, and other-health-related initiatives, as well as for legislative action and research funding. In the words of Hanzlick (1997), “the importance of accurate and complete death certification cannot be overstated” (p. 266).

The cause of death may represent the most important item of information documented on the death certificate. Certainly, appropriate recording procedures are essential for quality measures and for establishing the validity of national vital statistics. Not all interested parties adhere to a common set of criteria for adding to this storage of knowledge, however. For example, coverage of this topic in West’s Encyclopedia of American Law (“Death Certificates” 1998) only adds to a recognized confusion over the term death certificate (see Maudsley and Williams 1996:60) in that the general focus of this discussion deals with blame, such as whether a homicide (criminal law) or a wrongful death (tort law) occurred through the actions of another. Although such data details a form of causal information important in the detection of crime, many other types of information also are recorded on the death certificate. One of the most important of these is the cause of death, generated through application of the International Classification of Diseases (ICD), a three-digit whole number with one or more decimal points added. This schema has proven increasingly useful to government agencies, commercial interests, and researchers (Kastenbaum 1989; Hanzlick 1997:266). Specificity of wording on the death certificate is of obvious importance in order that the ICD coding procedure can be accurately applied. On the other hand, such accuracy may be a function of how well the decedent was integrated into the health care system.

In addition to recording information that may hold global consequences, in the United States, the death certificate must include certain data, and it is required that this document be signed by a certifier such as a physician or, in the case of equivocal death (such as homicide, contagious disease, suicide, or accident), a medical examiner. Such requirements include information pertaining to the decedent, the names of parents, the disposition of the body, circumstances of death, and contributing cause(s) of death.

The underlying cause of death is yet another important part of the mortality statistics that can be derived from information recorded on the death certificate. This information has in the past and continues in the present to serve well the needs of public health officials whose interests lie in determining the direct and indirect cause(s) of death. In the past, this information was useful for the study of acute infectious diseases. Although establishing a similar underlying cause of death involving chronic disease and illness is more challenging, the need is no less important. Simply put, the death certificate represents a public surveillance document that fulfills important legal and social needs (Messite and Stellman 1996).

Recording the Underlying Cause of Death

At the beginning of the 20th century, infective and parasitic diseases constituted the major causes of death in the world population. As noted by Moriyama (1968:502), documented cases of pneumonia and influenza, tuberculosis, diarrhea, enteritis, and childhood diseases represented the major causes of death in 1900. In the wake of the vast positive effects of the medical health revolution of the past century, chronic diseases have replaced infective and parasitic diseases as the most important public health problem.

Accurate documentation of the cause of death has far-reaching consequences. As noted by Grace Brooke Huffman (1997:1288), precise documentation of the immediate cause of death, the intervening cause of death, and the underlying cause of death holds many implications of a legal and public health nature. In the first instance, information gleaned from the death certificate is used to settle estates and to determine beneficiaries of insurance policies and pension funds. In the second instance, information based on death certificate data provides statistics that are useful to those in the health care professions to assist these individuals in making determinations as to patterns of disease, developing prevention strategies, and identifying weaknesses or deficiencies in health care delivery policies.

Known as diagnostic coding, the immediate causes of death differ from mechanisms of death. Recent initiatives to educate practitioners on how to fill out the death certificate is focused on coding categories that include immediate cause, underlying cause, and other significant conditions contributing to death (Franke 2000). Anderson T. Kircher (1987) offers what on the surface appears to be a straightforward set of definitions to guide death recorders. According to Kircher, the following definitions should serve to guide these efforts:

Immediate cause of death: the condition or complication that immediately precedes the death

Intervening cause of death: other conditions brought about by the underlying cause and contributing to the death

Underlying cause of death: the disease or injury that initiated the events resulting in death, or the circumstances that caused the fatal injury

At present, these distinctions are responsible for generating a substantial literature intended to call attention to a pressing need to produce accurate public health data documented by those whose task it is to certify death. Before this important call for refinement of the data can be accomplished, however, increased education of and more diligent effort by certifiers are required (Messite and Stellman 1996). This call to action on behalf of the medical community may not easily be achieved, according to Maudsley and Williams (1996). Education per se may be necessary, but it is not sufficient to solve the problems involved in recording the underlying cause(s) of death. These analysts argue that a traditional perspective on improving the accuracy of death certification is lacking in substance, proposing a complete reorientation in thinking is needed. Such a reorientation would lead to what Maudsley and Williams (1996) and D’Amico et al. (1999) refer to as “evidence-based educational interventions.” The orientation toward proper certification is strong among researchers, particularly those who pose that physicians need to be aware that certification is a fundamental requirement for building an appropriate base of data for epidemiological study.

The Certifiers of Death

In the historical past, certifiers of death included unregistered medical practitioners, midwives, and chemists. The evolution of the death registration process and legal mandates changed this. But who now legally certifies the cause of death? The answer to this question may be cause for concern, given that community health physicians, private practitioners, justices of the peace/coroners, medical examiners, and funeral directors are eligible to conduct this task. But people in occupations such as justice of the peace, coroner, and funeral director are medically untrained. The result is that recording of inaccurate causes of death is a frequent occurrence. Such recording errors appear to be consistent with those noted long ago by analysts such as Breed (1963) and Douglas (1967) who, among many others, observed that the inherent sources of error found on the death certificate for many categories greatly diminishes the quality of data available from official registration records.

A cure for this recording problem is not unwarranted. In addition, many medically trained personnel regularly complete death certificates. Unfortunately, many of these individuals are unfamiliar with the basic principles in writing accurate and appropriate causes of death statements as required by World Health Organization (WHO) protocol (D’Amico et al. 1999; Franke 2000:31). In other instances, the certifying physician may be unfamiliar with a patients’ medical history. This is especially the case when emergency room personnel are involved in making a diagnosis of the cause(s) of death.

In one study conducted in Naples, Italy, D’Amico and his colleagues (1999) found that more than half of the ill-defined causes and multiple causes of death may have been misclassified. According to these analysts, when the cause of death recorded on a random sample of death certificates (N = 372) was compared with a modified code developed from interviews conducted with the certifying physician or from clinical records, discordance was found to be quite high (54.5%). A similar study, conducted in the Netherlands (Mackenbach et al. 1997), found that competing causes of death were found for more than 46.2% of the deaths (N = 5,975). Competing causes were especially high when respiratory diseases were involved. An even more recent study conducted in Japan (Sharp et al. 2001) further underscores the significance of improved technologies and the certification process. Comparison of death certificate causes of death recorded for the 1958 to 1994 period with information gleaned from a review of pathology records led Sharp et al. (2001) to conclude that “errors in death certificate causes of death and changes in liver cancer diagnostic techniques have had an important impact on the reported incidence of this disease” (p. 751).

On the other hand, when it occurs, improvements in certification can have positive effects. In a U.S.-based study of Michigan State death certificates covering the 1989 to 1991 period, researchers found that when multiple causes of death are more accurately documented, the certifier may be aware of most if not all health problems experienced by the decedent. Individuals who are well integrated into the health care system, as opposed to those who are not, are more apt to benefit from preventive and curative medicine. Such factors weigh heavily in the accuracy of the death certification process (Johnson and Christenson 1998).

Death Certificate Information: Diagnosis versus Wording

Each state requires that the manner of death be determined and recorded on a death certificate. In some jurisdictions, anyone who knowingly falsifies a statement on a death certificate can be sentenced to prison for 2 to 10 years and fined up to $10,000 (Franke 2000). A properly documented death certificate lists only one condition for each line, and these are required to be in accordance with the guidelines established by the WHO ICD guideline of coding (D’Amico et al. 1999). But the cause of death stated by the certifier is, according to Hanzlick (1997), an opinion based on information available at the time of certification. Thus the certifier may not be legally accountable for the accuracy of the documented cause of death if the opinion “has a factual basis with reasonable medical probability” (p. 268).

It is expected that a certifier will certify the cause of death to the best of his or her ability. This expectation may be diminished, however, according to Maudsley and Williams (1996), because the “accuracy of causes of death is influenced by diagnostic and semantic considerations, themselves influenced by various characteristics of the certifier, the [death] certificate, the [status of the] deceased, and the ‘true’ cause of death” (p. 61). Some support for this contention is found in a report authored by Tsung-Hsueh Lu et al. (2001). Interested in exploring the diversity among physicians in death certification and its relationship with the complexity of the causal sequence of death, these analysts reviewed physician (N = 121) responses to the request to document cause of death on four dummy death certificates. The exercise, based on four case vignettes, represents fours levels of diversity: These are (a) wording and diagnostic semantics, (b) combinations of expression, (c) correct certification format, and (d) concordance with referent Universal Certification of Death (UCOD). The results indicate that the rates of agreement with the UCOD varied greatly with the level of complexity of the case.

Yet another important consideration in the certification process is the current de-emphasis on the use of the autopsy to determine the underlying cause of death. This concern is, according to Cordner and Loff (1994), Maudsley and Williams (1996), and Hanzlick (1997), that diagnostic information is lost. The increasing dependence on technological devices for analysis may enhance the clinicians ability to determine a diagnose disease, but this diagnosis also may mask the “true” cause of death. In sum, the recording of the underlying cause of death can be influenced by many factors, not the least important being medical recording procedures, the deductive reasoning process of the certifier, and an inappropriate use of WHO guidelines.

The Death Certificate and Types of Disease

A large medically related literature has been generated within the past decade regarding the need to specify certain diseases as the underlying cause of death. Included in this literature are articles pertaining to, for example, diabetes (Tierney et al. 2001), types of cancer registration (Brenner 1995; Sharp et al. 2001), dementia and Alzheimer disease (Olichney et al. 1995), concussion and alcohol (Shkrum and Ramsey 2000), sudden infant death syndrome (SIDS) (Nishida and Swaguchi 2000), coronary heart disease (Coady et al. 2001), patterns of occupational death (Conroy and Sciortino 1997), and tuberculosis (Washko and Frieden 1996). Certainly, each of these causes is important, but the proliferation of this literature seems to be much more than fortuitous at best. Indeed, it is possible that the vast amount of research money available for medical research may indeed be correlated with this increased interest in ensuring accuracy in the death certification process. For this reason, it is imperative that death certification diagnoses be complete, accurate, and current. As Maudsley and Williams (1996) observe, “Inaccuracies in documenting the underlying cause(s)] will ultimately affect health-related decisions for both individuals and populations” (p. 60).

Research conducted by sociologists and other policy-oriented researchers have much to offer when access to death certificates is allowed. For example, Johnson and Christenson (1998) demonstrated that during a 3-year period, age at death and underlying multiple causes of death were the most important factors in the sociodemographic and epidemiological correlates of the total number of deaths reported on the Michigan State death certificate. Older individuals died from underlying causes such as diabetes, hypertension, or septicemia, whereas young adults were more likely to have homicide, suicide, or accident recorded as the cause of death. It is obvious that such findings hold important implications for public health policy.

Social Consequences and Implications

Positive Aspects of a Registration Document

The death certificate is, as Quigley (1994) notes, “A document by which death is legally recognized and which is permanently filed with the state or district health department” (p. 49). This legal source of information contributes to the needs of the body politic that continues in function and structure after the individual no longer is a factor.

Beginning with the brief history of the registration process, positive aspects relative to the appropriate recording of vital data on the death certificate have been addressed throughout this chapter. Among these is the accumulation of sociodemographic data useful to governments, the private sector, and researchers; information pertaining to causes of death; the debate over appropriate recording procedures; and the process of refining knowledge of disease. Aside from these vital functions, the death certificate also provides information useful to population forecasting, adding to our knowledge of life expectancy, and to those interested in genealogical study. As an historical document that recounts significant portions of an individual’s life, it thus adds, if only in a small way, to the history of a community.

One critical component of the certificate of death is that of medical certification. But other aspects of this document serve to inform us of the overall ethnic and racial composition of a population, the levels of education, how people respond to the need to provide specimens for scientific study or the donation of organs, and the level of acceptance of nontraditional methods of body disposal such as cremation. Although less direct, a modest indication of movement patterns of the population also can be suggested from information recorded on this document. Finally, if certified by a medical examiner, information and testimony relating to this document can be used in criminal and civil courts of law.

Despite the numerous positive indicators surrounding the death certificate, some less than positive aspects corresponding to social attitudes and values can be identified. In the final sections, social deficiencies relating to information documented or not documented on the death certificate are addressed.

Deficiencies of the Death Certificate

More than 100 years ago, in 1901, Samuel W. Abbott made a strong appeal for reliable death registration book-keeping (as cited in Shneidman 1976). In the early years of the new millennium, this appeal is no less compelling. As noted above, valid death certification is essential to society, and the present call for refinement of the certification of death is equal to or even more challenging than in the past. In addition to a need to foster the creation of accurate public health information, there are scientific, legal, and social issues to consider. The public health and legal factors have already been addressed. In this section, the focus will be on social issues relating to the certification of death.

The accurate recording of data important to the research conducted by social and behavioral scientists has long been challenged. In a study conducted by Schmid and Van Arsdol (1955) the authors, noting the limited nature of official data on occupation, could not justify making any significant comparisons or conclusions. Similarly, Powell (1958) reported occupational data recorded on the death certificate were sparse and often inaccurate. Official data deficiencies of a similar nature are reported throughout the United States. For example, Warren Breed (1963) noted numerous errors pertaining to almost all categories for which sociodemographic characteristics are reported. Breed found that information pertaining to occupation, employment status, and marital status often is incorrectly documented. And in a more recent review of the suicide literature, Bedeian (1982) wrote,

Evidence suggests that a substantial number of deaths due to suicide are not recorded on death certificates because of religious, social, or moral stigma. Relatives or colleagues may attempt to disguise the true cause of a decedent’s death. Officials have been known to protect an influential person by classifying the cause of death as natural or accidental rather than as suicide. (Pp. 219-20)

Although death is an individual event that has social consequences, the implications of the recorded information can, in the words of Shneidman (1976), “enhance or degrade the reputation of the decedent and set its stamp on his post-self career” (p. 241). Social information, or information an individual conveys about himself or herself, creates a public persona. The erosion of one’s public image and replacing it with a degraded postself diminishes the social worth of the decedent while adding to the negative social consequences surrounding death experienced by relatives and friends. Suicide and the AIDS pandemic serve as examples of public health problems that hold this kind of social implication.

For religious, legal, and financial reasons, relatives may challenge a cause of death classification. Catholic orthodoxy denies burial in consecrated ground to those who end their life at their own hand. For this reason alone, there exists a tendency to have the cause recorded as other than suicide. While gathering data for the second time in a medical examiners office located in a large Midwest city, I asked the newly appointed medical examiner why I was unable to identify any new cases of suicide of individuals of a particular religious background. This individual declared that Jewish people do not commit suicide. Such an orientation holds obvious implications for the recording of cause of death classification.

To protect the financial interests of the family, a similar misclassification effect masks the true cause of death. Life insurance policies contain a clause that voids the full policy benefit should an individual die by suicide within a 2-year period dating from the time a life insurance policy is issued. Accurate classification of the true cause of death thus holds significant legal and financial consequences for the family of the deceased. I am reminded of an exchange with a medical examiner located in yet another large city; this individual stated that he would not allow anything to be recorded on an investigative report on equivocal death that could be held against him in a court of law. The same issue may be said to intervene for classifying cause of death.

Suggested Revisions Still an Unfulfilled Proposal

It is well-known among behavioral and social scientists that official documents offer scant assistance in establishing unusual or extenuating circumstances or individual motives that lead to death. More than a generation ago, noted thanatologist and suicidalogist Edwin S. Shneidman (1976) joined a rather long list of critics of official statistics in proposing that the death certificate, a document that he states memorializes death, be changed. What Shneidman identified is the need to provide a more useful definition of the current dimensions involving death. The death certificate, as Shneidman points out, is more than an official document. In his words,

The impact of the death certificate is considerable. It holds a mirror to our mores; it reflects some of our deepest taboos; it can directly affect the fate and fortune of a family, touching both its affluence and its mental health; it can enhance or degrade the reputation of the decedent and set its stamp on his post-self career. But if the impact of the death certificate is great, its limitations are of equal magnitude. In its present form the death certificate is a badly flawed document. (P. 241)

Shneidman’s call is for creation of a document that provides the requisite information to establish the dimensions that would allow construction of a “psychological autopsy.” Because the intent of the decedent may have a direct affect on life, Shneidman argued on behalf of recording the intention, subintention, or the unintention, believing that such information can offer information for determining whether the individual had a part in hastening his or her demise. Whether Shneidman’s admonition to action has merit outside psychological circles, the primary issue raised is not forgotten. The need to advance scientific knowledge of disease and quality-of-life factors and the physiological effects on human bodies has led to a renewed emphasis toward enhancing the death certificate categories to provide more definitive information.

As can readily be observed from the National Vital Statistics Report released by the Centers for Disease Control, numerous categories of death are currently documented. Just as in the past, trends are of obvious import. But the current emphasis is for a more informed professional cadre of certifiers of death to support the need to accurately certify the cause(s) of death. In this area then, Shneidman’s (1976) interest in creating psychological autopsies with the addition of categories in which social perspectives of death will be recognized represents a mirror image of the current charge to more fully understand the physiological dimensions of death on an official document. These dimensions recognize the enhanced state of knowledge contributed by the medical research professions. Thus Shneidman’s contention of the past that revision of this document is needed may hold merit for health care contemporaries. Certainly, the decedent’s documented history is brief, and it may also be argued that this history is too brief, given the social, economic, and legal significance of this legal document.

More recent acclaim for enhanced documentation on the death certificate emanates from public health officials and medical researchers whose need for specificity of the death-related causes is prompted by a different but no less important need to know.

Despite this need, not all certifiers adhere to the premise that enhanced standardization is essential to the well being of a population. Arthur G. Bedeian (1982), for example, claims that to protect the deceased’s reputation, “many suicides are reported as respiratory failure—which they are, but only by interpretation” (p. 220).

The death certificate and the implications corresponding to the social stigma related to the AIDS epidemic are not immune from the certification of cause of death process. It is well known that the AIDS virus is responsible for the onset of other maladies that directly bear on the infected person’s quality of life and ultimate death. These include immediate causes such as myelogenous leukemia and intervening causes of death, including infection and diseases such as pneumocystis carinii pneumonia, mycobacterium tuberculosis, and lymphomas (Halbach 2002). AIDS would, according to the ICD-10, be recorded as the underlying cause on the death certificate. But the social stigma attributed to those infected with the AIDS virus intervenes into the recording process to the extent that causes such as leukemia rather than AIDS often are reported as the underlying cause of death.

The need to educate medical practitioners to accurately report the appropriate cause of death is predicated on meeting requirements established by the Centers for Disease Control, and a state’s ability to receive federal funding depends on complying with these requirements. The national trend during the past several years is to disallow use of the “natural causes” classification. As observed by one public health official, allowances of this category as a primary cause of death holds little meaning for the vital statistics collected for the nation. Simply put, it does not offer enough information pertaining to cause of death (Franke 2000). Given that the cause of death is the most important item of public health information to be gleaned from the death certificate, specific data are useful to epidemiologists and are used by state legislatures to determine an appropriate level of funding for state public health agencies and to create new public policy. Such policies, intended to prevent disease, could be misdirected without an accurate base of information.

Perhaps the effort to draw attention to the importance of accurate recording is having or will have the intended effect. Victoria Stagg Elliott (2002), staff writer for the American Medical News, reports a resolution passed at the 2002 annual meeting of the American Medical Association calling for “affirming that cause of death is an integral part of patient care and that vital records should be a part of medical education at all levels” (p. 18).

One thing is certain: Appropriate mortality statistics holds crucial importance for the future quality of life and health of a nation.


Although death certification has undergone extensive revision in the past, this process continues. The U.S. standard death certificate has long been used as a guide for states to follow. The latest version of the U.S. Standard Certificate of Death, implemented January 1, 2003, provides an opportunity to create additional information, such as date of injury and description of how death-induced injuries occur, information pertaining to traffic fatalities, the effect of tobacco products in the cause of death, and whether female decedents were pregnant (Davis and Onaka 2001). These changes are certain to hold important implications in the future.

It is also certain that revision of the ICD will have a marked affect on certifying the cause of death. As noted by Shneidman (1976), “When the Seventh Revision [proposed in 1955] was put into effect for the data year 1958, the death rate for suicides [and accidents] increased markedly over 1957” (p. 242). Similar effects on both the definitions and incidents of suicides and accidents, according to Shneidman, took place with the introduction of the ICD-8 (and the later ICD-9) that led to international system changes for death classification. The same observation is made for the data from 1999. According to John L. McIntosh (2002), suicide, ranked among the top 10 causes of death since the 1950s, is no longer so highly ranked for the general population. In 1999, the 10th revision (ICD-10) became effective in establishing a new classification of death. Prior to 1999 and throughout the decade of the 1990s, the 9th revision (ICD-9) was employed with a different method and rules for grouping diseases and classifying the underlying cause of death. Such system changes affect the numbers, if only in a modest way, but the comparison of disease and other causes of death over time also are affected.

Numerous analysts recognize that death certificates are the foundation of cause of death and guide decision making for the allocation of medical resources (e.g., Lu et al. 2001) as well as for various social policies related to what is perceived to be the public good. Given the widespread interest in competing causes of death, perhaps the politics of death certification are inevitable. With increasing competition for available health-related research dollars and social service activities, researchers and other interested analysts have a stake in the accuracy of the certifiers’ recorded underlying cause of death. Recoding refinement will undoubtedly lead to changes in the causal ranking of the most important causes of death, and this kind of information serves well the needs of those in charge of research funding priorities. The amount of funding and for which set of diseases funding is made available refer to questions that can be answered only by the vital epidemiological database currently being accumulated. Such issues may also provide a partial explanation as to why so much disease-related literature is available on the death certificate and vital statistics topics.

Finally, the social dysfunctions of the recording procedure must be taken into consideration. Inadequate or inaccurate information may indeed protect the vested interests of families, but such self-serving information can only lead to conflicting conclusions as to the validity of knowledge, which, in this context, is the validity of death statistics. Such statistics are essential for assessing the social, psychological, and public health of our communities.