A Short History of Diagnostic Instruments

When researchers in psychiatric epidemiology embarked on the development of diagnostic measures of psychological
problems for use in community studies, they were faced with a problem. Diagnoses were not standardized. Clinicians
making diagnoses of major depression, for example, used different criteria and made different decisions (Wakefield
1999). There was nothing obvious in what constituted the boundaries between one disorder and another, or between
psychiatric illness and wellness. In reality, depression shades into anxiety, which shades into thought problems; people
have more than one type of problem; and people have symptoms that range along a continuum of severity rather than
being entirely present or entirely absent (Mirowsky and Ross 1983b; WiIson 1993). It is not obvious from clinical work
with people who have psychological problems how to make a standardized diagnosis in which some people are cate-
gorized as depressed and others not depressed (Kutchins and Kirk 1986). The crisp "syndromes" described in diagnostic
manuals are not distinct and characteristic collections of symptoms and signs that immediately impress themselves on the
minds of clinical observers (Aronowitz 2001). They are official classifications laboriously negotiated over decades, and
promulgated by the American Psychiatric Association (Horwitz 2002). Before the push to institute official standards in
the late 1970s, there was little agreement on concepts, definitions, and diagnostic criteria (Mirowsky 1990; WiIson 1993)

The first attempts to develop standardized diagnostic instruments collected information on the symptoms of patients in
psychiatric treatment. Factor analyses of the data revealed problems that clustered empirically, but they did not
correspond to clinical diagnoses (Endicott and Spitzer 1972; Spitzer et al. 1967; Spitzer et a1. 1970). Rather than accept
the implications of their results, psychiatric epidemiologists abandoned factor analysis as a means of shaping or
validating diagnostic categories. They shifted to demonstrating "procedural validity" (Robins 1986; Robins et a1. 1979),
or consensus among professionals, measured as inter-rater reliability rather than internal consistency (alpha reliability).
When the National Institutes of Mental Health (NIMH) contracted to develop a diagnostic protocol for use in large,
multi-city surveys called the Ecological Catchment Area (ECA) studies, they considered four research diagnostic
protocols. Only one had been developed and validated using factor analysis. It was immediately eliminated from any
further consideration because "its scales were based on internal consistency rather than approximating traditional clinical
syndromes" (Robins 1986:415). When empirical results did not conform to psychiatric preconceptions, psychiatric
epidemiology abandoned the results.

Ultimately, the ECA studies used a diagnostic instrument called the Diagnostic Interview Schedule (DIS), which
evolved from the Schedule for Affective Disorders and Schizophrenia (SADS), which we will describe below.

We will describe the transition from a situation of little consensus about diagnoses in the sixties and seventies to the
fully reified diagnostic system of today. The detailed examples will show how this system was constructed.

Although diagnostic instruments vary somewhat, all are based on descriptions in the Diagnostic and Statistical Manllal
of Mental Disorders (DSM III-R and IV) of the American Psychiatric Association (1987, 1994). The general criteria for
diagnosis are the presence of symptoms, co-existence of symptoms, and prolonged duration of the symptoms. Each
criterion has a cut-off point below which the person does not qualify for a diagnosis of a particular disorder. There are
also exclusion criteria. For example, depression with a medical cause, such as infection, anemia, or life-threatening
disease, is excluded from the diagnostic category of depression, as is depression caused by a death in the family.
Earlier instruments also included impaired functioning as a criterion, but later ones do not. Earlier instruments
considered diagnostic categories to be mutually exclusive, so that alternate diagnoses had to be ruled out, but later ones
allow for multiple diagnoses, which they call comorbidity (see, for example, Kessler and Zhao 1999).

We describe an earlier instrument called the SADS/RDC and a later one called the DIS because comparison of the
two shows the evolution of diagnosis. The early instruments show the decision rules more explicitly; the later ones
obscure them. There are four steps used. in making a diagnosis using the SADS/RDC. The first is assessing the level of
symptoms, and duration of problems. Note that level, extent, and duration all refer to assess ments of degree or amount.
The second is splitting each assessed amount at some cutoff point, so that differences in degree are col lapsed into two
categories: amounts that meet the criterion and amounts that do not. The third is toting up so that all possible
combinations of met/unmet on the three criteria are represented in a single overarching split. The fourth is excluding
cases that also meet other criteria considered preeminent (such as recent bereavement).

The later instrument, the DIS, combines the first two steps. It skips the assessment of symptom levels. Instead, the
split is built into the question. All questions are asked in a yes/no format, such as, " Have you ever had a period of two
weeks or more when you had trouble falling asleep, staying asleep or with waking up too early? Yes or No."
Respondents do not report how frequently they had trouble falling asleep, how long the period lasted, or when it
happened. Thus, the information is not assessed and then ignored, as in earlier instruments; it is never assessed
(Mirowsky and Ross 1989a).

Diagnosis combines assessment with judgment. The earlier instruments make this clear. Questioning, observing, and
recording symptoms, functioning, and duration is assessment. Using the answers, observations, and records to assign a
case to a category is judgment. The two kinds of tasks can be divided in time or between actors (e.g., nurse and
physician). In the SADS/RDC they are divided into a questionnaire or protocol for assessment and. an "algorithm" or set
of rules for making a judgment. Later instruments combine the two, going right to the judgment.

Diagnosing Major Depression

To illustrate psychiatric diagnosis, we'll see how the two diagnostic instruments are used to diagnose major
depression. The Schedule for Affective Disorders and Schizophrenia (SADS) is used to make a diagnosis based on the
Research Diagnostic Criteria (RDC) (Endicott and Spitzer 1979; Spitzer and Endicott 1978). The SADS is a
questionnaire and protocol for assessing symptoms, functioning, and duration. The RDC is a set of criteria for deciding
on a diagnosis given the information in the SADS. Together these are often called the SADS/RDC. Later it was replaced
by the Diagnostic Interview Schedule (OIS) (Robins et a1. 1979; Robins 1986). Both instruments were developed for
research, and both have been used in community surveys using trained interviewers who are not psychiatrists (Weissman
and Myers 1978), including large community surveys done in various metropolitan areas throughout the United States,
called Epidemiologic Catchment Area (ECA) surveys (Baton et a1. 1986). Somewhat later, the National Comorbidity
Survey used a modification of the DIS called the Composite International Diagnostic Interview (CIDI) (Robins et a1.
1988; Kessler and Zhao 1999). The logic of diagnostic instruments remained the same; the major difference is that the
CIDI is an international instrument, so that it is now used throughout the world. (For brief descriptions of other similar
diagnostic instruments see Switzer, Dew, and Bromet 1999).
 The earlier instrument, the SADSIRDC, is more standardized than the Diagnostic and Statistic Manual (DSM), and it
is more explicit and detailed in describing how a diagnosis is made than the later research diagnostic instruments, the
Diagnostic Interview Schedules (DIS and its successor, CID!), but the SADSIRDC, the DSM, and the DIS have common
roots (Horwitz 2002). Spitzer and Endicott based the RDC on the Feighner criteria (Feighner et a1. 1972). As head of the
task force to develop the DSM-III, Spitzer based the DSM-III on the RDC to the extent that committee decision-making
would allow (Robins 1986). The DIS, developed for the ECA studies, also has its roots in the Feighner criteria and the
diagnostic instrument used to make diagnoses based on these criteria, called the Renard Diagnostic Interview.

The SADS interviewer first assesses dysphoric (or depressed) mood on a continuum by asking the respondent, "How
have you been feeling? Describe your mood. Have you felt depressed _sad, blue, moody, down, empty, hopeless, as if
you didn't care)? How often? Does it come and go? How long does it last? How bad is the feeling?" Based on the
person's response, the interviewer records the frequency and severity of depressed mood from not at all depressed to
constant, unrelenting, extremely painful feelings of depression. Similar scales record the frequency and inten sity of self-
reproach, feelings of inadequacy, suicidal tendencies, sleep problems, trouble concentrating, loss of interest or pleasure,
loss of energy, and other elements of depression (Spitzer and Endicott 1978).

After the questions about specific symptoms, the interviewer makes a global assessment of the subject's level of
functioning. The assessment ranges from "good functioning in all areas, many interests, socially effective," to "major
impairment in several areas such as work or family relations," on up to "needs constant supervision for several days to
prevent hurting self or others, or makes no attempt to maintain minimal personal hygiene."

Once the assessments are made and recorded, the Research Diagnostic Criteria are applied. The symptom criterion has
two conditions, which must both be met: (1) dysphoric mood characterized by feeling depressed, blue, sad, hopeless,
irritable, and (2) at least five of the following eight symptoms: poor appetite or increased appetite, sleep problems, loss
of energy, psychomotor retardation or agitation, loss of interest or pleasure in usual activities, self-reproach, diminished
ability to concentrate, and thoughts of suicide. Although the SADS assessment records the frequency and intensity of
each symptom, the criteria only refer to the presence or absence of symptoms. The duration criterion is that dysphoric
mood has lasted at least one week. The functioning criterion is that the person sought or was referred for help, took
medication for the problem, or had impaired functioning with family, at home, at school, at work, or socially. Again,
although the SADS assessment records the frequency and intensity of dysfunction, the criterion only refers to its
presence. If all three criteria are met, if the person is not bereaved, if the person is not suffering from a medical ill ness
that could cause the symptoms, and if there are no signs of schizophrenia, the person is diagnosed as having major
depression.

The example illustrates the fact that diagnosing current major depression was a twopart process of collecting
information and then ignoring most of it. Information is thrown out in the splitting, toting, and excluding process es.
Splitting dispenses with much of it. Differences in the frequency and intensity of each symptom are ignored by counting
only the presence or absence of each. Information on the number of symptoms is thrown out by ignoring less than 5, and
treating any number greater than or equal to 5 as alike. Information on the duration of symptoms is thrown out by
ignoring episodes of less than a week, and treating all episodes of more than a week as alike. Differences in the type of
dysfunction are ignored by lumping them together, and differences in the extent of dysfunction are ignored by counting
only the presence or absence of some sign of dysfunction.
 The toting and excluding processes throw out information by ignoring the distinction between an emotional state and
its causes arid consequences (Mirowsky and Ross 1989a). When symptoms, functioning, and duration are toted up into
one global judgment, the distinctions among the three dimensions are ignored, as well as the distinctions within each. For
example, the practice of excluding depression caused by loss of a loved one from the diagnosis of depression obscures
social patterns. Exclusion rules ignore the distinction between the emotional state and its cause or concomitant. The rules
typically presuppose that each distinct cause necessarily produces a distinct disorder, even if the disorders are not
otherwise distinguishable. Psychiatrists cannot tell the difference between "grief" and "major depression" without
knowing if a patient has lost a close friend or relative. As a practical matter, it is worth noting whether a depressed
person has recently lost a loved one, just as it is worth noting other possible causes of the depression. There is no reason
to assume that each cause produces a unique emotional state. One consequence of excluding depression caused by
medical illness or loss of a loved one is that social patterns are obscured. The elderly and poor have more illness and
higher death rates among their loved ones than do the young and well to do. The emotional distress caused by this illness
and loss is discounted in studies of diagnosed major depression. As an example, Boyd et a!. (1982) describe the case of
an 85year-old woman who lives alone, without friends or family nearby. She is afmid of being robbed and will not leave
her apartment. Her husband recently died. She is incapacitated by arthritis. She scores very high on any measure of
depressive symptoms. She does not get a diagnosis of depression. Diagnostic exclusion criteria that discount depression
due to death of a loved one, illness, injury, drugs or medications sometimes account for the fact that research diagnostic
instruments count fewer people as depressed than do clinicians (Eaton et a1. 2000), disproportionately discounting the
depression of the elderly and poor (Bruce 1999).

Diagnoses of depression arbitrary eliminate some causes of depression. No cause should bc eliminated a priori. The
research questions of interest are mostly about the causes of depression. Eliminating a few causes, such as death of a
loved one and illness, confounds cause and effect, and arbitrarily eliminates some causes, but not others. Why not
eliminate depression caused by one's child's sickness, or poverty, or divorce? If, through measurement exclusion rules,
psychiatrists could eliminate all social causes of depression, they would have, by default, a measure whose causes were
unseen and unmeasured biological or genetic causes. This appears to be a goal of some (Wakefield 1999). For the
purpose is scientific investigation, however, it is better not to eliminate any purported cause, but rather to measure the
hypothesized causes separately and correlate them with depression. No causes should be assumed or ruled out; all should
be empirically examined.

Diagnoses also arbitrarily require the coexistence of symptoms. For example, in order to qualify for a diagnosis of
depression mood must co-exist with malaise, which is the physiological component of depression. Malaise includes
trouble sleeping, appetite problems, trouble concentrating, and feeling tired or rundown. The relationship of mood and
malaise should be an empirical question. Possibly, different ethnic groups, age groups, or genders express depression
differently, with more or less of a mood or a malaise component. Correlations between mood and malaise should be
examined, not assumed.

Historically, research diagnostic instruments ignored information on the level of a person's problems, instead splitting
the information into crude yes or no distinction, and they confounded infonnation on symptoms, their caus 5, and their
consequences; but the later diagostic instruments make these problems worse :lan the. earlier ones. The later diagnostic
nstrument used in the Epidemiologic :atchment Area (ECA) surveys and :omorbidity surveys is the Diagnostic [nterview
Schedule (DIS). The symptoms of
depression are the same in the DIS as in the SADS (feelings of depression, loss of interest or pleasure [in things usually
enjoyed, including sex], appetite problems, feelings of worthlessness, sleep problems, trouble thinking or concentrating,
loss of energy or restlessness, .
and suicidal tendencies); but the fonnat for assessing these problems in the DIS violates almost all psychometric
principles of good measurement. The DIS questions used to diagnose major depression are shown in Table 1.
The first major difference is that the DIS combines the assessment of problems with the cutoff rules, whereas the
SADSIRDC keeps them separate. The SADS first counts symptoms and then splits them into enough for a diagnosis
versus not enough. The DIS goes right to the split. People are first asked, "In your lifetime, have you ever had two weeks
or more during which you felt sad, blue, depressed or when you lost all interest and pleasure in things that you usually
care about or enjoyed? Yes or no." The intensity and duration of the feelings are compressed into a simple yes or no. All
other symptoms of depression are also recorded in a yeslno fonnat (see Table 1). The DIS forces respondents to answer
"yes" or "no" to questions about feeJings that are not easily dichotomized. Likert scales more accurately assess people's
feelings of mood and malaise.
The second major difference is the focus on lifetime problems, with current problems simply part of the lifetime
assessment. (The SADS assesses current and lifetime problems with two distinct schedules). It is well established in
survey research that people can most accurately report current or recent states. People have much more difficulty
remembering things that happened years ago, and memory is heaviIy
influenced by current feelings. Yet it is not until close to the end of the section on major depression (question 26; see
Table 1) that the

person is finally asked, "Are you in one of these spells of feeling low or disinterested and having some of these other
problems now? (yes or no)."
Recall is a serious problem in measures that are about emotions, thoughts, or feelings, yet the DIS begins by asking to
report whether they have ever been depressed in their lifetime. People tend to report the ways that they felt in the last
couple of weeks, or maybe the last month. When asked if there was ever in your lifetime a time that they felt sad,
depressed, and so on, people remember the most recent time they felt that way, and that is what they report. People can
report more objective things about their whole life, like whether their parents got divorced, when they got married, how
old they were when they had their first child, what their first job was; but reports on subjec tive things like feeling sad or
depressed cannot be reliably reported for one's lifetime.
A third major difference between the DIS and the SADS is that the DIS does not empha size impaired functioning as a
criterion for receiving a diagnosis. In the SADS, an important indicator of impaired functioning is the act of seeking or
being referred for treatment. Psychiatric epidemiologists recognized the problem of using this as a criterion for diagno-
sis: If, by definition, a person is not depressed ifhe or she does not seek treatment then there is no need for more
psychiatric services.
A minor difference in duration criteria illustrates that any cutoff is arbitrary. Diagnostic instruments make arbitrary
duration and severity cutoffs: the duration criteria for symptoms is one week in the SADS, while it is two weeks in the
DIS. Why would someone who felt depressed for 10 days not be considered depressed? Why would someone who felt
depressed for two weeks be considered equally depressed as someone who felt depressed for two years?
Why build in complex decision rules; dichotomous responses; frequency cutoffs, and so on? These are the same
questions used in scales of depression, just concatenated in obscure ways. The measurement of the diagnosis of major
depression is sometimes described as a "detailed computer algorithm" (Turner and Lloyd 1999:393), which we show in
Table 1. Why not just ask people about their symptoms, and let them answer on a scale of frequency or severity? We
think the answer is that diagnostic measures reify psychological prob
 pose, accomplishment, affection, and acceptance. Let the sociology of mental health observe, think, and speak in such
terms. Let us create a human science.

REFERENCES

American Psychiatric Association. 1987. Diagnostic and Statistical Manual of Mental

Disorders IlI-R. Washington DC: American' Psychiatric Association.
American Psychiatric Association 1994. Diagnostic and Statistical Manual of Mental Disorders IV. Washington: American Psychiatric
Association.
Aronowitz, Robert A.. 2001. "When Do Symptoms Become a Disease?" Annals of Internal Medicine 134:803-08.
Boyd, Jeffrey H., J.D. Burke, E. Gruenberg, C.E. Holzer, D.S. Rae, L.K. George, M. Kamo, R. Stolzman, L. McEnvoy, and G. Nestadt.
1984. "Exclusion Criteria ofDSM-Ill: A Study of Cooccurrence of Hierarchy-free Syndromes." Archives of General Psychiatry
41:983-89.
Boyd, Jeffrey H., Myna M. Weissman, Douglas Thompson, and Jerome K. Myers. 1982. "Screening for Depression in a Community
Sample: Understanding the Discrepancies between Depression Symptom and Diagnostic Scales." Archives of General Psychiatry 39:
1195-200.
Bruce, Martha Livingston. 1999. "Mental Illness as Psychiatric Disorder." Pp. 37-55 in Handbook of the Sociology of Mental Health,
edited by Carol S. Aneshensel and Jo C. Phelan. New York: Plenum.
Cohen, Jacob. 1983. "The Cost of Dichotomization." Applied Psychological Measurement 7:249-53.
Eaton, William W. and Larry G. Kessler. 1985. Epidemiologic Field Methods in Psychiatry: The NIMH Epidemiologic Catchment Area
Program. New York: Academic.
Eaton, William \v., Karen Neufeld, Li-Shiun Chen, and Guojun Cai. 2000. "A Comparison of Self report and Clinical Diagnostic
Interviews for Depression." Archh'es of General Psychiatry 57:217-22.
Eaton, William W., Darrel A. Regier, Ben Z. Locke, and Carl A. Taube. 1986. "The NIMH Epidemiologic Catchment Area Program."
Pp. 209-19 in Community Surveys of Psychiatric Disorders, edited by Myrna M. Weissman, Jerome K. Myers, and Catherine E. Ross.
New Brunswick, NJ: Rutgers University Press.
Endicott, Jean, and Robert L. Spitzer. 1972. "What! Another rating scale? The Psychiatric Evaluation Form" The Journal of Nervous and
Mental Disease 154:88-104.
Endicott, Jean and Robert L. Spitzer 1979. "Use of

the Research Diagnostic Criteria and Schedule for Affective Disorders and Schizophrenia to Study Affective Disorders." American
Journal of Psychiatry 136:52-56.
Feighner, lP, Eli Robins, S. B. Guze, R. A. Woodruff, G. Winokur, and R. Munoz. 1972. "Diagnostic Criteria for Use in Psychiatric
Research." Archives of General Psychiatry 26: 57-63.
Holzer, Charles E. Ill, B. M. Shea, Jeffery W. Swanson, Philip I Leaf, Jerome K. Myers, Unda George, Myrna M. Weissman, and P.
Bednarski. 1986. "The Increased Risk for Specific Psychiatric Disorders Among Persons of Low Socioeconomic Status: Evidence
From the Epidemiologic Catchment Area Surveys." American Journal of Social Psychiatry 6(4):259-71.
Horwitz, Allan V. 2002. Creating Mental Illness. Chicago: University of Chicago Press.
Horwitz, Allan V. and Teresa L. Sheid. 1999. "Approaches to Mental Health and Illness: Conflicting Definitions and Emphases" Pp. 1-11
in A Handbook for the Study of Mental Health: Social Contexts, Theories, and Systems, edited by AlIan V. Horwitz and Teresa L.
Scheid. Cambridge, UK: Cambridge University Press.
Kessler, Ronald C. 2000. "Psychiatric
 Epidemiology: Selected Recent Advances and Future Directions." Bulletin of the World Health Organization 78:464-74.
Kessler, Ronald C. and Shanyang Zhao. 1999. "The Prevalence of Mental Illness" Pp. 58-78 in A Handbookfor the Study of Mental
Health: Social Contexts, Theories. and Systems, edited by AlIan V. Horwitz and Teresa L. Scheid. Cambridge, UK: Cambridge
University Press.
Kirk, Stuart A. and Herb Kutchins. 1992. The Selling of DSM: The Rhetoric of Science in Psychiatry. New York: Aldine de Gruyter.
Kutchins, Herb and Stuart A. Kirk. 1986. "The Reliability of DSM-Ill: A Critical Review." Social Work Research and Abstracts
Winter:3-12.
Kutchins, Herb and Stuart A. Kirk. 1988. "The Business of Diagnosis: DSM-Ill and Clinical Social Work." Social Work 33: 215-20.
Langner, Thomas S. 1962. "A Twenty-two Item Screening Score of Psychiatric Symptoms Indicating Impairment." Journal of Health
and Human Behavior 3:269-76.
Levenstein, Susan. 1998. "Stress and Peptic Ulcer: Life Beyond Heliobacter." British Medical Journal 316:538-41.
Levenstein, Susan. 2000. "The Very Model of a Modem Etiology: A Biopsychosocial View of Peptic Ulcer. " Psychsomatic Medicine
62: 176-85.
Lin, E. and S. V. Parikh. 1999. "Sociodemographic, Clinical and Attitudinal Characteristics of the Untreated Depressed in
Ontario." JOIlr11al of Ajftctive Disorders 53:153-62.

Mirowsky, John. 1990. "Subjective Boundaries and Combinations in Psychiatric Diagnosis." JOIlr11al of Mind and Behavior 11(3):407-
24.
Mirowsky, John. 1994. "The Advantages of Indexes over Diagnoses in Scientific Assessment." Pp. 261-90 in Stress and Mental Health:
Contemporary Issues and Fllture Prospects, edited by William R. Avison and lan H. GotIib. New York: Plenum.
Mirowsky, John and Catherine E. Ross. 1983a. "Paranoia and the Structure of Powerlessness." American Sociological Review 48:228-39
Mirowsky, John and Catherine E. Ross. 1983b. "The Multidimensionality of Psychopathology in a Community Sample." American
Journal of Commll1!ity Psychology 11:573-91.
Mirowsky, John, and Catherine E. Ross. 1989a. "Psychiatric Diagnosis as Reified Measurement." Journal of Health and Social Behavior
30(1):11-25.
Mirowsky, John and Catherine E. Ross. 1989b. Social Causes of Psychological Distress. New York: Aldine de Gruyter.
Mirowsky, John and Catherine E. Ross. 1995. "Sex Differences in Distress: Real or Artifact?" American Sociological Review 60:449-68.
Mirowsky, John and Catherine E. Ross. 1996. "Fundamental Analysis in Research on WellBeing: Distress and the Sense of Contro!."
17ze Gerontologist 36:584-94.
Radloff, Lenore. 1977. "The CES-D Scale: A Selfreport Depression Scale for Research in the General Population. Applied Psychological
Measurement 1:385-401.
Robins, Lee N. 1986. "The Development and Characteristics of the NIMH Diagnostic Interview Schedule." Ppp. 403-427 in Community
Surveys of Psychiatric Disorders, edited by Myrna M. Weissman, Jerome K. Myers, and Catherine E. Ross. New Brunswick, NJ:
Rutgers University Press.
Robins, Lee N., John E. Helzer, Jack L. Croughan, Janet B. W. Williams, and Robert L. Spitzer. 1979. The NIMH Diagnostic
Interview Schedule (DIS). Washington DC: National Institute of Mental Health.
Robins, Lee N., J. Wing, H.-U. Witchen, J.E. Helzer, T.E Babor, JD. Burke, A. Farmer, A Jablenski, R. Pickens,D.A. Regier, N.
Sartorius, and L. H. Towle. 1988. "The Composite International Diagnostic Interview. An Epidemiological Instrument Suitable for
Use in Conjunction With Different Diagnostic Systems and in Different Cultures." Archives of General Psychiatry 45: 1069-17.
Ross, Catherine E. and John Mirowsky. 1984.
"Components of Depressed Mood in Married
Men and Women: The Center for Epidemiologic
 Studies' Depression Scale." American Journal of Epidemiology 119(6):997-1004.
Ross, Catherine E. and Marieke Van Willigen. 1996. "Gender, Parenthood and Anger." Jot/mal of Marriage and the Family 58:572-84.
Spitzer, Robert and Jean Endicott. 1978. Schedule
JOr Ajftctive Disorders and Schizophrenia. New
York: Biometric Research Division, Evaluation
Section, New York Psychiatric Institute.
Spitzer, Robert L., Jean Endicott, J.L. Fleiss, and 1. Cohen. 1970. "The Psychiatric Status Schedule: A Technique for Evaluating
Psychopathology and Impairment of Role Functioning." Archi_'es of General Psychiatry 23:41-55.
Spitzer, Robert L., 1. L. Fleiss, Jean Endicott, and 1. Cohen. 1967. "Mental Status Schedule: Properties of Factor-analytically Derived
Scales." Arclzives of General Psychiatry 16:479-93.
Srole, Leo and Anita Kassen Fischer. 1980. "To the
Editor." Archives of General Psyclziatry 37:
1424-26.
Sullivan, P.P., Ronald C. Kessler, and K.S. Kendler. 1998. "Latent Class Analysis of Lifetime Depressive Symptoms in the National
Comorbidity Survey." American JOllmal of Psychiatry 155: 1398-1406.
Switzer, Galen E., Mary Amanda Dew, and Evelyn J. Bromet. 1999. "Issues in Mental Health Assessment." Pp. 81-104 in Handbook of
the Sociology of Mental Health, edited by Carol S. Aneshensel and Jo C. Phclan. New York: Plenum.
Turner, R. Jay and Donald A. Lloyd. 1999. "The Stress Process and the Social Distribution of Depression." JOllmal of Health and Social
Behavior 40:374-404.
Wakefield, Jerome C. 1999. "The Measurement of Mental Disorder." Pp. 29-57 in A Handbookfor
the Swdy of Mental lIealth: Social Contexts. Theories, and Systems, edited by Allan V. Horwitz and Teresa L. Scheid. Cambridge, UK:
Cambridge University Press.
Weissman, Myrna M. 1987. "Advances in Psychiatric Epidemiology: Rates and Risks for Major Depression." American Jot/mal of Public
Health 77:445-51.
Weissman, Myrna. and Jerome K. Myers. 1978. "Affective Disorders in a U.S. Urban Community: The Use of the Research Diagnostic
Criteria in an Epidemiological Survey." Archives of General Psyclziatry 35:1304-11.
Wheaton, Blair. 1985. "Personal Resources and Mental Health." Pp. 139-184 in Research in Community and Mental Health, edited by
James R. Greenley. Greenwich, CT: JA!.
Wilson, Mitchell. 1993. "DSM-III and the Transformation of American Psychiatry." American Journal of Psyclziatry 150:399-410.