Ronald F. Kokes

Do Psychiatric Patients Fit Their Diagnoses? Patterns of Symptomatology as Described with the Biplot

Systems of psychiatric diagnosis have been regularly criticized for their low reliability and their inability to fit accurately the kinds of patients coming for treatment. To explore the reasons for these problems, this study utilizes a new method, the biplot, for defining groups of similar patients and the relationships of these groups to key symptom clusters. Using this technique to analyze data from a representative sample of first admissions for psychiatric disorder, results showed: a) symptom clusters representing the classical diagnostic categories, mania, schizophrenia, neurotic depression, and psychotic depression, were readily identified; b) however, only a few patients were clustered near these traditional syndromes. These findings suggest that although syndromes do exist that fit traditional diagnostic categories, the vast majority of patients fall between these syndromes, having characteristics from several of them. For most patients, forcing the diagnostician to choose among the categories requires an arbitrary decision that may contribute to dissatisfaction in the diagnostician who recognizes how misleading the diagnosis can be.

Patterns of disorder in first admission psychiatric patients.

Are there patterns of characteristics in psychiatric patients, different from traditional diagnostic considerations, that could provide important information for understanding, treatment, and research? To pursue this question, this report describes an investigation of clinical and demographic characteristics in a representative sample of first admissions for functional psychiatric disorder. Analyzing the patterns of these characteristics showed that social class had a particularly key role relating to a larger number of characteristics than did symptom and functioning measures. Symptoms when analyzed together revealed replicated factors not corresponding to diagnostic types. The implications of these and other findings for considering a broad perspective in conceptualizing, studying, and treating psychiatric disorder are discussed.

Child competence and psychiatric risk. IV. Relationships of parent diagnostic classifications and parent psychopathology severity to child functioning.

This study examined relationships between aspects of parent psychopathology and the behavioral competence of 101 children, as measured by peers and teachers in their school classrooms, from families in which at least one parent has been previously hospitalized for a functional psychiatric disorder. Since these children are at risk for present and future behavioral disorder and significant differences have already been established between this risk sample and control classmates, this study represented a further attempt to identify risk factors for future psychopathology in these children. Diagnostic classifications and continuous measures of overall psychological impairment were used for the hospitalized parent at the time of the most recent psychiatric hospitalization and at the time of the study, an average of 4.36 years later. These indices of parent psychiatric status were related to Fishers measures of child school competence. Results indicated that the continuous measure of overall impairment at the time of the risk study was the aspect of parent pathology most highly related to child functioning, particularly in those areas reflecting cognitive functioning. DSM-III diagnosis at both time periods showed fewer and more equivocal significant relationships with child variables. These findings suggested that the usual, ongoing level of parent impairment is the most important factor of parent psychopathology relating to the functioning of the child. Diagnosis was less important, and neither the number of lifetime psychiatric hospitalizations nor recency of hospitalization was associated with child functioning.

Child Competence and psychiatric risk. VI. Summary and integration of findings.

A review of the major findings of the five previous papers in the series is presented and points of integration of the results are noted. Particular emphasis is placed on the role of diagnosis in predicting child competence as well as on differences related to whether or not the mother or the father is the patient. Also, the need to design studies to tease out the relative contributions of patient and family variables is pointed out.

Child competence and psychiatric risk. III. Comparisons based on diagnosis of hospitalized parent.

The school competence of children whose parents were previously hospitalized for psychiatric disorder was studied. These children, at risk themselves for behavioral disorder, were compared to classmates as well as to each other based upon the diagnosis of the hospitalized parent. Four diagnostic groups were employed: narrowly defined schizophrenic, broadly defined schizophrenic, affective psychotic, and hospitalized nonpsychotic patients. Significant differences between the risk sample and classmates are reported along with differences among the diagnostic groups. Sons of nonpsychotic hospitalized patients were functioning most poorly and were in marked defiance of peer and adult norms. Parental and child IQ were found to have had little impact upon the data and were not considered major interactive variables. Issues concerning systems of diagnoses and the developmental implications of the findings are presented.

How diagnostic concepts of schizophrenia differ.

How do various diagnostic definitions of schizophrenia compare? As part of a larger study of diagnosis and prognosis, nine different systems used for diagnosing schizophrenia were applied to a sample of patients (N=272) from two geographical catchment areas who had been hospitalized for the first time in their lives for functional psychiatric illness. The size and composition of the groups diagnosed schizophrenic by each set of criteria were then compared with respect to five clinical variables hypothesized to be important for differentiating the diagnostic systems. All data were collected with structured interviews of demonstrated reliability. Results showed that several of the variables tested distinguished schizophrenics diagnosed by some systems from those considered schizophrenic by other systems and from those patients in the sample not considered schizophrenic by any system. High levels of delusions of passivity characterized the patients classified as schizophrenic by three systems: Schneider, Langfeldt, and the Flexible System. High psychosis scores characterized the patients considered schizophrenic by the New Haven Schizophrenia Index. Increased chronicity and low levels of affective symptoms characterized the patients considered schizophrenic by the Feighner criteria. Poor work and social function characterized the patients considered schizophrenic by the hospital physician using DSM-II guidelines. These and other findings reported indicate that the various systems for diagnosing schizophrenia have both characteristic differences and overlap. Considering these can help to delineate the extent to which research findings or clinical experience based on the system can be generalized to findings based upon another.

Child competence and psychiatric risk. V. Sex of patient parent and dimensions of psychopathology.

This study examines a) the effect of sex of index patient parent on the significant and positive relationship between severity of parent psychopathology and Fishers child competence measures reported in a previous study of this series; and b) the relationship of general and specific dimensions of parent symptomatology to Fishers child competence measures. Severity of psychopathology and type of symptomatology data were collected on parents of 101 intact families in which at least one parent had been previously hospitalized for a functional psychiatric disorder. Ratings on each psychopathological variable were made for two time frames, most recent hospitalization and time of study, which was a mean duration of 4.36 years later. Behavioral competence data were collected on one son in each family on the basis of peer and teacher ratings of school behaviors. Results indicated that the highly significant relationship between rating of severity of parent psychopathology at time of study and child competence measures was mainly a function of the index patient mothers psychopathology. Only the affective dimensions of symptomatology were consistently and significantly correlated with child competence measures. Lability of affect for index patient mother at most recent hospitalization was positively related to child competence. Depression, withdrawal, and incongruous affect rated for index patient mother at time of study were related to lower child competence. Severity or type of psychopathology of index father was not consistently related to child competence measures.

IQ as a prognostic indicator in adult psychiatric first‐admissions

This study used IQ, along with measures of premorbid adjustment, health-sickness, symptom level, diagnostic severity and demographic data, to predict to 2-year outcome measures of level of functioning, health-sickness, and symptoms for a sample of 145 adult psychiatric first-admissions. It was hypothesized that IQ as an indicator of cognitive ability, or of general ability to adapt, would predict positively to improvement over the 2-year period. Data analysis was conducted with bivariate correlations and multiple regressions, using both absolute-level and residualized outcome variables. IQ showed modest, significant relationships with all absolute outcome indices and six of seven residualized measures, especially for a subsample of those with non-average IQ scores. Regressions showed that IQ provided independent prediction of symptom outcomes.

Predictors of outcome among adult psychiatric first‐admissions

This study comparatively evaluated the prognostic potential of 10 demographic and clinical factors previously associated with psychiatric outcome. The longitudinal design employed intensive, structured, reliable interview techniques. One hundred forty-five subjects from a representative sample (N = 217) of community mental health catchment area first lifetime admissions were assessed at hospitalization and at 2-year follow-up. Predictors examined included premorbid functioning, overall health-sickness, diagnostic severity, social class, sex, age, IQ, race, and life events. Outcome was assessed multidimensionally with absolute-level and residualized indices of functioning and symptomatology. Phillips Premorbid Status and social class emerged from correlation and multiple regression analyses as the best predictors of functioning level and overall clinical status, while IQ was the most prominent indicator of symptomatology changes. Results support the notion that a general social competence factor predicts to psychiatric outcome across the entire spectrum of severe disorders.

The role of family interaction in mediating the effect of parental pathology upon the school functioning of the child.

This report has explored three sets of relationships: the relation of parental pathology to the school functioning of the child; the relation of parental pathology to the pattern of family interaction; and the relation of family interaction to school functioning. The previous chapters have shown that the general mental health of the parent at the time of URCAFS and the diagnosis of the patient at the key hospitalization both relate to the school competence of the child. The patient-parent with a higher global assessment (GAS score) has children who are judged more competent in school. Also, the parents with affective disorders (both psychotic and not), including schizo-affective, have children with better school functioning than children of a parent with a nonaffective disorder. These are the relationships we will try to explain. Second, we have shown that the characteristics of the parent who has been hospitalized relate to his or her behavior in the free-play family interaction. Patient-parents with a high GAS score were more active in the free play. Psychotics were less active than the nonpsychotics. The nonaffective psychotics (schizophrenics) were not significantly different from the affective psychotics in rate and proportion but were significantly lower in warmth. Third, we have shown that the school functioning of the child can be predicted from the family interaction. If the childs relation with either parent is characterized as warm, active, and balanced, the child is likely to be judged competent by his teachers and his peers. Of these three, balance is the most important. In this chapter we wish to explore to what extent the relationship between parental pathology and childs school competence can be accounted