Psychometric evaluation of the Spanish version of CONNECT: a measure of continuity of care in mental health services.

This article provides the results of the psychometric testing of the Spanish version of CONNECT(-S), a measure of continuity of care in mental health services. CONNECT-S is a multidimensional measure designed for use with seriously mentally ill respondents. Consisting of 12 scales and one single-item indicator, it addresses qualities of interaction in current relationships between mental health service providers and consumers in five conceptual domains: (1) practitioner knowledge of their clients, (2) creating flexibility, (3) practitioner availability, (4) practitioner co-ordination, and (5) smoothing transitions. One-hundred-and-fifty participants took part in the study. Participants were recruited from mental health outpatient clinics in both the Puerto Rican (n = 109) and the San Antonio (n = 41) samples. Internal consistency for scales in a combined site estimate ranged from 0.68 to 0.96. Test-retest reliability ranged from fair to substantial in all but one scale. Concurrent validity hypotheses based on a priori predictions were mostly supported. The Spanish translation and adaptation of CONNECT-S provided sound psychometric results across both sites. CONNECT-S addresses the gap in measurement of continuity of care for the two largest US Latino subgroups, Mexican Americans and Puerto Ricans; and provides an encouraging starting point for a measure that is both relevant and culturally sensitive.

Test-retest reliability of the Spanish version of the Diagnostic Interview Schedule for Children (DISC-IV).

The test-retest reliability of the Spanish Diagnostic Interview Schedule for Children (DISC-IV) is presented. This version was developed in Puerto Rico in consultation with an international bilingual committee, sponsored by NIMH. The sample (N = 146) consisted of children recruited from outpatient mental health clinics and a drug residential treatment facility. Two different pairs of nonclinicians administered the DISC twice to the parent and child respondents. Results indicated fair to moderate agreement for parent reports on most diagnoses. Relatively similar agreement levels were observed for last month and last year time frames. Surprisingly, the inclusion of impairment as a criterion for diagnosis did not substantially change the pattern of results for specific disorders. Parents were more reliable when reporting on diagnoses of younger (4-10) than older children. Children 11-17 years old were reliable informants on disruptive and substance abuse/dependence disorders, but unreliable for anxiety and depressive disorders. Hence, parents were more reliable when reporting about anxiety and depressive disorders whereas children were more reliable than their parents when reporting about disruptive and substance disorders.

Gender differences in the relationship of homelessness to symptom severity, substance abuse, and neuroleptic noncompliance in schizophrenia.

This study examined gender differences in the relationship of homelessness in schizophrenia to symptom severity, risk behaviors, and prognostic features. Four hundred subjects with schizophrenia were studied: 100 homeless men, 100 homeless women, 100 never homeless men, and 100 never homeless women. Assessments included derivation of five symptom factors by using the Positive and Negative Syndrome Scale (PANSS). Homelessness for the entire sample was associated with greater severity of positive, activation, and autistic preoccupation symptoms, younger age at first hospitalization, and substance abuse (SA). For men only, homelessness was associated with neuroleptic noncompliance (NN). When NN and SA were statistically controlled, symptom severity was not different between the homeless and never homeless. Women, independent of residential status, had more severe negative, activation, and autistic preoccupation symptoms that were not associated with prognostic features or risk behaviors. For both men and women, SA was associated with homelessness, but independent of residence, SA was less severe in women. Additionally, SA was less severe in homeless women than never homeless men. Thus, symptom severity in homeless individuals with schizophrenia appears as an interaction of symptom profiles and risk behaviors that are gender specific. Although cross-sectional analyses cannot distinguish cause from effect, these findings suggest gender-specific routes to homelessness among indigent urban adults with schizophrenia.

Risk factors for homelessness among indigent urban adults with no history of psychotic illness: a case-control study.

OBJECTIVES: This study identified risk factors for homelessness among indigent urban adults without dependent children and with no history of psychotic illness. METHODS: We conducted a matched case-control study, stratified by sex, of 200 newly homeless men and women and 200 indigent men and women with no history of homelessness. Newly homeless case subjects were recruited from shelter assessment centers in New York City. Never-homeless control subjects, selected from public assistance centers, were single adults applying for home relief. Control subjects were matched with case subjects according to ethnicity, age, and sex. Trained interviewers employed standardized research instruments to probe 3 domains of risk factors: symptom severity and substance use disorder, family support and functioning, and prior use of services. RESULTS: Significant interaction effects by sex were present for symptom severity, heroin use disorder, and prior service use. Greater numbers of the homeless of both sexes lacked a high school diploma and had less income from all sources, including from their families, than of the never homeless. CONCLUSIONS: Newly homeless men and women with no history of psychotic illness differed from their never-homeless counterparts in the 3 domains investigated, but socioeconomic factors were also important.

Measurement reliability and agreement in psychiatry.

Psychiatric research has benefited from attention to measurement theories of reliability, and reliability/agreement statistics for psychopathology ratings and diagnoses are regularly reported in empirical reports. Nevertheless, there are still controversies regarding how reliability should be measured, and the amount of resources that should be spent on studying measurement quality in research programs. These issues are discussed in the context of recent theoretical and technical contributions to the statistical analysis of reliability. Special attention is paid to statistical studies published since Kraemer's 1992 review of reliability methods in this journal.

Causal modeling of epidemiological data on psychiatric disorders.

This paper reviews the logic of causal inference from epidemiological data. I maintain that the clearest causal statements can be made when the philosophical causal principles of association, direction and isolation are upheld in epidemiological research. After reviewing the argument by Holland that only experimental manipulation affords clear causal claims, I examine the utility of structural equation models and longitudinal methods for making causal claims from non-experimental data. This examination leads to the conclusion that mental health epidemiologists should begin to incorporate intervention trials into the last phases of their research programmes when they want to make strong causal claims.

The effect of lead exposure on behavior problems in preschool children.

OBJECTIVES: Interpreting associations between lead exposure and child behavior problems is difficult because studies have not controlled for sociodemographic confounders or have used shed teeth to mark exposure. This study explored associations between blood lead and preschool behavior. METHODS: Children from a smelter town and a non-lead-exposed town in Yugoslavia were followed up prospectively from pregnancy through age 3. The Child Behavior Checklist was used to assess behavior problems in 379 3-year-olds, controlling for sociodemographic factors and difficult infant temperament. RESULTS: Multiple regression revealed the expected significant associations between checklist subscales and sociodemographic factors, which explained 7% to 18% of the variance on the subscales. Concurrent blood lead explained a significant 1% to 4% of the variance on the Destructive and Withdrawn subscales. Earlier difficult temperament explained an additional 2% to 5% of the checklist variance. Scores on the Destructive subscale were consistently associated with blood lead. As blood lead increased from 10 to 20 micrograms/dL, subscale scores increased by approximately 0.5 points. CONCLUSIONS: Lead/behavior associations are significant but small compared with the effects of social factors.

Major depressive disorder in the 6 months after miscarriage.

OBJECTIVE: To test a priori hypotheses that miscarrying women are at increased risk for a first or recurrent episode of major depressive disorder in the 6 months following loss and that this increased risk is greater for childless women, women with prior reproductive loss, and women aged 35 years or older; and to evaluate whether risk varies by time of gestation or by attitude toward the pregnancy. DESIGN: Cohort study. SETTING: The miscarriage cohort consisted of women attending a medical center for a spontaneous abortion (n=229); the comparison group was a population-based cohort of women drawn from the community (n=230). PARTICIPANTS: Miscarriage was defined as the involuntary termination of a nonviable intrauterine pregnancy before 28 completed weeks of gestation. Half of all participants were between 25 and 34 years of age; 40% were white and 35% Hispanic; 55% had more than a high school education. Participants constituted 60% of miscarrying women and 72% of community women who completed the first phase of this cohort study. MAIN OUTCOME MEASURE: Major depressive disorder was measured using the Diagnostic Interview Schedule. RESULTS: Risk for an episode of major depressive disorder among miscarrying women in the 6 months following loss was compared with the 6-month risk among community women who had not been pregnant in the preceding year. Among miscarrying women, 10.9% experienced an episode of major depressive disorder, compared with 4.3% of community women. The overall relative risk (RR) for an episode of major depressive disorder for miscarrying women was 2.5 (95% confidence interval [CI], 1.2-5.1) and was substantially higher for childless women (RR, 5.0; 95% CI, 1.7-14.4) than for women with children (RR, 1.3; 95% CI, 0.5-3.5) (P<.06). Among miscarrying women, 72% of cases of major depressive disorder began within the first month after loss; only 20% of community cases started during the comparable period. Among miscarrying women with a history of major depressive disorder, 54% experienced a recurrence. However, RR did not vary significantly by history of prior reproductive loss or by maternal age, nor did risk vary by time of gestation or attitude toward the pregnancy. CONCLUSIONS: Physicians should monitor miscarrying women in the first weeks after reproductive loss, particularly women who are childless or who have a history of major depressive disorder. Where appropriate, supportive counseling or psychopharmacologic treatment should be considered.

Jews and their intraethnic vulnerability to affective disorders, fact or artifact? II: Evidence from a cohort study.

This paper, which complements a prior review of published studies, reports findings from a community-based survey of 4,914 Israel-born offspring of immigrants from Europe (Ashkenazim) and North Africa. Respondents were examined by psychiatrists using the Schedule for Affective Disorders and Schizophrenia, Israel version, and diagnosed with the Research Diagnostic Criteria. Unlike previous studies, this investigation found that Israelis of North African origin had significantly higher rates of affective disorders, including major depressive and intermittent depressive disorders. The Ashkenazim, however, had higher rates of bipolar I disorder at the definite level of diagnosis. Differential patterns in help-seeking may account for the divergent findings between this community-based study and earlier treatment-based reports. These results suggest the need to further investigate social and genetic etiological factors that may explain the differential rates.

PTSD symptoms and comorbid mental disorders in Israeli war veterans.

BACKGROUND: The purpose of this study was to investigate patterns of mental disorders co-morbid with PTSD symptoms in young Israeli men exposed to combat. METHOD: Six hundred and seventeen subjects were selected via a general population sample and evaluated in a two-phase case-identification procedure, culminating in a modified SADS-L interview, administered by psychiatrists. RESULTS: Major depressive disorder (OR = 3.2), substance use disorders (OR = 1.9) and personality disorders (OR = 3.0) occurred more frequently in men reporting symptoms of PTSD than in men who had been under fire who did not report symptoms. With the possible exception of personality disorders, comorbid disorders did not constitute risk factors for PTSD. Comorbid PTSD and RDC disorders were associated with increased help-seeking. CONCLUSIONS: The results suggest similar rates and types of PTSD comorbidity in Israeli war veterans as in veterans in the US assessed in general population studies, and are consistent with shared risk factors for PTSD and comorbid disorders.

Association of stressful life events with chromosomally normal spontaneous abortion.

Spontaneous abortion is the most common adverse reproductive outcome. Despite evidence that negative life events increase risk for a number of medical disorders, their role in pregnancy disruption has not been investigated. The present study tested an a priori hypothesis that recent negative life events increase the odds of spontaneous abortion of a chromosomally normal conceptus. Between 1984 and 1986, 192 women aged 18-42 years who visited a medical center after spontaneous abortion were interviewed about positive and negative events that had occurred in the 4-5 months preceding the loss. Subsequently, women with chromosomally normal (n = 111) and chromosomally abnormal (n = 81) losses were identified on the basis of tissue culture after interview. The women with chromosomally abnormal loss provided an estimate of the expected frequency of life events against which to compare the event frequencies of women with chromosomally normal loss. Analyses were adjusted for duration of the recall period, payment status, maternal age, education, and ethnicity. Seventy percent of the women with chromosomally normal losses reported having had one or more negative life events in the months preceding loss, compared with 52% of the women with chromosomally abnormal losses (adjusted odds ratio = 2.6, 95% confidence interval (CI) 1.3-5.2). For private patients (n = 69), the adjusted odds ratio was 4.2 (95% CI 1.3-13.4); for public patients (n = 123), it was 1.9 (95% CI 0.8-4.8). The associations held for postconception events alone and were absent for positive events. Results were unaltered by adjustment for smoking, caffeine intake, and alcohol consumption. With recall bias precluded by the study design, the strength, timing, and specificity of these associations suggest that recent negative life events play a role in chromosomally normal spontaneous abortion. Efforts to replicate these results and to elucidate underlying biologic mechanisms are required.

The Diagnostic Interview Schedule for Children (DISC-2.1) in Spanish: reliability in a Hispanic population.

The reliability across time, informants and interviewers of the Spanish translation of the DISC-2.1 was tested on a Puerto Rican Hispanic sample using a test-retest design. Levels of reliability between clinic and community samples and between younger and older children were compared to explore the sources of low reliability for certain psychiatric disorders. Parents' reports tended to be more reliable than those of their children, although the difference was less obvious with older children. Reliability was generally higher for the externalizing disorders and when the second interviewer was a psychiatrist rather than a lay interviewer.

Risk factors for homelessness among women with schizophrenia.

A study of risk factors for homelessness among the severely mentally ill was extended to include women, and a case-control study of 100 indigent women with schizophrenia meeting criteria for literal homelessness and 100 such women with no history of homelessness was conducted. Subjects were recruited from shelters, clinics, and inpatient psychiatric programs in New York City. Clinical interviewers used standardized research instruments to probe three domains of risk factors: severity of mental illness, family background, and prior mental health service use. Findings adjusted for ethnicity revealed that homeless women had higher rates of a concurrent diagnosis of alcohol abuse, drug abuse, and antisocial personality disorder. Homeless women also had less adequate family support.

Evaluating a ten questions screen for childhood disability: reliability and internal structure in different cultures.

This paper uses five strategies to evaluate the reliability and other measurement qualities of the Ten Questions screen for childhood disability. The screen was administered for 22,125 children, aged 2-9 years, in Bangladesh, Jamaica and Pakistan. The test-retest approach involving small sub-samples was useful for assessing reliability of overall screening results, but not of individual items with low prevalence. Alternative strategies focus on the internal consistency and structure of the screen as well as item analyses. They provide evidence of similar and comparable qualities of measurement in the three culturally divergent populations, indicating that the screen is likely to produce comparable data across cultures. One of the questions, however, correlates with the other questions differently in Jamaica, where it appears to "over-identify" children as seriously disabled. The methods and findings reported here have general applications for the design and evaluation of questionnaires for epidemiologic research, particularly when the goal is to gather comparable data in geographically and culturally diverse settings.

Psychometric properties of parents and children as informants in child psychiatry epidemiology with the Spanish Diagnostic Interview Schedule for Children (DISC.2).

Parent and child reports were examined to study how epidemiological researchers can best use the information provided to describe childhood psychopathology. As part of a multisite methodologic study of mental disorders in children, a probability sample (N = 248) of children aged 9 to 17 years from the San Juan metropolitan area was selected. This sample was enriched with 74 clinic cases. Both parents and children were administered the DISC.2. Results showed that prevalence estimates were influenced by the informant. The clinicians' diagnosis is more concordant with children's reports of depression and with parents' reports of disruptive disorders. Parents and children provided unique information when interviewed with a structured psychiatric interview about child psychopathology. Their unique perspectives contributed to the observed discordance that emerged when DISC parent and DISC child results are compared. Combining the two perspectives with a simple "OR" rule at the symptom level did not seem to capture the unique perspectives.

Correlates of codisorders in homeless and never homeless indigent schizophrenic men.

We recently completed a case-control study of 100 literally homeless and 100 never homeless indigent schizophrenic men in New York City, in which concurrent substance abuse and antisocial personality disorder were widespread. In this paper we probe the correlates of 'pure' schizophrenia (single disorder, N = 60), schizophrenia and substance abuse (double disorder, N = 89), and schizophrenia, substance abuse, and antisocial personality disorder (triple disorder, N = 50), across the homeless/never homeless distinction. Subjects were recruited from a homeless shelter and mental health service programmes in Upper Manhattan. Psychologist and social worker interviewers administered the Structured Clinical Interview for DSM-III-R, I, and II and other structured instruments to explore social, family and illness history, the current illness, and aspects of treatment and family support. Codisorder subjects emerged from more disadvantaged family backgrounds, experienced greater school difficulties, began drug use in early adolescence, were more prone to hyperactivity in childhood, and were more likely to have spent time in jail. While codisorder groups did not differ on key aspects of schizophrenia, the triple disorder group was found to suffer from a more severe form of substance abuse than double disorder subjects, associated with an earlier age of onset and abuse of a wider array of substances. The widespread prevalence of codisorders among indigent schizophrenic men has major significance for clinical psychiatry. Study of the correlates of codisorders has revealed important differences in social, family, and illness history which may guide the development of more effective treatments and improved service delivery.

Minor depression in a cohort of young adults in Israel.

BACKGROUND: A diagnosis of minor depression was considered for DSM-IV. Mild depression is thought to be common in primary care settings and the community, but studies of the validity of minor depression as a separate diagnostic category are few. METHODS: Minor depression as defined by Research Diagnostic Criteria was assessed by psychiatrists using a modified Schedule for Affective Disorders and Schizophrenia-Lifetime version in a cohort of 5200 young adults in Israel. Subjects with year-prevalent minor depression were compared with subjects with major depression or generalized anxiety disorder and with controls on aspects of psychopathologic condition, psychosocial functioning, help-seeking behaviors, and demographic correlates. RESULTS: Symptomatically, minor depression appeared to be a mild version of major depression. Minor depression was associated with good teenage and general social functioning, but also with absence from work, separation or divorce, recent impairment in overall functioning, and help-seeking. CONCLUSIONS: The results lend support for including minor depression or expanding severity modifiers in future classifications to better capture the phenomenon of subthreshold depression.

Validity of the ten questions screened for childhood disability: results from population-based studies in Bangladesh, Jamaica, and Pakistan.

An international study to validate the Ten Questions screen for serious childhood disability was undertaken in communities in Bangladesh, Jamaica, and Pakistan, where community workers screened more than 22,000 children ages 2-9 years. All children who screened positive, as well as random samples of those who screened negative, were referred for clinical evaluations. Applying comparable diagnostic criteria, the sensitivity of the screen for serious cognitive, motor, and seizure disabilities is acceptable (80-100%) in all three populations, whereas the positive predictive values range from 3 to 15%. These results confirm the usefulness of the Ten Questions as a low-cost and rapid screen for these disabilities, although not for vision and hearing disabilities, in populations where few affected children have previously been identified and treated. They also show that the value of the Ten Questions for identifying disability in underserved populations is limited to that of a screen; more thorough evaluations of children screened positive are necessary to distinguish true- from false-positive results and to identify the nature of the disability if present.