Statistical methodologies in psychopharmacology: a review

There has been a greatly increased interest in statistical methods in the psychiatric research and its applications over the past few decades, in parallel with advances in computers and statistical software. This review aims to describe the main topics related to statistical methods in psychopharmacology, namely nature of statistics in medicine, problems in data analysis, statistical modelling, developments in statistical technology, statistical reporting and meta-analysis.     

Attention and behavioural problems of Finnish adolescents may be related to the family environment

BACKGROUND: Adolescents' attention and behavioural problems in relation to the family environment were studied in the Northern Finland 1986 Birth Cohort. METHOD: Fifteen-year-old adolescents (N=6888) completed the Youth Self-Report (YSR) questionnaire and their parents (N=6643) completed the Strengths and Weaknesses in ADHD and Normal Behaviours (SWAN) questionnaire. The ratings were compared in relation to gender and family characteristics. RESULTS: Girls scored higher than boys on the YSR attention problems subscale (means 4.61 vs. 3.41), the rule-breaking behaviour subscale (means 4.25 vs. 3.69) and the aggressive behaviour subscale (means 7.18 vs. 5.63). Parents reported more SWAN attention problems in their sons than in their daughters. Living in an other than intact family increased YSR and SWAN attention problems and YSR behavioural problems in both genders. Adolescents living in very large families (11-19 children) had least attention and behavioural problems. CONCLUSIONS: Attention and behavioural problems seem to be common among adolescents in Finland. For both genders, living in other than intact families increases especially behavioural problems. Additionally, a very large family seems to be a protective factor against those problems.     

Recovery from schizophrenic psychoses within the northern Finland 1966 Birth Cohort

BACKGROUND: Because of widely disparate findings from follow-up studies, the likelihood of recovery from schizophrenia remains controversial. We report the extent of recovery from schizophrenia in a population-based cohort. METHOD: Subjects with psychotic disorders were recruited from the Northern Finland 1966 Birth Cohort. Of the 91 subjects who agreed to participate, 59 were diagnosed with schizophrenia and 12 were diagnosed with schizophrenia spectrum disorders (schizophreniform psychosis, schizoaffective or delusional disorder) by DSM-III-R criteria. Diagnoses were established by interviewing the subjects, checking the Finnish Hospital Discharge Register, and reviewing their medical records. To assess recovery, we used the Clinical Global Impressions; the Positive and Negative Syndrome Scale; the Social and Occupational Functioning Assessment Scale; and information about psychiatric hospitalizations, use of antipsychotic medication, and occupational status. RESULTS: Only 1 subject (1.7%) with DSM-III-R schizophrenia and 3 subjects (25%) with schizophrenia spectrum disorders fully recovered; 1 schizophrenia subject (1.7%) and 2 schizophrenia spectrum subjects (16.7%) experienced partial recovery. CONCLUSION: The data indicate that, at least until age 35, complete recovery from schizophrenia is rare, and the prognosis for the disorder is far more serious than suggested by some follow-up studies.     

Psychiatric admissions at different levels of the national health care services and male criminality: the Northern Finland 1966 Birth Cohort study

  Background: It has earlier been suggested that one-third of violent offenders are recorded in psychiatric case registers. Our aim was to study whether violent and non-violent offenders differ with respect to admissions to any health care inpatient service due to psychiatric disorders. Methods: We used a genetically homogeneous, general population database from the Northern Finland 1966 Birth Cohort, together with the Finnish Hospital Discharge Register and national crime registers. Results: We found that one-third of violent and one-fourth of non-violent male offenders had at least one hospital admission due to a psychiatric disorder before the age of 32. Sixty-five violent criminals – 1.2% of all cohort males (n=5636) – accounted for 14.4% of all psychiatric treatment days. The admission rates among both violent and non-violent male offenders were significantly higher when compared with males with no criminal history. Among violent males, only half (55.5%) of the inpatient hospital days due to psychiatric disorders occurred in psychiatric hospitals. The corresponding percentages for non-violent criminals and non-criminals were 64.9% and 74.1%, respectively. Among the violent offenders, one-third of hospital inpatient days occurred in university hospitals or central hospitals, and only 1.9% of them occurred in a comprehensive community care system. Conclusions: Violent offenders' admission rates due to a psychiatric diagnosis are high, and they are frequently treated at an inappropriate health care level. Accepted: 23 February 2000     

Differential diagnosis in scrotal ultrasound. A review of 128 consecutive cases

One hundred and twenty-eight male patients were examined by ultrasound during a period of two years. The hospital records of the patients were reviewed one to three years after the first examination. The role of the ultrasound diagnosis in the differential diagnosis of suspected scrotal diseases was re-evaluated from a radiologist's point of view. Ultrasound is useful in the differentiation of intra- and extratesticular pathology, especially when palpation is difficult. All the extratesticular lesions were benign. The pathological intratesticular findings are described and pitfalls in the differential diagnosis discussed. It seems possible that the patient history greatly affects the ultrasound diagnosis when there is an intratesticular finding. Ultrasound should be used to provide reference information for the necessary follow-up studies of the effects of treatment.     

Infant motor development is associated with adult cognitive categorisation in a longitudinal birth cohort study

BACKGROUND: The relationship between the age of reaching infant developmental milestones and later intellectual function within the normal population remains unresolved. We hypothesised that the age of learning to stand in infancy would be associated with adult executive function and that the association would be apparent throughout the range of abilities, rather than confined to extremes. METHODS: The Northern Finland 1966 Birth Cohort is based upon 12,058 live-born children in a geographic and temporally defined population. Information on age at learning to stand without support was obtained at one year. At age 33-35 a random sample of 104 subjects underwent a neuropsychological test battery including tests of executive function (cognitive categorisation), visuo-spatial memory, verbal learning and visual object learning. We investigated associations between developmental data and adult neuropsychological test scores. RESULTS: There was a significant linear relationship between age of learning to stand and adult categorisation: the earlier the attainment of the milestone, the better was the categorisation. No such relationships were observed between infant neurodevelopment and adult cognition in other neuropsychological domains. CONCLUSION: Even within the normal range of development, early development in the gross motor domain is associated with better adult executive function (in tests of categorisation). Investigation of the determinants and sequelae of normal neural development will facilitate research into a variety of neurodevelopmental disorders.     

Psychometric properties of the Finnish 20-item Toronto Alexithymia Scale

The aim of this study was to examine the factor structure and the validity of the Finnish version of the 20-item Toronto Alexithymia Scale (TAS-20). As part of the Northern Finland 1966 Birth Cohort Project, the TAS-20 was presented to a sample of 5034 31-year old persons. A confirmatory factor analysis showed that the three-factor model, earlier established with the original TAS-20, was in agreement with the Finnish version of the scale. Three criteria of goodness-of-fit met the standards for adequacy of fit. For the total scale, internal reliability (Cronbach's alpha) was 0.83 and for the three subscales (factors 1, 2, and 3) it was 0.81, 0.77, and 0.66, respectively. Two- and one-factor models for TAS-20 were also examined, but the other models did not perform as well as the three-factor model. The factor model also worked well with a sample of 516 students with a mean age of 24.8 years. In conclusion, the TAS-20 scale is useful in the Finnish version, too.     

Use of antipsychotic medication and suicidality—the Northern Finland Birth Cohort 1966

In addition to psychoses, antipsychotic drugs are nowadays also prescribed for other psychiatric disturbances, such as mood disorders. We wanted to find out whether there is any association between the use of antipsychotic drugs and suicidality in cases of psychotic and non‐psychotic disorders. Our sample was the population‐based Northern Finland 1966 Birth Cohort. Information on the use of prescribed drugs was collected in 1997 from the nationwide medication register and with a postal questionnaire (N = 8218). The presence of suicidal ideation was assessed cross‐sectionally using the Symptom Check List‐25 questionnaire. We studied associations between suicidal ideation, adjusted for symptoms of depression and anxiety, and antipsychotic medication in different diagnostic groups (schizophrenia, other psychosis and no psychosis). Individuals receiving antipsychotic medication (n = 70, 0.9%) had in general more suicidal ideation regardless of diagnostic group, although the associations diminished when taking other symptoms into account. There were no statistically significant differences between those taking typical and atypical antipsychotics. In the non‐psychotic group, higher antipsychotic doses were associated with more suicidal ideation even when adjusted for symptoms of depression and anxiety (p < 0.05). In the cases of schizophrenia or other forms of psychosis, no such associations were observed. Our results suggest that one should take suicidal ideation into account when prescribing antipsychotic medication, especially for off‐label use. Copyright © 2012 John Wiley & Sons, Ltd.