Camil Fuchs

Maximum Likelihood Estimation and Model Selection in Contingency Tables with Missing Data

Abstract In many studies the values of one or more variables are missing for subsets of the original sample. This article focuses on the problem of obtaining maximum likelihood estimates (MLE) for the parameters of log-linear models under this type of incomplete data. The appropriate systems of equations are presented and the expectation-maximization (EM) algorithm (Dempster, Laird, and Rubin 1977) is suggested as one of the possible methods for solving them. The algorithm has certain advantages but other alternatives may be computationally more effective. Tests of fit for log-linear models in the presence of incomplete data are considered. The data from the Protective Services Project for Older Persons (Blenkner, Bloom, and Nielsen 1971; Blenkner, Bloom, and Weber 1974) are used to illustrate the procedures discussed in the article.

Obsessive-compulsive disorder in hospitalized patients with chronic schizophrenia.

Obsessive-compulsive (OC) symptoms have been observed in a substantial proportion of schizophrenic patients. In this study, we assessed the rate of occurrence of OC symptoms and the interrelationship between OC and schizophrenic symptoms in 68 hospitalized chronic schizophrenic patients. The patients were interviewed with the Structured Clinical Interview for Axis-I DSM-IV Disorders - Patient Edition (SCID-P) and the appropriate rating scales including the Scale for the Assessment of Positive Symptoms, the Scale for the Assessment of Negative Symptoms, the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS), the Barnes Akathisia Scale, the Abnormal Involuntary Movement Scale, and the Social Behaviour Schedule (SBS). Sixteen patients (23.5%) met the DSM-IV criteria for OCD. A comparison of schizophrenic patients with and without OCD showed that the schizo-obsessive patients were significantly (1.7-fold) more impaired in basic social functioning, as reflected by the SBS score. No significant between-group differences for any of the other clinical variables were found. There was no significant correlation between OC and schizophrenic symptoms within the schizo-obsessive subgroup. The mean Y-BOCS score for the patients with both schizophrenia and OCD was within the typical range (22.8+/-1.7) observed in OCD without psychosis. The findings provide further evidence for the importance of the OC dimension in schizophrenia and may have important implications for the application of effective treatment approaches in this difficult-to-treat subgroup of schizophrenic patients.

Association between obsessive-compulsive disorder and polymorphisms of genes encoding components of the serotonergic and dopaminergic pathways

Obsessive-compulsive disorder (OCD) is a severe and disabling anxiety disorder with a marked genetic contribution. Pharmacological data indicated involvement of the serotonergic and dopaminergic systems. We studied the association between OCD and six candidate genes encoding important components of the serotonergic and dopaminergic pathways in 75 biologically unrelated patients and 172 ethnically matched controls (Ashkenazi and non-Ashkenazi Jews). Polymorphisms in the following genes were studied: tryptophan hydroxylase (TPH), serotonin 2A receptor (HTR2A), serotonin 2C receptor (HTR2C), serotonin transporter (5-HTT), dopamine receptor D4 (DRD4), and dopamine transporter (DAT1). The genotypic and allelic distribution of all polymorphisms tested did not show statistically significant differences between patients and controls. Our results suggest that these polymorphisms do not play a major role in the genetic predisposition to OCD, although a minor contribution cannot be ruled out.

Effect of the 5-HT2 antagonist mianserin on cognitive dysfunction in chronic schizophrenia patients: an add-on, double-blind placebo-controlled study.

UNLABELLED Preponderance of serotonin 5-HT2A antagonism over dopamine D2 blockade exerted by atypical antipsychotics may contribute to their cognitive-enhancing effect. In a double-blind placebo-controlled study we examined the effect of add-on mianserin (15 mg/day), an agent with marked 5-HT2A antagonism, on cognitive functioning in 30 chronic hospitalized DSM-IV schizophrenia patients stabilized on typical antipsychotics. The Automated Neuropsychological Assessment Metrics (ANAM) battery was used to assess learning, memory and sustained attention; Wisconsin Card Sorting Test (WCST) to assess executive function at baseline and endpoint (4 weeks). Clinical assessment included appropriate rating scales. The mianserin group overperformed the placebo group on selective ANAM memory/learning tests, reflected in moderate-to-high effect size values. No between-group differences were revealed in WCST and clinical ratings. CONCLUSIONS Improved performance on selective neurocognitive tests with addition of the 5-HT2A antagonist mianserin to typical antipsychotics indicates a possible role of the 5-HT system in cognitive-enhancing effects. The effect of flexible doses of mianserin on cognitive deficits in a broader schizophrenia population merits further investigation.

Multivariate Profile Charts for Statistical Process Control

The multivariate profile (MP) chart is a new control chart for simultaneous display of univariate and multivariate statistics. It is designed to analyze and display extended structures of statistical process control data for various cases of grouping, reference distribution, and use of nominal specifications. For each group of observations, the scaled deviations from reference values are portrayed together as a modified profile plot symbol. The vertical location of the symbol is determined by the multivariate distance of the vector of means from the reference values. The graphical display in the MP chart enjoys improved visual characteristics as compared with previously suggested methods. Moreover, the perceptual tasks required by the use of the MP chart provide higher accuracy in retrieving the quantitative information. This graphical display is used to display other combined univariate and multivariate statistics, such as measures of dispersion, principal components, and cumulative sums

Heavy metals in the lichen Caloplaca aurantia from urban, suburban and rural regions in Israel (a comparative study)

The content of eight heavy metals: Mn, Zn, Fe, Pb, Ni, Cu, Cr and Cd in the lichen Caloplaca aurantia growing on roof-tiles in urban, suburban and rural settlements in Israel has been evaluated. The data obtained and their statistical analysis indicated the following: (1) The content of all the above listed metals was generally higher in the lichen growing in ‘town’ than in ‘village’ areas; among these metals Ni and Zn were found most suitable for the distinction between ‘town’ and ‘village’ settlements. (2) Comparisons of the coefficient of variation of metal content values in ‘town’ versus ‘village’ furnished indications on the dispersion capacity of the metal particles. (3) Correlation analysis among the metals in ‘town’ and ‘village’ resulted in suggestive information on the emission sources. (4) Leaching tests indicated the tenacity of metal retainment and incorporation efficiency into the lichen tissue. Lichen species like C. aurantia, which grow both in ‘clean’ and metal contaminated areas are suggested as comparative monitors and for assessing periodical changes in metal output and concentrations.

Secretory otitis media in adults : II. The role of mastoid pneumatization as a prognostic factor

We analyzed clinically 102 ears with secretory otitis media (SOM) belonging to 72 adult patients who during their adult life had not suffered previously from ear disease. As in children, most of the cases (63%) could be traced directly to an upper respiratory tract infection. The most striking finding was the preponderance of poorly pneumatized mastoids—which were measured planimetrically — among our SOM cohort. This was found in adult SOM ears compared to contralateral healthy ears (4.59 versus 7.88 cm2), as well as when all 102 SOM ears were compared with values of the normal population (5.41 versus 12.9 cm2). This study showed that poorly pneumatized mastoids are a significant risk factor as far as adult SOM is concerned.

Low-Dose Mirtazapine: A New Option in the Treatment of Antipsychotic-Induced Akathisia. A Randomized, Double-Blind, Placebo- and Propranolol-Controlled Trial

BACKGROUND Preliminary evidence indicates a beneficial effect of serotonin 2A (5-HT(2A)) receptor antagonists in antipsychotic-induced akathisia (AIA). We investigated the antiakathisia effect, safety, and tolerability of low-dose mirtazapine, an agent with marked 5-HT(2A) antagonism. METHODS In a 7-day double-blind trial, 90 antipsychotic-treated patients meeting DSM-IV criteria for AIA were randomly assigned to mirtazapine (n = 30; 15 mg), propranolol (n = 30; 80 mg), or placebo (n = 30). Primary outcome measures were between-group differences in Barnes Akathisia Scale (BAS) global scores and in the proportion of responders (reduction of > or = 2 points on BAS). Analysis was by intention to treat. RESULTS Twenty-four patients (26.6%) who were assigned treatment did not complete the study (7 mirtazapine, 8 propranolol, 9 placebo), due to lack of response (n = 19) and adverse events (n = 5). Both mirtazapine and propranolol significantly reduced AIA severity (BAS: -34% mirtazapine and -29% propranolol vs. placebo -11%; p = .012 and p = .023, respectively). Thirteen (43.3%) mirtazapine and 9 (30.0%) propranolol-treated patients versus 2 (6.7%) placebo-treated patients responded (the corresponding odds ratios 10.7 [95% confidence interval (CI), 2.1-53.3] and 6.0 [95% CI, 1.1-30.7]). Five (16.7%) of 30 propranolol-treated patients and none in the mirtazapine and placebo groups (p = .0195 for both) prematurely discontinued the study due to clinically significant hypotension or bradycardia. CONCLUSIONS The comparable efficacy and better tolerability of low-dose mirtazapine versus propranolol, the current first-line treatment for AIA, position mirtazapine as a favorable candidate for the treatment of acute AIA and may improve current therapeutic practices.

The effect of famotidine addition on olanzapine-induced weight gain in first-episode schizophrenia patients: a double-blind placebo-controlled pilot study

Olanzapine treatment is associated with substantial weight gain. In this double-blind placebo-controlled study we evaluated whether the H2 antagonist famotidine may prevent/attenuate olanzapine-induced weight gain. Fourteen first-episode DSM-IV schizophrenia patients were randomly allocated to receive either famotidine (40 mg/day, n=7) or placebo (n=7) in addition to olanzapine (10 mg/day) for 6 weeks. All patients completed the trial. Patients in both groups showed a similar increase in body weight (olanzapine/famotidine: 4.8 (3.2) kg and olanzapine/placebo: 4.9 (1.6) kg, respectively; a between-group difference of 0.14 (1.3) kg). Four of seven (57.1%) patients in the olanzapine/famotidine group and three of seven (42.9%) in the olanzapine/placebo group gained at least 7% of their initial body weight, a cut-off for clinically significant weight gain. Famotidine addition was safe and well tolerated and did not interfere with olanzapines therapeutic effect. In conclusion, famotidine (40 mg/day for 6 weeks) is not effective in preventing/attenuating weight gain in olanzapine-treated first-episode schizophrenia patients.

Familial aggregation of schizophrenia-spectrum Disorders and obsessive-compulsive associated disorders in schizophrenia probands with and without OCD

A substantial proportion of schizophrenia patients also has obsessive‐compulsive disorder (OCD). To further validate the clinical validity of a schizo‐obsessive diagnostic entity, we assessed morbid risks for schizophrenia‐spectrum disorders and OC‐associated disorders in first‐degree relatives of schizophrenia probands with and without OCD. Two groups of schizophrenia probands [with OCD (n = 57) and without OCD (n = 60)] and community‐based controls (n = 50) were recruited. One hundred eighty two first‐degree relatives of probands with OCD‐schizophrenia, 210 relatives of non‐OCD schizophrenia probands, and 165 relatives of community subjects were interviewed directly [59.3% (108/182), 51.9% (109/210), and 54.5% (90/165), respectively], using the Structured Clinical Interview for Axis‐I DSM‐IV Disorders and Axis II DSM‐III‐R Personality Disorders and the remaining relatives were interviewed indirectly, using the Family History Research Diagnostic Criteria. Relatives of OCD‐schiozphrenia probands had significantly higher morbid risks for OCD‐schizophrenia (2.2% vs. 0%; P = 0.033) and OCPD (7.14% vs. 1.90%; P = 0.014), and a trend towards higher morbid risk for OCD (4.41% vs. 1.43%; P = 0.08) compared to relatives of non‐OCD schizophrenia probands. When morbid risks for OCD, OCPD, and OCD‐schizophrenia were pooled together, the significant between‐group difference became robust (13.74% vs. 3.33%; P = 0.0002). In contrast, relatives of the two schizophrenia groups did not differ significantly in morbid risks for schizophrenia‐spectrum disorders, mood disorders, or substance abuse disorders. A differential aggregation of OC‐associated disorders in relatives of OCD‐schizophrenia versus non‐OCD schizophrenia probands, provides further support for the validity of a putative OCD‐schizophrenia (“schizo‐obsessive”) diagnostic entity.