Predictors of treatment response and length of stay for inpatients with major depression

Predictors of Remission in Acute and Continuation Treatment of Depressive Disorders

Objective

To identify factors predicting remission of depression during acute (12 weeks) and continuation treatment (12 months) using a 1-year, naturalistic prospective study design.

Methods

Patients with depressive disorders were recruited from Chonnam National University Hospital in South Korea from March 2012 to April 2017. At baseline, 1,262 patients received outpatient therapy, and sociodemographic and clinical data were obtained. Clinical visits took place every 3 weeks during the acute treatment phase (at 3, 6, 9, and 12 weeks; n = 1,246), and every 3 months during the continuation treatment phase (at 6, 9, and 12 months; n = 1,015). Remission was defined as a Hamilton Depression Rating Scale score ≤ 7.

Results

The remission rate was 43.3% at 12 weeks and 70.4% at 12 months. In multivariate analyses, remission during the acute treatment phase was more likely in patients with a shorter-duration present episode, higher functioning, and good social support. Remission during the continuation treatment phase was more likely in patients with fewer previous depressive episodes and/or a lower baseline stress score.

Conclusion

Factors predicting depressive disorder remission may differ between the acute and continuation treatment phases.

Effectiveness of electroconvulsive therapy on treatment-resistant depressive disorder: A population-based mirror-image study

The clinical effects of electroconvulsive therapy (ECT) in treatment-resistant depressive disorders need to be systemically examined in representative samples. This study aimed to examine whether ECT reduced re-hospitalization within one year after discharge. The authors used the Psychiatric Inpatients Medical Claim Dataset, a subset of a total population health claims database from Taiwan, to include 784 inpatients with treatment-resistant depressive disorders screened for ECT during hospitalization between 2001 and 2011. The same number of comparison subjects was selected by frequency matching on the demographic and clinical characteristics. Using a mirror-image comparison design, we compared group differences in re-hospitalization rate, number of hospital days, number of emergency department visits, and direct medical costs during the 1-year pre- and post-ECT periods. The modifying effects of patients' characteristics on these outcomes were also explored. The results showed that ECT was associated with a significant decrease in the rate of hospitalizations and emergency department visits over the 1-year follow-up period. However, there were no significant difference in the reduced rate of hospitalizations between ECT and comparison group. Demographic and clinical characteristic had no modifying effect on the odds of psychiatric hospitalization. In conclusions, ECT could reduce the rate of hospitalization and number of emergency department visits in patients with treatment-resistant depressive disorders. However, the study results might be biased by the inherent deficits of mirror-image design.

Results of the European Group for the Study of Resistant Depression (GSRD) - basis for further research and clinical practice

Objectives: The overview outlines two decades of research from the European Group for the Study of Resistant Depression (GSRD) that fundamentally impacted evidence-based algorithms for diagnostics and psychopharmacotherapy of treatment-resistant depression (TRD). Methods: The GSRD staging model characterising response, non-response and resistance to antidepressant (AD) treatment was applied to 2762 patients in eight European countries. Results: In case of non-response, dose escalation and switching between different AD classes did not show superiority over continuation of original AD treatment. Predictors for TRD were symptom severity, duration of the current major depressive episode (MDE), suicidality, psychotic and melancholic features, comorbid anxiety and personality disorders, add-on treatment, non-response to the first AD, adverse effects, high occupational level, recurrent disease course, previous hospitalisations, positive family history of MDD, early age of onset and novel associations of single nucleoid polymorphisms (SNPs) within the PPP3CC, ST8SIA2, CHL1, GAP43 and ITGB3 genes and gene pathways associated with neuroplasticity, intracellular signalling and chromatin silencing. A prediction model reaching accuracy of above 0.7 highlighted symptom severity, suicidality, comorbid anxiety and lifetime MDEs as the most informative predictors for TRD. Applying machine-learning algorithms, a signature of three SNPs of the BDNF, PPP3CC and HTR2A genes and lacking melancholia predicted treatment response. Conclusions: The GSRD findings offer a unique and balanced perspective on TRD representing foundation for further research elaborating on specific clinical and genetic hypotheses and treatment strategies within appropriate study-designs, especially interaction-based models and randomized controlled trials.

High occupational level is associated with poor response to the treatment of depression: A replication study

Major depressive disorder (MDD) is a leading cause of disability and inability to work. There is evidence that occupational factors may precipitate a MDD episode and interfere with the recovery process. In a previous investigation, we found that those employed in high occupational levels had a worse outcome after treatment for depression (Mandelli et al., 2016). The aim of the present study was to further investigate response to treatments for depression according to occupational status on an independent sample of MDD patients. Six hundred and forty-seven (647) subjects with a stable working occupation were taken from a larger independent sample of MDD patients evaluated for response and resistance to treatment for depression, after at least one adequate treatment trial. Three broad occupational categories were considered: 'manager', 'white-collar', 'blue-collar' and 'self-employed'. Managers had the highest rate of non-response and resistance to treatments. White-collar workers also had high non-response and resistance rates. At the opposite, Blue-collar workers had significantly lower rates of non-response and resistance. Self-employed were in between White- and Blue-collar workers and did not significantly differ from the other occupational categories. The findings of this replication study substantially support our previous observations. MDD patients employed in high-middle occupations may have a less favorable outcome after standard treatments of depression. Working stressful condition and other psychosocial factors at work should be investigated more closely in relation to treatment outcomes in MDD.

Opinion paper: poor response to treatment of depression in people in high occupational levels

The working environment may have a significant effect on response to treatment of depression and this issue has not yet been sufficiently addressed in the scientific literature. There is evidence showing that being engaged in high-level positions can be an obstacle to the success of treatment. This article discusses the few evidence in the literature and some of the possible mechanisms involved. Specific personality attributes and difficulties in adapting to depression may delay access to care and may also reduce treatment compliance. The presence of stress in jobs that require high cognitive function and lack of social support may be elements that hinder the recovery process. Residual symptoms that impact on cognitive functions may undermine adherence to treatment and adversely affect the response. The implications of these issues are potentially relevant for clinical practice in the treatment of depression and for future research.

Characterization, Categorization, and 5-Year Mortality of Medicine High Utilizer Inpatients

BACKGROUND

Patients who are frequently admitted to Medicine inpatient services comprise a distinct subset of readmitted patients about whom not much is known.

OBJECTIVE

We sought to characterize this group including mortality rates, with the goal of better understanding this population.

DESIGN

Observational study of frequently hospitalized patients defined as 4 or more admissions over a 6-month period, with hospitalization defined as nonelective admission to the hospital.

SETTING

Single large academic medical center.

PATIENTS

Adult inpatients on general medicine and medicine subspecialty services.

MEASUREMENTS

The number of nonelective medicine hospitalizations, age, clinical conditions and comorbidities, calculation of an age-adjusted Charlson Comorbidity Index (CCI), outpatient and emergency department visits, length of stay, costs of hospitalization, and mortality over a 5-year period. Descriptive statistics were used to characterize variables of interest.

RESULTS

We identified 153 patients with a total of 781 nonelective hospitalizations, totaling greater than 4000 hospital days and with charges of approximately US$9 million during the 6 months. Nearly all had insurance coverage and good outpatient follow-up (median of 7 appointments over the 6-month study period). Only 14% of those admissions qualified for observation status. Over 40% of patients had comorbid mental health disease or chronic narcotic dependence. Twenty-nine percent of patients died within 1 year; 50% were dead within 5 years. Age-adjusted CCI scores ranged annually from 3.00 to 3.58 among surviving patients versus 4.31 to 6.60 among deceased patients.

CONCLUSIONS

These findings point to distinct groups of patients who are frequently hospitalized, and therefore would benefit from tailored management strategies: Those with progression of end-stage disease comprised one-third of the group and targeting that subset with palliative care referrals could help decrease readmission rates. Those with recurrent exacerbations of a chronic medical condition could be managed through telemanagement programs. Those with exacerbations of chronic pain could be addressed through collaboration with pain management specialists. Individualized care management plans may be useful for all, especially the latter two groups. Based on differences between survivors and deceased patients, an age-adjusted CCI score of 4 or 5 could be valuable sensitive or specific cutoffs, respectively, for predicting those who would benefit most from palliative care consultation regarding end-of-life goals and management.

Nonremission and time to remission among remitters in major depressive disorder: Revisiting STAR*D

BACKGROUND

Some individuals with major depressive disorder do not experience a remission even after one or more adequate treatment trials. In some others who experience remission, it happens at variable times. This study sought to estimate the prevalence of nonremission in a large sample of patient participating in the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial and to identify correlates of nonremission and time to remission among remitters.

METHODS

Using data from 3,606 participants of STAR*D, the study used cure regression modeling to estimate nonremission and jointly model correlates of nonremission and time to remission among the remitters.

RESULTS

Overall, 14.7% of the STAR*D participants were estimated to be nonremitters. Among remitters, the rate of remission declined over time. Greater severity, poorer physical health, and poor adherence with treatments were associated with both nonremission and a longer time to remission among the remitters in multivariable analyses, whereas unemployment, not having higher education, and longer duration of current episode were uniquely associated with nonremission; whereas, treatment in specialty mental health settings, poorer mental health functioning, and greater impairment in role functioning with a longer time to remission among remitters.

CONCLUSION

Poor treatment adherence and poor physical health appear to be common risk factors for both nonremission and longer time to remission, highlighting the importance of integrated care models that address both medical and mental healthcare needs and interventions aimed at improving treatment adherence.

Variation in Major Depressive Disorder Onset by Place of Origin Among U.S. Latinos

Using a nationally representative sample of 2514 U.S. Latinos, this study examined the extent to which major depressive disorder (MDD) onset differs by place of origin and the factors associated with it. The Kaplan-Meier method estimated the survival and hazard functions for MDD onset by place of origin, and Cox proportional hazards models identified its associative factors. Approximately 13% of the sample had experienced MDD in their lifetimes. Cuban respondents showed the highest survival function, while Puerto Ricans showed the lowest. With the entire sample, the smoothed hazard function showed that the risk of MDD onset peaked in the late 20s and early 80s. Puerto Rican respondents showed the highest risk of MDD during their 20s and 30s, whereas Cuban respondents showed a relatively stable pattern over time. The results from the Cox proportional hazards model indicated that age, sex, and marital status were significantly related to MDD onset (p < .05). In addition, the effect of U.S.-born status on MDD onset was greater among Mexican respondents than among Puerto Ricans. Findings from the present study demonstrate that different Latino subgroups experience different and unique patterns of MDD onset over time. Future research should account for the role of immigration status in examining MDD onset.

High occupational level is associated with poor response to treatment of depression

Depression may be complicated by work-related stress and, in turn, depression is a leading cause of disability in workplaces. Though available effective treatments, only one third of patients reach full remission after a first treatment trial and nearly half of the patients are non-responders. Occupational level has been found to be a reliable predictor of health outcome in the general population. In the present study we tested the potential association of occupational level of those in work with response to treatment of depression in a large multinational sample. Major depressive disorder patients (n=654) stratified in three occupational levels (high, middle, low) were considered for the present study. Response to last treatment for current episode and treatment resistant depression, defined as non-response to 2 or more previous adequate treatment trials, were considered the outcome variables. Depressed patients from the high occupational level had a higher level of educational achievement. They showed a significantly poorer response to the last treatment with lower remission rates and more treatment resistance than the other occupational level groups. They were treated less with Serotonin Reuptake Inhibitors (SRIs). Potential confounding factors did not influence the main effect. The present findings indicate that those working at a high occupational level may be a risk factor for poor response to medication for depression and this has potential implications for clinicians and their patients, for future research, for employers and for public policy.

Hospital stay in patients admitted for acute bipolar manic episodes prescribed quetiapine immediate or extended release: a retrospective non-interventional cohort study (HOME)

BACKGROUND

Bipolar manic episodes often require hospital admission to ensure patient safety. The antipsychotic quetiapine is a common treatment for bipolar mania and is available in immediate release (IR) and extended release (XR) formulations; however, outcomes in patients receiving these different formulations have not been directly compared in an acute hospital setting.

METHODS

We conducted a multinational, observational, retrospective cohort study to describe and compare hospital stay in patients admitted for an acute bipolar manic episode treated with quetiapine IR or XR from 1 October 2009-1 October 2010. The primary outcome measure was comparison of length of stay (LOS) using zero-truncated negative binomial regression.

RESULTS

In total, 1230 patients were included (659 in the IR cohort; 571 in the XR cohort). The median LOS (interquartile range) was 18.0 days (12.0, 28.0) in the IR cohort and 20.0 days (12.0, 34.0) in the XR cohort, respectively. LOS was not significantly associated with quetiapine formulation irrespective of whether or not clinical characteristics were taken into account (p = 0.820 and p = 0.386, respectively). Overall, 84.2% and 84.4% of patients in the IR and XR cohorts, respectively, had not previously used quetiapine; of these patients, 78.7% and 68.9% received one total daily dose, and 14.4% and 23.9% received dose titration. Over half of patients received antipsychotic monotherapy (53.1% and 58.3% in the IR and XR cohorts, respectively) and most received a daily quetiapine dose ≥ 400 mg (64.9% and 71.8%, respectively, for quetiapine monotherapy and 59.9% and 80.3%, respectively, for combination treatment). As a secondary outcome, multivariate analysis was used to identify other factors that affect LOS. Factors associated with a longer hospital stay included public funding versus private, maximum number of new medications administered, did not receive lithium and did not receive anxiolytics, sedatives/hypnotics (all p < 0.0001). Factors associated with a shorter hospital stay included presence of drug/alcohol abuse, living accompanied and having a psychiatric medical history (all p < 0.05).

CONCLUSIONS

LOS was not found to be associated with quetiapine formulation. However, most patients received only one total daily dose of quetiapine without dose titration, which was unexpected and contrary to current recommendations.

TRIAL REGISTRATION

TRIAL REGISTRATION

NCT01239589.

Predictors of depressive symptoms at hospital discharge in patients with major depressive disorder

OBJECTIVE

Often patients with major depressive disorder (MDD) leave the hospital with continued significant symptomatology. This study sought to evaluate demographic, clinical, and psychosocial predictors of the presence of clinically significant depressive symptoms, defined as a Modified Hamilton Rating Scale for Depression score of ≥ 14, immediately following hospitalization for MDD.

METHODS

The study enrolled 135 patients with MDD as part of a larger clinical trial investigating the efficacy of post-hospitalization pharmacologic and psychosocial treatments for depressed inpatients. Structured clinical interview and self-report data were available from 126 patients at hospital admission and discharge.

RESULTS

Despite the significant decreases in depressive symptoms over the course of hospitalization, 91 (72%) displayed clinically significant depressive symptoms at discharge. Multivariate logistic regression analysis revealed that female sex, earlier age of onset, and poorer social adjustment were unique predictors of symptom outcome.

CONCLUSIONS

Results suggest that a large proportion of patients leave the hospital with continued significant symptomatology, and the presence of such symptoms following hospitalization for MDD is likely to be explained by a combination of factors.

Clinical outcome and length of stay in an Italian Psychiatric Emergency Service

PURPOSE

The aims of the present study were to analyze outcome and to evaluate diagnosis-specific pattern of improvement during a brief hospitalization in a Psychiatric Emergency Service (PES) in a catchment area in Turin, Italy.

METHODS

A sample of 848 acute patients, consecutively hospitalized between January 2007 and December 2008 in the PES of the San Giovanni Battista Hospital, with diagnoses of non-affective psychosis-affective psychosis, depressive disorder and mania-and personality disorder (DSM-IV-TR) was recruited. All patients were assessed with the Brief Psychiatric Rating Scale (BPRS). One-way analysis of variance was used to measure patients' individual reliable and clinically significant change speed between BPRS assessments, in which change speed was referred to the division of the gap between admission and discharge scores over the number of days of length of stay (LOS).

RESULTS

The overall sample showed a significant improvement of BPRS total score and each domain during a brief hospitalization (mean LOS 10.5 days), with a different pattern between the diagnostic groups. A significant difference in change speed of BPRS resulted in the whole sample and in each diagnostic groups; patients with mania showed a significantly faster improvement.

CONCLUSIONS

A brief hospitalization in our service was shown to be highly effective. A different and diagnosis-specific patients' individual reliable and clinically significant change speed was observed, with a significantly faster improvement in patients with mania.

A Nation-Wide Evidence-Based Data Analysis of Repeated Suicide Attempts

Background: Taiwan is a high-risk area for suicide. Repeated suicide attempts are an important factor of suicide mortality. Yet there has been little research on the factors associated with repeated suicidal behavior in Taiwan. Aims: To explore the characteristics of repeated suicide attempts. Methods: Data were obtained from the National Health Insurance Research Database for 2005 to 2008. We then classified repeated suicide attempts into single method, two different methods, and three different methods. Results: A total of 1,004 inpatients were admitted to Taiwanese hospitals due to repeated suicide attempts, and the use of a single suicide method constituted the majority of the cases (71%). Risk factors related to repeated suicide attempts included being female, aged 30–39, suicide by poisoning using solid and/or liquid substances, the fall season, living in Northern Taiwan, more psychiatric nature of injury codes (N-codes), seeking medical attention from a psychiatrist, receiving more surgeries or procedures, a longer length of stay in a hospital, coming from a low-income background, and having a serious illness. Conclusions: Consultation and treatment are particularly crucial for patients with mental illnesses and other concomitant diseases, so that the psychiatric symptoms such as auditory and visual hallucinations can be controlled.

Comprehensive review of factors implicated in the heterogeneity of response in depression

BACKGROUND

Heterogeneity in overall response and outcomes to pharmacological treatment has been reported in several depression studies but with few sources that integrate these results. The goal of this study was to review the literature and attempt to identify nongenetic factors potentially predictive of overall response to depression treatments.

METHODS

A comprehensive review of the literature from the last 10 years was performed using three key databases (PubMed, EMBASE, and Cochrane). All relevant studies that met the inclusion criteria were selected and scored for their levels of evidence using the NICE scoring method. A subjective assessment of the strength of evidence for each factor was performed using predefined criteria.

RESULTS

Our broad search yielded 76 articles relevant to treatment heterogeneity. Sociodemographic factors, disease characteristics, and comorbidities were the most heavily researched areas. Some of the factors associated with more favorable overall response include being married, other social support, and low levels of baseline depressive symptoms. Evidence relating to baseline disease severity as a factor predictive of antidepressant response was particularly convincing among the factors reviewed. The presence of comorbid anxiety and pain contributed to worse antidepressant treatment outcomes.

CONCLUSIONS

Several factors either predictive of or associated with overall response to antidepressant treatment have been identified. Inclusion of factors predictive of response in the design of future trials may help tailor treatments to depression patients presenting to the average clinical practice, resulting in improved outcomes.

A model to incorporate genetic testing (5-HTTLPR) in pharmacological treatment of major depressive disorders

OBJECTIVE

To evaluate the benefit of pharmacogenetics in antidepressant treatment.

METHODS

In a simulated trial 100,000 subjects in a current episode of major depressive disorder (MDD) received citalopram or bupropion based on the clinician's decision (algorithm A) or following indications from 5-HTTLPR genetic testing (algorithm B), which effect size of was estimated from a meta-analysis of pharmacogenetic trials. A and B were compared in a cost-utility analysis (12 weeks). Costs (international $, 2010) were drawn from official sources. Treatment effects were expressed as quality-adjusted life weeks (QALWs). Outcome was incremental cost-effectiveness ratio (ICER).

RESULTS

Under base-case conditions, genetic test use was associated with increases in antidepressant response (0.062 QALWs) and tolerability (0.016 QALWs) but cost benefit was not acceptable (ICER = $2,890; $1,800-$4,091). However, when the joint effect on antidepressant response and tolerability was analyzed in two recurrent episodes, ICER dropped to $1,392 ($837-$1,982). Cost-effectiveness acceptability curve (CEAC) showed a >80% probability that ICER value fell below the commonly accepted 3 times Gross Domestic Product (GDP) threshold (World Health Organization) and therefore suggesting cost-effectiveness.

CONCLUSION

Notwithstanding some caveats (exclusion of gene-gene and gene-environment interactions; simple 5-HTTLPR architecture), this simulation is favourable to incorporate pharmacogenetic test in antidepressant treatment.

Venlafaxine or a second SSRI: Switching after treatment failure with an SSRI among depressed inpatients: a retrospective analysis

BACKGROUND

Approximately 50% of patients with major depressive disorder (MDD) do not respond after adequate first-line treatment with a selective serotonin reuptake inhibitor (SSRI). Special interest is paid to whether specialist level inpatient psychiatric care results differ from community studies.

AIM

To compare switching alternatives after treatment failure with an SSRI; switching to venlafaxine (Dexcel Pharma Israel) versus switching to another SSRI in depressed inpatients.

METHOD

A retrospective register study of inpatients was undertaken in a psychiatric tertiary care university center serving an urban catchment area in Israel with a population of more than 900,000.

RESULTS

A total of 401 MDD inpatients were assigned to antidepressant treatment. Of these, 232 records (47 venlafaxine, 185 SSRI) were included in the analysis. Patients assigned to venlafaxine treatment were older (mean age 64.3 ± 15 years versus 53.6 ± 17; p<0.01) and had more comorbid physical disorders (80% versus 57%; p<0.001). In the primary analysis, there was no statistical difference between groups in reduction in CGI-S total scores. The secondary end point of achieving a CGI-S score of 2 or less (1 = normal, not at all ill or 2 = borderline mentally ill) was statistically significantly better for the venlafaxine treated inpatients (P=0.02). AEs were reported less than 10% of patients in both groups.

CONCLUSION

Patients who remain severely depressed following treatment with an SSRI may gain benefit from the dual-action drug venlafaxine, rather than switching to another SSRI. These findings need to be further supported by prospective studies.

Relationship between a history of a suicide attempt and treatment outcomes in patients with depression

This study aimed to investigate the sociodemographic and clinical correlates of a history of a suicide attempt in people with depression and their relationship with treatment outcomes and subsequent suicidal ideation and deliberate self-harm. Patients with depressive disorders according to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, were recruited from 18 hospitals across South Korea. Data on sociodemographic and clinical characteristics were obtained, and scales assessing depression, anxiety, and general functioning were administered during the first 12 weeks of antidepressant treatment. Baseline characteristics, responses to treatment, and suicidal ideation/deliberate self-harm during the follow-up period were compared according to a reported history or not of a suicide attempt. In a total of 723 participants, 143 (19.8%) with a history of a suicide attempt had an earlier age of onset, longer duration of illness, and a greater number of depressive episodes. Levels of depression and anxiety were significantly higher at baseline in this group who also experienced significantly lower remission and response rates, as well as longer time to remission. The case group was more likely to experience new suicidal ideation and carry out a deliberate self-harm act during the 12-week treatment period. In conclusion, a history of a suicide attempt in a Korean population with depression was characterized by more severe psychopathology, poorer treatment outcomes, and higher subsequent suicidal ideation and self-harm. Therefore, more intensive and longer-term treatment with particular ongoing clinical attention to risk is indicated in patients with these distinct, chronic, and severe forms of depression and ongoing high suicide risk.

[Predictors of treatment duration for inpatients with mental disorders--a systematic literature review]

OBJECTIVES

In a systematic literature review we examined patient-related predictors of inpatient treatment duration for mental disorders.

METHODS

The databases Medline, Embase, Psyndex, PsycINFO and EBMR were systematically reviewed for studies in the years 1990-2009 regarding treatment duration of inpatients with ICD-10 diagnoses F3-F6. Also, their methodological quality was evaluated.

RESULTS

Twenty-one studies examining 28 predictors (sociodemographic, clinical) were included. Longer treatment durations were found for patients living alone and with higher levels of initial symptom severity, duration of mental disorder, number of previous treatments, work absenteeism, and comorbidity. Obsessive-compulsive, eating and personality disorders were treated longer than depressive, anxiety, somatoform, and adjustment disorders. Results for age, gender, education, and profession were inconsistent.

CONCLUSIONS

The studies included heterogeneous analyses, treatments, and patients. Despite their sufficient methodical quality, these results are only tentative.

Outcome and length of stay in psychiatric hospitalization, the experience of the University Clinic of Turin

BACKGROUND

Given the current tendency to shorten psychiatric hospitalization and change its organization, an issue could be raised regarding its outcomes.

PURPOSE

To analyze features related to length of stay in a short-term inpatient treatment, to study outcomes and to evaluate the diagnosis-specific effects of hospitalization.

METHOD

A sample of 310 consecutive hospitalized patients, with psychotic disorder, depressive disorder and bipolar disorder (DSM IV-TR), was recruited at the University Psychiatric Clinic, Service for Cognitive Disorders, Department of Neuroscience, University of Turin. Severity of illness was rated using the brief psychiatry rating scale (BPRS). We evaluated relations between length of stay and clinical and socio-demographic features (linear regression) and possible differences confronting BPRS scores at admission and discharge in the different diagnostic subgroups (ANOVA for repeated measures).

RESULTS

All the sample of patients showed a significant improvement in symptomatology during hospitalization. Worse symptomatology in anxiety-depression domain of BPRS at admission in the whole sample was positively correlated with length of stay. A longer length of stay was also shown in patients with diagnosis of depressive disorder. Finally, a different pattern of improvement of BPRS (total score and domains) was shown between the different diagnostic groups.

CONCLUSION

Brief hospitalization in our service was shown to be highly effective. Different diagnostic groups had different response to hospitalization, showing faster improvement in characteristic symptomatology, but the anxiety-depression domain showed the highest percentage of change for all the diagnostic groups. We therefore suppose that hospitalization has two effects: a specific (due to tailored therapies) and a non-specific one (due to non-specific therapy and to a placebo-like effect).