Changes in Electric Brain Response to Affective Stimuli in the First Week of Antidepressant Treatment: An Exploratory Study.

INTRODUCTION: Asymmetrical alpha and frontal theta activity have been discussed as neurobiological markers for antidepressant treatment response. While most studies focus on resting-state EEG, there is evidence that task-related activity assessed at multiple time points might be superior in detecting subtle early differences. METHODS: This was a naturalistic study design assessing participants in a psychiatric in- and outpatient hospital setting. We investigated stimulus-related EEG asymmetry (frontal and occipital alpha-1 and alpha-2) and power (frontal midline theta) assessed at baseline and 1 week after initiation of pharmacological depression treatment while presenting affective stimuli. We then compared week 4 responders and nonresponders to antidepressant treatment. RESULTS: Follow-up analyses of a significant group × emotion × time interaction (p < 0.04) for alpha-1 asymmetry showed that responders differed significantly at baseline in their asymmetry scores in response to sad compared to happy faces with a change in this pattern 1 week later. Nonresponders did not show this pattern. No significant results were found for alpha-2, occipital alpha-1, and occipital alpha-2 asymmetry or frontal midline theta power. DISCUSSION: Our study addresses the gap in comparisons of task-related EEG activity changes measured at two time points and supports the potential value of this approach in detecting early differences in responders versus nonresponders to pharmacological treatment. Important limitations include the small sample size and the noncontrolled study design.

Test-Retest Reliability of Frontal and Parietal Alpha Asymmetry during Presentation of Emotional Face Stimuli in Healthy Subjects.

Resting-state and event-related frontal alpha asymmetry have been suggested as potential neurobiological biomarkers for depression and other psychiatric conditions. To be used as such, sufficient test-retest reliability needs to be demonstrated. However, test-retest reliability is underinvestigated for event-related alpha asymmetry. The objective of this study was to examine both short-term within-session and long-term between-session reliability of stimulus-related medial and lateral frontal as well as parietal alpha EEG asymmetry in healthy subjects during a simple emotional face processing task. Twenty-three healthy adults participated in two sessions with a test-retest interval of about 1 week. Reliability was estimated with Pearson's correlation coefficient and paired t test. Results revealed moderate to high within-session reliability of stimulus-related alpha asymmetry for all electrode sites and both conditions. Alpha asymmetry mean values did not change significantly within sessions. Between-session reliability was fair for frontomedial and moderate for frontolateral stimulus-related asymmetry. Exploratory exclusion of subjects with unstable between-session self-rating scores of emotional state and empathy toward stimuli resulted in some higher reliability values. Our results indicate that stimulus-related alpha asymmetry may serve as a useful electrophysiological tool given its adequate within-session reliability. However, long-term stability of stimulus-related frontal alpha asymmetry over 1 week was comparatively low and varied depending on electrode position. Influencing state factors during EEG recording, such as current mood or stimulus engagement, should be considered in future study designs and analyses. Further, we recommend to analyze alpha asymmetry from both frontomedial and frontolateral sites.

Reduced Sensitivity to Non-Fear-Related Stimulus Changes in Panic Disorder.

Panic disorder (PD) is associated with increased body vigilance and reduced cognitive resources directed at non-fear-related stimuli, particularly in the absence of stimulus-rich environments. To date, only few studies have investigated whether this deficit in PD is reflected in reduced mismatch negativity (MMN), an event-related potential indexing preattentive sensitivity to unexpected stimulus changes. We tested 35 patients affected by PD and 42 matched healthy controls in an oddball paradigm, using frequency and duration deviant stimuli to measure auditory MMN. PD patients displayed reduced duration MMN amplitudes in comparison to healthy controls. No group differences were detected for duration MMN latency, as well as frequency MMN indices. Results support the notion of reduced processing of non-fear-related stimuli in PD patients, particularly with regard to the preattentive processing of sound duration deviants. Additionally, our findings are in line with clinical studies reporting divergent deficits in preattentive processing of frequency and duration deviants.

Attentional dysfunction in abstinent long-term cannabis users with and without schizophrenia.

Long-term cannabis use may confer cognitive deficits and increased risk of psychosis. However, the relationship between cannabis use and schizophrenia is complex. In particular, little is known about the effects of chronic cannabis use on the attention-related electric brain response in schizophrenia. We investigated auditory novelty and oddball P300 evoked potentials in a mixed sample of first-episode and chronic schizophrenic patients and healthy controls with (SZCA, n = 20; COCA, n = 20, abstinence ≥28 days) or without (SZ, n = 20; CO, n = 20) chronic cannabis use. Duration of regular cannabis use was 8.3 ± 5.6 (SZCA) and 9.1 ± 7.1 (COCA) years. In general, schizophrenic patients showed reduced P300 amplitudes. Cannabis use was associated with both a reduced early and late left-hemispheric novelty P300. There was a significant 'diagnosis × cannabis' interaction for the left-hemispheric late novelty P300 in that cannabis use was associated with a reduced amplitude in the otherwise healthy but not in the schizophrenic group compared with their relative control groups (corrected p < 0.02; p > 0.9, respectively). The left-hemispheric late novelty P300 in the otherwise healthy cannabis group correlated inversely with amount and duration of cannabis use (r = -0.50, p = 0.024; r = -0.57, p = 0.009, respectively). Our study confirms attentional deficits with chronic cannabis use. However, cannabis use may lead to different cognitive sequelae in patients with schizophrenia and in healthy controls, possibly reflecting preexisting alterations in the endocannabinoid system in schizophrenia.

Pretreatment anterior cingulate activity predicts antidepressant treatment response in major depressive episodes.

Major depressive disorder leads to substantial individual and socioeconomic costs. Despite the ongoing efforts to improve the treatment for this condition, a trial-and-error approach until an individually effective treatment is established still dominates clinical practice. Searching for clinically useful treatment response predictors is one of the most promising strategies to change this quandary therapeutic situation. This study evaluated the predictive value of a biological and a clinical predictor, as well as a combination of both. Pretreatment EEGs of 31 patients with a major depressive episode were analyzed with neuroelectric brain imaging technique to assess cerebral oscillations related to treatment response. Early improvement of symptoms served as a clinical predictor. Treatment response was assessed after 4 weeks of antidepressant treatment. Responders showed more slow-frequency power in the right anterior cingulate cortex compared to non-responders. Slow-frequency power in this region was found to predict response with good sensitivity (82 %) and specificity (100 %), while early improvement showed lower accuracy (73 % sensitivity and 65 % specificity). Combining both parameters did not further improve predictive accuracy. Pretreatment activity within the anterior cingulate cortex is related to antidepressive treatment response. Our results support the search for biological treatment response predictors using electric brain activity. This technique is advantageous due to its low individual and socioeconomic burden. The benefits of combining both, a clinically and a biologically based predictor, should be further evaluated using larger sample sizes.

Impact of emotional valence on mismatch negativity in the course of cortical face processing.

Various aspects of cortical face processing have been studied by assessing event related potentials (ERP). It has been described in the literature that mismatch negativity (MMN), a well-studied ERP, is not only modulated by sensory features but also emotional valence. However, the exact impact of emotion on the temporo-spatial profile of visual MMN during face processing remains inconsistent. By employing a sequential oddball paradigm using both neutral and emotional deviants, we were able to differentiate two distinct vMMN subcomponents. While an early subcomponent at 150-250 ms is elicited by emotional salient facial stimuli, the later subcomponent at 250-400 ms seems to reflect the detection of regularity violations in facial recognition per se, unaffected by emotional salience. Our results suggest that emotional valence is encoded in vMMN signal strength at an early stage of facial processing. Furthermore, we assume that of facial processing consists of temporo-spatially distinct, partially overlapping levels concerning different facial aspects.

Early orbitofrontal hyperactivation in obsessive-compulsive disorder.

Dysfunctional activity in the orbitofrontal cortex (OFC) is one of the core features in the pathophysiology of obsessive-compulsive disorder (OCD). Neuroimaging studies indicate orbitofrontal hyperactivation during the resting state as well as during symptom provocation, whereas orbitofrontal hypoactivation has been reported during tasks designed to dissociate specific cognitive processes. Combined magnetoencephalic and functional magnetic resonance imaging studies show early involvement of the OFC in stimulus processing in healthy subjects. However, it is unclear whether OFC activation is dysfunctional at an early stage in patients with OCD. We investigated early electrical OFC activation evoked by reward and punishment feedback in a visual probabilistic object reversal task (pORT). Patients with OCD (n=23) and healthy controls (n=27), matched for gender, age and educational level, performed the pORT during a 29-channel electroencephalographic recording. Low resolution brain electromagnetic tomography was applied to localize orbitofrontal sources of neuronal activity at 80 to 200 ms post-stimulus. Group comparison showed significantly higher orbitofrontal activation in OCD patients at 100-120 ms after the reward stimulus. No group differences were found with respect to OFC activation in response to punishment stimuli and in task performance. Results substantiate dysfunctional OFC activity at a very early stage in the processing of reward stimuli in patients with OCD. Our results provide support for the assumption that the OFC plays a more active role in the processing of visual stimuli as previously supposed. As orbitofrontal hyperactivation following rewarding feedback occurred as early as 100 ms after receipt of the visual stimulus in patients with OCD, and as we did not find any OFC dysfunction following negative feedback, our findings may point towards a specific early disturbance of reward processing in OCD. This finding might have implications for cognitive behavioural therapy of this disorder.

COMT Val108/158Met genotype modulates human sensory gating.

BACKGROUND: The catechol-O-methyltransferase (COMT) Val(108/158)Met polymorphism of the dopamine system is essential for prefrontal cortex processing capacity and efficiency. In addition, dopaminergic neurotransmission is also associated with the sensory gating phenomenon protecting the cerebral cortex from information overload. It is however unclear if COMT genotype as a predictor of prefrontal efficiency modulates sensory gating on the level of the auditory cortex, i.e. the gating of the auditory evoked P50 and N100 components. METHODS: P50 and N100 gating and COMT Val(108/158)Met genotype were determined in 282 healthy subjects of German descent carefully screened for psychiatric or neurological disorders. RESULTS: A significant effect of the COMT genotype was observed for N100 gating (F=4.510, df=2, p=0.012) but not for P50 gating (F=0.376, df=2, p=0.687). Contrast analysis showed that Met/Met individuals had poorer N100 gating compared to Val/Met (F=-12.931, p=0.003) and the Val/Val individuals (F=-11.056, p=0.057). CONCLUSION: The results indicate that a high prefrontal efficiency as suggested by the COMT Met/Met genotype is associated with to a poor sensory gating of the N100 component. This would fit in a model where a high prefrontal processing capacity allows a pronounced afferent input of sensory information from the auditory cortex as reflected by a poor sensory gating. The more pronounced prefrontal contribution to the N100 compared to the P50 component may explain the exclusive genotype association with the N100 sensory gating. This preliminary model should be replicated and validated in future investigations.

Fronto-lateral alpha power asymmetry in panic disorder.

Resting state alpha power asymmetry in frontal and temporal regions has been reported in various clinical populations, possibly indicating deficits in prefrontal control. In panic disorder (PD), results regarding alpha asymmetric activity to date have been mixed. This study compared 55 PD patients and 42 healthy controls (HC) with regards to resting state alpha power asymmetry. Our results show more right-than-left fronto-lateral alpha power in PD, whereas at other sites and in HC no significant differences were detected. These results support the notion of altered neurobiological processes in PD that possibly represent a vulnerability to the experience of panic attacks. Further studies are needed to clarify potential causal implications of this finding in the genesis of PD, as well as to specify the functional significance of fronto-lateral alpha power asymmetry in PD.

EEG Frontal Asymmetry and Theta Power in Unipolar and Bipolar Depression.

BACKGROUND: Distinguishing between unipolar and bipolar depression is of high clinical relevance. However, there is sparse research directly comparing these groups in terms of EEG activity. METHOD: We investigated 87 participants' left and right EEG frontal alpha-1, alpha-2, and theta activity related to happy and sad face stimuli in unipolar (UD, n=33) and bipolar (BD, n=22) depressed participants, and controls without depression (HC, n=32). RESULTS: Post-hoc analysis of an observed hemisphere x group interaction (p< .037) showed significant differences in alpha-1 asymmetry only for the comparison of UD and HC (p< .006). Further analysis of a significant emotion x group interaction (p= .001) revealed a differential impact of stimulus valence on theta power between the groups (p< .001). The valence dependent theta power of the BD differed from that of the UD (p< .0002) and the HC (p< .004). Alpha-1 asymmetry classified HC and both depressed groups with an accuracy of .69. Valence-related theta classified BD from UD with an accuracy of .83. Leave-one-out cross validation resulted in slightly reduced accuracy. LIMITATIONS: Important limitations were the small sample size and that subjects were not medication-free. CONCLUSIONS: Our results demonstrate the value of simple, task related EEG activity for differentiating not only healthy individuals from those with depression, but also individuals with unipolar depression from those with bipolar depression.

Peripheral oxytocin is inversely correlated with cognitive, but not emotional empathy in schizophrenia.

Endogenous oxytocin has been associated with different aspects of social cognition in healthy subjects and patients with schizophrenia. In this pilot study, we investigated the relationship between plasma oxytocin and oxytocin level changes induced by empathy-eliciting, attachment-related movie scenes with correlates of cognitive and emotional empathy in patients and healthy controls. The Multifaceted Empathy Test (MET) and the Interpersonal Reactivity Index (IRI) were administered to patients with schizophrenia (N = 35, 12 females) and healthy controls (N = 35, 12 females) to estimate dimensions of cognitive and emotional empathy. Peripheral basal oxytocin concentrations and oxytocin responses to movie-based emotional stimuli were assessed using radioimmunoassay with sample extraction. In patients, induced oxytocin level changes were inversely correlated with MET cognitive empathy regarding negative emotional states. Controlling for non-social cognition and age revealed a significant negative association between basal oxytocin levels and MET cognitive empathy for positive emotions. In healthy subjects, oxytocin reactivity was inversely correlated with the IRI subscale "fantasy". Oxytocin was not related to any measure of emotional empathy. A hyper-reactive oxytocin system might be linked to impaired cognitive empathy as a part of a dysfunctional regulative circuit of attachment-related emotions and interpersonal stressors or threats by attribution of meaning. Healthy adults with a disposition to identify with fictional characters showed lower oxytocin reactivity, possibly indicating familiarity with movie-based stimuli. The oxytocinergic system may be involved in maladaptive coping mechanisms in the framework of impaired mentalizing and associated dysfunctional responses to interpersonal challenges in schizophrenia.

P50, N100, and P200 Sensory Gating in Panic Disorder.

Panic disorder (PD) has been linked to abnormalities in information processing. However, only little evidence has been published for sensory gating in PD. Sensory gating describes the brain's ability to exclude stimuli of low relevance from higher level information processing, thereby sustaining efficient cognitive processing. Deficits in sensory gating have been associated with various psychiatric conditions, most prominently schizophrenia. In this case-control event-related potential study, we tested 32 patients with PD and 39 healthy controls in a double click paradigm. Both groups were compared with regard to pre-attentive (P50), early-attentive (N100), and late-attentive (P200) sensory gating indices. Contrary to a hypothesized deficit, PD patients and healthy controls showed no differences in P50, N100 and P200 values. These results suggest that sensory gating seems to be functional across the pre-attentive, early-attentive, and late-attentive time span in this clinical population. Given this consistency across auditory sensory gating indices, further research aiming to clarify information processing deficits in PD should focus on other neurophysiological markers to investigate information processing deficits in PD (eg, P300, error-related negativity or mismatch negativity).

Treatment response of lithium augmentation in geriatric compared to non-geriatric patients with treatment-resistant depression.

BACKGROUND: Lithium augmentation (LA) of antidepressants is an effective strategy for treatment-resistant depression (TRD). Nevertheless, it is rarely used in geriatric patients. The purpose of this study was to investigate treatment response of LA in geriatric compared to non-geriatric patients. METHOD: In a prospective multicenter cohort study, severity of depression was measured weekly in 167 patients with unipolar depression (nage≥65years = 22; nage<65years = 145) at baseline and over at least four weeks of LA. RESULTS: Geriatric patients showed a significantly better response to LA compared to non-geriatric patients (Hazard Ratio = 1.91; p = 0.04). LIMITATIONS: An important limitation of our study is the lack of a control group of LA and the missing evaluation of side effects in both groups. CONCLUSIONS: This is the first study investigating the efficacy of LA for TRD in geriatric compared to non-geriatric patients. Our data suggest that LA is an effective treatment option in geriatric patients that clinicians might consider more frequently and earlier on in the course of treatment.

Effects of acute oral Delta9-tetrahydrocannabinol and standardized cannabis extract on the auditory P300 event-related potential in healthy volunteers.

Reduced amplitudes of auditory evoked P300 are a robust finding in schizophrenic patients, indicating deficient attentional resource allocation and active working memory. Delta9-Tetrahydrocannabinol (Delta9-THC), the main active constituent of Cannabis sativa, has been known to acutely impair cognitive abilities in several domains, particularly in memory and attention. Given the psychotic-like effects of Delta9-THC, a cannabinoid hypothesis of schizophrenia has been proposed. This prospective, double-blind, placebo-controlled cross-over study investigated the acute effects of cannabinoids on P300 amplitude in 20 healthy volunteers (age 28.2+/-3.1 years, 10 male) by comparing Delta9-THC and standardized cannabis extract containing Delta9-THC and cannabidiol (CBD). P300 waves were recorded during a choice reaction task. As expected, Delta9-THC revealed a significant reduction of P300 amplitude at midline frontal, central, and parietal electrodes. CBD has been known to abolish many of the psychotropic effects of Delta9-THC, but, unexpectedly, failed to demonstrate a reversal of Delta9-THC-induced P300 reduction. Moreover, there were no correlations between cannabinoid plasma concentrations and P300 parameters. These data suggest that Delta(9)-THC may lead to acute impairment of attentional functioning and working memory. It can be speculated whether the lack of effect of CBD may be due to an insufficient dose used or to an involvement of neurotransmitter systems in P300 generation which are not influenced by CBD.

Comparison of midlatency auditory sensory gating at short and long interstimulus intervals.

RATIONALE: Suppression of P50, N100 and P200 auditory evoked responses in a dual-click procedure is considered an index for the multistage sensory gating process. Whereas most studies use a protocol with long interstimulus intervals of 8-12 s between the stimuli pairs, there is also evidence that sensory gating occurs at much lower intervals. The aim of the study was to investigate whether a simple modified dual-click protocol with short interstimulus intervals elicts similar sensory gating ratios compared to the classic protocol. METHODS: P50, N100 and P200 amplitudes and sensory gating ratios were measured in 23 healthy subjects with 2 different dual-click protocols in 1 session: (1) a simple oddball modified with short interstimulus intervals of about 2.8 s (ISI2), and (2), the classic used with long intervals of about 8 s (ISI8). RESULTS: The amplitudes of the P50, N100 and P200 responses were mostly comparable and correlated between both protocols. Mean sensory gating ratios (ISI8/ISI2) were as follows: P50, 35.4/36.4%; N40P50, 36.1/39.9%; N100, 44.4/48.4%; P200, 46.8/43.3%; N100P200, 45.3/41.8%; all differences between protocols, p > 0.1. P50 ratio scores did not show a sufficient correlation between protocols [intraclass correlation coefficient (ICC) P50, 0.13; N40P50, 0.0] compared to N100 (ICC, 0.79), P200 (ICC, 0.6) and N100P200 (ICC, 0.61). CONCLUSION: Our results contradict the assumption that long interstimulus intervals of about 8 s are absolutely necessary to elicit a marked sensory gating phenomenon for P50, N100 and P200 auditory responses (at least when using a protocol with a simple attention task). However, because only healthy subjects were investigated, no prediction can be made for psychiatric patients, in whom neuronal processing may be different.

Test-retest reliability of P50, N100 and P200 auditory sensory gating in healthy subjects.

RATIONALE: Suppression of middle latency auditory evoked responses is considered an index for the multistage sensory gating process. This has been observed in sequentially occurring P50, N100 and P200 components in a dual-click procedure. Since P50 sensory gating deficits have been observed in schizophrenic patients and first degree relatives, this parameter was suggested as an intermediate phenotype of the disease. However, most studies only show a low reliability for P50 sensory gating and neither N100 nor P200 sensory gating have been sufficiently tested. METHODS: Reliability of P50, N100 and P200 sensory gating was measured in 41 healthy subjects in two sessions, four weeks apart, using intra-class correlation. Sensory gating was calculated as ratio-gating (second response magnitude/first response magnitude x100) as well as difference-gating (first response magnitude minus second response magnitude). RESULTS: The difference-gating showed good to excellent reliabilities independently of the amplitude-measurement method applied (P50 peak-to-peak 0.75 and baseline-to-peak 0.74, N100 peak-to-peak 0.63 and baseline-to-peak 0.70, P200 peak-to-peak 0.82 and baseline-to-peak 0.79). Regarding ratio-gating, best temporal stability was observed for the P200 (peak-to-peak 0.58 and baseline-to-peak 0.62). Reliability of P50 ratio-gating strongly depends on the amplitude-measurement method (peak-to-peak 0.0 and baseline-to-peak 0.46). CONCLUSION: Regarding long-term reliability in healthy subjects the difference-gating of all three evoked responses and the ratio-gating of the P200 component may be useful tools for clinical or intermediate phenotype studies measuring different stages of the auditory sensory gating process. In contrast, the reliability of the P50 and N100 ratio-gating component seems to be insufficient for this purpose. However, long-term reliability remains to be confirmed in clinical samples.

Nerve growth factor in serum is a marker of the stage of alcohol disease.

Long-term alcohol abuse has deleterious effects on the peripheral and central nervous system. Nerve growth factor (NGF) is a pleiotropic neurotrophic protein involved in development, maintenance of function and regeneration of nerve cells. We examined patients in different stages of alcohol disease and measured their NGF serum concentrations based on the hypothesis that these reflect the state of disease. We examined 57 patients suffering from alcohol-dependence for more than 2 years (DSM IV) on day 8 of a qualified withdrawal, 18 patients with Korsakoff's syndrome and 40 healthy controls. In addition to clinical examination, careful history taking and a standard neuropsychological test battery, serum NGF concentrations were measured by a highly sensitive enzyme-immunoassay. Of the 57 patients 9 had suffered from severe withdrawal delirium in the past, other clinical parameters were alike. Cognitive test performance did not differ from the control group. Mean NGF levels of controls amounted to 42.1pg/ml (S.D. 68.0); mean levels of patients with alcohol dependence were raised significantly to 401.5pg/ml (S.D. 932.6) without delirium in the past and even further to 3292.5pg/ml (S.D. 4879.6) with former withdrawal delirium. By contrast, patients with persistent amnestic disorder (Korsakoff's syndrome) showed values identical to the controls. NGF serum levels were significantly elevated in alcohol-dependent patients, more so in those with prior delirium. Their cognitive tests being normal, this possibly reflects the activity of NGF as an endogenous repair mechanism for damaged neurons. In accordance with this hypothesis, NGF values are "normal" in patients with persistent alcohol-related cognitive decline.

Cannabis induces different cognitive changes in schizophrenic patients and in healthy controls.

It is known that 60 to 80% of schizophrenic patients show deficits in cognition. There may be an increase in these deficits as a result of additional regular use of cannabis. The aim of the study was to evaluate the effect of chronic cannabis consumption on the cognitive functions of schizophrenic patients and healthy control subjects after a minimum abstinence time of 28 days. The study sample consisted of 39 schizophrenics (19 cannabis-abusers and 20 non-abusers) and 39 healthy controls (18 cannabis-abusers, 21 non-abusers). In a 2x2-factorial design (Diagnostic Groups [healthy controls, schizophrenic patients]xCannabis abuse [without, with]) with diagnostic group and cannabis consumption considered between-subject factors) we tested the hypothesis that dually diagnosed patients (i.e. suffering both from schizophrenia and cannabis abuse) perform worse in neuropsychological tests than schizophrenic patients without cannabis abuse. On the whole, schizophrenic patients performed worse than healthy control subjects. Surprisingly, rather than deteriorating neuropsychological performance, regular cannabis abuse prior to the first psychotic episode improved cognition in some tests. This was even more pronounced when regular consumption started before the age of 17. On the other hand, cannabis use deteriorated test performance in healthy controls, especially in cases when regular consumption started before the age of 17. To sum up, regular cannabis abuse has a different effect on cognitive function in schizophrenic patients and healthy controls.

[Coercive Interventions in Psychiatry: Prevention and Patients' Preferences]. / Psychiatrische Zwangsmaßnahmen: Prävention und Präferenzen aus Patientenperspektive.

Objective This study explores patients' preferences and measures of prevention of coercive methods in psychiatric treatment. Methods Structured interviews of 90 patients with psychotic disorders were undertaken, most of whom had previously experienced coercive methods. Results Participants saw preventive potential in a wider availability of individual non-pharmacological therapy, improvement of staff professional competence and communication skills, high staff-to-patient ratios and retreat facilities. The majority of participants preferred forced medication and manual restraint in case of self-endangerment, and forced medication and mechanical restraint in the event of endangerment of others. Conclusion Patients' suggestions relating to prevention of coercion are in line with most expert's opinions. In case coercive methods are required, manual restraint and application of forced medication is accepted as the treatment of choice according to most patients in case of self-endangerment.