Prevalence of suicide attempt and associations with deliberate self-harm, mental health problems, drug misuse and traumatic experiences - a cross sectional survey of the Norwegian population.

BACKGROUND: Monitoring self-reported suicide attempts (SA) with nationally representative surveys is important to initiate suicide prevention strategies. The aim of the study was to assess the prevalence of SA and compare deliberate self-harm, (DSH), mental health, drug misuse and traumas between SA and non-suicide attempters (NSA). METHODS: In this cross-sectional survey of a representative sample (N=1757) of the Norwegian population, we compared people with self-reported SA (n=54) to NSA (n=1703) regarding sociodemographic data, mental health problems, drug misuse and exposure to trauma. RESULTS: The prevalence of SA was 3.1 %. There was a higher proportion of welfare recipients and more deliberate self-harm, mental health problems, drug misuse and traumas in the SA group compared to NSA. CONCLUSION: This national study confirms the association between suicide attempt and deliberate self-harm, mental health problems, drug misuse and traumas.

Healthcare utilisation, induced labour and caesarean section in the pregnancy after stillbirth: a prospective study.

OBJECTIVE: To investigate healthcare utilisation, induced labour and caesarean section (CS) in the pregnancy after stillbirth and assess anxiety and dread of childbirth as mediators for these outcomes. DESIGN: Population-based pregnancy cohort study. SETTING: The Norwegian Mother and Child Cohort Study. SAMPLE: A total of 901 pregnant women; 174 pregnant after stillbirth, 362 pregnant after live birth and 365 previously nulliparous. METHODS: Data from questionnaires answered in the second and third trimesters of pregnancy and information from the Medical Birth Registry of Norway. MAIN OUTCOME MEASURES: Self-reported assessment of antenatal care, register-based assessment of onset and mode of delivery. RESULTS: Women with a previous stillbirth had more frequent antenatal visits (mean 10.0; 95% CI 9.4-10.7) compared with women with a previous live birth (mean 6.0; 95% CI 5.8-6.2) and previously nulliparous women (mean 6.3; 95% CI 6.1-6.6). Induced labour and CS, elective and emergency, were also more prevalent in the stillbirth group. The adjusted odds ratio for elective CS was 2.5 (95% CI 1.3-5.0) compared with women with previous live birth and 3.7 (1.8-7.6) compared with previously nulliparous women. Anxiety was a minor mediator for the association between stillbirth and frequency of antenatal visits, whereas dread of childbirth was not a significant mediator for elective CS. CONCLUSIONS: Women pregnant after stillbirth were more ample users of healthcare services and more often had induced labour and CS. The higher frequency of antenatal visits and elective CS could not be accounted for by anxiety or dread of childbirth. TWEETABLE ABSTRACT: Women pregnant after stillbirth are ample users of healthcare services and interventions during childbirth.

Lower suicide intention in patients with personality disorders admitted for deliberate self-poisoning than in patients with other diagnoses.

BACKGROUND: People with deliberate self-poisoning and personality disorders are in increased risk for suicide. Intention and psychiatric features are important factors in a psychiatric evaluation and for planning aftercare. METHODS: Patients admitted to medical departments after deliberate self-poisoning were studied (n = 117). Patients with personality disorder according to (ICD-10, F.60-69) were compared to patients with affective disorders, substance use disorders, and unknown psychiatric diagnosis on Beck Suicide Intention Scale (SIS), Beck Suicide Ideation Scale (BSI), Beck Hopelessness Scale (BHS), and Beck Depression Inventory (BDI). RESULTS: The mean suicide intention score (SIS) was significantly lower among patients with personality disorders compared with patients with other psychiatric diagnoses 10.2 (95% CI 8.1-12.4) vs. 14.6 (95% CI 12.7-16.4) (p = 0.040). The hopelessness scores (BHS) were significantly higher among patients with personality disorders 13.0 (95% CI 10.9-15.2) compared with patients with affective disorders 8.2 (95% CI 6.1-10.3) and substance use disorders 9.9 (95% CI 5.2-14.6) (p = 0.0014) and unknown psychiatric diagnoses 10.6 (95% CI 9.1-12.2). There were no significant differences between the groups on suicide ideation (BSI) and depression (BDI). CONCLUSIONS: Although patients with personality disorders had lower suicide intention compared to patients with other psychiatric diagnoses, they reported significantly more hopelessness. This distinction is an important implication in the clinical assessment and planning of further treatment of DSP patients.

Post-traumatic stress among rescue workers after terror attacks in Norway.

BACKGROUND: On 22 July 2011, Norway was struck by two terror attacks. Seventy-seven people were killed, and many injured. Rescue workers from five occupational groups and unaffiliated volunteers faced death and despair to assist victims. AIMS: To investigate the level of, and associations between, demographic variables, exposure and work-related variables and post-traumatic stress symptoms (PTSS). METHODS: A cross-sectional study of general and psychosocial health care personnel, police officers, firefighters, affiliated and unaffiliated volunteers were conducted ~10 months after the terror attacks. The respondents answered a self-reported questionnaire. Post-traumatic stress disorder (PTSD) Checklist - specific (PCL-S) assessed PTSS. RESULTS: There were 1790 participants; response rate was 61%. About 70% of the professional rescue workers had previous work experience with similar tasks or had participated in training or disaster drills. They assessed the rescue work as a success. Firefighters and unaffiliated volunteers reported more perceived threat compared with the other groups. Among the professional personnel, the prevalence of sub-threshold PTSD (PCL 35-49) was 2% and possible PTSD (PCL ≥ 50) 0.3%. The corresponding figures among the unaffiliated volunteers were 24% and 15%, respectively. In the multivariate analysis, female gender (ß = 1.7), witnessing injured/dead (ß = 2.0), perceived threat (ß = 1.1), perceived obstruction in rescue work (ß = 1.6), lower degree of previous training (ß = -0. 9) and being unaffiliated volunteers (ß = 8.3) were significantly associated with PTSS. CONCLUSIONS: In the aftermath of a terror attack, professional rescue workers appear to be largely protected from post-traumatic stress reactions, while unaffiliated volunteers seem to be at higher risk.

Structured follow-up by general practitioners after deliberate self-poisoning: a randomised controlled trial.

BACKGROUND: General Practitioners (GPs) play an important role in the follow-up of patients after deliberate self-poisoning (DSP). The aim was to examine whether structured follow-up by GPs increased the content of, adherence to, and satisfaction with treatment after discharge from emergency departments. METHODS: This was a multicentre, randomised trial with blinded assignment. Five emergency departments and general practices in the catchment area participated. 202 patients discharged from emergency departments after DSP were assigned. The intervention was structured follow-up by the GP over a 6-month period with a minimum of five consultations, accompanied by written guidelines for the GPs with suggestions for motivating patients to follow treatment, assessing personal problems and suicidal ideation, and availability in the case of suicidal crisis. Outcome measures were data retrieved from the Register for the control and payment of reimbursements to health service providers (KUHR) and by questionnaires mailed to patients and GPs. After 3 and 6 months, the frequency and content of GP contact, and adherence to GP consultations and treatment in general were registered. Satisfaction with general treatment received and with the GP was measured by the EUROPEP scale. RESULTS: Patients in the intervention group received significantly more consultations than the control group (mean 6.7 vs. 4.5 (p = 0.004)). The intervention group was significantly more satisfied with the time their GP took to listen to their personal problems (93.1% vs. 59.4% (p = 0.002)) and with the fact that the GP included them in medical decisions (87.5% vs. 54. 8% (p = 0.009)). The intervention group was significantly more satisfied with the treatment in general than the control group (79% vs. 51% (p = 0.026)). CONCLUSIONS: Guidelines and structured, enhanced follow-up by the GP after the discharge of the DSP patient increased the number of consultations and satisfaction with aftercare in general practice. Consistently with previous research, there is still a need for interventional studies. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT01342809. Registered 18 April 2011.

A bioinspired autonomous swimming robot as a tool for studying goal-directed locomotion.

The bioinspired approach has been key in combining the disciplines of robotics with neuroscience in an effective and promising fashion. Indeed, certain aspects in the field of neuroscience, such as goal-directed locomotion and behaviour selection, can be validated through robotic artefacts. In particular, swimming is a functionally important behaviour where neuromuscular structures, neural control architecture and operation can be replicated artificially following models from biology and neuroscience. In this article, we present a biomimetic system inspired by the lamprey, an early vertebrate that locomotes using anguilliform swimming. The artefact possesses extra- and proprioceptive sensory receptors, muscle-like actuation, distributed embedded control and a vision system. Experiments on optimised swimming and on goal-directed locomotion are reported, as well as the assessment of the performance of the system, which shows high energy efficiency and adaptive behaviour. While the focus is on providing a robotic platform for testing biological models, the reported system can also be of major relevance for the development of engineering system applications.

Acute poisonings treated in hospitals in Oslo: a one-year prospective study (I): pattern of poisoning.

OBJECTIVES: Prospective design is mandatory to study pattern of poisoning and suicidal intention of patients. MATERIAL AND METHODS: Prospective cross-sectional multi-center study of all patients contacting health care services because of acute poisoning during one year in Oslo, irrespective of intention. Data on the adult hospitalized patients (> or = 16 years) are presented here. RESULTS: Of a total of 3,775 such adult contacts (3,025 episodes), there were 947 (31 %) hospitalizations; annual incidence 1.9 (per 1,000) in males and 2.1 in females. Median age was 36 years (range 16-89); 54% females. Benzodiazepines (18%), ethanol (17%), paracetamol (12%), opioids (7%), and gamma hydroxybutyric acid (GHB) (7%) were most frequently taken. Patients stated suicidal intention in 29% of the admissions; physicians in 10%. CONCLUSION: Benzodiazepines and ethanol were the most common agents, but newer illicit drugs were frequent, especially GHB. Males often took ethanol and drugs of abuse; females often used prescription drugs with suicidal intention.

Clinical assessment compared to laboratory screening in acutely poisoned patients.

Acute poisonings may require identification of the toxic agents. It is impossible for routine laboratories to provide a full spectrum of toxicological analyses, and clinicians should know the reliability of the clinical diagnoses of toxic agents. We performed a 1-year study of hospitalized acute poisonings to determine the agreement between the clinical assessment on admission and serum laboratory tests for eight common toxic agents. Blood samples were drawn in 665 (70%) of the 947 admissions. The total number of laboratory found agents (967) exceeded the clinically suspected (871) by 11%. The agreement between the clinical assessment and laboratory analyses was good for ethanol and paracetamol (kappa = 0.70 for both), whereas only moderate or fair for other agents (kappa 0.22-0.51). Sensitivities of the clinical assessments compared to the laboratory results were better for common than rare agents, and better for higher than lower serum concentrations. The four most common agents (ethanol, benzodiazepines, paracetamol, and opiates) had overall sensitivity of 82% for higher-than-median serum concentrations, whereas the other agents had sensitivities ranging from 14% to 71% for higher-than-median concentrations. The reliability of the clinical diagnoses varied to such an extent that agents, which are important to recognize for specific treatment, should be tested for.

Neural and non-neural related properties in the spastic wrist flexors: An optimization study.

Quantifying neural and non-neural contributions to increased joint resistance in spasticity is essential for a better understanding of its pathophysiological mechanisms and evaluating different intervention strategies. However, direct measurement of spasticity-related manifestations, e.g., motoneuron and biophysical properties in humans, is extremely challenging. In this vein, we developed a forward neuromusculoskeletal model that accounts for dynamics of muscle spindles, motoneuron pools, muscle activation and musculotendon of wrist flexors and relies on the joint angle and resistant torque as the only input measurement variables. By modeling the stretch reflex pathway, neural and non-neural related properties of the spastic wrist flexors were estimated during the wrist extension test. Joint angle and resistant torque were collected from 17 persons with chronic stroke and healthy controls using NeuroFlexor, a motorized force measurement device during the passive wrist extension test. The model was optimized by tuning the passive and stretch reflex-related parameters to fit the measured torque in each participant. We found that persons with moderate and severe spasticity had significantly higher stiffness than controls. Among subgroups of stroke survivors, the increased neural component was mainly due to a lower muscle spindle rate at 50% of the motoneuron recruitment. The motoneuron pool threshold was highly correlated to the motoneuron pool gain in all subgroups. The model can describe the overall resistant behavior of the wrist joint during the test. Compared to controls, increased resistance was predominantly due to higher elasticity and neural components. We concluded that in combination with the NeuroFlexor measurement, the proposed neuromusculoskeletal model and optimization scheme served as suitable tools for investigating potential parameter changes along the stretch-reflex pathway in persons with spasticity.

Neuronal network models of motor generation and control.

Computer simulation of neuronal networks is rapidly becoming accepted as a powerful tool in neuroscience. We illustrate the trends in this field by looking at motor generation and control, with examples from recent modeling studies of different systems, including the spinal swimming rhythm generator of the lamprey.

Clinical vs. laboratory identification of drugs of abuse in patients admitted for acute poisoning.

OBJECTIVE: The extent of drug abuse in patients admitted for self-poisonings is uncertain. The aim of this study was to assess the pattern of drugs of abuse among patients admitted for acute poisoning according to age and gender, and to study the concordance between the clinical assessments by the physicians on duty and the drug analyses. METHODS: Prospective cross sectional study of all patients (n = 405, 52% males, median age 31 years) treated for acute poisoning in our department during one year (2001). The physician on-call classified type of drug of abuse by history and clinical assessment. This was later compared to urine and blood samples analysed for ethanol, benzodiazepines, opiates, cocaine, ecstasy, GHB, amphetamine and cannabis. RESULTS: In 320 admissions (79%), the comparison between clinical diagnosis and laboratory analyses could be performed. A total of 478 drugs were suspected and 621 were found. The main toxic agents found were benzodiazepines (49.7%), ethanol (40.3%), opiates (35.3%), cannabis (23.8%) and amphetamine (21.3%). Ninety-two had used drugs of abuse. The agreement between clinical assessments and laboratory findings was best for GHB and ethanol (kappa = 0.43), and for opiates (k = 0.38). For benzodiazepines and cannabis, the concordance was poor (k = 0.18 and 0.10, respectively). However, the correct clinical evaluation for these substances was 59% and 77%, respectively. CONCLUSIONS: Drugs of abuse were more frequently found than suspected clinically. Benzodiazepines, ethanol and opiates were most common. The agreement between clinical assessment and drug analyses was moderate to low. Physicians seem to underestimate the use of these drugs.

Hodgkin's disease survivors more fatigued than the general population.

PURPOSE: To estimate the level of fatigue and frequency of fatigue cases among Hodgkin's disease survivors (HDS) and compare them with normative data from the general population. PATIENTS AND METHODS: A cross-sectional follow-up study was done of 557 HDS (age range, 19 to 74 years) treated at the Norwegian Radium Hospital from 1971 to 1991. The sample was approached by mail, and their data were compared with normative data from 2,214 controls (age range, 19 to 74 years) representative of the general Norwegian population. Of the 557 HDS, 459 (82%) responded. The mean age (+/- SD) at the time of study was 44+/-12 years, and the mean observation time was 12+/-6 years. The Fatigue Questionnaire (11 items) measures physical and mental fatigue. Two systems of scoring were used, dichotomized (0, 0, 1, and 1) and Likert (0, 1, 2, and 3). Total fatigue (TF) constitutes the sum of all the Likert scores. Caseness was defined as a total dichotomized score of > or =4 and fatigue that lasted 6 months or longer. RESULTS: The HDS had significantly higher levels of TF than the controls (14.3 v 12.2) (P < .001). Fatigue among the HDS equaled that of the controls in poorest health. More HDS (61%) than controls (31%) reported fatigue symptoms lasting 6 months or longer (P < .001). Fatigue cases were more frequent among HDS (men, 24%; women, 27%) than among the controls (men, 9%; women, 12%) (P < .001). Disease stage/substage IB/IIB predicted fatigue caseness (P = .03). No significant associations were found between treatment characteristics and fatigue. CONCLUSION: Hodgkin's disease survivors are considerably more fatigued than the general population and report fatigue of a substantially longer duration.

Effects of beta 1- and beta 2-blockade on blood pressure and sympathetic responses to flight phobia stress.

Cardiovascular and sympathoadrenal effects of short-term oral treatment with beta 1-blockade (atenolol, 50 mg, administered two times) and beta 2-blockade (ICI 118,551, 50 mg, administered three times) were compared with placebo during actual flying in subjects with flight phobia (n = 34). beta 1-Blockade lowered resting blood pressure and heart rate and prevented a heart rate response but not a blood pressure response to this psychologic stress. beta 2-Blockade minimally lowered resting heart rate and prevented a heart rate response, but it failed to lower resting blood pressure or blood pressure response to the stress. Plasma epinephrine increased with all three treatments and more with beta 1-blockade than with placebo. Plasma norepinephrine decreased with administration of beta 2-blockade. Thus neither beta 1- nor beta 2-blockade prevents an increase in blood pressure during acute flight phobia stress. Increased plasma epinephrine seems to be the sympathetic variable that is closest related to this increase in blood pressure. Norepinephrine may be less consistently related to the blood pressure rise during flight phobia stress as shown by the decrease in plasma norepinephrine with administration of beta 2-blockade.

Attitudes toward informing the cancer patient--a survey of Norwegian physicians.

To examine Norwegian physicians' attitudes to informing patients of a diagnosis of cancer, a random sample of 1467 were surveyed. The respondents rated their level of agreement to 14 statements, and the responses were analysed by chi-squared statistics. 990 physicians responded (67%). Only 30.5% of the responding physicians had treated more than 10 cancer patients the previous year, which included 7.8% who had treated more than 50. 40.4% had treated none. The great majority (81%) preferred full information of the diagnosis. Physicians with increasing age preferred relatives not being present and gave priority to factual information and informing patients with the same diagnosis identically. Hospital physicians (39.5%) more often preferred other health professionals being present than physicians in private practice (18%) (P < 0.001). Number of cancer patients treated was not associated with attitudes toward the disclosure of information. Norwegian physicians prefer revealing the cancer diagnosis to patients, but have divergent opinions about how to do so. Some of these indicate suboptimal information-giving.

Young suicide attempters: a comparison between a clinical and an epidemiological sample.

OBJECTIVE: To compare risk factors for self-harm in 2 groups: hospitalized adolescents who had attempted suicide and adolescents reporting suicide attempts in a community survey. METHOD: All suicide attempters aged 13 to 19 years admitted to medical wards (n = 91) in a region of Norway were assessed and interviewed. Risk factors were identified by comparisons with a general population sample participating in a questionnaire study in the same community (n = 1,736). In this population sample, a separate analysis of risk factors for reporting deliberate self-harm (n = 141) was performed, applying bivariate and multivariate logistic regression models. RESULTS: Adjusted risk factors for suicide attempts in hospitalized adolescents were depression (odds ratio [OR] = 4.7), disruptive disorders (OR = 9.4), low self-worth (OR = 1.3), infrequent support from parents (OR = 3.3) or peers (OR = 3.3), parents' excessive drinking (OR = 4.3), and low socioeconomic status (OR = 2.4). For adolescents who self-reported self-harm, depression (OR = 3.1) and loneliness (OR = 1.13) were significant adjusted risk factors (p < .001). Low self-worth, low socioeconomic status, and little support from parents or peers characterized hospitalized suicidal adolescents compared with those who were not hospitalized. CONCLUSIONS: The risk factors were more powerful for hospitalized than for nonhospitalized adolescents. Prevention efforts should target the same factors for both groups, at a population level for nonhospitalized adolescents and at an individual level for hospitalized adolescents, with a focus on depression, low self-esteem, and family communication.

Suicide among children and younger and older adolescents in Norway: a comparative study.

OBJECTIVE: To compare characteristics and risk factors of suicide in early adolescence (younger than age 15 years) and in late adolescence. The authors examined whether differences in risk factors or resilience might explain the different suicide rates in the two age groups. METHOD: Information about all registered suicides of young people in Norway from 1990 through 1992 was gathered from several professional informants. Children younger than 15 years old who committed suicide (n = 14) were compared with late-adolescent suicides (15 through 19 years) (n = 115) and with controls (n = 889). RESULTS: Younger compared with older adolescent suicides more often hanged themselves (93% versus 35%). Suicidal ideation (7% versus 39%) and precipitating events were described less frequently (29% versus 49%). Older adolescents more often had psychiatric disorders (77% versus 43%). Compared with controls, the risk factors for suicide were affective disorders (young adolescents: odds ratio [OR] = 23.8, 95% confidence interval [CI] = 2.3 to 1,183; older adolescents: OR = 19.6, CI = 10.6 to 38.8); disruptive disorders (young adolescents: OR = 3.4, CI = 0.0 to 340; older adolescents: OR = 6.1, CI = 3.0 to 12.7); and not living with two biological parents (young adolescents: OR = 3.1, CI = 0.6 to 14.7; older adolescents: OR = 2.5, CI = 1.6 to 3.8). CONCLUSION: Children and young adolescents completing suicide were less exposed to known risk factors than older adolescents. The increased suicide risk was similar for both groups when they were compared with community controls. The low suicide incidence in childhood may be related to fewer risk factors, rather than to resilience to risk factors.

Use of alcohol to cope with tension, and its relation to gender, years in medical school and hazardous drinking: a study of two nation-wide Norwegian samples of medical students.

AIMS: To study the association between the use of alcohol to cope with tension and hazardous drinking, and the prevalence and the predictors of such drinking behaviours. DESIGN: Cross-sectional surveys of two nation-wide samples of medical students, one at the beginning and one at the end of medical training. PARTICIPANTS: Medical students entering Norwegian medical schools in 1993 (N = 379) or graduating in 1993 and 1994 (N = 522); 55.6% of the total sample were women. MEASUREMENTS: Postal questionnaires including SCL-5, Perceived Medical School Stress. FINDINGS: Use of alcohol to cope was reported by 10.5% of the students with no significant gender difference. Hazardous drinking ('binge drinking' at least 2-3 times per month) was reported by 14% of all the students, 24% among the men and 6% among the women. There was a strong association between use of alcohol to cope with tension and hazardous drinking, OR = 5.11, 95% CI (2.88-9.07) when controlling for other possible predictors. Use of alcohol to cope was also associated with increasing age, mental distress and lack of religious activity. The senior students used alcohol as a way of coping less often, but not hazardous drinking. Male gender, religious inactivity, high self-esteem and having no children were predictors of hazardous drinking. CONCLUSION: This study suggests that the use of alcohol to cope with tension is an independent risk factor connected with hazardous drinking among medical students, with no difference in prevalence between the genders.

Screening for breast cancer is associated with a low degree of psychological distress.

Psychological distress was measured among women on the day of recall after mammography in an official screening programme, and 4 weeks after the diagnosis was given. The Hospital Anxiety and Depression Scale (HADS) and a questionnaire to assess reactions to the examination were completed by 213 (97%) women aged 50-69 years. Among those without cancer, there were 45 (24.3%) who were anxiety cases and eight (4.3%) depression cases (HADS) on the day of recall mammography and 12 (7.3%) and three (1.8%) respectively after 4 weeks (P<0.001). Among those with cancer, the corresponding figures were seven (28.0%) and one (4.8%) before and six (24.0%) and three (12%) after screening (n.s.). A total of 194 (97%) women were definitely satisfied with the screening programme, and 195 (98%) would recommend others to participate. In conclusion, recall after mammography is associated with transiently increased levels of anxiety and depression in women without cancer. For those with cancer, the psychological distress continues. The women were almost unanimously content with participating in the screening programme.

Do women newly diagnosed with breast cancer and consulting surgeon assess decision-making equally?

This study assesses the perception of the decision-making process of 25 surgeons and 194 patients (aged 21-81 years) who had newly diagnosed breast cancer and had to undergo mastectomy or breast-conserving surgery (BCS). The majority of women wanted to participate in decision-making. When it was medically possible to give a choice between BCS and mastectomy, only 59% of women received a choice. The main reason that 11% of the women went against surgeon's recommendations was fear of cancer recurrence. The most influential factors for women were the surgeon's recommendation, and fear of cancer recurrence. Medical assessment and the cosmetic result were the most influential factors for the surgeons. Female surgeons were more influenced by their assessment of the women's need for security, than male. In conclusion, not all women who are eligible for BCS receive an option. Women and surgeons emphasize influencing factors differently. The surgeon's gender was found to influence recommendation given.

Suicidal ideation among medical students and young physicians: a nationwide and prospective study of prevalence and predictors.

BACKGROUND: Despite an increased risk of suicide among physicians we lack studies on prevalence and predictors of suicidal ideation among medical students and young doctors. METHOD: A prospective study of Norwegian medical students (n=522) re-examined after the first postgraduate year, comprising suicidal thoughts and attempts, perceived study stress, job stress, and personality. RESULTS: The previous year prevalence of suicidal thoughts was 14% at both points of time. The lifetime prevalence was 43%, while 8% had planned suicide, and 1.4% had attempted suicide. Suicidal ideation in medical school was predicted by lack of control, personality trait, single marital status, negative life events and mental distress (anxiety and depression). In the first postgraduate year, mental distress was the most important predictor, but before controlling for this variable, job stress, vulnerability (neuroticism), single status, and less working hours were independent predictors. Prospectively, suicidal thoughts and vulnerability as student predicted postgraduate suicidal ideation. CONCLUSIONS: The level of suicidal thoughts was high, but the level of attempts was low. CLINICAL IMPLICATIONS: Preventive efforts should be directed both at the students' abilities to cope with stress and at mental health services for young doctors. LIMITATIONS OF STUDY: The lower response rate at follow-up (57%) may reduce external validity.