We studied Swedish survey responders who reported regular treatment with hypnotic drugs, to find associations to perceived health problems, inpatient psychiatric diagnoses, and subsequent suicide. Among 32,679 sampled Swedes, 26,952 (83%) participated, 500 of which (2%) reported regular hypnotic drug treatment. The rate of treatment was higher in women, and increased by age in both sexes. The major findings were high odds of concurrent psychoactive drug treatments, nervous symptoms and insomnia, as well as high rates of circulatory and musculoskeletal conditions in both sexes, with indicators of disability and sleep-disturbing symptoms. During a 15-year period, 35% of the men and 21% of the women who reported regular hypnotic drug treatment had also been admitted to inpatient psychiatric care. Substance abuse was diagnosed in 20% of the men and 4.3% of the women reporting hypnotic drug treatment. In multiple logistic regression models, the highest odds for regular hypnotic drug treatment were incurred by recent/current insomnia, nervous symptoms, and other psychoactive drug treatment. We conclude that therapy was principally given according to some current peer guidelines. Yet, further research is needed into the risk/benefit ratio of sustained hypnotic drug therapy in patients with qualifying somatic and psychiatric disorders to obtain a more uniformly based consensus.

The aim of this study was to describe the life situation among,significant others' to patients with post-stroke depression, and to identify associations between the life situation of the significant others and patient characteristics. Seventy-one dyads consisting of patients with a diagnosed post-stroke depression and their significant others were included. The patients were assessed for depression with the Montgomery-angstrom sberg Depression Rating Scale and diagnosed according to the DSM-IV. The assessments of significant others included their own life situation and proxy assessments of the patients' state of depression, anger, change of personality, and need of assistance. Significant others of male stroke patients reported a more negative impact on their life situation, than did significant others of female stroke patients (p = 0.04). There was a significant association between the patient's level of depression and physical function [activities of daily living (ADL)], with those with less impaired ADL having more major depression than those with more impaired ADL (p = 0.007). This study indicates that major post-stroke depression is more common among patients with limited functional deficits. This highlights the importance of assessment for depression also among seemingly recovered stroke patients in order to treat and support those in need. This study also stresses the importance of identifying different needs of the significant others in order to provide appropriate support in their caregiving role.

This randomized study assessed if intravenous iron improves hemoglobin (Hb) response and permits decreased epoetin dose in anemic (Hb 9 - 11 g/dl), transfusion-independent patients with stainable iron in the bone marrow and lymphoproliferative malignancies not receiving chemotherapy. Patients (n = 67) were randomized to subcutaneous epoetin beta 30 000 IU once weekly for 16 weeks with or without concomitant intravenous iron supplementation. There was a significantly (P < 0.05) greater increase in mean Hb from week 8 onwards in the iron group and the percentage of patients with Hb increase >= 2 g/dl was significantly higher in the iron group (93%) than in the no-iron group (53%) (per-protocol population; P < 0.001). Higher serum ferritin and transferrin saturation in the iron group indicated that iron availability accounted for the Hb response difference. The mean weekly patient epoetin dose was significantly lower after 13 weeks of therapy (P < 0.029) and after 15 weeks approximately 10 000 IU (425%) lower in the iron group, as was the total epoetin dose (P = 0.051). In conclusion, the Hb increase and response rate were significantly greater with the addition of intravenous iron to epoetin treatment in iron-replete patients and a lower dose of epoetin was required.

Objective: To evaluate the long term prognosis of patients with stable angina pectoris.

Design: Registry based follow up ( median 9.1 years) of patients participating in the APSIS ( angina prognosis study in Stockholm), which was a double blind, single centre trial of antianginal drug treatment.

Patients: 809 patients (31% women) with stable angina pectoris < 70 ( mean (SD) 59 (7) years at inclusion) and an age and sex matched reference population from the same catchment area.

Interventions: Double blind treatment with metoprolol or verapamil during 3.4 years ( median), followed by referral for usual care with open treatment. Main outcome measures: Cardiovascular ( CV) death and non-fatal myocardial infarction (MI) in the APSIS cohort and total mortality in comparison with reference subjects.

Results: 123 patients died ( 41 MI, 36 other CV causes) and 72 had non-fatal MI. Mortality (19% v 6%, p< 0.001) and fatal MI (6.6% v 1.6%, p< 0.001) were increased among male compared with female patients. Diabetes, previous MI, hypertension, and male sex independently predicted CV mortality ( p, 0.001). Diabetes greatly increased the risk in a small subgroup of female patients. Male patients had higher mortality than men in the reference population during the first three years ( cumulative absolute difference 3.8%) but apparently not thereafter. Female patients had similar mortality to women in the reference population throughout the 9.1 years of observation.

Conclusions: Female patients with stable angina had similar mortality to matched female reference subjects but male patients had an increased risk. Diabetes, previous MI, hypertension, and male sex were strong risk factors for CV death or MI.

Objective: To evaluate the long term prognosis of patients with stable angina pectoris.

Design: Registry based follow up ( median 9.1 years) of patients participating in the APSIS ( angina prognosis study in Stockholm), which was a double blind, single centre trial of antianginal drug treatment.

Patients: 809 patients (31% women) with stable angina pectoris < 70 ( mean (SD) 59 (7) years at inclusion) and an age and sex matched reference population from the same catchment area.

Interventions: Double blind treatment with metoprolol or verapamil during 3.4 years ( median), followed by referral for usual care with open treatment. Main outcome measures: Cardiovascular ( CV) death and non-fatal myocardial infarction (MI) in the APSIS cohort and total mortality in comparison with reference subjects.

Results: 123 patients died ( 41 MI, 36 other CV causes) and 72 had non-fatal MI. Mortality (19% v 6%, p< 0.001) and fatal MI (6.6% v 1.6%, p< 0.001) were increased among male compared with female patients. Diabetes, previous MI, hypertension, and male sex independently predicted CV mortality ( p, 0.001). Diabetes greatly increased the risk in a small subgroup of female patients. Male patients had higher mortality than men in the reference population during the first three years ( cumulative absolute difference 3.8%) but apparently not thereafter. Female patients had similar mortality to women in the reference population throughout the 9.1 years of observation.

Conclusions: Female patients with stable angina had similar mortality to matched female reference subjects but male patients had an increased risk. Diabetes, previous MI, hypertension, and male sex were strong risk factors for CV death or MI.

In this article, we model electric power delivery networks as graphs, and conduct studies of two power transmission grids, i.e., the Nordic and the western states (U.S.) transmission grid. We calculate values of topological (structural) characteristics of the networks and compare their error and attack tolerance (structural vulnerability), i.e., their performance when vertices are removed, with two frequently used theoretical reference networks (the Erdos-Renyi random graph and the Barabasi-Albert scale-free network). Further, we perform a structural vulnerability analysis of a fictitious electric power network with simple structure. In this analysis, different strategies to decrease the vulnerability of the system are evaluated. Finally, we present a discussion on the practical applicability of graph modeling.

In this paper we discuss how to assess the vulnerability of electric power delivery systems with the help of standard power system performance indices. In two case studies, statistical analyses of disturbance data from Swedish power transmission and distribution are presented and discussed. The following indicators are used in the case studies: unserved energy, power loss, restoration time, and the number of disturbances with power loss. We demonstrate that disturbance size Q for larger values of indices shows a power-law distribution, i.e., P(Q>q)=1-F(q)∼q^-β (q→∞), and that the times of disturbances are distributed as a Poisson process

Background: Poststroke depression is a frequent condition and important to treat. The aim of this trial was to study the efficacy and tolerability of sertraline. Method: In 4 Swedish stroke centers, 123 patients (aged 70.7 +/- 9.9 years) were enrolled during the period September 1998 to January 2001 in a randomized, double-blind, placebo-controlled 26-week trial, at a mean of 128 +/- 97 days (range, 3-375 days) after stroke, if they fulfilled DSM-IV criteria of major depressive episode (N = 76) or minor depressive disorder (N = 47). The primary efficacy variable was a changed in depression assessed by the Montgomery-Asberg Depression Rating Scale. The Emotional Distress Scale (EDS) was administered and the occurrence of emotionalism and quality of life (QoL) were assessed, as well as neurologic recovery. Efficacy analyses were intention-to-treat, short-term (week 6) and long-term (week 26). Results: Of the 123 patients, 62 were treated with sertraline (50-100 mg/day) and 61 with placebo. Both groups improved substantially, with no differences between the treatments, either for major depressive episode or minor depressive disorder, or for short- or long-term antidepressant effect and neurologic outcome. EDS revealed a better outcome with sertraline at week 6 (p <.05). At week 26, the improvement in QoL was better in sertraline patients (p <.05) and there was a trend for emotionalism (p =.07). No serious side effects were seen. Conclusion: Poststroke depression as measured by a conventional depression rating scale improved over time irrespective of treatment. Positive effects specific to sertraline were identified in emotional distress, emotionalism, and QoL. The study indicates that poststroke emotional reactions comprise depression and other domains susceptible to pharmacologic therapy.

Background: There is indirect evidence that an increased infectious burden is associated with a decreased prevalence of IgE-mediated allergy during childhood. Objective: To determine whether there is a relation between the serostatus of 13 different viruses and parentally reported infections and IgE sensitization in 2-year-old children. To investigate whether there is an interaction between cytomegalovirus (CMV) and Epstein-Barr virus (EBV) in relation to IgE sensitization. Methods: A total of 246 infants were followed prospectively to 2 years of age with clinical examinations, skin prick test, and specific IgE analyses and through analysis of seropositivity against adenovirus, influenza, parainfluenza, respiratory syncytial virus, CMV, EBV, herpes simplex virus, human herpesvirus 6, and varicella-zoster virus. Results: There was some evidence that IgE sensitization (24%) tended to be more common among children who were seropositive against few compared with children who were seropositive against many viruses, but this was not statistically significant, and there was no consistent trend across the groups. IgE sensitization was statistically significantly less prevalent at 2 years of age among infants who were seropositive against EBV but not other viruses (adjusted odds ratio, 0.34; 95% CI, 0.14-0.86). The interaction of seropositivity against both CMV and EBV antibodies indicated a further reduction in the risk for IgE sensitization (adjusted odds ratio for interaction, 0.10; 95% CI, 0.01-0.92), indicating effect modification associated with seropositivity against CMV. Conclusion: Our results indicate that acquisition of EBV infection during the first 2 years of life is associated with a reduced risk of IgE sensitization, and this effect is enhanced by CMV coinfection.

In our daily life we are surrounded by different kind of risks and we constantly strive for better methods to quantify and in the prolongation manage these risks. Every activity involves some risks and there are some kinds of risks and some level of risks that we are unwilling to accept. We all like to live a life that is free from risks, but that is impossible.

The word risk has a lot of different interpretations. In this thesis we shall let risk stand for the combination of random or uncertain events with negative consequences for human health, life and welfare and for the environment together with some measures of the likelihood of such events. We believe this is the prevailing concept or understanding of risk; as the probability of an event followed by some negative consequences or activities of that event.

In risk analysis one tries to recognize the nature of various risks and to assess the magnitude of the risks. In the risk analysis it is very important to know what system to consider and this is not self evident in many cases. The situation is clearly different for planning and/or building a system compared with running the same system in a real time state. The system that is going to be the subject to the risk analysis must be clearly defined and the limitations and the boundaries of the system must be set. It is very important to ensure that all persons involved in a risk analysis have a common understanding of the system being considered, including relevant operations.

During the past decades many studies have been carried out on risk related topics and the society has showed a significant interest in the field of risk analysis. Risk analysis is the interdisciplinary field of science that combines results and knowledge of probability theory, mathematical statistics, engineering, medicine, philosophy, psychology, economics and other applied disciplines.

In this thesis we will give some examples of different risk analyses carried out basically within two areas. The first part of the thesis (paper 1- paper V) describes different risk analyses carried out in the area of transportation. This is an area with large differences between the different modes of transportation in respect to, for example number of users, number of accidents, magnitude of the accidents and accessible data. The latter part of the thesis (paper VI and paper VII) describes two risk analyses carried out in the field of medicine. Medicine is a science, which has used methods from the area of risk analysis for a long time. The different papers will be used to discuss risk analysis as a tool in decision-making.

Risk, risk analysis and decision-making are essential aspects of health care andmedicine, for patients as well as for physicians and for society as a whole, andthe concept of risk and risk analysis in decision-making has a long history. Theword risk has many different interpretations and has no commonly accepteddefinition. In this thesis, we shall let risk stand for the combination of randomor uncertain events with negative consequences for human health, life and/orwelfare and/or the environment together with some measures of the likelihoodof such events. We believe this is the dominant concept and understanding ofrisk, the risk being the likelihood or probability of an event followed by somenegative consequences or activities of that event.

In this doctoral thesis, we focus on biostatistics, risks and risk analysis in thefield of medicine, a science which has been using methods from the area of riskanalysis for a long time. The seven papers (paper I - paper VII) presented inthis thesis, together with a general introduction to risk, risk analysis anddecision-making, will be used to illustrate and discuss risk analysis as a tool fordecision-making in the field of medicine. From my point a view, risk analysisin the field of medicine aims to reduce pain, raise the quality of life, reduce therisk of adverse events, compare cost efficiency between different treatmentregimes and prolong a healthy life. Based on results presented in the thesis, weconclude that biostatistics, risks and risk analysis used in the field of medicineare valuable methods for evaluation of hypotheses within the health care areaand a good basis for decision-making.

Objective

To examine the perception of risk of smoking-related psychological and social outcomes, and the effect of pregnancy and intention to stop smoking on the perceived risk.

Methods

Eighty women were asked to make judgments about the probability of outcomes for smoking-related consequences. Four subgroups were created using the variables of pregnancy (pregnant versus not pregnant) and cessation of smoking (intention to stop versus no intention to stop). Judgments were based on the decision to stop and not stop smoking.

Result

Intention to stop smoking affected the estimated probabilities for the occurrence of consequences for both continuing and stopping smoking, whereas pregnancy did not affect the estimated probabilities. The estimated effect of stopping smoking was statistically significant.

Conclusion

Health messages about smoking for all population groups should consider both future risk of mortality and immediate quality-of-life effects of smoking.

The purpose of the study was to test a model based on the product of value and belief, called expected utility (EU), on the addictive behaviour of smoking. A total of 40 pregnant and 40 non-pregnant women over a period of 2 weeks performed judgements on values and beliefs about consequences related to smoking for the conditions of continuing and stopping smoking. There were no differences between pregnant and non-pregnant women in the EU of smoking. Differences in expected utility between the conditions of continuing and stopping smoking were larger for health consequences compared with psychological and social consequences and consequences related to pregnancy. Expected utility gives a good description of judgements over time. Values as well as beliefs related to health consequences should be stressed in smoking cessation programmes, especially among pregnant women.

Objective To study long-term treatment effects of metoprolol or verapamil on combined cardiovascular end points and psychological variables in patients with stable angina pectoris. Design Randomized, double-blind, double-dummy trial. Patients The study included 809 patients under 70 years of age with stable angina pectoris. The mean age of the patients was 59 +/- 7 years and 31% were women. Exclusion criteria were myocardial infarction within the previous 3 years and contraindications to beta-blockers and calcium antagonists. The patients were followed between 6 and 75 months (median 3.4 years and a total of 2887 patient years). Intervention The patients were treated with either metoprolol (Seloken ZOC 200 mg o.d.) or verapamil (Isoptin Retard 240 b.i.d.). Acetylsalicylic acid, ACE inhibitors, lipid lowering drugs and long acting nitrates were allowed in the study. End points Death, non-fatal cardiovascular events including acute myocardial infarction, incapacitating or unstable angina, cerebrovascular or peripheral vascular events. Psychological variables reflecting quality of life i.e. psychosomatic symptoms, sleep disturbances and an evaluation of overall life satisfaction. Results Combined cardiovascular events did not differ and occurred in 30.8% and 29.3% of metoprolol and verapamil treated patients respectively. Total mortality in metoprolol and verapamil treated patients was 5.4 and 6.2%, respectively. Cardiovascular mortality was 4.7% in both groups. Non-fatal cardiovascular events occurred in 26.1 and 24.3% of metoprolol and verapamil-treated patients, respectively. Psychosomatic symptoms and sleep disturbances were significantly improved in both treatment groups. The magnitudes of change were small and did not differ between treatments. Life satisfaction did not change on either drug. Withdrawals due to side effects occurred in 11.1 and 14.6%, respectively. Conclusion This long term study indicates that both drugs are well tolerated and that no difference was shown on the effect on mortality, cardiovascular end points and measures of quality of life.

The purpose of this article was to investigate how pregnant and non-pregnant women use various coping techniques when attempting to refrain from smoking. Eighty women with subgroups formed by the variables of pregnant/not pregnant and quitting/not quitting smoking were studied over a 2-week period. The general strategy was to follow smokers who had stated an intention to quit smoking. Smokers, pregnant and non-pregnant, who did not intend to quit were also followed with respect to the coping techniques they would hypothetically be using if they were trying to quit. Pregnant women used coping strategies more often than non-pregnant women. Differences found between pregnant and non-pregnant women were evenly distributed for behavioral and cognitive methods. The goal of becoming a non-smoker, especially during pregnancy, needs to be addressed to include psychological and physical factors. In these efforts, the framework introduced by the study involving a time-related approach could be useful.

Time-related problems interfere with treatment decisions and evaluations in health care, both from the perspective of the doctor and from the perspective of the patient. The compliance level of the patient and subsequent evaluations in clinical practice might be affected by a discrepancy in the time perspective. Context factors related to time and health perspectives are relevant to clinical decisions and quality management. A summary of evaluation factors in quality management is presented: (a) the time perspective of the patient is different from the time perspective of the doctor, both in an objective and in a subjective manner; (b) disease chronicity chronic affects the perception of time; (c) assessments need to extend over a period sufficiently long for variations in a disease activity to be noticed; (d) there is variation both in time for an outcome to occur and in time span for that outcome; (e) the number of patients benefiting from certain drugs and variability over time is valuable information; (f) the outcome of a specified treatment could be estimated for different periods in a sequence; and (g) changes occur in judgments and decisions over time both for the doctor and for the patient.

Rationale, aims and objectives Diagnostic reasoning and treatment decisions are a key competence of doctors. A model based on values and probability provides a conceptual framework for clinical judgments and decisions, and also facilitates the integration of clinical and biomedical knowledge into a diagnostic decision. Method Both value and probability are usually estimated values in clinical decision making. Therefore, model assumptions and parameter estimates should be continually assessed against data, and models should be revised accordingly. Introducing parameter estimates for both value and probability, which usually pertain in clinical work, gives the model labelled subjective expected utility. Estimated values and probabilities are involved sequentially for every step in the decision-making process. Results Introducing decision-analytic modelling gives a more complete picture of variables that influence the decisions carried out by the doctor and the patient. Conclusion A model revised for perceived values and probabilities by both the doctor and the patient could be used as a tool for engaging in a mutual and shared decision-making process in clinical work.

This study aimed to investigate factors predicting lapse among pregnant and non-pregnant women when trying to stop smoking. A total of 40 women, pregnant and non-pregnant, were investigated over a 2-week period when trying to stop smoking. One-quarter of the women lapsed every day. Not being pregnant was a significant predictor for the occurrence of any lapse during the time period, whereas age, number of years of smoking, number of earlier attempts to stop smoking, and number of cigarettes smoked per day did not predict lapse. There was a four times higher risk for lapse in non-pregnant compared with pregnant women. Being pregnant gives an opportunity to help stop smoking with a considerably lower risk of lapse compared with non-pregnant women.

There is an increased awareness about patients' involvement in the clinical decision process where uncertainty is an unavoidable condition. The impact of psychological factors like risk aversion, risk aversion and time, asymmetry in risk aversion, and risk and control on shared decision-making is discussed. In addition to differences in risk estimates, doctors and patients may exhibit a difference in perception of time perspectives, and losses versus gains. A summary of valuation factors in shared decision-making is presented: (a) the doctors tend to follow expected value combinations more closely, while the patient is more risk aversive; (b) unwillingness to take risks increases for rare outcomes; (c) there is an increased tendency to take risks with delayed outcomes of the decisions; (d) the doctor is generally well informed about risk and time aspects for different diseases, whereas this might not always be the case with the patient; (e) rheumatoid arthritis and diabetes mellitus are chronic diseases, and both create a vulnerability to a variety of complications over time; (f) rheumatoid arthritis demands different combinations of treatments sequentially over time, whereas diabetes mellitus is treated with insulin; (g) many diseases, like rheumatoid arthritis and diabetes mellitus, are not completely affected by control, as the disease may constantly progress.

The objective of this study was to test a framework based upon the value and the probability of outcomes related to smoking. Over a 2-week period, 80 women were asked to perform judgments of value and probability of the outcome for smoking-related consequences. Subgroups were formed by the two variables of pregnancy and intent to quit smoking. Judgments were performed given the conditions of quitting and not quitting smoking. The intent to quit or not quit smoking had an impact on expected utility of smoking. Moreover, there was a difference between expected utility given the conditions of not quitting smoking and quitting smoking. A framework based upon values and beliefs appears to be useful in describing the addictive behavior of smoking and can be applied in developing smoking-cessation education.

 Objectives: To study perceived smoking-related consequences of continuing and of quitting smoking. Methods: Eighty women, with subgroups formed by pregnant/nonpregnant women and trying/not trying to quit smoking, performed judgments of the probability for consequences to occur given the conditions of continuing or quitting smoking. Results: For both the pregnant and nonpregnant women, the probability that consequences will occur was rated as less likely given the condition of quitting smoking. The condition of quitting had its greatest effect on the probability that somatic consequences would occur. Conclusion: Consequences of smoking for somatic health should be stressed in health promotion, especially to pregnant women.

The objective was to examine the perception of smoking-related health consequences and its relationship to pregnancy and intent to quit. Over a two-week period, pregnant and non-pregnant women, intending and not intending to quit smoking, rated the probability for smoking-related health consequences to occur, given continuing to smoke and quitting smoking. Pregnant women who did not intend to quit smoking exhibited the lowest estimated probability for the smoking-related health consequences to occur if they continued smoking. For all women, there was a statistically significant estimated effect of quitting smoking. Renewed attention needs to be given to perceptions of health risks of smoking, especially among pregnant women.

Objective. To examine the association of pregnancy and intent to quit smoking with perception of the somatic health risk of smoking and the effect on risk of quitting smoking. Methods. Pregnant and non-pregnant women, intending and not intending to quit smoking, (n = 80) over a two-week period rated the probability for smoking-related health consequences to occur, given both conditions of not quitting smoking and of quitting smoking. Results. Groups were the determining variable accounting for differences in risk perception. For pregnant women who did not intend to quit smoking, the estimated probability for the consequences to occur was generally low given the condition of continuing to smoke. There was no effect for time. The estimated effect of quitting smoking was statistically significant. Conclusions. Future campaigns need to integrate risk information so that it can be grounded cognitively in order to increase the personal responsibility that women take for their own health and the health of the fetus.

Purpose: To examine the association between pregnancy and judgments about a range of somatic, psychological and social events and conditions related to smoking. Basic Procedures: Pregnant and non-pregnant women smokers, intending and not intending to quit, were contacted in family practices in Bulgaria where they were under ordinary medical observation. All 80 women in the study were asked to rate consequences related to smoking in terms of how good or bad they were. Results: Being pregnant or not pregnant had the main effect on rated values of smoking-related consequences, with non-pregnant women giving the least positive ratings across the time-span of the study and whether or not they intended to quit. No statistically significant difference was obtained between the quitting and non-quitting women, and there was only one statistically significant difference for different days. Conclusions: The physical variable of pregnancy has a larger impact on judgments about smoking compared to the motivational aspect of intending to quit.

We examined the number of lapses among pregnant and non-pregnant women when trying to quit smoking, number of coping techniques used, and the relationship between any lapse and usage of coping techniques. Forty women were followed over a two-week period. On day 14, the women rated how often eleven different coping techniques were used. One-fourth of the women lapsed every day. If non-pregnant, the odds ratio was eight times higher compared to being pregnant for any lapse during the period. Being pregnant gives a higher success rate in attempts to quit smoking.

Objectives: To study time estimates by women smokers for when smokingrelated consequences will occur given continuing or quitting smoking. The relationship of these estimates to pregnancy and intent to quit smoking was also investigated. Methods: Over a two-week period, eighty women, selected to constitute four subgroups formed by pregnant vs. non-pregnant and trying vs. not trying to quit smoking, rated times at which they would expect smoking-related consequences to occur given continuing or quitting smoking. Results: Somatic health consequences were estimated to occur later than consequences related to mood and social relations. All consequences were estimated to occur later given quitting smoking. Pregnancy had an effect on the estimated time that consequences would occur, with pregnant women estimating earlier occurrence of consequences related to mood and social relations than non-pregnant women did. Conclusion: Health messages should stress consequences for somatic health in quitting smoking, since outcomes later in time might have too low a value to exert a positive effect on decisions to quit smoking.