Search

ObjectiveTo examine the effect of sex on risk of bloodstream infections (BSI) and BSI mortality and to assess to what extent known risk factors for BSI mediate this association in the general population.ParticipantsThe prospective, population-based HUNT2 Survey (1995-97) in Norway invited 93,898 inhabitants ≥20 years in the Nord-Trøndelag region, whereof 65,237 (69.5%) participated. 46.8% of the participants were men.ExposuresSex and potential mediators between sex and BSI; health behaviours (smoking, alcohol consumption), education attainment, cardiovascular risk factors (systolic blood pressure, non-HDL cholesterol, body mass index) and previous or current comorbidities.Main outcome measuresSex differences in risk of first-time BSI, BSI mortality (death within first 30 days after a BSI), BSI caused by the most frequent bacteria, and the impact of known BSI risk factors as mediators.ResultsWe documented a first-time BSI for 1,840 (2.9%) participants (51.3% men) during a median follow-up of 14.8 years. Of these, 396 (0.6%) died (56.6% men). Men had 41% higher risk of any first-time BSI (95% confidence interval (CI), 28 to 54%) than women. An estimated 34% of the excess risk of BSI in men was mediated by known BSI risk factors. The hazard ratio (HR) with 95% CI for BSI due to S. aureus was 2.09 (1.28 to 2.54), S. pneumoniae 1.36 (1.05 to 1.76), and E. coli 0.97 (0.84 to 1.13) in men vs women. BSI related mortality was higher in men compared to women with HR 1.87 (1.53 to 2.28).ConclusionsThis large population-based study show that men have higher risk of BSI than women. One-third of this effect was mediated by known risk factors for BSI. This raises important questions regarding sex specific approaches to reduce the burden of BSI.

Objective We examined whether anxiety and depression symptoms constitute increased risk of bloodstream infection (BSI), as a proxy for sepsis. Methods A general population with self-reported anxiety and depression symptoms was followed prospectively for hospital-verified BSI. Using multivariable Cox regression analysis, we estimated hazard ratios (HR) with 95% confidence intervals (CI) of BSI and BSI mortality, with and without statistical adjustment for comorbidities, BMI, and life-style factors that may confound or mediate the associations. Results During 14.8 years median follow-up of 59,301 individuals, 1578 (2.7%) experienced BSI and 328 (0.55%) participants died within 30 days after a BSI. Severe depression symptoms were associated with a 38% increased risk of BSI, adjusted for age, sex, and education (HR = 1.38, 95% CI = 1.10-1.73). The HR was attenuated to 1.23 (0.96-1.59) after adjustment for comorbidities and to 1.15 (0.86-1.53) after additional adjustment for BMI and life-style factors. For severe anxiety symptoms, the corresponding HRs were 1.48 (1.20-1.83), 1.35 (1.07-1.70), and 1.28 (0.99-1.64). Moderate symptoms of depression and anxiety were not associated with increased BSI risk. The analysis of BSI mortality yielded imprecise results but suggested an increased risk of BSI mortality in participants with moderate depression symptoms. Conclusions Severe depression and anxiety symptoms were associated with a moderately increased risk of BSI. The association may, at least in part, be confounded or mediated by comorbidities, BMI, and life-style. Future research should investigate whether interventions targeting improved BMI and life-style may reduce the risk of BSI and sepsis in people with depression and anxiety symptoms.

Background We aimed to evaluate the clinical usefulness of qSOFA as a risk stratification tool for patients admitted with infection compared to traditional SIRS criteria or our triage system; the Rapid Emergency Triage and Treatment System (RETTS). Methods The study was an observational cohort study performed at one Emergency Department (ED) in an urban university teaching hospital in Norway, with approximately 20,000 visits per year. All patients >16 years presenting with symptoms or clinical signs suggesting an infection (n = 1535) were prospectively included in the study from January 1 to December 31, 2012. At arrival in the ED, vital signs were recorded and all patients were triaged according to RETTS vital signs, presenting infection, and sepsis symptoms. These admission data were also used to calculate qSOFA and SIRS. Treatment outcome was later retrieved from the patients’ electronic records (EPR) and mortality data from the Norwegian population registry. Results Of the 1535 admitted patients, 108 (7.0%) fulfilled the Sepsis2 criteria for severe sepsis. The qSOFA score ≥2 identified only 33 (sensitivity 0.32, specificity 0.98) of the patients with severe sepsis, whilst the RETTS-alert ≥ orange identified 92 patients (sensitivity 0.85, specificity 0.55). Twenty-six patients died within 7 days of admission; four (15.4%) of them had a qSOFA ≥2, and 16 (61.5%) had RETTS ≥ orange alert. Of the 68 patients that died within 30 days, only eight (11.9%) scored ≥2 on the qSOFA, and 45 (66.1%) had a RETTS ≥ orange alert. Discussion In order to achieve timely treatment for sepsis, a sensitive screening tool is more important than a specific one. Our study is the fourth study were qSOFA finds few of the sepsis cases in prehospital or at arrival to the ED. We add information on the RETTS triage system, the two highest acuity levels together had a high sensitivity (85%) for identifying sepsis at arrival to the ED - and thus, RETTS should not be replaced by qSOFA as a screening and trigger tool for sepsis at arrival. Conclusion In this observational cohort study, qSOFA failed to identify two thirds of the patients admitted to an ED with severe sepsis. Further, qSOFA failed to be a risk stratification tool as the sensitivity to predict 7-day and 30-day mortality was low. The sensitivity was poorer than the other warning scores already in use at the study site, RETTS-triage and the SIRS criteria. Electronic supplementary material The online version of this article (doi:10.1186/s13049-017-0399-4) contains supplementary material, which is available to authorized users.

Background: Bloodstream infections (BSI) cause considerable morbidity and mortality, and primary prevention should be a priority. Lifestyle factors are of particular interest since they represent a modifiable target. Methods: We conducted a prospective cohort study among participants in the population-based Norwegian HUNT2 Survey, where 64 027 participants were followed from 1995–97 through 2011 by linkage to prospectively recorded information on BSI at local and regional hospitals. The exposures were: baseline body mass index (BMI) measurements; and self-reported smoking habits, leisure time physical activity and alcohol intake. The outcomes were hazard ratios (HR) of BSI and BSI mortality. Results: During 810 453 person-years and median follow-up of 14.8 years, 1844 (2.9%) participants experienced at least one BSI and 396 (0.62%) died from BSI. Compared with normal weight participants (BMI 18.5–24.9 kg/m2), the age- and sex-adjusted risk of a first-time BSI was 31% [95% confidence interval (CI) 14–51%] higher at BMI 30.0–34.9 kg/m2, 87% (95% CI 50–135%) higher at BMI 35.0–39.9 kg/m2 and 210% (95% CI 117–341%) higher at BMI ≥ 40.0 kg/m2. The risk of BSI mortality was similarly increased. Compared with never-smokers, current smokers had 51% (95% CI 34–70%) and 75% (95% CI 34–129%) higher risks of BSI and BSI mortality, respectively. Physically inactive participants had 71% (95% CI 42–107%) and 108% (95% CI 37–216%) higher risks of BSI and BSI mortality, respectively, compared with the most physically active. © The Author 2017; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association. This is the authors' accepted and refereed manuscript to the article. Locked until 15 June 2018 due to copyright restrictions

Background: Invasive pneumococcal disease (IPD) is responsible for significant mortality and morbidity worldwide. There are however few longitudinal studies on the changes in case fatality rate of IPD in recent years. We carried out a prospective observational study of patients with IPD in Nord Trøndelag county in Norway from 1993 to 2011 to study the clinical variables and disease outcome. The main outcome was all-cause mortality after 30 and 90 days. Methods: Patients with positive blood cultures were registered prospectively by the microbiology laboratory and clinical variables were registered retrospectively from patients’ hospital records. The severity of sepsis was assigned according to the 2001 International Sepsis Definition Conference criteria. The association between mortality and predictive factors was studied using a logistic regression model. Results: The total number of patients was 414 with mean age of 67 years and 53 % were male. Comorbidity was assessed by the Charlson Comorbidity Index (CCI). A CCI-score of 0 was registered in 144 patients (34.8 %), whereas 190 had a score of 1–2 (45.9 %) and 80 (19.3 %) had a score ≥3. 68.8 % of the patients received appropriate antibiotics within the first 6 h. The 30-day mortality risk increased by age and was 3-fold higher for patients aged ≥80 years (24.9, 95 % CI 16.4–33.4 %) compared to patients aged