Missing data are ubiquitous in clinical epidemiological research. Individuals with missing data may differ from those with no missing data in terms of the outcome of interest and prognosis in general. Missing data are often categorized into the following three types: missing completely at random (MCAR), missing at random (MAR), and missing not at random (MNAR). In clinical epidemiological research, missing data are seldom MCAR. Missing data can constitute considerable challenges in the analyses and interpretation of results and can potentially weaken the validity of results and conclusions. A number of methods have been developed for dealing with missing data. These include complete-case analyses, missing indicator method, single value imputation, and sensitivity analyses incorporating worst-case and best-case scenarios. If applied under the MCAR assumption, some of these methods can provide unbiased but often less precise estimates. Multiple imputation is an alternative method to deal with missing data, which accounts for the uncertainty associated with missing data. Multiple imputation is implemented in most statistical software under the MAR assumption and provides unbiased and valid estimates of associations based on information from the available data. The method affects not only the coefficient estimates for variables with missing data but also the estimates for other variables with no missing data.

No abstract

Background: Among patients with chronic immune thrombocytopenia (cITP), little is known regarding risk factors for cardiovascular and bleeding outcomes and how these events influence mortality. Objectives:We examined the rate of cardiovascular events and bleeding requiring a hospital contact according to platelet count levels, as well as the prognostic impact of these events on all-cause mortality in adult patients with cITP. Methods: We identified all cITP patients registered in the Nordic Country PatientRegistry for Romiplostim during 1996 to 2015. Absolute risks and hazard ratios across platelet count levels based on Cox regression analysis were computed, adjusting for age, sex, prevalent/incident cITP, smoking, and comorbidities. We also compared all-cause mortality rates in cITP patients with and without cardiovascular and bleeding events.Results: Among 3 584 cITP patients, 1-year risks were 1.9% for arterial cardiovascular events, 1.2% for venous thromboembolism, and 7.5% for bleeding. Rates of cardiovascular events were similar across platelet counts. Patients with platelet counts <50 × 10 9 /L had >2-fold higher rates of bleeding than patients with normal platelet counts. These associations were unchanged in time-varying analyses that considered changes in platelet counts during follow-up. Occurrences of cardiovascular and bleeding events were associated with 4-fold to 5-fold increases in 1-year mortality. Conclusions:Among patients with cITP, the 1-year risks of cardiovascular events were 1% to 2%, while nearly 8% experienced a bleeding event within 1 year.Cardiovascular events occurred across all platelet levels, while low platelet counts were associated with increased hazards of bleeding. Cardiovascular and bleeding events were strong prognostic factors for mortality. K E Y W O R D Sbleeding, chronic ITP, epidemiology, mortality, stroke, myocardial infarction, venous thromboembolism | 913 ADELBORG Et AL.

No abstract

Background: Postoperative infection is a common complication in hip fracture patients and the risk appears to have increased during the last decade. However, the impact of infection on mortality after hip fracture surgery remains unclear. Purpose: We aimed to examine the association between infection (any, as well as specific infections), with all-cause mortality following hip fracture surgery. Methods: Using Danish nationwide registries, we conducted a population-based cohort study on 74,771 hip fracture patients ≥ 65 years old operated from 2005-2016. We included hospital-treated infection as a time-varying exposure, and calculated 30-days mortality rate per 1000 person-years (PY). We used time-varying Cox Proportional Hazard Regression to compute 30-days adjusted hazards ratios (aHRs) with 95 % confidence interval (CI) comparing the mortality of hip fracture patients with and without infections. We adjusted for sex, age, comorbidities, medication use, and marital status. Results: Within 30 days of surgery, 9,592 (12.8%) patients developed a hospital-treated infection. Among these, 30-days mortality was 8.43 per 1000 PY compared with 3.34 among patients without infection (aHR=2.72, 95 % CI: 2.56-2.88). For patients who developed pneumonia, aHR was 4.18 (95 % CI: 3.91-4.48), whereas the aHR was 8.86 (95 % CI: 7.88-9.95) for patients who developed systemic sepsis. For patients who sustained reoperation due to infection, aHR was 2.95 (95 %CI: 1.88-4.64). The mortality was higher in infected vs. noninfected patients irrespective of patients' age, sex and comorbidity. Conclusion: Infection within 30 days of hip fracture surgery is associated with substantially increased mortality risk. Further research should improve our knowledge about patients at increased risk and prevention measures for specific infections.