BackgroundPatient-reported outcome (PRO) measures may be used at a group level for research and quality improvement and at the individual patient level to support clinical decision making and ensure efficient use of resources. The challenges involved in implementing PRO measures are mostly the same regardless of aims and diagnostic groups and include logistic feasibility, high response rates, robustness, and ability to adapt to the needs of patient groups and settings. If generic PRO systems can adapt to specific needs, advanced technology can be shared between medical specialties and for different aims.ObjectiveWe describe methodological, organizational, and practical experiences with a generic PRO system, WestChronic, which is in use among a range of diagnostic groups and for a range of purposes.MethodsThe WestChronic system supports PRO data collection, with integration of Web and paper PRO questionnaires (mixed-mode) and automated procedures that enable adherence to implementation-specific schedules for the collection of PRO. For analysis, we divided functionalities into four elements: basic PRO data collection and logistics, PRO-based clinical decision support, PRO-based automated decision algorithms, and other forms of communication. While the first element is ubiquitous, the others are optional and only applicable at a patient level. Methodological and organizational experiences were described according to each element.ResultsWestChronic has, to date, been implemented in 22 PRO projects within 18 diagnostic groups, including cardiology, neurology, rheumatology, nephrology, orthopedic surgery, gynecology, oncology, and psychiatry. The aims of the individual projects included epidemiological research, quality improvement, hospital evaluation, clinical decision support, efficient use of outpatient clinic resources, and screening for side effects and comorbidity. In total 30,174 patients have been included, and 59,232 PRO assessments have been collected using 92 different PRO questionnaires. Response rates of up to 93% were achieved for first-round questionnaires and up to 99% during follow-up. For 6 diagnostic groups, PRO data were displayed graphically to the clinician to facilitate flagging of important symptoms and decision support, and in 5 diagnostic groups PRO data were used for automatic algorithm-based decisions.ConclusionsWestChronic has allowed the implementation of all proposed protocol for data collection and processing. The system has achieved high response rates, and longitudinal attrition is limited. The relevance of the questions, the mixed-mode principle, and automated procedures has contributed to the high response rates. Furthermore, development and implementation of a number of approaches and methods for clinical use of PRO has been possible without challenging the generic property. Generic multipurpose PRO systems may enable sharing of automated and efficient logistics, optimal response rates, and other advanced options for PRO data collection and processing, while still allowing adaptati...

ObjectiveMissing data is a ubiquitous problem in studies using patient-reported measures, decreasing sample sizes and causing possible bias. In longitudinal studies, special problems relate to attrition and death during follow-up. We describe a methodological approach for the use of multiple imputation (MI) to meet these challenges.MethodsIn a cohort of patients treated with percutaneous coronary intervention followed with use of repetitive questionnaires and information from national registers over 3 years, only 417 out of 1,726 patients had complete data on all measure points and covariates. We suggest strategies for use of MI and different methods for dealing with death along with sensitivity analysis of deviations from the assumption of missing at random, all with the use of standard statistical software. The Mental Component Summary from Short Form 12-item survey was used as an example.ConclusionIgnoring missing data may cause bias of unknown size and direction in longitudinal studies. We have illustrated that MI is a feasible method to try to deal with bias due to missing data in longitudinal studies, including attrition and nonresponse, and should be considered in combination with analysis of sensitivity in longitudinal studies. How to handle dropout due to death is still open for debate.

Different measures revealed some differences in proportions of RTW. However, high agreement between RTW-definitions was found. Choice of RTW-definitions should depend on study purpose; simple cross-sectional methods are sufficient in prediction of RTW and analysis of risk factors, while methods capturing relapses are recommended when sustainability, prognosis and vulnerability are in focus.

Objective This review aimed to examine systematically the epidemiological evidence linking work-related exposure to violence and threats thereof with risk of mental disorders and mental ill-health symptoms. Methods We searched PubMed, EMBASE, PsycINFO and Web of Science to identify original studies that provide quantitative risk estimates. The evidence was weighted according to completeness of reporting, potential common method bias, and bias due to differential selection and drop out, selective reporting, and misclassification of exposure and outcome. Results We identified 14 cross-sectional and 10 cohort studies with eligible risk estimates, of which 4 examined depressive disorder and reported an elevated risk among the exposed [pooled relative risk (RR) 1.42, 95% confidence interval (CI) 1.31–1.54, I 2 =0%]. The occurrence of depressive and anxiety symptoms, burnout and psychological distress was examined in 17 studies (pooled RR 2.33, 95% CI 3.17, I 2 =42%), and 3 studies examined risk of sleep disturbance (pooled RR 1.22, 95% CI 1.09–1.37, I 2 =0%). In most studies, common method bias and confounding could not be ruled out with confidence and strong heterogeneity in most outcome definitions invalidate the strict interpretation of most pooled risk estimates. Conclusion The reviewed studies consistently indicate associations between workplace violence and mental health problems. However, due to methodological limitations the causal associations (if any) may be stronger or weaker than the ones reported in this study. Prospective studies with independent and validated reporting of exposure and outcome and repeated follow-up with relevant intervals are highly warranted.