Bmc Med Res Methodol
-
Bmc Med Res Methodol · Feb 2017
Performance of Firth-and logF-type penalized methods in risk prediction for small or sparse binary data.
When developing risk models for binary data with small or sparse data sets, the standard maximum likelihood estimation (MLE) based logistic regression faces several problems including biased or infinite estimate of the regression coefficient and frequent convergence failure of the likelihood due to separation. The problem of separation occurs commonly even if sample size is large but there is sufficient number of strong predictors. In the presence of separation, even if one develops the model, it produces overfitted model with poor predictive performance. Firth-and logF-type penalized regression methods are popular alternative to MLE, particularly for solving separation-problem. Despite the attractive advantages, their use in risk prediction is very limited. This paper evaluated these methods in risk prediction in comparison with MLE and other commonly used penalized methods such as ridge. ⋯ The logF-type penalized method, particularly logF(1,1) could be used in practice when developing risk model for small or sparse data sets.
-
Bmc Med Res Methodol · Jan 2017
ReviewMeasuring health-related quality of life in cervical cancer patients: a systematic review of the most used questionnaires and their validity.
Data on health-related quality of life (HRQoL) is paramount for shared and evidence based decision-making. Since an overview of cervical cancer HRQoL tools and their validity appears to be lacking, we performed a systematic review on usage of disease specific HRQoL instruments in cervical cancer patients and their psychometric properties to identify the most suitable cervical cancer specific HRQoL tool. ⋯ The validity of the often used EORTC QLQ-CX24 questionnaire for cervical cancer patients remains uncertain as 5 out of 9 psychometric properties were doubtful or not reported in current literature. Cervical cancer specific HRQoL tools should therefore always be used in conjunction with validated generic cancer HRQoL tools until proper validity has been proven, or a more valid tool has been developed.
-
Bmc Med Res Methodol · Jan 2017
Multicenter StudyMultiple triangulation and collaborative research using qualitative methods to explore decision making in pre-hospital emergency care.
Paramedics make important and increasingly complex decisions at scene about patient care. Patient safety implications of influences on decision making in the pre-hospital setting were previously under-researched. Cutting edge perspectives advocate exploring the whole system rather than individual influences on patient safety. Ethnography (the study of people and cultures) has been acknowledged as a suitable method for identifying health care issues as they occur within the natural context. In this paper we compare multiple methods used in a multi-site, qualitative study that aimed to identify system influences on decision making. ⋯ Combining multiple qualitative methods with a collaborative research approach can facilitate exploration of system influences on patient safety in under-researched settings. The paper highlights empirical issues, strengths and limitations for this approach. Feedback workshops were effective for verifying findings and prioritising areas for future intervention and research.
-
Bmc Med Res Methodol · Nov 2016
ReviewRecommendations for the analysis of individually randomised controlled trials with clustering in one arm - a case of continuous outcomes.
In an individually randomised controlled trial where the treatment is delivered by a health professional it seems likely that the effectiveness of the treatment, independent of any treatment effect, could depend on the skill, training or even enthusiasm of the health professional delivering it. This may then lead to a potential clustering of the outcomes for patients treated by the same health professional, but similar clustering may not occur in the control arm. Using four case studies, we aim to provide practical guidance and recommendations for the analysis of trials with some element of clustering in one arm. ⋯ A partially clustered approach, modelling the clustering in just one arm, most accurately represents the trial design and provides valid results. Modelling homogeneous variances between the clustered and unclustered arm is adequate in scenarios similar to the case studies considered. We recommend treating each participant in the unclustered arm as a single cluster. This approach is simple to implement in R and Stata and is recommended for the analysis of trials with clustering in one arm only. However, the case studies considered had small ICC values, limiting the generalisability of these results.
-
Bmc Med Res Methodol · Nov 2016
Observational StudyPredictors of in-hospital mortality following major lower extremity amputations in type 2 diabetic patients using artificial neural networks.
Outcome prediction is important in the clinical decision-making process. Artificial neural networks (ANN) have been used to predict the risk of post-operative events, including survival, and are increasingly being used in complex medical decision making. We aimed to use ANN analysis to estimate predictive factors of in-hospital mortality (IHM) in patients with type 2 diabetes (T2DM) after major lower extremity amputation (LEA) in Spain. ⋯ Elixhauser Comorbidity Index is a superior comorbidity risk-adjustment model for major LEA survival prediction in patients with T2DM than Charlson Comorbidity Index model using ANN models. Female sex, congestive heart failure, and renal failure are strong predictors of mortality in these patients.