METHODS: A gender-matched case-control study was conducted in the largest public sector cardiac hospital of Pakistan, and the data of 460 subjects were collected. The dataset comprised of eight nonclinical features. Four supervised ML algorithms were used to train and test the models to predict the CVDs status by considering traditional logistic regression (LR) as the baseline model. The models were validated through the train-test split (70:30) and tenfold cross-validation approaches.
RESULTS: Random forest (RF), a nonlinear ML algorithm, performed better than other ML algorithms and LR. The area under the curve (AUC) of RF was 0.851 and 0.853 in the train-test split and tenfold cross-validation approach, respectively. The nonclinical features yielded an admissible accuracy (minimum 71%) through the LR and ML models, exhibiting its predictive capability in risk estimation.
CONCLUSION: The satisfactory performance of nonclinical features reveals that these features and flexible computational methodologies can reinforce the existing risk prediction models for better healthcare services.
OBJECTIVE: To develop an adherence prediction model for CKD patients.
METHODS: This multi-centre, cross-sectional study was conducted in 10 tertiary hospitals in Malaysia using simple random sampling of CKD patients with ≥1 medication (sample size = 1012). A questionnaire-based collection of patient characteristics, adherence (defined as ≥80% consumption of each medication for the past one month), and knowledge of each medication (dose, frequency, indication, and administration) was performed. Continuous data were converted to categorical data, based on the median values, and then stratified and analysed. An adherence prediction model was developed through multiple logistic regression in the development group (n = 677) and validated on the remaining one-third of the sample (n = 335). Beta-coefficient values were then used to determine adherence scores (ranging from 0 to 7) based on the predictors identified, with lower scores indicating poorer medication adherence.
RESULTS: Most of the 1012 patients had poor medication adherence (n = 715, 70.6%) and half had good medication knowledge (n = 506, 50%). Multiple logistic regression analysis determined 4 significant predictors of adherence: ≤7 medications (constructed score = 2, p
METHODOLOGY/PRINCIPAL FINDINGS: Available evidence was evaluated using a step-wise process that included systematic literature reviews, policymaker and stakeholder interviews, a study to assess dengue contingency planning and outbreak management in 10 countries, and a retrospective logistic regression analysis to identify alarm signals for an outbreak warning system using datasets from five dengue endemic countries. Best practices for managing a dengue outbreak are provided for key elements of a dengue contingency plan including timely contingency planning, the importance of a detailed, context-specific dengue contingency plan that clearly distinguishes between routine and outbreak interventions, surveillance systems for outbreak preparedness, outbreak definitions, alert algorithms, managerial capacity, vector control capacity, and clinical management of large caseloads. Additionally, a computer-assisted early warning system, which enables countries to identify and respond to context-specific variables that predict forthcoming dengue outbreaks, has been developed.
CONCLUSIONS/SIGNIFICANCE: Most countries do not have comprehensive, detailed contingency plans for dengue outbreaks. Countries tend to rely on intensified vector control as their outbreak response, with minimal focus on integrated management of clinical care, epidemiological, laboratory and vector surveillance, and risk communication. The Technical Handbook for Surveillance, Dengue Outbreak Prediction/ Detection and Outbreak Response seeks to provide countries with evidence-based best practices to justify the declaration of an outbreak and the mobilization of the resources required to implement an effective dengue contingency plan.
SUBJECTS: Patients who were admitted to the University of Malaya Medical Centre due to cardiac events.
METHODS: Eight different machine learning models were evaluated. The models included 3 different sets of features: full features; significant features from multiple logistic regression; and features selected from recursive feature extraction technique. The performance of the prediction models with each set of features was compared.
RESULTS: The AdaBoost model with the top 20 features obtained the highest performance score of 92.4% (area under the curve; AUC) compared with other prediction models.
CONCLUSION: The findings showed the potential of using machine learning models to predict return to work after cardiac rehabilitation.
METHODS: Eleven case-control studies within the International Pancreatic Cancer Case-control Consortium took part in the present study, including in total 2838 case and 4748 control women. Pooled estimates of odds ratios (ORs) and their 95% confidence intervals (CIs) were calculated using a 2-step logistic regression model and adjusting for relevant covariates.
RESULTS: An inverse OR was observed in women who reported having had hysterectomy (ORyesvs.no, 0.78; 95% CI, 0.67-0.91), remaining significant in postmenopausal women and never-smoking women, adjusted for potential PC confounders. A mutually adjusted model with the joint effect for hormone replacement therapy (HRT) and hysterectomy showed significant inverse associations with PC in women who reported having had hysterectomy with HRT use (OR, 0.64; 95% CI, 0.48-0.84).
CONCLUSIONS: Our large pooled analysis suggests that women who have had a hysterectomy may have reduced risk of PC. However, we cannot rule out that the reduced risk could be due to factors or indications for having had a hysterectomy. Further investigation of risk according to HRT use and reason for hysterectomy may be necessary.
OBJECTIVE: To compare the fall history between older adults with and without a previous stroke and to identify the determinants of falls and fear of falling in older stroke survivors.
DESIGN: Case-control observational study.
SETTING: Primary teaching hospital.
PARTICIPANTS: Seventy-five patients with stroke (mean age ± standard deviation, 66 ± 7 years) and 50 age-matched control participants with no previous stroke were tested.
METHODS: Fall history, fear of falling, and physical, cognitive, and psychological function were assessed. A χ2 test was performed to compare characteristics between groups, and logistic regression was performed to determine the risk factors for falls and fear of falling.
MAIN OUTCOME MEASURES: Fall events in the past 12 months, Fall Efficacy Scale-International, Berg Balance Scale, Functional Ambulation Category, Fatigue Severity Scale, Montreal Cognitive Assessment, and Patient Healthy Questionnaire-9 were measured for all participants. Fugl-Meyer Motor Assessment was used to quantify severity of stroke motor impairments.
RESULTS: Twenty-three patients and 13 control participants reported at least one fall in the past 12 months (P = .58). Nine participants with stroke had recurrent falls (≥2 falls) compared with none of the control participants (P < .01). Participants with stroke reported greater concern for falling than did nonstroke control participants (P < .01). Female gender was associated with falls in the nonstroke group, whereas falls in the stroke group were not significantly associated with any measured outcomes. Fear of falling in the stroke group was associated with functional ambulation level and balance. Functional ambulation level alone explained 22% of variance in fear of falling in the stroke group.
CONCLUSIONS: Compared with persons without a stroke, patients with stroke were significantly more likely to experience recurrent falls and fear of falling. Falls in patients with stroke were not explained by any of the outcome measures used, whereas fear of falling was predicted by functional ambulation level. This study has identified potentially modifiable risk factors with which to devise future prevention strategies for falls in patients with stroke.
LEVEL OF EVIDENCE: III.
Please provide feedback to Administrator (firstname.lastname@example.org)