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.
DESIGN: This post hoc descriptive exploration of data from the large international very early rehabilitation trial (A Very Early Rehabilitation Trial (AVERT)) examined the four common post acute rehabilitation pathways (inpatient rehabilitation, home with community rehabilitation, inpatient rehabilitation then community rehabilitation and home with no rehabilitation) experienced by participants in the 3 months post stroke and describes their 12-month outcomes.
SETTING: Hospital stroke units in AUS, UK and SE Asia.
PARTICIPANTS: Patients who had an acute stroke recruited within 24 hours who were ≤65 years.
RESULTS: 668 participants were ≤65 years; 99% lived independently, and 88% no disability (modified Rankin Score (mRS)=0) prior to stroke. We had complete data for 12-month outcomes for n=631 (94%). The proportion receiving inpatient rehabilitation was higher in AUS than other regions (AUS 52%; UK 25%; SE Asia 23%), whereas the UK had higher community rehabilitation (UK 65%; AUS 61%; SE Asia 39%). At 12 months, 70% had no or little disability (mRS 0-2), 44% were depressed, 28% rated quality of life as poor or worse than death. For those working prior to stroke (n=228), only 57% had returned to work. A noteworthy number of working age survivors received no rehabilitation services within 3 months post stroke.
CONCLUSIONS: There was considerable variation in rehabilitation pathways and post acute service use across the three regions. At 12 months, there were high rates of depression, poor quality of life and low rates of return to work.
TRIAL REGISTRATION NUMBER: Australian New Zealand Clinical Trials Registry (ACTRN12606000185561).
METHOD: A retrospective analyses of 370 medical reports written for patients who sustained traumatic brain injury from motor vehicle accidents was conducted. To establish the employment pattern, the pre-injury employment history was compared to the latest employment status documented. Types and severity of concomitant injuries were rated according to Abbreviated Injury Scale criteria. All significant variables were further analyzed using logistic regression to explore predictors of employment.
RESULTS: Up to 87% of the patients sustained concomitant injuries, with more than two-thirds (72%) scoring ≤ 2 on the Abbreviated Injury Scale. One hundred and eighty-two patients (49.2%) successfully returned to work. Among those who returned to work, 34% returned to former employment with pre-injury job description. Severity of traumatic brain injury, length of acute hospital stay, ambulation status and cognitive status were found to be significant predictive factors for employment status post traumatic brain injury. Presence of concomitant extremity injuries was found to influence the employment pattern among traumatic brain injury survivors.
CONCLUSION: The return to work rate was somewhat low and was not influenced by presence of concomitant injuries. .
MATERIALS AND METHODS: Six focus groups were conducted using a semi-structured interview guide on 40 informants (employed multiethnic survivors). Survivors were stratified into three groups for successfully RTW, and another three groups of survivors who were unable to return to work. Each of the three groups was ethnically homogeneous. Thematic analysis using a constant comparative approach was aided by in vivo software.
RESULTS: Participants shared numerous barriers and facilitators which directly or interactively affect RTW. Key barriers were physical-psychological after-effects of treatment, fear of potential environment hazards, high physical job demand, intrusive negative thoughts and overprotective family. Key facilitators were social support, employer support, and regard for financial independence. Across ethnic groups, the main facilitators were financial-independence (for Chinese), and socialisation opportunity (for Malay). A key barrier was after-effects of treatment, expressed across all ethnic groups.
CONCLUSIONS: Numerous barriers were identified in the non-RTW survivors. Health professionals and especially occupational therapists should be consulted to assist the increasing survivors by providing occupational rehabilitation to enhance RTW amongst employed survivors. Future research to identify prognostic factors can guide clinical efforts to restore cancer survivors to their desired level/type of occupational functioning for productivity and wellbeing.
METHODS: A cross-sectional study design was used in this research. The data was collected via interview using a validated questionnaire. Logistic regression models were developed to discover the significant determinants of RTW and of wage loss among BC survivors.
RESULTS: A total of 256 BC survivors were included in this study. The analysis showed that there was a 21% loss of or reduction in mean income within 1 year after diagnosis. The significant predictors of RTW are being a government employee, having reduced wages or wage loss, and if the case had been diagnosed 1 year or more ago. Being a private sector employee and having a late stage of cancer was a barrier to RTW. The main risk factors for reduced wages or wage loss were belonging to the age group of 40-59 years, being of Chinese or Indian ethnicity, having low educational status, and not returning to work. However, belonging to the higher monthly income group (earning > RM 2000) is a protective factor against the risk of reduced wages or wage loss.
CONCLUSIONS: Non-RTW and wage loss after diagnosis of BC may result in the survivors experiencing a significant financial burden. Assessment of these patients is becoming more crucial because more women participate in the workforce in Malaysia nowadays and because BC is managed using multiple treatment modalities with their consequences could lead to long absences from work.
OBJECTIVE: The aim was to understand the local prevalence and factors associated with returning to work in Malaysia after a cardiac event.
METHODS: A cross sectional design was used. All patients attending the cardiac rehabilitation program after major cardiac event during an 11-months period (2011-2012) were included. Data relating to socio-demographic, work-related, risk factors and acute myocardial infarction were collected. The SF-36 questionnaire was used to assess quality of life. Regression analysis was used to determine the predicting factors to return to work.
RESULTS: A total of 398 files were screened, 112 respondents agreed to participate giving a response rate of 47.3%. The prevalence of returned to work (RTW) was 66.1% [95% CI: 57.2-75.0]. Factors associated with work resumption were age (Adj. OR: 0.92 (95% CI: 0.84-0.99), diabetes mellitus (Adj. OR: 3.70, 95% CI: 1.35-10.12), Mental Component Summary (MCS) score (Adj. OR: 1.05 (95% CI: 1.01-1.09) and baseline angiography findings. Patients with single vessel and two vessel disease were 8.9 times and 3.78 times more likely to return to work compared to those with 3 vessels (Adj. OR: 8.90 (95% CI: 2.29-34.64) and Adj. OR: 3.78, (95% CI: 1.12, 12.74).
CONCLUSIONS: We proposed a cardiac rehabilitation program to emphasize mental health as it may improve successful return to work after cardiac event.