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.
BACKGROUND: Despite efforts to promote utilisation of cardiac rehabilitation (CR), participation among patients remains unsatisfactory. Little is known of patient decision to participate Phase II CR in a multi-ethnic country.
DESIGN: A cross-sectional study design.
METHODS: A consecutive sampling of 240 patients with coronary heart disease completed Coronary Artery Disease Education Questionnaire (CADE-Q) II, Hospital Anxiety and Depression Scale (HADS), Multidimensional Scale of Perceived Social Support (MSPSS) and Cardiac Rehabilitation Barriers Scale (CRBS).
RESULTS: Seventy per cent of patients (mean age 60.5 [SD = 10.6] years, 80.8% male) participated in phase II cardiac rehabilitation. Self-driving to cardiac rehabilitation centres, higher barriers in perceived need/health care and logistical factors were significantly associated with decreased odds of participation. Patients with more barriers from comorbidities/functional status, higher perceived social support from friends, and anxiety were more likely to participate. Chinese and Indians were less likely to participate when compared with Malays. More than 80% of patients used both home and mobile broadband internet, and 72.9% of them would accept the usage of technologies, especially educational videos, instant messenger, and video calls to partially replace the face-to-face, centre-based cardiac rehabilitation approach.
CONCLUSION: Several barriers were associated with non-participation in phase II cardiac rehabilitation. With the high perceived acceptance of technology usage in cardiac rehabilitation, home-based and hybrid cardiac rehabilitation may represent potential solutions to improve participation.
RELEVANCE TO CLINICAL PRACTICE: By addressing the barriers to cardiac rehabilitation, patients are more likely to be ready to adopt health behaviour changes and adhere to the cardiac rehabilitation programme. The high perceived acceptance of using technologies in cardiac rehabilitation may provide insights into new delivery models that can improve and overcome barriers to participation.
PURPOSE: The purposes of this study were to translate, adapt, and evaluate the Malay-language version of the Cardiac Rehabilitation Barriers Scale (CRBS) and to measure the psychometric properties of the Malay-version CRBS to justify its use in Sarawak.
METHODS: A forward and back-translation method was used. Content validity was assessed by three experts. Psychometric testing was conducted on a sample of 283 patients who were eligible to participate in cardiac rehabilitation. A construct validity test was performed using factor analysis. Cronbach's alpha was used to examine the internal consistency. The test-retest reliability was calculated using the intraclass correlation coefficient on 22 participants. Independent-samples t test and analysis of variance were conducted to assess the criterion validity. Mean scores for total barriers of the scale and each individual factor were compared among the different patient characteristics.
RESULTS: The Malay-version CRBS showed an item level of content validity index of 1.00 for all of the items after improvements were made based on the experts' suggestions. The factor analysis, using principal component analysis with direct oblimin rotation, extracted four factors that differed from the original study. These four factors explained 52.50% of the cumulative percentage of variance. The Cronbach's alphas ranged from .74 to .81 for the obtained factors. Test-retest reliability was established using the intraclass correlation coefficient value of .78. Criterion validity was supported using the significant differences in the mean score for total barriers among educational level, driving distance, travel time to the hospital, and cardiac rehabilitation attendance.
CONCLUSIONS/IMPLICATIONS FOR PRACTICE: This study found the Malay-version CRBS to be a valid and reliable instrument. It may be used with inpatients to identify barriers to participation in cardiac rehabilitation to promote rehabilitation attendance and improve patient care.
METHODS: We recruited 112 patients who were newly diagnosed with ACS and treated at the referral hospital, Sarawak General Hospital, Malaysia. In the intervention group (modified CRP), all medication was reviewed by the clinical pharmacists, focusing on drug indication; understanding of secondary prevention therapy and adherence to treatment strategy. We compared the "pre-post" quality of life (QoL) of three groups (intervention, conventional and control) at baseline, 6 months and 12 months post-discharge with Malaysian norms. QoL data was obtained using a validated version of Short-Form 36 Questionnaire (SF-36). Analysis of variance (ANOVA) with repeated measure tests was used to compare the mean differences of scores over time.
RESULTS: A pre-post quasi-experimental non-equivalent group comparison design was applied to 112 patients who were followed up for one year. At baseline, the physical and mental health summaries reported poor outcomes in all three groups. However, these improved gradually but significantly over time. After the 6-month follow-up, the physical component summary reported in the modified CRP (MCRP) participants was higher, with a mean difference of 8.02 (p = 0.015) but worse in the mental component summary, with a mean difference of -4.13. At the 12-month follow-up, the MCRP participants performed better in their physical component (PCS) than those in the CCRP and control groups, with a mean difference of 11.46 (p = 0.008), 10.96 (p = 0.002) and 6.41 (p = 0.006) respectively. Comparing the changes over time for minimal important differences (MICD), the MCRP group showed better social functioning than the CCRP and control groups with mean differences of 20.53 (p = 0.03), 14.47 and 8.8, respectively. In role emotional subscales all three groups showed significant improvement in MCID with mean differences of 30.96 (p = 0.048), 31.58 (p = 0.022) and 37.04 (p
METHODS: Data collected from patients with ACS between 2008 and 2009 for a study on cardiac rehabilitation at the Sarawak General Hospital were used for this study. QOL data were obtained using a validated version of the EuroQol five-dimensional questionnaire at baseline and at 12 months. Health utility scores were calculated using visual analogue scale scores and utility tariffs from Malaysia and the United Kingdom.
RESULTS: Data from 104 subjects from the earlier study was used. The mean age was 56.1 years, with 88.5% being men. The mean hospitalization duration was 6.3 days. The mean utility score was 0.75 at baseline and 0.82 at 12 months. There was a statistically significant improvement in utility from baseline to 12 months based on the Malaysian tariff (P = 0.014) but not with the UK tariff (P = 0.086). The QOL of patients was associated with sex and diagnosis of ST-segment elevation myocardial infarction.
CONCLUSIONS: Our results showed that there was a significant improvement in the QOL from baseline to 12 months. Only sex and diagnosis affected the QOL score at baseline because of limited variables available for testing. It also reconfirms the importance of applying the appropriate, country-specific utility tariffs in QOL studies. Despite limitations, the study is useful toward describing QOL among a group of Malaysian patients with ACS.
Methods: At the baseline of this cross-sectional study, 156 patients from Western Iran participated during April-November 2016. People 20 years-80 years able to recall the emotional content of dreams after cardiac surgery entered the study. The Beck depression inventory (BDI), Beck anxiety inventory (BAI), Buss and Perry's aggression questionnaire (BPAQ) and Schredl's dream emotions manual were used for collecting data. A binary logistic regression analysis used for the study of the relationship between risk factors and emotionally negative dreams.
Results: The mean age of participants was 59 (SD = 9) years (men: 64.1%). The results showed that 25% of patients have negative emotional content. After adjustment for demographic variables, the results showed that increased anxiety [adjusted odds ratio (adj OR) = 1.08 [1.01-1.16], P = 0.020] and anger (adj OR = 1.03 [1.00-1.06], P = 0.024) and hypertension (adj OR = 2.71 [1.10-6.68], P = 0.030) can predict the dreams with negative content significantly.
Conclusion: The increasing rates of anxiety and anger and history of hypertension are related to increasing dreams with the negative emotional load. The control of risk factors of dreams with negative emotional load can be the target of future interventions.