Using natural language processing for automated classification of disease and to identify misclassified ICD codes in cardiac disease.

Aims: ICD codes are used for classification of hospitalizations. The codes are used for administrative, financial, and research purposes. It is known, however, that errors occur. Natural language processing (NLP) offers promising solutions for optimizing the process. To investigate methods for automatic classification of disease in unstructured medical records using NLP and to compare these to conventional ICD coding. Methods and results: Two datasets were used: the open-source Medical Information Mart for Intensive Care (MIMIC)-III dataset (n = 55.177) and a dataset from a hospital in Belgium (n = 12.706). Automated searches using NLP algorithms were performed for the diagnoses 'atrial fibrillation (AF)' and 'heart failure (HF)'. Four methods were used: rule-based search, logistic regression, term frequency-inverse document frequency (TF-IDF), Extreme Gradient Boosting (XGBoost), and Bio-Bidirectional Encoder Representations from Transformers (BioBERT). All algorithms were developed on the MIMIC-III dataset. The best performing algorithm was then deployed on the Belgian dataset. After preprocessing a total of 1438 reports was retained in the Belgian dataset. XGBoost on TF-IDF matrix resulted in an accuracy of 0.94 and 0.92 for AF and HF, respectively. There were 211 mismatches between algorithm and ICD codes. One hundred and three were due to a difference in data availability or differing definitions. In the remaining 108 mismatches, 70% were due to incorrect labelling by the algorithm and 30% were due to erroneous ICD coding (2% of total hospitalizations). Conclusion: A newly developed NLP algorithm attained a high accuracy for classifying disease in medical records. XGBoost outperformed the deep learning technique BioBERT. NLP algorithms could be used to identify ICD-coding errors and optimize and support the ICD-coding process.

Validity and reliability of the Chinese version of digital health readiness questionnaire among hypertension patients in rural areas of China.

Introduction: Digital health has the potential to support health care in rural areas by overcoming the problems of distance and poor infrastructure, however, rural areas have extremely low use of digital health because of the lack of interaction with technology. There is no existing tool to measure digital health literacy in rural China. This study aims to test and validate the digital health readiness questionnaire for assessing digital readiness among patients in rural China. Methods: Due to the different Internet environments in China compared to Belgium, a cultural adaptation is needed to optimize the use of Digital Health Readiness Questionnaire in China. Then, a prospective single-center survey study was conducted in rural China among patients with hypertension. Confirmatory factor analysis was computed to test the measurement models. Results: A total of 330 full questionnaires were selected and included in the analysis. The model-fit measures were used to assess the model's overall goodness of fit (Chi-square/degrees of freedom = 5.060, comparative fit index = 0.889, Tucker-Lewis index (TLI) = 0.869, root mean square error of approximation (RMSEA) = 0.111, standardized root mean square residual (SRMR) = 0.0880). TLI is a little bit lower than the borderline (more than 0.9) and RMSEA is higher than it (less than 0.08 means good model fit). We deleted two items 2 and 4 and the result shows a better goodness of fit (Chi-square/degrees of freedom = 4.897, comparative fit index = 0.914, TLI = 0.895, RMSEA = 0.109, SRMR = 0.0765). Conclusion: To increase applicability and generalizability in rural areas, it should be considered to use the calculation of only the parts Digital skills, Digital literacy and Digital health literacy which are equally applicable in a Belgian population as in a rural Chinese population.

Motivational communication skills to improve motivation and adherence in cardiovascular disease prevention: A narrative review.

Lifestyle optimization is one of the most essential components of cardiovascular disease prevention. Motivational counseling provided by health care professionals could promote lifestyle modification. The purpose of the review is to identify possible evidence-based psychological principles that may be applicable to motivational counseling in the prevention of cardiovascular disease. These motivational communication skills promote behavioral change, improved motivation and adherence to cardiovascular disease prevention. A personal collection of the relevant publications. The review identified and summarized the previous evidence of implementation intentions, mental contrasting, placebo effect and nocebo effects and identity-based regulations in behavior change interventions and proposed their potential application in cardiovascular disease prevention. However, it is challenging to provide real support in sustainable CVD-risk reduction and encourage patients to implement lifestyle changes, while avoiding being unnecessarily judgmental, disrespectful of autonomy, or engaging patients in burdensome efforts that have little or no effect on the long run. Motivational communication skills have a great potential for effectuating sustainable lifestyle changes that reduce CVD-related risks, but it is also surrounded by ethical issues that should be appropriately addressed in practice. It is key to realize that motivational communication is nothing like an algorithm that is likely to bring about sustainable lifestyle change, but a battery of interventions that requires specific expertise and long term joint efforts of patients and their team of caregivers.

Smartphone-based gamification intervention to increase physical activity participation among patients with coronary heart disease: A randomized controlled trial.

INTRODUCTION: Despite proven benefits, patients with coronary heart disease (CHD) typically fail to participate in sufficient physical activity (PA). Effective interventions should be implemented to help patients maintain a healthy lifestyle and modify their present behavior. Gamification is the use of game design features (such as points, leaderboards, and progress bars) to improve motivation and engagement. It shows the potential for encouraging patients to engage in PA. However, empirical evidence on the efficacy of such interventions among patients with CHD is still emerging. PURPOSE: The aim of the study is to explore whether a smartphone-based gamification intervention could increase PA participation and other physical and psychological outcomes in CHD patients. METHODS: Participants with CHD were randomly assigned to three groups (control group, individual group, and team group). The individual and team groups received gamified behavior intervention based on behavioral economics. The team group combined gamified intervention with social interaction. The intervention lasted for 12 weeks, and the follow-up was12 weeks. The primary outcomes included the change in daily steps and the proportion of patient days that step goals were achieved. The secondary outcomes included competence, autonomy, relatedness, and autonomous motivation. RESULTS: For the individual group, smartphone-based gamification intervention significantly increased PA among CHD patients over the 12-week period (step count difference 988; 95% CI 259-1717; p < 0.01) and had a good maintenance effect during the follow-up period (step count difference 819; 95% CI 24-1613; p < 0.01). There are also significant differences in competence, autonomous motivation, body mass index (BMI), and waist circumference in 12 weeks between the control group and individual group. For the team group, gamification intervention with collaboration didn't result in significant increases in PA. But patients in this group had a significant increase in competence, relatedness, and autonomous motivation. CONCLUSION: A smartphone-based gamification intervention was proven to be an effective way to increase motivation and PA engagement, with a substantial maintenance impact (Chinese Clinical Trial Registry Identifier: ChiCTR2100044879).

An Electrocardiographic Characterization of Left Bundle Branch Area Pacing-Induced Right Ventricular Activation Delay: A Comparison With Native Right Bundle Branch Block.

Background: Left bundle branch area pacing (LBBAP) induces delayed RV activation and is thought to be harmless, since the electrocardiographic signature is reminiscent to native RBBB. However, to what extent the delayed RV activation during LBBAP truly resembles that of native RBBB remains unexplored. Methods: This study included patients with incomplete RBBB (iRBBB), complete RBBB (cRBBB) and patients who underwent LBBAP. Global and right ventricular activation times were estimated by QRS duration and R wave peak time in lead V1 (V1RWPT) respectively. Delayed RV activation was further characterized by duration, amplitude and area of the terminal R wave in V1. Results: In patients with LBBAP (n = 86), QRS duration [120 ms (116, 132)] was longer compared to iRBBB patients (n = 422): 104 ms (98, 110), p < 0.001, but shorter compared to cRBBB (n = 223): 138 ms (130, 152), p < 0.001. V1RWPT during LBBAP [84 ms (72, 92)] was longer compared to iRBBB [74 ms (68, 80), p < 0.001], but shorter than cRBBB [96 ms (86, 108), p < 0.001]. LBBAP resulted in V1 R' durations [42 ms (28, 55)] comparable to iRBBB [42 ms (35, 49), p = 0.49] but shorter than in cRBBB [81 ms (68, 91), p < 0.001]. During LBBAP, the amplitude and area of the V1 R' wave were more comparable with iRBBB than cRBBB. V1RWPT during LBBAP was determined by baseline conduction disease, but not by LBBAP capture type. Conclusion: LBBAP-induced delayed RV activation electrocardiographically most closely mirrors the delayed RV activation as seen with incomplete rather than complete RBBB.

Willingness to participate in cardiac telerehabilitation: results from semi-structured interviews.

Aims: Cardiac rehabilitation (CR) is indicated in patients with cardiovascular disease but participation rates remain low. Telerehabilitation (TR) is often proposed as a solution. While many trials have investigated TR, few have studied participation rates in conventional CR non-participants. The aim of this study was to identify the percentage of patients that would be willing to participate in a TR programme to identify the main perceived barriers and facilitators for participating in TR. Methods and results: Two groups of patients were recruited: CR non-participants and CR participants. Semi-structured interviews were conducted. Thirty non-participants and 30 participants were interviewed. Of CR non-participants, 33% would participate in TR and 10% would participate in a blended CR programme (combination of centre-based CR and TR). Of CR participants, 60% would participate in TR and 70% would be interested in a blended CR programme. Of those that would participate in TR, 44% would prefer centre-based CR, 33% would prefer a blended CR programme, and 11% would prefer a full TR programme. In both groups, the main facilitating aspect about TR was not needing transport and the main barrier was digital literacy. Conclusion: For CR non-participants, TR will only partly solve the problem of low participation rates and blended programmes might not offer a solution. Cardiac rehabilitation participants are more prepared to participate in TR and blended CR. Digital literacy was in both groups mentioned as an important barrier, emphasizing the challenges for healthcare and local governments to keep educating all types of patients in digital literacy.