Development of a patients' satisfaction analysis system using machine learning and lexicon-based methods.

BACKGROUND: Patients' rights are integral to medical ethics. This study aimed to perform sentiment analysis and opinion mining on patients' messages by a combination of lexicon-based and machine learning methods to identify positive or negative comments and to determine the different ward and staff names mentioned in patients' messages. METHODS: The level of satisfaction and observance of the rights of 250 service recipients of the hospital was evaluated through the related checklists by the evaluator. In total, 822 Persian messages, composed of 540 negative and 282 positive comments, were collected and labeled by the evaluator. Pre-processing was performed on the messages and followed by 2 feature vectors which were extracted from the messages, including the term frequency-inverse document frequency (TFIDF) vector and a combination of the multifeature (MF) (a lexicon-based method) and TFIDF (MF + TFIDF) vectors. Six feature selectors and 5 classifiers were used in this study. For the evaluations, 5-fold cross-validation with different metrics including area under the receiver operating characteristic curve (AUC), accuracy (ACC), F1 score, sensitivity (SEN), specificity (SPE) and Precision-Recall Curves (PRC) were reported. Message tag detection, which featured different hospital wards and identified staff names mentioned in the study patients' messages, was implemented by the lexicon-based method. RESULTS: The best classifier was Multinomial Naïve Bayes in combination with MF + TFIDF feature vector and SelectFromModel (SFM) feature selection (ACC = 0.89 ± 0.03, AUC = 0.87 ± 0.03, F1 = 0.92 ± 0.03, SEN = 0.93 ± 0.04, and SPE = 0.82 ± 0.02, PRC-AUC = 0.97). Two methods of assessment by the evaluator and artificial intelligence as well as survey systems were compared. CONCLUSION: Our results demonstrated that the lexicon-based method, in combination with machine learning classifiers, could extract sentiments in patients' comments and classify them into positive and negative categories. We also developed an online survey system to analyze patients' satisfaction in different wards and to remove conventional assessments by the evaluator.

A prospective survey of atrial fibrillation management in Iran: Baseline results of the Iranian Registry of Atrial Fibrillation (IRAF).

BACKGROUND: Atrial fibrillation. (AF) is the most common sustained arrhythmia globally and its prevalence is likely to increase in the next decades as a result of increasing age and co-morbidities. There are no data on demographic features, clinical characteristics, associated comorbidities, and practice patterns of AF in Iran. METHODS: The Iranian Registry of Atrial Fibrillation (IRAF) is a hospital-based prospective survey of AF patients with a 12-month follow-up. Data were collected on a standardized case report form and entered into a web-based electronic database. This paper reports the baseline characteristics of the IRAF cohort. RESULTS: Between February 2018 and March 2020, a total of 1300 patients (57% Male, mean age, 60 ± 14 years) were enrolled. Palpitations were the most common presenting symptom (66%). The most common cardiac comorbidities were hypertension (52%), heart failure (23.7%), and valvular heart disease (21.8%). AF mainly presented as a paroxysmal pattern (44.6%). Seventy-eight percent of the patients with non-valvular AF had CHA2 DS2 -VASc score ≥1 and most (97%) were at low risk for bleeding (HAS-BLED score <3). Rhythm control was given to 55.1% of the patients. Anticoagulation for stroke prevention was provided to 69.5% of the eligible patients, while aspirin was used in 35%. CONCLUSION: The IRAF Registry has provided a systematic collection of contemporary data regarding the management and treatment of AF in Iran. Oral anticoagulant was used in 69.5%, but aspirin use was still common.

Applying a Computer-based Warfarin Management System at a Large Tertiary Cardiovascular Center in Iran.

BACKGROUND: Regarding adjustments to warfarin dosage, numerous studies have shown that computerized methods are superior to those based on personal experience. OBJECTIVES: To report the efficacy of a computer-based warfarin management system (WMS) in the Iranian population. METHODS: By utilizing the existing dosing algorithms and obtaining expert opinions, we developed a computer-based WMS at a large tertiary cardiovascular center. The time in therapeutic range (TTR) and the number of international normalized ratio (INR) tests of clinic patients were compared before and after the implementation of WMS. RESULTS: Overall, 803 patients with 5407 INR tests were included in the before phase and 679 patients with 4189 INR tests in the after phase. The mean TTR was 57.3% before and 59% after WMS implementation (mean difference, 1.64, 95% CI: -1.12 to 4.40). In the before phase, the mean number of INR tests was 6.7, which dropped to 6.1 tests in the after phase (mean difference, -0.61, 95% CI: -0.97 to -0.24). Only 54.5% of the warfarin dosing prescriptions were consistent with the dosing recommendations of the WMS, and adherence to the WMS was poorest in the highest INR target range. CONCLUSIONS: For the first time in Iran, we demonstrated that a computerized system was as effective as a traditional experience-based method to monitor INR in VKA-anticoagulated patients. Furthermore, it could reduce both the number of INR tests and that of visits.

Comparison of the accuracy of inpatient morbidity coding with ICD-11 and ICD-10.

BACKGROUND: One of the challenges when transitioning from International Statistical Classification of Diseases and Related Health Problems, 10th Revision (ICD-10) to International Classification of Diseases, 11th Revision (ICD-11) is to ensure clinical coding accuracy. OBJECTIVE: To determine the accuracy of clinical coding with ICD-11 in comparison with ICD-10 and identify causes of coding errors in real clinical coding environments. METHOD: The study was conducted prospectively in two general hospitals. Medical records of discharged inpatients were coded by hospital clinical coders with both ICD-11 and ICD-10 on different days. These medical records were recoded by five mentors. Codes assigned by mentors were used as the gold standard for the evaluation of accuracy. RESULTS: The accuracy of ICD-10 and ICD-11 coding for 1578 and 2168 codes was evaluated. Coding accuracy was 89.1% and 74.2% for ICD-10 and ICD-11. In ICD-11, the lowest accuracy was observed in chapters 22 (injuries), 10 (ear) and 11 (circulatory) (51.1%, 53.8% and 62.7%, respectively). In both ICD-10 and ICD-11, the most important cause of the coding errors was clinical coders' mistakes (79.5% and 81.8% for ICD-10 and ICD-11, respectively). CONCLUSION: Accuracy of clinical coding with ICD-11 was lower relative to ICD-10. Hence, it is essential to carry out initial preparations, particularly the training of clinical coders based on their needs, as well as the necessary interventions to enhance the documentation of medical records according to ICD-11 before or simultaneous with the country-wide implementation. IMPLICATIONS: Clinical coders need complete training, especially in using extension codes and post-coordination coding. Local ICD-11 guidelines based on the needs of local users and reporting policies should be developed. Furthermore, documentation guidelines based on ICD-11 requirements should be developed.

Epidemiological, clinical, and economic burden of myocardial infarction patients in Iran during the COVID-19 pandemic.

Background: To define changes in AMI case rates, patient demographics, cardiovascular comorbidities, treatment approaches, in-hospital outcomes, and the economic burden of COVID-19 during the pandemic. Methods: We conducted a multicenter, observational survey with selected hospitals from three medical universities in Tehran city. A data collection tool consisting of three parts. The first part included socio-demographic information, and the second part included clinical information, major complications, and in-hospital mortality. Finally, the third part was related to the direct medical costs generated by AMI in COVID-19 and non-COVID-19 patients. The study cohort comprised 4,560 hospitalizations for AMI (2,935 for STEMI [64%] and 1,625 for NSTEMI [36%]). Results: Of those hospitalized for AMI, 1,864 (76.6 %) and 1,659 (78 %) were male before the COVID-19 outbreak and during the COVID-19 era, respectively. The length of stay (LOS), was significantly lower during the COVID-19 pandemic era (4.27 ± 3.63 vs 5.24 ± 5.17, p = 0.00). Results showed that there were no significant differences in terms of patient risk factors across periods. A total of 2,126 AMIs were registered during the COVID-19 era, with a 12.65 % reduction (95 % CI 1.5-25.1) compared with the equivalent time in 2019 (P = 0.179). The risk of in-hospital mortality rate for AMI patients increased from 4.9 % in 2019 to 7.0 % in the COVID-19 era (OR = 1.42; 95 % CI 1.11-1.82; P = 0.004). Major complications were registered in 9.7 % of cases in 2020, which is higher than the rate of 6.6 % reported in 2019 (OR = 1.46, 95 % CI 1.11-1.82; P = 0.000). Total costs in hospitalized AMI-COVID patients averaged $188 more than in AMI patients (P = 0.020). Conclusion: This cross-sectional study found important changes in AMI hospitalization rates, worse outcomes, and higher costs during the COVID-19 periods. Future studies are recommended to examine the long-term outcomes of hospitalized AMI patients during the COVID-19 era.

Predictors of readmission in hospitalized heart failure patients.

Introduction: Heart failure(HF) related hospitalization constitutes a significant proportion of healthcare cost. Unchanging rates of readmission during recent years, shows the importance of addressing this problem. Methods: Patients admitted with heart failure diagnosis in our institution during April 2018to August 2018 were selected. Clinical, para-clinical and imaging data were recorded. All included patients were followed up for 6 months. The primary endpoints of the study were prevalence of early readmission and the predictors of that. Secondary end points were in-hospital and 6-month post-discharge mortality rate and late readmission rate. Results: After excluding 94 patients due to missing data, 428 patients were selected. Mean age of patients was 58.5 years (±17.4) and 61% of patients were male. During follow-up, 99patients (24%) were readmitted. Early re-admission (30-day) occurred in 27 of the patients(6.6%). The predictors of readmission were older age (P = 0.006), lower LVEF (P <0.0001), higher body weight (P = 0.01), ICD/CRT implantation (P = 0.001), Lower sodium (P = 0.01), higher Pro-BNP(P = 0.01), Higher WBC count (P = 0.01) and higher BUN level (P = 0.02). Independent predictors of early readmission were history of device implantation (P = 0.007), lower LVEF (P = 0.016), QRS duration more than 120 ms (P = 0.037), higher levels of BUN (P = 0.008), higher levels of Pro-BNP(P = 0.037) and higher levels of uric acid (P = 0.035). Secondary end points including in-hospital and 6-month post-discharge mortality occurred in 11% and 14.4% of patients respectively. Conclusion: Lower age of our heart failure patients and high prevalence of ischemic cardiomyopathy, necessitate focusing on more preventable factors related to heart failure.

The Utility of rRT-PCR in Diagnosis and Assessment of Case-fatality rates of COVID-19 In the Iranian Population. Positive Test Results are a Marker for Illness Severity.

The utility of PCR-based testing in characterizing patients with COVID-19 and the severity of their disease remains unknown. We performed an observational study among patients presenting to hospitals in Iran who were tested for 2019-nCoV viral RNA by rRT-PCR between the fourth week of February 2020 to the fourth week of March 2020. Frequency of symptoms, comorbidities, intubation, and mortality rates were compared between COVID-19 positive vs. negative patients. 96103 patients were tested from 879 hospitals. 18754 (19.5%) tested positive for COVID-19. Positive testing was more frequent in those 50 years or older. The prevalence of cough (54.5% vs. 49.7%), fever (49.5% vs. 44.7%), and respiratory distress (43.0% vs. 39.0%) but not hypoxia (46.9% vs. 56.7%) was higher in COVID-19 positive vs. negative patients (p<0.001 for all). More patients had cardiovascular diseases (10.6% vs. 9.5%, p<0.001) and type 2 diabetes mellitus (10.8% vs. 8.7%, p<0.001) among COVID-19 positive vs. negative patients. There were fewer patients with cancer (1.1%, vs. 1.4%, p<0.001), asthma (1.9% vs. 2.5%, p<0.001), or pregnant (0.4% vs. 0.6%, =0.001) in COVID-19 positive vs. negative groups. COVID-19 positive vs. negative patients required more intubation (7.7% vs. 5.2%, p<0.001) and had higher mortality (14.6% vs. 6.3%, p<0.001). Odds ratios for death of positive vs negative patients range from 2.01 to 3.10 across all age groups. In conclusion, COVID-19 test-positive vs. test-negative patients had more severe symptoms and comorbidities, required higher intubation, and had higher mortality. rRT-PCR positive result provided diagnosis and a marker of disease severity in Iranians.

Echocardiographic predictors of worsening renal function in acute heart failure: observations from the RASHF registry.

AIMS: Echocardiography is known as the most useful diagnostic test in the assessment of patients with heart failure (HF), and the prognostic significance of echocardiographic findings in HF is well known. In this report, we aim to present the prognostic significance of a limited set of echocardiographic parameters obtained within 24 h of admission of patients enrolled in the Rajaie Acute Systolic Heart Failure registry. METHODS AND RESULTS: A total of 230 patients with the diagnosis of acute systolic HF (left ventricular ejection fraction ≤ 35%) were enrolled into the study. Transthoracic echocardiography was performed for all study population within 24 h of admission. The primary endpoint of the study was the occurrence of worsening renal function (WRF) during the hospitalization course.Acquiring data of transthoracic echocardiography within 24 h of admission was feasible in all study participants. The median (inter-quartile range) of left ventricular ejection fraction was 20% (15-23%). Severe right ventricular dysfunction was observed in 21.5% of patients. The grade of inferior vena cava collapse and right ventricular systolic dysfunction were associated with WRF. In multivariable analysis, right ventricular systolic dysfunction was among the independent predictors of WRF [ß = 0.8, P = 0.01, odds ratio (OR) = 2.4 (1.2-4.9)] and in-hospital mortality [ß = 0.6, P = 0.04, OR = 1.5 (0.5-4.6)]. CONCLUSIONS: Echocardiographic parameters are useful for baseline assessment and provide additional information besides other clinical variables for prognostication. Right ventricular dysfunction is the most important risk factor in developing WRF and in-hospital mortality in patients with acute HF.

Shift working and risk of lipid disorders: a cross-sectional study.

BACKGROUND: previous studies have indicated on association between shift work and lipid profile disturbances. Lipid profile disturbances could be due to internal desynchronization. The aim of this study was to analyze whether there is relationship between shift work and serum lipids, fasting blood glucose and hypertension. RESULTS: A total of 424 rail road workers between the ages of 21 and 64 years in this study filled out a questionnaire, and total cholesterol, triglyceride and HDL-C concentration were measured after 12-hours fasting. Association between shift work and biochemical variables and blood pressure were measured. The X2 and fisher's exact test was used for comparing the qualitative variables and for quantitative variables with normal distribution we used the parametric tests. Odds ratio (OR) with the 95% confidence interval (95% CI) was used for comparing the proportions of risk variables.Sub-populations in this study were consisting of 158 (37.3%) shift workers and 266 (62.7%) day workers. High levels of total cholesterol (> 200 mg/dl) and LDL-cholesterol (> 130 mg/dl) were significantly more prevalent in nearly all groups of shift workers irrespective of age. But there is no differences in the serum levels of triglyceride, HDL-C, fasting blood glucose and blood pressure between shift workers and day workers.Adjusted Odd's ratio for the effect of shift working on high serum total cholesterol and LDL-C level were 2.11(95%CI: 1.33-3.36) and 1.76(95%CI: 1.09-2.83), respectively. CONCLUSION: This study showed that high serum total cholesterol and LDL-C level were more common in shift workers than in day workers. This finding persisted after adjustment was made for age and food type. But there was no difference in the prevalence of HDL-C, triglyceride, fasting blood glucose and hypertension between shift working and day working. It was concluded that shift work is a risk factor for lipid profile disturbances.

Comparison between Prothrombin Complex Concentrate (PCC) and Fresh Frozen Plasma (FFP) for the Urgent Reversal of Warfarin in Patients with Mechanical Heart Valves in a Tertiary Care Cardiac Center.

Fresh frozen plasma (FFP) and prothrombin complex concentrate (PCC) reverse oral anticoagulants such as Warfarin. We compared the standard dosage of FFP and PCC in terms of efficacy and safety for patients with mechanical heart valves undergoing interventional procedures while receiving Warfarin. Fifty patients were randomized (25 for each group) with mechanical heart valves [international normalized ratio (INR) >2.5]. FFP dosage was administered based on body weight (10-15 mL/Kg), while PCC dosage was administered based on both body weight and target INR. INR measurements were obtained at different time after PCC and FFP infusion. The mean ± SD of INR pre treatment was not significantly different between the PCC and FFP groups. However, over a 48-hour period following the administration of PCC and FFP, 76% of the patients in the PCC group and only 20% of the patients in the FFP group reached the INR target. Five (20%) patients in the PCC group received an additional dose of PCC, whereas 17 (68%) patients in the FFP group received a further dose of FFP (P=0.001). There was no significant difference between the two groups in Hb and Hct before and during a 48-hour period after PCC and FFP infusion. As regards safety monitoring and adverse drug reaction screening in the FFP group, the INR was high (INR > 2.5) in 86% of the patients. There was no report of hemorrhage in both groups. PCC reverses anticoagulation both effectively and safely while having the advantage of obviating the need to extra doses.