Population-split-based risk assessment model of venous thromboembolism in Chinese medical inpatients

Objective Inpatients with high risk of venous thromboembolism (VTE) usually face serious threats to their health and economic conditions. Many studies using machine learning (ML) models to predict VTE risk neglected an important statistical phenomenon, "fuzzy feature", and achieved inferior results. Considering the effect of "fuzzy feature", our study aims to develop a VTE risk assessment model suitable for Chinese medical inpatients. Materials and Methods Inpatients in the medical department of Peking Union Medical College Hospital (PUMCH) from January 2014 to June 2016 were collected. A new ML VTE risk assessment model was built through population splitting. First patients were classified into different groups based on values of VTE risk factors, then trustless groups were filtered out, and finally ML models were built on training data in unit of groups. Predictive performances of our method, five traditional ML models, and the Padua model were compared. Results The "fuzzy feature" was verified on the whole dataset. Compared with the Padua model, the proposed model showed higher sensitivities and specificities on training data, and higher specificities and similar sensitivities on test data. Standard deviations of predictive validity of five ML models were larger than the proposed model. Discussion The proposed model was the only one which showed advantages on both sensitivity and specificity over Padua model. Its robustness was better than traditional ML models. Conclusion This study built a population-split-based ML model of VTE for Chinese medical inpatients and it may help clinicians stratify VTE risk and guide prevention more efficiently.

Antimicrobial Susceptibility of Neisseria gonorrhoeae in Adult Patients Seeking Care at Military Hospitals in Thailand From 2014 to 2020

ABSTRACT Introduction The effective dual antibiotics ceftriaxone (CRO) and azithromycin (AZM) have successfully treated Neisseria gonorrhoeae (GC) infection, however, the CRO- and AZM-resistant strains have been sporadically detected globally and in Thailand. Furthermore, there are no currently antimicrobial susceptibility profiles of the GC isolates obtained from soldiers reported in Thailand. Hence, this is the first study to describe the antimicrobial susceptibility profiles of GC isolates obtained from predominately soldiers who seeking care at Military Camp Hospitals, in Thailand from 2014 to 2020. Materials and Methods A total of 624 symptomatic gonococcal samples were received from 10 military hospitals during 2014-2020. They were collected from urethral swabs and inoculated into selective media. The suspected GC isolates were subcultured and presumptively identified using conventional microbiology techniques. Antimicrobial susceptibility test was performed by Etest to determine minimal inhibitory concentration (μg/mL) against AZM, benzylpenicillin, cefepime, cefixime, ceftriaxone (CRO), ciprofloxacin, spectinomycin, and tetracycline using the criteria outlined in the Clinical and Laboratory Standards Institute guidelines. This study was approved by Institutional Review Board, Royal Thai Army Medical Department under protocol number S036b/56 and Walter Reed Army Institute of Research, and Silver Spring, MD under protocol number WR #2039. Results A total of 624 samples were collected from symptomatic gonococcal infectious patients with 91.5% (571/624) of samples obtained from soldiers. Of those, 78% (488/624) were identified as GC and 92% (449/488) of them were isolated from soldiers. All GC samples collected were susceptible to CRO (first-line treatment) with only one GC isolate identified as non-susceptible to cefepime and three isolates identified as non-susceptible to AZM. Conclusion The recommended dual treatment of GC infections with CRO and AZM is currently an effective empirical treatment for patients who are seeking care at military hospitals in Thailand. Nevertheless, cefepime is a fourth-generation cephalosporin with documented high activity against GC strains equal to other “third-generation” cephalosporins such as CRO. Due to the active duty of military personnel, they concerned about the confidentiality and frequently seek treatment at civilian clinics. Additionally, due to the availability of antibiotics over the counter in Thailand, many choose the option to self-medicate without a physician’s prescription. These could be subsequently driven the gradual increase of multidrug-resistant gonococcal strains throughout the country. Thus, the GC surveillance would be needed for further Force Health Protection and public health authorities in response to the drug-resistant GC threats.

Contraband Hospital, Freedmen’s Hospital, and Alexander Augusta

The tens of thousands of enslaved Blacks liberated by union forces during the American Civil War were considered seized property and thus were referred to as contraband. As wartime refugees they sought protection in federal military installations, popularly known as contraband camps, located throughout the occupied South. One of the largest was Camp Baker in the rural northwestern sector of Washington, DC, where about 40,000 persons were sheltered. To provide basic medical care, the military outfitted, in 1863, an infirmary called the Contraband Hospital, later renamed Freedmen’s Hospital. From its founding in 1867 the medical department of Howard University was attached to Freemen’s Hospital, which in 1975 was renamed the Howard University Hospital, the two institutions establishing a long partnership of medical education and hospital care that continues to the present day.

Imparting medical knowledge to pharmaceutical sales force

Keycustomers of the pharmaceutical industry are qualified medical practitioners. To be able to stand their ground the sales representative needs training about medical terminology, the relevant disease, the molecule being marketed and some competitor information. A short induction could be followed by ongoing refresher training either face to face or online. The medical department has a major role to play here. Product launch training is planned by medical and marketing team based on the feature benefits of the product being launched. The training unit of the medical department of pharma companies are an inseparable part of the whole marketing effort.

Disease Named Entity Recognition in Medical Records: Analysis of Transfer Learning between Different Hospital Departments (Preprint)

BACKGROUND Electronic records provide rich clinical information for biomedical text mining. However, a system developed on one hospital department may not generalize to other departments. Here, we use hospital medical records as a research data source and explore the heterogeneous problem posed by different hospital departments. OBJECTIVE We use MIMIC-III hospital medical records as the research data source. We collaborate with medical experts to annotate the data, with 328 records being included in analyses. Disease named entity recognition (NER), which helps medical experts in consolidating diagnoses, is undertaken as a case study. METHODS To compare heterogeneity of medical records across departments, we access text from multiple departments and employ the similarity metrics. We apply transfer learning to NER in different departments’ records and test the correlation between performance and similarity metrics. We use TF-IDF cosine similarity of the named entities as our similarity metric. We use three pretrained model on the disease NER task to valid the consistency of the result. RESULTS The disease NER dataset we release consists of 328 medical records from MIMIC-III, with 95629 sentences and 8884 disease mentions in total. The inter annotator agreement Cohen’s kappa coefficient is 0.86. Similarity metrics support that medical records from different departments are heterogeneous, ranges from 0.1004 to 0.3541 compare to Medical department. In the transfer learning task using the Medical department as the training set, F1 score performs in three pretrained models average from 0.847 to 0.863. F1 scores correlate with similarity metrics with Spearman’s coefficient of 0.4285. CONCLUSIONS We propose a disease NER dataset based on medical records from MIMIC-III and demonstrate the effectiveness of transfer learning using BERT. Similarity metrics reveal noticeable heterogeneity between department records. The deep learning-based transfer learning method demonstrates good ability to generalize across departments and achieve decent NER performance thus eliminates the concern that training material from one hospital might compromise model performance when applied to another. However, the model performance does not show high correlation to the departments’ similarity.

Patient Vs. Soldier-Centered Medical Home: Comparing Access, Continuity, and Communication in the U.S. Army

ABSTRACT Introduction Over the last 40 years, patient-centered medical home (PCMH) has evolved as the leading primary care practice model, replacing traditional primary care models in the United States and internationally. The goal of PCMH is to improve chronic condition management. In the U.S. Army, the scope of the medical home, which encompasses various care delivery platforms, including PCMH and soldier-centered medical home (SCMH), extends beyond the management of chronic illnesses. These medical home platforms are designed to support the unique health care needs of the U.S. Army’s most vital asset—the soldier. The PCMHs and SCMHs within the U.S. Army employ patient-centered care principles while incorporating nationally recognized structural attributes and care processes that work together in a complex adaptive system to improve organizational and patient outcomes. However, U.S. Army policies dictate differences in the structures of PCMHs and SCMHs. Researchers suggest that differences in medical home structures can impact how organizations operationalize care processes, leading to unwanted variance in organizational and patient outcomes. This study aimed to compare 3 care processes (access to care, primary care manager continuity, and patient-centered communication) between PCMHs and SCMHs. Materials and Methods This was a retrospective, cross-sectional, and correlational study. We used a subset of data from the Military Data Repository collected between January 1, 2018, and December 31, 2018. The sample included 266 medical home teams providing care for active duty soldiers. Only active duty soldiers were included in the sample. We reviewed current U.S. Army Medical Department policies to describe the structures and operational functioning of PCMHs and SCMHs. General linear mixed regressions were used to evaluate the associations between medical home type and outcome measures. The U.S. Army Medical Department Center and School Institutional Review Board approved this study. Results There was no significant difference in access to 24-hour and future appointments or soldiers’ perception of access between PCMHs and SCMHs. There was no significant difference in primary care manager continuity. There was a significant difference in medical home team continuity (P < .001), with SCMHs performing better. There was no significant difference in patient-centered communication scores. Our analysis showed that while the PCMH and SCMH models were designed to improve primary care manager continuity, access to care, and communication, medical home teams within the U.S. Army are not consistently meeting the Military Health System standard of care benchmarks for these care processes. Conclusions Our findings comparing 3 critical medical home care processes suggest that structural differences may impact continuity but not access to care or communication. There is an opportunity to further explore and improve access to appointments within 24 hours, primary care manager and medical home team continuity, perception of access to care, and the quality of patient-centered communication among soldiers. Knowledge gained from this study is essential to soldier medical readiness.

Organization of Medical Support for the Military Service Call-Up of Citizens during New Coronavirus Infection COVID-19 Pandemic

The aim of the study was to evaluate the organization of screening of conscripts for COVID-19 infection in one of the subjects of the Russian Federation — Orenburg region. Materials and methods of the study. The study of organization of examination of conscripts for COVID-19 infection included study of the experience of the Center of Military Medical Examination of the Military Commissariat of Orenburg Region in carrying out military conscription under conditions of the COVID-19 pandemic; methodological recommendations approved by the Head of the Main Military Medical Department of the Defense Ministry of Russia as well as the analysis of the activities carried out at different stages of medical sorting. Results of the study and their analysis. The results of the study of the organization of COVID-19 screening of conscripts in Orenburg Region showed that sufficient and effective barrier medical screening was organized at all stages. Conscripts in good health condition were sent to the Armed Forces for military service under conscription. No claims were received from military units about poor quality selection of conscripts. In 2020, during the COVID-19 pandemic, recruitment commissions of Orenburg Region coped successfully with the task of drafting citizens for military service.

Microbial spectrum and antibiotic sensitivity in cirrhotic patients with spontaneous bacterial peritonitis

Objective: There are reports of changing microbial profile of ascitic fluid in spontaneous bacterial peritonitis (SBP) and developing resistance of these bacteria to commonly used antibiotics. This study was done to determine the micro-organism causing SBP and their sensitivity to various antibiotics. Setting and Methods: This observational study was done in Medical department of Sheikh Zayed Medical College/Hospital, Rahim Yar Khan from March to November 2019. Thirty five cirrhotic patients with ascites admitted in ward for various reasons and having positive ascitic fluid culture were included in the study. Exclusion criteria was presence of secondary bacterial peritonitis. Ten ml of ascitic fluid was inoculated in blood culture bottle and sent immediately to hospital laboratory. Results: The mean age of these 35 patients was 48.94 ± 13.51 years with a range of 19 to 80 years. Twenty (57.1 %) patients were male and 15 (42.9 %) were female. Bacteria that had caused SBP were E.coli (62.9 %), staph aureus (11.4 %), klebsiella (8.6 %), streptococci (8.6 %), gram positive cocci (5.7 %) and pseudomonas aeruginosa (2.8 %). Imipenem had high sensitivity rate (100 %) along with amikacin (82.9 %) and cefoperazone-sulbactam (68.6 %). Sensitivity of these organisms to other commonly used antibiotics were: ciprofloxacin 57.1 %, ofloxacin 40 %, norfloxacin 37.1 %, ceftazidime 34.3 %, ceftriaxone 31.4 % and piperacillin-tazobactam 25.7 %. Conclusion: We found that E.coli was the commonest bacteria causing SBP, and ceftriaxone and ciprofloxacin have significantly high resistance rate in these patients. Key words: cirrhosis, ascitic fluid, spontaneous bacterial peritonitis, antibiotic sensitivity

Improving Medical Handover: An SHO-Led Approach to Implementing A Formalized Handover System

Handover is a high-risk exercise. As the number of doctors in hospitals vary between shifts, continuity of care must be secured from many professionals down to a few. Literature has repeatedly shown that handover time is when the greatest number of medical errors occur [1]. Tasks may be inefficiently handover over, mis-represented or forgotten entirely. We have shared perspectives from a quality improvement (QI) project undertaken at the William Harvey Hospital, Kent, United Kingdom towards ensuring safe and smooth handover of medical tasks between shifts at the acute medical department in this hospital.