Finding undiagnosed patients with hepatitis C virus: an application of machine learning to US ambulatory electronic medical records.

AIMS: To develop and validate a machine learning (ML) algorithm to identify undiagnosed hepatitis C virus (HCV) patients, in order to facilitate prioritisation of patients for targeted HCV screening. METHODS: This retrospective study used ambulatory electronic medical records (EMR) from January 2015 to February 2020. A Gradient Boosting Trees algorithm was trained using patient records to predict initial HCV diagnosis and was validated on a temporally independent held-out cross-section of the data. The fold improvement in precision (proportion of patients identified by the algorithm who are HCV positive) over universal screening was examined and compared with risk-based screening. RESULTS: 21 508 positive (HCV diagnosed) and 28.2M unlabelled (lacking evidence of HCV diagnosis) patients met the inclusion criteria for the study. After down-sampling unlabelled patients to aid the algorithm's learning process, 16.2M unlabelled patients entered the analysis. Performance of the algorithm was compared with universal screening on the held-out cross-section, which had an incidence of HCV diagnoses of 0.02%. The algorithm achieved a 101.0 ×, 18.0 × and 5.1 × fold improvement in precision over universal screening at 5%, 20% and 50% levels of recall. When compared with risk-based screening, the algorithm required fewer patients to be screened and improved precision. CONCLUSIONS: This study presents strong evidence towards the use of ML on EMR data for the prioritisation of patients for targeted HCV testing with potential to improve efficiency of resource utilisation, thereby reducing the workload for clinicians and saving healthcare costs. A prospective interventional study would allow for further validation before use in a clinical setting.

Machine learning using longitudinal prescription and medical claims for the detection of non-alcoholic steatohepatitis (NASH).

OBJECTIVES: To develop and evaluate machine learning models to detect patients with suspected undiagnosed non-alcoholic steatohepatitis (NASH) for diagnostic screening and clinical management. METHODS: In this retrospective observational non-interventional study using administrative medical claims data from 1 463 089 patients, gradient-boosted decision trees were trained to detect patients with likely NASH from an at-risk patient population with a history of obesity, type 2 diabetes mellitus, metabolic disorder or non-alcoholic fatty liver (NAFL). Models were trained to detect likely NASH in all at-risk patients or in the subset without a prior NAFL diagnosis (at-risk non-NAFL patients). Models were trained and validated using retrospective medical claims data and assessed using area under precision recall curves and receiver operating characteristic curves (AUPRCs and AUROCs). RESULTS: The 6-month incidences of NASH in claims data were 1 per 1437 at-risk patients and 1 per 2127 at-risk non-NAFL patients . The model trained to detect NASH in all at-risk patients had an AUPRC of 0.0107 (95% CI 0.0104 to 0.0110) and an AUROC of 0.84. At 10% recall, model precision was 4.3%, which is 60× above NASH incidence. The model trained to detect NASH in the non-NAFL cohort had an AUPRC of 0.0030 (95% CI 0.0029 to 0.0031) and an AUROC of 0.78. At 10% recall, model precision was 1%, which is 20× above NASH incidence. CONCLUSION: The low incidence of NASH in medical claims data corroborates the pattern of NASH underdiagnosis in clinical practice. Claims-based machine learning could facilitate the detection of patients with probable NASH for diagnostic testing and disease management.

Finding undiagnosed patients with hepatitis C infection: an application of artificial intelligence to patient claims data.

Hepatitis C virus (HCV) remains a significant public health challenge with approximately half of the infected population untreated and undiagnosed. In this retrospective study, predictive models were developed to identify undiagnosed HCV patients using longitudinal medical claims linked to prescription data from approximately ten million patients in the United States (US) between 2010 and 2016. Features capturing information on demographics, risk factors, symptoms, treatments and procedures relevant to HCV were extracted from patients' medical history. Predictive algorithms were developed based on logistic regression, random forests, gradient boosted trees and a stacked ensemble. Descriptive analysis indicated that patients exhibited known symptoms of HCV on average 2-3 years prior to their diagnosis. The precision was at least 95% for all algorithms at low levels of recall (10%). For recall levels >50%, the stacked ensemble performed best with a precision of 97% compared with 87% for the gradient boosted trees and just 31% for the logistic regression. For context, the Center for Disease Control recommends screening in an at-risk sub-population with an estimated HCV prevalence of 2.23%. The artificial intelligence (AI) algorithm presented here has a precision which is substantially higher than the screening rates associated with recommended clinical guidelines, suggesting that AI algorithms have the potential to provide a step change in the effectiveness of HCV screening.

Utilising artificial intelligence to determine patients at risk of a rare disease: idiopathic pulmonary arterial hypertension.

Idiopathic pulmonary arterial hypertension is a rare and life-shortening condition often diagnosed at an advanced stage. Despite increased awareness, the delay to diagnosis remains unchanged. This study explores whether a predictive model based on healthcare resource utilisation can be used to screen large populations to identify patients at high risk of idiopathic pulmonary arterial hypertension. Hospital Episode Statistics from the National Health Service in England, providing close to full national coverage, were used as a measure of healthcare resource utilisation. Data for patients with idiopathic pulmonary arterial hypertension from the National Pulmonary Hypertension Service in Sheffield were linked to pre-diagnosis Hospital Episode Statistics records. A non-idiopathic pulmonary arterial hypertension control cohort was selected from the Hospital Episode Statistics population. Patient history was limited to ≤5 years pre-diagnosis. Information on demographics, timing/frequency of diagnoses, medical specialities visited and procedures undertaken was captured. For modelling, a bagged gradient boosting trees algorithm was used to discriminate between cohorts. Between 2008 and 2016, 709 patients with idiopathic pulmonary arterial hypertension were identified and compared with a stratified cohort of 2,812,458 patients classified as non-idiopathic pulmonary arterial hypertension with ≥1 ICD-10 coded diagnosis of relevance to idiopathic pulmonary arterial hypertension. A predictive model was developed and validated using cross-validation. The timing and frequency of the clinical speciality seen, secondary diagnoses and age were key variables driving the algorithm's performance. To identify the 100 patients at highest risk of idiopathic pulmonary arterial hypertension, 969 patients would need to be screened with a specificity of 99.99% and sensitivity of 14.10% based on a prevalence of 5.5/million. The positive predictive and negative predictive values were 10.32% and 99.99%, respectively. This study highlights the potential application of artificial intelligence to readily available real-world data to screen for rare diseases such as idiopathic pulmonary arterial hypertension. This algorithm could provide low-cost screening at a population level, facilitating earlier diagnosis, improved diagnostic rates and patient outcomes. Studies to further validate this approach are warranted.

Impact of Transcendental Meditation on Psychotropic Medication Use Among Active Duty Military Service Members With Anxiety and PTSD.

The purpose of the study was to determine whether the regular practice of Transcendental Meditation (TM) decreased the need for psychotropic medications required for anxiety and post-traumatic stress disorder (PTSD) management and increased psychological wellbeing. The sample included 74 military Service Members with documented PTSD or anxiety disorder not otherwise specified (ADNOS), 37 that practiced TM and 37 that did not. At 1 month, 83.7% of the TM group stabilized, decreased, or ceased medications and 10.8% increased medication dosage; compared with 59.4% of controls that showed stabilizations, decreases, or cessations; and 40.5% that increased medications (p < 0.03). A similar pattern was observed after 2 (p < 0.27), 3 (p < 0.002), and 6 months (p < 0.34). Notably, there was a 20.5% difference between groups in severity of psychological symptoms after 6 months, that is, the control group experienced an increase in symptom severity compared with the group practicing TM. These findings provide insight into the benefits of TM as a viable treatment modality in military treatment facilities for reducing PTSD and ADNOS psychological symptoms and associated medication use.

Clinical case series: treatment of PTSD with transcendental meditation in active duty military personnel.

Active duty U.S. Army Service Members previously diagnosed with post-traumatic stress disorder (PTSD) were selected from review of patient records in the Traumatic Brain Injury Clinic at the Department of Defense Eisenhower Army Medical Center at Fort Gordon in Augusta, Georgia. Patients agreed to practice the Transcendental Meditation (TM) technique for 20 minutes twice a day for the duration of a 2-month follow-up period. Three cases are presented with results that show the feasibility of providing TM training to active duty soldiers with PTSD in a Department of Defense medical facility. Further investigation is suggested to determine if a TM program could be used as an adjunct for treatment of PTSD. Impact of this report is expected to expand the complementary and alternative evidence base for clinical care of PTSD.

Concussions and the military: issues specific to service members.

Since October 2001, more than 1.6 million American military service members have deployed to Iraq and Afghanistan in the Global War on Terrorism. It is estimated that between 5% and 35% of them have sustained a concussion, also called mild traumatic brain injury (mTBI), during their deployment. Up to 80% of the concussions experienced in theater are secondary to blast exposures. The unique circumstances and consequences of sustaining a concussion in combat demands a unique understanding and treatment plan. The current literature was reviewed and revealed a paucity of pathophysiological explanations on the nature of the injury and informed treatment plans. However, through observation and experience, a theoretical but scientifically plausible model for why and how blast injuries experienced in combat give rise to the symptoms that affect day-to-day function of service members who have been concussed has been developed. We also are able to offer treatment strategies based on our evaluation of the current literature and experience to help palliate postconcussive symptoms. The purpose of this review is to elucidate common physical, cognitive, emotional, and situational challenges, and possible solutions for this special population of patients who will be transitioning into the civilian sector and interfacing with health professionals. There is a need for further investigation and testing of these strategies.

Perioperative epidural analgesia for major abdominal surgery for cancer and recurrence-free survival: randomised trial.

OBJECTIVE: To compare long term recurrence of cancer and survival of patients having major abdominal surgery for cancer. DESIGN: Long term follow-up of prospective randomised controlled clinical trial in which patients were randomly assigned to receive general anaesthesia with or without epidural block for at least three postoperative days. Setting 23 hospitals in Australia, New Zealand, and Asia. PARTICIPANTS: 503 adult patients who had potentially curative surgery for cancer. MAIN OUTCOME MEASURE: Cancer-free survival (analysis was by intention to treat). RESULTS: Long term follow-up data were available for 94% (n=446) of eligible participants. The median time to recurrence of cancer or death was 2.8 (95% confidence interval 0.7 to 8.7) years in the control group and 2.6 (0.7 to 8.7) years in the epidural group (P=0.61). Recurrence-free survival was similar in both epidural and control groups (hazard ratio 0.95, 95% confidence interval 0.76 to 1.17; P=0.61). CONCLUSION: Use of epidural block in abdominal surgery for cancer is not associated with improved cancer-free survival. TRIAL REGISTRATION: Australian New Zealand Clinical Trials Registry ACTRN12607000637448.

Neurorehabilitation in traumatic brain injury: does it make a difference?

Each year, 1.4 million people in the United States are seen in a hospital for a traumatic brain injury. Those with moderate-to-severe traumatic brain injury frequently go through a course of inpatient neurorehabilitation prior to discharge back into the community. A broad overview of neurorehabilitation is presented, including the standards for admission to inpatient rehabilitation and the members and roles of the neurorehabilitation team. Common medical complications that are managed after moderate-to-severe traumatic brain injury are reviewed. The spectrum of arousal issues is summarized. The evidence regarding neurorehabilitation is then reviewed. Future studies that are underway to better understand the utility of neurorehabilitation are then discussed.

Child health: evidence on the roles of family income and maternal mental health from a UK birth cohort.

There is a growing literature showing an association between higher family income and better child health. This paper uses cohort data with rich information on mother's early life events, her health, child-health-related behaviours, and her child's health to examine this association for the UK and to identify some of the mechanisms through which income affects child health. The paper examines the cross-sectional association between income and health, finds the expected association, but concludes that the association with current income cannot be distinguished from one between permanent income and child health. It then focuses on the mechanisms by which income translates into better child health; these include parental behaviours that may affect child health and parental health, including maternal mental health. Controlling for these factors, there is almost no direct impact of income. A significant role is played by mother's own health, particularly her mental health. No clear role is played by child-health production behaviours of the mother. Examining the maternal mental health-child health link in more detail suggests a role for maternal anxiety and somaticism.

Perioperative epidural analgesia and outcome after major abdominal surgery in high-risk patients.

In a primary analysis of a large recently completed randomized trial in 915 high-risk patients undergoing major abdominal surgery, we found no difference in outcome between patients receiving perioperative epidural analgesia and those receiving IV opioids, apart from the incidence of respiratory failure. Therefore, we performed a selected number of predetermined subgroup analyses to identify specific types of patients who may have derived benefit from epidural analgesia. We found no difference in outcome between epidural and control groups in subgroups at increased risk of respiratory or cardiac complications or undergoing aortic surgery, nor in a subgroup with failed epidural block (all P > 0.05). There was a small reduction in the duration of postoperative ventilation (geometric mean [SD]: control group, 0.3 [6.5] h, versus epidural group, 0.2 [4.8] h; P = 0.048). No differences were found in length of stay in intensive care or in the hospital. There was no relationship between frequency of use of epidural analgesia in routine practice outside the trial and benefit from epidural analgesia in the trial. We found no evidence that perioperative epidural analgesia significantly influences major morbidity or mortality after major abdominal surgery.

Epidural anaesthesia and analgesia and outcome of major surgery: a randomised trial.

BACKGROUND: Epidural block is widely used to manage major abdominal surgery and postoperative analgesia, but its risks and benefits are uncertain. We compared adverse outcomes in high-risk patients managed for major surgery with epidural block or alternative analgesic regimens with general anaesthesia in a multicentre randomised trial. METHODS: 915 patients undergoing major abdominal surgery with one of nine defined comorbid states to identify high-risk status were randomly assigned intraoperative epidural anaesthesia and postoperative epidural analgesia for 72 h with general anaesthesia (site of epidural selected to provide optimum block) or control. The primary endpoint was death at 30 days or major postsurgical morbidity. Analysis by intention to treat involved 447 patients assigned epidural and 441 control. FINDINGS: 255 patients (57.1%) in the epidural group and 268 (60.7%) in the control group had at least one morbidity endpoint or died (p=0.29). Mortality at 30 days was low in both groups (epidural 23 [5.1%], control 19 [4.3%], p=0.67). Only one of eight categories of morbid endpoints in individual systems (respiratory failure) occurred less frequently in patients managed with epidural techniques (23% vs 30%, p=0.02). Postoperative epidural analgesia was associated with lower pain scores during the first 3 postoperative days. There were no major adverse consequences of epidural-catheter insertion. INTERPRETATION: Most adverse morbid outcomes in high-risk patients undergoing major abdominal surgery are not reduced by use of combined epidural and general anaesthesia and postoperative epidural analgesia. However, the improvement in analgesia, reduction in respiratory failure, and the low risk of serious adverse consequences suggest that many high-risk patients undergoing major intra-abdominal surgery will receive substantial benefit from combined general and epidural anaesthesia intraoperatively with continuing postoperative epidural analgesia.