Using Artificial Intelligence With Natural Language Processing to Combine Electronic Health Record's Structured and Free Text Data to Identify Nonvalvular Atrial Fibrillation to Decrease Strokes and Death: Evaluation and Case-Control Study

BACKGROUND

Nonvalvular atrial fibrillation (NVAF) affects almost 6 million Americans and is a major contributor to stroke but is significantly undiagnosed and undertreated despite explicit guidelines for oral anticoagulation.

OBJECTIVE

The aim of this study is to investigate whether the use of semisupervised natural language processing (NLP) of electronic health record's (EHR) free-text information combined with structured EHR data improves NVAF discovery and treatment and perhaps offers a method to prevent thousands of deaths and save billions of dollars.

METHODS

We abstracted 96,681 participants from the University of Buffalo faculty practice's EHR. NLP was used to index the notes and compare the ability to identify NVAF, congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65 to 74 years, sex category (CHA₂DS₂-VASc), and Hypertension, Abnormal liver/renal function, Stroke history, Bleeding history or predisposition, Labile INR, Elderly, Drug/alcohol usage (HAS-BLED) scores using unstructured data (International Classification of Diseases codes) versus structured and unstructured data from clinical notes. In addition, we analyzed data from 63,296,120 participants in the Optum and Truven databases to determine the NVAF frequency, rates of CHA₂DS₂‑VASc ≥2, and no contraindications to oral anticoagulants, rates of stroke and death in the untreated population, and first year's costs after stroke.

RESULTS

The structured-plus-unstructured method would have identified 3,976,056 additional true NVAF cases (P<.001) and improved sensitivity for CHA₂DS₂-VASc and HAS-BLED scores compared with the structured data alone (P=.002 and P<.001, respectively), causing a 32.1% improvement. For the United States, this method would prevent an estimated 176,537 strokes, save 10,575 lives, and save >US $13.5 billion.

CONCLUSIONS

Artificial intelligence-informed bio-surveillance combining NLP of free-text information with structured EHR data improves data completeness, prevents thousands of strokes, and saves lives and funds. This method is applicable to many disorders with profound public health consequences.

A prospective, observational study of the safety and effectiveness of intramuscular psychotropic treatment in acutely agitated patients with schizophrenia and bipolar mania

This naturalistic, observational pan-European study assessed the safety and early effectiveness of intramuscular (IM) psychotropic treatments in patients with acute agitation suffering from schizophrenia or bipolar mania. One thousand nine hundred and forty of 1945 patients completed the 24-hour observation period after initial IM treatment. Patients from 12 European countries were included (mean age 39 years; 58% male, 66% schizophrenia). IM treatment was at the physician's discretion. The primary objective was to describe the acute tolerability of IM psychotropic therapies in clinical practice, with particular emphasis on EPS. At baseline, 68% of the patients received IM monotherapy, with IM olanzapine most commonly prescribed (36%). During the first 24hours, 190 (9.8%) patients experienced EPS. The occurrence of EPS was statistically significantly lower in patients treated with IM olanzapine compared to those treated with other IM psychotropic medications (mainly typical antipsychotics and benzodiazepines): acute dystonia: 1.1%, 95% CI 0.5–2.3 and 2.9%, CI 2.0–4.0; akathisia: 2.3%, CI 1.3–3.7 and 5.5%, CI 4.3–6.9; Parkinsonism: 2.9%, CI 1.8–4.4 and 7.8%, CI 6.4–9.4, respectively. Anticholinergic treatment was given to 12% IM olanzapine versus 31% non-olanzapine treated patients. Acute agitation after 24hours was reduced by 1.68 (95% CI 1.46–1.91) points on the Clinical Global Impression of Severity (CGI-S) in IM olanzapine patients and 1.51 (95% CI 1.30–1.73) points in non-olanzapine patients. Additional psychotropic medication was required for 90% of the patients during the first 24hours of treatment. Results provide naturalistic evidence for low EPS rates and improvement of agitation with IM psychotropic medications during acute states of patients suffering from acute mania or schizophrenia.

Barriers, Facilitators, and Solutions to Optimal Patient Portal and Personal Health Record Use: A Systematic Review of the Literature

Patient portal and personal health record adoption and usage rates have been suboptimal. A systematic review of the literature was performed to capture all published studies that specifically addressed barriers, facilitators, and solutions to optimal patient portal and personal health record enrollment and use. Consistent themes emerged from the review. Patient attitudes were critical as either barrier or facilitator. Institutional buy-in, information technology support, and aggressive tailored marketing were important facilitators. Interface redesign was a popular solution. Quantitative studies identified many barriers to optimal patient portal and personal health record enrollment and use, and qualitative and mixed methods research revealed thoughtful explanations for why they existed. Our study demonstrated the value of qualitative and mixed research methodologies in understanding the adoption of consumer health technologies. Results from the systematic review should be used to guide the design and implementation of future patient portals and personal health records, and ultimately, close the digital divide.

Referral to nephrologists for chronic kidney disease care: is non-diabetic kidney disease ignored?

BACKGROUND

Late referral to nephrologists is common and associated with increased morbidity and mortality. We aimed to analyze the prevalence rates, predictors and consequences of late referral to nephrologists by primary care physicians for chronic kidney disease (CKD) care.

METHODS

A retrospective analysis of 204 patients started on dialysis for CKD in two community hospitals between March 2003 and March 2005 was conducted. Relevant clinical and laboratory data were obtained from the patient records of the nephrology clinics and dialysis units. Patients referred in CKD stage 5 (estimated glomerular filtration rate <15 ml/min) were defined as late referral and patients in CKD stage 1-4 (estimated glomerular filtration rate >15 ml/min) as early referral.

RESULTS

Forty-five (22%) of the 204 patients were referred late. In the multivariate analysis, non-diabetic kidney disease (odds ratio = 2.46, p = 0.02) and Charlson comorbidity index (odds ratio = 1.17, p = 0.009) were significantly associated with late referral. The late referral group had lower hematocrit and serum calcium levels, and higher serum phosphorus and parathyroid hormone levels than the early referral group (p < or =0.05) at the time of referral. Late referral resulted in less permanent vascular access for initiation of dialysis (p = 0.03). Even though there was twice the number of deaths in the late referral group in 1 year (18 vs. 9%), this was not statistically significant (p = 0.07).

CONCLUSION

Referring physicians should pay special attention to patients with non-diabetic kidney disease and patients with multiple comorbidities since delayed referral to nephrologists may result in poorer patient-related outcomes. Larger and long-term prospective studies analyzing the long-term consequences of late referral to nephrologists are needed.