This monthly digest of recent peer-reviewed publications in Applied Clinical Informatics starts with notable articles in the field of telemedicine.
The imbalance in healthcare between urban and rural areas is still a problem in China. With development of telemedicine services, there are new possibilities to close that gap. West China Hospital established a telemedicine center in 2001. During a 12-year study period, almost 12,000 teleconsultations were performed. Consultations led to change of treatment in 55% of cases and had estimated savings of around $3 million. Telemedicine is a cost-effective solution and has great potential in developing countries.1
Mobile health (mHealth) is often cited as a cost-effective or cost-saving technology. However, evidence supporting this statement is limited. A systematic review of economic evaluations of mHealth solutions included 39 articles from 19 countries. Most of the studies (90%) were behavior change and communication interventions, such as improving adherence and rates. Results show that 75% of the studies reported that the mHealth intervention was cost effective.2
Interruption from mobile devices is a problem in all aspects of life. Such interruptions in the healthcare setting contribute greatly to information overload. During an 18-month study, almost 200,000 interruptions (calls, text messaging, emails) were sent to hospital-assigned smartphones, as documented at Academic General Internal Medicine Wards at two tertiary care centers in Toronto, Canada. On average, four interruptions per hour were recorded. Excessive interruptions may have implications for patient safety.3
Electronic sepsis detection has been a topic for research over the past decade. There are a number of locally built solutions as well as commercial ones. A recent study of one commercial system showed comparable sensitivity and specificity to other systems.4 That is predictable, as all systems are based on programmed established diagnostic criteria. Further advancement of such systems required carefully modified workflow. An uncontrolled before-after study reported that the outcome improvement in this article could be attributed to the Hawthorne effect—the quality improvement initiative drew the attention of providers and caused them to pay more attention to such patients.
Communication to families of patients in the ICU is important. A recent study published in Annals of the American Thoracic Society identified clinician perspectives on the current state of communication among patients, families, and clinicians in the ICU and the role of an electronic portal at a single academic medical center. The conclusions are that clinicians were cautiously supportive of an electronic portal to enhance communication in the ICU, and they anticipated several potential benefits.5
The use of machine learning algorithms is on the rise. An article in Critical Care Medicine reported a study comparing the accuracy of different machine learning techniques for detecting clinical deterioration on the wards in five hospital multicenter databases. The random forest model was the most accurate model in the validation dataset (AUC 0.80), and all models were more accurate than MEWS score. In conclusion, such techniques could improve the identification of critically ill patients on the wards.6
Another study published in JAMA used deep learning algorithm and photography as methods of detecting diabetic retinopathy. In the validation cohort, deep learning algorithms had high 97% sensitivity and 93% specificity for detecting diabetic retinopathy and macular edema in retinal fundus photographs.7
The adoption of electronic medical records allows for easier access of information and order mechanism. The objective of the current study was to determine if access to EMR influences the number of laboratory and imaging tests ordered. Five years of data from non-federally employed office-based physicians was analyzed. The finding is that physicians who actively used an EHR ordered more complete blood count tests (OR, 1.34; P <.001) as well as more computerized tomography scans (OR, 1.41; P <.001) and x-rays (OR, 1.39; P <.001) than physicians who did not.8 This viewpoint is echoed in an article in JAMA that discussed the contribution of technologies to “overuse” and “unnecessary health care.”9
Last on today’s review is an article in Journal of Clinical Monitoring and Computing, "Reduction of clinically irrelevant alarms in patient monitoring by adaptive time delays." To address the problem of alerts fatigue, the authors achieved a reduction in false alerts of 73%. Effect of this approach and safety should be evaluated in a prospective study.10
1. Wang T-T, Li J-M, Zhu C-R, et al. Assessment of Utilization and Cost-Effectiveness of Telemedicine Program in Western Regions of China: A 12-Year Study of 249 Hospitals Across 112 Cities. Telemed J E Health. 2016;22(11):909-920. PMID:27314300.
2. Iribarren SJ, Cato K, Falzon L, Stone PW. What is the economic evidence for mHealth? A systematic review of economic evaluations of mHealth solutions. PLoS One. 2017;12(2):e0170581. PMID:28152012.
3. Vaisman A, Wu RC. Analysis of Smartphone Interruptions on Academic General Internal Medicine Wards. Frequent Interruptions may cause a “Crisis Mode” Work Climate. Appl Clin Inform. 2017;8(1):1-11. PMID:28066851.
4. Westra BL, Landman S, Yadav P, Steinbach M. Secondary Analysis of an Electronic Surveillance System Combined with Multi-focal Interventions for Early Detection of Sepsis. Appl Clin Inform. 2017;8(1):47-66. PMID:28097288.
5. Bell SK, Roche SD, Johansson AC, et al. Clinician Perspectives on an Electronic Portal to Improve Communication with Patients and Families in the Intensive Care Unit. Ann Am Thorac Soc. 2016;13(12):2197-2206. PMID:27700144.
6. Churpek MM, Yuen TC, Winslow C, Meltzer DO, Kattan MW, Edelson DP. Multicenter Comparison of Machine Learning Methods and Conventional Regression for Predicting Clinical Deterioration on the Wards. Crit Care Med. 2016;44(2):368-374. PMID:26771782.
7. Gulshan V, Peng L, Coram M, et al. Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs. JAMA. 2016;316(22):2402-2410. PMID:27898976.
8. Hakim I, Hathi S, Nair A, Narula T, Bhattacharya J. Electronic health records and the frequency of diagnostic test orders. Am J Manag Care. 2017;23(1):e16-e23. PMID:28141935.
9. Rumball-Smith J, Shekelle PG, Bates DW. Using the Electronic Health Record to Understand and Minimize Overuse. JAMA. 2017;317(3):257-258. PMID:28114561.
10. Schmid F, Goepfert MS, Franz F, et al. Reduction of clinically irrelevant alarms in patient monitoring by adaptive time delays. J Clin Monit Comput. 2017;31(1):213-219. PMID:26621389.
About the Contributor
Vitaly Herasevich, MD, PhD, MSc, FCCM, CPHIMS is Associate Professor of Anesthesiology and Medicine in Department of Anesthesiology at Mayo Clinic. His interest in the area of medical informatics extends back to 1995 with specific concentration on the applied clinical informatics in critical care and science of healthcare delivery. Dr. Herasevich has interest in studying and development clinical syndromic surveillance alerting systems ("sniffers"), clinical data visualization (novel patient-centered EMR) and complex large data warehousing for healthcare predictive and prescriptive analytics as well as outcome reporting. He is author of more than 75 Pubmed cited articles and wrote two editions of book "Computer for Physician".
As a part of education effort Dr. Herasevich serves Clinical Informatics Fellowship program as Associate Program Director, appointed with full faculty privileges in Mayo Graduate School and teaching medical informatics classes to medical school and master students.
He is active within informatics and professional societies serving number of committees.
More information at profile web page - http://www.mayoclinic.org/biographies/herasevich-vitaly-m-d-ph-d/bio-20055468