Computer-generated patient histories are more thorough, organized, and useful than physician-written histories.
Computer-generated history of present illnesses (HPIs) are more effective at collecting information about patients than are HPIs written by physicians during usual care in gastrointestinal (GI) clinics, according to a study published in the American Journal of Gastroenterology.
Researchers from Cedars-Sinai Medical Center in Los Angeles, CA performed a cross-sectional study to compare the quality of computer-generated (Automated Evaluation of Gastrointestinal Symptoms [AEGIS]) versus physician-documented HPIs in outpatient adult GI clinics.
Seventy-five patients were identified for the study. The patients reported a variety of symptoms, including abdominal pain, heartburn, reflux, nausea, vomiting, constipation, and diarrhea. The physicians initially entered the patient history data into the electronic health record (typing or dictation). The patients then subsequently completed AEGIS on a website called My GI Health.
Forty-eight blinded physicians also participated in the study. Using five-point scales, they assessed HPI quality across six domains:
1. Overall impression
2. Thoroughness
3. Usefulness
4. Organization
5. Succinctness
6. Comprehensibility
The findings showed that the mean overall impression score for computer-generated HPIs was higher than physician HPIs (3.68 versus 2.80), even after adjusting for physician and visit type, location, mode of transcription, and demographics. Computer-generated HPIs were also judged more complete (3.70 versus 2.73), more useful (3.82 versus 3.04), better organized (3.66 versus 2.80), more succinct (3.55 versus 3.17, and more comprehensible (3.66 versus 2.97).
"The computer-generated narratives were of higher quality overall," lead author Christopher V. Almario, MD, gastroenterology fellow, Cedars-Sinai, said in a release.
The researchers said that computers offer a solution to the problem of doctors entering incomplete or inaccurate information into patients' records. They also noted that the technology also frees physicians to focus more on patients during office visits and to catch important bits of information and nonverbal cues that might otherwise be missed.
Where Things Stand with the Radiologist Shortage
June 18th 2025A new report conveys the cumulative impact of ongoing challenges with radiologist residency positions, reimbursement, post-COVID-19 attrition rates and the aging of the population upon the persistent shortage of radiologists in the United States.
How to Successfully Launch a CCTA Program at Your Hospital or Practice
June 11th 2025Emphasizing increasing recognition of the capability of coronary computed tomography angiography (CCTA) for the evaluation of acute and stable chest pain, this author defuses common misperceptions and reviews key considerations for implementation of a CCTA program.
The Reading Room: Artificial Intelligence: What RSNA 2020 Offered, and What 2021 Could Bring
December 5th 2020Nina Kottler, M.D., chief medical officer of AI at Radiology Partners, discusses, during RSNA 2020, what new developments the annual meeting provided about these technologies, sessions to access, and what to expect in the coming year.
Mammography AI Platform for Five-Year Breast Cancer Risk Prediction Gets FDA De Novo Authorization
June 2nd 2025Through AI recognition of subtle patterns in breast tissue on screening mammograms, the Clairity Breast software reportedly provides validated risk scoring for predicting one’s five-year risk of breast cancer.