PubMed · 8935997
The source did not provide an abstract. Follow the original record for more information.
Explore connections, maps & timelines
C Safran, P Gertman. An update on speech recognition.. https://pubmed.ncbi.nlm.nih.gov/8935997/
Cite the original work for its findings. Save a collection to share your selection of sources.
KEEP EXPLORING
OBJECTIVES: 1) To describe the content of medical data obtained from parents' use of free text during a computer interview; 2) To assess whether differences in parents' demographics or experience with computers differentiate their production of free text; and 3) To compare parents' entry of free text with the final physician-generated electronic medical record (EMR) regarding the chief complaint. METHODS: Data entered by parents during a prospective trial of electronic data entry were examined. Content analysis of parents' free text entries examined absolute word count and categorized entries by type of information. Demographic and technology-related variables were analyzed against the type of information and the absolute word count. Two reviewers independently judged agreement between parents' and physicians' data for the chief complaint. RESULTS: Of 100 parents enrolled, 30 entered data for the chief complaint using free text. Amount of free text entered ranged from 1 to 142 words with a median of 8 words. Parents' free text was categorized into 4 types of information: simple descriptive (n = 16), problem-focused (n = 7), descriptive plus a question agenda (n = 2), and descriptive with reference to past medical history (n = 5). Parental demographics and computer-related experience did not predict differential production of free text. Agreement between parents' and physicians' entries for the chief complaint was 80% (95% CI 66-94) [reviewer A] and 73% (95% CI 57-89) [reviewer B]. Seven of 30 parental entries (23%) provided data not contained in the EMR. CONCLUSIONS: When entering a chief complaint during an electronic interview, a majority of parents report descriptive data using short phrases. Good agreement between parents' and physicians' electronic report was demonstrated for the chief complaint.
OBJECTIVE: To describe emergency medicine residents' (EMRs') personal computer (PC) use and educational needs and to compare their perceived and actual PC skills. METHODS: This was a prospective, cross-sectional study. Subjects were all EMRs at seven midwestern Accreditation Council for Graduate Medical Education (ACGME) residency programs. The EMRs completed a questionnaire about their PC use and ability to perform 23 tasks derived from two national retail-training programs. The tasks covered word processing, slide making, and Internet use. The EMRs then took a three-part test performing the skills in the questionnaire. Two independent raters scored the tests. Frequencies with 95% confidence intervals (95% CIs) were calculated for categorical data. Positive and negative predictive values were used to report information comparing residents' performance with their self-assessment of skills. Cohen's kappa was used to test agreement between raters. RESULTS: One hundred twenty-four of 158 (79%) eligible EMRs participated. Since not all participants engaged in all parts of the study, the sample size varies between 121 and 124. One hundred one of 122 (83%; 95% CI = 75 to 89) owned a PC. The EMRs use home PCs a mean of 3.8 hours/week for physician duties and use residency PCs 1.9 hours/week (range 0-20). Ninety-six of 122 (79%; 95% CI = 70 to 86) EMRs reported no formal PC training during residency. Thirty-five percent (43/122; 95% CI = 27 to 44) passed the word-processing test and 50% (62/123; 95% CI = 41 to 60) passed the slide-making test. Reasons for failure were because of errors and not having a presentable product. Thirty-eight of 122 (31%; 95% CI = 23 to 40) failed the literature search, including 33 who said they could perform it. One hundred fifteen of 123 (94%; 95% CI = 88 to 98) EMRs were able to find an Internet address, including ten who stated they could not. Twenty-one percent of the residents who attempted any test (26/124; 95% CI = 14 to 29) passed all three tests. There was no association between year of training and success on the tests (p = 0.374). Thirty-seven of 115 (32%; 95% CI = 24 to 42) EMRs said they had insufficient PC training to meet their physician needs. CONCLUSIONS: Emergency medicine residents have much access to computer technology and possess some computer skills; however, many are unable to produce a usable product or conduct a literature search. Emergency medicine residents have not had sufficient computer training prior to residency. The computer skills of EMRs should be assessed through skills testing rather than self-assessment, and computer training during residency should be improved.
Explore the source record for details and available documents.