A study of the difference in volume of information in chief complaint and present illness between electronic and paper medical records.
The introduction of an electronic medical record (EMR) has been
rapidly accelerating in South Korea. The EMR was expected to improve
quality of care, readability, availability, and the quality of data.
However, the reluctance of healthcare providers to use the EMR may have
caused a reduction of information recorded in EMRs. The purpose of this
study was to identify whether there was any loss of information
following the introduction of a narrative text-based EMR in the
recording of chief complaint and present illness in inpatient medical
records. Inpatient medical records of a university hospital were
retrospectively evaluated for one month before and one month after the
introduction of the EMR in June 2006. The volume of information for
chief complaint and present illness was measured by number of words in
Korean and normalised bytes. Change in volume of information was
measured by two-way ANOVA and multiple regression analyses, controlling
for doctors' gender, age, and grade/year of residents,
patients' readmission status, reasons for admission and service
department to assess any effect of the introduction of an EMR. Total
numbers of paper-based medical records (PMRs) and EMRs for analysis were
1,159 and 1,122, respectively. Forty-three doctors participated in the
study. Thirty-one (72%) doctors were less than 30 years of age. Number
of words proved a better outcome measure ([R.sup.2]=.22 for CC,
[R.sup.2]=.36 for PI) than normalised bytes ([R.sup.2]=.18 for CC,
[R.sup.2]=.35 for PI) for measuring volume of information. Results
showed that the volume of information in the chief complaint and present
illness was not decreased after the introduction of the EMR, except when
the dependent variable was measured by number of words in the present
illness. The study showed that the introduction of the EMR did not
reduce the volume of information documented for chief complaint and
present illness in inpatient medical records. However, further studies
are needed to identify how to control the probable loss of information
as showed in present illness measured by number of words.
Keywords (MeSH): Electronic Medical Records; Electronic Clinical Documentation; Hospital Information Systems; Evaluation; Quality of Health Care; Medical Record Systems, Computerised
Supplementary keyword: Health Information Management
Physicians (Information management)
Physicians (Technology application)
Medical care (Quality management)
Medical care (Management)
Noh, Young A.
|Publication:||Name: Health Information Management Journal Publisher: Health Information Management Association of Australia Ltd. Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2012 Health Information Management Association of Australia Ltd. ISSN: 1833-3583|
|Issue:||Date: Feb, 2012 Source Volume: 41 Source Issue: 1|
|Topic:||Event Code: 200 Management dynamics; 260 General services Computer Subject: Company business management; Company systems management; Technology application|
|Product:||Product Code: 8011000 Physicians & Surgeons NAICS Code: 621111 Offices of Physicians (except Mental Health Specialists)|
|Geographic:||Geographic Scope: South Korea Geographic Code: 9SOUT South Korea|
Paper-based medical records (PMR) are an essential tool used by clinicians for documentation and communication in relation to patient care delivery. They are also used for medical research and by hospital administrators for management purposes. However, illegible handwriting, incomplete data, and data fragmentation have caused problems in both quality and continuity of care (Dick, Steen & Detmer 1997). If accurate and complete information is not recorded, communication between doctors can be disrupted during transfer of patients across and within care settings (Jha et al. 2009; Branger et al. 1998).
The electronic medical record (EMR) is considered to be part of the solution to solve the problems identified above by allowing doctors to enter and retrieve data directly. Efficient management of medical information, with or without electronic decision support, also contributes to improvement in the quality of care through reliable medical records (Patel et al. 2000, Chaudhry et al. 2006, Lo et al. 2007, Wright et al. 2009, Andrew & Dick 1995) Nationwide EMR projects are ongoing not only in developing countries (Williams & Boren 2008) but also in several developed countries such as the United States, United Kingdom, Australia and Canada (Jha et al. 2008). The introduction rate of the EMR has been increasing in Korea in recent years, and by 2005, inpatient EMR and ambulatory EMR had been introduced into 19.6% and 20.7% of Korean hospitals, respectively (Chae 2005).
Physician data entry is known to be one of the major issues for inputting data and utilisation of EMRs (Retchin & Wenzel 1999; Gilbert 1998; Kaplan 1994). If the quality of data in the EMR is poor this may compromise the patient and may threaten the quality of care (Chiang et al. 2003). This is particularly true in the earlier stages of implementing the EMR, when doctors who are familiar with a PMR might not enter data in the EMR (McDonald 1997; Embi et al. 2004; O'Connell et al. 2004). Unlike the PMR, entering data into the EMR is complex and onerous to some. Typing with keyboards has itself its own limitations, and reuse of pre-saved sentences or copy-and-pasting other parts of the record are major sources of errors in the EMR (Hammond et al. 2003). In order to overcome these problems, some EMRs have used a structured data entry system for the interim history, risk assessment, developmental screening and guidance sections of the record, which has resulted in better documentation in the EMR when compared with the PMR (Adams, Mann & Bauchner 2003; Johnson & Cowan 2002; Shiffman, Brandt & Freeman 1997; Bauchner Kanegaye et al. 2005). However, the patient's chief complaint and present illness were freely documented in the legacy PMR and were recorded with narrative text rather than the structured data as seen in the EMR. The use of narrative text may cause loss of information in the EMR (Patel, Arocha & Kushniruk 2002).
The aim of this study was to evaluate whether there is loss of information in the recording of chief complaint and present illness in inpatient medical records before and after the implementation of the EMR.
The study hospital
The setting for study was a Korean teaching hospital with 700 beds. The hospital has approximately 300 doctors in total, including interns and residents, with annual discharges of more than 24,000. A computerised physician order entry (CPOE) system was developed in-house and has been in use since 1995; since the introduction of the EMR in July 2006, no inpatient clinical documents have been written on paper except for the Emergency Room and Intensive Care Unit. The EMR has an integrated user interface to view all the patient information in a screen. The picture archiving system (PACS) is also connected to the EMR and diagnostic images are automatically retrieved when the patient record is open. There were two types of recording in the EMR: template based and narrative text based. The templates were authored, added and modified by each department of the hospital. The Microsoft Word tool was used for the narrative text-based recording. This tool also supports special characters and images.
[FIGURE 1 OMITTED]
Medical records of inpatients discharged one month before and one month after the implementation of the EMR were reviewed and the amount of information in the chief complaint and present illness was evaluated. The chief complaint and present illness were entered in narrative text in the EMR. All medical records with the following criteria were excluded: (a) incomplete admission notes; (b) records written on both paper and EMR; (c) medical records of patients admitted before May 2006 and discharged in May 2006; (d) medical records of normal babies because the content in chief complaint and present illness is empty; and (e) all medical records of psychiatric patients because in these the distinction between chief complaint and present illness was not clear. The total number of records that satisfied the exclusion criteria was 1,159 in May (PMR) and 1,122 in June (EMR) (total n=2,281) (See Figure 1).
One registered Health Information Manager (HIM) typed the chief complaint and the present illness recorded in the paper-based medical records in Microsoft Visual Foxpro. In order to capture the exact information in the medical documents from the doctors, and to make validation easy, syntactic errors such as incorrect sentences, spelling errors, and inappropriate abbreviations were typed as they had been entered. The chief complaint and present illness with extra spaces and word breaks in EMR were also imported exactly as they appeared. Sample characteristics collected included: department type (e.g. medical or surgical department); reason for admission to differentiate diseased from injured patients; type of admission by first admission or readmission; length of stay; age; gender; and residential grade/year of doctors.
The volume of information in the chief complaint and present illness were measured by both the number of words and the normalised bytes. When counting the number of words, postpositions that are attached at the end of words in Korean were not counted as a word because the Korean syntax does not allow a blank between noun and the postpositions, therefore they were allowed to be omitted without loss of meaning. The normalised bytes were calculated as a sum of bytes in the expression. Duplicates of blanks and carriage return with line feed were counted as a blank.
Documentation of medical records in Korean hospitals is generally completed by first or second year resident doctors who have initially seen the patient after admission in an assigned department. The total number of working residents during the study period was 115. Forty-three of the residents participated in the study.
Unpaired t-tests were used to compare volume of information according to type of medical record, type of department, disease status, and readmission. A two-way ANOVA was used to test the difference in volume of information according to the type of medical record, type of admission, reason for admission and department. Multiple regression analyses were used to find the effect of EMR on volume of information. All factors that influenced the outcome measures in ANOVA, the participating doctor's age, gender and grade of residents were included in the multiple regression model as control factors. SAS version 9.12 was used for statistical analyses.
The sample included 1,159 PMR and 1,122 in EMR. There were no significant differences in type of admission, reason for admission, or type of department according to the type of medical record (Table 1). Forty-three doctors participated in the study. Only 12 participants (28%) were above 30 years of age. The number of male doctors was 28 (65%). Thirty-seven (86%) doctors were first- or second-year residents.
Volume of information by type of medical record
The average number of words in present illness decreased significantly from 25.95 in the PMR to 24.70 in the EMR (p=.03). However, the average number of words in chief complaint and normalised bytes in both chief complaints and present illness showed no statistical differences (Table 2).
Type of medical record had no effect on volume of information in the chief complaint when controlling for type of admission, reason for admission, and department. However, type of admission (p<.001) and department (p<.001) influenced volume of information regardless of outcome measure (Table 3).
The number of words in present illness was decreased in the EMR when the type of admission (p = .046) and department (p = .041) were controlled. However, the type of medical record had no effect on the volume of information measured by normalised bytes, regardless of controlling factors. All controlling factors influenced volume of information in every case (Table 4).
Effect of EMR on volume of information controlling extraneous factors
The multiple regression model for chief complaint had better [R.sup.2] when the volume of information was measured as the number of words ([R.sup.2]=.22) compared to normalised bytes ([R.sup.2]=.18). In both models, the type of medical record had no influence on the volume of information.
The volume of information was significantly decreased by readmission (p = .001, p = .001), medical department (p<.001), male doctor (p<.001) and length of stay (p<.01) and increased by injury (p=.01) regardless of outcome measures.
The model for present illness as with the chief complaint had a better [R.sup.2] when the volume of information was measured as the number of words ([R.sup.2]=.36) rather than normalised bytes ([R.sup.2]=.35). 1.02 words as the volume of information decreased by EMR (p=.03). However, EMR had no effect on the volume of information in normalised bytes.
Results for the other factors indicated there was a decrease in the volume of information in the surgical department (p<.001), also for female doctors (p<.002, p <.001), and disease (p<.001), with an increase in length of stay (p<.001). The readmission factor decreased in the volume of information when measured as normalised bytes (p=.005). As the age of the doctor increased, the number of words decreased (p=.04) but normalised bytes did not show a significant change (Table 5).
Volume of information written by doctors can vary according to each doctor's documentation in using the EMR (Treweek 2003). Most of the medical records evaluated in this study were documented by first- and second-year residents. Residents in Korea change their rotations and training institutions regularly. It was decided to restrict the study period to one month before and one month after the introduction of the EMR, in order to keep the composition of resident staff, which changes regularly, constant before and after the introduction of the EMR. We expected the composition of doctors in the study to have limited influence on the observed difference in volume of information following the implementation of the EMR. However, had the observation period been longer, the true effect of the EMR could have been extrapolated more precisely. In addition, an observation period of two months in an inpatient setting is not long enough to enable results to be generalised other hospitals.
The type of medical record and the department for chief complaints showed significant interactions in the two-way ANOVA. However, we accepted the results in spite of the interactions because the average number of words in both the EMR and the PMR of surgical department inpatients (10.16 and 9.62 respectively) had a greater average number of words than inpatients in the medical department (5.86 and 6.98 respectively). The same interpretations were applied to interactions found in present illness.
Two outcome measures were explored in multiple regression models for both chief complaint and present illness; the total number of words and normalised bytes. When the dependent variable was measured by the total number of the words, the chief complaint and present illness ([R.sup.2]=.22, [R.sup.2]=.36) provided a better explanation of the model than when it was measured by normalised bytes ([R.sup.2]=.18, [R.sup.2]=.35).
There was no significant difference in the volume of information between the PMR and the EMR except when the present illness was measured by number of words. There seem to be several reasons for this: the younger age of the doctors, the use of templates, and the increased accessibility of the medical records. The number of words in the present illness was consistently decreased in the EMR from t-test to multiple regression analyses. Because the words as terms are representations of concepts, the number of concepts to describe a patient's present illness was decreased (Patel et al. 2000; Hamilton et al. 2003). However, the average bytes per word was increased from 6.05 to 6.33 (p<.01) in the EMR (i.e. the number of characters composing each word was significantly increased in EMR). Further studies are needed to clarify whether both a decrease in the number of words and an increase in the number of characters in a word have a positive or negative effect on the quality of medical records.
While other studies have examined documentation in electronic and paper-based health records (Mikkelsen & Aasly 2000; Stausberg et al. 2003), our study is the first to compare the volume of documentation in free text input of chief complaint and present illness between electronic and paper-based medical records. When a patient's subjective information was entered with narrative text, more than twice the number of chief complaints were entered in the EMR, and fewer present illnesses were entered in the EMR than the PMR (Patel, Arocha & Kushniruk 2002). When the medical records are inputted with narrative text, a higher volume of information was entered (Apkon & Singhaviranon 2001; Callen, Alderton & McIntosh 2008) and less time was required (Apkon & Singhaviranon 2001) than the PMR. Even when comparing results of the PMR with the EMR equipped with structured data entry format, the volume of information in EMR was more than the PMR (Tang, Larosa & Gorden 1999; Rosenbloom et al. 2004; Roukema et al. 2006; Tsai & Bond 2008).
Limitations of the study
This study was conducted in one Korean teaching hospital. If the same study was to be conducted in other hospitals, the characteristics of the EMR and the settings of different hospitals might provide different results. The medical records are written in Korean. Even though English terms are used together with Korean in the medical records, the syntax of sentences in this study was based on the Korean language (e.g. the combination of preposition and noun as objective in English is counted as a word composed of a noun with postposition in Korean). These issues should be considered when the results of this study are compared to other studies with non-Korean languages.
The volume of information in chief complaints was the same in both the PMR and EMR regardless of outcome measure. It also remained the same when measured by the normalised bytes. However, the volume of information in present illness decreased by 1.02 words when measured by number of words. Therefore, we can conclude that the volume of information in the EMR was not decreased. Further studies are needed to explore how to control the probable loss of information in the recording of present illness as measured by the number of words and how to improve the quality of electronic documentation of care generally.
Adams, W.G., Mann, A.M. and Bauchner, H. (2003). Use of an electronic medical record improves the quality of urban pediatric primary care. Pediatrics 111: 626-32.
Andrew, W.F. and Dick, R.S. (1995). Applied information technology: a clinical perspective. Feature focus: the computer-based patient record (Part 2). Computers in Nursing 13: 118-22.
Apkon, M. and Singhaviranon, P (2001). Impact of an electronic information system on physician workflow and data collection in the intensive care unit. Intensive Care Medicine 27: 122-30.
Branger, PJ., van't Hooft, A., van der Wouden, J.C., Duisterhout, J.S. and van Bemmel, J.H. (1998). Shared care for diabetes: supporting communication between primary and secondary care. Studies in Health Technology and Informatics 52(1): 412-6.
Callen, J.L., Alderton, M. and McIntosh, J. (2008). Evaluation of electronic discharge summaries: a comparison of documentation in electronic and handwritten discharge summaries. International Journal of Medical Informatics 77: 613-620.
Chae, Y.M. (2005). Report on Hospital Information System Survey. Journal of Korean Society of Medical Informatics 11: 323-336.
Chaudhry, B., Wang, J., Wu, S., Maglione, M., Mojica, W., Roth, E., Morton, S.C. and Shekelle, P.G. (2006). Systematic review: impact of health information technology on quality, efficiency, and costs of medical care. Annals of Internal Medicine 144: 742-52.
Chiang, M.F., Cao, H., Sharda, P, Hripcsak, G. and Starren, J.B. (2003). An experimental system for comparing speed, accuracy, and completeness of physician data entry using electronic and paper methods. AMIA Annual Symposium Proceedings: 812.
Dick, R., Steen, E. and Detmer, D. (1997). The computer-based patient record: an essential technology for health care. National Academies Press, Washington, DC.
Embi, PJ., Yackel, T.R., Logan, J.R., Bowen, J.L., Cooney, T.G. and Gorman, P.N. (2004). Impacts of computerized physician documentation in a teaching hospital: perceptions of faculty and resident physicians. Journal of the American Medical Informatics Association 11: 300-9.
Gilbert, J.A. (1998). Physician data entry: providing options is essential. Health Data Management 6: 84-6, 88, 90-2.
Hamilton, W.T., Round, A.P., Sharp, D. and Peters, T.J. (2003). The quality of record keeping in primary care: a comparison of computerised, paper and hybrid systems. British Journal of General Practice 53: 929-33; discussion 933.
Hammond, K., Helbig, S., Benson, C. and Brathwaite-Sketoe, B. (2003). Are electronic medical records trustworthy? Observations on copying, pasting and duplication. AMIA Annual Symposium Proceedings 2003: 269.
Jha, A.K., Desroches, C.M., Campbell, E.G., Donelan, K., Rao, S.R., Ferris, T.G., Shields, A., Rosenbaum, S. and Blumenthal, D. (2009). Use of electronic health records in U.S. hospitals. The New England Journal of Medicine 360: 1628-38.
Jha, A.K., Doolan, D., Grandt, D., Scott, T. and Bates, D.W. (2008). The use of health information technology in seven nations. International Journal of Medical Informatics 77: 848-54.
Johnson, K.B. and Cowan, J. (2002). Clictate: a computer-based documentation tool for guideline-based care. Journal of Medical Systems 26: 47-60.
Kanegaye, J.T., Cheng, J.C., McCaslin, R.I., Trocinski, D. and Silva, P.D. (2005). Improved documentation of wound care with a structured encounter form in the pediatric emergency department. Ambulatory Pediatrics 5: 253-7.
Kaplan, B. (1994). Reducing barriers to physician data entry for computer-based patient records. Topics in Health Information Management 15: 24-34.
Lo, H.G., Newmark, L.P., Yoon, C., Volk, L.A., Carlson, V.L., Kittler, A.F., Lippincott, M., Wang, T. and Bates, D.W. (2007). Electronic health records in specialty care: a time-motion study. Journal of the American Medical Informatics Association 14: 609-15.
McDonald, C.J. (1997). The barriers to electronic medical record systems and how to overcome them. Journal of the American Medical Informatics Association 4: 213-21.
Mikkelsen G. and Aasly J. (2000). Concordance of information in parallel electronic and paper based patient record. International Journal of Medical Informatics 63:123-31.
O'Connell, R.T., Cho, C., Shah, N., Brown, K. and Shiffman, R.N. (2004). Take note(s): differential EHR satisfaction with two implementations under one roof. Journal of the American Medical Informatics Association 11: 43-9.
Patel, V.L., Arocha, J.F. and Kushniruk, A.W. (2002). Patients' and physicians' understanding of health and biomedical concepts: relationship to the design of EMR systems. Journal of Biomedical Informatics 35: 8-16.
Patel, V.L., Kushniruk, A.W., Yang, S. and Yale, J.F. (2000). Impact of a computer-based patient record system on data collection, knowledge organization, and reasoning. Journal of the American Medical Informatics Association 7: 569-85.
Retchin, S.M. and Wenzel, R.P (1999). Electronic medical record systems at academic health centers: advantages and implementation issues. Academic Medicine 74: 493-8.
Rosenbloom, S.T., Kiepek, W., Belletti, J., Adams, P., Shuxteau, K., Johnson, K.B., Elkin, P.L. and Shultz, E.K. (2004). Generating complex clinical documents using structured entry and reporting. Studies in Health Technology and Informatics 107: 683-687.
Roukema, J., Los, R.K., Bleeker, S.E., Van Ginneken, A.M., Van Der Lei, J. and Moll, H.A. (2006). Paper versus computer: feasibility of an electronic medical record in general pediatrics. Pediatrics 117: 15-21.
Shiffman, R.N., Brandt, C.A. and Freeman, B.G. (1997). Transition to a computer-based record using scannable, structured encounter forms. Archives of Pediatrics & Adolescent Medicine 151: 1247-53.
Stausberg J., Koch D., Ingenerf J. and Betzler M. (2003). Comparing paper-based with electronic patient records: Lessons learned during a study on diagnosis and procedure codes. Journal of the Medical Informatics Association 10: 470-477.
Tang, P.C., Larosa, M.P. and Gorden, S.M. (1999). Use of computer-based records, completeness of documentation, and appropriateness of documented clinical decisions. Journal of the American Medical Informatics Association 6: 245-51.
Treweek, S. (2003). The potential of electronic medical record systems to support quality improvement work and research in Norwegian general practice. BMC Health Services Research 3: 10.
Tsai, J. and Bond, G. (2008). A comparison of electronic records to paper records in mental health centers. International Journal for Quality in Health Care 20: 136-43.
Williams, F. and Boren, S.A. (2008). The role of the electronic medical record (EMR) in care delivery development in developing countries: a systematic review. Informatics in Primary Care 16: 139-45.
Wright, A., McGlinchey, EA., Poon, E.G., Jenter, C.A., Bates, D.W. and Simon, S.R. (2009). Ability to generate patient registries among practices with and without electronic health records. Journal of Medical Internet Research 11: e31.
Yookyung Boo PhD
Eulji University of Korea
College of Health Industry
Department of Healthcare Management
Young A Noh MPH
The Catholic University of Korea
Graduate School of Public Health
Min-gyung Kim MBA
The Catholic University of Korea
College of Medicine
Department of Preventive Medicine
Sukil Kim MD, PhD, MSc
Department of Preventive Medicine
The Catholic University of Korea, College of Medicine
505 Banpo-dong Seocho-gu, Seoul 137-701 Korea
Tel: +82 2 2258 736 7
Fax: +82 2 532 3820
Mobile: +82 10 9911 0605
Table 1: General characteristics of medical records according to type of medical records. FACTORS TYPE OF MEDICAL RECORD (%) PMR EMR TOTAL Type of admission First admission 734(51.0%) 705(49.0%) 1,439(100%) Readmission 425(50.5%) 417(49.5%) 842(100%) Reason for admission Disease 1,067(50.9%) 1,029(49.1%) 2,096(100%) Injury 92(49.7%) 93(50.3%) 185(100%) Type of department Medical 680(51.6%) 639(48.4%) 1,319(100%) Surgical 479(49.8%) 483(50.2%) 962(100%) Length of stay 4.98 days 5.05 days Total 1,159(50.8%) 1,122(49.2%) 2,281(100%) PMR: Paper-based medical record, EMR: Electronic medical record Table 2: Average volume of information according to type of medical records and outcome measures OUTCOME TYPE OF MEDICAL RECORD MEASURES (MEAN [+ or -] SD) CONTENT PMR EMR Number of words Chief Complaints 8.07 [+ or -] 5.88 7.71 [+ or -] 5.81 Present Illness 25.95 [+ or -] 13.84 24.70 [+ or -] 14.16 Normalized bytes Chief Complaints 54.68 [+ or -] 44.21 54.86 [+ or -] 45.25 Present Illness 162.80 [+ or -] 98.22 162.46 [+ or -] 100.78 OUTCOME MEASURES CONTENT t p Number of words Chief Complaints -1.47 .14 Present Illness -2.15 .03 Normalized bytes Chief Complaints .10 .92 Present Illness -.08 .93 PMR: Paper-based medical record, EMR: Electronic medical record Table 3: Volume of information in chief complaint according to related factors by outcome measures OUTCOME MEASURES (MEAN [+ or -] SD) TYPE OF NUMBER MEDICAL OF FACTORS (A) RECORD n WORDS Type of admission First admission EMR 705 8.07 [+ or -] 5.61 PMR 734 8.67 [+ or -] 6.13 Readmission EMR 417 7.10 [+ or -] 6.09 PMR 425 7.04 [+ or -] 5.28 [P.sub.A] < .001, [P.sub.b] = .29, [P.sub.A*B] = .19 Reason for Admission Disease EMR 1,029 7.65 [+ or -] 6.00 PMR 1,067 8.07 [+ or -] 6.09 Injury EMR 93 8.35 [+ or -] 2.73 PMR 92 8.03 [+ or -] 2.44 [P.sub.A] = .46, [P.sub.B] = .91, [P.sub.A*B] = .41 Department Medical EMR 639 5.86 [+ or -] 4.64 PMR 680 6.98 [+ or -] 4.91 Surgical EMR 483 10.16 [+ or -] 6.27 PMR 479 9.62 [+ or -] 6.75 [P.sub.A] < .001, [P.sub.B] = .22, [P.sub.A*B] < .001 OUTCOME MEASURES (MEAN [+ or -] SD) TYPE OF MEDICAL NORMALISED FACTORS (A) RECORD BYTES Type of admission First admission EMR 57.83 [+ or -] 43.81 PMR 58.86 [+ or -] 45.16 Readmission EMR 49.84 [+ or -] 47.21 PMR 47.47 [+ or -] 41.58 [P.sub.A] < .001, [P.sub.b] = .73, [P.sub.A*B] = .38 Reason for Admission Disease EMR 54.92 [+ or -] 46.75 PMR 54.65 [+ or -] 45.74 Injury EMR 54.15 [+ or -] 22.9 PMR 55.07 [+ or -] 19.03 [P.sub.A] = .96, [P.sub.B] = .93, [P.sub.A*B] = .86 Department Medical EMR 43.93 [+ or -] 41.43 PMR 50.07 [+ or -] 41.81 Surgical EMR 69.32 [+ or -] 46.06 PMR 61.22 [+ or -] 46.68 [P.sub.A] < .001, [P.sub.B] = .60, [P.sub.A*B] < .001 [P.sub.A] : p-value for each factor A, [P.sub.B] : p-value for factor B (type of medical records) [P.sub.A*B] : p-value for interaction between factor A and factor B Table 4: Volume of information in present illness according to related factors by outcome measures OUTCOME MEASURES (MEAN [+ or -] SD) TYPE OF NUMBER MEDICAL OF FACTORS (A) RECORD (B) n WORDS Type of admission First admission EMR 705 24.02 [+ or -] 14.16 PMR 734 25.45 [+ or -] 14.25 Readmission EMR 417 25.83 [+ or -] 13.59 PMR 425 26.82 [+ or -] 13.01 [P.sub.A] = .01, [P.sub.B] = .046, [P.sub.A*B] = .71 Reason for Admission Disease EMR 1,029 25.66 [+ or -] 14.26 PMR 1,067 26.83 [+ or -] 13.93 Injury EMR 93 13.99 [+ or -] 6.72 PMR 92 15.82 [+ or -] 7.29 [P.sub.A] < .001, [P.sub.B] = .15, [P.sub.A*B] = .75 Department Medical EMR 639 31.91 [+ or -] 12.57 PMR 680 32.86 [+ or -] 12.34 Surgical EMR 483 15.14 [+ or -] 9.85 PMR 479 16.15 [+ or -] 9.15 [P.sub.A] < .001, [P.sub.B] = .04, [P.sub.A*B] = .96 OUTCOME MEASURES (MEAN [+ or -] SD) TYPE OF MEDICAL NORMALISED FACTORS (A) RECORD (B) BYTES Type of admission First admission EMR 157.62 [+ or -] 102.22 PMR 161.20 [+ or -] 102.15 Readmission EMR 170.63 [+ or -] 97.89 PMR 165.57 [+ or -] 91.08 [P.sub.A] = .04, [P.sub.B] = .86, [P.sub.A*B] = .32 Reason for Admission Disease EMR 169.25 [+ or -] 101.53 PMR 168.69 [+ or -] 99.07 Injury EMR 87.25 [+ or -] 48.46 PMR 94.53 [+ or -] 51.57 [P.sub.A] < .001, [P.sub.B] = .65, [P.sub.A*B] = .60 Department Medical EMR 212.70 [+ or -] 90.31 PMR 210.06 [+ or -] 90.04 Surgical EMR 95.98 [+ or -] 71.07 PMR 95.70 [+ or -] 64.53 [P.sub.A] < .001, [P.sub.B] = .67, [P.sub.A*B] = .73 [P.sub.A] : p-value for each factor A, [P.sub.B] : p-value for factor B (type of medical records) [P.sub.A*B] : p-value for interaction between factor A and factor B Table 5: Effect of EMR on the volume of information in chief complaint and present illness controlling related factors by outcome measures NUMBER OF NORMALISED WORDS BYTES INDEPENDENT VARIABLES [beta] P [beta] P Chief complaint EMR(PMR=0) -.17 .43 1.75 .31 Readmission(First admission=0) -.65 .004 -5.72 .001 Medical dept.(Surgical dept.=0) -4.71 <.001 -27.03 <.001 Male doctor(Female doctor=0) -3.58 <.001 -25.71 <.001 Injury(disease=0) 1.51 <.001 10.87 .001 Age of doctor -.40 <.001 -3.61 <.001 Length of stay .07 0.01 .53 .01 ([R.sup.2]=.22) ([R.sup.2]=.18) Present illness EMR(PMR=0) -1.02 .03 .53 .88 Readmission(First admission=0) -.96 .05 -10.00 .005 Medical dept.(Surgical dept.=0) 16.71 <.001 119.52 <.001 Man doctor(Woman doctor=0) 1.68 <.002 24.86 <.001 Injury(disease=0) 4.13 <.001 30.55 <.001 Age of doctor -.23 .04 -.17 .83 Length of stay .22 <.001 1.74 <.001 ([R.sup.2]=.36) ([R.sup.2]=.35)
|Gale Copyright:||Copyright 2012 Gale, Cengage Learning. All rights reserved.|