Does a Hybrid Electronic-Paper Environment Impact on Health Professional Information Seeking?

Transcription

1 ehealth Beyond the Horizon Get IT There S.K. Andersen et al. (Eds.) IOS Press, Organizing Committee of MIE All rights reserved. 505 Does a Electronic- Environment Impact on Health Professional Information Seeking? E. M. BORYCKI RN PhD a,1 and L. LEMIEUX-CHARLES PhD b a School of Health Information Science, University of Victoria, British Columbia, Canada b Department of Health Policy, Management and Evaluation, University of Toronto Toronto, Ontario, Canada Abstract. The purpose of this study was to investigate the effects of a hybrid electronic-paper patient record environment upon health professional information seeking (i.e. amount of information accessed, choice of key sources of information, type of information and use of information seeking tactics). A within group, laboratory, experimental study was conducted using two simulation environments (i.e. a paper patient record and a hybrid or electronic-paper environment). Thirtyfive novice nurses participated in this within group, laboratory based study. Findings revealed significant differences between the paper and hybrid environments in terms of their effects upon information seeking. Subjects: (1) accessed less data in the hybrid than the paper environment, (2) accessed more non-electronic sources than electronic sources of information in the hybrid environment, and (3) used more passive information seeking tactics in the hybrid than the paper environment. Findings from the cued recall data revealed subjects experienced increased cognitive load in the hybrid environment. Implications for the design of hybrid environments are discussed. Keywords. EPR-CPR-EMR, Health professional workstation, human interfaces Introduction Health services and health informatics researchers are recognizing that the ability of clinicians to acquire, interpret, and use patient information is a critical aspect of clinician [1-3], team [5] and organizational decision making [4,5]. Yet, even as the process of information seeking remains a critical aspect of health care work [10], clinicians spend a considerable amount of their time seeking information [1,11]. Research suggests up to 70% of clinician information needs relevant to patient care decision making are unmet [1, 6]. Furthermore, up to 42% of workers accidentally use the wrong information, and 57% of workers go to many information sources to obtain needed information [4, p. 1]. Given the urgency associated with providing timely and 1 Corresponding Author: E. M. Borycki, School of Health Information Science, University of Victoria, P. O. Box 3050 STN CSC, Victoria, British Columbia, V8W 3P5, Canada; emb@uvic.ca

2 506 E.M. Borycki and L. Lemieux-Charles / A Electronic- Environment Impact relevant information to clinicians, information seeking has become an important area of concern for health care organizations. 1. Literature Review Health care organizations are investing in information technology (IT) to support clinician information seeking [2, 4]. A number of differing types of IT are being deployed in health care organizations to support clinician information seeking, among them the electronic patient record (EPR) [2, 4, 7]. Health care organizations are implementing EPRs incrementally (i.e. implementing components of the EPR over a period of years) [7, 8]. Even in technologically advanced countries where greater than 50% of EPR components have been implemented, the evidence suggests that a high proportion of organizations are implementing hybrid (i.e. electronic-paper) environments in their progression towards a full EPR. This has led some researchers to conclude that a full EPR may be difficult to implement, that paper may continue to exist, and that hybrid environments are best suited to the clinical setting [9]. As a result, some researchers have concluded the vision of a full EPR will be replaced by a hybrid environment. Similar conclusions have been made by researchers in other industries (e.g. banking) who have noted that it may not be possible to migrate to a fully paperless environment [10]. The discussion of hybrid environments in the research literature and their impact upon clinician information seeking has been limited. In studies, where there has been some attention drawn to aspects of information seeking (e.g. information needs) involving patient records, only full EPRs have been studied and not in the context of a clinical environment [11-12]. Furthermore, research has largely been descriptive, outlining the information needs of clinicians rather than studying the effects of the EPR upon information seeking. These studies have also failed to include other sources of information that are present in a typical clinical environment (e.g. textbooks, health professionals). Lastly, no attempts have been made to test for differences in clinical behaviour between paper and hybrid environments. Therefore, the purpose of this study was to investigate the effects of a hybrid environment upon health professional (i.e. clinician) information seeking (i.e. amount of information accessed, choice of key sources of information, types of information and use of information seeking tactics). 2. Methods 2.1. Sample and Materials Thirty-five novice nurses participated in a laboratory based, experimental study (power = 92%; effect size = 93 for the primary outcome variable: information accessed). In this study, two representative, hospital-based environments were simulated. One was paper-based (i.e. all of the information in the environment was provided to subjects in paper form). The paper environment represented a nursing station where subjects are provided with paper-based information (i.e. a paper patient record, health professional reference texts, a communication book and a list of health professional colleagues that could be contacted by phone). The second simulated environment or the hybrid

3 E.M. Borycki and L. Lemieux-Charles / A Electronic- Environment Impact 507 environment (i.e. electronic-paper) provided subjects with access to a patient record, where part of the patient record was electronic and accessible via a workstation and the other part was paper-based and accessible via a paper chart. Patient demographic information, doctor s orders and documentation records were in the EPR and all other parts of the patient record and sources of information were paper-based (e.g. reference texts, a communication book and a list of health professional colleagues). Two patient records were used in the study (i.e. A and B). The patient records were taken from a hospital during a time period when the hospital was fully paperbased. The patient records represented patients who were early in their hospital stay and had been admitted to a medical unit for one day (i.e. there was a high degree of uncertainty associated with their conditions to stimulate information seeking) [14]. Both the patient records represented older adults diagnosed with multiple chronic and acute illnesses. The patient records were selected by a Masters prepared, expert nurse who deemed them to be typical patients a novice nurse would encounter. Each record was in paper form and was also recreated in hybrid form for the hybrid environment. Both environments (i.e. the paper and hybrid) were evaluated for their ecological validity (by six experts: three health informatics and three clinician) and were deemed typical of the types of environments that are present in North American hospitals [8] Study Design A within group, laboratory, experimental study was conducted using the two simulated environments (i.e. paper and hybrid). Subjects were asked to perform the task of assessment and planning of a new patient s care involving the review of a new patient s record at the beginning of a nursing shift. Assessment and planning involving a review of a patient s chart stimulates nurse information seeking [13]. Subjects were presented with a paper and then a hybrid environment. Research has demonstrated the presence of long term carry over effects following the use of EPRs [12]. Therefore, subjects were block randomized to receiving either patient A s or B s information first (i.e. AB or BA) to prevent learning, order effects and effects associated with differential difficulty between the patients. Subjects were video and audio taped thinking aloud in both environments. Subjects participated in a cued recall session following the simulations. During the cued recall session, subjects were asked to clarify their verbalizations and/or their actions [14] Data Analysis Protocol analysis was used to code the data [14]. Think aloud data from the simulations was transcribed, and divided into segments according to natural pauses in speech. The data (i.e. audio and video) was then coded using Transana according to information seeking concepts defined by Miller and Jablin s [3] and Johnson s [2] work (See Table 1). The unit of analysis for the data consisted of utterances, words, phrases, sentences, paragraphs, and video frames (i.e. the smallest unit of information that could be understood [11, 14] and could be represented by concepts defined by Miller and Jablin [3] and Johnson [2]). Ericsson and Simon refer to this coding approach as Model Based Coding [14]. A Kappa of was calculated, indicating high in-coder agreement. The coded audio and video data from the simulations was then quantified [12]. The numbers of occurrences of each code were counted for each subject in both environments. Paired sample t-tests were conducted to test for differences between the

4 508 E.M. Borycki and L. Lemieux-Charles / A Electronic- Environment Impact environments for the amount of information accessed, type of information, sources of information and use of information seeking tactics (i.e. aspects of information seeking). This method has been used by researchers who have quantified think aloud data to conduct inferential statistics [e.g , 14]. Table 1. Coding Scheme for Theoretically Based Concepts Code Definition Data Accessed Item of data attended to or heeded by the subject [14]. Types of Information Referent Relational Appraisal Information that tells the worker what is required of him or her to function successfully in their work [14, p.48]. Information that tells the worker about the nature of their relationship with others in the workplace [14, 1978]. Information that tells the worker how well they are functioning at work [14, p.48] Information Seeking Tactics Active Tactics Passive Tactics Active information seeking tactics involve direct interaction with the information source or target (i.e. overt questions) [2] Passive information seeking tactics involve indirect interaction with the information source (i.e. testing, indirect questioning, surveillance, observing, disguising conversations, third party questioning [2]. Sources of Information A paper source of information (e.g. a paper patient record, a text book) [7] Electronic An electronic source of information (e.g. an EPR) [7]. 3. Results There were a number of statistically significant differences in terms of subject information seeking between the two environments. In terms of the primary outcome variable (i.e. information accessed), subjects accessed less information in the hybrid than the paper environment. The eta squared statistic for information accessed was calculated as 0.3 which indicated a small to moderate effect size. Paired t-tests were conducted to test for differences between the paper and hybrid environment for subject use of referent, relational and appraisal information. There were no statistically significant differences between the two environments in terms of the use of these types of information. Subjects accessed more non-electronic than electronic sources of information in the hybrid environment. Tests for differences in subject use of active and passive information seeking tactics in both environments were also conducted. There were no statistically significant differences in active information seeking between the two environments. Alternatively, there was a statistically significant difference in passive information seeking. Passive information seeking tactics were used more frequently in the hybrid than the paper environment (See Table 2 for more details).

5 E.M. Borycki and L. Lemieux-Charles / A Electronic- Environment Impact 509 Table 2. Results Aspect of Information Seeking Environment Mean P value Data Accessed *** Types of Information Referent Relational Appraisal Information Seeking Tactics Active tactics Passive Tactics *** Sources of Information Electronic versus Non-electronic Electronic Non-Electronic *** *** indicates statistical significance Cued recall data suggested 25 subjects (i.e. 74%) experienced increased cognitive load or difficulty integrating information from differing sources of information (i.e. the paper patient record, the EPR, textbooks, communication book and/or a list of health professionals that could be contacted) in the hybrid condition. Subjects described the experience of obtaining information from information sources in the hybrid environment as frustrating and confusing when going back and forth between the EPR and the paper record and other paper-based information resources; for example: I would like to have all of my information in one place then I am not transferring back and forth. It was frustrating to have to check on the computer. Check everywhere on the computer. Oh it is not there. Then go back to the chart and oh its not there either and then try it again [29, Cued Recall] In summary subjects accessed more information in the paper environment. Subjects also accessed more non-electronic sources of information than electronic sources in the hybrid environment. Lastly, passive information seeking tactics were used more frequently in the hybrid environment than the paper environment by subjects. 74% of subjects experienced cognitive load in the hybrid environment. 4. Conclusion The purpose of this study was to determine if hybrid environments can influence aspects of clinician information seeking. In order to determine the effects of a hybrid

6 510 E.M. Borycki and L. Lemieux-Charles / A Electronic- Environment Impact environment upon information seeking, two typical hospital environments were simulated. Study findings revealed the hybrid environment altered the amount of information accessed by information seekers. Subjects accessed less information in the hybrid than the paper environment. Cued recall data suggested subjects may have experienced increased cognitive load arising from the need to consult more sources of information in the hybrid environment. This may have placed excess demands upon subjects cognitive processes such as memory causing them to access less information. The implications of this work are significant. Health informaticians will need to take into consideration the way they implement and train health professionals to use hybrid environments as such environments may lead to increased cognitive load and the use of less information involving patient care. Future research needs to explore how the order of implementation of the components of the EPR impact upon cognitive load and clinician ability to seek information in hybrid health care environments. References [1] L. M. Currie, M. Graham, M. Allen, S. Bakken, V. Patel, J. J. Cimino, Clinical information needs in context: An observational study of clinicians while using a clinical information system. In M. A. Musen, C. P. Friedman, J. M. Teich (Eds.), AMIA 2003 Symposium: Foundations of Informatics: Vol. 27. (pp ). American Medical Informatics Association, Washington, D. C., [2] J. D. Johnson, Information seeking: An organizational dilemma, Quorum, Westport, Connecticut, [3] V. D. Miller, F. M. Jablin, Information seeking during organizational entry: Influences, tactics and a model of process, The Academy of Management Review 16 (1991), [4] Accenture, Managers say the majority of information obtained for their work is useless: Accenture survey finds, Retrieved March 17, [5] J. G. March, H. A. Simon, Organizations, John Wiley and Sons, New York, [6] D. G. Covel, G. C Ulman, P. R. Manning, Information needs in office practice: Are they being met? Annals of Internal Medicine 103(4) (1985), [7] E. H. Shortliffe, J. J. Cimino, Biomedical informatics: Computer applications in health care and biomedicine, Springer Verlag, New York, [8] Branham Group, ehealth in Canada: Crossing the chasm. Branham Group, Ottawa, [9] R. Pizzi, medical records an option for some hospitals. Healthcare IT News, New Glouchester, Maryland, 6 (2007),1. [10] A. J. Sellen, H. R. Harper, The myth of the paperless office. MIT Press, Cambridge, Massachusetts, [11] A. W. Kushniruk, D. Kaufmann, V. L. Patel, Y. Levesque, P. Lottin, Assessment of a computerized patient record system: A cognitive approach to evaluating medical technology, MD Computing, 13(5) (1996), [12] V. L. Patel, A. W. Kushniruk, S. Yang, J. F. Yale, Impact of a computer-based patient record system on data collection, knowledge organization, and reasoning, JAMIA, 7(6) (2000), [13] S. Salentera, E. Eriksson, T. Junnola, E. K. Salminen, S. Lauri, Clinical judgment and information seeking by nurses and physicians working with cancer patients. Psycho-onology, 12 (2003), [14] K. A. Ericsson, H. A. Simon, Protocol analysis: Verbal reports as data (2 nd ed.). MIT Press, Cambridge, Massachusetts, [15] L. M. Hanser, P. M. Muchinsky, Work as an information environment. Organizational Behaviour, 21 (1978),