Embedded Guideline Information without ... - Wiley Online Library

Embedded Guideline Information without Patient Specificity in a Commercial Emergency Department Computerized Order-entry System Phillip V. Asaro, MD, Amy L. Sheldahl, BS, Douglas M. Char, MD

Abstract Background: Clinical practice guidelines and computerized provider order entry (CPOE) have potential for improving clinical care. Questions remain about feasibility and effectiveness of CPOE in the emergency department (ED). However, successful implementations in other settings typically incorporate decision support functions that are lacking in many commercially available ED information systems. Objectives: To compare acute coronary syndrome (ACS) guideline compliance before and after implementation of a locally implemented ACS guideline, first on paper and then in a commercially available ED information system without patient-specific clinical decision support. Methods: Clinical data were abstracted retrospectively on patients seen before and after introduction of paper and, subsequently, CPOE versions of ACS guideline-based order-sets. Order-set use was determined. Risk category assignments were made retrospectively using guideline criteria and compliance with the guideline regarding b-blockers, heparin, and aspirin was determined. Association between order-set use and compliance was determined. Results: The authors found increasing use of order-sets over the period of study. However, there was poor association between the order-sets used and risk stratification category. Some association between ED b-blocker use and use of CPOE order-sets was found, but there was no improvement in overall compliance with any of the guideline recommendations. Conclusions: Adherence to an ACS guideline did not improve with implementation of a commercial ED information system without provision for patient-specific decision support. This suggests that the lack of patient-specific decision-support functionality in most current ED information system products may hamper progress in the development of effective decision support. ACADEMIC EMERGENCY MEDICINE 2006; 13:452–458 ª 2006 by the Society for Academic Emergency Medicine Keywords: clinical decision-support systems, practice guidelines, emergency medicine, computerized medical record systems

C

linical practice guidelines widely are recognized as having potential for improving clinical care. The National Guideline Clearinghouse, maintained by the Agency for Healthcare Research and Qual-

From the Emergency Medicine Division, Washington University School of Medicine in St. Louis (PVA, DMC); and Washington University School of Medicine in St. Louis (ALS), St. Louis, MO. Received April 22, 2005; revision received September 25, 2005; accepted September 26, 2005. Address for correspondence and reprints: Phillip V. Asaro, MD, Emergency Medicine Division, Washington University School of Medicine in St. Louis, Campus Box 8072, 660 S. Euclid Avenue, St. Louis, MO 63110. Fax: 314-362-0478; e-mail: asarop@msnotes. wustl.edu.

452

ISSN 1069-6563 PII ISSN 1069-6563583

ity (AHRQ), contains more than 1,200 current clinical practice guidelines. On the other hand, low rates of adoption of published guidelines in clinical practice are well documented.1 Great effort has been directed at better understanding how to develop, disseminate, and implement practice guidelines.2–5 It is known that educational efforts alone do little to change physician practice, even with a guideline as straightforward as the Ottawa Ankle Rules.6 There is a long history of attempts at using information technology to enhance guideline implementation,7,8 and it is known that the most effective methods of implementation involve delivery of guideline information at the point of care.9–11 Widespread implementation of computerized provider order entry (CPOE) has become an increasingly highpriority goal of healthcare policy-makers,12–14 accelerated

ª 2006 by the Society for Academic Emergency Medicine doi: 10.1197/j.aem.2005.09.015

ACAD EMERG MED



April 2006, Vol. 13, No. 4



www.aemj.org

by the 1999 IOM report ‘‘To Err is Human,’’15 and further is encouraged by reports of benefit in some settings.11,16,17 Few healthcare organizations have fully implemented CPOE, with frequently cited barriers, including physician and organizational resistance, the immaturity of available commercial products, and the cost of implementation.18,19 However, many sites are at least contemplating implementation at some point in the future.20 Most of the evidence for benefits of CPOE systems arises from studies with homegrown systems in academic environments. There remains uncertainty regarding the generalizability of these results to CPOE implementation with commercially available systems,21 but the patient safety gains appear to be dependent upon the incorporation of decision support.22,23 The emergency department (ED), with its hectic pace, unpredictability, and highly variable patient mix, is thought to be a particularly challenging environment in which to introduce CPOE with decision support, and little has been published regarding efforts in this setting.24 ED information system vendors have provisions for CPOE, but provisions for decision support have generally been limited to allergy and interaction checking, and problems with excessive frequency of interaction alerts have limited the use of this functionality.25 Many questions remain about the feasibility and effectiveness of CPOE with decision support in the emergency department. Computerized provider order entry is an attractive vehicle for delivery of evidence-based clinical guideline information at the point of care.26,27 Various methods of presentation of guideline information have been reported, including context-specific links to intranet or Internet information sources,28 immediate feedback during the ordering process,29 printed reminders during routine outpatient visits,30 alerts and reminders triggered by orders,16 and the use of order-sets that are consistent with guidelines.31 Full realization of the potential for integrated guidance with CPOE remains limited by the difficulty of representing guideline knowledge in computerized order entry systems,32,33 lack of structured, patient-specific information available for computer analysis at the point of care, and provider perceptions that the computer generated suggestions are strictly related to attempts at cost-reduction rather than quality of care.34 Our objective was to compare acute coronary syndrome (ACS) guideline compliance before and after implementation of a locally implemented ACS guideline, first on paper and then in a commercially available ED information system without patient-specific clinical decision support. METHODS Study Design This was a descriptive, retrospective study of the effectiveness of implementation of an ACS guideline, initially as paper order forms and subsequently in CPOE ordersets; the effectiveness was measured as degree of order-set use and compliance with guideline recommendations. The study was approved by the Human Studies Committee of Washington University with waiver of informed consent.

453

Study Setting and Population A paper-based guideline with order-sets for ED patients with ACS was developed by a multidisciplinary team over a six-month period. This guideline closely followed the recommendations of the 2002 ACS guideline update of the American College of Cardiology–American Heart Association Task Force on Practice Guidelines.35 The guideline was finalized a few months before the planned implementation of CPOE in the ED. CPOE was to be added, along with full nursing documentation to the previously implemented electronic tracking board, nursing triage documentation, and physician-generated discharge instructions. The guideline was initially introduced as a set of four preprinted paper forms, consisting of check-off orders with some fill-in-the-blank provision, along with embedded guideline information (see Appendices A through D, available as data supplements at http://www.aemj.org/cgi/content/full/j.aem.2005.09.015/ DC1). The first order-set included initial orders for all chest pain patients and a chart containing criteria for the four ACS risk categories (Appendix A): 1) acute ST elevation myocardial infarction (MI; STEMI), 2) high-risk ACS (including non-ST elevation MI), 3) intermediate-risk ACS, and 4) low-risk ACS. The risk assignment directed the provider to one of three risk-specific order forms: 1) STEMI, 2) High/Intermediate Risk ACS, or 3) Low Risk ACS. The order-sets, designed in a paper format, then were adapted for inclusion as CPOE order-sets (see Appendix E, available as a data supplement at http://www.aemj. org/cgi/content/full/j.aem.2005.09.015/DC1). The ED information system (Healthmatics ED by A4 Health Systems, Cary, NC) supports problem-oriented order sets for common ED presentations. Default selection of orders is not supported (i.e., each order must be individually selected by the ordering provider). Organization of orders within an order-set is limited to grouping under headings such as ‘‘Lab’’ or ‘‘Medications.’’ There is a width constraint of about 20 characters in the order selection display, requiring abbreviation for longer orders (the full orders are written to the chart). Stratification criteria for ACS risk were incorporated into the system in two ways. First, PDF (portable document format) versions of the paper forms were made available online, accessible via a link at the bottom of the order entry screen. Second, we created ‘‘information-only order-sets’’ listing the criteria for each of the risk stratification categories. We also created an information-only order-set with the recommended criteria for each of the reperfusion–thrombolysis options in the guideline. Additional guidance, based on the original guideline, was added in the form of ‘‘information-only’’ lines, inserted between order items (e.g., ‘‘consider clopidogrel if ASA allergic’’ or ‘‘preferred for age > 75’’). Apart from allergy and drug interaction alerting, patient-specific decision-support functions such as rules-based alerting and automated calculations are not supported by the information system. Study Protocol Chart abstractions were performed retrospectively by the authors on all patients seen in our ED during each of four one-month periods and who had an ED diagnosis of unstable angina, acute coronary syndrome, or acute MI. Patients were identified by querying ED physician

454

diagnoses in the ED information system. The periods, designated as phases of implementation, are represented as an ordinal measure of degree of implementation as follows: Phase 1 (pre-CPOE) was just before the introduction of the new paper guideline and order-sets, when a simple preprinted chest pain order form was available. Phase 2 (pre-CPOE) began several weeks after the introduction of the paper guideline and ended before CPOE began. Phase 3 began several weeks after CPOE implementation. Phase 4 began approximately three months after the CPOE implementation, after an additional educational effort regarding the ACS order-sets. Data collected included elements of the ED history and physical examination, laboratory results, ED medication orders, and final hospital diagnoses. Evaluation of electrocardiograms (ECGs) was performed by one reviewer (DC) and consisted of simply listing the findings present without making an ECG diagnosis. Patients were retrospectively given risk-category assignments based on the extracted data from the charts and ECG reviews using the guideline criteria without regard for summary assessments or diagnoses of treating physicians. Order-set use was determined for each patient. For patients in Phases 1 and 2, the presence of completed preprinted order sheets was noted. For patients in Phases 3 and 4, order-set use was measured by whether or not the order-set had been opened as recorded in the CPOE system. Because full benefit from the guideline order-sets depends upon sequenced use of both the initial order-set and an advanced (risk-specific) order-set, order-set use was categorized as none, partial (when only either the initial or advanced order-sets were used), or full (when both the initial and at least one advanced order-set were used). Measures We determined whether b-blocker, heparin, and aspirin (or clopidogrel) were administered to each patient. These interventions were chosen because of wide acceptance of the recommendations and the broad applicability to the higher risk ACS patients whom we were considering in this study. In the analysis of b-blocker use, we considered only patients who were not already on a b-blocker (as a home medication) and who did not have a contraindication (heart block, hypotension, or bradycardia). We considered administration of either fractionated or unfractionated heparin as equivalent for the purpose of this study. We considered administration of aspirin either before (at home or administered by emergency medical services) or during the ED visit as being consistent with the guideline recommendation. The administration of clopidogrel to aspirin-allergic patients was considered to fulfill the aspirin recommendation. Final hospital diagnosis was obtained during chart abstraction. The diagnosis of coronary artery disease (CAD) was assigned for patients without current evidence of acute cardiac ischemia who had a history of CAD or who were found to have hemodynamically insignificant lesions during the current hospital stay. We collapsed final diagnoses into the categories of ACS (unstable angina and MI) and non-ACS (chest pain, CAD, or other). Data Analysis We used SPSS (SPSS Inc., Chicago, IL) in the analysis of data. Associations between categorical variables were

Asaro et al.



EMBEDDED GUIDELINE INFORMATION IN CPOE

performed by using chi-square. Somers’ d was used for ordinal variables such as risk stratification category, final diagnosis categories, and order-set use. RESULTS Case Selection A total of 218 patients met inclusion criteria by ED diagnosis. Of these, 10 patients were excluded from the study because of significant noncardiac medical complications, atypical presentations, or lack of complete records, leaving 208 patients for analysis. By retrospective risk stratification, 29 (13.9%) were in the intermediate-risk group, 143 (68.8%) were in the high-risk group (includes nonST elevation MI), and 36 (17.3%) were in the STEMI category. Order-set Use, Risk Stratification, and Final Diagnosis Order-sets were used increasingly through the implementation phases as shown in Table 1. The percentage of cases in which no order-sets were used decreased from 73% in Phase 1 to 16% in Phase 4. The percentage of cases in which there was full use of the currently available order-sets increased from 27% to 55%. We determined which was the highest-risk order set opened and compared that with the retrospectively assigned risk stratification category. Given that this risk stratification is based on information available during patient care in the ED, following the guideline should lead to opening of the relevant risk-specific advanced orderset. The asterisked cells in Table 2 identify cases in which the highest-risk order-set opened matched the patient’s risk category. Although there appears to be a trend toward increasingly risk-specific use of the order-sets between Phase 3 (19 of 48 cases) and Phase 4 (23 of 49 cases), the highest-risk order-set opened matched the risk stratification category in only 47% of cases in Phase 4. Providers opened the information-only criteria sets in just six cases between Phases 3 and 4. In only three of these cases did the provider go on to open the relevant risk-specific order-set. Higher risk category at presentation was associated with ACS-consistent final hospital diagnoses (Somers’ d of 0.27; p < 0.001; Table 3). Guideline Adherence We did not find overall improvement in compliance with guideline recommendations for b-blockers, heparin, or aspirin/clopidogrel across the phases of the study (Table 4). For patients not already on b-blockers and without

Table 1 Order-set Use by Phase None Phase Phase Phase Phase

1 2 3 4

33/45 36/66 12/48 8/49

(73) (55) (25) (16)

Partial 20/66 (30) 17/48 (35) 14/49 (29)

Full 12/45 10/66 19/48 27/49

(27) (15) (40) (55)

All data are n (%). In Phase 1, only simple reprinted order sheets were available. In Phase 2, the full paper guideline–order-sets were available. In Phases 3 and 4, the order-sets were available as computerized provider order entry order-sets.

ACAD EMERG MED



April 2006, Vol. 13, No. 4



www.aemj.org

455

Table 2 Highest-risk Computerized Provider Order Entry Order-set Opened by Risk Stratification Highest-risk Order-set Opened, n (%) Risk Category Phase 3 Intermediate risk High risk STEMI All Phase 4 Intermediate risk High risk STEMI All

None 3 8 1 12

Initial

(38) (24) (14) (25)

Low Risk

2 (25) 10 (30) 12 (25) 4 5 1 10

5 (21) 3 (19) 8 (16)

(44) (21) (6) (20)

1 (11)

1 (2)

Intermediate and High Risk 3 13 3 19

(38)* (39)* (43) (40)

3 12 4 19

(33)* (50)* (25) (39)

STEMI

Total

2 (6) 3 (43)* 5 (10)

8 33 7 48

1 2 8 11

(11) (8) (50)* (22)

9 24 16 49

All data are n (%). STEMI = acute ST elevation myocardial infarction. * Highest-risk order-set opened matched the risk stratification.

contraindications, we did find an association between ED b-blocker use and order-set use in the fourth phase of the study (Table 5), but this did not translate into an increase in b-blocker administration as compared with baseline (Table 4). We did not find associations between use of heparin or aspirin and order-set opening. DISCUSSION Despite good evidence and consensus on the emergency care of ACS patients36,37 and evidence that acute coronary syndrome pathways can be effective,38 it remains difficult to implement them,39 even in (perhaps especially in) an academic environment. We found this to be true when the guideline and orders were available as paper forms and also when they were available as CPOE order-sets. Clinicians used order-sets more as they became familiar with them, but the pattern of use suggests that use was driven more by the ease of placing orders from collections of pre-constructed orders than by the clinical guidance offered. Low rates of full and risk-specific use

Table 3 Risk Stratification vs. Final Hospital Diagnosis Chest Pain or Other Category

Stable CAD

Unstable Angina

Non-ACS

Intermediate risk (n = 29)

51

High risk (n = 143)

32

STEMI (n = 36)

11

ACS 21

21

24

23

14

11

72

7 28

56

25

MI

21 44 64 75

All data are percentages of patients in each hospital discharge category for each ED risk stratification category (all study phases combined). Rows total to 100%. CAD = coronary artery disease; ACS = acute coronary syndrome; STEMI = acute ST elevation myocardial infarction.

of order-sets (last column in Table 1 and asterisked cells of Table 2), and minimal utilization of the online stratification criteria lists, indicate little effort to follow the guideline explicitly. In many cases, more than one of the advanced order sets was opened, consistent with ‘‘hunting’’ for preconstructed orders. In contrast, riskcategory–driven use of the order-sets would result in opening of the one relevant advanced order-set containing all guideline-recommended orders. These findings are disappointing from a guideline implementation perspective, particularly in light of the demonstrated association between risk stratification and final diagnosis (Table 3). Although the general push for CPOE relates primarily to patient safety and quality of care, the chief concern of clinicians is their ability to accomplish the tasks at hand efficiently.40–44 It is apparent that CPOE must be tightly integrated with provider workflow.11 Local customization and preconstructed order-sets have emerged as key factors in CPOE implementation success.31,45–47 Although the possibility of time-neutrality has been demonstrated,48 most studies to date have shown that placing orders through a CPOE system takes providers more time than writing orders on paper.49,50 To be acceptable to clinicians, an implementation of CPOE must produce benefits in efficiency and quality of care that offset the added order entry time. In our ED implementation, CPOE contributes to efficiency by enabling the tracking of orders electronically, retrieving results through the tracking board, and allowing simultaneous access to the chart by multiple caregivers from anywhere in the ED. Although these benefits and our order-sets have made the transition to CPOE palatable for clinicians, our attempt at guideline implementation using the limited functionality available in the system did not produce demonstrable improvement in quality of care. In contrast to what clinicians may naturally look for in an information system, functions that have been shown to support improved patient care typically do so through interruption of the clinician’s workflow. There is a growing corpus of descriptive works regarding the requirements for effective decision-support systems.17,24,44 Alerts and reminders as well as presentation of guideline

456

Asaro et al.



EMBEDDED GUIDELINE INFORMATION IN CPOE

Table 4 Use of Recommended Treatments (Percentage Receiving Treatment) Treatment b-Blocker Heparin Aspirin or clopidogrel

Included Patients Not on b-blocker, no contraindication High risk/STEMI All

Phase 1 81% of 21 84% of 44 93% of 45

Phase 2 68% of 37 78% of 55 91% of 66

Phase 3 69% of 26 75% of 40 83% of 48

Phase 4 77% of 26 63% of 40 92% of 49

Data show number of patients eligible for each treatment and the percentage of eligible patients receiving the treatment by phase of study. STEMI = acute ST elevation myocardial infarction.

information at the point of care slow the clinician to some extent. These functions must be used selectively and specifically if clinicians are to view the system as improving quality of care and patient safety. The information delivered to the clinician must be patient specific and context specific. In other words, it must fit the patient and must come at the correct point in the patient’s care, at least enough of the time to avoid being viewed as noise. Our implementation could not provide patient-specific information to the provider. Although it placed the information close at hand, it relied upon the clinician to select and apply the information relevant to the individual patient. To provide patient-specific functionality, an information system must hold essential information as structured data in an information model that is designed with decision support in mind. ED information system products that have been designed without consideration for decision support may require major changes to their information model and redesign of the entire product if such functionality is to be added. The discrepancy between what purchasers of ED information systems are demanding and what is truly effective with regard to improvement in patient care and patient safety may be limiting the development of important functionality. Clinical information system vendors tailor their products to the purchasers, typically administrators with input from clinicians. Many clinicians do not believe that they need guidance and will not make decision support a priority in system selection. Further research on the efficacy of decision support in the ED is needed, but given the efficacy shown in the inpatient setting, it is likely that some degree of decision support will be feasible and efficacious in the ED. Unfortunately, the lack of patient-specific decision-support functionality in many current ED systems creates a real barrier to decisionsupport research and development. Hopefully, purchasers of ED systems, particularly in academic settings,

Table 5 b-Blocker Use by Order-set Use Phase

n

1 2 3 4

21 37 26 26

None 88% 68% 33% 67%

of of of of

Partial

16 22 70% of 10 6 88% of 8 3 43% of 7

Full 60% 60% 75% 94%

of of of of

Somers’ d (Significance) 5 ÿ0.28 (p = 0.26) 5 ÿ0.03 (p = 0.85) 12 0.20 (p = 0.22) 16 0.36 (p = 0.02)

Data are number of patients by phase of study and order-set use who were eligible for b-blocker treatment and the percentage of eligible patients who received a b-blocker. Somers’ d is a measure of the association between order-set use and receiving a b-blocker for each study phase.

will increasingly push vendors toward the needed functionality.

LIMITATIONS With the limited number of patient cases in this study, only relatively large effects would have demonstrated statistical significance. However, we found no overall improvement in the recommended interventions studied. This manuscript reports on the experience with one guideline at one institution with one information system product. However, the lack of decision-support functionality that we discuss is common to most current standalone ED information system products. Although we cannot conclude that other attempts at guideline implementation in other settings without patient-specific decision-support functionality would also fail to improve guideline compliance, we believe that our results suggest a need for incorporation of the type of patient-specific decision-support functionality that has been shown helpful in the inpatient setting.

CONCLUSIONS Adherence to an ACS guideline did not improve with implementation in a commercial ED information system without provision for patient-specific decision support. Our experience suggests that the lack of patient-specific decision-support functionality in most current standalone ED information system products may hamper progress in the development of effective decision support. Vendors of commercial ED information systems should be encouraged to incorporate patient-specific decisionsupport functionality in their products.

References 1. Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians follow clinical practice guidelines? A framework for improvement. JAMA. 1999; 282: 1458–65. 2. Davis DA, Taylor-Vaisey A. Translating guidelines into practice. A systematic review of theoretic concepts, practical experience and research evidence in the adoption of clinical practice guidelines. CMAJ. 1997; 157:408–16. 3. Grimshaw JM, Thomas RE, MacLennan G, et al. Effectiveness and efficiency of guideline dissemination and implementation strategies. Health Technol Assess. 2004; 8:iii–iv, 1–72.

ACAD EMERG MED



April 2006, Vol. 13, No. 4



www.aemj.org

4. Weingarten S. Translating practice guidelines into patient care: guidelines at the bedside. Chest. 2000; 118(2 Suppl):4S–7S. 5. Andrews EJ, Redmond HP. A review of clinical guidelines. Br J Surg. 2004; 91:956–64. 6. Holroyd BR, Wilson D, Rowe BH, Mayes DC, Noseworthy T. Uptake of validated clinical practice guidelines: experience with implementing the Ottawa Ankle Rules. Am J Emerg Med. 2004; 22:149–55. 7. Gallagher EJ. How well do clinical practice guidelines guide clinical practice? Ann Emerg Med. 2002; 40:394–8. 8. Weingarten S, Ellrodt AG, Riedinger MS, Huang C. A computerized expert system for outcome-validated medical practice guidelines. Proc Annu Symp Comput Appl Med Care. 1993;198–202. 9. Geissbuhler A, Miller RA. Clinical application of the UMLS in a computerized order entry and decisionsupport system. Proceedings of the AMIA Annual Symposium, 1998, pp 320–4. 10. Rood E, Bosman RJ, van der Spoel JI, Taylor P, Zandstra DF. Use of a computerized guideline for glucose regulation in the intensive care unit improved both guideline adherence and glucose regulation. J Am Med Inform Assoc. 2005; 12:172–80. 11. Teich JM, Glaser JP, Beckley RF. The Brigham integrated computing system (BICS): advanced clinical systems in an academic hospital environment. Int J Med Inform. 1999; 54:197–208. 12. Sarudi D. The leapfrog effect. Hosp Health Netw. 2001; 75:32–4, 36. 13. Delbanco SF. Suzanne F. Delbanco on the Leapfrog Group and employer purchasing power. Interview by Pamela K. Scarrow. J Healthc Qual. 2004; 26:18–21, 28. 14. Eikel C, Delbanco S, John M. Eisenberg Patient Safety Awards. The Leapfrog Group for Patient Safety: rewarding higher standards. Jt Comm J Qual Saf. 2003; 29:634–9. 15. Kohn LT, Corrigan JM, Donaldson MS, eds. To err is human: building a safer health system. Washington, DC: National Academies Press, 1999. 16. Teich JM, Merchia PR, Schmiz JL, Kuperman GJ, Spurr CD, Bates DW. Effects of computerized physician order entry on prescribing practices. Arch Intern Med. 2000; 160:2741–7. 17. Kuperman GJ, Gibson RF. Computer physician order entry: benefits, costs, and issues. Ann Intern Med. 2003; 139:31–9. 18. Haugh R, Gearon CJ, Serb C, Carpenter D, Scalise D. Point and click. (If only it were that easy). Hosp Health Netw. 2002; 76:36–50. 19. Poon EG, Blumenthal D, Jaggi T, Honour MM, Bates DW, Kaushal R. Overcoming barriers to adopting and implementing computerized physician order entry systems in U.S. hospitals. Health Aff (Millwood). 2004; 23:184–90. 20. Ash JS, Gorman PN, Seshadri V, Hersh WR. Computerized physician order entry in U.S. hospitals: results of a 2002 survey. J Am Med Inform Assoc. 2004; 11: 95–9. 21. Scanlon M. Computer physician order entry and the real world: we’re only humans. Jt Comm J Qual Saf. 2004; 30:342–6.

457

22. Giuse DA. Provider order entry with integrated decision support: from academia to industry. Methods Inf Med. 2003; 42:45–50. 23. Finch E, Mayne C. Thinking beyond CPOE to integrated IT strategy and management. J Healthc Inf Manage. 2004; 18:24–9. 24. Handler JA, Feied CF, Coonan K, et al. Computerized physician order entry and online decision support. Acad Emerg Med. 2004; 11:1135–41. 25. Glassman PA, Simon B, Belperio P, Lanto A. Improving recognition of drug interactions: benefits and barriers to using automated drug alerts. Med Care. 2002; 40:1161–71. 26. Chin HL, Wallace P. Embedding guidelines into direct physician order entry: simple methods, powerful results. Proc AMIA Symp. 1999; 221–5. 27. Rosenbloom ST, Talbert D, Aronsky D. Clinicians’ perceptions of clinical decision support integrated into computerized provider order entry. Int J Med Inform. 2004; 73:433–41. 28. Cimino JJ, Li J. Sharing infobuttons to resolve clinicians’ information needs. AMIA Annu Symp Proc. 2003; 815. 29. Bindels R, Hasman A, Derickx M, Van Wersch JW, Winkens RA. User satisfaction with a real-time automated feedback system for general practitioners: a quantitative and qualitative study. Int J Qual Health Care. 2003; 15:501–8. 30. Maviglia SM, Zielstorff RD, Paterno M, Teich JM, Bates DW, Kuperman GJ. Automating complex guidelines for chronic disease: lessons learned. J Am Med Inform Assoc. 2003; 10:154–65. 31. Payne TH, Hoey PJ, Nichol P, Lovis C. Preparation and use of preconstructed orders, order sets, and order menus in a computerized provider order entry system. J Am Med Inform Assoc. 2003; 10: 322–9. 32. de Clercq PA, Blom JA, Korsten HH, Hasman A. Approaches for creating computer-interpretable guidelines that facilitate decision support. Artif Intell Med. 2004; 31:1–27. 33. Greenes RA. Decision support at the point of care: challenges in knowledge representation, management, and patient-specific access. Adv Dent Res. 2003; 17:69–73. 34. Murray MD, Harris LE, Overhage JM, et al. Failure of computerized treatment suggestions to improve health outcomes of outpatients with uncomplicated hypertension: results of a randomized controlled trial. Pharmacotherapy. 2004; 24:324–37. 35. Braunwald E, Antman EM, Beasley JW, et al. ACC/ AHA guideline update for the management of patients with unstable angina and non-ST-segment elevation myocardial infarction—2002: summary article: a report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee on the Management of Patients With Unstable Angina). Circulation. 2002; 106: 1893–900. 36. Vikman S, Airaksinen KE, Tierala I, et al. Improved adherence to practice guidelines yields better outcome in high-risk patients with acute coronary syndrome without ST elevation: findings from

458

37.

38.

39.

40.

41.

42.

43.

Asaro et al.

nationwide FINACS studies. J Intern Med. 2004; 256(4):316–23. Chen MS, Bhatt DL. Highlights of the 2002 update to the 2000 American College of Cardiology/American Heart Association acute coronary syndrome guidelines. Cardiol Rev. 2003; 11(3):113–21. Cannon CP. Treatment algorithms and critical pathways for acute coronary syndromes. Semin Vasc Med. 2003; 3:425–32. Scott IA, Denaro CP, Bennett CJ, Mudge AM. Towards more effective use of decision support in clinical practice: what the guidelines for guidelines don’t tell you. Intern Med J. 2004; 34:492–500. Lee F, Teich JM, Spurr CD, Bates DW. Implementation of physician order entry: user satisfaction and self-reported usage patterns. J Am Med Inform Assoc. 1996; 3:42–55. Weiner M, Gress T, Thiemann DR, et al. Contrasting views of physicians and nurses about an inpatient computer-based provider order-entry system. J Am Med Inform Assoc. 1999; 6:234–44. Murff HJ, Kannry J. Physician satisfaction with two order entry systems. J Am Med Inform Assoc. 2001; 8:499–509. Blanchard J, Clinton P, De Lorimer K, Dulay B, Hackmeyer P, Hallman E. FMEA Utilization as part of the implementation process of computerized physician

44.

45.

46. 47.

48.

49.

50.



EMBEDDED GUIDELINE INFORMATION IN CPOE

order entry in a procedure area. Medinfo 2004; 2004(CD):1529. Bates DW, Kuperman GJ, Wang S, et al. Ten commandments for effective clinical decision support: making the practice of evidence-based medicine a reality. J Am Med Inform Assoc. 2003; 10:523–30. Franklin MJ, Sittig DF, Schmiz JL, et al. Modifiable templates facilitate customization of physician order entry. Proc AMIA Symp. 1998; 315–9. Traynor K. Customization key to successful CPOE. Am J Health Syst Pharm. 2004; 61:1087. Ahmad A, Teater P, Bentley TD, et al. Key attributes of a successful physician order entry system implementation in a multi-hospital environment. J Am Med Inform Assoc. 2002; 9:16–24. Overhage JM, Perkins S, Tierney WM, McDonald CJ. Controlled trial of direct physician order entry: effects on physicians’ time utilization in ambulatory primary care internal medicine practices. J Am Med Inform Assoc. 2001; 8:361–71. Bates DW, Boyle DL, Teich JM. Impact of computerized physician order entry on physician time. Proc Annu Symp Comput Appl Med Care. 1994; 996. Shu K, Boyle D, Spurr C, et al. Comparison of time spent writing orders on paper with computerized physician order entry. Medinfo. 2001; 10:1207–11.