General practitioners' attitudes and preparedness towards Clinical Decision Support in e-Prescribing (CDS-eP) adoption in the West of Ireland: a cross sectional study
- Chee Peng Hor†1,
- James M O'Donnell†2,
- Andrew W Murphy†3,
- Timothy O'Brien†1 and
- Thomas JB Kropmans†1Email author
© Hor et al; licensee BioMed Central Ltd. 2010
Received: 17 June 2009
Accepted: 12 January 2010
Published: 12 January 2010
Electronic clinical decision support (CDS) is increasingly establishing its role in evidence-based clinical practice. Considerable evidence supports its enhancement of efficiency in e-Prescribing, but some controversy remains. This study evaluated the practicality and identified the perceived benefits of, and barriers to, its future adoption in the West of Ireland.
This cross sectional study was carried out by means of a 27-part questionnaire sent to 262 registered general practitioners in Counties Galway, Mayo and Roscommon. The survey domains encompassed general information of individual's practice, current use of CDS and the practitioner's attitudes towards adoption of CDS-eP. Descriptive and inferential analyses were performed to analyse the data collected.
The overall response rate was 37%. Nearly 92% of respondents employed electronic medical records in their practice. The majority acknowledged the value of electronic CDS in improving prescribing quality (71%) and reducing prescribing errors (84%). Despite a high degree of unfamiliarity (73%), the practitioners were open to the use of CDS-eP (94%) and willing to invest greater resources for its implementation (62%). Lack of a strategic implementation plan (78%) is the main perceived barrier to the incorporation of CDS-eP into clinical practice, followed by i) lack of financial incentives (70%), ii) lack of standardized product software (61%), iii) high sensitivity of drug-drug interaction or medication allergy markers (46%), iv) concern about overriding physicians' prescribing decisions(44%) and v) lack of convincing evidence on the systems' effectiveness (22%).
Despite favourable attitudes towards the adoption of CDS-eP, multiple perceived barriers impede its incorporation into clinical practice. These merit further exploration, taking into consideration the structure of the Irish primary health care system, before CDS-eP can be recommended for routine clinical use in the West of Ireland.
The introduction of electronic prescribing (e-Prescribing) a decade ago, whether adopted alone or functionally incorporated into electronic medical record (EMR) regimens, has transformed prescribing practice. The proposed incorporation of clinical decision support (CDS) mechanisms such as formulary prescription, drug-drug interaction checking and drug allergy checking, into e-Prescribing provides a means of optimizing the prescribing process, including stewardship for prescribing decisions and enhancement of the safety and appropriateness of a prescription . It potentially improves the communication pathway between prescribers and dispensers, as well as augmenting the cost-effectiveness of national healthcare planning [1, 2].
There is considerable evidence supporting the role of CDS within e-Prescribing (CDS-eP) in enhancing prescribing efficiency [1, 3–6], however some controversy over its design, operational functions and national implementation remain [1, 7–9].
In the Irish primary care setting, despite increasing application of electronic medical records (EMR), the functions of e-Prescribing have only been partially adopted and utilized. Currently, the majority of general practitioners (GPs) utilize the paper mechanisms such as the British National Formulary (BNF), Monthly Index of Medical Specialties Ireland (MIMS) and international or local guidelines, as a reference to support their prescribing decision. Meanwhile, some practitioners employ electronic mechanisms such as prescribing websites and software products as an additional reference. To date, an acceptable and user-friendly CDS-eP has yet to be developed for the Irish primary care system. However, recent announcements from the Irish Health Service Executive (HSE) indicated an extension in its Information Technology (IT) strategy to include IT-related projects, a national electronic prescription system and EMR implementation [10, 11]; this reflects a continued progress in the Irish e-Health initiative and may provide opportunity for further research in developing an optimal CDS-eP for primary care.
- 1.to assess, among GPs in the West of Ireland
the prevalence of EMR adoption
the current use of CDS mechanism(s) in their prescribing practice
- 2.to identify
their perceived benefits of CDS-eP adoption in future
the potential barriers impeding its implementation
their presumptive responses towards potential alerts flagged by CDS-eP
This study was carried out using a cross sectional survey among GPs in the West of Ireland. A 27-question survey was developed and evaluated through a pilot study with five GPs to assess its comprehension and appropriateness. These questions were formulated with reference to the selected publications [8, 9, 12, 13], reviewed for their relevance and subsequently categorized into two sections in the survey.
A short explanatory note was incorporated into the survey to provide a standardized reference for the respondents in understanding the two main concepts central to this study. They were(1) "e-Prescribing (eP) allows prescribers to utilize electronic systems to facilitate and enhance the communication of a prescription, aiding the choice, administration or supply of a medicine through decision support and providing a robust audit trail for the entire medicine use process," and (2) "Clinical Decision Support within the e-Prescribing context (CDS-eP) is an algorithm that establishes the safety and appropriateness of a prescription, with links to a third party information system employed to enhance the safety of the prescription. Such systems may include clinical checks which allow alerts to be flagged up to the prescribers on drug-drug interactions or formulary status" .
Section I evaluated the individual GP's practice, including their practice duration in general practice, practice type and practice premise, prevalence of EMR adoption and current use of CDS mechanism(s) in their prescribing practice.
Section II comprised three components to assess the GPs for (i) the benefits they perceived towards CDS-eP adoption in their clinical practice, (ii) the potential barriers they perceived impeding its implementation in the primary care setting and (iii) their presumptive responses to potential alerts flagged by CDS-eP. The respondents were asked to rate their responses for each statement using ordinal five-level Likert items . This survey did not specifically refer to any software currently in use for e-Prescribing or CDS-eP.
The survey was ended with an open question for any free comments on CDs-eP from the respondents.
The questionnaire together with a one-page of personalized introductory cover letter was sent to 262 registered public and private GPs in Counties Galway, Roscommon and Mayo, representing a complete sample group for this survey. The postal details were obtained from the HSE database, which captured all GPs in the counties. However, the data source did not include any information that would provide any description of the GPs who would respond to this survey. The respondents were given an option whether to remain anonymous or to identify themselves in this survey. A pre-paid return envelope was provided to enhance responses . This study took place between 10th July and 31st August 2008. The study was granted the ethical approval from the University Research Ethics Committee of the National University of Ireland, Galway.
All the collected data were compiled into a database and further analysed using Statistical Package for the Social Sciences (SPSS) version. 14.0. In addition to the descriptive analyses, different inferential statistical analyses were performed. The ordinal five-level Likert items (strongly agree, agree, neither, disagree and strongly disagree) were collapsed to three data points (agree, neither, disagree) for inferential statistical analyses. Using Pearson Chi square (χ2) tests, we examined the associations (a) between prevalence of EMR adoption and practice premise, (b) between current use of CDS mechanisms and perceived value of different CDS mechanisms in assisting GPs' prescribing decision, and (c) between the GPs' preparedness towards future CDS-eP adoption, and duration in general practice (data were categorized into two categories according to median general practice), practice premise and current use of CDS mechanisms. If the assumptions for a χ2 test were not met, Fisher's exact or linear by linear association test would be applied as an alternative test. Besides, Spearman's correlation tests were performed to examine the relationship between the different aspects of GPs' preparedness, and their perceived barriers towards future CDS-eP adoption. In addition, we investigated whether the duration of individual GP's practice influenced the frequency of EMR adoption, their preparedness and their perceived barriers towards CDS-eP adoption using one-way Anova tests. The level of statistical significance for all inferential analyses was defined at p-value less than 0.05.
An additional component to survey "the GPs' feelings when the alerts flagged by CDS-eP during the prescribing process" was reported incomprehensive and not answered by majority of the respondents. It was hence omitted from data analysis.
We obtained an overall response rate of 37% (98 out of 262) in this study. Nearly 45% (44 out of 98) of respondents identified themselves in this survey.
Overview of individual practice and EMR prevalence
The practice durations of the respondents were variable, ranging from one to 39 years. Half of them had at least 19 years of experience in the field of general practice. There were nearly equal proportions of GPs currently working in either city (30.9%) or rural (32%) areas, while 37.1% had a mixed practice. Approximately 60% of them practiced in a group of at least two whole-time equivalent GPs, with the remaining 40% operating single-handedly. Among those practising in groups, half practiced in a group of two whole-time equivalent practitioners, followed by 31% in a group of three, 13% in a group of four, 3% in a group of five and 2% in a group of eight.
Current use of CDS mechanisms and their values in prescribing process
With regard to the types of CDS mechanisms presently used by the GPs in their prescribing practice, 45% of them relied solely on the conventional paper mechanisms, such as BNF, MIMS, and local or international guidelines. Only 5% of GPs referred to computer or internet-based information exclusively. The remaining half of the respondents utilized combinational electronic and paper mechanisms in supporting their prescribing decisions. In addition, two respondents reported referring to the specialists in hospital for prescribing advice occasionally.
Attitudes and preparedness towards CDS-eP adoption
GPs' attitudes and preparedness towards CDS-eP
I am familiar with what CDS-eP is and how it is used in clinical prctice (N = 93)
I believe that CDS-eP has the capacity to improve prescribing quality (N = 91)
I believe that using CDS-eP may reduce prescribing errors (N = 91
I believe that CDS-eP may reduce my decision making power in prescribing (N = 90)
I am open to learning/using new CDS-eP (N = 95)
My practice is willing to invest greater resources in CDS-eP in the future (N = 92)
Potential barriers impeding CDS-eP adoption
Potential barriers impeding implementation of CDS-eP in general practice
Lack of convincing evidence regarding its effectiveness (N = 88)
High sensitivity of drug -- drug interaction or drug allergy markers (N = 85)
Concern about the degree of flexibility for the physician to override CDS-eP (N = 88)
Lack of financial incentives (N = 87)
Lack of acceptable, standardized product software (N = 88)
Lack of a strategic plan for implementation (N = 89)
Presumptive responses towards alerts from CDS-eP
General comments on CDS-eP
A total of 13 comments were provided by the respondents with some illustrated in the Additional File 1. Concern was expressed regarding the potential interference with their prescribing independence, as well as the system's efficiency as an up-to-date clinical decision support tool. With regards to hypersensitivity of drug-drug interaction markers, a GP commented that "really depend(s) on the specificity e.g. I know that angiotensin-converting enzyme (ACE) inhibitors and diuretics interact. This may be why I prescribe them (e.g. for enhanced antihypertensive effect or to avoid ACEI-induced hyperkalaemia). I probably would get frustrated if the alert is flagged every time I attempt to co-prescribe these medication and especially if the alert substantially delayed my prescribing (e.g. by 20 seconds)".
In this study, majority of the GP respondents displayed positive attitudes towards the potential benefits brought by CDS-eP. They expressed their readiness to embrace the new mechanism through their willingness to learn and to invest greater resources in adopting this mechanism. In addition, the high prevalence of EMR adoption in the clinical practice has set a potential platform for future incorporation of CDS-eP into e-Prescribing system in order to optimize its functions. Moreover, there is an increased utilization of electronic CDS mechanisms in the Irish primary care setting. The positive value of such mechanisms in aiding their prescribing decision-making process has been acknowledged by the GPs. An increased confidence in the application, together with improved availability and accessibility to electronic sources may have contributed to emergence of this transitional phase, from relying solely on paper mechanisms to start exploring the electronic mechanisms. The assessment of the source of the specific electronic mechanism was beyond the scope of this study, and remains as an interesting aspect to explore further. Also, the high rates of appreciation for the value of CDS mechanisms and willingness for further investment may be specific to the sub-group of GPs who were likely to respond to this survey.
Worldwide, inappropriate prescribing in the community setting, particularly among elderly populations has been reported, with some resulting in adverse medication events [16–18]. Currently, there is no national programme or policy in place to collect and evaluate data related to medication errors in the Irish primary care setting. A report regarding the medication safety scheme from Tallaght Hospital in Dublin revealed that nearly 15% of 102 prescribing errors in the community setting reported involved patient harm . Besides, prescribing errors accounted for 10 - 20% of legal claims against GPs in the United Kingdom and Ireland [20, 21].
Since its inception, incorporation of CDS-eP into e-Prescribing systems has encountered challenges ranging from national standardized implementation to individual prescriber endorsement for its application in the clinical setting. The absence of a strategic national action plan was perceived as the greatest barrier of all. Further progress is awaited with the recent HSE announcements on its plan to invest in a national e-Prescribing system and IT-related projects [10, 11]. Financial incentives often act as a driving force to accelerate adoption of the new initiative. Its long term implication for cost-saving would be significant. An example is a recent initiative by the Department of Health and Human Services in United States to launch an incentive payments scheme to eligible professional successful electronic prescribers, via Medicare over a five-year period from 2009. It is expected to save up to US $156 m by avoiding adverse medication events over the course of the programme . McMullin et. al. reported that application of e-Prescribing systems with an integrated CDS mechanism in primary care had shifted the prescribing behaviour away from the high cost therapies and lowered prescription costs. The savings from the altered prescribing behaviours offset the subscription cost of the system . More studies are needed to illustrate the cost-and-benefit analysis of CDS-eP adoption to convince authorities that investment is worthwhile in the primary care setting. Designation of standardized, acceptable and user-friendly software products for CDS-eP is a demanding process, especially with increasing emphasis on evidence-based prescribing .
Hypersensitivity of the drug allergy or drug-drug interaction alerts and concern over flexibility as a prescriber to override CDS-eP are the GP-specific barriers. Hypersensitivity of alerts is often attributed to frequent flagging of trivial or unnecessary alerts and in high volume. For example, running an allergy or drug-drug interaction check against medication history instead of current medication regimen . Studies have indicated 40 - 96% of electronic medication safety alerts were overridden by physicians [24–26]. Factors contributing to overriding the alerts were poor specificity and their low significance in clinical context, alerts overload interrupting workflows and constant time pressure in clinical practice. Moreover, known interactions with justifiable benefits greater than risks and patients' resistance to change have also led to non-adherence to the alerts [8, 24–29]. Concern over CDS-eP interfering with the prescribers' decision making power, as expressed by one-fifth of GP respondents may become an inherent factor contributing to the resistance to adopting the mechanism in individual practice. It is proposed that allowing prescribers to set their desirable threshold of alerts severity may be expected to improve the alerts acceptance rate [24, 28]. Also, a "smarter" system designed to stratify various medication alerts to their clinical relevance and utilize a set of mandatory alerts may improve prescribing safety . In this study, more than 80% of GP respondents reported that they would accept drug allergy and drug-drug interaction alerts most of the time, if CDS-eP is in place.
There is considerable evidence supporting the roles of CDS-eP in enhancing prescribing quality [1, 3–6]. The lack of training and exposure towards e-Prescribing-related issues may have contributed to the high degree of unfamiliarity among the GPs and their perception of lacking convincing evidence on the mechanism's effectiveness.
This study should be viewed with the following limitations is mind. A 37% response rate may have provided an unrepresentative sample that has limited views in this area. Besides, the survey was carried out among GPs in three out of 26 Irish counties, hence the results may not be generalizable to reflect the views of the entire Irish primary care. However this is the first study of its kind in Ireland and, at the very least, provides a valuable benchmark. Although there is a high EMR adoption rate (92%) among our respondents, future research should attempt to assess the attitudes and opinions of the non-respondents towards e-Health initiative.
We propose a multidisciplinary and multidimensional approach to make further progression with CDS-eP. At national level, establishment of an expert review panel specific to e-Prescribing and CDS-eP is needed to evaluate the current Irish e-Health initiatives (e.g. e-Prescribing system, e-transfer of medical information), assess in depth the existing barriers and draft practical recommendations for implementation. With a score of 42 points out of 100 for the e-Health component in the recent Euro Health Consumer Index (EHCI) 2008, more concerted efforts are needed by all stakeholders . At local level, such as in the West of Ireland, a pilot feasibility study of CDS-eP adoption in selected practices is recommended.
Prescribers' opinions and attitudes are a pre-requisite to determine the success of CDS-eP adoption in clinical practice. The results from this study reflect encouraging attitudes of GPs in the West of Ireland towards CDS-eP, potentially paving the way for its future adoption, as part of a better integrated care pathway for patients. Major barriers identified require to be overcome for future progression through multidisciplinary collaboration.
We acknowledged funding from the Healthcare Informatics Society of Ireland (HISI) research bursary (2007 - 2009) for CPH. We are grateful to all the general practitioner respondents for their participation. Thanks to Una St. John from the Department of General Practice, National University of Ireland, Galway for her assistance in handling the questionnaires.
- Kuperman GJ, Bobb A, Payne TH, Avery AJ, Gandhi TK, Burns G, Classen DC, Bates DW: Medication-related clinical decision support in computerized provider order entry systems: a review [see comment]. J Am Med Inform Assoc. 2007, 14 (1): 29-40. 10.1197/jamia.M2170.View ArticlePubMedPubMed CentralGoogle Scholar
- United States DoHaHS: HHS Takes New Steps to Accelerate Adoption of Electronic Prescribing. 2008, HHS Press OfficeGoogle Scholar
- Tamblyn R, Huang A, Perreault R, Jacques A, Roy D, Hanley J, McLeod P, Laprise R: The medical office of the 21st century (MOXXI): effectiveness of computerized decision-making support in reducing inappropriate prescribing in primary care. CMAJ Canadian Medical Association Journal. 2003, 169 (6): 549-556.PubMedPubMed CentralGoogle Scholar
- Kawamoto K, Houlihan CA, Balas EA, Lobach DF: Improving clinical practice using clinical decision support systems: a systematic review of trials to identify features critical to success [see comment]. BMJ. 2005, 330 (7494): 765-10.1136/bmj.38398.500764.8F.View ArticlePubMedPubMed CentralGoogle Scholar
- Berner ES, Houston TK, Ray MN, Allison JJ, Heudebert GR, Chatham WW, Kennedy JI, Glandon GL, Norton PA, Crawford MA: Improving ambulatory prescribing safety with a handheld decision support system: a randomized controlled trial. J Am Med Inform Assoc. 2006, 13 (2): 171-179. 10.1197/jamia.M1961.View ArticlePubMedPubMed CentralGoogle Scholar
- Bates DW, Teich JM, Lee J, Seger D, Kuperman GJ, Ma'Luf N, Boyle D, Leape L: The impact of computerized physician order entry on medication error prevention. J Am Med Inform Assoc. 1999, 6 (4): 313-321.View ArticlePubMedPubMed CentralGoogle Scholar
- Teich JM, Osheroff JA, Pifer EA, Sittig DF, Jenders RA, The CDSERP: Clinical decision support in electronic prescribing: recommendations and an action plan: report of the joint clinical decision support workgroup [see comment]. J Am Med Inform Assoc. 2005, 12 (4): 365-376. 10.1197/jamia.M1822.View ArticlePubMedPubMed CentralGoogle Scholar
- Sittig DF, Krall MA, Dykstra RH, Russell A, Chin HL: A survey of factors affecting clinician acceptance of clinical decision support. BMC Med Inf Decis Mak. 2006, 6: 6-10.1186/1472-6947-6-6.View ArticleGoogle Scholar
- Short D, Frischer M, Bashford J: Barriers to the adoption of computerised decision support systems in general practice consultations: a qualitative study of GPs' perspectives. Int J Med Inf. 2004, 73 (4): 357-362.View ArticleGoogle Scholar
- Griffin K: HSE planning to resurrect controversial PPARS system. Sunday Tribune. 2008Google Scholar
- Griffin K: HSE to spend €78 m replacing failed Ppars system. 2008, Tribune News. DublinGoogle Scholar
- Wright M-O, Knobloch MJ, Pecher CA, Mejicano GC, Hall MC: Clinical decision support systems use in Wisconsin. WMJ. 2007, 106 (3): 126-129.PubMedGoogle Scholar
- Anderson JG: Social, ethical and legal barriers to e-health. Int J Med Inf. 2007, 76 (5-6): 480-483. 10.1016/j.ijmedinf.2006.09.016.View ArticleGoogle Scholar
- Matell MS, Jacoby J: Is There an Optimal Number of Alternatives for Likert Scale Items? Study I: Reliability and Validity. Educational and Psychological Measurement. 1971, 31 (3): 657-674. 10.1177/001316447103100307.View ArticleGoogle Scholar
- Edwards P, Roberts I, Clarke M, DiGuiseppi C, Pratap S, Wentz R, Kwan I: Increasing response rates to postal questionnaires: systematic review [see comment]. BMJ. 2002, 324 (7347): 1183-10.1136/bmj.324.7347.1183.View ArticlePubMedPubMed CentralGoogle Scholar
- Steven ID, Malpass A, Moller J, Runciman WB, Helps SC: Towards safer drug use in general practice. Journal of Quality in Clinical Practice. 1999, 19 (1): 47-50. 10.1046/j.1440-1762.1999.00302.x.View ArticlePubMedGoogle Scholar
- Soendergaard B, Kirkeby B, Dinsen C, Herborg H, Kjellberg J, Staehr P: Drug-related problems in general practice: results from a development project in Denmark. Pharmacy World & Science. 2006, 28 (2): 61-64.View ArticleGoogle Scholar
- Howard M, Dolovich L, Kaczorowski J, Sellors C, Sellors J: Prescribing of potentially inappropriate medications to elderly people. Fam Pract. 2004, 21 (3): 244-247. 10.1093/fampra/cmh305.View ArticlePubMedGoogle Scholar
- New Safety Scheme in Tallaght Hospital highlights common drug errors. Irish Medical News. 2006
- Green S, Goodwin H, Moss J: Problems in general practice: medication incident. The MDU Risk Management Medical Defence Union. 1996, 1-8.Google Scholar
- Silk N: An Analysis of 1000 Consecutive General Practice Negligence Claims. Medical Protection Society. 2000, 1-18.Google Scholar
- McMullin ST, Lonergan TP, Rynearson CS: Twelve-month drug cost savings related to use of an electronic prescribing system with integrated decision support in primary care [see comment]. Journal of Managed Care Pharmacy. 2005, 11 (4): 322-332.View ArticlePubMedGoogle Scholar
- Robert J, Fortuna DR-DJFFZFXCSRS: Clinician attitudes towards prescribing and implications for interventions in a multi-specialty group practice. Journal of Evaluation in Clinical Practice. 2008, 14 (6): 969-973. 10.1111/j.1365-2753.2007.00913.x.View ArticleGoogle Scholar
- Lapane KL, Waring ME, Schneider KL, Dube C, Quilliam BJ: A mixed method study of the merits of e-prescribing drug alerts in primary care. Journal of General Internal Medicine. 2008, 23 (4): 442-446. 10.1007/s11606-008-0505-4.View ArticlePubMedPubMed CentralGoogle Scholar
- Sijs van der H, Aarts J, Vulto A, Berg M: Overriding of drug safety alerts in computerized physician order entry. J Am Med Inform Assoc. 2006, 13 (2): 138-147. 10.1197/jamia.M1809.View ArticlePubMedPubMed CentralGoogle Scholar
- Weingart SN, Toth M, Sands DZ, Aronson MD, Davis RB, Phillips RS: Physicians' Decisions to Override Computerized Drug Alerts in Primary Care. Arch Intern Med. 2003, 163 (21): 2625-2631. 10.1001/archinte.163.21.2625.View ArticlePubMedGoogle Scholar
- Taylor LK, Tamblyn R: Reasons for physician non-adherence to electronic drug alerts. Medinfo. 2004, 11 (Pt 2): 1101-1105.Google Scholar
- Tamblyn R, Huang A, Taylor L, Kawasumi Y, Bartlett G, Grad R, Jacques A, Dawes M, Abrahamowicz M, Perreault R: A Randomized Trial of the Effectiveness of On-demand versus Computer-triggered Drug Decision Support in Primary Care. J Am Med Inform Assoc. 2008, 15 (4): 430-438. 10.1197/jamia.M2606.View ArticlePubMedPubMed CentralGoogle Scholar
- Payne TH, Nichol WP, Hoey P, Savarino J: Characteristics and override rates of order checks in a practitioner order entry system. Proceedings/AMIA Annual Symposium. 2002, 602-606.Google Scholar
- Spina JR, Glassman PA, Belperio P, Cader R, Asch S, Primary Care Investigative Group of the VALAHS: Clinical Relevance of Automated Drug Alerts From the Perspective of Medical Providers. American Journal of Medical Quality. 2005, 20 (1): 7-14. 10.1177/1062860604273777.View ArticlePubMedGoogle Scholar
- Björnberg A, Uhlir M: Euro Health Consumer Index 2008. 2008, Brussels, Belgium: Health Powerhouse, 1-69.Google Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1472-6947/10/2/prepub