Perspective
Towards creating the perfect electronic prescription
Ajit A Dhavle,1 Michael T Rupp2
1
Surescripts LLC, Arlington,
Virginia, USA
2
Midwestern University,
Glendale, Arizona, USA
Correspondence to
Dr Ajit A Dhavle, Surescripts
LLC, 2800 Crystal Drive,
Arlington, VA 22202, USA;
[email protected]
The views expressed in this
article are those of the authors
and do not necessarily
represent those of Surescripts
LLC or Midwestern University
Received 20 February 2014
Revised 5 June 2014
Accepted 3 July 2014
ABSTRACT
Signiﬁcant strides have been made in electronic (e)prescribing standards and software applications that
have further fueled the adoption and use of eprescribing. However, for e-prescribing to realize its full
potential for improving the safety, effectiveness, and
efﬁciency of prescription drug delivery, important work
remains to be carried out. This perspective describes the
ultimate goal of all e-prescribing stakeholders including
prescribers and dispensing pharmacists: a clear,
complete, and unambiguous e-prescription order that
can be seamlessly received, processed, and fulﬁlled at
the dispensing pharmacy without the need for additional
clariﬁcation from the prescriber. We discuss the
challenges to creating the perfect e-prescription by
focusing on selected data segments and data ﬁelds that
are available in the new e-prescription transaction as
deﬁned in the NCPDP SCRIPT Standard and suggest
steps that could be taken to move the industry closer to
achieving this vision.
THE PERFECT ELECTRONIC (E)-PRESCRIPTION
ORDER
To cite: Dhavle AA,
Rupp MT. J Am Med Inform
Assoc Published Online First:
[please include Day Month
Year] doi:10.1136/amiajnl2014-002738
In a perfect world, prescribers would validate their
patients’ personal information and reconcile their
medication history during every ofﬁce visit.1 2 Prior
to issuing an e-prescription, the prescriber would
have reviewed the patient’s prescription beneﬁt
coverage and discussed the most clinically appropriate and cost-effective therapy options with the
patient.3
The perfect e-prescription issued by the prescriber would be limited to data that are structured
and codiﬁed using recognized industry standards
and data formats, with strict limitations on free-text
information. These structured and codiﬁed data
enable the e-prescription-generating software application to apply clinical decision support tools that
alert the prescriber to possible conﬂicts prior to
transmission of the e-prescription to the pharmacy.4
The resulting perfect prescription would be a complete, interpretable, and safe prescription order that
arrives at the dispensing pharmacy electronically
having already been screened for therapeutic and
drug beneﬁt conﬂicts. The perfect e-prescription
would seamlessly populate all required data ﬁelds
in the pharmacy’s practice management system
without the need for additional edits and would
generate
no
conﬂicts
during
subsequent
in-pharmacy and online drug utilization review. In
short, the perfect e-prescription order would
require only routine processing and fulﬁllment,
thereby fully utilizing the production efﬁciencies
the pharmacy has engineered into its policies, procedures, technologies, workﬂow, and management
practices (ﬁgure 1). Receipt of such a complete,
unambiguous, and accurate e-prescription will
Dhavle AA, et al. J Am Med Inform Assoc 2014;0:1–4. doi:10.1136/amiajnl-2014-002738
eliminate mundane manual pharmacy tasks such as
data entry and order clariﬁcations, thus allowing
the pharmacist to focus on delivering much needed
patient counseling and medication therapy management services.5–8
The good news is that the perfect world we envision above is achievable. Impressive strides have
been made in the quality of e-prescribing in recent
years. However, the ultimate goal of a perfect
e-prescription remains elusive. Contributing factors
include: the absence and implementation of coding
formats that accurately and completely communicate prescriber intent in all circumstances; inadequate end-user interface designs; limited end-user
system knowledge, training, and support; and insufﬁcient integration of user workﬂow and preferences
into e-prescribing systems.9–14 This paper discusses
key data segments and ﬁelds within the widely
implemented National Council for Prescription
Drug Programs’ (NCPDP) SCRIPT Standard that
we believe would move us towards achieving the
goal of the perfect e-prescription.
DRUG DESCRIPTION DATA FIELD
Purpose
The Drug Description data ﬁeld is contained within
the SCRIPT Drug segment of the e-prescription
message. This free-text data ﬁeld is designed to
communicate the medication name, strength, and
dosage form.15 The receipt of a complete and
unambiguous drug description clearly communicates the prescriber’s intent to the pharmacy.
Problems/challenges
Some e-prescribing applications allow users to enter
free-text drug descriptions. In many cases, these
descriptions are incomplete and/or open to multiple
interpretations. The absence of standardized drug
descriptions is an industry-wide problem.16 Most
physician and pharmacy vendor systems subscribe to
one of several commercial drug databases to fulﬁll
their drug terminology needs. Each drug database
company maintains its own proprietary editorial
naming convention policy. Moreover, physician and
pharmacy vendors that use these drug database products also create their own in-house editorial policies,
and healthcare systems may implement their own
naming conventions beyond what is provided to
them by their system vendor and the vendorsubscribed drug database product. Finally, some physician practices create their own local medication
name lists (ie, favorites list), thereby further confusing
drug description names. As a result, a pharmacy can
receive dozens of different variations for a particular
drug description. The receipt of a non-standardized
drug description string prevents the pharmacy system
from programmatically reconciling with its internal
drug description using the standardized drug
1
Perspective
Figure 1 The perfect e-prescription.
identiﬁers in the message. This, in turn, precludes them from
populating internal pharmacy screens, thereby requiring staff to
manually match incoming drug descriptions to internal descriptions or rekey data.
Recommendations
NCPDP has created recommendations to help alleviate this
problem, including: (a) encouraging drug database companies to
create recommended e-prescribing drug description names and
train clients on proper usage; (b) encouraging pharmacy and
physician technology vendors to send only recommended
e-prescription descriptions in all outbound messages; and (c)
urging prescriber and pharmacy system vendors to prohibit ‘free
texting’ of drug descriptions.17 We believe industry-wide implementation of these recommendations would eliminate most
observed drug description problems.
DRUG IDENTIFIER DATA FIELD
Purpose
The Drug Identiﬁer data ﬁeld carries a codiﬁed data element in
the SCRIPT Drug segment of the e-prescription message. Early
in e-prescribing, lacking a unique medication identiﬁer, the
e-prescribing industry began using the concept of a ‘representative National Drug Code (NDC)’.17 Drug identiﬁer values such
as representative NDCs and the National Library of Medicine’s
(NLM) RxNorm concept unique identiﬁers (RxCUIs) are used
to communicate the identity of the prescribed drug to enable
clinical decision support checks.17–19 The accurate and consistent population of drug identiﬁers in the e-prescription also
enables the receiving pharmacy system to positively identify the
incoming free-text drug description by validating it against the
textual drug description associated with these identiﬁers. This
2
allows the pharmacy system to pre-populate internal user
screens.
Problems/challenges
The NDC directory has not been properly maintained, and the
absence of an accurate and complete central repository to identify and cross-reference representative NDCs is an acknowledged industry problem.19 20 The accuracy of the representative
NDC identiﬁer received in the new e-prescription message is
sometimes questionable. If the drug description associated with
the representative NDC does not match the textual drug
description sent in the message, then required screens in the
pharmacy system cannot be auto-populated.11 The use of the
NLM’s RxNorm clinical terminology is recommended.19 While
RxNorm codes are available directly from NLM, most physician
technology vendors depend on commercial drug databases for
their terminology mapping needs. The industry-wide adoption
and consistent use of this terminology is still lacking.
Meaningful use criteria speciﬁed by the Health Information
Technology for Economic and Clinical Health Act, enacted as
part of the American Recovery and Reinvestment Act of 2009,
mandates the use of RxNorm, which may propel adoption.21
Recommendations
Prescriber technology vendors should restrict end-user access by
allowing only authorized personnel to map representative
NDCs and RxCUIs to drug descriptions. Prescriber technology
vendors should also consistently send accurate RxCUIs when
available and validate representative NDC and RxCUI identiﬁers
at the point of care before transmitting prescriptions. Prescriber
and pharmacy vendors must maintain accurate and timely drug
database ﬁles and ensure that they are accessible to end users.
Long term, the e-prescribing industry should move away from
Dhavle AA, et al. J Am Med Inform Assoc 2014;0:1–4. doi:10.1136/amiajnl-2014-002738
Perspective
using representative NDCs completely and use RxNorm as the
main prescribing drug identiﬁer.
PATIENT DIRECTIONS (SIG) DATA FIELD AND STRUCTURED
AND CODIFIED SIG SEGMENT
Purpose
The Sig text is a 140-character free-text data ﬁeld in the
SCRIPT Drug segment. The Sig is also available within the
SCRIPT Standard as an independent structured and codiﬁed
segment. A complete and unambiguous Sig (ie, directions for
use) communicates the prescriber’s intent for how the patient
should take the medication. Communication of clear and complete directions for use is essential to ensure proper prescription
labeling, appropriate pharmacist counseling, and optimal medication use.22
Problems/challenges
As with paper prescriptions, considerable variability is found in
the electronic Sig text strings that are generated by prescribers.23–
25
A seemingly simple Sig such as ‘take one tablet every day’ can
be represented in numerous ways, and the absence of standardization can lead to requests for clariﬁcation by the pharmacy. In
the current SCRIPT version used by the industry, prescribers cite
a problem when some directions are greater than 140 characters.
A standard codiﬁed Sig structure has been a recognized industry
need for many years. Although this functionality is available in
the SCRIPT Standard, feedback from pharmacy and physician
technology vendors suggests that the approach taken in the structure is complex and cumbersome to implement. However, our
initial research indicates that most commonly used ambulatory
Sigs are fairly simple in construction and hence can be more
easily mapped to the standard codiﬁed Sig structure.
Recommendations
Universal Medication Schedule (UMS) is a methodology that
simpliﬁes medication administration instructions for the patient
and/or their caregiver and is intended as an optimal way to
convey Sigs for use to the patient.26 Variability in the Sig text
strings observed in e-prescriptions has the potential to introduce
quality defects; industry-wide implementation of UMS can
result in standardization of Sigs, thus reducing variations as
described in table 1.27 To this effect, NCPDP, in April 2013,
published a white paper to help the industry implement UMS
using NCPDP standards.26 Industry-wide deployment of a
Structured and Codiﬁed Sig standard in SCRIPT is needed. In
September 2013, NCPDP created a task group to develop
detailed guidance around the implementation of the Structured
and Codiﬁed Sig standard as available in the 10.6 NCPDP
SCRIPT and future SCRIPT versions, and create examples to
help facilitate adoption. Finally, the length of the free-text Sig
Table 1 Example of Universal Medication Schedule Sig
Sig string variations found in
e-prescriptions
Take one tablet twice a day
One tablet twice daily
Twice daily
Take one tablet by mouth twice
a day
Take tablet twice a day
Take twice daily
Universal Medication Schedule sig
Take one tablet in the morning and take
one tablet in the evening
Dhavle AA, et al. J Am Med Inform Assoc 2014;0:1–4. doi:10.1136/amiajnl-2014-002738
ﬁeld as available within the Drug segment of SCRIPT has been
increased from 140 to 1000 characters in SCRIPT Standard
V.2012, and it is recommended that the industry implement a
future version of SCRIPT as soon as possible to accommodate
all legitimate use cases.
NOTES DATA FIELD
Purpose
The Notes data ﬁeld in the Drug segment is a free-text optional
ﬁeld in the SCRIPT standard. This ﬁeld allows prescribers to
include additional patient-speciﬁc information that is related to,
but not part of, the prescription.28 An example is a prescriber’s
request to the pharmacist to reinforce key aspects of therapy
(eg, ‘Reinforce to mother that medication must be given until it
is all gone’). Importantly, the Notes ﬁeld should not be used for
prescription and other types of information that has a designated, standard ﬁeld.17 28
Problems/challenges
The Notes ﬁeld is often populated with information that should
be included in one of the other designated ﬁelds, most commonly patient Sig.11 This often results in confusing or contradictory information to that included in other parts of the
e-prescription message such as Sig or Quantity, thereby necessitating contact with the prescriber to resolve. The Notes ﬁeld is
also sometimes populated with instructions that are to be
handled in other available transactions (such as a new prescription containing a note to cancel a previous prescription).
Recommendations
Better user training, end-user interface enhancement, and
ongoing content monitoring by physician technology vendors is
needed to reduce inappropriate use of the Notes ﬁeld.17 29
Industry-wide adoption, utilization, and end-user education on
other NCPDP SCRIPT transactions—speciﬁcally ‘Change
Prescription’ and ‘Cancel Prescription’ messages—are key
towards ensuring complete resolution of this problem.
NCPDP’s Electronic Prescribing Best Practices Task Group is
working to structure this Notes data ﬁeld and its use.
Presentation of structured notes strings and creation of new data
ﬁelds in the SCRIPT Standard will result in machine-readable,
actionable data that can then be put to meaningful use at the
pharmacy. When available, prompt industry-wide deployment of
this enhancement will help mitigate the problems. However, the
Notes ﬁeld should continue to allow prescribers to enter free
text only in cases where it is needed.
ADDITIONAL REQUIREMENTS
In addition to the data segments and ﬁelds discussed above, realizing our vision of a perfect e-prescription will also require a
strategy that addresses other components of the e-prescribing
process including implementation of continual end-user training
programs, interface improvements targeted towards improving
usability, and integration of end-user input into system
designs.9 30–32 Achievement of this goal will require the collective efforts of standards organizations such as NCPDP, with
industry involvement from database suppliers, e-prescribing networks, and physician and pharmacy technology vendors.
CONCLUSION
Properly implemented and used, e-prescribing has demonstrated
improved safety, effectiveness, and efﬁciency of patient care.
However, some work remains to be carried out before the
potential beneﬁts of this transformative technology are fully
3
Perspective
realized. In this paper we have discussed the need for implementation of several key SCRIPT data ﬁelds and segments and adoption of best practice recommendations. While the current
SCRIPT Standard does contain important features, it is important that the e-prescribing industry continues to move forward to
adopt and implement the most recent version of the SCRIPT
Standard and encourage improvements in interface designs and
end-user training.
Acknowledgements The authors are indebted to Lynne Gilbertson, Vice President
of Standards at NCPDP, for her critical review of the manuscript.
Contributors The corresponding author, AAD, takes responsibility for the integrity
of this manuscript. Study concept and design: AAD, MTR. Drafting of manuscript:
AAD, MTR. Critical revision of manuscript for important intellectual content: AAD,
MTR. Administrative, technical, or material support: AAD. Supervision: AAD.
Competing interests None.
14
15
16
17
18
19
20
Provenance and peer review Not commissioned; externally peer reviewed.
21
REFERENCES
1
2
3
4
5
6
7
8
9
10
11
12
13
4
Lapane KL, Whittemore K, Rupp MT, et al. Maximizing the effectiveness of
e-prescribing between physicians and community pharmacies. Final Progress Report
submitted to AHRQ on January 28, 2007.
Grossman JM, Gerland A, Reed MC, et al. Physicians’ experiences using commercial
e-prescribing systems. Health Aff (Millwood) 2007;26:w393–404.
Fischer MA, Vogeli C, Stedman M, et al. Effect of electronic prescribing with
formulary decision support on medication use and cost. Arch Intern Med
2008;168:2433–9.
National Research Council. Preventing medication errors: quality chasm series.
Washington, DC: The National Academies Press, 2007.
Crosson JC, Etz RS, Wu S, et al. Meaningful use of electronic prescribing in 5
exemplar primary care practices. Ann Fam Med 2011;9:392–7.
Kilbridge P. E-prescribing. Oakland, CA: California HealthCare Foundation, 2001.
Institute for Safe Medicine Practices. A call to action: eliminate handwritten
prescriptions within three years! http://www.ismp.org/newsletters/acutecare/articles/
whitepaper.asp (accessed April 2014).
Harvey J, Avery AJ, Waring J, et al. The socio-technical organisation of community
pharmacies as a factor in the Electronic Prescription Service Release Two
implementation: a qualitative study. BMC Health Serv Res 2012;12:471.
Singh H, Mani S, Espadas D, et al. Prescription errors and outcomes related to
inconsistent information transmitted through computerized order entry: a
prospective study. Arch Intern Med 2009;169:982–9.
Palchuk MB, Fang EA, Cygielnik JM, et al. An unintended consequence of electronic
prescriptions: Prevalence and impact of internal discrepancies. J Am Med Inform
Assoc 2010;17:472–6.
Grossman JM, Cross DA, Boukus ER, et al. Transmitting and processing electronic
prescriptions: experiences of physician practices and pharmacies. J Am Med Inform
Assoc 2012;19:353–9.
Rupp MT, Warholak TL. Evaluation of e-prescribing in chain community pharmacy:
best-practice recommendations. J Am Pharm Assoc (2003) 2008;48:64–70.
Warholak TL, Rupp MT. Analysis of community chain pharmacists’ interventions on
electronic prescriptions. J Am Pharm Assoc (2003) 2009;49:59–64.
22
23
24
25
26
27
28
29
30
31
32
Koppel R, Metlay JP, Cohen A, et al. Role of computerized physician order entry
systems in facilitating medication errors. JAMA 2005;293:1197–203.
National Council for Prescription Drug Programs (NCPDP). SCRIPT Standard
Implementation Guide. October 2008 Version 10.6.
Levin MA, Krol M, Doshi AM, et al. Extraction and mapping of drug names
from free text to a standardized nomenclature. AMIA Annu Symp Proc
2007:438–42.
National Council for Prescription Drug Programs (NCPDP). SCRIPT Implementation
Recommendations. April 2014 Version 1.25. http://www.ncpdp.org/NCPDP/media/
pdf/SCRIPT-Implementation-Recommendations-V1-25.pdf (accessed April 2014).
Teich JM, Osheroff JA, Pifer EA, et al. CDS Expert Review Panel. Clinical decision
support in electronic prescribing: recommendations and an action plan: report of the
joint clinical decision support workgroup. J Am Med Inform Assoc
2005;12:365–76.
Bell DS, O’Neill S, Reynolds K, et al. Evaluation of RxNorm in Ambulatory Electronic
Prescribing. TR-941-CMS. Santa Monica, CA: Rand Corporation, 2011. http://www.
rand.org/pubs/technical_reports/TR941 (accessed Feb 2013).
Levinson DR. The Food and Drug Administration’s National Drug Code Directory.
Department of Health and Human Services, Ofﬁce of Inspector General, 2006.
http://oig.hhs.gov/oei/reports/oei-06-05-00060.pdf (accessed Feb 2014).
United States Department of Health and Human Services (DHHS). Health
Information Technology: Initial Set of Standards, Implementation Speciﬁcations,
and Certiﬁcation Criteria for Electronic Health Record Technology; Final Rule 45 CFR
Part 170. http://edocket.access.gpo.gov/2010/pdf/2010-17210.pdf (accessed Feb
2014)
Hernandez LM ed. Standardizing Medication Labels: Confusing Patients Less.
Washington, DC: National Academies Press, 2008.
Bailey SC, Persell SD, Jacobson KL, et al. Comparison of handwritten and
electronically generated prescription drug instructions. Ann Pharmacother
2009;43:151–2.
Wolf MS, Shekelle P, Choudhry NK, et al. Variability in pharmacy interpretations of
physician prescriptions. Med Care 2009;47:370–3.
Shrank WH, Agnew-Blais J, Choudhry N, et al. The variability and poor quality of
medication container labels: a prescription for confusion. Arch Intern Med
2007;167:1760–5.
National Council for Prescription Drug Programs (NCPDP). Universal Medication
Schedule (UMS) April 2013. http://www.ncpdp.org/Education/Whitepaper?page=2
(accessed Feb 2014).
Wolf MS, Curtis LM, Waite K, et al. Helping patients simplify and safely use
complex prescription regimens. Arch Intern Med 2011;171:300–5.
Whittemore Ken Jr. “Guidelines to Creating High-Quality Electronic Prescriptions in
the Ambulatory Healthcare Setting,” Surescripts LLC, 2012. http://surescripts.com/
docs/default-source/Resources/quality_e-prescription_guidelines_surescripts.pdf?
sfvrsn=0 (accessed April 2014)
Dhavle , Ajit A. “Best Practices and Implementation Clinical Quality Checklist v1.0c”
Surescripts LLC, 2012. http://surescripts.com/docs/default-source/Resources/best_
practices_implementation_clinical_quality_checklist_surescripts.pdf (accessed Apr
2014).
Campbell EM, Guappone KP, Sittig DF, et al. Computerized provider order entry
adoption: implications for clinical workﬂow. J Gen Intern Med 2009;24:21–6.
Aarts J, Ash J, Berg M. Extending the understanding of computerized physician
order entry: implications for professional collaboration, workﬂow and quality of care.
Int J Med Inform 2007;76(suppl 1):S4–S13.
van der Sijs H, Aarts J, Vulto A, et al. Overriding of drug safety alerts in
computerized physician order entry. J Am Med Inform Assoc 2006;13:138–47.
Dhavle AA, et al. J Am Med Inform Assoc 2014;0:1–4. doi:10.1136/amiajnl-2014-002738