Are National Efforts to Reduce Drug Name Confusion Paying Off?
One in every 1,000 medication orders in a hospital, and one in every 1,000 prescriptions in a pharmacy, have been associated with selecting the wrong drug while prescribing, transcribing, dispensing, or administering medications.1-4 Drug name similarities are a primary cause of these errors.5 Orthographic (spelling) factors that increase visual resemblance among drug names include similarities in the length of the names and the number of groups of similar or the same characters within the names.6 Phonological (sound) factors that increase auditory resemblance among drug names include similarities in the number of syllables, the stressed syllable, the initial or terminal syllable, and the stressed vowel.6 Other factors that increase the risk of drug name confusion include similarities in strength, dosing, route of administration, dosage forms, indication, and other factors, such as the environment in which the drugs are used, the frequency of use, and product labeling.7
Sources of name confusion. Sources of drug name confusion errors include: memory, perceptual, and motor control errors.6 Memory errors can arise when practitioners make a mistake during recall or recognition of a drug name. Perceptual errors occur when practitioners misread or mishear a drug name. Motor control errors occur upon selection of a drug. For example, this type of error occurs when an adjacent drug with a similar name is selected in error from a list, such as a drop-down set of choices on a computer screen.
Drug naming processes. Generic (nonproprietary) drug names are based upon a collection of standard stems used as prefixes, suffixes, and infixes to identify the pharmacologic property and/or chemical structure of the medication. In the US, generic names are assigned by the United States Adopted Names (USAN) Council. In the global arena, the World Health Organization (WHO) International Nonproprietary Name (INN) members work with international naming authorities like USAN to harmonize generic names between different countries. Proposed generic names are released for public review and comment. Sometimes, the drug name stems embedded in generic names contribute to mix-ups among names with the same stem. However, the stem helps position an unfamiliar drug with others in a class and provides clues as to its use and effects.
Brand (proprietary) names for drugs are selected by the manufacturer. As part of the drug product approval process, the US Food and Drug Administration (FDA) reviews the proposed brand name and determines its acceptability. Brand names are intended to be unique and memorable to identify products and distinguish one manufacturer’s product from its competitors. However, brand names that look or sound alike can contribute to name confusion errors.
Name safety testing. For generic drug names, USAN Council members (one each from the American Medical Association, USP, American Pharmacists Association, FDA, and a member at large) conduct an evaluation during the naming process to reduce the risk of similarities with existing brand or generic drug names. One of the guiding principles associated with the USAN Council naming process includes criteria that the name should not conflict, mislead, or be confused with other nonproprietary or proprietary drug names.
Over the past decade or more, the pre-market safety evaluation of proposed brand names has become more extensive and structured. FDA issued voluntary proprietary naming guidances in 20088 and 2014,9 and the role of regulatory authorities in the review of brand names has increased significantly.10 Today, before launching a new drug, many pharmaceutical companies voluntarily use external safety testing companies to evaluate potential risks associated with proposed brand names, including name similarities. Med-ERRS, a wholly owned subsidiary of ISMP, is one of the companies that conducts this testing. The testing, which often involves practicing healthcare practitioners who may prescribe, dispense, and/or administer the new drug, identifies similarities with existing drug or medical product names, medical terms, and abbreviations that may lead to confusion. A computer software program, POCA (phonetic and orthographic computer analysis), is also used by external safety testing companies and FDA to evaluate name similarity.
It is important to note that pharmaceutical companies are NOT required by regulation to test and evaluate their proposed brand names for potential name similarities, so many pharmaceutical and biotech companies, including generic manufacturers and distributors, have not adopted this practice. According to FDA, in 2017, only 57% of the submissions for new drug approvals were accompanied by testing and evaluation results supporting the proposed new brand name.11 Still, the FDA Division of Medication Error Prevention and Analysis (DMEPA) evaluates ALL brand names presented with products submitted for approval using practitioner name simulation studies, POCA results, and input from the FDA review team for that product. DMEPA conducts this independent evaluation of brand name similarity, even if the company conducts its own safety evaluation of a brand name and submits the results to FDA.
Is name confusion declining? With more than 27,000 drug products currently on the US market, creating new drug names that are not similar to existing drug names is challenging.10 But, has a decade of efforts to evaluate brand names for possible name confusion prior to launch made a difference? Have significant advances in technology, including electronic prescribing, barcoding, and other practices during this time contributed to fewer errors associated with name confusion? Because little is known about the true incidence of drug name confusion,6 ISMP conducted a retrospective analysis of name-related medication errors voluntarily reported to ISMP to examine the percent of change in reporting over time to begin to answer these important questions.
Since 1994, ISMP has operated the voluntary, practitioner-based ISMP National Medication Errors Reporting Program (ISMP MERP). Today, the ISMP MERP receives more than a thousand medication-related error reports annually from physicians, pharmacists, nurses, and other healthcare practitioners who prescribe, dispense, and/or administer medications to patients in a wide variety of settings, including hospitals, long-term care facilities, infusion centers, community pharmacies, and other treatment locations.
Two chronological samples of ISMP MERP error reports (submitted between 2000 and 2004, and between 2012 and 2016) were extracted for analysis and compared to determine whether drug name confusion reports had increased or decreased between the two time periods, and whether the types of name confusion reports had changed over time. The narrative in each report was reviewed, and based on the description of the hazard or error, the reports in each data set were categorized into four groups: 1) name confusion between two proprietary drug names (brand-brand); 2) name confusion between a proprietary and nonproprietary name (brand-generic); 3) name confusion between two nonproprietary drug names (generic-generic); and 4) all other reports that were not associated with drug name confusion (other reports). Errors associated with labeling and packaging (not related to name confusion), drug name modifiers (e.g., LA, ER, XL), brand name extensions, or differences in formulations were categorized as other reports. Reports involving the same name pair were each included separately.
A Pearson’s Chi-squared test of independence with Yates’ continuity correction was used to test: 1) whether the frequency of name- and non-name-related error reports was different between time periods; and 2) whether the frequency of the three different types of name-related reports (brand-brand, brand-generic, and generic-generic) differed between time periods. The Pearson’s Chi-squared test of independence was used to assess if observed frequencies deviated from expected, random frequencies for categorical data. The Yates’ continuity correction conservatively estimates the P values in Chi-squared analyses, and is recommended when analyses have low degrees of freedom, as in this study. Pearson’s residuals in post-hoc tests were also calculated to assess which of the specific associations in each Chi-squared test was statistically significant. The null hypothesis for the first test was that there was no association between name- and non-name-related error reports and the time period. The null hypothesis for the second test was that there was no association between the type of name-related error reports and the time period.
A total of 4,091 reports were submitted between 2000 and 2004, of which 816 were related to drug name confusion and 3,275 were classified as other types of events (Figure 1). A total of 6,206 reports were submitted between 2012 and 2016, of which 603 were related to drug name confusion and 5,603 were classified as other types of events. Among the 816 reports of drug name confusion submitted between 2000 and 2004, 507 involved brand-brand name confusion, 91 involved brand-generic name confusion, and 218 involved generic-generic name confusion (Figure 2). Among the 603 reports of drug name confusion submitted between 2012 and 2016, 183 involved brand-brand name confusion, 51 involved brand-generic name confusion, and 369 involved generic-generic name confusion.
We found strong evidence that name-related reports were significantly less common in 2012-2016, while all other types of reports significantly increased over this time span (Figure 1). There were 52% more total reports in the 2012-2016 period than in the 2000-2004 period. Name-related reports declined by 26% in the years 2012-2016, but the number of non-name-related reports increased by 71%. There was a significant association between the reporting frequency of name- and non-name-related reports and the time period (χ2 = 216.31; df = 1; P < 2.2 x 10-16). The Pearson’s residuals (Table 1) indicated that the reporting frequency of name-related reports was significantly lower than expected in 2012-2016 (R = -8.62), and significantly greater than expected in 2000-2004 (R = 10.62). Furthermore, there were significantly fewer non-name-related reports than expected in 2000-2004 (R = -4.24).
We also found strong evidence that reporting of brand-brand name-related confusion significantly decreased over time, while reporting of generic-generic name-related confusion significantly increased over time (Figure 2). There were 26% less name-related reports submitted in 2012-2016 than in 2000-2004, primarily due to drops in brand-brand name and brand-generic name reports, which declined by 64% and 44% respectively. In contrast, generic-generic name-related reports increased by 69%. There was a significant association between name-related reports and time period (χ2 = 174.20; df = 2; P < 2.2 x 10-16). Specifically, there was a lower proportion of brand-brand name confusion reports than expected in 2012-2016 (R = -6.43), and a greater proportion of generic-generic name confusion reports than expected in 2012-2016 (R = 7.57). However, the proportion of brand-generic name confusion reports did not differ from expectation in 2012-2016 (Table 2).
Similarity between drug names has been a frequent cause of medication errors. As expected, all types of drug name confusion—brand-brand name, brand-generic name, and generic-generic name—were reported in all the years studied. However, we observed a decrease in the reporting of all types of name confusion in 2012-2016 when compared to 2000-2004. We are unable to determine the reason for the change in reporting frequency between the two time periods. However, it is unlikely that practitioners were less motivated between 2012-2016 than between 2000-2004 to report name- related confusion and errors to ISMP. So it may be plausible that the overall reduction in the reporting of name confusion errors of all types in 2012-2016 is due to national efforts to reduce drug name-related confusion, including advances in technology such as electronic prescribing (which eliminates handwritten prescriptions that risk misinterpretation) and barcode scanning (which can help detect and correct an error due to drug container name confusion). Practice improvements such as reducing verbal orders, tagging problem name pairs in computer databases to aid clinical decision support, expanding the use of tall man letters, and including an indication on prescriptions, also may have impacted the occurrence (and subsequently the reporting) of drug name confusion.
We also found that name-related confusion reporting appears to have switched from predominantly brand-brand name confusion in 2000-2004, to predominantly generic-generic name confusion in 2012-2016. The change from brand-brand to generic-generic name confusion error reporting may be due to the evolution of FDA and manufacturer testing of brand names prior to approval to ensure they have a low potential for confusion and are safe to use in the healthcare environment. The ever-increasing market share for generic medications, which accounts for the bulk of outpatient prescriptions in the US, also may have played a role in the increased reporting of generic-generic name confusion. In 2002, only about half of outpatient prescriptions were for generic medications; in 2016, generic medications accounted for 90% of all outpatient prescriptions.12 Other factors that may have contributed to an increase in generic-generic name confusion reporting include an increase in the use of generic drug names, the expanding number of generic drug names that utilize the same stem within a therapeutic class, assignment of similar stem names (e.g., -umab and -ximab), and the use of longer USAN stems.
Managing the risks associated with name similarity is an industry-wide obligation. It begins with pharmaceutical companies that propose generic and/or brand names, and with regulatory and standards organizations that approve the names. The increase in reporting of generic-generic drug name confusion suggests possible vulnerabilities in the way generic drug names are assigned, evaluated, and approved. While USAN has remained open to changing the generic name of a product if post-marketing surveillance shows harmful or potentially harmful confusion with another generic drug name, changing a name is a complex and lengthy process and should not be relied upon as a risk mitigation strategy. Instead, FDA, USP, and USAN should work with industry leaders to develop a more robust, standard evaluation method for nonproprietary names to be employed prior to generic name assignment. Growth in generic drugs and biopharmaceutical products will likely require new funding methods to allow for computerized screening of proposed generic names and field testing with practitioners.
Although FDA evaluates brand names prior to drug approval, requiring ALL pharmaceutical companies to use an independent source to test proposed brand names to identify and remedy potential look- and sound-alike confusion with existing drug names, and to submit their results to FDA when seeking new drug approval, can further reduce name similarities that may cause serious errors. In addition, there should be a consistent and standardized approach regarding the methods employed to determine the acceptability of a brand name.10 Furthermore, FDA should require companies to develop a risk management program that includes a name change provision for newer brand names if post-marketing surveillance (including error reports) shows harmful or potentially harmful confusion with an existing brand or generic name.
Healthcare providers can also reduce the risk of drug name confusion by implementing strategies to prevent errors (e.g., indication-based prescribing, computer listing of both brand and generic names, separate storage, tall man letters, electronic alerts for look- and sound-alike names). Accreditors should provide assistance to ensure this is successful.
This study examined the differences in reporting of name confusion to the voluntary, practitioner-based ISMP MERP during two different time periods. The data sets did not include all name confusion events occurring in the US during the study time periods. Thus, our study was only able to detect variations in the voluntary reporting of name-related confusion events to ISMP, and our results are not generalizable to all US reporting programs. We were also limited in our ability to determine why the changes in reporting occurred. Thus, plausible explanations for the changes were based upon expert opinion and not scientific evidence.
The volume of medication error reporting to the ISMP MERP has increased over time. While the reporting of drug name confusion of all types, particularly brand-brand name confusion, has decreased over time, the reporting of generic-generic drug name confusion has increased and is likely to continue increasing as the US market share of generic medications rises. Future work should attempt to reduce the risk of generic-generic drug name confusion through better pre-market evaluation of generic names along with post-market monitoring and action if serious or potentially serious drug name confusion errors occur.
ISMP thanks Alexander Radovanovich, PharmD, Medication Safety Fellow at Novartis Pharmaceutical Corporation, for analysis of the data while on rotation at ISMP. We also extend our appreciation to Dorothy Linvill-Neal, Global Head, Name Creation & Regulatory Strategy at Novartis for suggestions made regarding this project.
- Lambert BL, Lin SJ, Tan H. Designing safe drug names. Drug Saf. 2005;28(6):495–512.
- Lambert BL, Schroeder SR, Galanter WL. Does tall man lettering prevent drug name confusion errors? Incomplete and conflicting evidence suggest need for definitive study. BMJ Qual Saf. 2016;25(4):213-7.
- Flynn EA, Barker KN, Carnahan BJ. National observational study of prescription dispensing accuracy and safety in 50 pharmacies. J Am Pharm Assoc (Wash). 2003;43(2):191–200.
- Cina JL, Gandhi TK, Churchill W, et al. How many hospital pharmacy medication dispensing errors go undetected? Jt Comm J Qual Patient Saf. 2006;32(2):73–80.
- DeHenau C, Becker MW, Bello NM, Liu S, Bix L. Tallman lettering as a strategy for differentiation in look-alike, sound-alike drug names: the role of familiarity in differentiating drug doppelgangers. Appl Ergon. 2016;52:77-84.
- Lambert BL, Chang KY, Lin SJ. Effect of orthographic and phonological similarity on false recognition of drug names. Soc Sci Med. 2001;52(12):1843-57.
- ISMP Canada. Application of TALLman lettering for selected high-alert drugs in Canada. ISMP Canada Safety Bulletin. 2015;15(10):1-3.
- FDA. PDUFA Pilot Project. Proprietary name review. Concept paper. September 2008.
- US Department of Health and Human Services, FDA, Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research. Guidance for industry: best practices in developing proprietary names for drugs. Draft Guidance. May 2014.
- Linvill-Neal D. Naming your pharmaceutical. In: DiSpirito NP, Hall RF, Hill MJ, eds. Bringing your Pharmaceutical Drug to Market. Washington, DC: Food and Drug Law Institute; 2015:219-33.
- Bridges T, Harris D. US Food and Drug Administration update: proprietary name review program. Division of Medication Error Prevention and Analysis, FDA. Presentation at the Pharmaceutical Research and Manufacturers of America (PhRMA) meeting, July 24, 2017.
- Kesselheim AS, Avorn J, Sarpatwari A. The high cost of prescription drugs in the United States. Origins and prospects for reform. JAMA. 2016;316(8):858-71.