Hofstra Horizons Research

Health Care Professionals and Wrong Drug Mistakes: Can Language Research Help?

Scott R. Schroeder, PhD, Assistant Professor of Speech-Language-Hearing Sciences, Hofstra University
Speech-Language-Hearing Sciences Programs

The next time you go to the pharmacy to pick up a prescription, check to make sure that you have received the right medicine.

If you take Hydroxyzine for your allergies, you might have mistakenly been given Hydralazine, a drug that treats high blood pressure. If you take Celebrex® for your arthritis, you might have accidently received Celexa®, which is an antidepressant. These drug names (Celebrex® and Celexa® and Hydroxyzine and Hydralazine) are called look-alike sound-alike drug names, of which there are many, thousands even. (See Table 1.) Fortunately, health care professionals, such as doctors, nurses, pharmacists, and technicians, are trained to not mistake similarly named drugs, so, most likely, they will not make a mistake on your next prescription. Nevertheless, as the proverb goes, to err is human, and sometimes wrong-drug errors occur. In fact, each year up to 3.9 million of these errors may be made in community pharmacies (Lambert, Lin, & Tan, 2005).

When wrong-drug errors occur, they are often quite harmful. For example, a patient once received Durasal® (a wart remover) instead of Durezol® (an eye ointment) and put the wart remover in their eye. Ouch! (See Figure 1.) Recently, patients taking Brilinta® (a blood thinner) received Brintellix®, an antidepressant that can have many unpleasant side effects, including vomiting, dizziness, and sexual dysfunction. (Note that because of these errors, Brintellix® has been changed to Trintellix®). Errors like Brilinta®-Brintellix® are harmful in two ways. One is rather obvious – the patient is taking a drug that they do not need, and taking this drug can lead to highly undesirable consequences (such as vomiting, dizziness, and sexual dysfunction). The second, perhaps less obvious, consequence of the error is that the patient fails to receive a drug that they need. For example, a patient who takes the blood thinner Brilinta® might suffer a heart attack if they do not receive the blood thinner.

How can this happen? How can well-trained and well-intentioned health care professionals make errors of this magnitude, and do so millions of times each year? Mistakes like these can be made in multiple ways. A doctor might enter in a prescription for Brilinta® by typing in the first three letters (b-r-i), and then the name Brintellix® pops up, and the doctor accidentally selects Brintellix®. Or suppose the doctor calls in a prescription for Brilinta® to the pharmacy, and the pharmacist mis-hears the name of the drug as Brintellix® because the pharmacy is busy, and there is a lot of noise in the background. Or perhaps the pharmacy technician goes to retrieve Brilinta® from the shelf, becomes preoccupied on the way over there, and then mistakenly picks up Brintellix.

Table 1:
A Short List of Look-Alike
Sound-Alike Drug Names

Tramadol – Toradol

Cymbalta® – Symbyax®

Vinblastine – Vincristine

Clonidine – Klonopin®

Metformin – Metronizadole

Prednisone – Prednisolone

Zyrtec® – Zantac®

Famotidine – Furosemide

Researching why a patient received an antidepressant like Brintellix® instead of a blood thinner like Brilinta® seems odd for someone like me, who teaches in the Speech-Language-Hearing department and does psycholinguistics research. But actually these errors are psycholinguistic errors – in other words, the brain (the “psycho” part of the word psycholinguistics) has confused two similar words (the “linguistic” part of psycholinguistics). As psycholinguistics researchers, my colleagues and I have been studying look-alike sound-alike drug name confusion errors for the last few years, with the ultimate goal of decreasing the rates of these errors and increasing patient safety.

Before I started researching the problem of drug name errors, a promising solution to the problem had already been put forward. The idea was simple yet very clever, and most importantly, very promising. The idea was (and still is) called Tall Man Lettering, and it entails capitalizing the letters that differ between similarly named drugs. For example, Hydroxyzine and Hydralazine would appear as hydrOXYzine and hydrALAzine. The difference between the two drug names (in other words, the OXY and ALA) is emphasized, drawing your eyes to the particular letters that crucially distinguish these names. (See Figure 2.) This idea struck many health care professionals as incredibly smart, and so the idea was put into practice across the United States. In fact, the Joint Commission, which is the organization that accredits a very large number of health care organizations and programs in the United States, has strongly encouraged the use of Tall Man Lettering, and many organizations and programs have complied, making Tall Man Lettering highly prevalent.

Figure 1: A patient once mistakenly received Durasal®, a wart remover, instead of Durezol®, an eye ointment, resulting in the patient putting the wart remover in their eye.
Retrieved from:

But does Tall Man Lettering actually reduce the number of drug name errors? My colleagues and I argued in an editorial in the British Medical Journal: Quality and Safety that, to date, there is no compelling evidence that Tall Man Lettering truly works (Lambert, Schroeder, & Galanter, 2016). There have been many artificial laboratory experiments that have examined whether people make fewer errors when Tall Man lettering is used. While some of these studies are suggestive of an advantage of Tall Man Lettering over normal lettering, many other studies fail to show a benefit of Tall Man Lettering.

Even when a benefit is shown, it is unclear whether results from an artificial laboratory setting translate into real-world clinical practice. Fortunately, a much-needed study recently addressed the real-world effectiveness of Tall Man Lettering, by assessing rates of drug name errors across 42 children’s hospitals from 2004 to 2012 (Zhong, Feinstein, Patel, Dai, & Feudtner, 2016). What did they find? They found that Tall Man Lettering did not reduce the number of drug name errors. Of course, no single research study provides a definitive ruling, and future studies might find that Tall Man Lettering is indeed effective, but at this point, my colleagues and I believe that major skepticism is warranted.

Figure 2: To decrease the chances of drug name mistakes, labels are often printed in Tall Man Lettering. This special type of lettering involves capitalizing letters that differ between similarly named drugs. For example, Hydroxyzine and Hydralazine are similarly named, so the differentiating letters (OXY and ALA) are capitalized, resulting in hydrOXYzine and hydrALAzine.
Retrieved from:

If the effectiveness of Tall Man Lettering is questionable, what can we do to reduce drug name errors? My colleagues and I are taking a proactive approach to the problem. In other words, we are trying to prevent confusing drug names from entering the market in the first place (i.e., before they can do any harm). While the U.S. Food and Drug Administration (FDA) has a process for evaluating proposed drug names, confusing names still slip through the cracks and make it onto the market. My colleagues and I have created a battery of tests for proposed drug names that federal regulators (such as the FDA) can utilize to improve their ability to detect bad names that are prone to confusion and thus should not be approved.

Our battery of tests consists of simple video game-style cognitive tests that assess people’s perception and memory of drug names. For example, in a visual perception test, a drug name (such as Hydroxyzine) is flashed on the computer screen extremely rapidly (less than a tenth of a second). Then, two names pop up on the screen (Hydroxyzine and Hydralazine), and the person has to choose which drug name they believe they saw. In an auditory version of the perception test, a person hears a drug name like Hydroxyzine but at the same time, they hear 20 people babbling in the background. Then, as in the visual perception test, two names pop up on the screen (Hydroxyzine and Hydralazine), and the person has to choose which drug name they believe they heard. In a memory test, a drug name is displayed on the screen, and then the person solves a math problem (for example, 2455 + 1392). After solving the math problem, the person has to try to remember which drug name they saw.

The claim we are making about these cognitive tests is that if a given drug name is mistaken quite frequently on the tests, then that drug name should not go on the market. But, of course, a claim needs evidence if it is to be taken seriously. To assess the claim, my colleagues and I tested 80 participants (both health care professionals and patients) on this battery of tests, and then we analyzed their error rates on particular drug name pairs (such as Hydroxyzine-Hydralazine). Specifically, we looked at how the error rates on particular drug name pairs compared to the error rates of those same drug names in two large community pharmacy chains in the United States. We found that error rates on our tests could predict the error rates in both community pharmacy chains. In other words, if participants made a lot of Hydroxyzine-Hydralazine errors on our test, then Hydroxyzine-Hydralazine had a high error rate in real-world pharmacy chains, a result that we published in the British Medical Journal: Quality and Safety (Schroeder et al., 2017). The important implication of this result is that when a drug name is proposed and goes up for evaluation by federal regulators, the regulators can insert the proposed drug name into these cognitive tests and determine whether the drug name is likely to cause real-world errors and thus should not be allowed to enter the market. Recently, we conducted a follow-up study with different participants and different drug names, and we replicated the primary finding: Cognitive tests predict real-world drug name errors. These new results, which have yet to be submitted for publication, lend further credence to our claim.

But what about the drug names that are already on the market and are likely to continue to be confused, such as Hydroxyzine and Hydralazine and Celebrex® and Celexa®? In addition to our proactive efforts to prevent bad names from entering the market, reactive efforts are also being made to reduce errors involving names that are already on the market. In this vein, my colleagues and I recently conducted a very small pilot study that assessed an unusual approach to reduce drug name errors. The approach is based on the notion that the mind has two distinctive cognitive systems, an idea popularized in the best-selling book Thinking, Fast and Slow by Daniel Kahneman (Kahneman, 2011). System 1 is a fast system that involves automatic processing, whereas System 2 is a slow system that involves effortful processing. My colleagues and I hypothesize that pharmacists are often functioning in System 1 mode when hearing or seeing a drug name. Pharmacists encounter the same drug names over and over, day in and day out, which can lead them to mentally process these names in auto-pilot mode. The problem with operating in automatic mode is that errors are likely to occur. Perhaps, if we can nudge pharmacists from System 1 into the more deliberate System 2, they will make fewer errors.

How then can we make the pharmacist’s routine and automatic job of reading drug names from bottles, packages, and computer screens more deliberate?

One possible solution is to put the drug names in harder-to-read fonts, such as Hydroxyzine or Hydroxyzine or Hydroxyzine. By making the drug names harder to read, the pharmacist is forced to slow down and engage System 2. In this more deliberate mind frame, pharmacists may be less likely to make mindless mistakes. In a very small pilot study, our results were promising, but until a full study is conducted and passes through peer review, it is premature to claim that changing fonts is likely to reduce drug name errors.

My colleagues and I are continuing our fight against wrong-drug errors. With the belief that difficult problems are often best solved with a diverse group of people, we are taking an interdisciplinary approach. Our team includes language researchers, health communication researchers, medical doctors, pharmacists, computer scientists, and statisticians from several institutions across the United States, including Northwestern University in Illinois, University of Illinois at Chicago, Brigham and Women’s Hospital in Massachusetts, the Institute for Safe Medication Practices in Pennsylvania, and of course, Hofstra University in New York. Together, we are making progress on the problem of wrong-drug errors, but until the problem is solved, please check that you receive the right medication on your next visit to the pharmacy.


  • Kahneman, D. (2011). Thinking, fast and slow. Macmillan.
  • Lambert, B. L., Lin, S.-J., & Tan, H. (2005). Designing safe drug names. Drug Safety, 28(6), 495-512.
  • Lambert, B. L., Schroeder, S. R., & Galanter, W. L. (2016). Does Tall Man lettering prevent drug name confusion errors? Incomplete and conflicting evidence suggest need for definitive study. BMJ Qual Saf, 25(4), 213-217.
  • Schroeder, S. R., Salomon, M. M., Galanter, W. L., Schiff, G. D., Vaida, A. J., Gaunt, M. J., … Lambert, B. L. (2017). Cognitive tests predict real-world errors: The relationship between drug name confusion rates in laboratory-based memory and perception tests and corresponding error rates in large pharmacy chains. BMJ Qual Saf, 26(5), 395-407.
  • Zhong, W., Feinstein, J. A., Patel, N. S., Dai, D., & Feudtner, C. (2016). Tall Man Lettering and potential prescription errors: A time series analysis of 42 children’s hospitals in the USA over 9 years. BMJ Qual Saf, 25(4), 233-240.


Hofstra in the News

More In The News
Hofviews - See More Photos

Hofstra Weather

° F
Heat Index ° F / Wind Chill ° F
Humidity %
Wind mph / Direction °
Rain in
See More from the Project WX Weather Stations

Recent Faculty News