Nursing Research

Research Article

Using QR Code Technology to Reduce Self-Administered Medication Errors

Johanna Svensk, BS1, and Scott E. McIntyre, PhD1

Abstract Background: Data indicate there are tens of thousands of self-administered medication errors each year in the United States alone. Objective: The aim of this study was to determine whether information embedded in Quick Response (QR) codes could reduce self-administered medication errors compared to current medication labeling among older and younger age groups. Methods: Two population samples (Arizona State University undergraduates and senior citizens over 70; n¼ 55) were recruited for participation. Participants were randomly assigned to 2 groups: one with access to QR code-based information (graphic and text) and a second group with only bottle label information. Participants were allowed 30 minutes to answer 17 scenario-based questions about administering their medications. Results: Statistically significant main effects of more correct answers when using QR code than current bottle labeling, F1, 51¼ 181.57, P < .001, Z2¼ 0.78, and for younger adults compared to older, F1, 51¼ 24.4, P < .001, Z2 ¼ 0.33. Conclusion: The study supports the use of QR code technology to increase patient safety of self- administered medications in both older and younger age groups. Future research is needed to address the technological and usability aspects of implementation (eg, phone app, voice, graphic, and text presentation).

Keywords medication labeling, patient safety, self-administration, QR code technology

Introduction

Over the years in the health-care profession, medication

errors, including missed dosages, wrong dosages, and unin-

tended drug interactions, have mitigated the effects of high-

quality patient care. Today, patient error and confusion when

reading and understanding medication labels is prevalent and

results in a significant amount of disabilities, hospitalizations,

and even death. In fact, the US Food and Drug Administration

(FDA)1 receives tens of thousands of reported medication

errors every year of which more than one-third are from

self-administered medications.

According to the National Coordinating Council for Medi-

cation Error Reporting and Prevention, a medication error is

defined as “any preventable event that may cause or lead to

inappropriate medication use or patient harm while the medi-

cation is in the control of the healthcare professional, patient or

consumer.”2 This would include, for example, taking a medi-

cation at the wrong time, at the wrong dosage, or inadvertently

with another medication that creates a harmful side effect.

One of several factors contributing to outpatient medication

error is insufficient labeling on both prescribed and over-the-

counter drugs and supplements.3 This was confirmed in a study

indicating that “up to 25% of all medication errors are attrib-

uted to name confusion and 33% to packaging and labeling

confusion.”3 A similar study conducted by Miriam Klein and

Henry Cohen4 identified risks associated with poor labeling of

intravenous drug administration and emphasized the impor-

tance of implementing new labeling systems to improving

patient safety. It is this type of perceptual label confusion that

can cause patients, or even caregivers, to maladminister med-

ication with serious consequences.

In addition, interactions, especially for those taking several

prescription drugs and/or supplements, unaware of the conse-

quences, can create serious, harmful patient complications.5

For example, warfarin, which is a prescription blood anticoa-

gulant, and Ginkgo biloba, a commonly used herbal supple-

ment, can each reduce blood clotting; consequently, “taking

any of these products together may increase the potential for

internal bleeding or stroke.”5 This is especially problematic in

older patients who generally have more “complex clinical

problems and take multiple treatments,” as their cognitive abil-

ities may be impaired, increasing the likelihood of errors.6

Overall, it is evident that clear, concise, and readily available

information is essential to reducing the risk of misadministra-

tion and interactions of medications. This would also assist in

1 Arizona State University, Lake Havasu City, AZ, USA

Corresponding Author:

Scott E. McIntyre, Arizona State University, 100 University Way, Lake Havasu

City, AZ, USA.

Email: scott.mcintyre@asu.edu

Journal of Pharmacy Practice 2021, Vol. 34(4) 587-591 ª The Author(s) 2019 Article reuse guidelines: sagepub.com/journals-permissions DOI: 10.1177/0897190019885245 journals.sagepub.com/home/jpp

raising awareness of the unintended consequences of mixing

and combining certain dietary supplements and medications.

Furthermore, people tend to ignore or discard the sometimes

complex and potentially confusing printed information pro-

vided with their prescribed medications.7 For instance, in one

study conducted in community pharmacies and patients’

homes in England, 456 patients were asked questions about

their patient information leaflets (PILs), which are written

instructions received with every prescribed medication bottle.

The results indicated that 97% of all patients were aware of

the leaflets; however, 87% of frequent users of prescriptions

“had never or rarely looked at the leaflet after the first time.”8

In fact, these PILs tend to be thrown away after a patient

returns home from the doctor’s office.7 This has serious impli-

cations, since important information may change over time or

new medications or supplements may be added to a person’s

daily intake after the medication has been prescribed by a

doctor. This also underscores the importance of developing

new ways to improve patient access to critical information

and thereby improving administration of medication for the

best possible outcome.

Several studies point to promising new technological solu-

tions that can reduce medication errors and improve patient

outcomes. A vast amount of literature and research have sug-

gested that the use of barcoding on medications, by health-care

professionals in hospitals and other health-care settings,

improves patient care.9-13 For example, in one study conducted

in 13 nursing homes, it was found that the use of a hand-held

device, called a pharmacy-led barcode medication system,

improved registered nurses and care staff awareness of drug

errors and reduced stress and confusion.13 The barcode system

included specific information about the patient, dosages, pre-

scribed times, and so on. Overall, it was found that this system

made it more efficient for trained health-care professionals to

properly administer medications to their patients.

Another study discussed the barcode medication adminis-

tration system, which allowed educated and trained nurses to

scan “the patient’s wristband barcode and the barcode on the

medication to be administered.”9 If an issue was flagged

regarding this administration of medication, the system would

alert the nurse to not administer the medication; this, as a result,

reduced errors. However, currently, this barcoding system is

currently only used by less than 25% of US hospitals.9

Although procedures utilizing barcoding are becoming more

and more common in health-care settings, the effectiveness of

barcoding on self-management warrants further research. In

one previous study, 61 patients taking multiple medications

were asked to evaluate the usefulness of information provided

by an app reading a barcode on their medications packaging.12

They were provided 2 different questionnaires, where they

responded subjectively on perceptions and experiences utiliz-

ing this technology. The results indicated that the patients were

favorable to this intervention, finding it “useful for safer use of

medicines.”12

With this in mind, this current study seeks to build on and

complement previous literature and studies by proposing that

implementation of Quick Response (QR) code technology,

which can provide substantially more embedded information

than bar coding, could benefit not only trained health-care

professionals but also patients and caregivers self-

administering medication by providing convenient access to

important information in much more detail than is available

on current prescription and over-the-counter labels. Accord-

ingly, the focus of the current research is to determine the

extent to which QR code technology can improve medication

labels to reduce medication errors among self-administered

individuals. It seemed prudent to first see if such information

would be of any benefit before developing an app for a QR

code reader and deciding on the mode of information delivery

(voice, text, and visualizations) that could be pursued in a

human factors–type usability study. Thus, in this initial study,

participants received text output that could be contained in a

QR code and did not use an actual QR reader.

Hypotheses

� Hypothesis 1: Regardless of age, the QR code will result

in less errors due to conveniently available medication

information.

� Hypothesis 2: Older age groups will make more errors

than younger age groups in both the experimental and

the control groups.

� Hypothesis 3: Correct answers per minute will be higher

in the experimental groups who receive additional

access to a QR code relative to the control groups.

Methods

Participants

Thirty-three (both men and women, aged 18 years or older)

younger adults attending Arizona State University (ASU) in

Lake Havasu City, Arizona, were recruited with flyers posted

on the ASU campus and via e-mail. Twenty-two (both men and

women, aged 70 years or older) older adults were recruited by

word of mouth and flyers at the local senior center. The college

student participants were compensated by receiving course

credit if approved by their instructor; seniors were not given

any compensation. The study was approved by the Arizona

State University Institutional Review board, Study 9505.

Materials

A questionnaire, consisting of 17 questions, was given to each

subject (see Supplementary Appendix A). It covered a range of

scenarios to determine whether an independently living patient

had sufficient knowledge to safely administer a combination of

prescribed medications and supplements. The subjects were

asked to circle the correct answers that corresponded to the

information provided to them.

Empty prescribed medication bottles (Warfarin) and over-

the-counter supplement bottles (Ginkgo biloba and CBD:

588 Journal of Pharmacy Practice 34(4)

Cannabidiol oil; See Supplementary Appendix B) were used in

the control group to simulate real-life medications and supple-

ments taken by a patient. The experimental group, on the other

hand, was given printed versions of pertinent information for

all 3 medications (Warfarin, Ginkgo biloba, and CBD oil) that

could be stored in QR codes (see Supplementary Appendix C).

These printed QR codes resembled real barcodes scanned from

a phone or laptop; this allowed the researcher to design an

experiment with QR codes in an artificial environment.

Design

This study was an experimental, between-subjects design,

where 2 population samples, one from Arizona State University

undergraduates and another from a population over 70 at a local

senior center (n ¼ 55), were recruited for participation. Each

sample was randomly assigned to 1 of 2 groups: 1 experimental

and 1 control group. In the experimental group, subjectswere

given a printed QR code for each medication (Warfarin,

Ginkgo biloba, and CBD oil). This allowed the subjects to

receive additional information about prescription and over-

the-counter supplements and drugs, including personally

relevant information on a prescription drug, as well as general

contraindications, precautions, and side effects, normally not

found on regular medication labeling. The control group, how-

ever, only had access to the information available on the pre-

scription or over-the-counter bottle labeling of the 3 chosen

medications. Both groups were measured by the same dependent

variable, which in this study was accuracy and correct answers

per minute. Correct answers per minute was used to account for

speed accuracy trade-offs common in tasks similar to those in

this study. Additionally, there were no statistical differences in

time to complete the questionnaire between conditions.

Procedure

The college student population was tested in a quiet confer-

ence room in a building at the ASU campus, where each

subject walked into the room either separately or in groups

for approximately 30 minutes each. The senior citizen popu-

lation was tested in a similar setting, a lunch room at the local

senior center.

Prior to the beginning of the experiment, a verbal informed

consent form was read to all subjects, emphasizing that all

responses would be anonymous in order to protect the sub-

ject’s privacy. In addition, they were informed of their right

to decline to answer any of the questions and to stop partic-

ipation at any time.

Then, each subject was randomly assigned to either the

condition with QR codes or the bottle labeling only. The ques-

tionnaire was handed out, consisting of questions about a

hypothetical person using 1 prescribed medication (warfarin)

and 2 supplements (Ginkgo biloba and CBD oil; see Supple-

mentary Appendix A). Each subject in the control group was

provided 3 medication bottles, with labels of warfarin, Ginkgo

biloba, and CBD oil, whereas the subjects in the experimental

group received pertinent information on a paper document that

would be available from a scanned QR code from each bottle:

Warfarin, Ginkgo biloba, and CBD oil.

Finally, once the subjects completed answering the ques-

tions, the questionnaires were collected, and the time it took

to finish the task was recorded. In addition, the nature of the

experiment was fully explained; they were told that it was not a

study intended to measure their intelligence but to examine the

effectiveness of medication labeling. The subjects were also

thanked for participating in the research.

Results

In all analyses, P values <.05 were considered as statistically

significant. All data from the questionnaires were analyzed

with multivariate analysis of variance using SPSS version 25.

As seen in Figure 1, the main effect (Fisher’s F) of condition

(label vs QR) indicated that both students and senior citizens

in the experimental group, the QR code condition, had signif-

icantly more correct answers (Students mean [M] ¼ 14.4,

standard error [SE] ¼ 0.61 and Seniors M ¼ 9.0, SE ¼ 0.74)

than those in the control group (Students M ¼ 2.9, SE ¼ 0.66

and Seniors M ¼ 1.3, SE ¼ 0.81), F1, 51 ¼ 181.57, P < .001,

Z2 ¼ 0.78.

Similarly, as seen in Figure 2, a significant main effect of

labeling was also found for correct answers per minute, as both

Figure 1. Correct answers for medical information with bottle labeling only or Quick Response code for students and senior citizens (bars are standard error).

Figure 2. Correct answers per minute for medical information with bottle labeling only or Quick Response code for students and senior citizens (bars are standard error.).

Svensk and McIntyre 589

age groups in the experimental group had significantly more

correct answers per minute (Students M ¼ 1.4, SE ¼ 0.08 and

Seniors M ¼ 0.73, SE ¼ 0.09) than those in the control group

(Students M ¼ 0.31, SE ¼ 0.09 and Seniors M ¼ 0.08,

SE ¼ 0.11), F1, 51 ¼ 87.9, P < .001, Z2 ¼ 0.63.

In addition, there was a significant main effect of age, where

students had significantly more correct answers than seniors,

F1, 51¼ 24.4, P < .001, Z2 ¼ 0.33, and also when measured as

correct answers per minute, F1, 51¼ 23.6, P < .001, Z2 ¼ 0.32.

Thus, there was significant interaction between age and label-

ing for both correct answers, F1, 51 ¼ 6.98, P < .001, Z2 ¼ 0.12, and for correct answers per minute, F1, 51¼ 5.7, P < .05,

Z2 ¼ 0.1.

Discussion

The results of this study were considered against the initial

hypotheses. All 3 hypotheses were supported. A significant

main effect of labeling (QR code or bottle label only) was

seen for both students and older populations, showing

increased benefits of QR code labeling for both age groups.

This supported the prediction that those using the QR code

would make less errors due to the additional medication infor-

mation. Overall, the very poor scores (<3 out of 17 possible)

based on information obtained from medication labels in this

study indicates that the current bottle labeling for prescribed

and over-the-counter drugs and supplements seems insuffi-

cient. Using a knowledge test seems an adequate proxy for

predicting medication error, as the FDA has indicated that

“ambiguities in product names, directions for use, medical

abbreviations or writing . . . patient misuse because of poor

understanding of the directions for use of the product” are

contributing factors in medication errors.14

Also, older age groups tended to make more errors than

younger age groups in both the QR code and the bottle label

condition. This supports the hypothesis that older age groups

are more prone to make errors and have difficulty comprehend-

ing and understanding the available information regardless of

the amount of additional written information provided to them.

This is not surprising, as cognitive and sensory abilities tend to

decline with age.6

Finally, the QR code condition proved to be more time

efficient than the bottle labeling condition, resulting in less

time consumption and less errors from both students and

seniors in the experimental group.

By measuring correct answers and correct answers per min-

ute on various age groups, such as undergraduate students and

senior citizens, specific factors effecting dependent variables

became evident. If all participants were measured without

regard to age differences, it would have been difficult to ascer-

tain a main effect of age on the efficacy of QR code technology

as a tool to reduce self-administered medication errors. While

speed of determining medication administration may not seem

important, it could be a factor that affects if and when assis-

tance is sought by a patient due to frustration or confusion.

However, the results showed that information stored in QR

codes help young and old alike. Thus, the benefits are not due

solely to sensory or cognitive limitations of older adults. Con-

sequently, this innovation has widespread application beyond

geriatric populations.

This study indicated that QR code technology could play a

significant role in reducing medication errors regardless of age;

however, it has its limitations. For example, due to a small

sample size (33 ASU undergraduates and 22 senior citizens), the

results should be tentatively generalized, and larger samples

could be tested. On the other hand, the main effect of QR

code versus current labeling had large effect sizes when mea-

sured be either correct answers alone, F1, 51¼ 181.57, P < .001,

Z2¼ 0.78 or correct answers per minute, F1, 51¼ 87.9, P < .001,

Z2 ¼ 0.63, and the calculated power was ¼ 1.

Another notable limitation with this present study was that

the participants who enrolled were not representative of

patients taking these specific medications. If it would have

been noted that the participants would have been taking these

specific medications themselves on a regular basis, it may have

provided different results. However, caregivers helping admin-

ister medications may not be as familiar with medications and

supplements and the additional information available through

the QR codes could assist in those situations.

Finally, younger age groups (college students) may be more

familiar and comfortable using technology, such as QR codes,

than older age groups (senior citizens older than 70 years); this

could have contributed to the lower correct answers in the older

age-group relative to the higher correct answers in the younger

age groups, specifically in the experimental group. Further

studies could examine the best modalities (visual [text and

images] and auditory) to deliver the information and perhaps

doing so redundantly through multiple modalities.

Based on the reported data, this kind of research can be

applied in both inpatient and outpatient settings, where it could

improve the administration of both prescribed medications and

over-the-counter drugs and supplements. Since it is evident that

information currently available on medication bottles is insuf-

ficient and requires further clarification, the application of QR

codes on prescription drug and supplement packaging could

reduce common patient errors.

One could argue that the information used in the QR code

condition of this experiment can be found in the supplemental

information supplied with prescribed medications. However,

previous research mentioned earlier shows this information is

often discarded, so having that information readily available on

the medication container at all times is advantageous. It is impor-

tant to recognize that prescribed medications are often only part

of a patient’s health regimen, and over-the-counter supplements

and medications that could interact with those prescription med-

ications have little if any packaging information like that con-

tained in this study’s QR condition. This study is suggesting that

both prescription and nonprescription labeling should be consid-

ered in any attempt to reduce medication errors.

It is important to note that since the results of this study

indicated that QR codes do not completely eliminate errors,

further improvements need to be explored. Rather than simply

590 Journal of Pharmacy Practice 34(4)

providing additional barcode information, patients could ben-

efit from novel technological improvements related to QR code

technology. For example, patients could use a smartphone or

app to scan the code in order to receive information in a gra-

phic, text, or audible format.

Current successful innovations, such as voice-assistant

devices, could be utilized to reduce medication errors. Through

a digital assistant, the patient could receive alerts and remin-

ders tailored to their specific medical background and

prescribed treatment. For example, it could remind self-

administered patients to take their medications and supple-

ments as prescribed. This technology could not only track

prescribed medications but any supplements a patient is taking

and sound the alert for any potentially dangerous interactions.

Furthermore, this could reduce reliance on health-care providers

by giving people easy access to important medication informa-

tion concerning potential side effects and other relevant warn-

ings from an alternate yet reliable and trusted source. This can

help reduce the amount of hospital visits, because people will be

able to get credible information about side effects and warnings

from a reliable and trusted source. QR Code implementation on

medication labels would be beneficial for all age groups regard-

less of diagnosis; however, it would especially benefit older

patients who are more likely prescribed multiple medications.

Conclusion

This study investigated whether the implementation of QR

code technology could increase the effectiveness of medication

labeling and the reduction in self-administered medication

errors. The results indicated that regardless of age, bottle labels

cannot provide enough information to avoid information errors

that could lead to medication errors. However, QR code tech-

nology could be implemented to deliver vital information via

smartphone apps in audible, graphic, or text format. Accord-

ingly, future research is warranted to explore the feasibility and

implementation of this technology with the goal of improving

health-care outcomes.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to

the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, author-

ship, and/or publication of this article.

ORCID iD

Scott E. McIntyre, PhD https://orcid.org/0000-0002-8751-2177

Supplemental Material

Supplemental material for this article is available online.

References

1. U.S. Food and Drug Administration [FDA]. Strategies to Reduce

Medication Errors: Working to Improve Medication Safety. 2017.

Accessed May 7, 2019. https://www.fda.gov/drugs/drug-informa

tion-consumers/strategies-reduce-medication-errors-working-

improve-medication-safety

2. U.S. Food and Drug Administration [FDA]. Medication Errors

Related to CDER-Regulated Drug Products. 2018. Accessed

May 7, 2019. https://www.fda.gov/drugs/drug-safety-and-avail

ability/medication-errors-related-cder-regulated-drug-products

3. Jeetu G, Girish T. Prescription drug labeling medication erorrs: a

big deal for pharmacists. J Young Pharm. 2010;2(1):107-111.

4. Klein M, Cohen H. Survey of feasibility of a peelable and point-

of-use labelling system. Eur J Hosp Pharm. 2014;21(1):22-28.

5. U.S. Food and Drug Administration [FDA]. Mixing Medications

and Dietary Supplements can Endanger Your Health. 2014.

Accessed February 7, 2019. https://www.fda.gov/consumers/con

sumer-updates/mixing-medications-and-dietary-supplements-

can-endanger-your-health

6. Fialová D, Onder G. Medication errors in elderly people: contri-

buting factors and future perspectives. Br J Clin Pharmacol.

2009;67(6):641-645. doi:10.1111/j.1365-2125.2009.03419.x

7. Lockwood W. QR codes: a digital avenue for PMI. Comput.

Pharm. 2013;33(3):12-14.

8. Raynor D, Silcock J, Knapp P, Edmondson H. How do patients

use medicine information leaflets in the UK? Int J Pharm Pract.

2007;15(3):209-218. doi:10.1211/ijpp.15.3.0008

9. Bubalo J, Warden BA, Wiegel JJ, et al. Does applying technology

throughout the medication use process improve patient safety

with antineoplastics? J Oncol Pharm Pract. 2014;20(6):

445-460. doi:10.1177/1078155213514469

10. Fowler SB, Sohler P, Zarillo DF. Bar-code technology for med-

ication administration: Medication errors and nurse satisfaction.

Medsurg Nurs. 2009;18(2):103-109.

11. Hensley S. Pfizer to place bar codes on pills—new labels will be

used in hospitals, with the aim of cutting medical errors. Wall

Street J. 2003:D.3. http://login.ezproxy1.lib.asu.edu/login?

url¼https://search-proquest-com.ezproxy1.lib.asu.edu/docview/3

98944058?accountid¼4485

12. Mira JJ, Guilabert M, Carrillo I, et al. (2015). Use of QR and

EAN-13 codes by older patients taking multiple medications for

a safer use of medication. Int J Med Inform. 2015;84(6):406-412.

doi:10.1016/j.ijmedinf.2015.02.001

13. Wild D, Szczepura A, Nelson S. New barcode checks help reduce

drug round errors in care homes. Nurs Manag (Harrow). 2011;

18(5):26-30.

14. 14. U.S. Food and Drug Administration [FDA]. Medication Error

Reports. 2017. Accessed August 6, 2019. https://www.fda.gov/

drugs/medication-errors-related-cder-regulated-drug-products/

medication-error-reports

Svensk and McIntyre 591

Copyright of Journal of Pharmacy Practice is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.