0% found this document useful (0 votes)
11 views9 pages

Gotlieb 2022 Oi 221211 1669128872.89706

The study assessed how well the general public understands common medical phrases that use words with different meanings in medicine than in everyday language. It found that many phrases were misunderstood, with the interpreted meaning often opposite of what was intended by medical professionals. Understanding of phrases was not significantly associated with demographics. Using plain language instead of jargon improved comprehension.

Uploaded by

Ale Melli
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
11 views9 pages

Gotlieb 2022 Oi 221211 1669128872.89706

The study assessed how well the general public understands common medical phrases that use words with different meanings in medicine than in everyday language. It found that many phrases were misunderstood, with the interpreted meaning often opposite of what was intended by medical professionals. Understanding of phrases was not significantly associated with demographics. Using plain language instead of jargon improved comprehension.

Uploaded by

Ale Melli
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 9

Original Investigation | Public Health

Accuracy in Patient Understanding of Common Medical Phrases


Rachael Gotlieb, MD; Corinne Praska, MD; Marissa A. Hendrickson, MD; Jordan Marmet, MD; Victoria Charpentier, MD; Emily Hause, MD;
Katherine A. Allen, MD; Scott Lunos, MS; Michael B. Pitt, MD

Abstract Key Points


Question How well do adults
IMPORTANCE Despite acknowledging that medical jargon should be avoided, health care
understand common phrases clinicians
practitioners frequently use it when communicating with patients.
use when communicating with patients?

OBJECTIVE To characterize the understanding of common medical jargon terms by surveying a Findings In this cross-sectional study
cross section of the general public and studying phrases that have established meanings in regular that surveyed 215 adults, participants
usage but different meanings in a medical context (eg, negative and positive test results). frequently misunderstood and often
assigned meaning opposite to what the
DESIGN, SETTING, AND PARTICIPANTS In this cross-sectional study, participants indicated their clinician intended.
understanding of phrases that may have different meanings in medicine than in colloquial English via
Meaning These findings suggest that
a mix of short answer and multiple choice questions. Several questions included paired phrases to
use of common medical phrases may
assess for differences in understanding with or without jargon. Volunteers were recruited at the 2021
lead to confusion among patients
Minnesota State Fair near St Paul, Minnesota. An electronic survey was given to a volunteer sample
affecting health outcomes.
of 215 adults (>18 years) who did not work or train to work in the medical field and spoke and
read English.
+ Supplemental content
EXPOSURES Completing a written or verbal survey. Author affiliations and article information are
listed at the end of this article.

MAIN OUTCOMES AND MEASURES The main outcome was an accurate understanding of the
medical terminology. Free-text responses were coded by 2 researchers for comprehension.
Secondary outcomes looked for associations between volunteer demographics and understanding.

RESULTS The 215 respondents (135 [63%] female; mean [SD] age, 42 [17] years) demonstrated a
varied ability to interpret medical jargon phrases. For example, most participants (207 [96%]) knew
that negative cancer screening results meant they did not have cancer, but fewer participants (143
[79%]) knew that the phrase “your tumor is progressing” was bad news, or that positive lymph nodes
meant the cancer had spread (170 [67%]). While most (171 [80%]) recognized that an unremarkable
chest radiography was good news, only 44 participants (21%) correctly understood that a clinician
saying their radiography was impressive was generally bad news. In each of the paired phrases
comparing jargon vs nonjargon approaches, the nonjargon phrase was understood significantly
better (P < .001).

CONCLUSIONS AND RELEVANCE These findings suggest that several common phrases are
misunderstood when used in a medical setting, with the interpreted meaning frequently the exact
opposite of what is intended.

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972

Open Access. This is an open access article distributed under the terms of the CC-BY License.

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 1/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022


JAMA Network Open | Public Health Accuracy in Patient Understanding of Common Medical Phrases

Introduction
Health care professionals regularly use jargon when communicating with patients, despite
acknowledging that it should be avoided.1-7 Though this medical language may facilitate
communication between health care professionals, its use with patients can introduce confusion that
may have serious consequences.8-11 The mismatch between our intent to avoid jargon and the reality
of our frequent use of it has been called jargon oblivion.12 One potential reason for this disconnect
is that, as health care professionals, we simply assume our patients understand the terminology we
are using. No matter how intentional we are about minimizing jargon, we will not avoid using words
and phrases that we fail to recognize as jargon in the first place. Accordingly, by better understanding
what medical terms and phrases patients do or do not comprehend, we can expand our jargon
identification toolkit and ultimately improve our communication with patients.
Previous studies have shown that while technical terminology, abbreviations, and acronyms are
the most commonly used forms of jargon, several other types of jargon are also used frequently.1,2
These include terms that have a well-understood meaning in common usage but often have a
different meaning in medicine. For example, in most contexts, negative typically indicates something
bad, such as negative feedback, negative viewpoints, or negative reviews. However, in the medical
context, negative typically has a different or even opposite meaning, whereas a negative test result
often signifies a favorable outcome.12 In this study, we aimed to characterize the understanding of
common medical jargon terms and phrases by surveying a cross section of the public at the
Minnesota State Fair. While others have demonstrated how the public rarely understands technical
terminology and acronyms used in medicine,6-8,11,13 we aimed to assess understanding of the types of
jargon that include words or phrases that have common meanings in regular usage and different
meanings in medicine because these phrases may be particularly confusing to patients. We also
wished to understand if certain demographic factors (ie, age, gender, or education) were associated
with differences in understanding and if the method of administering the survey (ie, written vs
verbal) changed how well respondents understood the phrases. Our hypothesis was that many of
these terms will be poorly understood by adults in the general public regardless of age, gender,
education, or survey method.

Methods
We report our findings of this cross-sectional study using the Strengthening the Reporting of
Observational Studies in Epidemiology (STROBE) reporting guideline. This study was approved by
the University of Minnesota institutional review board (IRB). Participants were given a printed study
information form that described the study objectives and risks and benefits of participating. Given
that obtaining a signature would have been the only identifying information in the study, the IRB
determined it was a lower risk to have verbal consent alone.

Setting and Survey Design


We surveyed a cross section of the public during 3 days at the 2021 Minnesota State Fair. We
developed a 13-question survey with a mix of open-ended and multiple-choice questions assessing
jargon understanding of common phrases used in medicine as well as brief demographics (age,
gender, education). To look for changes over time, we included some questions that had been
studied previously, such as13:
Imagine you are a patient who has just had surgery for cancer. After surgery your doctor gives
you your test results and tells you the following: “Your nodes are positive.”
What do you believe your doctor is telling you?
A) You are clear of cancer
B) The cancer has spread

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 2/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022


JAMA Network Open | Public Health Accuracy in Patient Understanding of Common Medical Phrases

C) The cancer has not spread


D) Don’t know
Additionally, we included several new questions assessing understanding of phrases, which
include words with meanings outside of medicine that differ from their meaning when used as a
medical context, such as “Your chest X-ray was unremarkable,” “The findings on the X-ray were quite
impressive,” or “Neuro exam is grossly intact.” Respondents were asked to indicate if they felt these
statements indicated good news, bad news, or they were unsure.
Free-text response questions asked participants to write what they felt a doctor would mean by
phrases, such as “occult infection” or “bugs in the urine,” because these also include words that have
different meanings outside of medicine. We also assessed understanding of the word “febrile”
because we hear this used often with patients but were uncertain of how broadly it is understood.
We also chose to assess understanding of alternative phrases for 2 terms that, in our experience,
can be confusing to patients: “negative” and “NPO”—the latter of which has been previously shown
to be poorly understood by patients.14 To compare the understanding of different ways of sharing
this information, we developed paired items and presented them at separate points in the survey. For
example, in a paired set of questions, participants were first asked to write what they felt it would
mean if a doctor said, “You will need to be NPO at 8 AM,” followed later by a question with the
phrasing “You are to have nothing by mouth at 4 PM.” The full survey text is available in the eAppendix
in the Supplement. After each question, respondents were asked how confident they were that they
were correct (ie, not at all confident, somewhat confident, quite confident, or certain). Statistical
analysis took place from September 2021 to December 2021.

Survey Delivery Method


We developed both a written and a verbal version of the survey. For the written survey, all content
was presented in written form on REDCap, accessed through an iPad.15 In the verbal versions, the
question prompts were provided in written format on an iPad, but the phrases being studied were
read aloud by a voice actor, accessed by clicking on an audio file icon and listening through
headphones. Participants who received the verbal version could replay a sound file as often as they
wished. Two phrases were written out in both versions of the survey because the question prompt
asked what a patient would think if they read a certain phrase in their medical notes (eg, “neuro exam
is grossly intact” and “I am concerned this patient has an occult infection”).

Data Collection
Attendees entering or passing by the University of Minnesota’s Driven to Discover research building16
were invited to participate in the survey in exchange for a backpack with the University logo.
Participants were eligible if they were at least 18 years old, had no history of medical or nursing
training, and were comfortable participating in an English-language survey. The survey was
anonymous and voluntary, and participants were able to stop at any point during the survey and still
receive the incentive. After consenting to participate, volunteers were randomized by a throw of a
die to either a written or verbal form of the survey.

Statistical Analysis
Multiple choice responses were coded as correct or incorrect. Free-text responses were coded for
accuracy by 2 independent researchers (R.G. and C.P), adding a third researcher if consensus was not
reached.17 We used descriptive statistics to summarize the survey results, calculating means and
standard deviations for continuous variables, and counts and percentages for categorical variables.
Demographics and survey questions were compared between written and verbal groups using a t
test for age, Fisher exact tests for categorical variables, and Wilcoxon rank sum test for confidence
questions. McNemar test was used to compare correct responses for the paired questions with
different options for sharing the same information. The association of a correct response with
demographics (ie, age, gender, education, and group) and confidence was examined with

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 3/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022


JAMA Network Open | Public Health Accuracy in Patient Understanding of Common Medical Phrases

multivariable logistic regression models. Adjusted odds ratio (aOR) and 95% CIs were reported from
these models. P values <.05 were considered statistically significant. SAS V9.4 (SAS Institute) was
used for the analysis.

Results
In this cross-sectional study, 215 volunteers completed the survey (116 written, 99 verbal).
Respondents had a mean (SD) age of 42 (17) years, 140 (65%) had a bachelor's degree or higher, and
135 (63%) were female. All respondents completed the entire survey. The demographics were
statistically similar across the survey type (Table 1). Because there was no significant difference in
correct understanding between those who were given the written and verbal surveys for all but 2
questions (Table 2), the responses are grouped together regardless of the method of survey
administration for the remainder of the analysis.
There was mixed understanding of which phrases were meant to convey good news vs bad
news (Table 2). For example, of the 215 respondents, most respondents (207 [96%]) knew that

Table 1. Summary of Demographics by Groupa

No. (%)
Characteristic Total (N = 217) Verbal (n = 101) Written (n = 116) P valueb
Age, mean (SD) [range] 42.4 (17.1) [18-88] 42.7 (16.5) [18-73] 42.3 (17.8) [18-88] .86
Gender
Female 135 (62.8) 58 (57.4) 77 (67.5) .16
Male 80 (37.2) 43 (42.6) 37 (32.5)
Education
Some HS 3 (1.4) 0 3 (2.6)
HS or GED 20 (9.2) 5 (5.0) 15 (12.9)
Associates 18 (8.3) 9 (8.9) 9 (7.8) Abbreviations: GED, general educational
.13 development; HS, high school.
Some college 35 (16.1) 16 (15.8) 19 (16.4)
a
No statistically significant differences in
Bachelor’s 70 (32.3) 39 (38.6) 31 (26.7)
demographics between the 2 groups.
Grad or professional 70 (32.3) 32 (31.7) 38 (32.8)
b
Two group t test for age and Fisher exact test for
Other 1 (0.5) 0 1 (0.9)
categorical variables.

Table 2. Accurate Understanding of Jargon Phrases by Cross Section of the Public

Answered correctly by survey type, No. (%)


Overall Verbal Written P value (verbal vs
Phrase (N = 215) (n = 99) (n = 116) written survey)a
Your blood tests showed me that you do not 208 (97.7) 99 (100) 109 (94) .02a
have an infection in your blood.
Your cancer screening test came back and the 207 (96.3) 96 (97.0) 111 (95.7) .73
results are negative.
Your blood culture was negative. 186 (86.5) 88 (88.9) 98 (84.5) .42
Your chest x-ray was unremarkable. 171 (79.5) 80 (80.8) 91 (78.4) .74
We are halfway through your chemotherapy 170 (79.1) 78 (78.8) 92 (79.3) >.99
treatment and your tumor is progressing.
You are to have nothing by mouth after 4 PM. 162 (75.3) 76 (76.8) 86 (74.1) .75
Your nodes are positive. 143 (66.5) 64 (64.6) 79 (68.1) .66
Patient’s neuro exam is grossly intact.b 89 (41.4) NA NA NA
Your urine tests are back and there were bugs 62 (28.8) 32 (32.3) 30 (25.9) .37
in your urine.
The findings on the x-ray were quite 44 (20.5) 21 (21.2) 23 (19.8) .87 Abbreviations: NA indicates not applicable; NPO,
impressive. nothing by mouth.
You will need to be NPO at 8 AM. 24 (11.2) 11 (11.1) 13 (11.2) >.99 a
P value <.05 considered statistically significant.
Have you been febrile? 20 (9.3) 9 (9.1) 11 (9.5) >.99 b
This question was written out in both surveys
I am concerned the patient has 4 (1.9) 4 (4.0) 0 (0) .04a because the prompt was about reading this
an occult infection.
statement in a medical record.

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 4/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022


JAMA Network Open | Public Health Accuracy in Patient Understanding of Common Medical Phrases

negative cancer screening results meant they did not have cancer. However, fewer respondents
knew that “your tumor is progressing” was bad news (170 [79%]) or that positive nodes meant their
cancer had spread (143 [67%]). Only 89 (41%) of respondents correctly interpreted “neuro exam is
grossly intact” as good news. Additionally, while most respondents (171 [80%]) recognized that an
unremarkable chest radiography was good news, only 44 respondents (21%) correctly understood
that a clinician saying their radiography was impressive was generally bad news. Few respondents
accurately understood the prompts that required a free-text response. Sixty-two respondents (29%)
correctly interpreted “bugs in the urine” as intending to convey a urinary tract infection, 20 (9%)
knew what febrile meant, and 4 (2%) of respondents understood the phrase occult infection.
Full findings from the paired items are presented in Table 3. Significantly more respondents
correctly interpreted the phrase nothing by mouth compared with the use of the acronym NPO (162
respondents [75%] vs 24 respondents [11%], respectively; P < .001). When comparing the
understanding of the same concept (blood infection) through nonjargon (“blood test shows no
infection”) vs jargon (“your blood culture was negative”), significantly more respondents correctly
interpreted the nonjargon phrase than the jargon phrase (208 [98%] vs 186[87%],
respectively; P < .001).
In multivariable logistic regression models, there were a few statistically significant associations
between demographics and understanding (Table 4). Notably, increasing age was associated with
increased understanding of nothing by mouth and negative blood cultures but decreased
understanding of the term impressive in the context of radiography findings. Two questions showed
an association with increased understanding if the respondent had a graduate degree: the phrases
nothing by mouth and unremarkable, and the acronym NPO and the term febrile were better
understood by women.

Table 3. Comparative Understanding of Paired Phrases by Cross Section of the Public

Phrase Answered correctly (n = 215), No. (%) P valuea


NPO vs nothing by mouth
You will need to be NPO at 8 AM. 24 (11.2) <.001
You are to have nothing by mouth after 4 PM. 162 (75.3)
Negative vs do not have
Your blood culture was negative. 186 (86.5)
<.001
Your blood test showed me that you do not have an infection 208 (96.7) Abbreviation: NPO indicates nothing by mouth.
in your blood. a
P value <.05 considered statistically significant.

Table 4. Statistically Significant Demographic Associations With Correct Understanding of Jargon Phrases
on Multivariable Logistic Regressiona
Adjusted odds ratio
Phrase Demographic association with correct understanding (95% CI) P valueb
Your blood culture Older age (each year) associated with 1.03 (1.00-1.06) .03
was negative increased understanding
The findings on the x-ray Younger age (each year) associated with 0.96 (0.94-0.99) .002
were quite impressive increased understanding
You are to have nothing by Older age (each year) associated with 1.03 (1.01-1.06) .002
mouth after 4 PM increased understanding
Graduate degree associated with increased 3.33 (1.39-7.99) .007
understanding compared with associate’s
degree or lower
Bachelor’s degree associated with increased 2.23 (1.00-4.95) .049
understanding compared with associate’s Abbreviation: NPO indicates nothing by mouth.
degree or lower a
Associations between the 3 surveyed demographics
Your chest x-ray was Graduate degree associated with increased 3.45 (1.35-8.87) .01 (ie, age, gender, and education) and each of the 13
unremarkable understanding compared with associate’s
degree or lower questions were assessed with multivariable logistic
You will need to be NPO Female gender associated with 5.65 (1.59-20.13) .008 regression models yielding 39 comparisons. The 7
at 8 AM increased understanding associations which reached statistical significance
Have you been febrile? Female gender associated with 5.90 (1.31-26.71) .02 are depicted here.
increased understanding b
P value <.05 considered statistically significant.

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 5/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022


JAMA Network Open | Public Health Accuracy in Patient Understanding of Common Medical Phrases

In general, the more confident the respondents were in their answers, the more likely they were
to be correct. Additionally, respondents were more confident about both of the approaches to
sharing blood culture results if they heard the clinician say them vs if they read the phrase. The
confidence in the accuracy of verbal vs written questions for “blood culture was negative” was mean
(SD) of 3.0 (0.9; median, 3.0) vs 2.7 (0.8; median, 3.0; P = .04). The confidence in accuracy of verbal
vs written questions for “blood test showed me you do not have an infection” was mean (SD) 3.6 (0.7;
median, 4.0) vs 3.3 (0.8; median, 3.0; P = .003).

Discussion
In this cross-sectional study, we found that terms and phrases commonly used in clinical settings
remain frequently misunderstood. Many studies of jargon comprehension take place in a medical
setting, either as observational studies1-3,6,9 or surveys of patients in clinics.8,18,19 We aimed to better
capture a less clinically biased sample by surveying a cross section of the public at the Minnesota
State Fair to determine their understanding of commonly used medical jargon. To our knowledge,
this is the largest study of patients’ understanding of jargon and the first to compare the
understanding of jargon vs nonjargon phrases.
Our testing of several phrases that had been studied previously yielded several notable
differences, in most cases, with a higher proportion of our study sample demonstrating
understanding. For example, in 2001, Chapman et al13 found that among a sample of 105 adults in the
UK, only 52% understood that the phrase “the tumor was progressing” signified bad news. They
noted that progress is interpreted as a good thing in most settings. In our study, 79% of respondents
correctly understood this phrase as bad news, with an absolute increase in understanding of 27%.
Similarly, 43% of the respondents in the Chapman et al study13 correctly understood that having
positive nodes meant their cancer had spread vs 67% correct among our sample, an absolute
increase of 24%.
Some of these differences may be accounted to cultural differences between the UK-based
sample in the Chapman study13 and our sample in the US. Additionally, the high number of college
graduates in our sample (65% with a bachelor’s degrees or higher) compared with the general
population of the United States (35%)20 may be a factor in that difference, although Chapman et al13
also reported that “a large proportion of the sample was well educated.” However, it should be noted
that we found no statistically significant association with the level of education in the accuracy of
interpreting the jargon in our study sample for all but 2 survey questions, so this is unlikely to account
for the differences fully.
Some changes in understanding may be the result of the COVID-19 pandemic. For example, we
hypothesize that the widely used designations of negative and positive in the context of viral testing
during the pandemic have increased the public’s understanding of these terms in the medical
context, accounting for the near-universal understanding of negative cancer screening being
considered good news in our study. However, it is worth noting that when comparing the
understanding of the phrase “your blood test shows no infection” and “your blood culture was
negative,” significantly more respondents correctly interpreted the phrase that avoided the word
negative altogether.
The use of terms that mean something different in common usage than in a medical context—or
medicalized English1,2,12—was a frequent cause of confusion in our study. More people believed that
the phrase “had an occult infection” had something to do with a curse than understood that this
meant that they had a hidden infection. Fewer than half knew that their neuro examination being
“grossly intact” was a good thing, possibly because the word “gross” more often means “unpleasant”
than “in general” in common usage. These terms may not necessarily be recognized by clinicians as
jargon because they do not land in the commonly understood category of technical, medical
terminology. However, they have been shown to be used frequently in clinical settings.1,2

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 6/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022


JAMA Network Open | Public Health Accuracy in Patient Understanding of Common Medical Phrases

Most published results of jargon understanding by adults involve having the respondents read
a prompt a doctor might say and indicate their understanding.7,8,13,18,19,21 Given that, in most cases,
jargon is spoken aloud by clinicians during patient encounters, we opted to assess understanding of
both written and spoken forms of communication to determine if survey methods affected
understanding. To our knowledge, our study was the first to directly compare understanding of
medical jargon in written vs audio form. Overall, we did not observe any significant difference in
understanding between these 2 delivery methods, which may support the less time-intensive, more
frequently used written survey approach to assessing jargon understanding. We hope that future
studies on medical jargon will further explore this observation at a larger scale, which may help
provide important insights for optimizing the study of patient communication, particularly as
patients increasingly have real-time access to their medical documentation.22,23
Given that increasing age comes with more opportunities to have heard these terms used in a
medical context, it is somewhat surprising that older age was only associated with better
understanding of 2 of the 13 phrases. In fact, the lack of consistent predictors of understanding by the
demographics we studied (ie, age, gender, and education) highlights the importance of using clear
communication with all patients.
We hypothesized that respondents would interpret plain-language descriptions of medical
events more correctly than jargon-based descriptions. Indeed, in both cases for which we assessed
comparisons the nonjargon phrase was significantly more widely understood. However, it is worth
noting that while nothing by mouth was better understood than NPO by nearly 7-fold, 1 in 4
respondents also did not understand the phrase nothing by mouth. Given that in everyday language
we do not talk about the act of eating or drinking as taking something by mouth, perhaps the clearest
way to indicate that a patient should abstain from oral intake is to simply say, “You should not have
anything to eat or drink.”

Limitations
Our study has several limitations. Though participation was open to all adult fairgoers who
volunteered, there is likely a naturally occurring bias in selecting individuals who would visit a
university research building during their visit to a fair. Additionally, this research building had a mask
mandate to mitigate the spread of SARS-CoV2 at the time of study, whereas many other areas of the
fair did not. This requirement may have further selected a nonrepresentative sample. Although
education demographics are not available for visitors to the 2021 Minnesota State Fair, in the state as
a whole, 50% of those older than 25 years are reported to hold an associate degree or higher, which
is higher than in many other states.24 In our sample, 77% reported that level of attainment,
demonstrating a bias toward participants with more education. However, it is worth noting that only
2 of the survey questions showed increased understanding with increasing education. Furthermore,
if education is associated with a better understanding of medical jargon, our results likely represent
an overrepresentation of the actual comprehension at a societal level, which indicates these phrases
may be even less understood in the population as a whole. Additionally, adding a control group of
clinicians taking the survey would have been helpful to validate the agreement behind what clinicians
intend when they use these phrases, though the only phrase we can hypothesize where this may be
used differently depending on the context is the term “impressive” when describing radiography,
which conceivably a clinician might use to describe how quickly something healed. Finally, though
our survey questions were thoroughly assessed for bias, in some cases, the answer choices were
multiple choice (eg, good news, bad news, or don’t know), allowing a survey respondent to guess and
provide answers that may not reflect their true understanding.

Conclusions
Medical jargon remains a common source of confusion for patients, and care should be taken to avoid
using it with patients to prevent misunderstanding. Many commonly used jargon phrases are

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 7/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022


JAMA Network Open | Public Health Accuracy in Patient Understanding of Common Medical Phrases

associated with poor understanding by the general public, and more people understood jargon-free
versions of common medical phrases than expressions using jargon. No significant differences were
found between an audio and a written version of the survey indicating that future studies of jargon
understanding may support the less time-intensive written survey approach. Future studies should
continue characterizing the understanding of jargon among the public and testing recommended
alternatives to improve our communication with patients.

ARTICLE INFORMATION
Accepted for Publication: September 30, 2022.
Published: November 30, 2022. doi:10.1001/jamanetworkopen.2022.42972
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Gotlieb R
et al. JAMA Network Open.
Corresponding Author: Michael Pitt, MD, Department of Pediatrics, University of Minnesota, 2450 Riverside Ave,
AOB 1, Minneapolis, MN 55454 ([email protected]).
Author Affiliations: University of Minnesota Medical School, Minneapolis (Gotlieb, Praska, Charpentier);
Department of Pediatrics, University of Minnesota, Minneapolis (Hendrickson, Marmet, Hause, Allen, Pitt); M
Health Fairview Masonic Children’s Hospital, Minneapolis, Minnesota (Hendrickson, Marmet, Pitt); Biostatistical
Design and Analysis Center, Clinical and Translational Science Institute, University of Minnesota,
Minneapolis (Lunos).
Author Contributions: Dr Pitt and Mr Lunos had full access to all of the data in the study and take responsibility for
the integrity of the data and the accuracy of the data analysis. Drs Gotlieb and Praska are co-first authors.
Concept and design: Gotlieb, Praska, Hendrickson, Marmet, Charpentier, Hause, Pitt.
Acquisition, analysis, or interpretation of data: Gotlieb, Praska, Hendrickson, Marmet, Hause, Allen, Lunos, Pitt.
Drafting of the manuscript: Gotlieb, Praska, Charpentier, Pitt.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Gotlieb, Allen, Lunos, Pitt.
Obtained funding: Marmet, Pitt.
Administrative, technical, or material support: Marmet, Charpentier, Hause, Allen, Pitt.
Supervision: Marmet, Pitt.
Conflict of Interest Disclosures: None reported.
Funding/Support: This research was supported by grant UL1TR002494 from the National Institutes of Health’s
National Center for Advancing Translational Sciences and an internal grant from the Driven to Discover Research
Facility at the University of Minnesota.
Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection,
management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and
decision to submit the manuscript for publication.
Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official
views of the National Institutes of Health’s National Center for Advancing Translational Sciences.

REFERENCES
1. Charpentier V, Gotlieb R, Praska CE, Hendrickson M, Pitt MB, Marmet J. Say what: quantifying and classifying
jargon use during inpatient rounds. Hosp Pediatr. 2021;11(4):406-410. doi:10.1542/hpeds.2020-002790
2. Miller AN, Bharathan A, Duvuuri VNS, et al. Use of seven types of medical jargon by male and female primary
care providers at a university health center. Patient Educ Couns. 2022;105(5):1261-1267. doi:10.1016/j.pec.2021.
08.018
3. Lehmann LS, Brancati FL, Chen MC, Roter D, Dobs AS. The effect of bedside case presentations on patients’
perceptions of their medical care. N Engl J Med. 1997;336(16):1150-1155. doi:10.1056/NEJM199704173361606
4. Howard T, Jacobson KL, Kripalani S. Doctor talk: physicians’ use of clear verbal communication. J Health
Commun. 2013;18(8):991-1001. doi:10.1080/10810730.2012.757398
5. Sevinc A, Buyukberber S, Camci C. Medical jargon: obstacle to effective communication between physicians and
patients. Med Princ Pract. 2005;14(4):292. doi:10.1159/000085754

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 8/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022


JAMA Network Open | Public Health Accuracy in Patient Understanding of Common Medical Phrases

6. Farrell M, Deuster L, Donovan J, Christopher S. Pediatric residents’ use of jargon during counseling about
newborn genetic screening results. Pediatrics. 2008;122(2):243-249. doi:10.1542/peds.2007-2160
7. LeBlanc TW, Hesson A, Williams A, et al. Patient understanding of medical jargon: a survey study of US medical
students. Patient Educ Couns. 2014;95(2):238-242. doi:10.1016/j.pec.2014.01.014
8. Pieterse AH, Jager NA, Smets EMA, Henselmans I. Lay understanding of common medical terminology in
oncology. Psychooncology. 2013;22(5):1186-1191. doi:10.1002/pon.3096
9. Links AR, Callon W, Wasserman C, Walsh J, Beach MC, Boss EF. Surgeon use of medical jargon with parents in
the outpatient setting. Patient Educ Couns. 2019;102(6):1111-1118. doi:10.1016/j.pec.2019.02.002
10. Castro CM, Wilson C, Wang F, Schillinger D. Babel babble: physicians’ use of unclarified medical jargon with
patients. Am J Health Behav. 2007;31(suppl 1):S85-S95. doi:10.5993/AJHB.31.s1.11
11. Rau NM, Basir MA, Flynn KE. Parental understanding of crucial medical jargon used in prenatal prematurity
counseling. BMC Med Inform Decis Mak. 2020;20(1):169. doi:10.1186/s12911-020-01188-w
12. Pitt MB, Hendrickson MA. Eradicating jargon-oblivion-a proposed classification system of medical jargon. J Gen
Intern Med. 2020;35(6):1861-1864. doi:10.1007/s11606-019-05526-1
13. Chapman K, Abraham C, Jenkins V, Fallowfield L. Lay understanding of terms used in cancer consultations.
Psychooncology. 2003;12(6):557-566. doi:10.1002/pon.673
14. Fields AM, Freiberg CS, Fickenscher A, Shelley KH. Patients and jargon: are we speaking the same language?
J Clin Anesth. 2008;20(5):343-346. doi:10.1016/j.jclinane.2008.02.006
15. Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap)—a
metadata-driven methodology and workflow process for providing translational research informatics support.
J Biomed Inform. 2009;42(2):377-381. doi:10.1016/j.jbi.2008.08.010
16. D2D: the driven to discover research facility. University of Minnesota. Accessed May 3, 2022. http://d2d.
umn.edu/
17. Hsieh HF, Shannon SE. Three approaches to qualitative content analysis. Qual Health Res. 2005;15(9):
1277-1288. doi:10.1177/1049732305276687
18. Peckham TJ. ‘Doctor, have I got a fracture or a break’? Injury. 1994;25(4):221-222. doi:10.1016/0020-1383(94)
90065-5
19. O’Connell RL, Hartridge-Lambert SK, Din N, St John ER, Hitchins C, Johnson T. Patients’ understanding of
medical terminology used in the breast clinic. Breast. 2013;22(5):836-838. doi:10.1016/j.breast.2013.02.019
20. Educational attainment distribution in the United States from 1960 to 2020. Statista. Accessed March 5,
2022. https://www.statista.com/statistics/184260/educational-attainment-in-the-us/
21. Barker KL, Reid M, Minns Lowe CJ. What does the language we use about arthritis mean to people who have
osteoarthritis:a qualitative study. Disabil Rehabil. 2014;36(5):367-372. doi:10.3109/09638288.2013.793409
22. Fernández L, Fossa A, Dong Z, et al. Words matter: what do patients find judgmental or offensive in outpatient
notes? J Gen Intern Med. 2021;36(9):2571-2578. doi:10.1007/s11606-020-06432-7
23. Wolff JL, Darer JD, Berger A, et al. Inviting patients and care partners to read doctors’ notes: OpenNotes and
shared access to electronic medical records. J Am Med Inform Assoc. 2017;24(e1):e166-e172. doi:10.1093/jamia/
ocw108
24. Educational Attainment Data. Minnesota Office of Higher Education. Accessed March 5, 2022. http://www.ohe.
state.mn.us/sPages/educ_attain.cfm

SUPPLEMENT.
eAppendix. Survey Used in Study
eTable. Associations Between Correct Responses of Jargon Terminology and Demographics in Cross Section of
General Public

JAMA Network Open. 2022;5(11):e2242972. doi:10.1001/jamanetworkopen.2022.42972 (Reprinted) November 30, 2022 9/9

Downloaded From: https://jamanetwork.com/ on 12/02/2022

You might also like