Observations
and Interviews of Triage Nurses
Ethnographic Study of Triage Nurses
Abstract
Triage is often defined as the process of prioritizing. A word that originated
in the battlefield, triage was originally done so as to allocate the limited
resources of the field to only patients who stood the highest chance of
recovery. Today, while triage has developed to take different forms, it’s
main goal of resource allocation and optimization remains the same.
Introduction
This report documents the findings from an ethnographic study of a group
of triage nurses at a Family Practice Clinic. Each nurse is equipped and
trained with a set of telephone and treatment protocols they use with
each in-coming patient call. The focus of the study is to examine whether
the usage of telephone protocols alone is sufficient for effective triage.
It also seeks to articulate the differences between phone triage vs. face-to-face.
The findings from this study will inform the design requirements of online
medical triage software that automates the nurses’ protocols, presenting
patients with presumptive diagnoses based on their answers to the questions.
Methods
Participants
The group of triage nurses observed was the Advice Nurse Team at the Palo
Medical Foundation Family Practice Clinic. The phone calls of two particular
advice nurses were listened in on Vera* nurse was a new hire for one month
and Irene* had been there for 3 years. Both were registered nurses with
triage in their work experience. Vera had previously worked as a face-face
triage nurse in a clinic and Irene had worked primarily as a phone triage
nurse for most of her nursing experience. While observing these two nurses,
notes were taken of comments made by other nurses.
Procedures
This study was conducted in two sessions -- each 2 hours in duration.
In the first session, only observations were done and no contextual inquiry
or interviewing was conducted. A list of questions were compiled from
the observations and integrated with a list of questions compiled prior
to the observation. In the second session, a combination of contextual
inquiry and observations were done. Conversations between that one nurse
and all in-coming calls were listened in on via headsets. Interviews were
conducted in between the calls.
Questions that arose during the phone calls were documented and asked
after the call. Brief interviews were also carried out with nearby nurses.
Hypothesis to investigate
Broad issues of investigation include, what is the actual process of triage,
how much of it pertains to the telephone protocols provided for each triage
nurse, what other skills and processes are involved in a successful triage?
How does the nurse’s questions influence or direct the patient’s
articulation? (For actual questions asked see Appendix)
Findings (report only presents highlights of the
study, for full notes please see Appendix)
Interviews and observations of the triage nurses at work revealed other
facets of the triage process apart from simply asking the questions in
the telephone protocols. In the sessions observed, none of the nurses
physically referred to their protocol tools at all. When asked, Irene
said that she knows the protocols by heart after using them so many times.
Observations of the two nurses and conversations of the other nurses revealed
a pattern of questions that inquire about the duration of symptoms, the
presence of fever, diarrhea and vomiting before triaging more specifically.
The four questions seem to be the key questions asked by most nurses in
the group when trying to ascertain an emergent situation. When asked to
define triage, Irene said it is the process of deciding which facility
of healthcare to channel a patient to. " When I mean, "decide
whether he should come in" I mean into the emergency room, the urgent
care, the family practice clinic or the specialty clinic." She explains.
" I try to deduce whether they should come in right away, wait a
week or practice home care". " So it’s not only where to
come in but when".
Many of the questions asked were elaboration or clarification of the symptoms
e.g. " How long ago was your last bowel movement?" " How
loose is your stools (asked to ascertain diarrhea), " Are your children
shying away from you (asked to ascertain foul breath), " What color
was your spittoon?" Physical symptoms were clarified with questions
such as " On a scale of one to ten with ten being the worst pain
you’ve ever felt, how would you rate your pain right now?" "Is
the pain in one legs or two legs?" "Is it between the knee and
ankle?"
When asked about these observations, Irene* said the hardest thing about
her job is trying to take vitals over the phone. " It is like trying
to use your nose, your eyes, your ears and your hands through the phone".
She talked about different methods she used to get a patient to listen
to their own symptoms and vitals. E.g. In order to answer " Do you
have foul breath? ", she suggests having a patient’s spouse
or friend answer that question instead of themselves since patients rarely
can smell their own breath. She also tries to listen to their breathing
rhythms over the phone.
Irene comments about the limitations of the phone by stressing the importance
of visual cues. She mentions two little girls brought in by their moms
for the similar symptoms of abdominal pain. She said that she took one
look at one little girl and she knew that she had to be taken into the
ER right away. "The child had acute appendicitis, I mean you could
see it on her face! " The other child, the nurses ruled out all kinds
of diseases with different protocols and finally she took the little girl
into a separate room and asked her " Did you eat Cheetos yesterday?
" "She had eaten a whole bag of Cheetos! "Irene exclaimed.
" I know it’s terrible to make assumptions and all, but you
can tell a lot by just looking at the child and the mother." "
The little girl who had appendicitis, her mother didn’t look like
the kind who would feed her Cheetos, I mean Odwalla maybe but not Cheetos!"
Irene said laughing.
Patients also tend to minimize their pain. Irene tells of one lady who
called up about simple viral, flu-like symptoms, she had called to inquire
about what over-the-counter medicine she could be using. Irene asked her
about flank pain, fever and chills and the patient answered negative for
them all. When asked about back pain, the patient suddenly paused and
said, "Yeah, I do have back pain" "It turns out she had
a full blown kidney infection! " she said incredulously.
Discussion
Analysis of the data confirmed my initial hypothesis that a large part
of the triage process is spent helping the patient articulate his symptoms
accurately. Via the telephone, the nurse helps the patient isolate and
identify physical symptoms; it’s severity and the duration of the
symptoms. Questions like "When was the last time you vomited? "
" What did you vomit? " and " Do your kids shy away from
you? (to deduce bad breath ) are examples of the nurse helping the patient
remember and isolate his symptoms? The study also reveals how the patient’s
threshold of pain greatly influences the "story" told to the
nurses. Findings also tell how the absence of visual cues removes valuable
clues in the process of triage and diagnosis. In short the observations
and interviews demonstrate the limitations of the protocols, and the need
for the nurse to explain, interpret and customize each protocol for individual
patients.
Critique
Validity
While observations of nurses in this triage group customized the telephone
protocols to each patient, and did not formally use them during the time
of observation, other triage groups might stress using the protocols more
formally. Experience of the nurses influence the instances in which the
triage nurses follows their instinct and experience instead of the strict
protocols from the books. The fact that these nurses book only same day
appointments and not next day, might incent the nurses to offer more customized
and personal care; since appointments get booked up earlier in the day,
and home care is the only alternative to the patient besides coming into
the emergency room and urgent care facility. However, data collected in
the two sessions confirmed my hypothesis that symptom articulation was
an integral part of the triage process.
Reliability
Conclusion and deductions made were from interviews with only two nurses.
Each interview was only 30 –45 minutes long. Triangulating those
perspectives with other nurses on the team might provide a more complete
picture of triage at this center. Both nurses operated from different
work experiences, Irene’s extended experience with phone triage allowed
her to develop skills and ways to work around the visual limitations of
using phone as a means of triage. Vera’s face-to-face triage experience
influenced her using the protocols in a different way.
Observations were also done during the hours of 9-12pm, the busiest times
in the clinic where most of the appointments are booked. Observations
done during the latter part of the day might reveal different data. Also,
one observation session occurred on a weekday and the other on Saturday
where the clinic hours are from 8.30am to 12pm. Patients who call that
cannot get appointments can only see their doctor after the weekend. That
might warrant more detailed interviewing and more meticulous home care
advice.
Generalizability
Information gathered was only from one triage setting. Studies of triage
nurses in other departments such as urgent care, internal medicine or
pediatrics might reveal different patterns.
