International Emergency Nursing 23 (2015) 294–298

Contents lists available at ScienceDirect

International Emergency Nursing
j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / a a e n

Descriptions and presentations of sepsis – A qualitative content
analysis of emergency calls
Katarina Bohm RN, PhD (Lecturer) a,*, Lisa Kurland MD, PhD (Associate Professor, Senior
Lecturer) a,b, Sofia Bartholdson (Medical Student) a, Maaret Castrèn MD (Professor) a,b
a
b

Karolinska Institutet, Institution of Clinical Science and Education, Södersjukhuset, Stockholm, Sweden
Section of Emergency Medicine, Södersjukhuset, Stockholm, Sweden

A R T I C L E

I N F O

Article history:
Received 2 December 2014
Received in revised form 13 April 2015
Accepted 15 April 2015
Keywords:
Sepsis
Communication
Emergency medical services
Dispatching

A B S T R A C T

Background: Sepsis is a serious condition which requires early treatment. We often fail to recognize sepsis
patients in the chain of prehospital care. Knowledge of how sepsis is expressed in calls to the emergency medical communication centre (EMCC) is limited. An increased understanding could lead to earlier
identification of patients with sepsis.
Objective: The aim of this study was to describe the descriptions of sepsis used during communication
between the caller and the emergency medical dispatcher (EMD).
Methods: To achieve the aim of the study, an inductive approach of qualitative content analysis was used.
In total, 29 consecutive patients, who arrived at the emergency department by ambulance and received
a diagnosis of sepsis according to the International Classification of Diseases (ICD)-10, were included in
the study. For each case, the corresponding emergency call recording from the EMCC was transcribed
verbatim. Main categories and subcategories from the text were abstracted.
Results: From fifteen subcategories, three main categories were abstracted: “Deterioration”, “Physical signs
and symptoms” and “Difficulties establishing satisfactory cont; act with the patient.” The way laymen and
professionals expressed themselves seemed to differ.
Conclusions: Sepsis was described in terms of the physical symptoms, changes of condition and communication abilities of the patient. This knowledge could lead to the identification of keywords which
could be incorporated in the decision tool used by the EMD to increase sepsis identification, but further
research is required.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction
Sepsis is a serious condition with a mortality rate of approximately 20% (Wenzel, 2002). For septic shock the mortality rate can
be as high as 45% (Wenzel, 2002). The incidence of sepsis in the
United States of America (USA) is approximately 240/100 000 citizens (Martin et al., 2003). The incidence is increasing and possible
explanations for the increase are the population growth, the ageing
of the population, more patients being treated with invasive procedures, a greater use of immunosuppressant drugs and an increase
in HIV and antibiotic-resistant infections (Rangel-Frausto, 1999).
Time is critical for the outcome of septic patients. A delay in antibiotic administration has been shown to increase mortality (Kumar
et al., 2006). Half of all emergency department (ED) patients with
severe sepsis, and one-third of patients treated for an infection, are

* Corresponding author. Karolinska Institutet, Institution of Clinical Science and
Education, Södersjukhuset, Stockholm, Sweden. Tel.: +46736403013; fax:
+4686162933.
E-mail address: katarina.bohm@ki.se (K. Bohm).
http://dx.doi.org/10.1016/j.ienj.2015.04.003
1755-599X/© 2015 Elsevier Ltd. All rights reserved.

transported by the emergency medical services (EMS) (Wang et al.,
2007, 2010). Patients with severe sepsis arriving with the EMS receive
antibiotics more quickly in the ED. A documented sepsis impression by the EMS provider decreases the time interval even more
(Studnek et al., 2012). Additionally, protocolized sepsis identification in the ambulance appears to be feasible (Wallgren et al., 2014).
Thus, identification of sepsis in the prehospital setting is both possible and important for patient care, but due to the often nonspecific presentations of sepsis, remains challenging.
The emergency medical dispatcher (EMD) plays an important role
in the continuum of emergency care. Early identification of serious
conditions by the EMD has been shown to shorten the ambulance
arrival time on scene and reduce mortality in cardiac arrest
(Berdowski et al., 2009), while no such studies exist for the septic
patient. There is no specific mention of sepsis in the protocol used
by Swedish EMDs today (SOS Alarm AB, 2001) and to our knowledge, no previous publication concerning how sepsis might be
described during the emergency call or how the EMD can identify
the septic patient (Herlitz et al., 2012). The aim of this study is to
describe the presentations of sepsis during the communication
between the caller and the EMD.

K. Bohm et al./International Emergency Nursing 23 (2015) 294–298

2. Methods

295

Consecutively included by
ICD-codes for sepsis

2.1. Design and analysis

N=65

A qualitative descriptive approach was chosen because of the lack
of previous knowledge regarding sepsis presentation when contacting the EMCC (Fevang et al., 2011; Herlitz et al., 2012). To start
with an open mind helped widen the picture and made it possible
to find descriptions that might otherwise have been overlooked
(Graneheim and Lundman, 2004).
Authentic emergency call recordings were collected and transcribed verbatim for every included patient. Data were analyzed
using content analysis (Graneheim and Lundman, 2004). Multiple
authors were involved in the data analysis to decrease the chance
of individual biases influencing the research findings. Authors reviewed each transcript, used open coding to code the key messages
in each passage, organized the codes into common groupings, and
began to identify the categories. The thematic development is exemplified in Table 1.
The study was conducted with the approval of the ethics committee in Stockholm, dnr: 2011/2013-31/5.

Excluded
Did not arrive at the ED by ambulance

N=24

Excluded
Did not have sepsis in the ED

N=7

Excluded
No recording found
Not adequate Swedish
Called ambulance for other
reasons
Presented with undistinguishable
co-disease

2.2. Data collection and setting
All patients were treated at the ED of Södersjukhuset, Stockholm, Sweden which is one of the busiest EDs in northern Europe,
with more than 115 000 patients per year.

N=5

Included in analysis

N=29

2.2.1. Inclusion and exclusion criteria
All patients with an ICD-10 code compatible with sepsis at
discharge from hospital were included consecutively between November 30th 2011 and February 12th 2012. Since the aim of the
study was to evaluate sepsis presentations to the emergency medical
call centre, it was important to include patients who had signs of
sepsis, during their ED visit (as described below), which was a
proxy for having signs and symptoms during their call to the
medical call centre. Therefore, a patient was assumed to have
ongoing sepsis in the ED if the term “sepsis” was used in the ED
medical record or if broad spectrum antimicrobial agents used to
treat sepsis had been administered. The patient was also assumed
to have sepsis in the ED if the patient had vital signs fulfilling the
sepsis criteria (Bone et al., 2009). A patient was excluded if he or
she arrived at the hospital by other means than by ambulance, or
if the patient did not have a presentation compatible with sepsis
during the ED visit, as defined above. If the recorded call was not
found or the caller did not speak Swedish (i.e. the qualitative
analysis), it was also excluded. Finally, 29 patients were included
(Fig. 1).

Fig. 1. Flowchart of included and excluded patients in the study.

2.3. Patient population
2.3.1. The patients
The age span of the included patients was between 50 and 95
years. Median and mean age were both 80 years. The patient was
a female in 55% (n = 16) of the calls.
2.3.2. The callers
Most commonly, the caller was either a nurse (n = 15) from a
nursing home, geriatric care centre, or sheltered housing, or a relative (n = 10). Other people calling were the police (n = 1), geriatrician
(n = 1), home care service (n = 1) and the Swedish medical care
hotline (n = 1). In none of the cases, did the patient call by him or
herself.

Table 1
Examples of condensation and abstraction of the analysis unit in this study, using qualitative content analysis.
Meaning unit

Condensed meaning unit

Code

Sub-category

Main category

This is apparently, according to the staffs who knows her, a quite
lively old retiree who is normally ambulant and the one
who. . . She is the most vivid there in the department. . . Yes, like
that and. . . up and walking around and helping and yes, you
know. And now she has just been lying down.
EMD: I understand. So you notice a significant difference, there?
Caller: Yes, absolutely. Since yesterday, there is definitely a change
and I feel that. . . he probably needs a bit more advanced care.

She is the most vivid in the department,
but now she has just been lying down.

Just lying down

Weakness

Deterioration

Since yesterday, there is definitely a
change.

New change for
the worse

Sudden
deterioration

EMD = Emergency medical dispatcher.

296

K. Bohm et al./International Emergency Nursing 23 (2015) 294–298

3. Results
The symptoms and descriptions for sepsis generated three main
categories. These were Deterioration, Physical signs and symptoms, and Difficulties in establishing contact with the patient.
3.1. Deterioration
A common description by the caller was the description of
changed condition of the patient, which often seemed to reflect a
concern of the patient’s condition or asthenia. The sub-categories
abstracted were Concern for serious condition, Weakness, Sudden deterioration and Falling and collapsing. Descriptions did not differ
substantially between laymen and health professionals (HP). In particular the descriptions in the subcategories Falling and Collapsing
and Sudden deterioration were often connected with anxiety of the
caller, and this anxiety had made the caller contact the EMCC.
3.1.1. Concern for serious condition
The caller described that the patient felt bad, not only general
worsened condition but something more. The caller put it often as:
Caller (relative): He’s really sick.
Caller (nurse): She is really bad here.
3.1.2. Weakness
There was a broad spectrum of descriptions of patients who
seemed to have lost their normal energy. Most commonly, the caller
described the patient as being feeble or passive:
Caller (nurse): This is apparently, according to the staff who know
her, a quite lively old retiree who is normally ambulant and one
who. . . She is the most lively there in the department. . . Yes, like that
and. . . up and walking around and helping and yes, you know. And
now she has just been lying down.
3.1.3. Falling and collapsing
This subcategory contains all kinds of descriptions of falling, including collapsing and fainting. Most often, these symptoms were
used by relatives in out-of-hospital settings. Many callers did not
distinguish between descriptions such as falling and fainting.
Caller (relative): Well it. . . uhm she passed out yesterday and she
has been in bad shape for quite some time now. Passes out and falls
and so I called the ambulance . . .
3.1.4. Sudden deterioration
Sudden deterioration describes a disease evolving over time,
usually in the last hour or hours:
Caller (relative): He feels worse during the last . . . then only the last
hour.
Caller (nurse): He was healthy this morning . . .
3.2. Physical signs and symptoms
Breathing difficulties were commonly described, and the other subcategories were Nausea and Vomiting, Pain, Pallor and Measurable
signs of illness. Pallor was mentioned only in a few cases, where the
EMD specifically asked about the appearance of the patient. Laymen
and professionals both used these descriptions, but differed in descriptions of pain.
3.2.1. Pain
Professionals sometimes told the EMD about pain on palpation
when they examined the patient.

Caller (nurse): “And she has so much pain when I come and visit
her so I try to palpate the stomach and then, well. . . yes. She
feels. . . pain. . .”
Laymen usually talked about pain expressed by the patient:
Caller (relative): “Yes, she has had a huge headache, but now she
is taking pain killers, so she says that . . .”

3.2.2. Breathing difficulties
The caller mentioned heavy or strained breathing or wheezing
in more than half of the cases. Laymen often mentioned strained
breathing or just answered “No” to the EMD asking if the patient
was breathing normally.
Professionals usually described the breathing very early in the
conversation (often with a measured respiratory rate) and without
the need for attendant questions:
Caller (nurse): “We have a patient here, and the breathing is very
strained right now”

3.2.3. Measurable signs of illness
Fever was the most prominent description in this subcategory.
Both laymen and professionals measured body temperature.
However, only the latter reported the blood pressure level.
Measureable signs of illness showed the most distinct disparity
between the different types of callers, since professionals explained the specific values of vital signs to the EMD in every single
case and often early in the conversation. The measurements of
the vital signs were often the first information given by the
professionals:
Caller (nurse): “His temperature has risen fast and is 38.8. His
breathing is a bit fast, the blood pressure is 180/85, the pulse 88.
There is no indication of pulmonary oedema or anything now but . . .”
In comparison, laymen seldom told the EMD about fever unless
they were asked:
EMD: “Okay. What is her temperature?”
Caller (relative): “Yes, well. . . going to check here. . . [silence] 39.3.”

3.3. Difficulties establishing satisfactory contact with the patient
In almost half of the recordings, the caller mentioned problems talking to the patient, sometimes explained as the patient
being feeble or drowsy, but more often only vague descriptions
were used. Subcategories were confusion, drowsiness and simply not
answering. Professionals sometimes used more medical language,
with terms such as “decreased level of alertness.” Confusion was the
smallest sub-category, appearing in only a few calls. It differed substantially from the other subcategories by conveying that the patient
did answer, but not appropriately. Most often, the descriptions
of contact difficulties arose when the EMD asked if the patient was
unconscious.

3.3.1. Drowsiness
Many callers said sleepiness was the problem. Some said they
were able to wake up the patient but that the patient soon went
back to sleep and could not keep up a conversation.
Caller (relative): She can speak and. . .. So on, then she awakens
when I say that. . . she is barely able to contact, but it. . . it is not
much more than that.

K. Bohm et al./International Emergency Nursing 23 (2015) 294–298

3.3.2. Not answering
In several calls there was no explanation at all. Some callers
seemed unable to suggest any reason for the problem with the nonspeaking patient. The EMD often did not ask any subsequent
questions that could help the caller explain why the patient could
not speak.
EMD: Does she not respond?
Caller (relative): No . . .
EMD: Do you mean she is unconscious?
Caller: No but. . . she cannot say anything.
4. Discussion
In the current qualitative study we found that sepsis was described in terms of deterioration, physical symptoms, and
communication abilities of the patient.
There was a wide spectrum of descriptions used by the callers.
The main category ‘Difficulties in establishing contact with the
patient’ included vague descriptions of worsened general condition, while difficulties breathing, and difficulties establishing contact
with the patient were the most commonly mentioned symptoms.
This may reflect the fact that the EMD always asks about breathing and unconsciousness, since these are covered in the criteriabased protocol in the Swedish EMCC. An implication of the current
study is that the results can be added to a decision tool/protocol
but also to the EMD-education. Prior studies have shown that identification of sepsis in the EMS leads to less delay before antibiotic
treatment and protocol-driven treatment in the ED (Studnek et al.,
2012; Wang et al., 2010). Also, identification of the septic patient
by the EMD opens the possibility of initiating treatment in the ambulance, i.e. moving the treatment outside the hospital.
4.1. Different callers, different descriptions
Our results indicate an interesting difference between laymen
and professionals. The latter used more specific medical terminology when talking to the EMD than laymen. However, this did not
influence the analysis and coding process. In the current study it
was more common that the caller was a professional than a relative. Interestingly it was never the patient who called. It is shown
to be more difficult for the EMD to identify the symptoms when
there is a third party calling the EMCC (Karlsten and Elowsson, 2004).
Most of the patients presenting with sepsis in this study lived in
geriatric or nursing homes, and this explains the high rate of nurses
calling the EMCC. The work as an EMD differs substantially from
the work of other health care professionals, since the phone call provides only the caller’s interpreted version of the situation.

297

trigger follow-up questions to work out if the condition is acute,
and should also lead to suspicion of severe infection, but more research is needed to confirm this.

4.3. Physical signs and symptoms
Typical symptoms of sepsis (Levy et al., 2003; Martin et al., 2003)
were described in a clear way (e.g. fever was called fever), which
implies that callers were used to these descriptions. Pallor was only
mentioned in a few calls, despite being a common symptom of sepsis
(Martin et al., 2003). The EMD often asked about fever, but not
always. Since there is no specific EMD protocol for infection in the
Swedish EMCC – only for “fever” or “unknown problem” – not asking
about fever may slow down the identification process of sepsis. All
EMDs asked about breathing, but not whether the patient was
breathing faster than normal or not. More commonly they only asked
if the patient “was breathing normally”. Since laymen are probably not familiar with the concept of increased respiratory rate as a
symptom of sepsis the EMD should ask specific questions regarding this. Hypotension was never directly asked about by the EMD
and never mentioned by laymen. A patient’s fall could be caused
by hypotension, but could also have other reasons. Low blood pressure might not be known to laymen as a disease indicator (Bone
et al., 2009).

4.4. Difficulties establishing satisfactory contact with the patient
Difficulties establishing satisfactory contact with the patient, most
often used by laymen, described how the caller was unable to communicate with the patient as he or she normally did, and often came
in response to the EMD’s question about consciousness. The exact
term “lowered level of consciousness” was seldom used by laymen.
These descriptions differed substantially from the other categories because the difficulties establishing contact with the patient
seemed to be more subjective. Sometimes it was possible to hear
the patient talking in the background, even though the caller answered “yes” when the EMD asked if the patient was unconscious.
The EMD should listen to the patient through the telephone or even
better, ask if it is possible to talk with the patient, to get a better
understanding of the situation. A description of unsatisfactory contact
should be taken seriously.
The results of the current study could be used as the basis for
future quantitative studies. Central keywords could be incorporated into the decision tool, i.e. the protocol used by the EMD, once
identification of these keywords has been done.

5. Conclusions
4.2. Deterioration
Professionals used medical terminology when expressing symptoms, but when expressing presentations categorized as deterioration
they used descriptions that were as vague as those of the laymen.
An explanation for this could be that neither laymen nor professionals know of any “correct way” of expressing behavioural changes
in terms of change of condition. These changes are useful as a part
of pattern recognition for the EMD. A previous study suggests that
the activity level and body position of a patient might help the EMD
to make a correct identification (Clawson et al., 2008). Considering the symptoms in the subcategory falling and collapsing, it was
uncertain whether these falls were caused by general weakness, hypotension or something else. However, falling was mentioned for
patients with sepsis, which indicates that this symptom might be
of value for recognizing a septic patient. These symptoms could

Sepsis was described in terms of the physical symptoms, changes
of behaviour and communication abilities of the patient. This knowledge could lead to the identification of keywords which could be
incorporated into the decision tool used by the EMD to increase
sepsis identification, but further research is needed. The form of description seemed to differ between laymen and health care
professionals calling the EMCC, which challenges the adaption ability
and flexibility of the EMD.

Acknowledgements
We want to thank the dispatchers and the people who phoned
112. One of the authors (KB) receives unrestricted grants from SOS
Alarm AB.

298

K. Bohm et al./International Emergency Nursing 23 (2015) 294–298

References
Berdowski, J., Beekhuis, F., Zwinderman, A.H., Tijssen, J.G.P., Koster, R.W., 2009.
Importance of the first link description and recognition of an out-of-hospital
cardiac arrest in an emergency call. Circulation. 119 (15), 2096–2102.
Bone, R.C., Balk, R.A., Cerra, F.B., Dellinger, R.P., Fein, A.M., Knaus, W.A., et al., 2009.
Definitions for sepsis and organ failure and guidelines for the use of innovative
therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American
College of Chest Physicians/Society of Critical Care Medicine. 1992. Chest. 136
(5 Suppl.).
Clawson, J., Olola, C., Heward, A., Patterson, B., Scott, G., 2008. Ability of the medical
priority dispatch system protocol to predict the acuity of “unknown problem”
dispatch response levels. Prehospital Emergency Care. 12 (3), 290–296.
Fevang, E., Lockey, D., Thompson, J., Lossius, H.M., Collaboration, T.R., 2011. The top
five research priorities in physician-provided pre-hospital critical care:
a consensus report from a European research collaboration. Scand J Trauma
Resusc Emerg Med. 19, 57.
Graneheim, U.H., Lundman, B., 2004. Qualitative content analysis in nursing research:
concepts, procedures and measures to achieve trustworthiness. Nurse Education
Today. 24 (2), 105–112.
Herlitz, J., Bång, A., Wireklint-Sundström, B., Axelsson, C., Bremer, A., Hagiwara, M.,
et al., 2012. Suspicion and treatment of severe sepsis. An overview of the
prehospital chain of care. Scand J Trauma Resusc Emerg Med. 20, 42.
Karlsten, R., Elowsson, P., 2004. Who calls for the ambulance: implications for decision
support. A descriptive study from a Swedish dispatch centre. European Journal
of Emergency Medicine. 11 (3), 125–129.

Kumar, A., Roberts, D., Wood, K.E., Light, B., Parrillo, J.E., Sharma, S., et al., 2006.
Duration of hypotension before initiation of effective antimicrobial therapy is
the critical determinant of survival in human septic shock. Critical Care Medicine.
34 (6), 1589–1596.
Levy, M.M., Fink, M.P., Marshall, J.C., Abraham, E., Angus, D., Cook, D., et al., 2003.
2001 SCCM/ESICM/ACCP/ATS/SIS international sepsis definitions conference.
Critical Care Medicine. 31 (4), 1250–1256.
Martin, G.S., Mannino, D.M., Eaton, S., Moss, M., 2003. The epidemiology of sepsis
in the United States from 1979 through 2000. The New England Journal of
Medicine. 348, 1546–1554.
Rangel-Frausto, M.S., 1999. The epidemiology of bacterial sepsis. Infectious Disease
Clinics of North America. 13 (2), 299–312, vii.
SOS Alarm AB, 2001. Svenskt Medicinskt Index för Akutmedicinsk Larmmottagning
2nd ed. Åsmund S. Laerdal A/S, Stavanger.
Studnek, J.R., Artho, M.R., Garner, C.L., Jones, A.E., 2012. The impact of emergency
medical services on the ED care of severe sepsis. The American Journal of
Emergency Medicine. 30 (1), 51–56.
Wallgren, U.M., Castrén, M., Svensson, A.E., Kurland, L., 2014. Identification of adult
septic patients in the prehospital setting: a comparison of two screening tools
and clinical judgment. European Journal of Emergency Medicine. 21 (4), 260–265.
Wang, H.E., Shapirom, N.I., Angus, D.C., Yealy, D.M., 2007. National estimates of severe
sepsis in United States emergency departments. Critical Care Medicine. 35 (8),
1928–1936.
Wang, H.E., Weaver, M.D., Shapiro, N.I., Yealy, D.M., 2010. Opportunities for emergency
medical services care of sepsis. Resuscitation. 81 (2), 193–197.
Wenzel, R.P., 2002. Treating sepsis. The New England Journal of Medicine. 347 (13),
966–967.