EthnographicStudy_Depression.pdf

RESEARCH ARTICLE Open Access

An ethnographic study of the effects of
cognitive symptoms in patients with major
depressive dis : the IMPACT study
Bjarke Ebert1*, Kamilla Miskowiak2, Morten Kloster3, Jon Johansen1, Cara Eckholm3, Torbjörn Wærner1,
Mads Holme3 and Louise Meldgaard Bruun4

Abstract

Background: The manifestation of major depressive dis (MDD) may include cognitive symptoms that can precede
the onset of MDD and persist beyond the resolution of acute depressive episodes. However, little is known about how
cognitive symptoms are experienced by MDD patients and the people around them.

Methods: In this international (Brazil, Canada, China, France, and Germany) ethnographic study, we conducted semi-structured
interviews and observations of remitted as well as symptomatic MDD patients (all patients self-reported being diagnosed by an
HCP and self-reported being on an antidepressant) aged 18–60 years with self-reported cognitive symptoms (N = 34).
In addition, participating depressed patients’ close family or friends (N = 31) were interviewed. Separately recruited
from depressed participants, work colleagues (N = 21) and healthcare providers (HCPs; N = 13) of depressed individuals
were interviewed.

Results: Key insights were that: (1) patients were generally unaware that their cognitive symptoms were linked to their
depression and, instead, attributed these symptoms to negative aspects of their person (e.g., age, separate disease,
laziness, exhaustion); (2) cognitive symptoms in MDD appeared to negatively impact patients’ social relationships and
patients’ ability to handle daily tasks at work and at home; (3) patients’ cognitive symptoms also impacted relationships
with family members and coworkers; (4) patients’ cognitive symptoms increased stress and feelings of failure, which in
turn seemed to worsen the cognitive symptoms, thereby creating a destructive cycle; and (5) although HCPs recommended
that patients re-engage in everyday activities to help overcome their depression, cognitive symptoms seemed to impede
such functional recovery.

Conclusions: Taken together, these findings highlight a negative impact of patients’ cognitive symptoms on their social
functioning, work performance, and quality of life on the people close to them, and consequently on the degree of
functional recovery after depression.

Keywords: Major depressive dis , Cognitive symptoms, Qualitative study, Brazil, Canada, China, France, Germany

Background
A case study
Clara, a 23-year-old student, had just begun a master’s
degree at a French university when she suddenly fell into
what she describes as a “third dimension.” Her mood
lowered, she was stressed and irritable, and she started
getting into fights with her mother. Her brain seemed to
slow down and everyday tasks, such as homework, became

insurmountable. “I felt like I was stoned or drunk, or some-
thing,” says Clara, reflecting on her mental state throughout
her first months at university. Clara fretted but took no
action—until one day when she woke up with her head
spinning. The vertigo was so severe that she could not get
out of bed, prompting her to make a rush appointment
with her general practitioner. After a couple of sessions,
Clara was diagnosed with major depressive dis
(MDD) and prescribed medication.
For the following 2 months, Clara did not attend clas-

ses; she quit her part time job as a secretary at a bank. “I
* Correspondence: [email protected]
1H. Lundbeck A/S, Ottiliavej 9, 2500 Valby, Denmark
Full list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Ebert et al. BMC Psychiatry (2017) 17:370
DOI 10.1186/s12888-017-1523-8

http://crossmark.crossref.org/dialog/?doi=10.1186/s12888-017-1523-8&domain=pdf

mailto:[email protected]

http://creativecommons.org/licenses/by/4.0/

http://creativecommons.org/publicdomain/zero/1.0/

was exhausted physically and intellectually, at every
level,” she stated. She was bedridden with a terrible
mood and with little mental capacity or willpower. By
the time of our visit 6 months later, Clara was back at
university and at work, but her return to normality had
been riddled with frustration. Her mood stabilized, but
her cognitive abilities were still plagued by abnormal
lapses. Her memory was shoddy: “It still happens that I
ask a question, the person answers, and I ask it again 2
minutes later,” and her ability to concentrate was in a
state of disrepair: “I was in a 4-hour exam, and for 3 of
the hours I didn’t write anything at all,” she lamented.
How do we make sense of Clara’s struggle to recover

from depression? Historically, MDD has been typified by
its affective symptoms, related to lowered mood. How-
ever, recent studies have shown that a range of cognitive
symptoms is often present during MDD and can persist
after symptomatic remission from depression [1, 2].

Burden of depression and cognitive symptoms
MDD is a common psychiatric dis , with an estimated
point prevalence of 4.7% of the global population [3]. The
negative consequences are profound and impact health-
related quality of life, workplace performance and sick days
(presenteeism and absenteeism, respectively), and interper-
sonal relationships [4–9]. The global economic toll of
MDD was estimated to be US$800 billion in 2010 and is
rising [10]. In the US alone, the incremental economic
burden of individuals with MDD in 2010 was estimated to
be US$210.5 billion, with approximately half of the total
incremental costs associated with the workplace [11].
Cognitive symptoms are a common feature of MDD

that contributes to this large economic burden. In par-
ticular, cognitive complaints in MDD can be broad and
non-specific, with deficits observed across processing
speed, attention, learning and memory, and executive
functioning [1, 2]. Evidence suggests that cognitive
symptoms directly influence workplace performance
(perhaps to an even greater extent than affective symp-
toms) and directly contribute to socio-occupational dis-
ability independently of mood symptoms [12, 13]. Subtle
cognitive symptoms may also precede the onset of the
formal major depressive episode (MDE) and increase the
risk of illness onset [1, 14]. Furthermore, deficits in
processing speed, attention, learning and memory, and
executive functioning may persist beyond depressive
remission as residual cognitive symptoms [1, 15, 16].
Despite evidence of the inherent harms and long-lasting

effects of cognitive symptoms, MDD treatment remains
primarily directed at managing mood symptoms, with
limited attention given to the management of associated
cognitive complaints [17, 18]. Nevertheless, only a few
studies have investigated how patients with MDD experi-
ence cognitive symptoms and the consequent implications

of those experiences for the people around them. This
ethnographic study aims to evaluate the perceived impact
of cognitive symptoms on the everyday life of MDD
patients and of those with whom patients are regularly in
contact. The insights from this study will help inform the
development of educational material for people with
depression and their families, friends, and work colleagues
and help lead to the development of new tools to help
healthcare providers (HCPs) better identify and treat
people who experience cognitive symptoms in their every-
day life in connection with MDD.

Methods
Study design
This qualitative ethnographic study—the IMPACT study
(Investigating MDD Patients’ Accounts of Cognitive
Symptoms During Treatment)—was conducted between
May and June 2014 in major cities in Brazil, Canada,
China, France, and Germany. The study utilized a patient-
centric research model to capture the experience of cogni-
tive symptoms in MDD in daily life, including their impact
on patients and their partners/relatives/friends, colleagues,
and HCPs. The protocol and research themes were
informed by an earlier pilot study conducted in Denmark
from December 2013 to January 2014. In the pilot study,
we conducted 36 semi-structured interviews and observed
depressed patients with cognitive symptoms (N = 14),
their family or friends (N = 14), and HCPs (N = 8). A key
observation was that participants consistently stated that
their cognitive symptoms were impairing their work life.
Based on this observation, this study included interviews
with coworkers.
The IMPACT study was conducted following the ethical

standards outlined by the ICC/ESOMAR International
Code on Market and Social Research [19], which sets out
global standards for self-regulation for researchers and data
analysts, as well as relevant national standards for partici-
pating countries [20–23]. Patients signed consent forms
and trained anthropologists conducted all interviews.

Participants and recruitment strategies
Patients
Patients with MDD aged 18–60 years were recruited and
screened for study participation in Brazil, Canada, China,
France, and Germany. To find patients, ReD Associates, a
consultancy that specializes in anthropological research,
worked with experienced, local recruitment agencies that
specialize in identifying patients to partake in medical
studies. Once the recruitment agency confirmed a patient
met all criteria, ReD conducted an additional telephone
screening to re-confirm each patient’s fit for the study. In
this conversation, ReD researchers checked that patients
had received a relevant diagnosis by a qualified HCP and
that patients were taking relevant medication for MDD

Ebert et al. BMC Psychiatry (2017) 17:370 Page 2 of 13

(excluding those known to cause/affect cognitive symp-
toms). During the daylong in-home interview, the investi-
gator confirmed the presence of the antidepressant
identified during screening. At least one partner, relative,
or close friend of each MDD patient accompanied the
patient for a portion of the in-home interview (≥1 h).
Criteria for patient participation included self-reported (1)
history of an MDE within the previous year, with moder-
ate or severe symptoms and without a diagnosis of any
mental illness other than MDD; (2) memory and concen-
tration complaints while depressed, ranging in severity
from “some” to a “large” extent; and (3) to be on an anti-
depressant medication and be capable of naming the anti-
depressant. Investigators made no formal, independent
diagnosis of MDD; the presence of MDD was determined
by a patient’s self-reported diagnosis from an HCP, self-
reported depression history, and self-reported usage of
antidepressant medication. Seven patients were recruited
in each country. Recruitment targets applied to each
country are listed in Table 1 and were chosen to produce
a diverse pool of patient experiences. Patients received a
modest stipend in compensation for their time.

Coworkers and workplaces
Workplace participants (i.e. individuals who had worked
closely with someone who had depression) were identi-
fied and recruited separately from patient participants.
Recruitment targets (for each country) are listed in Table 2
and were chosen to produce a diverse pool of coworker
experiences and workplace environments. Both the work-
place and the employee affected by depression needed to
fit the same work-type category (e.g. a knowledge worker
in a company that generally does knowledge work).

Healthcare providers
HCPs were also identified and recruited separately from
patient participants. Recruitment targets (for each coun-
try) are listed in Table 3 and were chosen to produce a
diverse, yet knowledgeable pool of HCP experiences and

practices. All HCPs were required to be active in the
management of patients with MDD and to have special-
ized in the field. All were required to have ≥5 years at
their current practice and to diagnose and treat MDD
patients on a regular basis, with ≥20% of their MDD pa-
tients identified as having cognitive symptoms. Psychiatrists
and general practitioners had to prescribe antidepressants
regularly and had to have prescribed a new brand or type
of antidepressant ≤18 months prior to study participation.

Recruitment in Germany
All countries recruited equally for the IMPACT study
with the exception of Germany, where recruitment was
affected by issues related to study timelines and local
review board approval. In Germany, only six depressed
patients were recruited, three of whom were accompan-
ied by a secondary participant (partner or friend) for
part of the interview. In addition, no workplaces, col-
leagues, or HCPs were recruited in Germany.

Data collection
The IMPACT study used a patient-centric research model to
capture the experiences of patients with cognitive symptoms
in MDD in everyday life. The implemented ethnographic
methodologies included semi-structured interviews, observa-
tions of relevant activities, mapping exercises, and situation
card exercises. Research was conducted in the home or
office of the participant, depending on the participant type.
Interviews were conducted by trained ethnographers,

with university degrees in the social and human sciences,
from ReD Associates and consisted of a daylong visit to the
patient’s home, with the secondary participant present for
≥1 h. HCP interviews lasted ~2 h and were conducted at
their place of work. Coworker interviews lasted ≥1.5 h and
were conducted at their place of work where possible. In-
terviews explored a range of themes dependent on the par-
ticipant type: patients (Additional file 1: Table S1a), HCPs
(Additional file 2: Table S1b), or colleagues (Additional file 3:
Table S1c). Mapping exercises involved the patient drawing

Table 1 IMPACT study entry criteria for patient participants in each country

Criteria Targets

Condition-related Depression type Mix of moderate to severe typologies

Memory problems Self-reported memory problems, ranging from “some” to a “large” extent

Concentration problems Self-reported concentration problems, ranging from “some” to a “large” extent

Socio-demographic Gender ≥3 men ≥3 women

Age, years Minimum age, 18 ≥3 patients between 30–44 Maximum age, 60

Living situation Mix of living: alone, with parents or siblings, with partner without children,
or with partner with children

Education ≥4 patients with at least a college degree ≥1 patient with no more than high school education

Work-related Work type Mix of workers: administrative, knowledge, people-centered, or physical

Sick leave ≥2 patients on sick leave ≥2 patients who have returned to work

Ebert et al. BMC Psychiatry (2017) 17:370 Page 3 of 13

the evolution and components of their depression and the
development of their cognitive symptoms; situation card
exercises focused on determining the situations in which
cognitive symptoms are most obvious by having the patient
describe their cognitive complaints before and after the on-
set of their depression in the context of the situation pre-
sented on the card.
All interviews were recorded with the permission of

the participant and field notes were written following
each interview. All participant names and identifying
information have been changed to protect privacy.

Data analysis
We utilized the “sense making” method [24], which begins
with the study of a phenomenon—in this case the experi-
ence of cognitive symptoms in depression. The data
driving the study of the phenomenon are not quantitative
in nature, but consist of pictures, emotions, artifacts, ob-
served behavior, and conversations. A model of phenom-
enological explication inspired by Heidegger’s concept of a
“formal indication” is then used to make sense and
discover patterns in the data that have been gathered [25].
Our analysis began by a series of open, data-driven

story-telling sessions. During these sessions, each re-
searcher gave a thorough presentation of each of their
respondents to the full project team, offering “thick de-
scriptions” of their respondents’ behavior (i.e., detailed
and specific), which illustrated the nuances of their cul-
turally complex gestures [26]. These presentations drew
on each researcher’s field notes, as well as audio and
visual material from the interview. At the conclusion of
the story-telling sessions, the researchers stepped away
and independently made decisions about the key themes
represented. The team then compared responses and
collectively agreed on thematic categories into which the

data would be coded, ensuring that the analysis was
firmly grounded in data.
The researchers systematically coded all interviews accord-

ing to these key thematic areas. When the coding was con-
cluded, the researchers rigorously tested the thematic
content of the codes by exploring similarities and differences
between participants, and across research markets. Special
attention was paid to examining unexpected findings, and
active efforts were made to disprove prior interpretations.
The findings are thus a result of an iterative process, in
which the team routinely challenged its own assumptions
and refined the analysis until consensus was achieved.
Using the ethnographic interviews, patient observa-

tion, mapping exercises, and situation card exercises,
analysts derived four work profiles representing a variety
of challenges for working professionals with cognitive
symptoms in MDD. We spoke with each patient at
length about their job, rating its cognitive requirements
in nine key areas (concentrating/staying focused, pro-
cessing information, memory, overview, verbal abilities,
problem solving/decision making, attention control,
motor skills and perception, and ability to initiate
actions). After experimenting with different groupings to
find meaningful correlations between job type (e.g.
“people-centered” versus “knowledge-centered”), chal-
lenge areas (i.e. the largest discrepancies between the
job’s cognitive requirements and the depressed patient’s
cognitive performance), and consequences (i.e. what
happened, both internally and externally, for the patient
and others as a result of the challenge areas), clear pat-
terns emerged by grouping along two axes: (1) whether
the focus of the work is social or is based on content,
and (2) whether the work involves leading or following
processes (Fig. 1). Social focus jobs are those with sig-
nificant interaction with others (e.g. account director,
public relations specialist), whereas content focus jobs

Table 2 IMPACT study entry criteria for workplaces in each country

Criteria Targets

Size of workplace ≥1 company with 6–49 employees ≥1 company with ≥50 employees

Work typea Mix of workplaces: administrative, knowledge, people-centered, or physical

Relation to depressed patient Mix of colleagues, superiors, and HR managers

HR department ≥1 company with a dedicated HR department ≥1 company with no dedicated HR department

HR human resources
aBoth the workplace and the employee affected by depression needed to fit the same category (e.g. a knowledge worker in a company that generally does
knowledge work)

Table 3 IMPACT study entry criteria for healthcare providers in each country

HCP Clinical experience Proportion of patients with MDD Frequency of prescribing
antidepressants

Last time prescribing new brand
or type of antidepressants

Psychiatrist ≥5 years in current practice ≥20% of patients with MDD ≥4 times a week ≥18 months previously

General practitioner ≥2 times a week

Psychologist N/A N/A

HCP healthcare provider, MDD major depressive dis , N/A not applicable

Ebert et al. BMC Psychiatry (2017) 17:370 Page 4 of 13

would involve less interaction and more focus on the
project or deliverable at hand (e.g. architect, program-
mer). Leading versus following processes refers to
whether an individual drives a team/project (e.g. medical
director, police commissioner) or someone else in that
team/project provides them direction (e.g. medical
writer, police officer).

Results
Participant characteristics
In all, 34 depressed patients; 31 partners, relatives, or
friends; 21 colleagues from 18 workplaces; and 13 inde-
pendent HCPs participated in the study (Table 4). More
than half of the participating MDD patients were women
(19/34; 56%) and most were employed (30/34; 88%). The
mean (±standard deviation) age of the depressed patients

was 40 (±9.1) years. Demographic characteristics were
similar across countries. Of the partners/relatives/friends
interviewed, there were 13 spouses/partners, 5 parents, 2
children, 2 siblings, and 11 other close acquaintances
(i.e. friend, neighbor, roommate, or extended family).
Workplace colleagues included 14 coworkers, 3 supe-
riors, 2 HR managers, and 2 HR representatives. The in-
dependent HCPs interviewed included 5 psychiatrists, 4
psychologists, and 4 general health practitioners.

Patient perception of cognitive symptoms and the impact
on daily life
The missing link between cognitive symptoms and
depression
Most depressed patients had experienced cognitive
symptoms before realizing that these might be related to
their depression. Before discovering this relationship,

The independent

Patients who must begin and
finish tasks on their own accord

The team player

Patients who must keep up with
others and be sociable

The expert

Patients who must reliably solve
problems on demand

The organizer

Patients who must keep an
overview and think on their feet

S
O

C
IA

L
F

O
C

U
S

S
U

C
O

F
T

N
E

T
N

O
C

LEADS PROCESS

FOLLOWS PROCESS

Overview

Memory

Attention

Problem
solving

Concentration

Initiation

Overview

Memory

Attention

Problem
solving

Concentration

Initiation

Overview

Memory

Attention

Problem
solving

Concentration

Initiation

Overview

Memory

Attention

Problem
solving

Concentration

Initiation

Fig. 1 Work profiles

Table 4 Study population of the IMPACT study by patient type and country (city)

Population interviewed, n Brazil (Sao Paolo) Canada (Toronto) China (Shanghai) France (Paris) Germany (Berlin, Hamburg) Total

Patients with MDD 7 7 7 7 6 34

Partners/Relatives/ Friends 7 7 7 7 3 31

Colleagues (Workplaces) 4 (4) 5 (4) 6 (4) 6 (6) 0 21 (18)

HCP 3 3 4 3 0 13

HCP healthcare provider; MDD major depressive dis

Ebert et al. BMC Psychiatry (2017) 17:370 Page 5 of 13

many had attributed their symptoms to innate negative
aspects of their personality, stress in the work environ-
ment, or family problems. Some even feared the pres-
ence of other dis s, such as Alzheimer’s disease.
When Clara (age 23, France) experienced mental foggi-
ness and vertigo, she “didn’t realize that depression could
have a physical and intellectual impact at the same time.”
When another patient, Maya (age 38, Canada), was first
diagnosed with depression, “people began to ask [her] if
[she] had ADHD.” She assumed that her cognitive symp-
toms “might have to do with age,” but this perception
was corrected by her HCP.
Some patients remembered that discovering the rela-

tionship between their cognitive symptoms and their
depression was a point of relief. As one patient stated:

“It was a turning point for me when I found out that
the problems [cognitive symptoms] were due to my
depression. … [I was terrified of having something]
much more serious—I thought I was going to die.”
(Sara, age 49, Brazil)

Impact on activities of daily living and close relatives
Patients in all five countries indicated their cognitive
symptoms of depression (e.g. leaving the water running,
forgetting the keys in the door, losing orientation, mixing
up words) interfered with their activities of daily living
(ADL). In addition to making mistakes and losing track
of their activities during ADL, patients experienced
mental exhaustion from activities they had formerly con-
sidered routine and were often unable to turn intentions
into actions.
For many patients, difficulties with prioritizing tasks

and keeping organized had a large impact on ADL, caus-
ing the tidiness of their homes to deteriorate. At one
point, a single mother with two children found it too
mentally overwhelming to shop for groceries and cook
food for her children, so the family stopped eating
together and everyone began to feed themselves. This
depressed patient reported feeling like a bad mother and
worried about how she was “hurting” her children by
not being able to prepare meals or help them with
homework and allowing them to form bad habits. Another
patient’s inability to initiate and sustain action led him to
do less work around the house, including planned home
repairs and renovations. He used to enjoy and excel at
home improvement, but after his cognitive symptoms
emerged he had difficulty prioritizing and did not get
around to doing anything unless he was explicitly asked.
He and his girlfriend saw their home deteriorate around
them and knew it to be a financial loss:

“I’ll just sit there and do nothing. … We bought the
house to fix it up. I looked forward to it. I haven’t done

anything in more than a year. The windows need
renovation, I know, but… it’s worth less now than
when we bought it.” (Steve, age 51, Canada)

The partners and relatives of patients with MDD in all
five countries reported significant effects on interper-
sonal relationships. These individuals indicated they
often had to take on additional tasks formerly performed
by the patients and to contend with unpredictable emo-
tions and behavior. In one case, the long-term girlfriend
of a depressed patient experiencing cognitive symptoms
related to initiation and concentration had taken over
much of the housework that her boyfriend used to
complete. She began feeling that her boyfriend had
become someone she had to care for, like a child. He no-
ticed how the relationship had changed and felt guilty
about being a bad partner and a burden to his girlfriend.

Impact on social life and recreation
Many patients talked about the negative social and
recreational consequences of having strained work life
due to cognitive symptoms. One patient described how,
exhausted by the effort it took to function with her cog-
nitive symptoms at work, she had been isolating herself
from social and leisure activities. To her, and to many
other patients, going out with friends was not as import-
ant as work. In other cases, recreational activities were
directly hindered by cognitive symptoms. For example,
one patient explained that his cognitive symptoms
hampered his ability to engage in his favorite hobby,
playing multiplayer online games with his university
friends. He now found gaming more difficult and could
no longer anticipate other gamers’ moves when playing.
This made playing less enjoyable and he had therefore
withdrawn from this otherwise enjoyable and social
hobby. A friend of this patient reported feeling that he
had seemingly lost interest in the game and in them. He
did not understand why his friend had become such a
bad team member and felt rejected:

“He used to be a really solid player. I don’t know
what’s up with him. If he’s tired of playing, he should
just quit. It’d be a shame to lose him; he’s a good
friend, but it’d be better than it is now.” (China)

Impact on professional life
In working patients, cognitive symptoms often interfered
with their work performance. More than half of the
patients interviewed (18/34; 53%) indicated they had
changed jobs due to cognitive symptoms. Four other
patients (12%) indicated they had considered changing
jobs because of cognitive symptoms.
When describing their work performance, patients

from all countries used words relating to forgetfulness,

Ebert et al. BMC Psychiatry (2017) 17:370 Page 6 of 13

distractibility, and reduced problem solving—terms that
are key indicators of cognitive symptoms in MDD. As
one patient stated: “At one point, I overlooked a meeting
with a client. I simply stood him up. I can’t have something
like that happen again.” (Ana, age 36, Brazil).
According to another patient: “If I’m in the middle of

something and someone walks up to me and asks a
question, then I have to start all over with the thing I
was doing. I can’t just stop and then pick up where I left
off. I have to start from the beginning; otherwise I’m lost.”
(Beathe, age 43, Germany).
Colleagues of depressed patients reported that the indi-

viduals with depression were less productive and made
mistakes that damaged business outcomes. These effects
created negativity and resentment, as well as fostered
additional tensions among other workers in the same
business unit. Allie, an administrator at a public health
organization who used …