Available via license: CC BY-NC-ND 3.0
Content may be subject to copyright.
Original article
Signs and Symptoms of Impending Death in End-of-life
Elderly Dementia Sufferers: Point of View of Formal
Caregivers in Rural Areas
–A Qualitative Study–
Yoshihisa Hirakawa and Kazumasa Uemura
Center for Postgraduate Clinical Training and Career Development, Nagoya University Hospital, Japan
Abstract
Objective: The aim of the present study was to clarify the signs and
symptoms of impending death in end-of-life senile dementia from
the point of view of formal caregivers in rural areas.
Patient/Materials and Methods: We used qualitative data based
on retrospective analyses. The data was gathered following a work-
shop on end-of-life care of the elderly with dementia attended by
formal caregivers that was held in Iga City, Mie Prefecture, Japan,
in September 2011. There was a total of 29 workshop participants.
The workshop products were created in the rst session of the
workshop entitled “Signs of death.” During the session, we used the
brainstorming method, and participants took turns stating at least
two signs, symptoms or premonitions of death. In the end, there
were 93 cards in total displaying signs of impending death observed
in the end stage of dementia. These 93 entries were then classied
into clear categories.
Results: The categories dened were breathing disorder, conscious-
ness decline, vital power decline, reduced oral intake, feces disor-
der, calm and peaceful character, blood pressure decline, change in
skin color, patient odor, edema, preagonal vital power, body tem-
perature decline, bedsore/wound deterioration, body weight reduc-
tion, cyanosis, and oliguria. The most frequently cited symptoms
fell in the breathing disorder category (12 cards), followed by con-
sciousness decline (9 cards), vital power decline (9 cards), reduced
oral intake (6 cards), and feces disorder (6 cards). Also frequently
mentioned were symptoms falling in the calm and peaceful charac-
ter, patient odor and preagonal vital power categories.
Conclusion: The results show that formal caregivers in rural areas
identied breathing disorder as a top indicator of impending death
in end-of-life senile dementia cases. The results also highlight some
other characteristic signs of impending death, such as preagonal
vital power and calm and peaceful character. This research could
help develop formal caregivers’ observational skills in the end-of-
life care settings.
Key words: dementia, symptom, end-of-life, qualitative study, for-
mal caregiver
(J Rural Med 2012; 7(2): 59–64)
Introduction
It is generally believed that, in Japan, elderly people
would rather pass away at a long-term care facility or in the
home where they have spent most of their life, although in
recent years, a growing number of Japanese people spend
their last days of life in hospitals
1, 2)
. However, hospitals no
longer have the spare bed capacity to cope with the increas-
ing demands of an aging population
1)
. Therefore, institu-
tional and home end-of-life care for the elderly has received
much attention in Japan, especially in rural areas, where
there has been a dramatic rise in elderly patients and a se-
rious shortage of hospital physicians
3)
. In community set-
tings, such as long-term care facilities or the home, the role
of formal caregivers is more signicant in end-of-life care
of the elderly than in hospitals
4, 5)
. Formal caregivers also
develop a closer relationship with dying patients and their
families than hospital staff. Moreover, in cooperation with
family members, formal caregivers are required to prepare
for death and to notify local medical services of any sudden
change in the patients’ condition. As a result, they inevitably
need to deepen their understanding of the signs and symp-
toms of advanced dementia through education.
The main diseases of aged people in end-of-life care are
cancer, dementia, and internal organ failure
6)
. Literature on
the pain and suffering of aged cancer sufferers is widely
J Rural Med 2012; 7(2): 59–64
©2012 The Japanese Association of Rural Medicine
Correspondence to: Yoshihisa Hirakawa, Center for Postgraduate Clini-
cal Training and Career Development, Nagoya University Hospital, 65
Tsuruma-cho, Showa-ku, Nagoya City, Aichi 466-8560, Japan
E-mail: y.hirakawa@med.nagoya-u.ac.jp
60
available
7–10)
. The research shows the following dominant
signs and symptoms: loss of appetite, pain, whole body mal-
aise, dyspnea, fever, edema, coma, delirium, pleural effu-
sion and ascites.
However, further research on the signs and symptoms of
advanced dementia in the elderly is necessary, especially in
light of the fact that end-of-life dementia is increasing, partic-
ularly in recent years due to the sharp rise in the aging popu-
lation
11–13)
. Research on the topic is truly lacking, and relevant
data has not yet been adequately collected
13)
. This is in part
because it is difcult to garner clear indications of the signs
and symptoms from elderly people who suffer from dementia
due to their impaired and lowered communication and cogni-
tion skills
13–15)
. Also, there may be a lack of acknowledgment
and understanding among nonmedical professionals that de-
mentia is a syndrome causing death
14, 16, 17)
. This is actually a
reection of the opinions of medical professionals who have
a strong inuence on nonmedical professionals.
Moreover, those caregivers that are at the forefront of
end-of-life care of dementia sufferers are not always accus-
tomed to medical terminology, and as a result, their obser-
vations may differ from the opinions of physicians and nurs-
es. Thus, we recognize the need for the opinions of formal
caregivers regarding the symptoms and signs of impending
death of elderly people suffering from end-of-life dementia
to be formulated clearly. Therefore, the present study aims
at clarifying the signs and symptoms of impending death in
end-of-life senile dementia from the point of view of formal
caregivers.
Materials and Methods
Formal caregivers are dened as paid nonmedical care
providers (including certied care workers) from a private
agency or a government or nonprot organization in the
present study. We used qualitative data based on retrospec-
tive analyses. The data was gathered following a workshop
on end-of-life care of the elderly with dementia attended
by formal caregivers that was held in Iga City, Mie Prefec-
ture, Japan, in September 2011. The workshop was entitled
“Thinking about Dementia Palliative Care and Behavioral
and Psychological Symptoms of Dementia (BPSD).” Iga
City, which located in the northwestern part of Mie Prefec-
ture, has a population of 100,000. The city is in a typical
rural area, roughly positioned between Osaka and Nagoya.
About 27% of the people in Iga are over 65 years of age.
Workshop participants were selected with the help of the
social welfare council ofce of Iga City from a wide range
of people involved in caregiving. Iga suffers from an acute
shortage of hospital physicians, and we therefore believed
that the workshop products would be very instructive, since
many caregivers experienced with end-of-life care outside
the hospital would contribute to the research. We also be-
lieved that there would be many caregivers in this area with
bedside experience of death from dementia. These are the
main reasons why our research was based in Iga City. There
were 29 workshop participants in total. The average age of
the participants was 46.9 years old. There were 24 female
and 5 male participants. Eighteen people held administra-
tive positions in their workplaces.
The workshop products were created in the rst session
of the workshop entitled “Signs of death.” During the ses-
sion, the participants were divided into smaller groups of 7
to 8 people. We used the brainstorming method in the ses-
sion. Because brainstorming works best with a varied group
of people, each group consisted of participants who came
from various departments across the organization. The
rules of brainstorming were “focus on quantity,” “withhold
criticism,” “welcome unusual ideas” and “combine and im-
prove ideas.” Also, to stimulate idea generation, the partici-
pants were required to take turns stating at least one sign,
symptom or premonition of death. This continued until no
participants had any original ideas, and at least two signs,
symptoms or premonitions of death were required per per-
son. Each group elected a record keeper, who made a list of
all signs, symptoms or premonitions of death stated in each
group.
The workshop products were analyzed in December
2011. First, we transferred all the signs, symptoms or pre-
monitions of death from the lists given by each group to
individual cards, for a total of 89 cards. Next, in cases where
more than two symptoms were listed on a single card, we
separated the symptoms and listed them on separate cards.
For instance, we had cards for “hypothermia/hypotension,”
“melena/hematemesis,” “hallucination/twilight state” and
“no reduction in fever/pneumonia,” which we subsequent-
ly listed individually on 8 cards. In the end, there were 93
cards in total displaying signs of impending death observed
at the end stage of dementia.
These 93 entries were then classied into clear catego-
ries by the head author of this report and 3 assistant re-
searchers familiar with the purpose of the research. This
was accomplished by the head author reading out the cards
one by one and then the assistant researchers grouping the
cards, which were laid out in front of them, together with
other cards that looked like they belonged with them. The
head author and the 3 assistant researchers discussed the
groupings and then gave them an appropriate title. Effort
was made to avoid technical terms so as to allow caregiv-
ers to relate to them quickly. Careful consideration was also
taken not to adhere to the usual preconceptions, ensuring a
fresh approach to the research topic. Furthermore, we listed
61
separately a number of symptoms that did not t into any
of the outlined categories. These “solitary cards” were not
excluded from the research but were classied as separate
entries. Finally, we reviewed the entries in each category
carefully until we reached a consensus among our research
team members. This research was carried out with the con-
sent of the Nagoya University School of Medicine Ethics
Committee (Approval number 82).
Results
The cards were numbered and collated according to cat-
egory. Table 1 lists the various categories and symptoms
listed on the cards. The categories dened were breathing
disorder, consciousness decline, vital power decline, re-
duced oral intake, feces disorder, calm and peaceful char-
acter, blood pressure decline, change in skin color, patient
odor, edema, preagonal vital power, body temperature de-
cline, bedsore/wound deterioration, body weight reduction,
cyanosis and oliguria. The most frequently cited symptoms
fell in the breathing disorder category (12 cards), followed
by consciousness decline (9 cards), vital power decline (9
cards), reduced oral intake (6 cards) and feces disorder (6
cards). Also frequently mentioned were symptoms falling
in the calm and peaceful character, patient odor and preago-
nal vital power categories. Other symptoms such as ascites
and presence of water in the lungs that did not t into other
categories were listed in the solitary cards category. Other
symptoms difcult to classify such as family members be-
come kinder and get along better with family, whose inter-
pretations are difcult to ascertain, were also placed in the
solitary cards category, since they were related to the theme
of this research.
Discussion
This research has an original approach to its topic, in
that it attempts to clarify the signs and symptoms of death in
end-of-life dementia cases from the point of view of formal
caregivers in rural areas. Since the setting of the research
was limited to a single rural area in Mie Prefecture, Japan,
care needs to be taken regarding the interpretation and gen-
eralization of the results; nevertheless, we believe that this
research could help develop formal caregivers’ observation-
al skills in the end-of-life care settings.
The results show that caregivers in rural areas identi-
ed symptoms falling into the breathing disorder category
with the greatest frequency, conrming it as a top indicator
of impending death. Dyspnea was sometimes expressed as
breathing hard, panting and shoulder breathing, and breath-
ing pattern disorders were described as lower jaw breathing.
These symptoms are widely observed in end-of-life patients
with cancer, dementia and internal organ failure.
However, regarding breathing pattern disorders, vari-
ous classications exist, such as Cheyne-Stokes breathing,
Biot’s breathing and lower jaw breathing. Because each
breathing disorder has different causes, caregivers should
be able to identify breathing disorders more accurately.
Caregivers need to acquire the medical knowledge about
breathing patterns and observation points.
Other symptoms falling in the categories of conscious-
ness decline, vital power decline and reduced oral intake
were frequently observed among end-of-life dementia el-
derly in formal daily care settings
16)
. It is widely known that
daily life actions, volition and swallowing abilities decline
as dementia progresses
16)
. Formal caregivers easily recog-
nized reduced oral intake symptoms as signs of impending
death.
Regarding feces disorder, a unique symptom, the Japa-
nese word kanibaba means meconium but is also known
by the general public as death-bed feces. Tarry feces and
bloody feces are symptoms that commonly appear in the
end-of-life phase
18)
, but they can also appear in curable cases
such as peptic ulcer. Therefore, they should not necessarily
be associated with impending death.
The calm and peaceful character category was consid-
ered to provide new insight into the research topic. Although
prior research indicates that BPSD such as hallucination,
anxiety, irritation and shouting loudly are commonly ob-
served among end-of-life dementia patients
16)
, there is hard-
ly any documented reference on mood improvement. It is
possible that this could be a distinct sign of impending death
in elderly people with senile dementia.
Regarding the preagonal vital power category, we think
that this is the same condition that Barbara Karnes describes
in her publication
19)
, referring to terminal cancer patients
who become temporarily active a few days before death.
Our results suggest the possibility that the same symptom
appears in cases of elderly people with senile dementia, but
there is a lack of sufcient corroborative research to conrm
this. Further research on this topic is therefore needed.
A number of caregivers mentioned odor as a forebod-
ing symptom of death, especially the smell of a dead body.
In dementia end-of-life cases, daily-life independence de-
clines gradually; incontinence usually happens as a result
of cognitive function decline, causing a foul smell in el-
derly people’s living environments. Patient odor in places
where the elderly are cared for has been well documented
in literature
1, 20)
. As far as we know, however, the particular
smell of a dead body-the odor that the dead body gives off-
has not been documented. It is possible that the workshop
participants assumed that the smell of a dead body meant
62
Table 1 Signs and symptoms at the end stage of dementia listed by
formal caregivers
Category
*
(number of cards)
List of words written on cards
**
(number of cards)
Breathing disorder (12) breathing hard (2)
deep breathing
increased state of apnea
increased yawning
keeping mouth open
longer state of apnea
lower jaw breathing (2)
panting (2)
shoulder breathing
Consciousness decline (9) decreased vital reaction
fall into delirium
loss of consciousness
not responding (2)
slower to react
slow to respond
twilight state (2)
Vital power decline (9) cannot talk
do not talk
increased sleep time
loss of desire
more frequently somnolent
no movement
no longer act violently
no conversation
sleep longer
Reduced oral intake (6) anorexia
cannot eat
cannot eat from mouth
cannot even take water
cannot take water (2)
Feces disorder (6) appearance of black feces
kanibaba
***
melena
tarry feces (3)
Calm and peaceful charac-
ter (5)
become gentle, obedient
change from obstinate to calm and ap-
preciative
express thanks
put affairs in order
suddenly become gentle
Blood pressure decline (5) blood pressure cannot be measured
blood pressure decline
decline in blood pressure
decline in blood pressure/impossible to
measure
hypotension
Category
*
(number of cards)
List of words written on cards
**
(number of cards)
Change in skin color (4) skin becomes deadly pale/earth-like
color
skin color changes
skin color drains/turns white
skin turns pale
Patient odor (4) exude dead body smell (2)
exude sour smell
terrible smell
Edema (4) begin to swell
edema of extremities
edema of instep
edema of underside of foot
Preagonal vital power (3) ash of vigor/last ame?
more talkative
sudden appetite improvement
Body temperature decline
(3)
body becomes cold/hypothermia
body temperature decline
hypothermia
Bedsore/wound deteriora-
tion (3)
bedsores get worse
presence of bedsores
wounds do not heal
Body weight reduction (2) weight loss (2)
Cyanosis (2) blood ow to the distal end of the body
blocked/becomes purple
legs become cold
Oliguria (2) difculty passing urine
oliguria
Solitary cards (14) ascites
bradycardia
chocolate-colored phlegm
constant fever
family members become kind
feel lonely and call their family
get along better with family
hallucinate
hematemesis
incontinence
lips round and thoroughly dried out
pneumonia
presence of water in lungs
pupillary dilation
* The head author and assistant researchers classied the cards and gave
all appropriate title to the groups. ** The responses of the formal caregiv-
ers are listed in alphabetical order. *** Kanibaba means death-bed feces in
the text.
Table 1 (continued)
63
that the body’s odor undergoes a transition just before death.
Understandably, both the smell of feces and the smell of a
dead body exert a negative inuence on comfortable bed-
side care surroundings. For a better outlook on this issue,
further research about odor in elderly care locations needs
to be carried out.
Body weight reduction is often observed among the el-
derly with end-of-life dementia
21)
. In fact, a previous study
emphasizes the deep factor connection between body weight
reduction and the prognosis of elderly patients
22)
. At the de-
mentia end-of-life stage, even if the appropriate amounts of
nutrition are given, body-weight reduction still occurs
23)
.
Our study has a number of important limitations. First,
because our study was retrospective in nature, we were un-
able to ascertain the places of employment of the partici-
pants, such as elderly nursing homes or home care settings,
which may have been unevenly balanced. Second, there is
a high possibility that the symptoms at end-of-life vary de-
pending on where the elderly people are taken care of, and
we therefore think that further investigation is warranted,
with a narrower focus on the attributes of the target group,
such as research targeting long-term care facility caregivers.
Third, in this research, a number of typical symptoms were
not mentioned, such as death rattle, that appear at the time
of death. For a more objective interpretation of the results,
we think it is necessary to do additional research with an
increased number of participants and number of issues.
Conclusions
This research had the original approach of clarifying the
signs and symptoms of death in end-of-life senile dementia
cases from the point of view of caregivers in rural areas.
The results show that caregivers in rural areas identied
breathing disorder as a top indicator of impending death.
The results also highlight some other characteristic signs of
impending death in end-of-life senile dementia cases, such
as preagonal vital power and calm and peaceful character.
This research could help develop formal caregivers’ obser-
vational skills in the end-of-life care settings.
Acknowledgment
This study was supported by a grant from the Sasagawa
Memorial Health Foundation. We declare that we have no
conicts of interest.
References
1. Hirakawa Y. Palliative care for the elderly: a Japanese per-
spective. In: Contemporary and Innovative Practice in Pallia-
tive Care. Chang E, Johnson A, Eds. InTech, 2012; Available
from: http://www.intechopen.com/articles/show/title/pallia-
tive-care-for-the-elderly-a-japanese-perspective.
2. Hirakawa Y, Masuda Y, Uemura K, et al. End-of-life care at
group homes for patients with dementia in Japan –ndings
from the analysis of policy-related differences. Arch Gerontol
Geriatr 2006; 42: 233–245. [Medline]
3. Matsumoto M, Inoue K, Kajii E, et al. Retention of physi-
cians in rural Japan: concerted efforts of the government, pre-
fectures, municipalities and medical schools. Rural Remote
Health 2010; 10: 1432 (online); Available from: http://www.
rrh.org.au. [Medline]
4. Hirakawa Y. End-of-life care at long-term care facilities for
the elderly in Japan. Hallym International Journal of Aging
(HIJA) 2009; 11: 1–12.
5. Hirakawa Y, Kuzuya M, Uemura K. Opinion survey of nurs-
ing or caring staff at long-term care facilities about end-of-
life care provision and staff education. Arch Gerontol Geriatr
2009; 49: 43–48. [Medline]
6. Lunney JR, Lynn J, Foley DJ, et al. Patterns of functional de-
cline at the end of life. JAMA 2003; 289: 2387–2392. [Med-
line]
7. Twycross RG. The terminal care of patients with lung cancer.
Postgrad Med J 1973; 49: 732–737. [Medline]
8. van den Beuken-van Everdingen MH, de Rijke JM, Kes-
sels AG, et al. Prevalence of pain in patients with cancer: a
systematic review of the past 40 years. Ann Oncol 2007; 18:
1437–1449. [Medline]
9. Hirakawa Y, Masuda Y, Kuzuya M, et al. Symptoms and care
of elderly dying at home with lung, gastric, colon, and liver
cancer. Jpn Med Assoc J 2006; 49: 140–145 (JMAJ).
10. Morita T, Tsunoda J, Inoue S, et al. Characteristics of pal-
liative care for elderly cancer patients –a pilot study at our
hospital. Geriat Med 1997; 35: 1505–1511 (in Japanese).
11. Morrison RS, Siu AL. Survival in end-stage dementia follow-
ing acute illness. JAMA 2000; 284: 47–52. [Medline]
12. Sampson EL, Ritchie CW, Lai R, et al. A systematic review
of the scientic evidence for the efcacy of a palliative care
approach in advanced dementia. Int Psychogeriatr 2005; 17:
31–40. [Medline]
13. Hirakawa Y, Masuda Y, Kuzuya M, et al. End-of-life expe-
rience of demented elderly patients at home: ndings from
DEATH project. Psychogeriatrics 2006; 6: 60–67.
14. Hirakawa Y, Masuda Y, Kimata T, et al. Terminal care for el-
derly patients with dementia in two long-term care hospitals.
Nihon Ronen Igakkai Zasshi 2004; 41: 99–104 (in Japanese,
Abstract in English). [Medline]
15. Scherder E, Herr K, Pickering G, et al. Pain in dementia. Pain
2009; 145: 276–278. [Medline]
16. Sampson EL. Palliative care for people with dementia. Br
Med Bull 2010; 96: 159–174. [Medline]
17. Sachs GA, Shega JW, Cox-Hayley D. Barriers to excellent
end-of-life care for patients. J Gen Intern Med 2004; 19:
1057–1063. [Medline]
18. Kuebler KK, Berry PH, Heidrich DE. End-of-life Care: Clini-
64
cal Practice Guidelines. Berry PH, Heidrich DE, Eds. W.B.
Saunders, Philadelphia, 2002.
19. Karnes B. Gone from My Sight –the Dying Experience. Bar-
bara Karnes Books, Vancouver, 2008.
20. Hirakawa Y, Suzuki S, Uemura K. Trial to use persimmon leaf
tea as air freshener at a long-term care facility. Hospice and
Home Care 2010; 18: 53–55 (in Japanese).
21. Henderson ML, Hanson LC, Reynolds KS. Improving Nurs-
ing Home Care of the Dying: a Training Manual for Nursing
Home Staff. Springer, New York, 2003.
22. Flacker JM, Kiely DK. A practical approach to identifying
mortality-related factors in established long-term care resi-
dents. J Am Geriatr Soc 1998; 46: 1012–1015. [Medline]
23. Palliative CDI. Enhancing Community Capacity Project. De-
mentia: Information for Carers, Families and Friends of Peo-
ple with Severe and End Stage Dementia. 2nd ed. University
of Western Sydney, Penrith South, 2007.
