Heart & Lung 50 (2021) 491 498

Contents lists available at ScienceDirect

Heart & Lung

journal homepage: www.heartandlung.com

Perceived stress and social support needs among primary family

caregivers of ICU Patients in Taiwan
Pei-Yu Changa,b,*, Tsai-Hsiu Changc, Jung-Min Yua
a
Taichung Tzu Chi Hospital, Buddhist Tzu Chi Medical Foundation, Taichung, Taiwan
b
School of Nursing, College of Medicine, National Taiwan University, Taipei, Taiwan
c
School of Nursing, Hungkuang University, Taichung, Taiwan

A R T I C L E I N F O A B S T R A C T

Article History: Background: Family caregivers of intensive care unit (ICU) patients may experience distress due to their care
Received 27 August 2020 recipients’ unexpected ICU hospitalization. Social support in coping with stress has been discussed from dif-
Revised 3 March 2021 ferent cultural perspectives, but social support does not seem to buffer stress for Chinese people.
Accepted 5 March 2021
Objectives: The purpose of this study was to explore stress perception and social support needs and their
Available online 6 April 2021
associations among Taiwanese primary family caregivers of patients admitted to the ICU for the first time.
Methods: This descriptive correlational study used the Perceived Stress Scale (PSS), the Impact of Events
Key words:
Scale-Revised (IES-R), the Norbeck Social Support Questionnaire (NSSQ), and structured interviews to
Family caregiver
Stress
explore stress perception and social support needs among Taiwanese primary family caregivers of patients
Posttraumatic stress disorder admitted to the ICU.
Social support Results: The Taiwanese primary family caregivers (N = 71) perceived considerable social support, but they still
Family needs experienced high stress, either daily life stress or ICU-related event stress. Most of them required extra sup-
Intensive care port, such as discussion of medical conditions, disease treatment information and psychological support,
even though they had social resources to assist them.
Conclusion: ICU health professionals should actively inquire about family caregivers’ actual needs, even when
family caregivers perceive considerable social support. Tailor-made interventions should be provided to
assist family members in coping with stress. Further research should also explore the role of social support
in stress and coping processes in Asian contexts due to cultural variance.
© 2021 Elsevier Inc. All rights reserved.

Introduction addition, culture can affect the coping strategies that individuals pre-
fer and cause individuals to use different forms of social support to
Having a family member hospitalized in the intensive care unit cope with stress.12,13 Previous studies have shown that family care-
(ICU) is a stressful event for family caregivers.1,2 The stress level could givers of ICU patients need various forms of social support from their
be even higher among patients unexpectedly admitted to the ICU and family, friends and health care providers. Family caregivers of
their family members.3 For family caregivers of ICU patients, the patients in the ICU needed financial support, religious or spiritual
stressors could be derived from unexpected hospitalization, the ICU support, and emotional support, as they must visit patients and form
environment, the uncertainty of the disease progress, medical deci- opinions to make medical decisions for them.14,15 However, health
sion-making, insufficient care information, and their own daily care providers tend to underestimate the actual needs of family care-
lives.1,4,5 The stressors perceived by individuals are influenced by givers, and neglected family caregivers cannot receive appropriate
their cultural context, which further affects their strategies for coping support during their family members’ hospitalization periods.10,16,17
with stress.6 8 Social support has been identified as an important Providing timely communication and informational and emo-
buffer and resource for stressed families.9 tional support could increase family caregivers’ comprehension of
Social support, which refers to the resources or assistance pro- patients’ medical conditions and treatment18 20; further, this support
vided by individuals or groups, has various types and functions, and may reduce their stress and depressive symptoms.21 To date, most
the sources of social support may vary across one’s lifespan.10,11 In research in the areas of stress and coping among family caregivers of
ICU patients has been conducted in Western cultural contexts. Due to
differences in culture, Westerners and Asians might expect different
* Corresponding author at: No. 88, Sec. 1, Fengxing Rd., Tanzi Dist., Taichung City types of social support. For example, Asians seek less social support
427019, Taiwan (R.O.C.) than Westerners because of their cultural norms of group harmony
E-mail address: [email protected] (P.-Y. Chang).

https://doi.org/10.1016/j.hrtlng.2021.03.001
0147-9563/© 2021 Elsevier Inc. All rights reserved.
492 P.-Y. Chang et al. / Heart & Lung 50 (2021) 491 498

and avoidance of public disgrace.22,23 Unsolicited support was shown The PSS was adopted to measure the primary family caregiver’s
to yield more effective outcomes in Asian culture, as Asians tend to global life stress level. The PSS includes 10 items rated on a Likert-
seek help only when experiencing extreme difficulties and feeling type scale from 0 (never) to 4 (very often). The internal consistency
distressed.22,24 However, knowledge of the association between per- reliability for the PSS were found to range from 0.83 0.86,27,30 and it
ceived stress and social support in Asians is limited; thus, this study was 0.80 in the current study. Higher total scores indicate a perceived
aimed to 1) comprehend the perceived stress and social support higher stress level, and the cutoff point used in this study was 12.9.30
needs and utilization among Taiwanese family members of ICU The IES-R 31 was used to measure the primary family caregiver’s
patients; 2) explore the association between perceived stress and perceived stress due to his or her care recipient’s ICU hospitalization.
social support; 3) understand what types of social support are The IES-R is a 22-item Likert-type scale that is composed of three sub-
demanded most; and 4) discuss the current needs and kinds of needs scales (intrusion, avoidance and hyperarousal). The five-point
that have been sought among Taiwanese primary family caregivers response scale ranges from 0 (not at all) to 4 (extremely).31,32 A total
of patients hospitalized in ICUs. score greater than 33 or a total score of each subscale greater than
1.5 indicates that the respondent is at risk of posttraumatic stress dis-
Method order (PTSD) symptoms.31,33 The internal consistency reliability for
the total scale and subscales were found to range from 0.79
Study design 0.91.25,31 In this study, the Cronbach’s alpha for the total scale was
0.94, and those for the hyperarousal, avoidance and intrusion sub-
This cross-sectional descriptive study, approved by the Institu- scales were 0.90, 0.75 and 0.85, respectively.
tional Review Board (IRB) of the study hospital, was conducted in The 9-item NSSQ includes three subscales (affect, affirmation and
two adult ICUs (a 28-bed medical ICU and a 10-bed surgical ICU) in a aid), with 6 items measuring the quantity of social support on a Lik-
teaching hospital in central Taiwan. These two ICUs allowed family ert-type scale from 0 (not at all) to 4 (a great deal) and another 3
members to make three visits per day, and each visit lasted only items measuring the length of acquaintance, contact frequency and
30 min. This is a common visiting policy in Taiwan. There were no loss of social support resources.34,35 Respondents are asked to list
clinical psychologists to provide emotional consultation to family members of their significant networks in their lives, up to 24 individ-
members in these ICU settings, but social workers provided minor uals, before they answer the questions.34,35 An average score equal to
assistance if the family members had financial crises due to their care or above 3, indicating quite a bit of social support, was utilized for
recipients’ ICU admission. The IRB protocol number was REC-105 1. data analysis to prevent network size variation and consequent mea-
All participants signed informed consent after the study design and surement error.36 The Cronbach’s alphas of the total scale in previous
process were explained to them in detail. The interviews were research26 and the current study were 0.98 and 0.99, respectively.
stopped at any time at the participants’ request. In addition to the above standardized questionnaires, two struc-
tured interview questions were used to increase understanding of
Samples the primary family caregivers’ social support. One question asked
what types of assistance and resources the primary family caregivers
A convenience sample of 71 primary family caregivers who met sought during their care recipients’ hospitalization periods. Another
the criteria and were approached during the ICU morning and after- asked what assistance they needed urgently and from what resources
noon visiting hours (10:30~11:00 and 14:00~14:30) participated in they expected to receive this assistance.
this study. The participants were greater than 20 years old and had
care recipients who had been admitted to the ICU for the first time
Data analysis
between 24 h and one week prior to being surveyed. The primary
family caregivers were either nominated by their care recipients or
SPSS Statistics 22.0 (IBM Corp, Armonk, NY, USA) was used to ana-
listed as emergency contact persons in the patients’ medical records.
lyze the data. The assumptions for each statistical test were verified
All participants were able to read or speak Mandarin or Taiwanese.
first. Descriptive statistics were used to analyze demographic data
The participants were excluded if they were paid caregivers or had a
and measurements. Inferential statistics, t-tests and one-way ANOVA
history of psychological disorder (i.e., depression, bipolar disorder or
were used to define the variance between groups. Bivariate associa-
schizophrenia).
tions among the different variables were assessed by using Pearson
or Spearman correlation analyses. The data for the structured inter-
Measurements
view questions were categorized and analyzed for overlapping words
and similar content to determine common themes.
Data were collected through (1) the review of patient medical
records; (2) the collection of primary family caregiver demographic
data, including severity of patient illness reported by family care- Results
givers; (3) the Perceived Stress Scale (PSS); (4) the Impact of Events
Scale-Revised (IES-R); (5) the Norbeck Social Support Questionnaire Characteristics of primary family caregivers and patients
(NSSQ); and (6) structured interviews. The Chinese versions of struc-
tured questionnaires with good validity and reliability were adopted Eighty-five primary family caregivers of ICU patients were
from the literature25 27; in addition, two structured interview ques- approached, but three did not meet the enrollment criteria, and 11
tions were asked to the participants when their care recipients were declined to participate (Fig. 1). A total of 71 primary family caregivers
still in the ICU and after the researcher had explained the study pro- with a mean age of 49.3 § 10.9 years were enrolled. Most of the pri-
cess face to face. Patient gender, age, primary diagnosis, lengths of mary family caregivers were female (n = 44, 62.0%), were the patients’
ICU stay and hospital stay, medical history and the Acute Physiology children (n = 48, 67.6%), were medical surrogates for the patients
and Chronic Health Evaluation II (APACHE II) score were collected (67.6%), and lived with their care recipients (62.0%). Among the male
from the patients’ medical records. The APACHE II score is based on primary family caregivers (n = 27; 38%), 85.2% of them were medical
patient physiological parameters and is commonly used to predict surrogates. On the other hand, the majority of the care recipients had
the risk of death for ICU patients in Taiwan. A score greater than 20 a history of chronic medical conditions and took medication (n = 62,
has been associated with mortality rates of >29% and >40% among 87.2%), and most had hypertension (n = 40, 56.3%). The average
nonsurgical and postoperative patients, respectively.28,29 length of the ICU stays of the care recipients was 3.8 (SD=1.7) days.
 P.-Y. Chang et al. / Heart & Lung 50 (2021) 491 498 493

Fig. 1. Flow chart of study enrollment.

The characteristics of the primary family caregivers and their care decision was not associated with a higher level of social support
recipients are summarized in Table 1. (p>.05).
The primary family caregivers who perceived higher life stress
Stress and social support among primary family caregivers also experienced higher ICU event stress as well as higher intrusion,
avoidance and hyperarousal symptoms (p < 0.05) (Table 3). The pri-
The primary family caregivers perceived high global life stress mary family caregivers perceived similar stress levels from daily life
(M = 19.9§6.0), measured by the PSS, and more than 90% of them and ICU event stress regardless of their gender, occupation, level of
scored above the cutoff point of 12.9. They also reported high ICU- social support, perceived patient severity of illness or any type of
related stress (M = 33.61§15.56), measured by the IES-R, indicating a social support (p > 0.05) (Table 3). In addition, the primary family
risk of PTSD. On average, the primary family caregivers had highly caregivers who had more chronic diseases perceived more ICU event
intrusive thoughts (M = 1.77§0.90) but not avoidance or hyper- stress (rs = 0.296, p < 0.05), but the number of chronic diseases of the
arousal; however, more than 40% of them still scored above the cutoff primary family caregivers was not significantly associated with the
point in avoidance experiences and hyperarousal responses (Table 2). stress level from daily life (p > 0.05). Neither the level nor type of
Moreover, ICU-related stress explained approximately 43% of global social support was significantly associated with any dimension of ICU
life stress (Table 3), indicating that half of the primary family care- event-related stress (p > 0.05); however, the level of social support
givers’ stress was from daily life. was positively associated with perceived patient severity of illness
Most of the primary family caregivers had known the individuals (rs = 0.296, p < 0.05) (Table 3). The younger primary family caregivers
who acted as social support resources for more than five years experienced higher global life stress (rs = 0.238, p < 0.05), and
(94.9%) and had been contacting them on a daily basis (50.2%). The females (M = 3.38; SD=0.51) perceived statistically significantly more
caregivers perceived quite a bit of support (M = 3.21; SD=0.62) from social support than male primary family caregivers (M = 2.95;
their support resources in all three dimensions: affect (M = 3.21; SD = 0.70) (t[69] = 2.76, p < 0.05).
SD=0.62), affirmation (M = 3.10; SD=0.70) and aid (M = 3.30;
SD = 0.66) (Table 2). In general, females perceived more social sup- Family caregivers’ current needs social support resources
port than males in terms of affect, affirmation and aid (p < 0.05).
Nearly half (n = 29, 40.8%) of the participants expressed that their Nearly half of the primary family caregivers (n = 29; 40.8%) had
loved one in the ICU was one of their important social support actively sought assistance. The majority of them (79.3%) reported
resources. On average, the primary family caregivers had 6 7 social that they requested help in the following areas: discussion of medical
support resources, and these resources included family (58.6%), condition, disease treatment information (51.7%) and psychological
friends (13.7%) and spouses (7.6%); only a few primary family care- support (24.1%) (Table 4). Almost all primary family caregivers who
givers (n = 7, 9.9%) had experienced the loss of social support resour- had sought assistance (n = 28) still had current needs. Half of them
ces during the past year. The average number of persons on whom required updated information on patient medical conditions and
the primary family caregivers could rely to discuss the care recipi- treatment plans (54.6%), and others needed optimal care for their
ent’s medical condition was approximately 3 4 persons; however, care recipients to be provided (43.6%) or needed psychological sup-
having more persons with whom to discuss the patient’s medical port for themselves (18.2%). The details of social support resources
494 P.-Y. Chang et al. / Heart & Lung 50 (2021) 491 498

Table 1 Table 2
Characteristics of primary family caregivers and patients (N = 71). Perceived stress and social support (N = 71).

Variables (range) Mean (SD) Frequency (%) Variables (cut-off) Mean § SD Above cut-off (%)

Primary family caregivers PSS (>12.9) 19.90 § 6.0 91.5

Age (years) 49.3(10.9) IES-R (33) 33.61 § 15.56 49.3
Gender Intrusion (1.5) 1.77 § 0.90 60.6
Female 44 (62.0) Avoidance (1.5) 1.40 § 0.69 40.8
Relationship to patient Hyperarousal (1.5) 1.44 § 0.92 42.3
Child 48(67.6) NSSQ (3) 3.21 § 0.62 69.0
Daughter 22 (31.0) Affect (3) 3.10 § 0.70 66.2
Son 20 (28.2) Affirmation (3) 3.30 § 0.66 76.0
Daughter-in-law 6 (8.4) Aid (3) 3.25 § 0.69 71.8
Spouse 15 (21.1)
PSS: Perceived Stress Scale; IES-R: Impact of Events Scale-
Wife 9 (12.7)
Revised; NSSQ: Norbeck Social Support Questionnaire.
Husband 6 (8.4)
Parent 4 (5.6)
Mother 4 (5.6)
Father 0 assistance but mentioned their current needs when interviewed. In
Sibling 1 (1.4) addition, they perceived similar global life stress (t[69] = 1.835,
Other 3 (4.2) p > 0.05) and ICU event stress (t[69] = 1.853, p > 0.05) as those
Perceived patient’s severity of illness (0 10) 7.92(2.21)
Live with patient 44 (62.0)
(n = 28; 39.4%) who had sought assistance and still reported current
History of chronic disease 29(40.8) needs when interviewed. The primary family caregivers (n = 29),
One chronic disease 24(33.8) who had proactively looked for assistance before the interviews
Above two chronic disease 5(7) experienced either higher life stress (t[69] = 3.385, p < 0.05) or ICU
Patients
event stress (t[69] = 2.880, p < 0.05) than those who did not seek
Age 68.4(18.5)
Gender assistance but perceived a similar level of social support (t
Female 35 (49.3) [69] = 0.328, p > 0.05).
APACHE II on admission 14.3 (6.7)
APACHE II on survey date 12.0 (5.9)
Primary diagnosis Discussion
Cardiovascular 14 (19.7)
Neurological 14 (19.7) The primary family caregivers in this study reported high global
Respiratory 12 (16.9) life stress, similar to the findings in the general population37 41;
Gastrointestinal 10 (14.1)
Sepsis 9 (12.7)
however, in addition to global stress, primary family caregivers also
Major trauma 7 (9.9) experienced stress related to the ICU hospitalization event.1,42 Over
Others 5 (7.0) 40% of primary family caregivers in the current study reported intru-
Days in ICU stay 3.8 (1.7) sive thoughts, avoidance experiences and hyperarousal responses,
APACHE II: Acute Physiology and Chronic Health Evaluation II. with intrusive thoughts being the most common among primary
family caregivers, consistent with the findings from a previous
study.43 Stress-related symptoms might increase over time and
that the primary family caregivers sought and expected are listed in impact cognitive function.1,44 Therefore, further research should
Tables 4 and 5; there were no significant gender differences regard- focus on the management of stress-related symptoms to improve the
ing the types of assistance that the primary family caregivers sought well-being of primary family caregivers.
and their current needs (p>.05). On the other hand, the primary fam- In the current study, younger age was significantly associated
ily caregivers (n = 55; 77.5%) who expressed their current social sup- with more global life stress, which may be due to the multiple roles,
port needs during the care recipient’s ICU hospitalization period lower maturity level, and lack of coping skills and resources among
(Table 5) perceived similar global life stress (t[69] = 1.852, p > 0.05) younger individuals.45 The severity of ICU event stress was negatively
and ICU event stress (t[69] = 1.167, p > 0.05) as those who did not associated with age, although not significantly; this association might
express current needs. Some primary family caregivers (n = 27; 38%), indicate that young primary family caregivers, particularly primary
half of whom were male (n = 14) and most of whom were above the family caregivers whose main social support resources are their care
mean age of the primary family caregivers (n = 18), never sought any recipients, are more vulnerable to PTSD. In addition, nearly half of the

Table 3
Correlation between variables and PSS, IES-R and NSSQ (N = 71).

1 2 3 4 5 6 7

1. PFC age
2. Patient’s ages .337**
3. Perceived patient’s severity of illnesss 0.018 0.048
4. APACHE II on admission 0.115 .292** .169
5. Persons to discuss medical decision with 0.145 .199 0.098 .281*
6. PSS 0.238** 0.345** 0.024 .016 .091
7. IES-R 0.142 0.164 .075 .139 .286* .656**
8. NSSQs .099 .151 .296** .151 0.088 0.081 0.099
S
: Spearman correlation.
* p < 0.05.
** p < 0.01; *** p < 0.001; APACHE II: Acute Physiology and Chronic Health Evaluation II
PSS: Perceived Stress Scale; IES-R: Impact of Events Scale-Revised; NSSQ: Norbeck Social Support Questionnaire.
 P.-Y. Chang et al. / Heart & Lung 50 (2021) 491 498 495

Table 4 many concerns regarding patient safety, visiting hours remain very
Assistance and social support resources sought prior the interview (n = 29). limited in Taiwan and certain countries, which warrants further
Assistance sought (%) Social support (%) exploration of the pros and cons of flexible visiting hours. In addition,
resources systematic family visit guidelines can increase ICU nurse staff skills in
communication for family education, which can reduce ICU nurse
1 Discussion of medical condition 79.3 Family 34.6
2 Disease treatment information 51.7 Healthcare 33.7 staff stress in confronting families of ICU patients and lessen the anxi-
professionals ety of families of ICU patients.53 Family-centered rounds and family
3 Psychological support 24.1 Friends 20.0 meetings might be another option to communicate with family care-
4 Patient visiting assistance 10.3 Friends with medical 10.9
givers if the visiting policy poses challenges.49,54 Family members
background
5 Financial support 6.9 Religions 10.9
can be invited to participate in rounds or the interdisciplinary team,
6 Alternative therapy consultation 3.5 Websites 9.1 and they can also be encouraged to ask questions and comment on
7 Paid-caregiver support 3.5 Colleagues or 7.3 the patient’s treatment and care plan. Family participation in rounds
classmates and family meetings could increase family satisfaction with decision
8 Others 3.6
making and communication with health professionals.49,54 Thus, fre-
This table is presented as the percentage of the PFCs who addressed the assistance and quent communication could provide more opportunities to realize
resources for each item.
and meet the family’s needs.
Although the current study did not find a statistically significant
Table 5 association between gender and perceived stress levels, attention
Current Needs and Social support resources expected (n = 55). needs to be paid to gender differences in stress perception and sup-
Current need (%) Social support (%) port seeking, as Chinese males have more role obligations than
resources expected females in providing family support and care.23 Gender differences in
perceived social support can be revealed by socialization experiences
1 Timely updated medical condition 54.6 Doctors 63.6
& treatment plan and social roles,55 and females were found to perceive more social
2 Optimal care granted 43.6 Nurses 50.9 support than males.56 This study found that nearly 90% of the male
3 Psychological support 18.2 Hospital facilities 10.9 primary family caregivers were medical surrogates and that they per-
4 Flexible visiting hours 9.1 No ideas 9.1
ceived less social support than females; furthermore, they reported
5 Others 5.5 Family 7.3
6 Safety of hospital accommodation 3.6 Friends 5.5
that they needed more assistance. Chinese males seldom express
7 Chores assistance 3.6 Religions 36 their actual feelings and tend to hide their vulnerability due to self-
8 Patient visiting assistance 1.8 Volunteers 3.6 esteem, social reputation and Chinese social expectations.23,57 For
9 Financial support 1.8 Others 1.8 this reason, healthcare professionals may consider the specific cul-
This table is presented as the percentage of the PFCs who addressed the current needs tural expectations or acculturation of Chinese male primary family
and resources for each item. caregivers, who believe they should fulfill male gender roles rather
than express and acknowledge what they need to respond to a criti-
cal hospitalization event; such an approach would allow the provi-
primary family caregivers mentioned that their care recipients in the sion of support and facilitation of changes in men’s responses to
ICU were important social resources for them, which indicated that better assist them in coping with a stressful event.
these primary family caregivers lost their most important support Coping is crucial for primary family caregivers to prevent PTSD
resources during this period. The current study showed that the per- symptoms.42,58 Social support can be a strategy to cope with stress;
ceived severity of patients’ illness was positively associated with per- however, the primary family caregivers in the current study per-
ceived social support, indicating that Taiwanese primary family ceived high stress even though they had received various types of
caregivers tended to seek help in accordance with their own percep- considerable social support, which is consistent with previous find-
tion of the severity of patients’ illness rather than the objective ings in Japan.3 The Taiwanese primary family caregivers in this study
APACHE II score. Therefore, the perceived severity of patients’ illness perceived a reasonable level of social support from people with
might be a better way to understand primary family caregivers’ whom they had been acquainted for over five years and had daily
intention to seek help or need for help. In addition, healthcare pro- contact. Despite this, they still felt distressed, which indicates that
viders should collect self-reported data instead of relying on APACHE the support might not have been effective. Cultural variance might
II scores to estimate primary family caregivers’ needs. explain this unique phenomenon. Asians focus more on group inter-
The study participants expressed that they most needed health- action and harmony, and they prioritize communal relationships,
care providers to update them on their care recipients’ medical prog- interdependence and group outcomes more than Westerners,22 indi-
ress and care plans and provide the best care for their care recipients, cating that Asians tend to be more collectivist than individualist and
which indicates that effective communication between primary fam- act in a communally acceptable way in certain group situations.59,60
ily caregivers and healthcare professionals and family education are In addition, Asians tend to devote all resources (emotional support,
both lacking.46,47 Rather than providing additional educational hand- information support and tangible assistance) to others to express
outs to increase primary family caregivers’ knowledge,48,49 frequent their kindness, sympathy, and obligation in family, community or a
family presence and sufficient communication time might also help group.23 As a result, sometimes support providers might lack bound-
health professionals to provide sufficient information and clarify aries and not respect privacy, which could cause the recipient to
medical ambiguity; further, care quality from health professionals receive unnecessary aid, negative emotions and criticisms from their
might be discerned by primary family caregivers.46,47 Thus, flexible social support resources, further increasing the secondary bur-
ICU visiting hours might be important for primary family caregivers den.61 63 This secondary burden for Asians could result from the reci-
since it has been found to increase primary family caregivers’ satis- procity principle, with support recipients feeling that they owe favors
faction with health professionals for patient care and reduce their to support providers and that they should return the favors.23 In
anxiety and depressive symptoms.50 52 Some literature has shown addition, family conflicts commonly occur when family members of
that ICU nurse staff might experience burnout and that their work- ICU patients have different opinions regarding the patient’s treat-
load may increase with flexible ICU visiting hours 5;50 however, flexi- ment plan, which could generate additional stress.63 This reasoning
ble visiting hours were also found to build a better interaction applies to the Taiwanese primary family caregivers in the current
between ICU staff and primary family caregivers.51 Because of the study, who proactively sought assistance and still experienced higher
496 P.-Y. Chang et al. / Heart & Lung 50 (2021) 491 498

global life stress and ICU event stress. Social support has not been Implications for practice
found to buffer stress in most collectivist cultures.64,65 Thus, health
professionals should identify the specific social support needs of Healthcare professionals in intensive care units should consider
Asian primary family caregivers. Because social support might family-centered care as part of patient care. The family caregivers of
become a stressor in certain circumstances, further research should ICU patients in this study experienced high perceived stress and
explore the social support that Taiwanese primary family caregivers needed updated information on their care recipients’ medical condi-
receive and what they expect to receive and explore the associations tions and treatment and better care for their care recipients. Social
among stress, social support and coping strategies among Asian pri- support did not lessen stress among these Taiwanese primary family
mary family caregivers. caregivers of ICU patients. Therefore, healthcare professionals in the
Culture shapes people’s values, attitudes and beliefs and even ICU should proactively ask primary family caregivers about their
their appraisals of stress, stress reactions and coping strategies.12 Chi- needs for support, and cultural differences should be taken into
nese people tend to passively express their actual feelings and needs account when family-centered care is provided.
because silence is considered a virtue, and they tend to resolve their
problems alone until they are truly distressed because the collectivist
Conclusion
culture leads Chinese people to think they have an obligation to not
burden their social resources and to talk less about their stress to
The family caregivers in this study were distressed not only by the
maintain group interaction and harmony.22 24, 66,67 In addition, Chi-
ICU hospitalization event but also by their daily lives. To provide fam-
nese people prefer not to lose face (i.e. not to lose respect, dignity, or
ily-centered care, healthcare providers should also assess primary
prestige)23,60; therefore, they tend to accept a situation and responsi-
family caregivers’ perceived stress in their daily lives. Social support
bility rather than confront other people.23,66 Similarly, this study
can be a strategy to cope with stress; however, although Taiwanese
showed that nearly 40% of the primary family caregivers never
primary family caregivers perceived various types of social support,
sought any assistance but acknowledged their current needs for
this social support did not lessen their stress because culture can
social support when asked. The results also indicated that the Taiwa-
affect the stress and coping process or the social support system. On
nese primary family caregivers might have been exhausted and dis-
the other hand, the primary family caregivers’ perception of the
tressed. Effective communication between primary family caregivers
severity of the patient’s illness could affect their perception of seek-
and health professionals and good comprehension of primary family
ing social support. Thus, healthcare providers should pay attention to
caregivers’ actual needs to support the provision of appropriate infor-
primary family caregivers’ subjective perception of disease severity,
mation can reduce primary family caregivers’ stress.68 71 Therefore,
not only to objective assessments.
in addition to exploring flexible visiting hours, further research
Taiwanese primary family caregivers needed health professionals
should develop interventions to encourage Asian primary family
to update them on their care recipients’ medical conditions in a
caregivers to express their actual needs. For example, ICU diaries
timely manner, to provide better care for their care recipients’ and
have been used to promote multidimensional interaction and com-
support them psychologically, and they also required flexible visiting
munication to meet primary family caregivers’ actual needs and pro-
hours. In addition, Asians avoid bothering others and seldom utilize
vide adequate support and emotional understandings.72,73 Moreover,
social support22; therefore, primary family caregivers might not
healthcare professionals can assist primary family caregivers in cop-
directly express their actual feelings and request help. Thus, health
ing with stress by better comprehending primary family caregivers’
professionals should actively inquire about primary family caregivers’
needs and developing tailor-made interventions to help them man-
needs, and further studies in Asian contexts should consider cultural
age stress-related symptoms and prevent PTSD.
specificity to develop interventions to assist primary family care-
givers.

Limitations
Acknowledgments
Apart from the few Asian studies that support the results of this
study and small sample size in this study, there were some limita- The authors thank the family members of the ICU patients who
tions of this study. Due to the nature of the cross-sectional design, participated in this study and provided us with important informa-
the data were collected only at a certain point in time; therefore, the tion. We also would like to thank the ICU nurses who assisted with
ability to understand stress and the seeking of social support from a the study recruitment process. Additionally, the first author would
transactional perspective was limited. The participants were like to express her gratitude to Dr. Hsiang-Ping Wang for helping
recruited only during the morning and afternoon visiting hours, with this study and Dr. Shih-Yu Lee for preparing this manuscript.
which might have introduced some study bias. In addition, some pri-
mary family caregivers of ICU patients refused to participate in this Declaration of Competing Interest
study because they felt too stressed to express their feelings, which
might have led to the underestimation of caregivers’ stress in the None.
results. The findings from this study have limited generalizability
because the investigation was conducted in a developing regional
hospital and with relatively limited resources. All the patients in this References
study had been admitted to the ICU for the first time; however, data
1. Chang P-Y, et al. Stress, stress-related symptoms and social support among Taiwa-
on whether the primary family caregivers had any previous ICU expe- nese primary family caregivers in intensive care units. Intens Crit Care Nurs.
riences were not collected. Therefore, this confounding variable 2018;49:37–43. https://doi.org/10.1016/j.iccn.2018.05.002.
might have threatened both internal and external validity; further 2. De Beer J, Brysiewicz P. The experiences of family members during critical illness
of a loved one admitted to an intensive care unit. Afr J Nurs Midwifery. 2017;19
research should collect information on primary family caregivers’
(1):56–68. https://doi.org/10.25159/2520-5293/1446.
previous ICU experiences and examine their impact on the current 3. Komachi M, Kamibeppu K. Acute stress symptoms in families of patients admitted
ICU hospitalization event. Some participants found the NSSQ to be to the intensive care unit during the first 24 hours following admission in Japan.
time consuming and laborious, which might have threatened the Open J Nurs. 2015;5(04):325–335. https://doi.org/10.4236/ojn.2015.54035.
4. Iverson E, et al. Factors affecting stress experienced by surrogate decision makers
internal validity of the data on social support; therefore, further for critically ill patients: implications for nursing practice. Intens Crit Care Nurs.
researchers might consider another, more efficient measure. 2014;30(2):77–85. https://doi.org/10.1016/j.iccn.2013.08.008.
 P.-Y. Chang et al. / Heart & Lung 50 (2021) 491 498 497

5. Matt B, et al. Relatives' perception of stressors and psychological outco- 34. Norbeck JS, Lindsey AM, Carrieri VL. The development of an instrument to measure
mes Results from a survey study. J Crit Care. 2017;39:172–177. https://doi.org/ social support. Nurs Res. 1981;30(5):264–269. https://doi.org/10.1097/00006199-
10.1016/j.jcrc.2017.02.036. 198109000-00003.
6. Lazarus RS, Folkman S. Stress, appraisal, and coping. New York: Springer Publishing 35. Norbeck JS, Lindsey AM, Carrieri VL. Further development of the Norbeck Social
Company; 1984. Support Questionnaire: normative data and validity testing. Nurs Res. 1983;32
7. English AS, Zeng ZJ, Ma JH. The stress of studying in China: primary and secondary (1):4–9. https://doi.org/10.1097/00006199-198301000-00002.
coping interaction effects. Springerplus. 2015;4(1):755–768. https://doi.org/ 36. Gigliotti E, Samuels WE. Use of averaged norbeck social support questionnaire
10.1186/s40064-015-1540-3. scores. Int Scholarly Res Netw Nurs. 2011;232(3):957–973. https://doi.org/10.5402/
8. Kuo BC. Collectivism and coping: current theories, evidence, and measurements of 2011/567280.
collective coping. Int J Psychol. 2013;48(3):374–388. https://doi.org/10.1080/ 37. Falcon LM, Todorova I, Tucker K. Social support, life events, and psychological dis-
00207594.2011.640681. tress among the Puerto Rican population in the Boston area of the United States.
9. Turner-Cobb J, et al. The acute psychobiological impact of the intensive care expe- Aging Mental Health. 2009;13(6):863–873. https://doi.org/10.1080/
rience on relatives. Psychol Health Med. 2016;21(1):20–26. https://doi.org/ 13607860903046552.
10.1080/13548506.2014.997763. 38. Hynynen E, et al. The incidence of stress symptoms and heart rate variability dur-
10. Blom H, Gustavsson C, Sundler AJ. Participation and support in intensive care as ing sleep and orthostatic test. Eur J Appl Physiol. 2011;111(5):733–741. https://doi.
experienced by close relatives of patients—A phenomenological study. Intens Crit org/10.1007/s00421-010-1698-x.
Care Nurs. 2013;29(1):1–8. https://doi.org/10.1016/j.iccn.2012.04.002. 39. Kell G, Rao A, Katsikitis M. A randomised placebo controlled clinical trial on the
11. Pedro ICdS, Rocha SMM, Nascimento LC. Social support and social network in fam- efficacy of Caralluma fimbriata supplement for reducing anxiety and stress in
ily nursing: reviewing concepts. Rev Lat Am Enfermagem. 2008;16(2):324–327. healthy adults over eight weeks. J Affect Disord. 2019;246:619–626. https://doi.
https://doi.org/10.1590/S0104-11692008000200024. org/10.1016/j.jad.2018.12.062.
12. Aldwin C M, Culture, coping and resilience to stress. Paper presented at the Pro- 40. Taylor DJ, et al. The role of sleep in predicting college academic performance: is it a
ceedings of the First International Conference on Operationalization of Gross unique predictor? Behav Sleep Med. 2013;11(3):159–172. https://doi.org/10.1080/
National Happiness, Thimphu. 2004. Retrieved from http://www.bhutanstudies. 15402002.2011.602776.
org.bt/gross-national-happiness-and-development/. 41. Yeung NC, Lu Q. Perceived stress as a mediator between social support and post-
13. Kuo BC, et al. The mediating role of cultural coping behaviours on the relationships traumatic growth among Chinese American breast cancer survivors. Cancer Nurs.
between academic stress and positive psychosocial well-being outcomes. Int J Psy- 2018;41(1):53–61. https://doi.org/10.1097/NCC.0000000000000422.
chol. 2018;53:27–36. https://doi.org/10.1002/ijop.12421. 42. Petrinec AB, et al. Coping strategies and posttraumatic stress symptoms in post-
14. Casarini KA, Gorayeb R, Basile Filho A. Coping by relatives of critical care patients. ICU family decision makers. Crit Care Med. 2015;43(6):1205–1212. https://doi.org/
Heart Lung. 2009;38(3):217–227. https://doi.org/10.1016/j.hrtlng.2008.05.003. 10.1097/CCM.0000000000000934.
15. Schenker Y, et al. It hurts to know. . . and it helps”: exploring how surrogates in the 43. Trevick SA, Lord AS. Post-traumatic stress disorder and complicated grief are com-
ICU cope with prognostic information. J Palliat Med. 2013;16(3):243–249. https:// mon in caregivers of neuro-ICU patients. Neurocrit Care. 2017;26(3):436–443.
doi.org/10.1089/jpm.2012.0331. https://doi.org/10.1007/s12028-016-0372-5.
16. Hutchison PJ, et al. Dimensions and role-specific mediators of surrogate trust in the 44. Petersen H, et al. Stress vulnerability and the effects of moderate daily stress on
ICU. Crit Care Med. 2016;44(12):2208–2214. https://doi.org/10.1097/ sleep polysomnography and subjective sleepiness. J Sleep Res. 2013;22(1):50–57.
CCM.0000000000001957. https://doi.org/10.1111/j.1365-2869.2012.01034.x.
17. Khatri Chhetri I, Thulung B. Perception of nurses on needs of family members of 45. Chen Y, et al. Age differences in stress and coping: problem-focused strategies
patient admitted to critical care units of teaching hospital, Chitwan Nepal: a Cross- mediate the relationship between age and positive affect. Int J Aging aHum Dev.
Sectional Institutional Based Study. Nurs Res Pract. 2018;2018. https://doi.org/ 2018;86(4):347–363. https://doi.org/10.1177/0091415017720890.
10.1155/2018/1369164. 46. Au SS, et al. A multicenter observational study of family participation in ICU rounds. Crit
18. Hinkle LJ, Bosslet GT, Torke AM. Factors associated with family satisfaction with Care Med. 2018;46(8):1255–1262. https://doi.org/10.1097/CCM.0000000000003193.
End-of-life care in the ICU. Chest. 2015;147(1):82–93. https://doi.org/10.1378/ 47. Jacob M, et al. Needs of patients’ family members in an intensive care unit with
chest.14-1098. continuous visitation. Am J Crit Care. 2016;25(2):118–125. https://doi.org/10.4037/
19. Hwang DY, et al. Assessment of satisfaction with care among family members of survi- ajcc2016258.
vors in a neuroscience intensive care unit. The Journal of neuroscience nursing. J Am 48. Gaeeni M, et al. Informational support to family members of intensive care unit
Assoc Neurosci Nurs. 2014;46(2):106–116. https://doi.org/10.1378/chest.14-1098. patients: the perspectives of families and nurses. Glob J Health Sci. 2015;7(2):8–19.
20. Moore CD, et al. The effect of a family support intervention on physician, nurse, https://doi.org/10.5539/gjhs.v7n2p8.
and family perceptions of care in the surgical, neurological, and medical intensive 49. Kynoch K, et al. The effectiveness of interventions to meet family needs of critically
care units. Crit Care Nurs Q. 2012;35(4):378–387. https://doi.org/10.1097/ ill patients in an adult intensive care unit: a systematic review update. JBI Evid
CNQ.0b013e318268fde3. Synth. 2016;14(3):181–234. https://doi.org/10.11124/JBISRIR-2016-2477.
21. Carlson EB, et al. Care and caring in the intensive care unit: family members' dis- 50. Junior APN, et al. Flexible versus restrictive visiting policies in ICUs: a systematic
tress and perceptions about staff skills, communication, and emotional support. J review and meta-analysis. Crit Care Med. 2018;46(7):1175–1180. https://doi.org/
Crit Care. 2015;30(3):557–561. https://doi.org/10.1016/j.jcrc.2015.01.012. 10.1097/CCM.0000000000003155.
22. Kim HS, Sherman DK, Taylor SE. Culture and social support. Am Psychol. 2008;63 51. Khaleghparast S, et al. New visiting policy: a step toward nursing ethics. Nurs
(6):518–526. https://doi.org/10.1037/0003-066X. Ethics. 2019;26(1):293–306. https://doi.org/10.1177/0969733017703701.
23. Hwang K-K. Foundations of Chinese psychology. New York: Springer Publishing 52. Mitchell ML, Aitken LM. Flexible visiting positively impacted on patients, families
Company; 2012. and staff in an Australian Intensive Care Unit: a before-after mixed method study.
24. Mojaverian T, Kim HS. Interpreting a helping hand: cultural variation in the effec- Aust Crit Care. 2017;30(2):91–97. https://doi.org/10.1016/j.aucc.2016.01.001.
tiveness of solicited and unsolicited social support. Personality and Social Psychol- 53. Kozub E, et al. Improving Nurse Satisfaction With Open Visitation in an Adult
ogy Bulletin. 2013;39(1):88–99. https://doi.org/10.1177/0146167212465319. Intensive Care Unit. Crit Care Nurs Q. 2017;40(2):144–154. https://doi.org/
25. Wu KK, Chan K. The development of the Chinese version of Impact of Event Sca- 10.1097/CNQ.0000000000000151.
le Revised (CIES-R). Soc Psychiatry Psychiatr Epidemiol. 2003;38(2):94–98. https:// 54. Ludmir J, Netzer G. Family-centered care in the intensive care unit—what does best
doi.org/10.1007/s00127-003-0611-x. practice tell us? In seminars in respiratory and Critical Care Medicines. Thieme Medi-
26. Chan CW, et al. Social support and coping in Chinese patients undergoing cancer cal Publishers; 2019. https://doi.org/10.1055/s-0039-1697957.
surgery. Cancer Nurs. 2004;27(3):230–236. https://doi.org/10.1097/00002820- 55. Matud MaP, et al. Structural gender differences in perceived social support. Pers Indi-
200405000-00009. vid Dif. 2003;35(8):1919–1929. https://doi.org/10.1016/S0191-8869(03)00041-2.
27. Leung DY, Lam T-h, Chan SS. Three versions of perceived stress scale: validation in 56. Burkert S, et al. Gender differences in social support in persons with epilepsy. Epi-
a sample of Chinese cardiac patients who smoke. BMC Public Health. 2010;10 lepsy Behav. 2015;46:205–208. https://doi.org/10.1016/j.yebeh.2015.02.041.
(1):513. https://doi.org/10.1186/1471-2458-10-513. 57. Chui WYY, Chan SWC. Stress and coping of Hong Kong Chinese family members
28. Knaus WA, et al. APACHE II: a severity of disease classification system. Crit Care during a critical illness. J Clin Nurs. 2007;16(2):372–381. https://doi.org/10.1111/
Med. 1985;13(10):818–829. j.1365-2702.2005.01461.x.
29. Niewin  ski G, Starczewska MH, Kan  ski A. Prognostic scoring systems for mortality 58. Petrinec AB, Martin BR. Post-intensive care syndrome symptoms and health-
in intensive care units—The APACHE model. Anaesthesiol Intensive Ther. 2014;46 related quality of life in family decision-makers of critically ill patients. Palliat Sup-
(1):46–49. https://doi.org/10.5603/AIT.2014.0010. port Care. 2018;16(6):719–724. https://doi.org/10.1017/S1478951517001043.
30. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health 59. Hwang A, Francesco AM, Kessler E. The relationship between individualism-collec-
Soc Behav. 1983;24(4):385–396. https://doi.org/10.2307/2136404. tivism, face, and feedback and learning processes in Hong Kong, Singapore, and
31. Weiss and Marmar. In: Wilson JP, Keane TM, eds. Assessing Psychological Trauma the United States. J Cross Cult Psychol. 2003;34(1):72–91. https://doi.org/10.1177/
and PTSD: A Practitioner’s Handbook. New York: Guilford Press; 1997. 0022022102239156.
32. Horowitz M, Wilner N, Alvarez W. Impact of Event Scale: a measure of subjective 60. Chang J. The interplay between collectivism and social support processes among
stress. Psychosom Med. 1979;41(3):209–218. https://doi.org/10.1097/00006842- Asian and Latino American college students. Asian Am J Psychol. 2015;6(1):4–11.
197905000-00004. https://doi.org/10.1037/a0035820.
33. Creamer M, Bell R, Failla S. Psychometric properties of the impact of event scale— 61. Morling B, Uchida Y, Frentrup S. Social support in two cultures: everyday transac-
Revised. Behav Res Ther. 2003;41(12):1489–1496. https://doi.org/10.1016/j. tions in the US and empathic assurance in Japan. PLoS One. 2015;10:(6) e0127737.
brat.2003.07.010. https://doi.org/10.1371/journal.pone.0127737.
498 P.-Y. Chang et al. / Heart & Lung 50 (2021) 491 498


62. Sevinç S, et al. Expectations of relatives of Syrian patients in intensive care units in 68. Agren S, et al. The health promoting conversations intervention for families with a
a state hospital in Turkey. J Clin Nurs. 2016;25(15 16):2232–2241. https://doi.org/ critically ill relative: a pilot study. Intens Crit Care Nurs. 2019;50:103–110. https://
10.1111/jocn.13254. doi.org/10.1016/j.iccn.2018.04.007.
63. Wong P, et al. The impact of social support networks on family resilience in an aus- 69. Chhetri IK, Thulung B. Perception of nurses on needs of family members of patient
tralian intensive care unit: a constructivist grounded theory. J Nurs Scholarsh. admitted to critical care units of teaching hospital, Chitwan Nepal: a cross-sec-
2019;51(1):68–80. https://doi.org/10.1111/jnu.12443. tional institutional based study. Nurs Res Pract. 2018;2018:1–7. https://doi.org/
64. Shavitt S, et al. Culture moderates the relation between perceived stress, social 10.1155/2018/1369164.
support, and mental and physical health. J Cross Cult Psychol. 2016;47(7):956–980. 70. Fortunatti P, Felipe C. Most important needs of family members of critical patients
https://doi.org/10.1177/0022022116656132. in light of the critical care family needs inventory. Invest Educ Enferm. 2014;32
65. Wang S-w, Lau AS. Mutual and non-mutual social support: cultural differences in (2):306–316. https://doi.org/10.17533/udea.iee.v32n2a13.
the psychological, behavioral, and biological effects of support seeking. J Cross Cult 71. McKiernan M, McCarthy G. Family members’ lived experience in the intensive care
Psychol. 2015;46(7):916–929. https://doi.org/10.1177/0022022115592967. unit: a phemenological study. Intens Crit Care Nurs. 2010;26(5):254–261.
66. Tweed RG, White K, Lehman DR. Culture, stress, and coping: internally-and exter- 72. Egerod I, et al. Constructing the illness narrative: a grounded theory exploring
nally-targeted control strategies of European Canadians, East Asian Canadians, and patients' and relatives' use of intensive care diaries. Crit Care Med. 2011;39
Japanese. J Cross Cult Psychol. 2004;35(6):652–668. https://doi.org/10.1177/ (8):1922–1928. https://doi.org/10.1016/j.iccn.2010.06.004.
0022022104270109. 73. Johansson M, et al. Family members' experiences with intensive care unit diaries
67. Frías MT, Shaver PR, Díaz-Loving R. Individualism and collectivism as moderators when the patient does not survive. Scand J Caring Sci. 2018;32(1):233–240. https://
of the association between attachment insecurities, coping, and social support. J doi.org/10.1111/scs.12454.
Soc Pers Relat. 2014;31(1):3–31. https://doi.org/10.1177/0265407513484631.