| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
48 Hours) Mechanical Ventilation in the ICU*
* From the Epidemiology Data Center, Department of Epidemiology, Graduate School of Public Health (Ms. Im, and Drs. Belle and Mendelsohn), University Center for Social and Urban Research (Dr. Schulz), Department of Critical Care Medicine (Dr. Chelluri), University of Pittsburgh, Pittsburgh, PA.
Correspondance to: KyungAh Im, MS, Epidemiology Data Center, Department of Epidemiology, Graduate School of Public Health, University of Pittsburgh, 130 Desoto St, 127 Parran Hall, Pittsburgh, PA 15261; e-mail: im{at}edc.pitt.edu
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
Design: Prospective cohort study.
Study participants: Caregivers of 115 patients who received prolonged MV in an ICU.
Measurements: Sociodemographics, employment status, hours spent providing care, help from paid caregiving sources, caregiver burden, and caregiver depressive symptoms.
Results: The proportion of patients who survived at least 2 months and required caregiver support was 74.8%. The average age of caregivers was 52.9 years (SD, 14.2), 76.5% were women, and more than half were spouses (52.2%). Only 33 of the caregivers (28.7%) were working, and 30.3% had to reduce their time spent at work to provide care to the patient. The prevalence of risk of clinical depression (defined as Center for Epidemiological Studies depression scale [CES-D] score 
16) among caregivers was 33.9%. The mean caregiver CES-D score was 13.2 (SD, 11; median, 10). Multiple linear regression analysis showed that higher CES-D score was associated with more hours per day helping with patients’ activities of daily living and instrumental activities of daily living (p = 0.003).
Conclusions: Two months after being placed on MV for at least 48 h, a high proportion of patients need caregiver support. Approximately 34% of caregivers are at risk of clinical depression. Many caregivers report lifestyle changes and burden when providing care for the patients.
Key Words: caregiving • Center for Epidemiological Studies depression scale • critical care • mechanical ventilation • prevalence
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
Existing studies on the informal caregivers of patients receiving MV have focused almost exclusively on caregivers of patients receiving MV at home. Thomas et al10 studied needs and involvement in 44 caregivers of persons receiving ventilatory support at home. They reported that finances, handling emergencies, information, family relationships, and continuity of care were ranked the most important needs. Sevick et al11 studied 27 caregivers of home-based, ventilator-dependent patients and reported that home placement was not perceived to be a negative experience based on the caregiving appraisal scale.12 Findeis et al13 reported that most of the 13 family caregivers of individuals using home ventilators experienced a positive sense of mastery and satisfaction, but some believed, to a moderate degree, the burden and the negative impact of caregiving. Smith and Mayer14 identified 20 caregiver learning needs and reactions to providing care for adult patients receiving MV in the home. Using the caregiver reactions inventory instrument,15 they reported a positive correlation between the time patients spent on the ventilator per day and financial strain, schedule disruption, and negative reactions to caregiving. Others investigated economic aspects of managing ventilator-assisted individuals at home. Sevick and Bradham16 estimated caregiver effort in dollar value by converting time into cost estimates in a random sample of 277 caregivers. They found that home care for ventilator-assisted individuals is not always a less expensive alternative than care provided in institutional settings.
Few studies have examined the impact of caregiving in a more general sample of patients who survived prolonged (at least 48 h) MV in an ICU. Thus, the objectives of this study were as follows: (1) to assess the proportion of patients who experienced prolonged MV, and required a caregiver 2 months following the onset of prolonged (at least 48 h) ventilatory support; (2) to describe caregiver depressive symptomatology, burden, and caregiver limitations in activities; and (3) to identify factors associated with caregiver depressive symptoms at 2 months.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
An informal caregiver was defined as a nonpaid person, identified by the patient, who provided the most care for the patient. Caregivers were enrolled into the study during hospitalization of the patient and during the 2-month follow-up period. The current report describes caregiver outcomes 2 months following the onset of MV.
Data Collection and Measurements on Patients
Baseline data on patients were collected either by personal interviews with proxy or medical record review by trained interviewers, and included sociodemographics, hospital admission diagnoses, and Charlson comorbidity score unadjusted for age.17
Functional status at 2 months was assessed by the number of deficiencies in activities of daily living (ADL) and instrumental activities of daily living (IADL).18
19
ADL assess the need for help in performing basic activities of daily living (bathing, dressing, using the toilet, transfer in and out of bed/chair, continence, and feeding). ADL scores ranged from 0 to 6 indicating the number of activities for which help was required. IADL assess the ability to perform household chores, shopping, food preparation, housekeeping, doing laundry, traveling, taking medications, and handling finances. IADL scores ranged from 0 to 8. Higher scores for ADL and IADL are indicative of greater dependence.
Data Collection and Measurements on Caregivers
Caregiver sociodemographics, employment status, and relationship to the patient were collected 2 months after enrollment of the patient into the study. The following outcome measurements were examined:
Caregiver Depressive Symptomatology:
The primary outcome for this study was depressive symptomatology as measured by the Center for Epidemiologic Studies depression scale (CES-D).20
This instrument, developed for nonpsychiatric patients, is widely used in community surveys to measure depressive mood. Subjects are asked to respond to 20 items detailing depressive symptoms by indicating how often they experienced these symptoms during the past month. Scores have a potential range of 0 to 60. Higher scores indicate greater presence of depressive symptoms. A score of 16 was used as a screen for clinical depression by interviewers to report to the study principal investigator.21
The CES-D has been demonstrated to have acceptable levels of reliability and validity.20
22
23
24
The Cronbach coefficient 
for the CES-D was 0.90 in this study.25
Caregiver Changes in Role Function:
An 11-item scale was used to measure changes in caregiver activities due to caring for the patient. Caregivers were asked "Please tell me how much you have restricted your activities in each of the following areas." The areas assessed include the following: caring for him/herself, caring for others, eating habits, sleeping habits, doing household chores, going shopping for him/herself, visiting friends, working on hobbies, sports and recreation, going to work, and maintaining friendship. The four level responses were as follows: not restricted at all (1), slightly restricted (2), moderately restricted (3), and greatly restricted (4), yielding a potential range of scores from 11 to 44. Higher scores indicate more restrictions. The Cronbach coefficient for this measure was 0.89.25
Patient Behavior Frequency and Caregiver Reaction to These Behaviors:
An 18-item scale that includes items modified from the revised memory and behavior problems checklist (RMBPC)26
was used for assessing these domains. The RMBPC was originally designed to measure caregiver assessments of memory problems, disruptive behaviors, and depressive symptoms among demented care recipients. For the QOL-MV study,1
the RMBPC was modified by reducing the number of memory questions and adding items relevant to the population of people undergoing prolonged MV as determined by a psychologist (R.S.) and intensivist (L.C.) investigator. New items included sleeping, swallowing, pain, hearing, understanding people, following directions, and doing things for himself or herself. For behavior frequency, an overall score was generated by summing up all reported behaviors. This total score ranged from 0 to 18, with higher scores indicating more behavior problems. The Cronbach coefficient for this cohort was 0.73.25
This instrument also measured the amount of caregiver burden by scoring the amount of bother or upset experienced by the caregiver due to these patient behaviors. For caregiver burden, the five level responses to degree of bother or upset (not at all [0], a little [1], moderately [2], very much [3], and extremely [4]) were summed. A score of zero was assigned if a given behavior was not reported. The total burden score ranged from 0 to 72, with higher scores indicating more burden. The Cronbach coefficient was 0.78 for this measure.25
Statistical Analysis
Continuous variables are reported as mean ± SD, median, and 25th, 75th percentiles. Frequencies and percentages are reported for categorical variables. The primary outcome variable was caregiver 2-month depressive symptomatology (CES-D). Multiple linear regression was used for the multivariable modeling of caregiver 2-month CES-D. The distribution of CES-D scores was positively skewed so the square root transformation was used for normalization. The following independent variables were considered at the initial stage of modeling: caregiver age, sex, education (less than or equal to high school vs at least some college), relationship to patient (spouse vs nonspouse), hours per day spent helping patient, patient location at 2 months (home vs institution), and patient functional dependency at 2 months (ADLs and IADLs). Stepwise modeling beginning with all variables in the model was used to derive the preliminary main-effects model. The probability for entry and removal of variables from the model was 0.15. Due to possible colinearity among variables, the variance inflation factor27
and the consistency of direction of the associations were examined. Once a plausible main effects model was determined, all first-order interaction terms were tested for significance at = 0.05. Statistical significance was defined as a p value < 0.05. Statistical analyses were performed using SAS (Version 6.12; SAS Institute; Cary, NC) and Stata (Version 6.0; Stata Corporation; College Station, TX).
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
Characteristics of Patients in Analysis Cohort
There were 115 patients in the analysis cohort (Fig 1
). The median patient age was 57 years (Table 1
). More than half of the patients were male (62%), and the majority (89.6%) were white. The ICU admitting diagnoses were respiratory (41 patients, 35.7%), postsurgical (19 patients, 16.5%), trauma (19 patients, 16.5%), neurologic (18 patients, 15.7%), cardiovascular (3 patients, 2.6%), and "other" (15 patients, 13.0%). On average, patients spent approximately 14 days receiving MV. The median Charlson comorbidity score was 1.8. Median ADL and IADL scores at 2 months were 3 and 6, respectively. Approximately one half of patients were at home at 2 months, and the others were in an institution. Among those living in an institution, 41.7% were in a rehabilitation facility, 24.1% were in a nursing facility, 22.4% were in a hospital, and 12.1% were in a long-term ventilator facility.
|
|
|
Caregiver employment status is described in Table 2 . At 2 months, 33 caregivers (28.6%) were employed at a job either full-time or part-time. Of these 33 caregivers, 10 persons (30.3%) had to reduce their work hours (a median of 16 h/wk) in order to provide care to the patient. Forty-seven caregivers (40.9%) were either retired or not employed, 15 of whom (31.9%) stopped working because they needed to provide care to the patient.
Help From Friends/Family and Paid Caregivers
We asked caregivers about the number of days in the past week that other family members or friends (who are not being paid) provided help. Thirty-seven caregivers (32.2%) did not have any help from family or friends, though 19 of these caregivers (51.4%) reported receiving help from paid caregivers. Thirty-six caregivers (31.3%) received help from friends and family for 7 days in the past week, while the remaining 42 caregivers (36.5%) received help from 1 to 6 d/wk. Overall, 69 informal caregivers (69.3%) received help from paid caregivers.
Caregiver Assistance to Patient ADLs and IADLs by Patient Location
Tables 3
, 4
provide the percentages (defined as the number of caregivers providing assistance divided by the number of patients who needed assistance for that activity) of caregivers providing assistance for each ADL and IADL item according to patient location (home or institution). For each IADL, with the exception of helping with handling finances, caregivers were more likely to provide assistance (p < 0.05, Fisher exact test) with IADLs for patients living at home than for those who were at an institution. Although a smaller percentage of caregivers provided assistance for patients at an institution than at home, many of the informal caregivers were still involved in helping with ADLs and IADLs for institutionalized patients. The median time spent on providing care was 4 h/d for ADLs and IADLs, with half of the caregivers spending at least 5 h/d assisting the patient if the patient was at home, and half of the caregivers spending at least 3 h/d helping if the patient was at an institution (p = 0.026).
|
|
|
|
|
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
The patients in the QOL-MV cohort were recovering from an acute illness, although some had a history of one or more chronic conditions prior to the hospitalization. The result was that many patients had high levels of functional dependency at 2 months. So, more than half of the caregivers reported spending 4 h/d to help patients with ADL and IADL activities. In addition, 69% of the caregivers said that they received help from paid caregiving sources. The need for caregiver support continued even after admission to an institution.
Caregiver CES-D: Risk of Clinical Depression Among Caregivers
Collins et al24
reported that mean CES-D scores do not generally exceed 10 in the general population, and Aneshensel et al29
reported that the mean CES-D scores for 172 adult women was approximately 8.6 based on a randomly selected sample of households in a community in California. In the current study, the mean CES-D score was 13.2 for the caregivers (median, 10). Schulz et al30
reported that mean CES-D scores in dementia caregiver studies range between 14.5 and 34.4. Thus, the mean CES-D score for the caregivers in our cohort appears to be higher than that of the general population but lower than that of caregivers of dementia patients. However, the prevalence of risk of clinical depression in our caregiving cohort (34%) is comparable to the prevalence among informal caregivers of frail elderly people found by Tennstedt et al,31
and is at least double that of general population, typically noted as approximately 15%.30
Furthermore, the prevalence of risk of clinical depression in our cohort is similar to the reported proportion at risk for clinical depression of caregivers of Alzheimer disease and/or related dementia patients (between 28% and 55%).30
Caregiver CES-D: Risk factors by Different Type of Caregiving Populations
Caregiving studies on patients surviving critical illness and mechanical ventilatory support are rare. As a result, it is difficult to compare these results to other studies.
We begin with a brief summary of generally known risk factors for CES-D among different caregiving populations. The risk factors for higher CES-D score among caregivers include being female, experiencing stressful life events such as bereavement, lacking a supportive social network, physical health problems, economic hardships, and loss of control.30 Risk factors for the caregivers of patients with Alzheimer or dementia-related disease also include the relationship to patient (spouses were more likely to be distressed than other relatives), low self-rated health status, increased burden, and more patient behavioral problems. However, low levels of stress, high ratings of life satisfaction, high levels of self-esteem, and mastery were related to low CES-D scores.30 Others reviewed risk factors among the caregivers of terminally ill patients, such as patients with cancer, and reported that older age, caring for male patients,32 patient dependencies with regard to daily living activities, and patient depression levels33 34 were related to caregiver risk of depression.
Amount of Assistance
In our study, we examined caregiver demographic variables (age, sex, education, relationship to patient), caregiving context (such as the amount of assistance provided to patients and the location of patients), and patient functional dependency (ADLs and IADLs). Factors associated with caregiver depressive symptomatology in our study were the number of hours providing care for ADL or IADL and location of the patient.
Tennstedt et al,31 studying caregivers for impaired elderly, specifically pointed out that there was no relationship between the presence of depressive symptoms and caregiving tasks, measured by the hours per week of care and the number of ADLs with which the caregiver provides assistance. In the work of Schulz et al,30 however, the relationship between the amount of assistance and depressive symptomatology was found to be inconsistent throughout the different caregiving studies of Alzheimer disease.
We did not find that the effect of the amount of assistance on caregiver depressive symptoms depended on patient location (ie, home vs institution). However, our data show caregivers spent more hours helping patients if patients were at home. We speculated that caregivers may have overestimated the time spent caregiving at home because they had more contact with patient with no clear start and stop point, whereas time estimates for institutionalized care may be more accurate because caregivers know when they start and stop providing help.
Due to the cross-sectional nature of this study, we do not know whether longer hours helping patients cause the greater number of depressive symptoms or these symptoms lead to more time spent providing care. Due to limitations in data collection, we do not know whether these caregivers had these symptoms prior to providing care, or whether the symptoms would have developed even in the absence of a caregiving role. However, the results demonstrate that there exists an association between caregiving context (such as the amount of assistance provided to patients) and depressive symptoms.
Patient Location
There was not a statistically significant (p = 0.07) difference in depressive symptoms between caregivers of institutionalized patients and other caregivers. Nevertheless, this may underscore the practical significance of different strategies to help this group of caregivers.
Some studies,24 35 36 37 38 using a cross-sectional approach to detect an effect of location, reported similar findings: there is little difference in measures of mental health between residential and institutional caregivers. But Schulz et al30 noted that studies are inconsistent with regard to this question. Others have attempted to examine this relationship through longitudinal studies.25 However, these studies were done for dementia patients. More research is required that examines caregivers of patients receiving MV.
Informal Caregiving Does Not End Even After Institutionalization
Findings from our study indicate that informal caregivers provide caregiving assistance for patients at institutions. It was previously believed that admission to a nursing home or an institution is generally regarded as the termination of family care and the commencement of institutional care.39
However, Bowers in 1988 examined family perceptions of nursing home care and their own involvement in that care, and families assumed extensive responsibilities that went beyond the visible tasks they performed.35
Further, studies39
40
indicated that the family caregivers perceived themselves to have a greater role in caring for relatives than that perceived by the nursing home staff and pointed out they perceived themselves as having an important role in providing social and emotional care.
What Can We Do To Help Caregivers?
It was in the mid 1980s that researchers in Alzheimer caregiver studies examined the efficacy of interventions for caregivers. Despite previous skepticism among some researchers, there has been a positive attitude toward supporting intervention research for the caregivers. Knight et al41
conducted a systematic review and concluded that respite interventions and individual psychosocial interventions are moderately effective on caregiver distress (defined broadly to include subjective burden, depression, anxiety, hostility, and other measures of negative effect) compared with treatment as usual or uncontrolled use of community services. Also, interventions providing support for the caregivers in caring for the patients were shown to improve depression and reduce burden in caregivers of patients with Alzheimer disease and patients with dementia,42
43
although the amount of change found among study participants was relatively small. In contrast, Tennstedt et al31
suggested that caregiver depression is best treated directly (eg, by drug therapy and/or psychotherapy) rather than indirectly by attempting to alleviate the negative consequences of providing care, for example, by relieving the caregiver of care responsibilities or by introducing formal services for the care recipients.
Researchers in Alzheimer caregiving studies have already begun to implement multicomponent interventions for caregivers and are working on how to better classify different characteristics of interventions: emphasizing more pragmatic approaches to the method, ie, articulating the "who, what, and how" of interventions.44 Because so little is known about caregiving needs after critical illness, it is questionable which one of these approaches would be the best for this population. We need a better understanding of the impact of providing caregiver support for this group, so that the systematic strategies that are suitable for this population could be evaluated.
Limitations
Due to the absence of baseline data in this study, we were unable to examine whether the depressive symptoms increased, decreased, or remained constant over the course of caregiving. Nor do we know how many other caregivers had provided care in the past to either the patient or to other family members. Knowledge of previous caregiving experience in a similar setting, as well as caregiver experience with handling their depressive symptoms are important to measure in future, longitudinal research to help to understand the relationship of these two issues.
|
Conclusions |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
|
Footnotes |
|---|
Research partly supported by the National Institute on Aging, R01 AG11970-05.
Received for publication March 7, 2003. Accepted for publication August 12, 2003.
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionConclusionsReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  B. J. Daly, S. L. Douglas, C. G. Kelley, E. O'Toole, and H. Montenegro Trial of a Disease Management Program to Reduce Hospital Readmissions of the Chronically Critically Ill Chest, August 1, 2005; 128(2): 507 - 517. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Azoulay, F. Pochard, N. Kentish-Barnes, S. Chevret, J. Aboab, C. Adrie, D. Annane, G. Bleichner, P. E. Bollaert, M. Darmon, et al. Risk of Post-traumatic Stress Symptoms in Family Members of Intensive Care Unit Patients Am. J. Respir. Crit. Care Med., May 1, 2005; 171(9): 987 - 994. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. L. Douglas, B. J. Daly, K. Im, S. Belle, L. Chelluri, A. B. Mendelsohn, and R. Schulz Caregiving and Long-term Mechanical Ventilation Chest, October 1, 2004; 126(4): 1387 - 1388. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
The caregivers and patients were under severe stress because of the acute illness, and this underscores the difficulty in performing studies of this nature in this patient population. We believe this reflected on our refusal rate of 34.7%. However, the baseline patient characteristics such as age, sex, Charlson comorbidity score, as well as in-hospital severity of illness (eg, acute physiology score, APACHE [acute physiology and chronic health evaluation] 3, risk of death) did not differ significantly between the 115 caregivers who participated in the study and 61 caregivers who refused an interview at 2 months (2m).