Abstract
Aim:
To investigate the association between informal and formal care and stroke participants’ self-reported health-related quality of life and depressive symptoms one year after the stroke event.
Method:
We examined a national population-based sample of 123 stroke participants. Care received was defined as formal (outpatient therapy, care from home health aides, nurses or therapists), informal (family caregiver) or shared (formal and informal). Hours of care were extracted from Medicare claims and caregiver self-report. A general linear model was used to compare health-related quality of life and depressive symptoms one year after the stroke for those who received shared care, only informal care or only formal care to those with no post-stroke care.
Results:
Among stroke participants, 12.2% received only formal care, 35.0% only informal care, 38.2% shared care, and 14.6% no care. Those with only informal care had greater self-reported depressive symptoms than those who received no care at all.
Conclusions:
Shared care was the most common care configuration for stroke participants, but no significant associations were found between shared care and self-reported outcomes after adjusting for stroke participant characteristics. Further research is needed on the dose of informal and formal care and their coordination to better understand relationships with recovery.
Keywords: stroke, caregivers, rehabilitation, Medicare, depression, quality of life
Introduction
Stroke is a leading cause of disability in the United States.1 Of the 7 million stroke survivors across the country, two-thirds report some form of disability, half of which are moderate to severe [2,3]. One in three report depression at 1 year after the stroke event [4]. As such, ongoing post-stroke care is needed to mitigate the effects of these conditions. But despite the high prevalence of functional dependence and depressive symptoms, over 40% of stroke survivors are not referred to any post-acute care and 60% do not receive such care in the first 30 days after discharge [4,5].
Low healthcare utilization rates may suggest that informal caregivers including family relatives, partners, or friends play a significant role in stroke patients’ recovery. Nearly half of all stroke survivors report co-residing with an informal caregiver after discharge [6]. Rehabilitation in the home may be particularly beneficial as it facilitates greater emphasis on engaging in daily activities, and has shown greater improvement in activities of daily living (ADL) abilities [7]. Having an available informal caregiver might also facilitate post-acute care options such as early supported discharge [8].
Shared care is defined as receiving care from both informal caregivers and trained healthcare professionals, also referred to as formal care, a practice that is useful across settings with differential resource availabilities. For example, task-sharing has proven to be feasible and effective in low-resource settings in treating infectious diseases and performing surgical operations, though not yet found to be effective for post-stroke care [9,10,11]. In the Netherlands, a high-income country, the increasing number of informal caregivers who are also employed full-time or part-time show the need for a combination of care from different sources [12]. However, although studies have measured the provision of post-stroke care and support by informal caregivers [13], the extent to which this type of care is shared or shifted from formal healthcare providers to informal caregivers is unclear. Furthermore, while numerous observational studies have assessed the impact of caregiving on caregivers’ health [14,15], not many have focused on care recipients’ health-related outcomes.
The purpose of this study was to examine the provision of shared care for Medicare beneficiaries who were living in the community after a stroke, as well as to examine the association between shared care and patients’ self-reported functional outcomes and depressive symptoms one year after the stroke event.
Methods
Data sources
This study was a retrospective analysis of stroke survivors in the Reasons for Geographic and Racial Differences in Stroke project who had caregivers recruited for the Caring for Adults Recovering from the Effects of Stroke study and data available in Medicare fee-for-service claims.
Data were collected from participants enrolled in the national Reasons for Geographic and Racial Differences in Stroke study, which was designed to identify the causes of regional and racial disparities in stroke mortality. A total of 30,239 participants were enrolled into the project from 2003 to 2007. All participants were above 45 years of age at the time of enrollment and self-identified as either African American or Caucasian/white. African Americans and residents of “stroke belt” states (Alabama, Arkansas, Georgia, Kentucky, Louisiana, Mississippi, North Carolina, South Carolina, and Tennessee) were oversampled by design. Every six months, follow up telephone interviews were conducted that included queries about hospitalizations, outpatient visits, and symptoms that might indicate possible stroke events. Medical record information regarding the reported stroke event were retrieved and subsequently adjudicated by a team of trained reviewers including stroke neurologists. A data use agreement was established between the national study team and the Center for Medicare and Medicaid Services to obtain claims files for study participants. Stroke participants identified by the Reasons for Geographic and Racial Differences in Stroke project were then recruited along with a caregiver to participate in the ancillary Caring for Adults Recovering from the Effects of Stroke project.
Demographic information for the stroke participant includes age, race, gender, marital status, income, education, and whether or not stroke participant lives with their caregiver. Demographic information for the caregiver includes age, gender, and type of caregiver (spouse, significant other, children, other family, or non-family) were obtained. These variables were based on self-report and were obtained during a structured telephone interview conducted at enrollment into the Caring for Adults Recovering from the Effects of Stroke study, which took place on average 9 months after the stroke event. Informed consent was obtained before this interview. Caregivers who were enrolled were asked to estimate the number of hours of care per day that they provided to the stroke participant. Other variables were extracted from the Medicare claims files including the length of hospital stay (measured in days), days at home since hospital discharge, and whether or not the stroke participant stayed at an inpatient rehabilitation facility (IRF) or skilled nursing facility (SNF) since the stroke. Additional variables extracted from the adjudicated medical records included an estimated National Institutes of Stroke Scale (NIHSS) and the modified Rankin Scale (mRS) at hospital discharge.
Additional details about the recruitment, enrollment, and assessment procedures for the Reasons for Geographic and Racial Differences in Stroke study and the Caring for Adults Recovering from the Effects of Stroke study have been described elsewhere [16,17,18]. All procedures were reviewed and approved by the Institutional Review Board of the University of Alabama at Birmingham and Johns Hopkins University.
Study inclusion criteria
The patients in this study met the following inclusion criteria. First, 360 Reasons for Geographic and Racial Differences in Stroke participants were enrolled into the Caring for Adults Recovering from the Effects of Stroke study. Of those, 31 did not have a primary caregiver available for interview and were excluded from the analysis. This study only included patients with ischemic strokes to ensure that the clinical characteristics of the study population are sufficiently similar. As such, the 100 patients that did not have adjudicated ischemic strokes were also excluded. Next, the 101 patients who did not have continuous Medicare (A+B-C) coverage from the stroke event to the interview with the Reasons for Geographic and Racial Differences in Stroke study team were excluded, since no information about their conditions after discharge was obtained. Lastly, 3 patients were interviewed more than 13 months after the stroke event, and were thus excluded to ensure that the study retained focus on one-year outcomes after stroke. Of the 125 remaining, 2 did not have data on informal care hours. The final study population included 123 patients (figure 1).
Figure 1.
Flow chart of stroke participants who met inclusion/exclusion criteria for study population
Medicare claims data
Inpatient, outpatient, carrier, home health aide (HHA) and skilled nursing facility (SNF) base claims and revenue files were linked for each stroke participant by a unique beneficiary identification number. Revenue center codes in the outpatient and HHA files were examined to determine the number of Physical Therapy (PT), Occupational Therapy (OT), and Speech and Language Therapy (SLP) visits. The number of home health visits for skilled nursing, medical social service, home health aides and other home health visits not including PT, OT or SLP was calculated by counting the number of visits in the HHA revenue file within the defined time period of this study. The number of PT, OT and SLP visits were calculated by examining revenue center codes in the revenue files from the outpatient and HHA files and summing the total number of visits within the defined time period. PT, OT, SLP and HHA claims were extracted for the time interval between the stroke event date and the Caring for Adults Recovering from the Effects of Stroke study interview. Additional details about abstraction of these variables have been described elsewhere [6].
Outcome measures
Stroke participants were surveyed by staff from the Caring for Adults Recovering from the Effects of Stroke study for depressive symptoms as well as physical and mental functional outcomes within 13 months of the stroke event. Post-stroke depressive symptoms were measured with the 20-item Center for Epidemiologic Studies-Depression scale (CES-D). Frequency of each symptom was rated on a 0 (rarely) to 3 (most of the time) scale; higher scores indicate greater depression symptoms. The 12-item Short Form Health Survey assessed health-related quality of life, measured by the Physical Component Summary (PCS) and Mental Component Summary (MCS) scores. In the U.S. population, SF-12 measures have a mean of 50 and a standard deviation of 10, with higher scores indicating better health. In this study, we analyzed the change in SF-12 component scores by subtracting the pre-stroke score from the post-stroke score. Negative values indicate a decrease in health status and positive values indicate an increase in health status, as defined by the SF-12 component scores.
Care types and hours
Formal care was determined by Medicare claims indicating care was provided for the stroke participant during a home health visit, which included care by a home health aide (HHA), skilled nursing, medical social service or a rehabilitation therapist (PT, OT and SLP). Formal care also included outpatient rehabilitation (PT, OT and SLP). Healthcare utilization in Medicare claims was billed in quarter-hour intervals. The total number of hours of utilization was obtained by dividing the number of claims by four. The median number of total formal care hours received among all stroke participants was 20 hours, excluding those with no hours of care provided. We used the median as a cutoff for demarcating low/high formal care. Low formal care was defined as having less than or equal to 20 hours of formal care from hospital discharge to the date of the interview. High formal care was defined as having more than 20 hours.
Interviews with the stroke participant’s primary caregiver at enrollment into the Caring for Adults Recovering from the Effects of Stroke study on average 9 months after the stroke event estimated the number of informal care hours provided per day. Information about other caregivers who may have provided care for the stroke participant was not available. To determine the total hours of informal care over the course of the stroke participant’s stay at home, we multiplied the caregiver-reported number of hours per day by the number of days that the stroke participant was at home since the stroke. This excluded any IRF or SNF stay as well as any hospitalization. The median number of total informal care hours received among all stroke participants was 1200 hours, excluding those with missing data or no hours of care provided. We used the median as a cutoff for demarcating low/high informal care. Low informal care was defined as having less than or equal to 1200 hours of care between the time of discharge to the interview. High informal care was defined as having more than 1200 hours.
We conducted several exploratory analyses to define the shared care group. We constructed a 3x3 matrix with informal and formal care along each axis, stratified by none, low, and high number of hours (table 1). First, we compared “high-users”, defined as stroke participants who received a high number of hours for at least one type of care in addition to any amount (>0) of the other type of care, with everyone else. The small sample size of this group precluded further analysis. Next, we compared those who received both informal and formal care to everyone else. Due to differences in healthcare utilization and post-stroke disability between those who had no care hours and those who received formal or informal care only, these groups were separated. We ultimately defined shared care as receiving any combination of both formal and informal care. Care type was categorized as the following: no care, only informal care, only formal care, and shared care (figure 2).
Table 1.
Number of patients by care type and hours
| No formal care hours | Low formal care hours (≤20 hrs) | High formal care hours (>20 hrs) | |
|---|---|---|---|
| No informal care hours | 18 (29.51%) | 10 (35.71%) | 5 (14.71%) |
| Low informal care hours (≤1200 hrs) |
23 (37.70%) | 8 (28.57%) | 14 (41.18%) |
| High informal care hours (>1200 hrs) | 20 (32.79%) | 10 (35.71%) | 15 (44.12%) |
Figure 2.
Stroke participants with different levels of care hours in each group
Statistical analyses
The analysis focused on identifying differences between groups by care type (no care, informal care only, formal care only, shared care) and care hours (none, low, high) for both formal and informal care.
For each care group, descriptive analyses presented the means and standard deviations for continuous demographic and clinical variables including stroke participant age, mRS at discharge, NIHSS, length of hospital stay, the number of days at home at the time of interview, and caregiver age. The mRS is scored from 0–6, with 0–2 indicating no symptoms or slight symptoms of disability and capability for independent living, 3–5 indicating disability and dependence, and 6 indicating death. Univariate linear regressions were used to test the significance of associations between the types of care received (independent variable) and the continuous variables.
For each care group, descriptive analyses presented the counts and percentages for binary or categorical variables including stroke participant race, gender, relationship status or living situation, IRF or SNF stay, caregiver gender, and caregiver type. Fisher’s exact test was used to determine differences between care groups in income and education, which were both coded as categorical variables. Univariate logistic regressions were used to test the significance of associations between the types of care received (independent variable) and categorical variables.
Supplemental descriptive analyses using the same procedures were conducted to determine the variables associated with receiving none, low or high informal care. This analysis was repeated for formal care.
We used the general linear model to analyze the associations between the type of care received (independent variable) and the stroke participant’s functional outcomes and depressive symptoms (dependent variables), adjusting for the stroke participant’s age, race, mRS score after discharge, whether or not the stroke participant lives with the caregiver interviewed in the study, and whether or not the stroke participant stayed in an intensive rehabilitation facility (IRF) or a skilled nursing facility (SNF).
All analyses were conducted using SAS, version 9.4.
Results
Description of Cohort
The 123 participants from the Reasons for Geographic and Racial Differences in Stroke study with adjudicated ischemic stroke selected for this analyses were all enrolled in the Caring for Adults Recovering from the Effects of Stroke study within 13 months of their stroke, had a family caregiver and continuous Medicare coverage from the stroke event to the study interview. If no claim for a service was found, it was assumed that the stroke participant did not use it.
Type of care received
Of the 123 stroke participants in this study, 18 received no formal or informal care. Of the 58 who received only one of the two types of care, 43 received informal care and 15 received formal care. Forty-seven received shared care.
Of the 15 stroke participants who only received formal care, 10 received low formal care hours (<20 hours) and 5 received high formal care hours (>20 hours). Of those who received only informal care, 23 received low informal care hours (<1200 hours) and 20 received high informal care hours (>1200 hours). The 47 stroke participants who had shared care received four different combinations of formal and informal care hours. Eight received both low formal care and low informal care hours, 15 received both high formal care and high informal care hours, 14 received high formal care and low informal care hours, and 10 received low formal care and high informal care hours (table 1).
Stroke participant and caregiver characteristics by care type
Descriptive analysis showed that half of all stroke participants included in this study had an mRS score of 0 or 1 at discharge, indicating no symptoms or no significant disabilities despite minor symptoms. Shared care was associated with more severe strokes (higher NIHSS score, longer hospitalization) (P<0.01), longer hospitalizations (4.53 vs. 2.17 days; P=0.01), greater functional dependence at discharge (mRS[mean]=2.93 vs. 0.72; P<0.0001), IRF or SNF stay after hospitalization (34/47 vs. 6/18; P=0.006) and fewer days at home between the stroke event and the interview (260.74 vs. 303.83 days; P=0.002), compared to receiving no care. Compared to receiving no care, receiving only formal care was associated with greater functional dependence at discharge (mRS[mean]=2.00 vs. 0.72; P=0.016) and receiving only informal care was associated with fewer days at home since the stroke event (276.470 vs. 303.83 days; P=0.046). Groups also differed significantly in levels of educational attainment (P=0.016). No differences between groups were found for any of the other stroke participant or caregiver characteristics (table 2).
Table 2.
Demographics of the stroke participants by type of care
| Variable Mean (SD), n (%) |
Overall (n=123) | No care1 (n=18) | Formal care only (n=15) | Informal care only (n=43) | Shared care (n=47) |
|---|---|---|---|---|---|
| Stroke participant characteristics | |||||
| Age | 75.71 (6.57) | 75.75 (6.11) | 75.50 (8.05) | 76.24 (6.08) | 75.27 (6.83) |
| Race (African American) | 37 (30.08%) | 2 (11.11%) | 2 (13.33%) | 13 (30.23%) | 20 (42.55%) |
| Gender (female) | 58 (47.15%) | 8 (44.44%) | 8 (53.33%) | 18 (41.86%) | 24 (51.06%) |
| Married/co-residing | 72 (58.54%) | 10 (55.56%) | 9 (60.00%) | 25 (58.14%) | 28 (59.57%) |
| Lives with caregiver | 78 (63.41%) | 11 (61.11%) | 7 (46.67 %) | 32 (74.42%) | 28 (59.57%) |
| mRS at discharge | 1.97 (1.64) | 0.72 (0.89) | 2.00 (1.73) | 1.49 (1.45) | 2.93 (1.52) |
| NIHSS2 | 1.70 (2.29) | 0.39 (0.70) | 0.39 (0.70) | 1.12 (1.33) | 2.84 (3.01) |
| Days in hospital | 3.34 (3.33) | 2.17 (1.69) | 3.40 (2.95) | 2.51 (2.91) | 4.53 (3.92) |
| IRF/SNF stay | 62 (50.41%) | 6 (33.33%) | 6 (40.00%) | 16 (37.21%) | 34 (72.34%) |
| Days at home | 276.38 (49.68) | 303.83 (38.08) | 291.53 (43.15) | 276.70 (40.81) | 260.74 (57.47) |
| Education3 | |||||
| < High school | 17 (13.82%) | 2 (11.76%) | 0 (0.00%) | 9 (52.94%) | 6 (35.29%) |
| High school grad | 32 (26.02%) | 2 (6.25%) | 7 (21.88%) | 13 (40.63%) | 10 (31.25%) |
| Some college | 31 (25.20%) | 7 (22.58%) | 0 (0.00%) | 7 (22.58%) | 17 (54.84%) |
| > College | 43 (34.96%) | 7 (16.28%) | 8 (18.60%) | 14 (32.56%) | 14 (32.56%) |
|
Caregiver characteristics | |||||
| Age | 62.48 (13.03) | 65.84 (13.28) | 63.86 (14.56) | 63.85 (14.51) | 59.49 (10.61) |
| Gender (female) | 96 (78.05%) | 13 (72.22%) | 12 (80.00%) | 35 (81.40%) | 36 (76.60%) |
| Relationship to stroke participant | |||||
| Significant other | 64 (52.03%) | 11 (61.11%) | 6 (40.00%) | 25 (58.14%) | 22 (46.81%) |
| Children | 47 (38.21%) | 3 (16.67%) | 6 (40.00%) | 14 (32.56%) | 24 (51.06%) |
| Other family | 8 (6.50%) | 3 (16.67%) | 2 (13.33%) | 2 (4.65%) | 1 (2.13%) |
| Non-family | 4 (3.25%) | 1 (5.56%) | 1 (6.67%) | 2 (4.65%) | 0 (0%) |
Group with no care used as reference
NIHSS information not available for all patients
Fisher’s exact test compared all education across all 4 care groups
Stroke participant and caregiver characteristics by care hours
Analyses of informal and formal care hours showed that stroke participants with high informal care hours were more likely to be African American (48.89% vs. 12.12% P=0.002), living with their caregivers (77.78% vs. 54.55%; P=0.033), and to not have graduated from high school (24.44% vs. 6.06%) when compared to those with no informal care hours. They spent fewer days at home (274.64 days vs. 298.24; p=0.034) and have more formal care hours (21.98 hours vs. 7.77; p=0.028) than stroke participants with no informal care hours. Receiving high informal care hours was also associated with higher mRS at discharge (2.54 vs. 1.26; P=0.001) and higher NIHSS scores (2.39 vs. 0.80; P=0.004; table 3).
Table 3.
Demographics and healthcare utilization of the stroke participants by informal care hours
| Variable Mean (SD), n (%) |
No informal care hours (n=33) | Low informal care hours (n=45) | High informal care hours (n=45) |
|---|---|---|---|
| Stroke participant characteristics | |||
| Age | 75.64 (6.94) | 77.17 (6.41) | 74.29 (6.26) |
| Race (African American) | 4 (12.12%) | 11 (24.44%) | 22 (48.89%) |
| Gender (female) | 16 (48.48%) | 24 (53.33%) | 18 (40.00%) |
| Married/co-residing | 19 (57.58%) | 25 (55.56%) | 28 (62.22%) |
| Living w. caregiver | 18 (54.55%) | 25 (55.56%) | 35 (77.78%) |
| mRS at discharge | 1.26 (1.44) | 1.95 (1.63) | 2.54 (1.63) |
| NIHSS | 0.80 (1.21) | 1.66 (2.27) | 2.39 (2.69) |
| Days in hospital | 2.73 (2.39) | 3.00 (2.68) | 4.13 (4.29) |
| IRF or SNF stay | 12 (36.36%) | 25 (55.56%) | 25 (55.56%) |
| Days at home | 298.24 (40.29) | 262.09 (52.63) | 274.64 (48.18) |
| Formal care1 hours | 7.77 (13.05) | 15.21 (22.63) | 21.98 (38.61) |
|
Caregiver characteristics | |||
| Age | 64.94 (13.69) | 61.78 (11.68) | 61.37 (13.85) |
| Gender (female) | 25 (75.76%) | 34 (75.56%) | 37 (82.22%) |
| Relationship to stroke participant | |||
| Significant other | 17 (51.52%) | 20 (44.44%) | 27 (60.00%) |
| Children | 9 (27.27%) | 22 (48.89%) | 16 (35.56%) |
| Other family | 5 (15.15%) | 2 (4.44%) | 1 (2.22%) |
| Non-family | 2 (6.06%) | 1 (2.22%) | 1 (2.22%) |
Formal care includes: home health visits, outpatient therapy, skilled nursing care, medical social services
Stroke participants who received high formal care hours were more likely to be female (64.71% vs. 42.62%; P=0.041) and less likely to live with their caregivers (44.12% vs. 70.49%; P=0.013). Caregivers of stroke participants with higher formal care hours were more likely to be younger (age[mean]=58.77 vs. 64.44; P=0.044) and be the children of the stroke participants (52.94% vs. 27.87%; P=0.022). Compared to the group with no formal care hours, those with high formal care hours stayed for more days at the hospital (4.29 vs. 2.41; P=0.007), were more likely to stay at an IRF or SNF (85.29% vs. 36.07%; P<0.001), stayed for fewer days at home (261.71 vs. 284.70; P=0.031), had higher mRS (3.13 vs. 1.25; P<0.001) and NIHSS scores (3.03 vs. 0.90; P<0.001) at discharge (table 4).
Table 4.
Demographics and healthcare utilization of the stroke participants by formal care hours
| Variable Mean (SD), n (%) |
No formal care hours (n=61) | Low formal care hours (n=28) | High formal care hours (n=34) |
|---|---|---|---|
| Stroke participant characteristics | |||
| Age | 76.09 (6.04) | 75.85 (6.36) | 74.89 (7.68) |
| Race (African American) | 15 (24.59%) | 7 (25.00%) | 15 (44.12%) |
| Gender (female) | 26 (42.62%) | 10 (35.71%) | 22 (64.71%) |
| Married/co-residing | 35 (57.38%) | 19 (67.86%) | 18 (52.94%) |
| Living w. caregiver | 43 (70.49%) | 20 (71.43%) | 15 (44.12%) |
| mRS at discharge | 1.25 (1.35) | 2.23 (1.53) | 3.13 (1.57) |
| NIHSS | 0.90 (1.21) | 1.92 (2.53) | 3.03 (2.97) |
| Days in hospital | 2.41 (2.60) | 4.21 (4.34) | 4.29 (3.19) |
| IRF or SNF stay | 22 (36.07%) | 11 (39.29%) | 29 (85.29%) |
| Days at home | 284.70 (41.62) | 276.07 (48.45) | 261.71 (60.86) |
| Formal care1 hours | 954.92 (1293.49) | 1672.50 (2245.26) | 1515.59 (1579.59) |
|
Caregiver characteristics | |||
| Age | 64.44 (14.08) | 62.71 (11.85) | 58.77 (11.46) |
| Gender (female)1 | 48 (78.69%) | 22 (78.57%) | 26 (76.47%) |
| Relationship to stroke participant | |||
| Significant other | 36 (59.02%) | 15 (53.57%) | 13 (38.24%) |
| Children | 17 (27.87%) | 12 (42.86%) | 18 (52.94%) |
| Other family | 5 (8.20%) | 0 (0.00%) | 3 (8.82%) |
| Non-family | 3 (4.92%) | 1 (3.57%) | 0 (0.00%) |
For male stroke survivors, 98% of the caregivers were female; for female stroke survivors, 55% of the caregivers were female
Stroke participant functional outcomes and depressive symptoms
In our unadjusted analysis, we found that those receiving only informal care and those receiving shared caregiving had higher depressive symptoms compared to those with no care (table 5). After adjusting for stroke participant’s age, race, mRS score after discharge, whether or not the stroke participant lives with the caregiver interviewed in the study, and whether or not the stroke participant stayed in an IRF/SNF, there was no difference in physical and mental functional outcomes or depressive symptoms between those with no care versus those with shared care or with only formal care. Stroke participants receiving informal care only had greater depressive symptoms (CES-D[mean]=9.65 vs. 4.34; P=0.03) compared to those with no care. All groups showed decreases in physical component summaries of the SF-12 from pre- to post-stroke, and score increases on the CES-D scale. Stroke participants with no care or formal care only showed increases in mental component summaries of the SF-12, while the score decreased for those with informal care only or shared care (table 5).
Table 5.
Changes in SF-12 (Physical Component Score, PCS; Mental Component Score; MCS) and post-stroke Center for Epidemiological Studies-Depression (CES-D) score of study population
| Mean PCS (SE) | Difference with no care group (95%CI) | t | Mean MCS (SE) | Difference with no care group (95%CI) | t | Mean CES-D (SE) | Difference with no care group (95%CI) | t | |
|---|---|---|---|---|---|---|---|---|---|
| Unadjusted | |||||||||
| No care1 | −4.80 (2.42) | -- | -- | 1.03 (1.95) | -- | -- | 3.22 (1.86) | -- | -- |
| Formal care only | −5.73 (2.75) | −0.93 (−8.18, 6.32) | −0.25 (p=0.80) | 0.48 (2.22) | −0.55 (−6.40, 5.30) | −0.19 (p=0.85) | 6.13 (2.04) | 2.91 (−2.56, 8.39) | 1.05 (p=0.29) |
| Informal care only | −5.99 (1.58) | −1.19 (−6.92, 4.54) | −0.41 (p=0.68) | −0.28 (1.28) | −1.31 (−5.93, 3.32) | −0.56 (p=0.58) | 8.47 (1.21) | 5.24 (0.85, 9.64) | 2.36 (p=0.02) |
| Shared care | −6.21 (1.51) | −1.41 (−7.07, 4.25) | −0.49 (p=0.62) | −1.51 (1.22) | −2.54 (−7.10, 2.02) | −1.10 (p=0.27) | 8.49 (1.15) | 5.27 (0.93, 9.61) | 2.40 (p=0.02) |
| Adjusted2 | |||||||||
| No care1 | −4.71 (1.77) | -- | -- | 1.76 (2.25) | -- | -- | 4.34 (2.10) | -- | -- |
| Formal care only | −5.38 (3.21) | −0.65 (−8.56, 7.27) | −0.16 (p=0.87) | 2.01 (2.65) | 0.25 (−6.29, 6.78) | 0.08 (p=0.94) | 4.67 (2.37) | 0.33 (−5.64, 6.31) | 0.11 (p=0.91) |
| Informal care only | −5.09 (1.76) | −0.36 (−6.45, 5.73) | −0.12 (p=0.91) | −0.35 (1.45) | −2.11 (−7.14, 2.92) | −0.83 (p=0.41) | 9.65 (1.35) | 5.31 (0.66, 9.96) | 2.26 (p=0.03) |
| Shared care | −4.71 (1.77) | 0.02 (−6.61, 6.66) | 0.01 (p=0.99) | −1.37 (1.46) | −3.12 (−8.60, 2.36) | −1.13 (p=0.26) | 7.97 (1.34) | 3.63 (−1.46, 8.72) | 1.41 (p=0.16) |
No care used as reference group
Adjusted covariates: stroke participant age, race, mRS at discharge, lives with caregiver (Y/N), IRF/SNF stay (Y/N)
Discussion
Informal caregivers play an important part in stroke participants’ post-discharge recovery. In this population-based study, we found that stroke severity was associated with the type of care received without adjusting for other covariates. More severe strokes, longer hospitalizations and greater disability/dependence following the stroke were associated with the stroke participant later receiving shared care, compared to receiving no care at all. However, we found no associations between shared care and stroke survivors’ functional outcomes and depressive symptoms when compared to those who received no care. These findings suggest that stroke participants who received both formal and informal care may have benefited from this care in their post-discharge recovery since their physical and mental health outcomes were not significantly worse from the group that received no care, despite their greater impairment at hospital discharge.
Stroke participants with shared care, only formal care or only informal care all experienced more severe strokes than those who subsequently received no care at all. Patients receiving post-discharge care of some type stayed for more days at the hospital, were more likely to stay at an IRF or SNF, and as a result, were home for fewer days from the time of stroke to the telephone interview. This suggests that stroke participants who had greater functional dependence at discharge had greater need for inpatient as well as outpatient services. Higher healthcare utilization in the shared care and formal care-only groups was translated into functional outcomes and depressive symptoms that were not significantly different from the less impaired group. This may indicate that shared care and formal care were able to mitigate health deterioration, although the effect was not significant. However, those who received only informal care experienced significantly greater depressive symptoms. Given that there is a high prevalence of depression in stroke caregivers, these findings may be explained by previous literature showing that care recipient depression is associated with caregiver depression [19,20]. Furthermore, previous research has shown that stroke survivors and their caregivers differentially appraise stroke survivors’ cognitive, physical and psychosocial function, which may be detrimental to health-related outcomes for both the stroke survivor and the caregiver [21]. Since this phenomenon does not affect stroke survivors receiving no care, only formal care or shared care to the same extent, it may partially explain our finding of greater depressive symptoms in stroke survivors receiving only informal care.
Secondary analysis of subgroups stratified by the number of care hours showed that a higher proportion of female stroke participants received high formal care hours. Similarly, previous studies have shown that female participants are more likely to adopt stronger health-promotion behaviors and actively seek out care resources for themselves [22]. Interestingly, those who received high formal care hours were less likely to be living with their primary caregiver. The caregivers of stroke participants receiving high formal care hours were also younger than the caregivers of all other groups, and were more likely to be their children. This suggests that adult children who were caregivers may have been more involved in ensuring that their parents receive adequate post-discharge care compared to other caregivers, despite not living with them. Further study is needed to determine if different types of informal caregivers could differentially influence stroke participants’ care-seeking behavior or formal care utilization.
This study has a number of strengths, including the use of a population-based sample of stroke survivors and caregivers that enhances the generalizability of the results. This question has not been previously investigated and it is significant in its implications for stroke participants’ post-discharge care utilization. In particular, this study provides valuable information about which populations are receiving what kind of care, and it shows that the population with poor health at discharge and are receiving no care at all seem to be falling through a gap in the care continuum.
The large number of African American stroke participants and caregivers included in this study allow for greater generalizability than studies that primarily include Caucasian/white participants. However, other groups such as Hispanics are underrepresented, which limits generalizability to those populations. The caregivers in this study were identified as primary caregivers for the stroke survivors after the stroke event but may not have maintained the same hours of care for the entirety of the 10 months from hospital discharge until the structured telephone interview. There may be informal caregivers other than the primary caregiver enrolled in this study who provided care for the stroke participant and whose efforts were not included in the analysis. Furthermore, hours of informal care was self-reported by caregivers in the study to estimate the number of hours per day that they spent on providing care and may be subject to recall bias. Government medical insurance is standardized across the study population and all stroke participants have Medicare A and B, which include basic inpatient and outpatient services. However, due to the lack of data on privately paid caregivers outside of those paid by home health services, it is unclear if some stroke participants identified as receiving informal care may have, in fact, also received more formal and structured type of care. Therefore, although only 12.2% of the stroke participants in this study received only formal care, there are likely more stroke participants who received this type of care but were not identified in the Medicare claims data.
The small sample size of stroke participants and caregivers limited the possibilities for analyses. While it may have been more informative to separately examine functional outcomes and depressive symptoms of stroke participants receiving only informal care and only formal care, creating more subgroups for analysis would further decrease the size of each. Conducting population-based studies with larger samples of stroke participants and their caregivers may allow us to better understand how formal and informal care can best support patient care.
As healthcare costs continue to rise, it is imperative to utilize all the available resources to ensure that stroke patients are receiving the care that they need. Future work in shared care for patient populations with well-defined types of care will serve to improve the quality of care in the post-discharge period. Qualitative studies have proven to be useful and informative in the context of understanding stroke patient-caregiver dyads’ experiences [23]. A qualitative study of caregivers’ feedback analyzing open-ended interview data collected by the Caring for Adults Recovering from the Effects of Stroke study will be informative regarding challenges faced by informal caregivers and on how formal care services could best supplement their work. This could also reveal nuances in the types of care, such as the extent to which informal caregivers are involved in stroke participants’ care and whether private or paid caregivers supplement informal caregiving. Furthermore, interventions aiming to improve stroke patient outcomes such as the CARE4STROKE early supported discharge program often require collaboration from caregivers [24]. An understanding of the characteristics of stroke patients who receive formal and/or informal care will be essential to designing and implementing these trials.
Stroke is one of the leading causes of disability for adults in the United States and ongoing care is needed for survivors.
Informal caregivers such as family members play an important role in recovery in addition to formal care provided by home health aides, nurses and therapists.
This study showed that survivors with informal care had more severe depressive symptoms, while shared care provided by formal and informal caregivers may prevent further decline in patients with worse baseline health.
Acknowledgement:
The authors thank the other investigators, the staff, and the participants of the Caring for Adults Recovering from the Effects of Stroke study and Reasons for Geographic and Racial Differences in Stroke study for their valuable contributions. A full list of participating Reasons for Geographic and Racial Differences in Stroke investigators and institutions can be found at www.regardsstudy.org. This research project is supported by a cooperative agreement U01 NS041588 from the National Institute of Neurological Disorders and Stroke (NINDS), National Institutes of Health, Department of Health and Human Service. Additional funding for the Caring for Adults Recovering from the Effects of Stroke project was provided by an investigator-initiated grant (R01 NS075047) from NINDS. Additional funding was provided by Duke University. The content is solely the responsibility of the authors and does not necessarily represent the official views of NINDS or the National Institutes of Health. Representatives of the funding agency have been involved in the review of the manuscript but not directly involved in the collection, management, analysis, or interpretation of the data.
Footnotes
Declaration of interest: The authors report no conflicts of interest.
Data availability statement:
The data that support the findings of this study are available on request from the Reasons for Geographic and Racial Differences in Stroke executive committee. The data are not publicly available due to their containing information that could compromise the privacy of research participants.
References
- 1.Ovbiagele B, Goldstein LB, Higashida RT, et al. Forecasting the Future of Stroke in the United States: A Policy Statement From the American Heart Association and American Stroke Association. Stroke. 2013;44(8):2361–75. [DOI] [PubMed] [Google Scholar]
- 2.Wilkinson PR, Wolfe CD, Warburton FG, et al. A long-term follow-up of stroke patients. Stroke. 1997;28(3):507–12. [DOI] [PubMed] [Google Scholar]
- 3.Kelly-Hayes M, Beiser A, Kase CS, et al. The influence of gender and age on disability following ischemic stroke: the Framingham study. J Stroke Cerebrovasc Dis 2003;12(3):119–26. [DOI] [PubMed] [Google Scholar]
- 4.Benjamin EJ, Virani SS, Callaway CW, et al. Heart disease and stroke statistics—2018 update. Circulation. 2018;137:e67–492. [DOI] [PubMed] [Google Scholar]
- 5.Freburger JK, Li D, Johnson AM, Fraher EP. Physical and Occupational Therapy From the Acute to Community Setting After Stroke: Predictors of Use, Continuity of Care, and Timeliness of Care. Archives of Physical Medicine and Rehabilitation. 2018;99:1077–89. [DOI] [PubMed] [Google Scholar]
- 6.Roth DL, Sheehan OC, Huang J, et al. Medicare claims indicators of healthcare utilization differences after hospitalization for ischemic stroke: Race, gender, and caregiving effects. International Journal of Stroke. 2016;11(8):928–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Gladman JR, Lincoln NB, Barer DH. A randomised controlled trial of domiciliary and hospital-based rehabilitation for stroke patients after discharge from hospital. J Neurol Neurosurg Psychiatry. 1993;56(9):960–966. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Mas MA, Inzitari M. A critical review of early supported discharge for stroke patients: from evidence to implementation into practice. Int J Stroke 2015; 10: 7–12. [DOI] [PubMed] [Google Scholar]
- 9.The ATTEND Collaborative Group. Family-led rehabilitation after stroke in India (ATTEND): a randomised controlled trial. Lancet 2017;390(10094):588–599. [DOI] [PubMed] [Google Scholar]
- 10.Farley JE, Ndjeka N, Kelly AM, et al. Evaluation of a nurse practitioner-physician task-sharing model for multidrug-resistant tuberculosis in South Africa. PLoS ONE 2017;12(8):e0182780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bolkan HA, van Duinen A, Waalewijn B, et al. Safety, productivity and predicted contribution of a surgical task-sharing programme in Sierra Leone. BJS 2017;104:1315–1326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Oldenkamp M, Bültmann U, Wittek RPM, Stolk RP, Hagedoorn M, Smidt N. Combining informal care and paid work: The use of work arrangements by working adult-child caregivers in the Netherlands. Health Soc Care Community. 2018;26(1):e122–e131. [DOI] [PubMed] [Google Scholar]
- 13.Oliva-Moreno J, Peña-Longobardo LM, Mar J, et al. Determinants of Informal Care, Burden, and Risk of Burnout in Caregivers of Stroke Survivors. The CONOCES Study. Stroke. 2018;49(1):140–146. [DOI] [PubMed] [Google Scholar]
- 14.Haley WE, Allen LY, Grant JS, Clay OJ, Perkins M, Roth DL. Problems and benefits reported by stroke family caregivers: Results from a prospective epidemiological study. Stroke 2009;40:2129–2133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Haley WE, Roth DL, Hovater M, Clay OJ. Long-term impact of stroke on family caregiver well-being: A population based case-control study. Neurology. 2015;84:1323–1329. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Haley WE, Roth DL, Kissela B, et al. Quality of life after stroke: a prospective longitudinal study. Qual Life Res 2011;20(6):799–806. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Howard VJ, Cushman M, Pulley L, et al. The reasons for geographic and racial differences in stroke study: objectives and design. Neuroepidemiology. 2005;25(3):135–143. [DOI] [PubMed] [Google Scholar]
- 18.Perkins M, Howard VJ, Wadley VG, et al. Caregiving strain and all-cause mortality: evidence from the REGARDS study. J Gerontol B Psychol Sci Soc Sci 2013; 68(4):504–512. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Chai YC, Mahadevan R, Ng CG, Chan LF, Dai FM. Caregiver depression: The contributing role of depression in patients, stigma, soial support and religiosity. Int J Soc Psychiatry. 2018;64(6):578–8. [DOI] [PubMed] [Google Scholar]
- 20.Cameron JI, Cheung AM, Streiner DL, Coyte PC, Stewart DE. Stroke survivor depressive symptoms are associated with family caregiver depression during the first 2 years poststroke. Stroke. 2011;42:302–306. [DOI] [PubMed] [Google Scholar]
- 21.McCarthy MJ, Bakas T, Schellinger J, Stapleton K, Kissela BM. Association between incongruence about survivor function and outcomes among stroke survivors and family caregivers. Top Stroke Rehabil 2018;25(8):569–575. [DOI] [PubMed] [Google Scholar]
- 22.Hunt K, Adamson J, Hewitt C, et al. Do women consult more than men? A review of gender and consultation for back pain and headache. Journal of Health Services Research and Policy, 2011. 16(2):109–117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.McCurley JL, Funes CJ, Zale EL, et al. Preventing Chronic Emotional Distress in Stroke Survivors and Their Informal Caregivers. Neurocrit Care. 2019;30(3):581–589. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Vloothuis JDM, Mulder M, Nijland RHM, et al. Caregiver-mediated exercises with e-health support for early supported discharge after stroke (CARE4STROKE): A randomized controlled trial. PLoS One 2019;14(4):e0214241. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available on request from the Reasons for Geographic and Racial Differences in Stroke executive committee. The data are not publicly available due to their containing information that could compromise the privacy of research participants.