Home Telehealth and Caregiving Appraisal in Chronic Illness

Abstract

Background: Remote health monitoring applications are being adopted to improve the health of chronically ill individuals. Little work has focused on the effects of these technologies on informal caregivers (CG) of patients with chronic illnesses. Objectives: To examine differences in caregiving appraisal between CG of enrolled and nonenrolled Veterans in the home telehealth (HT) program. Methods: Cross-sectional survey methodology in 244 dyads (Veteran and CG) from 6 rural Midwestern Veterans Affairs Medical Centers. Survey variables were derived from the 2004 National Alliance for Caregiving survey, along with measures of caregiving strain, burden, and satisfaction. Results: We found no differences when comparing HT and non-HT CG. In multivariate analyses combining the two groups, CG characteristics associated with CG strain included younger age, providing help with activities of daily living and instrumental activities of daily living, use of coping skills, depressive symptoms, and less use of unpaid help (all p ≤ 0.001). Burden was associated with CG use of coping skills, caregiving confidence, and relationship quality with the Veteran (all p < 0.0001). CG satisfaction was associated with presence of social support (p < 0.0001). High CG strain was associated with Veteran hospitalization in the combined group (p = 0.03). Burden (p = 0.0002) was significantly associated with CG satisfaction. Discussion: Existing HT infrastructure provides an opportunity to incorporate training and support programs for CG of chronically ill patients. Such programs could improve CG confidence and use of positive coping skills, lower strain and burden, and potentially improve the health of both the care recipient and CG.

Introduction

Chronic illness management involves a number of daily management tasks that must be incorporated into one's life. Especially for older individuals with chronic illness, much of this care is coordinated or delivered by family members, referred to as informal caregivers (CG).¹ Much of the current literature on informal caregiving has focused on CG strain and burden^2

–6 in dementia. More recent work has recognized that individuals with chronic illnesses such as diabetes, heart failure (HF), and lung disease also rely on informal CG who provide a significant amount of support and care.^7

–10 Informal CG assist with medication management, schedule appointments, provide transportation, perform or assist with activities of daily living (ADL), and assess their family member's health status between medical appointments. A majority of family members willingly help with chronic illness management,¹¹and social support can have a significant positive effect on patient outcomes.^12,13 While caregiving can be satisfying, informal CG can also experience opportunity costs (e.g., lost wages¹⁴) and may suffer their own health decline due to the burden of caregiving.⁶

Remote and mobile health monitoring applications are being promoted to improve the health of chronically ill individuals and facilitate chronic illness management outside of formal healthcare settings. To address the needs of a large geographically dispersed population with high rates of chronic illness, the Veterans Health Administration (VA) has implemented a home telehealth (HT) program nationally. Almost 90,000 Veterans are enrolled in the HT program nationally¹⁵; the majority of these patients are monitored for diabetes, hypertension, or HF.¹⁶ Using the HT technology, these programs focus on health education, behavior management, and symptom recognition and control. Peripheral devices, such as blood pressure cuffs, can connect directly to devices for easy data downloading and transmission to a nurse. Frequently, especially for older Veterans, family members assist with data collection and entry in these devices, as well as monitoring health status between healthcare visits. In an earlier study evaluating HT in Veterans with HF, anecdotal evidence indicated a strong influence of family members (typically the spouse) on whether the Veteran agreed to and participated in the trial, for example, if the spouse felt the program would be helpful, the Veteran was more likely to participate.¹⁷ Thus, we believe that the family unit evaluates programs such as HT in terms of the effect on the whole family, not just the Veteran alone.

Little work has addressed the potential burden of informal CG of patients with chronic illnesses enrolled in HT programs. Integrating an out-of-home support person for patients with diabetes enrolled in an HT program significantly improved patient adherence,^18,19 however, the study did not address burden on the support person. Incorporating an out-of-home support person for patients with HF strengthened the relationship between patient and partner²⁰ and decreased strain and depression in the CG.²¹ Our prior work examined caregiving in Veterans enrolled in the HT program. We found that over one-third of informal CG in that sample reported high strain and a moderate level of caregiving satisfaction.²² However, we did not compare them to Veterans with chronic illnesses, who were not enrolled in HT. Thus, it is unclear whether programs that provide additional monitoring in the home improve or exacerbate caregiving strain and burden for in-home CG.

As noted, our prior study only included patients who were enrolled in the HT program; Veterans not enrolled in the HT program were not included in the sample. The purpose of this exploratory study was to examine differences in strain, burden, and satisfaction between CG of Veterans enrolled in the HT program and CG of Veterans who are not enrolled in the HT program. A secondary aim was to examine the relationship between CG' reports of strain/burden and Veteran use of VA healthcare resources and CG satisfaction.

Methods

Conceptual Framework

The conceptual model guiding this study is adapted from Kramer²³ who builds on work drawn from theoretical and conceptual models of stress, social exchange, role theory, and work motivation. The caregiving context (characteristics of the care recipient [CR], CG, and caregiving demands) plays a central role in understanding all aspects of the caregiving experience. CG-specific resources also play a central role in understanding CG outcomes and are hypothesized to be important in explaining the variation in strain, burden, and satisfaction found among CG. These resources can be internal, for example, coping skills, or external (e.g., social support or tangible resources such as income). Of interest in this study is if health system resources (i.e., HT programs) directly influence responses to caregiving (strain, burden, satisfaction, and healthcare utilization).

Design

The study was approved by the University of Iowa Institutional Review Board and local VA Research Committee. The study reported here used the same design, survey instrument, recruitment, and enrollment approach as our prior study.²² This study used a cross-sectional design and survey methodology to examine CG strain, burden, caregiving satisfaction, Veteran hospitalization, and Veteran and CG characteristics associated with CG' responses to caregiving.

Setting and Sample

Veterans who were Primary Care enrollees of the regional VA healthcare system located in Minnesota, Iowa, western Illinois, North Dakota, South Dakota, and Nebraska were eligible to enroll. All Veteran participants in this study had a diagnosis of either diabetes mellitus (DM) or HF and identified an informal CG who agreed to participate; spoke English; and had a working telephone. CG were relatives or friends 18 years of age or older; they could either live with the Veteran or visit regularly (out of home CG), and provided unpaid help to the Veteran.

Data Collection

Names and contact information of Veterans enrolled in the HT program for DM or HF were obtained from VA telehealth databases. Once HT enrollees were enrolled, the Veterans Health Information Systems and Technology Architecture (VistA) database was queried to find (up to 30) potential non-HT Veteran matches for each completed HT Veteran interview. Veterans were matched on facility, age category (±5 years), and disease (DM or HF).

Data were collected using structured telephone interviews from September 2009 through March 2010. For both groups, the Veteran received an invitation letter that described the study. If both the Veteran and CG were interested in participating, they returned a form that included their contact information. Trained research assistants interviewed the Veteran and CG separately by telephone. Each participant received a $10 shopping card mailed after they completed the survey ($20/dyad).

Measures

As in our prior study, survey questions were derived from the 2004 National Alliance for Caregiving (NAC) survey report “Caregiving in the US” (www.caregiving.org/pdf/research/04questionnaire.pdf), along with other standardized instruments, described in detail in our earlier report.²² The NAC survey contains 98 questions in 8 domains: characteristics of the relationship between the CR and CG; characteristics of recipient needs; medications; other support; stress on working CG; physical, emotional, and financial stress; information services; and demographics. We used 48 items from the NAC survey for our CG survey and 39 items for our Veteran survey (reworded slightly to address the question from the CR perspective).

Demographic characteristics include age in years, gender, marital status, years of formal education, race/ethnicity, and current employment status. CR-CG relationship was measured by asking the person's relationship (e.g., spouse, child). A single-item screen using a 5-point scale was used to measure self-rated health: “In general, how would you rate your health? Excellent, very good, good, fair, or poor?” This single item has been shown to have a strong association with mortality even after adjustment of key covariates.²⁴ Depression was measured using the Geriatric Depression Scale (GDS) Short Form, which contains 15 statements assessing depressed mood²⁵ (Cronbach's alpha 0.54). Respondents answer each question with a “yes” or “no” for how they have felt over the past week. Used in VA patients, the GDS has a sensitivity of 92%, specificity of 89%, and a negative predictive value of 99%. A score >5 points is suggestive of depression; scores >10 indicate depression.

Coping style was assessed using a series of eight questions from the NAC survey asking about use of coping strategies. A coping style score was calculated by summing “yes” responses to each question, resulting in a possible score range of 0–8 (with higher scores indicating greater use of coping strategies). Relationship quality was assessed by asking both the Veteran and informal CG to rate, using a 5-point scale, the quality of the relationship between the CR and CG. A single question asked only the CG whether they felt they had a choice in taking on the informal care role for the Veteran.

Regarding caregiving assistance, the Veteran was asked how much help they needed and the CG were asked how much help they provided to the Veteran for 6 ADL, 12 instrumental activities of daily living (IADL), medications, and other assistance. CG were asked to estimate how many hours per week were spent helping the Veteran. CG resources and support were assessed, including confidence in providing assistance (“How confident are you in your ability to provide assistance to [care recipient]?”), use of paid or unpaid assistance, and social support. Social support was measured using the Personal Resource Questionnaire (PRQ2000)²⁶ (Cronbach's alpha 0.67), which includes 15 questions scored on a 7-point Likert scale where 1 = strongly disagree to 7 = strongly agree; responses are summed to calculate a total score. Reliability estimates across eight studies ranged from 0.87 to 0.93. The range of total scores is 15–105, with higher sores indicating higher levels of perceived social support.

Outcome variables included number of hospitalizations in the past 12 months (Veteran) collected from VA administrative datasets, CG strain using the Caregiver Strain Index (CSI),²⁷ caregiving burden using the Zarit Burden Inventory (ZBI) Short Version,²⁸ and caregiving satisfaction using the Caregiving Satisfaction Scale (CSS).^29,30 The CSI measures employment, financial, physical, social, and time strain on the CG³¹ and contains 13 items, each scored as “yes” or “no.” Higher scores indicate a greater level of strain.²⁷ The ZBI is widely used in dementia caregiving research. Three dimensions of burden are represented: effect on social and personal life; psychological burden; and feelings of guilt.³² Bedard et al.²⁸ validated the shorter 12-item version in a sample of 413 CG of community-dwelling older adults with cognitive impairment. Cronbach's alpha was 0.88 and correlations between the short and full versions were excellent (0.96 and 0.97). A score of 17 or higher indicates high burden. The CSS includes 11 positive aspects of caregiving scored on a 5-point Likert-type scale where 0 = agree a lot and 4 = disagree a lot. Scores range from 0 to 44, with higher scores indicating less satisfaction with caregiving. In a sample of 1,229 CG of persons with Alzheimer's disease, Cronbach's alpha was 0.89³⁰ In this study, Cronbach's alpha for the CSI, ZBI, and CSS was 0.84, 0.87, and 0.94, respectively.

Data Analysis

First, bivariable analyses compared sociodemographic, relationship, health status, and support characteristics, and outcomes of CG strain, burden, satisfaction, and Veteran hospitalizations between HT and non-HT participants. Subsequently, we conducted multivariable analyses to examine differences in outcomes, while controlling for important Veteran and CG demographic, relationship, health status, and support characteristics that may confound the relationship between HT and outcomes. To construct the multivariable models, we first examined bivariable relationships between each Veteran and CG characteristic and outcomes of CG strain, burden, and satisfaction, and Veteran hospitalizations. Spearman's rank correlation coefficient and two-sample t tests were used for ordinal and continuous dependent variables, respectively. To find an optimal set of predictors for each outcome measure, we then used a bootstrapping procedure in which 1,000 random samples were generated with replacement. The 1,000 samples were used to develop separate linear regression (for CG strain, burden, and satisfaction), or logistic regression (for Veteran hospitalization) models for each outcome using stepwise variable selection. For each outcome, variables that were statistically significant in at least 60% of the models were identified. Once the final models were derived, final multivariable models were generated using the original sample, the HT participation indicator, and the selected participant characteristics identified through the bootstrapped samples. Finally, we examined relationships between the Veteran and CG outcome variables using linear regression. The significance level for all analyses was set at p ≤ 0.05.

Results

In the HT group, of the 998 Veterans contacted, 30 were deceased, 49 reported not having a CG, and 31 letters were undeliverable. Of the remaining 888, 506 did not return the response form after 2 mailings. Of the 382 who returned the form, 37.6% agreed to participate. Of these, 129 Veterans and 125 CG completed the survey, resulting in 121 dyads (where both the Veteran and CG completed the survey). In the non-HT group, of the 2,376 Veterans contacted, 64 were deceased, 244 reported not having a CG (did not have anyone or need anyone), 55 reported hearing problems precluding a telephone survey, and 41 letters were undeliverable. Of the remaining 1,972, 1,193 did not return the response form after 2 mailings. Of the 779 who returned the form, 18.9% (n = 148 dyads) agreed to participate. Of these, 128 Veterans and 129 CG completed the survey, resulting in 123 dyads (where both the Veteran and CG completed the survey).

Sample Description

Demographic characteristics and significant differences between HT and non-HT CG and Veterans on participant characteristics and outcomes are presented in Table 1. CG strain, burden, and satisfaction did not differ significantly for HT and non-HT participants in these unadjusted analyses. In contrast, HT Veterans were significantly more likely to have one or more hospitalizations, compared to non-HT Veterans (p = 0.001).

Table 1.

Participant Characteristics and Outcome Variables for Home Telehealth and Non-Home Telehealth Veterans and Caregivers

	HT VETERAN (N = 121)	NON-HT VETERAN (N = 123)	p	HT CG (N = 121)	NON-HT CG (N = 123)	p
Participant characteristics
Age in years, mean (SD)	69.4 (10.8)	70.4 (9.4)	0.45	63.3 (11.7)	64.9 (10.3)	0.25
Gender (male), %	97.5	91.9	0.05	95.0	97.6	0.29
Married	80.0	87.8	0.33	86.8	92.7	0.56
Education, %			0.25			0.59
Less than 12 years	15.8	8.9		4.1	6.5
Completed high school	32.5	33.3		39.7	35.0
Post-high school education/training	51.7	57.7		56.2	58.5
Race, %			0.40			0.32
Caucasian	98.0	95.8		99.2	96.8
Black	0.8	3.3		0.0	1.6
Other	0.8	0.8		0.8	1.6
ADL, mean (SD)	3.0 (3.5)	2.3 (3.0)	0.009	2.7 (3.2)	2.2 (2.7)	0.19
IADL, mean (SD)	7.8 (2.7)	6.9 (2.5)	0.06	7.6 (2.7)	7.1 (2.3)	0.16
Depression, mean (SD)	4.6 (3.4)	3.1 (2.9)	0.0001	2.1 (2.0)	1.9 (2.1)	0.43
Help assistance, mean (SD)				3.6 (1.2)	3.2 (1.2)	0.006
Provide medical support (yes), %	55.8	30.9	<0.0001	41.3	26.2	0.01
Taking any kind of medication (yes), %	52.5	39.8	0.05
Talking with a VA nurse, doctor, or social worker (yes), %	89.2	71.5	0.0006
Cigarette use in the past year (yes), %	20.0	8.9	0.01
Currently working (yes), %	11.7	22.0	0.03
Help with medical equipment (yes), %				32.2	19.5	0.03
Unpaid help in last 12 months (yes), %				36.4	18.7	0.003
Paid help in last 12 months (yes), %				24.0	9.0	0.002
Children/grandchildren in house under 18 (yes), %				14.9	6.5	0.04
Outcomes
CG strain, mean (SD)				4.6 (3.6)	4.0 (4.0)	0.19
CG burden, mean (SD)				5.8 (5.2)	5.2 (5.3)	0.35
Hospitalized past 12 months (yes), %	37.8	18.7	0.001	—	—
Caregiving satisfaction, mean (SD)	—	—	—	34.2 (5.2)	35.2 (6.1)	0.17

ADL, activities of daily living; CG, caregiver; HT, home telehealth; IADL, instrumental activities of daily living; SD, standard deviation; VA, Veterans Health Administration.

Bivariate Analyses: Participant Characteristics and Outcomes

In bivariate analyses of the relationship between participant characteristics and outcomes, variables significantly associated with CG strain included CG age, providing help with ADL and IADL, use of coping skills, depression, use of unpaid help, social support, and self-reported health status. Veteran variables include age, depression, and needing help with ADL. CG variables significantly associated with burden included CG age, providing help with ADL and IADL, use of coping skills, depression, confidence in skills, using unpaid help, and health status. Veteran depression was associated with burden. CG variables significantly associated with satisfaction include depression, confidence in caregiving skills, self-rated health, and social support. There were no Veteran variables associated with caregiving satisfaction. Finally, other than HT status, no participant characteristics were associated with Veteran hospitalization in the past 12 months (Table 2).

Table 2.

Bivariate Relationship Between Veteran and Caregiver Characteristics and Strain, Burden, and Satisfaction

CHARACTERISTIC	STRAIN, BETA (95% CI), p	BURDEN, BETA (95% CI), p	SATISFACTION, BETA (95% CI), p
CG outcomes
CG age	−0.07 (−0.11 to 0.03), 0.0002	−0.07 (−0.13 to 0.007), 0.03	NS
Veteran age	−0.04 (−0.09 to −0.001), 0.04	NS	NS
CG providing help with ADL	0.42 (0.29 to 0.56), 0.0001	0.27 (0.05 to 0.49), 0.02	NS
CG providing help with IADL	0.47 (0.30 to 0.63), 0.0001	0.28 (0.02 to 0.54), 0.03	NS
CG coping	0.80 (0.60 to 1.00), 0.0001	0.68 (0.34 to 1.0), 0.0001
CG depression	0.73 (0.54 to 0.93), 0.0001	0.69 (0.37 to 1.00), <0.001	−0.68 (−1.03 to −0.34), 0.001
CG confidence	NS	−3.69 (−5.50 to 1.89), <0.001	2.90 (0.90 to 4.90), 0.004
CG self-reported health	0.75 (0.30 to 1.20), 0.001	0.69 (0.007 to 1.37), 0.05	−1.07 (−1.81 to 0.33), 0.005
CG social support	−0.04 (−0.08 to −0.004), 0.029	NS	0.18 (0.12 to 0.24), 0.0001
Unpaid help in last 12 months	2.25 (1.30 to 3.20), 0.0001	1.99 (0.53 to 3.45), 0.008	NS
Veteran depression	0.41 (0.28 to 0.53), 0.0001	0.27 (0.07 to 0.46), 0.009	NS
Veteran needing help with ADL	0.28 (0.15 to 0.41), 0.0001	NS	NS

95% CI, 95% confidence interval; NS, not significant.

Multivariable Analyses

Final risk adjustment models are shown in Table 3. In the multivariable models, CG strain was higher in younger CG, those providing help with ADL and IADL, CG use of coping skills, depressive symptoms, and less use of unpaid help. CG burden increased with CG use of coping skills and decreased with CG confidence and CG-reported relationship quality with the Veteran. Two CG characteristics, relationship quality and greater social support, were significantly associated with satisfaction. Other than HT status, no variables were associated with Veteran hospitalization (beta 0.19; confidence interval 0.07–0.30; p < 0.002).

Table 3.

Veteran and Caregiver Characteristics Associated with Strain, Burden, and Satisfaction

CHARACTERISTIC	STRAIN, BETA (95% CI), p	BURDEN, BETA (95% CI), p	SATISFACTION, BETA (95% CI), p
CG outcomes
CG age	−0.05 (−0.08 to −0.02), 0.003	NS	NS
CG providing help with ADL	0.21 (0.09 to 0.33), 0.001	NS	NS
CG providing help with IADL	0.22 (0.09–0.37), <0.0013	NS	NS
CG coping	0.51 (0.34–0.69), <0.0001	0.63 (0.33–0.95), <0.0001	NS
CG depression	0.53 (0.33–0.66), <0.0001	NS	NS
CG confidence	NS	−3.23 (−4.92 to 1.54), <0.0001	NS
Quality of relation with veteran	NS	−2.79 (−4.12 to −1.46), <0.0001	2.99 (1.49 to 4.50), 0.0001
CG social support	NS	NS	0.14 (0.09 to 0.20), <0.0001
Unpaid help in last 12 months	−1.67 (−2.41 to −0.94), 0.0001	NS	NS

We then examined the relationship between CG' reports of strain or burden and Veteran hospitalization and CG satisfaction. Combining the HT and non-HT groups, both strain (p = 0.002) and burden (p = 0.04) were significantly associated with Veteran hospitalization in the past 12 months. To assess the effect of HT, we ran a models that also included strain, burden, and HT status. Only strain was associated with hospitalization (p = 0.03) Veterans enrolled in HT were more likely to have been hospitalized in the past 12 months (p = 0.03) even after controlling for strain. We also examined whether higher levels of CG strain and burden were associated with satisfaction with caregiving role. In separate linear regression models, both strain (p = 0.02) and burden (p = 0.03) were significantly associated with CG satisfaction. CG with the highest strain scored an average of 3.5 points lower on the satisfaction score, while CG with the highest burden scored an average of 4 points lower, although it is not clear that these are clinically meaningful results. In models that also included HT status, HT was not significantly associated with CG satisfaction.

Discussion

This study sought to examine whether there are differences in strain, burden, and caregiving satisfaction between CG of Veterans enrolled in the HT program and CG of chronically ill Veterans who are not enrolled in the HT program. We found no differences when comparing HT and non-HT CG in this study. Several characteristics in both bivariate and multivariate models were associated with CG strain, burden, and satisfaction in the overall group, but few Veteran-reported characteristics predicted these outcomes. Veterans enrolled in the HT program were more likely to have had a hospitalization in the past 12 months, compared to non-HT Veterans, although this finding is not surprising given the HT program enrollment criteria. High CG strain was associated with hospitalization in the combined group. Both strain and burden were significantly associated with CG satisfaction.

In our prior work, we conducted the same survey in a group of Veterans enrolled in the HT program.²² That study differed from this study in a number of ways: non-HT Veterans were not included; the sample was composed primarily of Veterans with DM (61%) and HF (19%), but other diagnoses were included; burden was not measured; and the HT program served a different geographic region than this study. Nevertheless, both samples were similar in demographics. In this and our prior study, higher strain was associated with CG depression, use of coping strategies, and receiving less unpaid help from family and friends. Veteran dependency in IADL was a predictor of CG strain in both studies. In the study reported here, burden (which was not measured in our prior study) was associated with greater use of coping strategies, decreased confidence in caregiving skills, and lower relationship quality with the veteran. Greater social support was associated with caregiving satisfaction in both studies.

Due to the historical legal restrictions on including the family in care services delivered by the VA, very little CG research has been conducted in users of VA healthcare. Legislation has been passed (Public Law: 111–163; Caregivers and Veterans Omnibus Health Services Act of 2010) to develop and implement services for Veteran's informal CG. The Program of Comprehensive Assistance for Family Caregivers is directed toward severely injured OEF/OIF/OND Veterans who need assistance and provides a broad range of support for both the Veteran and the informal CG, including such things as stipends, training, travel assistance, respite care, and mental health services. The less expansive Program of General Caregiver Support Services provides services for informal CG such as training and education. Information about these services is available on a publicly available Web site (www.caregiver.va.gov). However, the primary focus of information on this site is on Veterans with such diagnoses as post-traumatic stress disorder, traumatic brain injury, Alzheimer's, multiple sclerosis, and Parkinson's, which may or may not meet the needs of CG of Veterans with chronic conditions such as HF or DM. In part, this may be that the term “caregiver” is not often recognized as valid when referring to individuals who help patients manage their chronic illness. In both of our surveys, many Veterans declined to participate because they did not “need” a CG, although they described many activities their (usually) spouse did to help them.

Limitations of this study include the low participation rate, particularly among the non-HT Veteran group. We did not incorporate a severity of illness measure, and because HT Veterans had higher hospitalization rates, there may have been differences between the two groups in the severity of their diabetes or HF. The cross-sectional design limits our understanding of the trajectory of CG strain, burden, and satisfaction when caring for chronically ill family members. This study was conducted in a population of patients who use VA healthcare services, with a majority male population and the majority identified female CG. In a recent survey of a nationally representative sample of U.S. adults who were “potential” CG, 70% were female. However, only slightly more than half of those who actually provided support (“current disease management supporter”) were female (53%).¹¹ Furthermore, our sample was more than 95% Caucasian, and reported higher rates of post-secondary education than the Rosland et al. study¹¹ (>50% vs. 24%), thus our sample may not be representative of the general population of CG and CR in the United States. Finally, although we attempted to interview the CR and CG separately, on occasions where both were in the same room during the interview, we cannot estimate how that might have altered individual responses to questions.

Recognition of the role of CG in chronic illness is increasing, yet current practice does little to recognize the potential adverse effects on the CG. Little research has addressed family CG needs in adult patients with type 2 diabetes. Research that includes families or that test family interventions focus on outcomes in the patient with diabetes, rather than the family.^19,33,34 Far more work has been conducted in CG of patients with HF. A recent review identified needs of these CG as a need for “normalcy” and a social life, support for the daily needs of the CR, and support from the healthcare system, including tailored information and meeting information needs.¹⁰ Consistent with our findings, Grigorovich et al.³⁵ found that depressive symptoms and lower social support/engagement in social activities were associated with poor emotional outcomes in CG. Thus, a number of interventions for CG should be considered. CG should be encouraged to attend clinic visits with the CR³⁶; this provides an opportunity for the CR's clinicians to interact with individuals who may be helping the CR with illness management. As appropriate, CG can then be assessed for risk factors, for example, living with the CR, depressive symptoms, decreased social activity, and/or loss of employment due to caregiving demands. A question such as “what do you need so that you can care better for [CR]?” may then open the door for referrals to community services and legitimize the needs of the CG.³⁷ CG also need in-depth training to obtain skills necessary to support the CR in needed behavior changes,^11,38 which cannot be provided within the time limits of a primary care clinic visit, so community referrals are essential.

The VA has been an early adopter and embraced the use of technology to provide care for Veterans, including the electronic medical record, home- and clinic- based telemedicine, and a patient portal that incorporates a number of functions such as secure messaging. Using existing infrastructure for rural and remote settings, such as video to home, training, and assistance for CG of individuals with chronic illnesses like HF could be delivered to improve the health of both the CR and CG, and may lower strain and burden among CG. These programs could guide family members in goal setting, supportive communication strategies, and tools to assist with symptom monitoring and medication management.^21,33,39 Such programs could assist with improving CG confidence and use of positive coping skills. Given the impact of family support on CR outcomes, it is critical to expand our understanding of the needs of all CG and to begin to address those needs.

Footnotes

Disclosure Statement

No competing financial interests exist.

References

Salyers

, White

, Zubritsky

, Osin

. Family involvement in the care of healthy medical outpatients. Fam Pract, 2006; 23:317–324.

Zarit

, Femia

. Behavioral and psychosocial interventions for family caregivers. Am J Nurs, 2008; 108(Suppl 9):47–53; quiz 53.

Fitting

, Rabins

, Lucas

, Eastham

. Caregivers for dementia patients: A comparison of husbands and wives. Gerontologist, 1986; 26:248–252.

Zarit

, Todd

, Zarit

. Subjective burden of husbands and wives as caregivers: A longitudinal study. Gerontologist, 1986; 26:260–266.

Zarit

, Femia

. A future for family care and dementia intervention research? Challenges and strategies. Aging Ment Health, 2008; 12:5–13.

Pinquart

, Sorensen

. Correlates of physical health of informal caregivers: A meta-analysis. J Gerontol B Psychol Sci Soc Sci, 2007; 62:P126–P137.

Bryant

, Mansfield

, Boyes

, Waller

, Sanson-Fisher

, Regan

. Involvement of informal caregivers in supporting patients with COPD: A review of intervention studies. Int J COPD, 2016; 11:1587–1596.

Chen

, Chiou

, Yu

, et al. The unmet supportive care needs—What advanced lung cancer patients' caregivers need and related factors. Support Care Cancer, 2016; 24:2999–3009.

Awadalla

, Ohaeri

, Al-Awadi

, Tawfiq

. Diabetes mellitus patients' family caregivers' subjective quality of life. J Natl Med Assoc, 2006; 98:727–736.

10.

Doherty

, Fitzsimons

, McIlfatrick

. Carers' needs in advanced heart failure: A systematic narrative review. Eur J Cardiovasc Nurs, 2014; 15:203–212.

11.

Rosland

, Heisler

, Janevic

, et al. Current and potential support for chronic disease management in the United States: The perspective of family and friends of chronically ill adults. Fam Syst Health, 2013; 31:119–131.

12.

Rosland

, Kieffer

, Spencer

, et al. Do pre-existing diabetes social support or depressive symptoms influence the effectiveness of a diabetes management intervention?. Patient Educ Couns, 2015; 98:1402–1409.

13.

Graven

, Grant

. Social support and self-care behaviors in individuals with heart failure: An integrative review. Int J Nurs Stud, 2014; 51:320–333.

14.

Van Houtven

, Coe

, Skira

. The effect of informal care on work and wages. J Health Econ, 2013; 32:240–252.

15.

Darkins

, Foster

, Anderson

, Goldschmidt

, Selvin

. The design, implementation, and operational management of a comprehensive quality management program to support national telehealth networks. Telemed J E Health, 2013; 19:557–564.

16.

Darkins

, Ryan

, Kobb

, et al. Care Coordination/Home Telehealth: The systematic implementation of health informatics, home telehealth, and disease management to support the care of veteran patients with chronic conditions. Telemed J E-Health, 2008; 14:1118–1126.

17.

Wakefield

, Holman

, Ray

, et al. Outcomes of a home telehealth intervention for patients with heart failure. J Telemed Telecare, 2009; 15:46–50.

18.

Aikens

, Zivin

, Trivedi

, Piette

. Diabetes self-management support using mHealth and enhanced informal caregiving. J Diabetes Complications, 2014; 28:171–176.

19.

Aikens

, Trivedi

, Aron

, Piette

. Integrating support persons into diabetes telemonitoring to improve self-management and medication adherence. J Gen Intern Med, 2015; 30:319–326.

20.

Piette

, Striplin

, Marinec

, et al. A mobile health intervention supporting heart failure patients and their informal caregivers: A randomized comparative effectiveness trial. J Med Internet Res, 2015; 17:e142.

21.

Piette

, Striplin

, Marinec

, Chen

, Aikens

. A randomized trial of mobile health support for heart failure patients and their informal caregivers: Impacts on caregiver-reported outcomes. Med Care, 2015; 53:692–699.

22.

Wakefield

, Hayes

, Boren

, Pak

, Davis

. Strain and satisfaction in caregivers of veterans with chronic illness. Res Nurs Health, 2012; 35:55–69.

23.

Kramer

. Gain in the caregiving experience: Where are we? What next?. Gerontologist, 1997; 37:218–232.

24.

DeSalvo

, Fan

, McDonell

, Fihn

. Predicting mortality and healthcare utilization with a single question. Health Serv Res, 2005; 40:1234–1246.

25.

Sheikh

, Yesavage

. Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. In: Brink

, ed. Clinical gerontology: A guide to assessment and intervention. New York: The Hayworth Press, 1986:165–173.

26.

Weinert

. Measuring social support: PRQ2000. In: Strickland

, DiIorio

, eds. Measurement of nursing outcomes: Vol. 3. New York: Springer, 2003:161–172.

27.

Robinson

. Validation of a caregiver strain index. J Gerontol, 1983; 38:344–348.

28.

Bedard

, Molloy

, Squire

, Dubois

, Lever

, O'Donnell

. The Zarit Burden Interview: A new short version and screening version. Gerontologist, 2001; 41:652–657.

29.

Lawton

, Kleban

, Moss

, Rovine

, Glicksman

. Measuring caregiving appraisal. J Gerontol, 1989; 44:P61–P71.

30.

Tarlow

, Wisniewski

, Belle

, Rubert

, Ory

, Gallagher-Thompson

. Positive aspects of caregiving: Contributions of the REACH Project to the development of new measures for Alzheimer's caregiving. Res Aging, 2004; 26:429–453.

31.

Sullivan

. Caregiver strain index. J Gerontol Nurs, 2002; 28:4–5.

32.

Ankri

, Andrieu

, Beaufils

, Grand

, Henrard

. Beyond the global score of the Zarit Burden Interview: Useful dimensions for clinicians. Int J Geriatr Psychiatry, 2005; 20:254–260.

33.

Armour

, Norris

, Jack

Jr , Zhang

, Fisher

. The effectiveness of family interventions in people with diabetes mellitus: A systematic review. Diabet Med, 2005; 22:1295–1305.

34.

, Wallace

, McCoy

, Amirehsani

. A family-based diabetes intervention for Hispanic adults and their family members. Diabetes Educ, 2014; 40:48–59.

35.

Grigorovich

, Lee

, Ross

, Woodend

, Forde

, Cameron

. A longitudinal view of factors that influence the emotional well-being of family caregivers to individuals with heart failure. Aging Ment Health, 2016; 1–7.

36.

Rosland

, Piette

, Choi

, Heisler

. Family and friend participation in primary care visits of patients with diabetes or heart failure: Patient and physician determinants and experiences. Med Care, 2011; 49:37–45.

37.

Adelman

, Tmanova

, Delgado

, Dion

, Lachs

. Caregiver burden: A clinical review. JAMA, 2014; 311:1052–1059.

38.

Van Houtven

, Voils

, Weinberger

. An organizing framework for informal caregiver interventions: Detailing caregiving activities and caregiver and care recipient outcomes to optimize evaluation efforts. BMC Geriatr, 2011; 11:77.

39.

Rosland

A-M

, Piette

. Emerging models for mobilizing family support for chronic disease management: A structured review. Chronic Illn, 2010; 6:7–21.