Extensive research has identified substantial needs for care in patients with serious non-cancer illnesses, such as congestive heart failure (CHF) and chronic obstructive pulmonary disease (COPD), and their families. For Veterans whose goals for care focus exclusively on comfort, hospice offers a valuable set of palliative care services. However, hospice is not an option for patients who wish to receive life-prolonging treatment or who have a prognosis of greater than 6 months; and their needs for palliative care must be met in other ways. For this reason, there is rapidly growing interest in alternatives to hospice that meet the palliative care needs of Veterans living at home with serious illness, without the eligibility restrictions that hospice imposes. Design of ideal programs to meet these needs should be informed by evidence of Veterans' and families' preferences about palliative home care services.
The objectives of this study were: 1) To identify palliative home care services that are most important to patients and families; 2) To identify home care services that have greater value for certain subgroups of patients and families; 3) To identify differences in patients' vs. families' priorities for palliative home care services; and 4) To define unique patterns of priorities for palliative home care services.
This project used telephone interviews with 453 Veterans with CHF and COPD, and 131 family members at the Philadelphia and Greater Los Angeles VAMCs. The interviews used methods of preference assessment to define the importance that Veterans and families place on a range of palliative home care services, and to define subgroups with unique priorities. The list of palliative care options was endorsed by a panel of 10 experts. The value of these care options was assessed by an adaptive conjoint interview. In the adaptive conjoint procedure (implemented in Sawtooth Software), respondents rated the value of each palliative care service and then answered a series of trade-off questions between services to refine the estimates of the value or 'utility' of each service relative to the others. The Mplus version 6 software was used to look for potential 'latent classes' or 'market segments' among Veterans in their preferences for different bundles of related services. The interview also assessed demographic and additional background variables. All other analyses were performed in SAS version 9.2
The 453 Veterans (298 in Philadelphia and 155 in Los Angeles) were mostly males (97%), with a mean age of 69, 89% were college graduates, 42% African-American and 49% Caucasian. The 131 family members (83 Philadelphia and 48 Los Angeles) were 89% females with a mean age of 60, 65% were spouses or partners, 17% were adult children, and 79% of the family members were living with the Veteran.
On the Eastern Cooperative Oncology Group (ECOG) score, 22% were capable of only limited self-care (scores of 3 or higher) and 45% were unable to carry out any work activities despite being ambulatory and capable of self-care (score of 2). The Veteran's health was rated as poor by 28% of family members and 22% of Veterans, and rated as fair by 47% of family members and 43% of Veterans.
Among the 453 Veterans, the highest scoring service was transportation (utility score= 1.27), followed by a meal program (1.11), family care (.96), a nurse available up to five times a week (.96), a care coordinator (.95), peer support (.89), a counselor (.83), reimbursement to someone you chose for home health support (.80), respite care (.73), a chaplain (.52), and someone hired from a health care agency for home health support (.24).
Among the 131 family members, transportation (utility score=1.29) was highest, followed by a nurse available (1.07), family care (1.07), a care coordinator (1.02), a meal program (.90), peer support (.85), a counselor (.81), reimbursement to someone you chose for home health support (.80), respite care (.64), a chaplain (.59), and someone hired from a health care agency for home health support (.34).
Comparing the 131 family members to their matched 131 Veterans using a paired t-test showed that family members had lower utility values on the meal program (.90 vs 1.13, p= .02), but higher values on a chaplain (.59 vs .43, p= .03), and a nurse available up to 5 times a week (1.07 vs .92, p= .04).
Veterans aged 66 or older had higher utilities for having a nurse available (1.01 vs .87, p= .04) and for transportation (1.38 vs 1.13, p= .002). Younger Veterans had higher utilities for a counselor (.93 vs .75, p=.005) and a care coordinator (1.09 vs .84, p <.001). Those who said they did not have enough money to cover expenses had a lower utility for a chaplain (.39 vs .56, p=.016) and a higher utility for transportation (1.43 vs 1.22, p= .025).
Veterans with higher disability (an ECOG score of 3 or 4) showed little difference in utility from those with less disability. When asked to choose during the interview, 47% wanted primarily to extend life, 45% wanted palliative care, and 8% did not have a preference. The utilities for services of those wanting to extend life did not differ from those wanting palliative care.
Latent class analysis on the utilities for Veterans suggested that there are no distinct underlying classes among Veterans; that is, preferences are relatively uniform across the Veterans and they do not need to be treated as separate market segments or offered distinctly different programs for different groups of Veterans. However, family members do have somewhat different emphases than Veterans and Veteran preferences do vary slightly by age and other factors.
In the current environment of limited resources, coupled with an emphasis on patient-centered care, there are several specific ways that the findings from this study can be useful to policy makers and planners, program administrators and staff, and ultimately to Veterans and their families. First, the identified preferences for palliative home care services can guide policy-level decisions about the appropriate roles for hospice and palliative care. Second, our findings regarding patients' and families' priorities for services can help palliative care programs select those services that are most important to Veterans and their families, even when they may not be eligible for traditional hospice services. Third, we identified unique service preferences among older Veterans and those facing financial constraints; and this information can be used to tailor service offerings that incorporate the preferences of these subgroups of patients. Overall, this will enable the VA healthcare system to design and deliver patient-centered palliative home care to Veterans, including those with serious non-cancer illnesses.
External Links for this Project
- Smith D, Caragian N, Kazlo E, Bernstein J, Richardson D, Casarett D. Can we make reports of end-of-life care quality more consumer-focused? results of a nationwide quality measurement program. Journal of palliative medicine. 2011 Mar 1; 14(3):301-7. [view]
- Sudore RL, Casarett D, Smith D, Richardson DM, Ersek M. Family involvement at the end-of-life and receipt of quality care. Journal of pain and symptom management. 2014 Dec 1; 48(6):1108-16. [view]