590 Journal of Pain and Symptom Management Vol. 48 No. 4 October 2014

Original Article

Can Senior Volunteers Deliver Reminiscenceand Creative Activity Interventions? Results ofthe Legacy Intervention Family EnactmentRandomized Controlled TrialRebecca S. Allen, PhD, Grant M. Harris, MA, Louis D. Burgio, PhD,Casey B. Azuero, MPH, MA, Leslie A. Miller, BA, Hae Jung Shin, PhD,Morgan K. Eichorst, MA, Ellen L. Csikai, PhD, Jamie DeCoster, PhD,Linda L. Dunn, DSN, RN, CNL, Elizabeth Kvale, MD, and Patricia Parmelee, PhDCenter for Mental Health and Aging (R.S.A., G.M.H., L.D.B., C.B.A., L.A.M., H.J.S., M.K.E.,

E.L.C., L.L.D., P.P.), Department of Psychology (R.S.A., G.M.H., L.D.B., C.B.A., M.K.E., P.P.),

School of Social Work (L.A.M., E.L.C.) and Capstone College of Nursing (L.L.D.), The University of

Alabama, Tuscaloosa, Alabama, USA; Gwangmyeong Social Welfare Center (H.J.S.), Gwangmyeong

City, Gyeonggido, Republic of Korea; Center for Advanced Study of Teaching & Learning (J.D.),

University of Virginia, Charlottesville, Virginia; and Birmingham-Atlanta Geriatric Research

Education and Clinical Center (GRECC) (E.K.), Veterans Administration Medical Center,

Birmingham, Alabama, USA

Abstract

Context. Palliative care patients and their family caregivers may have a

foreshortened perspective of the time left to live, or the expectation of thepatient’s death in the near future. Patients and caregivers may report distress inphysical, psychological, or existential/spiritual realms.

Objectives. To conduct a randomized controlled trial examining the effectivenessof retired senior volunteers (RSVs) in delivering a reminiscence and creative activityintervention aimed at alleviating palliative care patient and caregiver distress.

Methods. Of the 45 dyads that completed baseline assessments, 28 completedpostintervention and 24 completed follow-up assessments. The interventiongroup received three home visits by RSVs; control group families received threesupportive telephone calls by the research staff. Measures included symptomassessment and associated burden, depression, religiousness/spirituality, andmeaning in life.

Results. Patients in the intervention group reported a significantly greaterreduction in frequency of emotional symptoms (P¼ 0.02) and emotionalsymptom bother (P¼ 0.04) than the control group, as well as improved spiritualfunctioning. Family caregivers in the intervention group were more likely thancontrol caregivers to endorse items on the Meaning of Life Scale (P¼ 0.02). Only

Address correspondence to: Rebecca S. Allen, PhD, Cen-ter for Mental Health and Aging, The University ofAlabama, Box 870315, Tuscaloosa, AL 35487-0315,USA. E-mail: rsallen@ua.edu

Accepted for publication: November 12, 2013.

� 2014 American Academy of Hospice and PalliativeMedicine. Published by Elsevier Inc. All rights reserved.

0885-3924/$ - see front matterhttp://dx.doi.org/10.1016/j.jpainsymman.2013.11.012

Vol. 48 No. 4 October 2014 591Therapy With Patients and Caregivers Near EOL

improvement in intervention patients’ emotional symptom bother maintained atfollow-up after discontinuing RSV contact (P¼ 0.024).

Conclusion. Delivery of the intervention by RSVs had a positive impact onpalliative care patients’ emotional symptoms and burden and caregivers’ meaningin life. Meaningful prolonged engagement with palliative care patients andcaregivers, possibly through alternative modes of treatment delivery such ascontinued RSV contact, may be necessary for maintenance of therapeuticeffects. J Pain Symptom Manage 2014;48:590e601. � 2014 American Academy ofHospice and Palliative Medicine. Published by Elsevier Inc. All rights reserved.

Key Words

Community-based RCT, palliative care dyads, volunteers, mixed methods

IntroductionPalliative care patients and other vulnerable

populations may have a foreshortenedperspective of the time left to live1 and reportdistress in multiple realms. Chochinov et al2

found that palliative care patients’ existentialdistress was associated with patient characteris-tics, but was even more strongly related to spir-ituality and a sense of meaning and purpose.Meaning reconstruction,3,4 spirituality, andreligious coping have been identified as poten-tial therapeutic tools to enhance interventioneffects.5,6 For example, Randall and Bishop6

found that the relationship between religionand valuation of life was partially mediated byforgiveness.

Interventions to improve the lives of pallia-tive care patients and caregivers frequentlyuse reminiscence, or the concept of preservingdignity, as a therapeutic tool.7e13 Throughfocusing individuals’ attention on lifetime ac-complishments, relationships, and values,techniques such as life review/reminiscencefacilitate meaning reconstruction and reducesymptoms of depression.3,14 The most widelystudied of these interventions is Chochinov’sDignity Therapy.10,12 McClement et al13 re-ported that 95% of the bereaved family mem-bers perceived their loved one’s participationin Dignity Therapy as beneficial, 78% reportedthat the ‘‘generativity document’’ helped theirgrief process, and 95% would recommend Dig-nity Therapy to others. In a randomizedcontrolled trial (RCT), Chochinov et al12

found that the Dignity Therapy improved spir-itual well-being and was perceived as morehelpful than standard palliative- or client-centered care.

A Cochrane review of 11 diverse interven-tions targeting palliative caregivers15 foundweak evidence for therapeutic effect and high-lighted the need for greater methodologicalrigor in future studies. In our prior work,7,8 so-cial work and psychology graduate studentswith bachelor’s or master’s degrees weretrained to deliver a three-session, home-basedreminiscence and creative activity interventionto palliative care patients and their family care-givers. The intervention improved palliativepatients’ sense of meaning, reduced their phys-ical symptom burden, and improved their fam-ily caregivers’ levels of stress relative to acontrol group. Fidelity data evaluated by aspecialist in evidence-based treatment out-comes revealed that the accuracy of treatmentdelivery averaged 91.55%.

Although hospice and palliative care socialworkers frequently use reminiscence and crea-tive activities with their patients,16 such inter-ventions need to be more accessible topatients and families transitioning from com-munity, hospital, and palliative care settings. Ifhospice or palliative care is not chosen as atreatment option, few means of delivering ther-apeutic reminiscence-based interventions exist.This represents a significant gap in practiceand in the psychosocial palliative care interven-tion literature. Kazdin and Blase17arguecogently that the community need for mentalhealth services far outstrips the number of pro-viders available to assist those in distress. Theycall strongly for new intervention deliverymodes targeting prevention and treatment toalleviate suffering. Hence, the purpose of thepresent study was to evaluate the effectivenessof retired senior volunteers (RSVs), who are

592 Vol. 48 No. 4 October 2014Allen et al.

available nationally through the national SeniorCorps program, to deliver a three-session remi-niscence and creative activity intervention previ-ously found effective in improving palliativecare patient and caregiver outcomes.7,8 We hy-pothesized that the palliative care patients andtheir caregivers in the RSV-delivered interven-tion group would demonstrate improvedemotional and spiritual functioning relative toa supportive contact control group. If success-ful, this mode of treatment delivery (RSV inter-vention) would represent a significant steptoward translation and greater access at earlierdisease stages of therapeutic psychosocial inter-ventions for individuals near the end of life andtheir family members.

MethodsData were collected between June 2009 and

December 2011 with approval from the Univer-sity of Alabama and the University of Alabamaat Birmingham Institutional Review Boards. Avariety of methods were used for palliativecare patient/caregiver dyad recruitmentthrough physicians/clinical champions at twouniversity medical centers, one home healthagency, five assisted living facilities, fourcongregate apartment sites for older adultsand those with low income, one continuingcare retirement community, and one rehabili-tation unit in a local skilled nursing facility.Around 66% of the dyads were physicianreferred. A community recruiter and advertise-ments in community and university print newsmedia also were used.

Patient/Caregiver Screening and GroupAssignment

After clinical champions at the recruitmentsites explained the main components of theproject, patients provided verbal consent tohave their contact information forwarded tothe research staff. Next, research staff tele-phoned the patients who had expressed inter-est; described the project in more detail; and,after getting verbal consent, administered thescreening measures (basic demographics, theVulnerable Elders Survey-13 [VES],18,19 andthe Telephone Interview for Cognitive Status-modified [TICS-m]20,21) over the telephone.Eligible patients directly nominated a ‘‘family’’

caregiver (related by blood or marriage orfictive kin) for study participation.Palliative care patients were eligible if they:

1) were aged 55 years or older, 2) were livingin the community or assisted living, 3) had anadvanced illness or combination of chronic ill-nesses, 4) received a score of three or greateron the VES,18,19 5) had no more than mildcognitive impairment as measured by a scoreof 17 or greater on the TICS-m,20,21 6) receivedan average of four hours per week of care froma caregiver, and 7) read and spoke English.Caregivers were eligible if they: 1) were iden-

tified by an eligible palliative care patient, 2)were aged older than 19 years (Alabama’s ageof majority), 3) were providing an average offour hours per week of care for the patient,4) were cognitively intact as measured by ascore of 28 or greater on the TICS-m,20,21 5)lived nearby, and 6) read and spoke English.Dyads were excluded if the patient was

receiving hospice care, or if either individualhad schizophrenia or bipolar disorder, or hada nursing home admission planned withinthree months.

Retired Senior VolunteersThe RSVs were participants in the national

Senior Corps program, part of the Corpora-tion for National and Community Service.They were recruited from FOCUS on SeniorCitizens in Tuscaloosa, AL, and PositiveMaturity in Birmingham, AL. The RSV pro-gram is open to all U.S. citizens, nationals,or lawful permanent resident aliens aged 55years or older (http://www.nationalservice.gov/programs/senior-corps/rsvp). Within thisprogram, the volunteers receive standard pre-service orientation and training; supplementalinsurance while on duty; and, potentially, reim-bursement for costs such as travel incurredduring service. The RSVs receive no pay forthe services they perform. In this study, theRSVs were included if they: 1) had a highschool education, 2) read and spoke English,and 3) had a car and drove independently.

MeasuresOther thanpatient screening instruments, pa-

tient and caregiver assessments were completedin the patient’s home at baseline, postinterven-tion, and follow-up. Demographics includedage, race/ethnicity, education and income

Vol. 48 No. 4 October 2014 593Therapy With Patients and Caregivers Near EOL

adequacy, and assessing difficulty paying forbasic needs with response options rangingfrom 1¼ not at all difficult to 4¼ very difficult.

Patient Health and Cognitive Screening by Phone.The VES-1318,19 assesses risk of functionaldecline and death. Items assess age, self-ratedhealth, limitation in physical function, andfunctional disabilities, with scores rangingfrom zero to 10. Higher scores indicate greaterrisk. Saliba et al16 found that 32% of olderadults in the nationally representative MedicareCurrent Beneficiary Survey were consideredvulnerable, with a score of three or greater of10, and this group had 4.2 times the risk ofdeath or functional decline over a two yearperiod. Further validation of the VES has shownthat, among community-dwelling older adultswith a VES score of three or higher, the esti-mated combined risk of death and functionaldecline within 11 months rose from 23%among older adults with VES¼ 3e60% amongolder adults with VES¼ 10.15

The TICS-m20,21 detects cognitive impair-ment, with scores ranging from zero to 50.Scores lower than 21 have been comparedwith scores lower than 2521 on the Mini-Mental State Examination,22 indicating clini-cally significant cognitive impairment.

Patient and Caregiver Assessments Administered inPerson. The Future Time Perspective Scale23

assesses individual’s motivation and approachto life based on time perspective and wasadministered to patients and caregivers onlyat baseline. Ten items with alpha value of0.92, assessed on a seven-point Likert-typescale, are summed; higher scores indicate anincreasingly expansive time perspective.

The Center for Epidemiological Studies-Depression Scale24 asks about the frequencyof depressive symptoms within the past week,with a score of 16 or greater indicating clini-cally significant levels of depression. The Cen-ter for Epidemiological Studies-DepressionScale has good internal consistency with Cron-bach’s alpha value of 0.73e0.81.25

The Positive and Negative Affect Schedule(PANAS)26 queries participants about theircurrent feelings regarding 10 positive and 10negative adjectives. Item responses rangefrom very slightly or not at all to extremely,with higher scores indicating more intense

affect. The PANAS has good internal consis-tency for the positive affect (PA; Cronbach’salpha¼ 0.84) and negative affect (Cronbach’salpha¼ 0.87) scales. The two scales havegood discriminant validity (r¼�0.09).

Patients and caregivers independently ratedthe patient’s physical and psychological symp-toms and associated distress using the Memo-rial Symptom Assessment Scale-Short Form.27

Higher scores in the symptom category indi-cate the presence of more patient symptoms,whereas higher bother scores indicate greaterpatient distress related to symptoms. Cron-bach’s alpha coefficients for the MemorialSymptom Assessment Scale-Short Form sub-scales range from 0.76 to 0.87.

The Brief Multidimensional Measure of Reli-giousness and Spirituality28 was used toexamine: 1) patient forgiveness (three items;a¼ 0.75), asking how often the patient hasforgiven themselves, others, or experiencedGod’s forgiveness. Higher scores indicategreater frequency of forgiveness; 2) religiousstruggle (one item): ‘‘I wonder whether Godhas abandoned me?’’ Higher scores indicategreater agreement with the item; and 3) reli-gious meaning (two items): ‘‘The events inmy life unfold according to a divine or greaterplan’’ and ‘‘I have a sense of mission or callingin my own life’’ (a¼ 0.64). Higher scores indi-cate greater agreement with the items.

The Meaning in Life Scale29,30 is an eight-item scale and it assessed four domains,namely values, purpose, goals, and reflections.Higher scores indicate greater meaning (Cron-bach’s alpha¼ 0.856).31

The caregivers completed CaregiverStressors Scale-Revised32 measure of percep-tions of competency, strain, role overload,role captivity, and emotional control, withhigher scores indicating greater caregivingstress. Internal consistency estimates rangefrom 0.60 to 0.82, with stability coefficientsranging from 0.61 to 0.74.

The Positive Aspects of Caregiving33 scale as-sesses caregivers’ perceived gains fromproviding care, with higher scores indicatingmore positive aspects of caregiving (Cron-bach’s alpha¼ 0.88).

ProceduresThis RCT compared the RSV-delivered inter-

vention to a supportive telephone contact

594 Vol. 48 No. 4 October 2014Allen et al.

control delivered by research staff. Prior expe-rience in dealing with terminal illness was notan eligibility criterion for RSVs or for researchstaff. However, the RSVs indicated if there wereany type of advanced chronic illness theywould prefer to avoid when being assigned apatient-caregiver dyad. Although the RSVswere aged older than 55 years and had at leasta high school education, all research staffmembers were aged younger than 40 yearsand had completed at least a bachelor’s degreein social/behavioral science or nursing. Ourobjective in this RCTwas to determine whetherthe RSVs could be trained to effectively deliverthe reminiscence and creative activity interven-tion. Thus, the RSVs and research staff did notneed to be similar. In translation to the com-munity, there will only be dyads receiving theintervention from RSVs.14

Patient- and caregiver-written informed con-sent was obtained at baseline. After baseline,dyads were stratified by race and site andrandomly assigned to the intervention or con-trol condition. The RSVs were asked to audio-record the final intervention session forevaluation of treatment acceptability by pallia-tive care patients and caregivers.

Baseline and postintervention assessmentswere completed in separate, concurrent inter-views with research staff not involved inproviding control calls. Postintervention as-sessments occurred approximately one weekafter completion of intervention or controlcontacts, 15e20 weeks after baseline assess-ment. Qualitative content analysis was used toexamine treatment acceptability. Follow-up as-sessments for both groups were completed13e14 weeks after the postintervention assess-ments. There was no difference between inter-vention and control groups on timing betweenbaseline, postintervention, and follow-up as-sessments. Patients and caregivers eachreceived $20 for completing baseline, postin-tervention, and follow-up assessments for a to-tal of $60 for the patient and $60 for thecaregiver.

RSV-Delivered Legacy Intervention Family Enact-ment Intervention Group. Before deliveringthe manualized intervention,8 the RSVsreceived intensive training (approximatelyfour to six hours over two sessions), whichincluded reading materials, structured role

play, and practice opportunities observed bya licensed clinical psychologist (R. S. A.).Research staff were on call for consultationwith the RSVs. The Legacy Intervention FamilyEnactment (LIFE) Volunteer InterventionistManual and LIFE Participant Notebook haveFlesch-Kincaid grade levels of 7.0 with a FleschReading Ease rating of 68.7% and were avail-able in 14-point font. The manual and accom-panying workbook comprise: 1) instructionsabout using the steps of problem solving34 todecide on a period of life and creative activityproject, 2) constructing a project, 3) evalua-tion of the activity, and 4) an appendix withlife review questions for dyads that find gener-ation of stories more difficult.With the help of the RSV, in Session 1, the

patient-caregiver dyad narrowed the focus toa time period in the patient’s life that couldbe adequately represented in one tangibleproject (e.g., scrapbook, cookbook, or audio-tapes) to represent the patient’s values andachievements in life (see Appendix for ques-tion prompts). The RSV and dyad activelybrainstormed ways to portray the life story;then the RSV helped the dyad further narrowthe focus to one meaningful project, directingthe dyad to the LIFE Participant Notebook fortips in project construction. The dyad wasasked to gather all necessary materials (suchas pictures, recipes, and stories from otherfamily members) and actively work oncompleting a portion of the project betweeneach session with the RSV.During the second session, the RSVs helped

the dyad work on the activity and further useproblem-solving skills. During the third visit,the patient and caregiver shared their activitywith the RSV, who discussed the dyad’s feelingsabout the process, including a qualitative eval-uation of the LIFE project and what the familylearned when creating it.The RSVs were carefully trained to monitor

dyads’ reactions. If either member of thedyad reacted in a persistently negative mannerto the intervention, the activity was discontin-ued. The RSVs were taught to look for depres-sion in the following ways, such as look for asaddened face, crying, shaky voice, or a state-ment from the caregiver that the care recipientseems depressed (e.g., downtrodden, sad, orblue). The LIFE Volunteer InterventionistManual includes ‘‘Feelings Checks’’ in which

Vol. 48 No. 4 October 2014 595Therapy With Patients and Caregivers Near EOL

the RSVs ask, ‘‘How does talking about thesememories make you feel?’’ The average treat-ment fidelity score achieved by the RSVs dur-ing training was 89.8%.

Patient-caregiver dyads received $25 to offsetthe costs of purchasing materials (e.g., photoalbums, scrapbooks, photograph printing pa-per, and so on) desired for the project.Research team members purchased and deliv-ered materials to mobility-restricted dyads.

Supportive Telephone Contact Control Group. Pa-tients and caregivers each received three sepa-rate, structured emotional support telephonecalls with research staff (mean� SD du-ration¼ 13� 6.5 minutes) to minimize differ-ential dropout with the RSV interventiongroup. Control callers asked questions of par-ticipants and then engaged in supportive con-versations using empathic listening andreflection. Topics discussed included family,intergenerational ties, and important aspects

Fig. 1. CONSORT fl

of the patient’s life, but structured reminis-cence and the creative and therapeutic natureof legacy activities were not discussed.

ResultsSample Characteristics

A total of 256 dyads were referred andreceived at least a letter or telephone contactto gauge eligibility and interest in participa-tion. About 76 dyads completed screeningand were eligible, 45 (59%) completed base-line, 28 completed postintervention (62% ofbaseline), and 24 completed follow-up assess-ments (86% of postintervention). As shownin Fig. 1, reasons for discontinuation frombaseline to postintervention and follow-upincluded patient health decline or death(43%), inability to re-contact (24%), caregiverabsence (19%), and loss of interest (14%). De-mographic characteristics at baseline and post-intervention are presented in Table 1; there

ow diagram.

Table 1Patient and Caregiver Characteristics (SD) at Baseline (N¼ 45) and Postintervention (N¼ 28)

Characteristics

Baseline Postintervention

Patient Caregiver Patient Caregiver

Age (yrs) 78.22 (11.59) 59.85 (11.76) 77.5 (12.13) 60.71 (11.02)Gender (%)

Women 82 71 82 68Race/ethnicity (%)

African American 40 44 43 43Education (y) 12.45 (4.05) 13.53 (2.74) 12.48 (4.25) 13.71 (3.24)Income adequacy d 2.84 (1.01) d 2.89 (.93)VES 7.29 (1.95) d 7.0 (2.31) dTICS-m 25.49 (4.85) 32.51 (4.57) 26.04 (4.80) 32.43 (4.20)

VES¼ Vulnerable Elders Survey-13; TICS-m¼ Telephone Interview for Cognitive Status-modified.There were no significant differences in baseline and postintervention patient or caregiver characteristics.

596 Vol. 48 No. 4 October 2014Allen et al.

were no significant differences between dyadswho began the study and completers. Patientsreported a median of two chronic illnesses atbaseline; 18% reported three or more chronicconditions. The most commonly reported ill-nesses were central nervous system disorders,pain/arthritis, diabetes, cancer, and heartdisease.

Caregivers had varied relationships with pa-tients including adult child (44.4%), spouse(17.8%), other family (i.e., sibling, grandchild,or niece/nephew; 11.1%), friend (4.4%), orpaid (2.2%). Nine caregivers (20%) did notreport their relationship. Caregivers ratedtheir health as ‘‘good,’’ on average.

Other than patient health (VES-13),18,19

there were no differences based on randomassignment to the intervention or controlgroup. Intervention group patients scoredhigher on the VES than control group pa-tients (t(26)¼ 2.89, P¼ 0.008) and wereconsidered more ill. Therefore, patient healthwas entered as a covariate in outcome ana-lyses. Patient and caregiver scores at baselineand postintervention assessments are pre-sented in Table 2.

Baseline AssociationsAmong palliative care patients, baseline cor-

relations reveal significant relations betweengreater future time perspective (i.e., theperception of more time left to live) and betterhealth (VES-13; r¼�0.30; P¼ 0.04), moremeaning in life (r¼ 0.39; P¼ 0.007), andmore PA (r¼ 0.41; P¼ 0.005). Measures ofphysical and emotional symptoms and botherwere significantly associated with negativeaffect and depressive symptoms.

Among caregivers, baseline correlationsreveal significant relations between greaterfuture time perspective and greater meaningin life (r¼ 0.41; P¼ 0.009), more PA(r¼ 0.56; P< 0.0001) and fewer depressivesymptoms (r¼�0.34; P¼ 0.02). Greaterfuture time perspective also was significantlyassociated with lower cognitive functioning(r¼�0.31; P¼ 0.04). Greater caregiving stressat baseline was significantly related to worsepalliative care patient health (VES-13;r¼ 0.36; P¼ 0.02), and caregiver report of agreater number of patient physical symptoms(r¼ 0.33; P¼ 0.03) and physical symptombother (r¼ 0.42; P¼ 0.004), as well as a greaternumber of patient emotional symptoms(r¼ 0.35; P¼ 0.02) and emotional symptombother (r¼ 0.39; P¼ 0.008).

Intervention Effects at Postintervention andMaintenance at Follow-UpControlling for patient health, patients in

the intervention group reported a significantlygreater reduction in number of emotionalsymptoms (F(1,25)¼ 6.18, P¼ 0.02, partialh2¼ 0.20) and emotional bother (F(1,25)¼4.81, P¼ 0.04, partial h2¼ 0.16) than patientsin the control group, whose emotional symp-toms and bother increased. Intervention pa-tients expressed more forgiveness thancontrol patients (F(1,25)¼ 5.24, P¼ 0.031;partial h2¼ 0.17) and less negative religiouscoping (e.g., feelings of abandonment byGod; F(1,25)¼ 4.88, P¼ 0.037; partialh2¼ 0.16) from baseline to postintervention.Thus, delivery of the intervention by RSVshad a positive impact on palliative care pa-tients’ emotional symptoms and burden and

Table 2Characteristics of Patients and Caregivers Who Completed LIFE

Participant Characteristics

Experimental Condition

Intervention Group (n¼ 10 dyads) Control Group (n¼ 18 dyads)

Baseline Postintervention Baseline Postinterventionr

CaregiverMSASdsymptom numberPhysical 9.86 (4.2) 10.90 (4.3) 9.57 (3.2) 9.00 (4.0)Emotional 1.86 (1.4) 1.80 (1.5) 2.43 (1.2) 1.83 (1.6)

MSASdsymptom botherPhysical 0.56 (0.4) 0.77 (0.6) 0.54 (0.3) 0.49 (0.5)Emotional 0.92 (0.8) 1.08 (1.2) 1.27 (0.8) 1.00 (1.0)

Positive affect 3.89 (0.6) 3.26 (0.8) 3.61 (0.7) 3.42 (0.7)Negative affect 1.67 (0.6) 1.62 (0.7) 1.54 (0.5) 1.48 (0.5)CESD 9.68 (7.6) 14.90 (13.5) 9.70 (7.2) 8.17 (6.4)Future time perspective 52.94 (8.9) d 48.48 (10.8) dBMMRSdmeaning 3.36 (0.6) 3.63 (0.4) 3.31 (0.6) 3.43 (0.7)Meaning in life 3.42 (0.4) 3.64 (0.4)a 3.53 (0.4) 3.53 (0.4)Caregiver stress 2.35 (0.2) 2.32 (0.2) 2.34 (0.2) 2.30 (0.2)PAC 4.67 (0.3) 4.67 (0.4) 4.61 (0.6) 4.65 (0.5)

PatientMSASdsymptom numberPhysical 9.27 (5.2) 9.40 (5.7) 8.30 (2.8) 9.06 (3.8)Emotional 2.22 (1.7) 1.80 (1.1)a 1.39 (1.3) 2.00 (1.2)

MSASdsymptom botherPhysical 0.74 (0.5) 0.78 (0.6) 0.61 (0.3) 0.64 (0.4)Emotional 1.15 (0.9) 1.0 (0.8)a 0.64 (0.7) 0.97 (0.7)

Positive affect 3.12 (0.7) 3.17 (0.8) 3.38 (0.7) 3.22 (0.7)Negative affect 1.80 (0.6) 1.80 (0.6) 1.61 (0.5) 1.66 (0.6)CESD 14.27 (8.8) 18.40 (10.0) 11.65 (8.0) 13.50 (9.6)Future time perspective 38.45 (14.1) d 44.26 (13.6) dBMMRSDaily spiritual experiences 5.12 (0.7) 5.23 (10.0) 4.85 (1.2) 4.90 (1.2)Forgiveness 3.59 (0.3) 3.80 (0.3)a 3.54 (0.7) 3.50 (0.7)Meaning 3.41 (0.7) 3.35 (0.7) 3.24 (0.8) 3.33 (0.8)Abandonment 3.73 (0.6) 3.80 (0.4)a 3.96 (0.2) 3.83 (0.5)

Meaning in life 3.39 (0.5) 3.43 (0.6) 3.57 (0.4) 3.58 (0.4)

MSAS¼Memorial Symptom Assessment Scale; CESD¼ Center for Epidemiological Studies Depression Scale; BMMRS¼ Brief MultidimensionalMeasure of Religiousness and Spirituality.Values represent means with standard deviation in parentheses.aValues in bold indicate significant differences from baseline to postintervention assessment at the P< 0.05 level.

Vol. 48 No. 4 October 2014 597Therapy With Patients and Caregivers Near EOL

spiritual health. Effect sizes observed weresimilar to those in previous research.8

However, only improvement in interventionpatients’ emotional symptom bother was main-tained at follow-up (F(2,19)¼ 4.56, P¼ 0.024;h2¼ 0.32). It is noteworthy that between post-intervention and follow-up, there was nocontact with the RSV.

Palliative caregivers also experienced initialgains from the RSV-delivered intervention.Controlling for group differences in patienthealth, caregivers in the intervention group re-ported increased meaning in life comparedwith control caregivers (F(1,25)¼ 6.20,P¼ 0.02; partial h2¼ 0.24). Once again, effectsizes were similar to those in previousresearch.8 No other statistically significantgroup differences for caregivers were found,

and this initial gain was not maintained atfollow-up, during which time there was no con-tact with the RSV.

Qualitative Analysis of TreatmentAcceptability

Descriptive qualitative content analysis35 ofaudiotaped and transcribed intervention Ses-sion 3 revealed several themes. Patients andcaregivers shared that the intervention hadbeen an enriching, energizing, or valued expe-rience. A family pride theme emerged, withimportance placed on ‘‘passing down’’ familyhistory. A theme of patient-caregiver connec-tion emerged, along with the concept of bettercommunication. Patients and caregiversstrongly expressed appreciation for the RSV.Some intervention dyads shared that their

598 Vol. 48 No. 4 October 2014Allen et al.

project was not completed, but expressed theintention of continuing work on it.

Unique to patients, the theme of beingthankful for, proud or appreciative of the care-giver arose. One patient stated, ‘‘She’s got adifferent perspective than I do,’’ to which thecaregiver replied, ‘‘I base it on. the thingsthat inspire me about you. the qualities youpossess.’’ Patients mentioned the idea of beingremembered through creation of the project.One patient stated, ‘‘I got some more great-great grans coming along now. And see, theydon’t even know me.’’ Another stated, ‘‘I loveremembering and talkin’ about all ofthem.and me. That’s good, too, the part ofme.’’ This theme was occasionally tied to familybeing geographically distant.

The theme of patient suffering arose. Onepatient stated, ‘‘I see that my stroke was aturning point in my life.’’ Another said, ‘‘Ihave always worried that I wouldn’t be able tohave my whole life flash before my eyes. I’dforget some of it.’’

Unique to caregivers, the theme of motiva-tion by the project and the RSV arose. Onecaregiver stated, ‘‘It’s one of those things thatyou always intend to get around to.’’ Anothersaid, ‘‘I probably would have just stopped hadyou not come back to see what else I haddone.’’ Caregivers expressed appreciation forthe Participant Notebook and brainstormingactivity, along with the project’s ability to‘‘keep [the patient’s] mind going a littlewhile.’’ Negative themes emerged concerningcosts of the project in time and money andcaregivers feeling overwhelmed.

DiscussionThe results of this RCTextend prior research

by demonstrating the effective delivery by RSVsof a manualized reminiscence and creative activ-ity intervention with positive emotional out-comes for palliative care patients andcaregivers. Effect sizes were similar to priorresearch wherein graduate students acted as in-terventionists.8 Palliative care patients experi-enced fewer emotional symptoms and lessemotional symptom bother, greater forgiveness,and less ‘‘abandonment by God,’’ and their care-givers experienced greater meaning in life incomparison with the control group. It is note-worthy that, as in Dignity Therapy,10,12 several

intervention effects reflected existential or spiri-tual realms. Unfortunately, only improvement inintervention patients’ emotional symptombother was maintained at follow-up after contactwith the RSV ceased. This finding suggests that,as recommended by Kazdin and Blase,17 alterna-tive treatment delivery modes such as continuedRSV contact, technology, or self-help must bebuilt into translation of interventions to relievepatient and caregiver suffering. Notably, care-givers reported that presence of the RSV-motivated continued work on the reminiscenceand creative activity project.Further translation and implementation of

this RSV-delivered intervention will necessitateprogramming for maintenance of therapeuticgains. Perhaps certain RSVs would be inter-ested in ‘‘adopting’’ a family for longer inter-vention periods. Alternatively, volunteerscould be recruited from faith-based or otherorganizations with which palliative care pa-tients and caregivers are actively involved. Re-cruiting volunteers from within the palliativecare patient or caregiver community to act asinterventionists also may be feasible undercertain conditions. Finally, technology-assisteddelivery methods for such interventionsshould be explored.17

How time perspective functions near deathhas been a matter of some debate.36e38 In thisstudy, future time perspective measured at base-line was associated with meaning in life and PAamong both patients and caregivers. As ex-pected, patients in better health, and hencewith lower risk of death or functional declinein the near future, reported a more expansiveperspective of future time. Among caregivers,a significant association between an expansivetime perspective and fewer depressive symp-toms was observed. Caregiver stress was associ-ated with worse patient physical andemotional functioning. Future qualitativestudies should explore the experience of time,suffering, forgiveness, and meaning amongpalliative care patients and caregivers39 toclarify situations in which existentially focusedintervention may be therapeutically beneficial.As with any research, this project had limita-

tions. First, community-based sample recruit-ment proved difficult, with only 45 of 76eligible dyads (59%)completingbaseline.More-over, generalizability is limitedby thegeographicregion and racial/ethnic characteristics of the

Vol. 48 No. 4 October 2014 599Therapy With Patients and Caregivers Near EOL

participants. Future research should targetdyads of diverse racial/ethnic groups, particu-larly in rural areas. Third, our patients’ prox-imity to death varied. Although thecommunity-based, pre-hospice nature of thisstudy is unique, the timing of intervention maybe better placed at patient diagnosis or healthstatus change to focus patients and caregiverson prevention of decline and increasemotivation.

Despite these limitations, these promisingresults have practice and policy implications.The RSV delivery of effective interventions ex-tends the potential scope of palliative mentalhealth care practice to patient and caregiverdyads that have not yet chosen hospicecare.17 Moreover, volunteer efforts reflect amonetary value of $161.7 billion,40 and volun-teering affords positive health benefits to thevolunteer.41,42 The results of this RCT demon-strate that RSVs can be trained to effectivelydeliver reminiscence and creative activity inter-ventions with similar patient and caregiver im-provements in outcome to that found inprevious research.8

Disclosures and AcknowledgmentsThis project was supported by funding from

the National Institute of Nursing Research(R21NR011112) to Dr. Allen. Although all au-thors received financial support for this proj-ect, no conflicts of interest were reportedand the National Institute of Nursing Researchwas not involved in the conduct of the study ordevelopment of the manuscript.

Special thanks are extended to Drs. ReginaHarrell, Richard Sims, Ali Ahmed, Andrew Dux-bury, and Kelli Flood, and to Ms. Mary Hardrickand Ms. Starr P. Culpepper for assistance inrecruitment. Most especially, the LIFE team ex-tends its thanks to FOCUS on Senior Citizensand Positive Maturity for supporting volunteerrecruitment and involvement, and to all of thepatients and caregivers who gave generously oftheir time and energy to this project.

References1. Carstensen L, Isaacowitz DM, Charles ST. Tak-ing time seriously. Am Psychol 1999;54:165e181.

2. Chochinov HM, Hassard T, McClement S, et al.The landscape of distress in the terminally ill. J PainSymptom Manage 2009;38:641e649.

3. Breitbart W, Applebaum A. Meaning-centeredgroup psychotherapy. In: Watson M, Kissane DW,eds. Handbook of psychotherapy in cancer care.New York: Wiley, 2011:137e148.

4. Neimeyer RA, Pennebaker JW, van Dyke JG.Narrative medicine: writing through bereavement.In: Chochinov HM, Breitbart W, eds. Handbook ofpsychiatry in palliative Medicine. New York: OxfordUniversity Press, 2009:454e469.

5. Harris GM, Allen RS, Dunn L, Parmelee P.‘‘Trouble won’t last always’’: religious coping andmeaning in the stress process. Qual Health Res2013;23:773e781.

6. Randall GK, Bishop AJ. Direct and indirect ef-fects of religiosity on valuation of life throughforgiveness and social provisions among older incar-cerated males. Gerontologist 2013;53:51e59.

7. Allen RS. The legacy project intervention toenhance meaningful family interactions: case exam-ples. Clin Gerontol 2009;32:164e176.

8. Allen RS, Hilgeman MM, Ege MA, Shuster JL,Burgio LD. Legacy activities as interventions ap-proaching the end of life. J Palliat Med 2008;11:1029e1038.

9. Ando M, Morita T, Akechi T, Okamoto T. Effi-cacy of short-term life-review interviews on the spir-itual well-being of terminally ill cancer patients.J Pain Symptom Manage 2010;39:993e1002.

10. Chochinov HM. Dignity therapy: Final words forfinal days. New York: Oxford University Press, 2012.

11. Chochinov HM, Hack T, Hassard T, et al. Dig-nity therapy: a novel psychotherapeutic interventionfor patients near the end of life. J Clin Oncol 2005;23:5520e5525.

12. Chochinov HM, Kristjanson LJ, Breitbart W,et al. Effect of dignity therapy on distress and end-of-life experience in terminally ill patients: a rando-mised controlled trial. Lancet Oncol 2011;12:753e762.

13. McClement S, Chochinov HM, Hack T, et al.Dignity therapy: family member perspectives.J Palliat Med 2007;10:1076e1082.

14. Scogin F, Welsh D, Hanson A, Stump J,Coates A. Evidence-based psychotherapies fordepression in older adults. Clin Psychol Sci Pract2005;12:222e237.

15. Candy B, Jones L, Drake R, Leurent B, King M.Interventions for supporting informal caregivers ofpatients in the terminal phase of a disease. Co-chrane Database Syst Rev 2011;CD007617.

16. Csikai EL, Weisenfluh S. Hospice and palliativesocial workers’ engagement in life review interven-tions. Am J Hosp Palliat Care 2013;30:257e263.

17. Kazdin AE, Blase SL. Rebooting psychotherapyresearch and practice to reduce the burden ofmental illness. Perspect Psychol Sci 2011;6:21e37.

600 Vol. 48 No. 4 October 2014Allen et al.

18. Min LC, Elliott MN, Wenger NS, Saliba D. High-er vulnerable elders survey scores predict death andfunctional decline in vulnerable older people. J AmGeriatr Soc 2006;54:507e511.

19. Saliba D, Elliott M, Rubenstein LZ, et al. TheVulnerable Elders Survey: a tool for identifyingvulnerable older people in the community. J AmGeriatr Soc 2001;49:1691e1699.

20. van den Berg E, Ruis C, Biessels GJ, Kappelle LJ,van Zandvoort MJ. The Telephone Interview forCognitive Status (Modified): relation with a compre-hensive neuropsychological assessment. J Clin ExpNeuropsychol 2012;34:598e605.

21. De Jager CA, Budge MM, Clarke R. Utility ofTICS-M for the assessment of cognitive function inolder adults. Int J Geriatr Psychiatry 2003;18:318e324.

22. Folstein MF, Folstein S, McHugh P. ‘‘Mini-Mental State’’: a practical method for grading thecognitive state of patients for the clinician.J Psychiatr Res 1975;12:189e198.

23. Lang FR, Carstensen LL. Time counts: Futuretime perspective, goals, and social relationships. Psy-chol Aging 2002;17:125e139.

24. Radloff LS. The CES-D scale: a self-reportdepression scale for research in the general popula-tion. Appl Psychol Meas 1977;1:385e401.

25. Carpenter JS, Ph D, Andrykowski MA, Wilson J,Rayens MK. Psychometrics for two short forms ofthe Center for Epidemiologic Studies-DepressionScale. Issues Ment Health Nurs 1998;19:481e494.

26. Watson D, Clark LA, Tellegen A. Developmentand validation of brief measures of positive andnegative affect: the PANAS scales. J Pers Soc Psychol1988;54:1063e1070.

27. Chang VT, Hwang SS, Feuerman M, Kasimis BS,Thaler HT. The memorial symptom assessmentscale short form (MSAS-SF). Cancer 2000;89:1162e1171.

28. Fetzer Institute. Multidimensional measure-ment of religiousness/spirituality for use in healthresearch. Kalamazoo, MI: John E. Fetzer Institute,1999.

29. Krause N. Longitudinal study of social sup-port and meaning in life. Psychol Aging 2007;22:456e469.

30. Krause N. Evaluating the stress-buffering func-tion of meaning in life among older people.J Aging Health 2007;19:792e812.

31. Krause N. Meaning in life and mortality.J Gerontol B Psychol Sci Soc Sci 2009;64B:517e527.

32. Zarit SH, Stephens MA, Townsend A, Greene R.Stress reduction for family caregivers: effects ofadult day care use. J Gerontol B Psychol Sci SocSci 1998;53:S267eS277.

33. Tarlow BJ, Wisniewski SR, Belle SH, et al. Posi-tive aspects of caregiving: contributions of theREACH project to the development of new mea-sures for Alzheimer’s caregiving. Res Aging 2004;26:429e453.

34. D’Zurilla T, Nezu A. Problem-solving therapy: Apositive approach to clinical intervention, 3rd ed.New York: Springer Publishing Company, 2007.

35. Sandelowski M. Whatever happened to qualita-tive description? Res Nurs Health 2000;23:334e340.

36. Chochinov HM. Death, time and the theoryof relativity. J Pain Symptom Manage 2011;42:460e463.

37. Chochinov HM. Relatively speaking. J PainSymptom Manage 2012;43:e6ee7.

38. Round J. Death, time, and the theory of relativ-ity: a brief reply? J Pain Symptom Manage 2012;43:e2ee6.

39. Monin JK, Schulz R. Interpersonal effects ofsuffering in older adult caregiving relationships.Psychol Aging 2009;24:681e695.

40. Johnson RW, Schaner SG. Value of unpaid activ-ities by older Americans tops $160 billion per year.Washington, DC: The Urban Institute’s RetirementProject, 2005.

41. Harris AHS, Thoresen CE. Volunteering is asso-ciated with delayed mortality in older people: anal-ysis of the longitudinal study of aging. J HealthPsychol 2005;10:739e752.

42. Martinez IL, Frick K, Glass TA, et al. Engagingolder adults in high impact volunteering that en-hances health: recruitment and retention in theExperience Corps Baltimore. J Urban Health 2006;83:941e953.

Appendix

Question Prompts Used in Session 1 From the Legacy Intervention Family EnactmentVolunteer Interventionist Manual

Patient questionsdFour items‘‘The things I care about and value most in my life are.’’‘‘The most important people in my life have been.’’‘‘I would like people to remember these things about me.’’‘‘The ideas, books, music, and poems that have most influenced my life are.

Caregiver questionsdThree items‘‘My favorite memories of times with my loved one are.’’‘‘The things I most want to remember about my loved one are.’’‘‘The lessons/values I have learned or most associate with my loved one are.’’

Tips for legacy activities1. Writing or recording stories on paper2. Recording on an audio cassette tape player3. Scrapbook or photo album4. Video tape recording5. Family cookbook6. Other ideas

Vol. 48 No. 4 October 2014 601Therapy With Patients and Caregivers Near EOL