Can Counseling and Support Reduce Burden and Depressive Symptoms in Caregivers of People with Alzheimer's Disease During the Transition to Institutionalization? Results from the New York University Caregiver Intervention Study

Can Counseling and Support Reduce Burden and Depressive Symptoms in Caregivers of People with Alzheimer's Disease During the Transition to Institutionalization? Results from the New York University Caregiver Intervention Study
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Can Counseling and Support Reduce Burden and DepressiveSymptomsinCaregiversofPeoplewithAlzheimer’sDiseaseDuringthe Transition to Institutionalization? Results from the New YorkUniversity Caregiver Intervention Study  Joseph E. Gaugler, PhD,  David L. Roth, PhD, w  William E. Haley, PhD, z  and Mary S. Mittelman, DrPH  §  k OBJECTIVES:  To determine whether counseling and sup-port reduce the burden and depressive symptoms of spousecaregivers of patients with Alzheimer’s disease (AD) duringthe transition to institutionalization. DESIGN:  A randomized, controlled trial of an enhancedcounseling and support program for spouse caregivers of persons with AD. Structured interviews were conductedwith spouse caregivers at baseline, every 4 months duringYear 1, and every 6 months thereafter for up to 16 years. SETTING:  Outpatient research clinic in the New YorkCity metropolitan area. PARTICIPANTS:  Referredvolunteersampleof406spousecaregivers of community-dwelling patients with AD en-rolled over a 9.5-year period. INTERVENTION:  Enhancedcounselingandsupportcon-sisting of six sessions of individual and family counseling,support group participation, and continuous availability of ad hoc telephone counseling. MEASUREMENTS:  Outcome measures included burden(modified Zarit Burden Interview) and depressive symp-toms (Geriatric Depression Scale). RESULTS:  Burden and depressive symptoms were signifi-cantly lower for caregivers in the treatment group than forcontrols receiving usual care at the time of and after insti-tutionalization. Nursing home admission itself significantlyreduced burden and depressive symptoms in the interven-tion and control groups. CONCLUSION:  Institutionalization alone can reducecaregiver burden and depressive symptoms, but enhancedcounseling provides additional long-term benefits. The re-sults offer some of the first clinical evidence of the benefitsof enhanced counseling during the transition to institution-alization for caregivers of people with AD.  J Am GeriatrSoc 56:421–428, 2008. Key words: caregiving; nursing home placement; nursinghome admission; informal long-term care; psychosocialintervention I t is estimated that 8.9 million family caregivers provideassistancetosomeoneaged50andolderwithAlzheimer’sdisease (AD) or a related dementia. 1 Families provide per-sonal care that is integral to helping disabled older adultsremain in their homes. 1,2 Family care is not without itscosts, particularly in the context of AD. As the disabilitiesand care needs of the person with dementia increase overtime, the accumulated financial, social, psychological, andphysiological effects of family caregiving also intensify. 3–8 The perspective that tends to dominate much of theliterature is that care by family members is provided solelyto older adults living at home. When caregivers are fol-lowed over longer periods of time, it becomes evident thatfamily caregiving responsibilities do not end with institu-tionalization of the disabled relative. Instead, this key tran-sition appears to affect the type and intensity of helpprovided. 9–11 Unlike earlier studies that treated institution-alization as an ‘‘endpoint’’ in family caregiving, recent re-search has emphasized the continued involvement of relatives in care and the effects of nursing home admission(NHA) on the stress and mental health of family mem-bers. 9,12–14 A version of this paper was presented at the 2006 Gerontological Society of America annual meeting in Dallas, Texas.Address correspondence to Joseph E. Gaugler, Center on Aging, Center forGerontological Nursing, School of Nursing, University of Minnesota, Min-neapolis, MN. E-mail: gaug0015@umn.eduDOI: 10.1111/j.1532-5415.2007.01593.xFrom the   Center on Aging, Center for Gerontological Nursing, School of Nursing, University of Minnesota, Minneapolis, Minnesota;  w Department of Biostatistics, School of Public Health, University of Alabama at Birmingham,Birmingham, Alabama;  z School of Aging Studies, College of Arts andSciences, University of South Florida, St. Petersburg, Florida;  § Departmentof Psychiatry and  k Psychosocial Research and Support Program, SilbersteinInstitute, School of Medicine, New York University, New York, New York.  JAGS 56:421–428, 2008 r 2008, Copyright the Authors Journal compilation  r 2008, The American Geriatrics Society 0002-8614/08/$15.00  Whether NHA leads to fewer depressive symptoms andless burden for caregivers is an important clinical questionthat has yet to be definitively answered. One early descrip-tive study suggested that NHA provides relief to familycaregivers and reduces stressors such as feelings of exhaus-tion and fatigue related to care provision (role overload), atleasttemporarily. 15 More-recentstudiesconcludethatNHAresults in guilt, anger, anxiety, and depression for caregiversof people dementia, although others suggest that NHA doeslittletoinfluencethetrajectoriesofstressornegativementalhealth before or after institutionalization. 13,14,16–20 Addi-tional descriptive research has focused on the new challeng-es that institutionalization poses for families, such as thenegative interactions between family caregivers and staff ascontrol of care is shifted to the facility and the family mem-ber is left with an ambiguous care role. 11,16,18 Parallel to descriptive analyses of dementia caregivingand institutionalization, a series of clinical efforts havesought to alleviate the psychological, emotional, and phys-iological distress that can occur as a function of intensivefamily care provision. Recent meta-analyses and multisite,randomized, controlled evaluations indicate that psycho-social interventions for caregivers such as skills training,education, therapeutic counseling, and information-basedservicesaregenerallyeffectiveinproducingclinicallymean-ingfulimprovementsinpsychologicalwell-being. 21–26 Nev-ertheless, there is little research to document the effects of these interventions in easing the transition of NHA forfamily caregivers. The goal of the current study is to unitethese two streams of research to determine whether theavailability of long-term counseling and support reducesburden (the emotional, psychological, physical, and emo-tional ‘‘load’’ of care provision 27 ) and depressive symptom-atology across the institutionalization transition.The New York University Aging and Dementia Re-search Center (NYU-ADRC) has been conducting a pio-neering randomized trial of a psychosocial intervention forspouse–caregivers of people with AD since 1987. 28–34 Be-cause of its prospective data collection strategy throughoutthe provision of dementia caregiving, the NYU CaregiverIntervention (NYUCI) is uniquely positioned to answer theresearch questions of interest in this study. Moreover, therandomized, controlled design of the NYUCI offers clinicalinsights beyond those offered in descriptive research.Based on the previously reported effects of the NYUCIin delaying institutionalization 29–31 and the equivocal find-ings of prior research on the ramifications of NHA forfamily caregivers, analyses were conducted of the NYUCIto answer the following questions:(1) DoestheNYUCIreduceburdenanddepressivesymptomsof caregivers of people with AD before and after NHA?(2) What are the long-term, post-NHA changes in burdenand depressive symptoms for caregivers of people withAD who do or do not receive the NYUCI? METHODSProcedure Participants in the parent NYUCI included 406 spouses of persons with a clinical diagnosis of AD. Slightly more than50% of participants were recruited through the NYU-ADRC (n 5 214), whereas the remaining participants wererecruited through a number of channels, including localAlzheimer’s Association chapters, physicians, public mediaannouncements, and referring community providers (e.g.,adult day centers, social service agencies, lawyers). Partic-ipants recruited through the NYU-ADRC had spouses whoreceived an AD diagnosis according to National Institute of Neurologic and Communicative Diseases and Stroke–Alzheimer’s Disease and Related Disorders Association cri-teria. 35 The other participants’spouses received a physiciandiagnosis of AD. All spouses were living with the personwith AD and were residing at home at the time of studyenrollment. Each participant (caregiver or spouse with AD)had to have at least one other family member or relativeliving in the New York City metropolitan area.Upon study intake, participants were administered anintensive baseline assessment protocol consisting of struc-tured questionnaires. Participants were then randomizedusing lottery to one of two conditions: the experimental,comprehensive counseling condition (n 5 203) or the con-trol group (n 5 203), which consisted of the usual care pro-vided to patients and their caregivers at the NYU-ADRC.All participants could use services outside of the study. Theassessment battery was re-administered quarterly duringthefirstyearofparticipationandtwiceayearinsubsequentyears. All follow-up assessments were conducted in personor over the telephone. After NHA, caregiver assessmentscontinued at the regularly scheduled follow-up intervalsunless the caregiver refused or was lost to follow-up. Sample Participants were followed for up to 15.9 years. At least 10years of follow-up data are available for 12% of the cases(n 5 48). During the course of the study, 210 care recipientswere institutionalized; 99 (48.8%) of those in the interven-tion condition placed their spouse in a nursing home atsome point, compared with 111 (54.7%) of those in theusual care condition. Previously published Cox propor-tional hazards models have indicated that there was asignificantly longer time to NHA in the interventiongroup. The median times to NHA were 4.8 and 3.3 yearsafter baseline for intervention and control care recipients,respectively. 31 Ten cases (4 intervention, 6 usual care) completed abaseline assessment but were lost to follow-up. In 11 cases(8intervention,3usualcare),thecarerecipientsdiedbeforethefirst(4-month)follow-upinterview.Becausethefocusof this analysis is on the effects of comprehensive counselingbeforeandafterNHA,these21individualsarenotincludedin subsequent analyses. This resulted in a longitudinal sam-pleof385caregivers.Comparisonsbetweenthese385casesand the 21 cases excluded from the longitudinal analysisindicated that these two groups did not differ significantlyon baseline values of burden or depressive symptoms( P 4 .21). Inclusion status also did not differ according totreatmentgrouporsex( P 4 .25).Consequently,noevidencefor selection bias was observed as a result of including 385cases for analysis from the larger pool of 406 randomizedcases. 422  GAUGLER ETAL.  MARCH 2008–VOL. 56, NO. 3 JAGS  Descriptive data for the sample are presented inTable 1. As detailed in previous publications from theNYUCI, 32 there was a random sex imbalance at baseline,with a higher proportion of husband caregivers being ran-domlyassignedtotheinterventioncondition.Consequently,sex and baseline scores on the outcome variables wereincluded as covariates in all subsequent analyses. Intervention The NYUCI consisted of three components: individual andfamily counseling, support group participation, and ad hoccounseling. During the 4 months after the baseline assess-ment, spouse caregivers participated in six individual andfamily sessions with the study counselor (two with thespouse caregiver only and four with the spouse caregiverand at least one other family member; the person with ADdid not participate in these sessions). The content of thesesessions was oriented to the stated needs of the caregiverand included information on AD, skills related to the man-agement of behavioral problems, and strategies to bolstercommunicationwithinvolvedandnoninvolvedfamilymem-bers. Caregivers agreed at intake that, after the 4-monthfollow-up, they would participate in a support group thatmet weekly under the auspices of the Alzheimer’s Associ-ation. These groups were widely available in the New YorkCity metropolitan area. The third component, providedthroughout the duration of the spouse caregiver’s partici-pation in the intervention, was ad hoc counseling; caregiv-ers and participating family members were free to contactthe study counselors by way of telephone to address anyconcerns, crises, or other significant changes that occurred.Counselors with advanced degrees in social work, psychol-ogy, counseling, or gerontology delivered the NYUCI. Thecontent of NYUCI is described in detail elsewhere. 36 Caregivers in the usual care group did not receive theformal counseling sessions provided to those in the inter-vention group, and their family members did not have con-tact with the study counselors. For ethical reasons,caregivers in the usual care group were free to identify anduse services on their own, and they were informed uponenrollment that they could contact study counselors for in-formationorreferralpurposes. Forthesereasons,caregiversin the ‘‘usual’’ care group were likely to have received moreservices than are typically provided in clinical care settings.A prior report has described methods of tracking en-actment used in the NYUCI. 37 Data on contact with par-ticipants were collected to provide counselors withinformation about use of recommended resources. TheNYUCI began in 1987, and the treatment fidelity measuresin current psychosocial interventions were not routinelyused 2 decades ago. The records indicate that eight of thefirst 11 participants in the treatment group who were in-stitutionalized did so on advice from the counselors, whoknew through their work with spouse caregivers and otherfamily members that NHA was in the best interest of allconcerned. 29 This is particularly relevant when interpretingthe findings of the current study, which are detailed below. Measures  Nursing Home Admission Dates of NHA were derived from follow-up interviews,NYU-ADRC records, or ad hoc telephone contacts withspouse caregivers or other family members.  Burden Caregiver burden was measured using a subset of questionsfrom the Zarit Burden Interview (ZBI). The ZBI is one of the most widely used instruments to assess caregiving bur-den. 27,38 The modified ZBI includes 15 questions that mea-sure areas of potential stress (e.g., perceived time pressure,emotional distress, financial strain, guilt, and overall bur-den) that could arise before and after NHA for spousecaregivers.  Depressive Symptoms The 30-item version of the Geriatric Depression Scale(GDS) 39 wasadministeredatbaselineandateachfollow-upinterval to measure spouse caregivers’ mood and psycho-logical well-being. Statistical Analyses TheeffectsoftheinterventionandNHAonthe15-itemZBIand the 30-item GDS were estimated and tested using ran-dom effects regression growth curve analyses. A multilevelchange model was used that estimated longitudinal trajec-tories for individual participants at one level, with the in-tercepts and slopes of these person-specific longitudinaltrajectories analyzed as the effects of between-subject pre-dictors at a higher-order second level. 40 All analyses wereconducted using restricted maximum likelihood estimationas provided by SAS PROC MIXED (SAS Institute, Inc.,Cary,NC). 41 Themodelswerebasedon4,193observationsover time. These included 3,055 observations during thecommunity caregiving phase and 1,148 observations of caregivers while the care recipients were in the nursinghome.The primary analytical models included four time-invariant variables: caregiver sex, treatment group, place-ment group, and the baseline (pretreatment) value on theoutcome being examined. Sex was coded as 1 for female Table 1. Descriptive Characteristics VariableEnhancedTreatment(n 5 191)Usual Care(n 5 194) Female caregiver, n (%) 103 (53.9) 126 (65.0) Age of caregiver, mean  SD 71.55  8.71 71.03  9.46 Age of care recipient, mean  SD 73.67  8.19 74.60  8.23Global Deterioration Scale score, n (%)4 71 (37.2) 63 (32.5)5 85 (44.5) 73 (37.6)  6 35 (18.3) 58 (29.9)Caregiver Geriatric Depression Scalescore, mean  SD8.97  5.73 10.33  7.11Caregiver Modified Zarit Burden Scalescore, mean  SD25.55  9.46 26.87  10.86 Note:  Sample is the subset of 385 caregivers who provided sufficient follow-up data for the analyses.SD 5 standard deviation. INTERVENTION AND INSTITUTIONALIZATION  423 JAGS MARCH 2008–VOL. 56, NO. 3  caregivers and 0 for male caregivers. Treatment group wascoded as 1 for the intervention group and 0 for the usualcare group. Placement group was coded as 1 for cases inwhich a nursing home admission occurred at some point inthe follow-up interval and 0 for cases in which no nursinghome admission was observed. The baseline value on theoutcomewasmean-centeredsuchthattheoverallmeanwassubtracted from each participant’s score, resulting in a cen-tered score that was a raw deviation from the mean andequal to 0 for someone who scored at the mean.The models included two predictors for time: one thatmeasured time elapsed since the date of the baseline assess-ment and one for the amount of time elapsed since NHA.Both predictors were calculated in actual days and thendivided by 365.25 to scale these measures in years. Theyears since baseline measure was further modified by sub-tracting 1, resulting in a centered measure that indicatedthe time since (or before) the 1-year post-baseline point.The purpose of this centering was to scale the model so thatthe main effect tests for the time-invariant predictors (e.g.,treatment group, sex) were comparisons between thegroups at the 1-year post-baseline point in time. The yearssinceNHAvariablewassetto0foranyobservationsbeforeplacement and for those cases who never placed. Usingthese two measures for time (years since baseline minus 1and years since NHA), the main effect for years since base-line tested whether the linear slope across time was signifi-cantly different from 0, and the main effect for time sinceNHA tested whether this linear slope across time changedsignificantly after NHA for caregivers who placed theirspouses.A time-varying indicator for institutionalization ortype of assessment was also included in the model as a pre-dictor.Thisindicatorwassetto0forallobservationsbeforeplacementandforthosecaseswhoneverplacedandto1forall observations after placement in those who were institu-tionalized. This time-varying indicator provided a test of whether NHA led to an abrupt change in the level of theoutcome (burden–ZBI or depressive symptoms–GDS) im-mediately after institutionalization. In addition to theseseven main effects (sex, treatment group, placement group,baseline, years since baseline, years after NHA, and type of assessment), six two-way interaction effects were includedin the primary analytical models. All four time-invariantpredictors were specified to have an interaction effect withyears since baseline. This included a treatment group-by–years since baseline effect that tested whether the linearrates of change before placement differed between inter-vention and usual care participants. A group-by–years afterNHAinteraction termtested whether thelinearslope oftheoutcome variable across time differed between the inter-ventionandusualcaregroupsafterplacement.Agroup-by–type of assessment interaction effect tested whether thechange in outcome observed from before to immediatelyafter placement differed as a function of treatment group.Prior research has emphasized the various indicatorsthat are associated with caregiver burden, caregiver de-pressive symptoms, and NHA. 2,4,15–17 The NYUCI consid-ered these factors in its design and data collection protocol.These factors include sociodemographic context and sever-ity of cognitive impairment. 15 In spite of the random as-signment of cases to conditions, some imbalances werefound between treatment and control conditions across keycaregiving indicators, such as sex (Table 1). For this reason,caregiversexandthebaselinevalueoftheoutcomevariablein question were included as covariates in all models whenpredicting trajectories of change in caregiver burden anddepressive symptoms across the NHA transition. Manyother potential covariates (e.g., Global Deterioration Scalescores, which measured dementia severity) would be ex-pected and were found to be correlated with baseline valueof the outcome. By including baseline outcome as a covari-ate, statistical control was approximated for many of theseother potential covariates that were not explicitly includedin the analytical model. RESULTS Model-predicted trajectories of the modified burden–ZBIscore for intervention and usual care participants are illus-trated in Figure 1. Solid lines are used to illustrate predictedvalues for caregivers where placements were observed, anddashed lines are used to illustrate predicted values for care-givers where no placements were observed. The NHA tran-sition points were the median times to placementestablished in prior evaluations of the NYUCI’s effect oninstitutionalization. 31 The statistical results of the randomeffects growth curve model for burden are presented inTable 2. The estimates in Tables 2 and 3 represent theamount of change in the outcome variable that is predictedfora1-unitchangeinthatpredictorvariable.Intheanalysisfor the modified burden score in Table 2, a strong imme-diate effect for NHAwas found. The estimate of    9.52 fortype of assessment indicates that, after accounting for theotherpredictorsascovariates,thepost-NHAmeasurementsof the modified ZBI were 9.52 units lower on average thanthe measurements obtained immediately before NHA forthe usual care participants and 10.26 units lower for inter-vention group participants.Although the estimates in Table 2 did not demonstratea significant treatment effect at 1 year after baseline, pair-wise comparisons of the treatment and usual care groups atother time points revealed significant differences. Directpairwise comparisons using the LSMEANS statement inSAS PROC MIXED (SAS Institute, Inc., Cary, NC) indi-cated that the burden of the intervention group was sig- Usual Care: Institutionalized Intervention: Institutionalized Usual Care: At-Home Intervention: At-Home Nursing home admission Figure 1.  Effects of the enhanced counseling intervention onburden trajectories (N 5 385). 424  GAUGLER ETAL.  MARCH 2008–VOL. 56, NO. 3 JAGS  nificantly lower than the burden of the usual care groupat each point after NHA ( P o .03), and the difference at3.3 years before placement closely approached statisticalsignificance (difference in estimates 5   1.56,  P 5 .06).The results of the random effects growth curve modelfor depressive symptoms are presented in Table 3. Therewas a significant effect for depressive symptom reductionafter NHA ( P o .001) and for the accelerated decrease of depressive symptoms after NHA ( P 5 .009). Treatmentgroup was also found to have a significant main effect,with a covariate-adjusted difference of 1.33 at the 1-yearpostbaseline assessment. In addition, a statistically signifi-cant effect of group-by–years since placement was found.These effects are evident in the trajectories for depressivesymptoms displayed in Figure 2. The linear slope for a de-crease in GDS scores over time was statistically significantonly for usual care participants after NHA. Covariate-ad-justed means were compared between intervention andusualcareparticipantsateachtimepoint.InterventionGDSscores were significantly lower than usual care GDS scoresat all points before NHA with the exception of baseline( P o .01). These significant differences were maintainedafter NHA for the treatment group for approximately4 months (128 days). By that point, GDS scores for theusual care group were statistically similar to GDS scores inthe treatment group for the remainder of study follow-up.The depressive symptoms of the treatment and usual caregroups were statistically similar because of the more rapiddecrease in depressive symptoms observed after NHA forthe usual care group. DISCUSSION The current analyses present a dynamic picture of howcomprehensive counseling and support can benefit spousecaregivers throughout the progression of AD and the tran- Table 2. Effects of Nursing Home Admission (NHA) and Intervention on Burden Effect Estimate Standard Error  t   Degrees of Freedom  P  -Value Intercept 27.00 0.75 35.96 380  o .001 Years after baseline (centered at 1 year)   0.87 0.24   3.59 355 .001 Years after NHA    1.15 0.30   3.90 3,069  o .001Type of assessment (NHA  5 1, community 5 0)   9.52 0.54   17.61 3,069  o .001Burden at baseline (mean centered) 0.57 0.03 16.85 3,069  o .001Baseline burden by years after baseline   0.10 0.01   9.94 3,069  o .001Caregiver sex (female 5 1, male 5 0) 0.69 0.70 0.99 3,069 .32Sex by years after baseline 0.32 0.20 1.56 3,069 .12Placement group (yes 5 1, no 5 0) 1.23 0.67 1.84 3,069 .07Placement group by years after baseline 1.08 0.24 4.54 3,069  o .001Treatment group (intervention 5 1, usual care 5 0)   1.08 0.67   1.63 3,069 .10Treatment group by years after baseline   0.22 0.24   0.90 3,069 .37Treatment group by years after NHA    0.31 0.36   0.85 3,069 .40Treatment group by type of assessment   0.74 0.72   1.02 3,069 .31 Note:  Estimate represents the amount of change in the outcome variable that is predicted for a 1-unit change in that predictor variable. Table 3. Effects of Nursing Home Admission (NHA) and Intervention on Symptoms of Depression Effect Estimate Standard Error  t   Degrees of Freedom  P  -Value Intercept 10.04 0.48 20.84 380  o .001 Years after baseline (centered at 1 year)   0.11 0.16   0.69 355 .50 Years after NHA    0.49 0.18   2.63 3,069 .009Type of assessment (NHA  5 1, community 5 0)   1.73 0.34   5.12 3,069  o .001Burden at baseline (mean centered) 0.69 0.03 20.26 3,069  o .001Baseline burden by years after baseline   0.06 0.01   5.55 3,069  o .001Caregiver sex (female 5 1, male 5 0) 0.22 0.45 0.48 3,069 .63Sex by years after baseline   0.11 0.13   0.81 3,069 .42Placement group (yes 5 1, no 5 0) 0.34 0.43 0.78 3,069 .43Placement group by years after baseline 0.16 0.15 1.07 3,069 .28Treatment group (intervention 5 1, usual care 5 0)   1.35 0.43   3.12 3,069 .002Treatment group by years after baseline   0.13 0.15   0.82 3,069 .41Treatment group by years after NHA 0.60 0.23 2.62 3,069 .009Treatment group by type of assessment 0.24 0.45 0.54 3,069 .59 Note:  Estimate represents the amount of change in the outcome variable that is predicted for a 1-unit change in that predictor variable. INTERVENTION AND INSTITUTIONALIZATION  425 JAGS MARCH 2008–VOL. 56, NO. 3
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!