Aging & Mental Health Vol. 14, No.

3, April 2010, 263273

Effectiveness of a psychoeducational skill training DVD program to reduce stress in Chinese American dementia caregivers: Results of a preliminary study
Dolores Gallagher-Thompsona*, Peng-Chih Wangab, Weiling Liuac, Vinnie Cheunga, Rebecca Pengad, Danielle Chinaac and Larry W. Thompsona
a

Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, CA; bFu Jen Catholic University, Taiwan, Peoples Republic of China; cPacific Graduate School of Psychology, Palo Alto, CA; dRichmond Area Multi-Services, Inc., San Francisco, CA (Received 16 November 2008; final version received 15 October 2009) Prior research (Gallagher-Thompson, D., Gray, H., Tang, P., Pu, C.-Y., Tse, C., Hsu, S., et al. (2007). Impact of in-home intervention versus telephone support in reducing depression and stress of Chinese caregivers: Results of a pilot study. American Journal of Geriatric Psychiatry, 15, 425434.) found that an in-home behavioral management program, derived conceptually from cognitive behavioral theories (CBT), was effective in reducing caregiver related stress and depressive symptoms in Chinese American dementia caregivers (CGs). Results were promising, but a more cost-effective intervention is needed to serve this growing population. Past work also found that a psychoeducational videotaped training program based on CBT was effective in reducing stress due to caregiving in Caucasian and African American dementia family CGs (Steffen, 2000, Anger management for dementia caregivers: A preliminary study using video and telephone interventions. Behavior Therapy, 31, 281299.). To date no research has been conducted using a technological medium to deliver a similar kind of intervention to Chinese American caregivers. The present study evaluated the effectiveness of a similar but culturally tailored program in which 70 CGs were randomly assigned to a 12-week CBT skill training program delivered on a DVD, or to a general educational DVD program on dementia. Both were available in Mandarin Chinese or English as preferred. Pre post change analyses indicated that CGs did not differ on change in level of negative depressive symptoms, but positive affect was higher, and patient behaviors were appraised as less stressful and bothersome, for CGs in the CBT skill training program. They were also more satisfied with the program overall and reported that they believed they were able to give care more effectively. Results encourage further development of theoretically based interventions, delivered using modern technology, for this ever increasing group of CGs. Keywords: dementia; caregiving; technology; cognitive/behavioral therapy; psychoeducational program; Chinese American

Chinese Americans are the largest Asian ethnic minority group in the USA. At present, there are roughly a quarter million individuals from this group over the age of 65 living in the United States (US Government Factfinder, 2005). One of the largest concentrations of Chinese Americans can be found in the San Francisco Bay Area, where this study was completed. This group is expected to continue to grow exponentially over the next three decades (US Government Factfinder, 2005). As a result, age-related health issues among Chinese Americans are becoming a major concern both for families and for the health care industry. Dementia has come to the forefront as a significant health problem in the older age range in nearly all ethnic and racial groups living in this country, and it is well-known that the prevalence of dementing disorders is both age-related and has a significant negative impact on the patient and on the primary family caregiver (CG) (Dilworth-Anderson, Williams, & Gibson, 2002; Schulz & Martire, 2004; Vitaliano, Zhang, & Scanlan, 2003). Although most research has been conducted with Caucasian caregivers, evidence is accumulating that Chinese family caregivers also often suffer from depression and
*Corresponding author. Email: dolorest@stanford.edu
ISSN 13607863 print/ISSN 13646915 online 2010 Taylor & Francis DOI: 10.1080/13607860903420989 http://www.informaworld.com

anxiety disorders and from caregiving-related stress (Chou, LaMontagne, & Hepworth, 1999; Pinquart & Sorensen, 2005; Yates, Tennstedt, & Chang, 1999). Cultural differences observed in the Chinese community are known to affect the CGs reaction to the stress of family caregiving (Adams, Aranda, Kemp, & Takagi, 2002; Patterson et al., 1998), CG coping styles (Haley, Roth, Coleton, Ford, & West, 1996), and service utilization practices (Haley, Roth, Coleton, Ford, & West, 2001). For example, stigmatization (Wang et al., 2006) and lack of knowledge (Guo, Levy, Hinton, Weitzman, & Levkoff, 2000; Zhan, 2004) about dementia, scarcity of linguistically appropriate community service facilities (Hinton, Franz, & Friend, 2004), familial obligation, referred to as filial piety (Hinton, Guo, Hillygus, & Levkoff, 2000), past uncomfortable relationships with health care providers (Chow et al., 2002; Hinton et al., 2004) and general attitudes about aging (Ayalon & Arean, 2004) all have impacted negatively on the accrual of accurate knowledge about caregiving among the Chinese. The proportion of those who identify as CGs is higher among Chinese households than Caucasian households (National Alliance for Caregiving and American Association for Retired Persons, 2004).

264

D. Gallagher-Thompson et al. more detail about the Coping with Caregiving intervention program). The groups recommended that: (1) the intervention be offered in the CGs homes rather than use a small group format to deliver the program in order to protect privacy and encourage disclosure of real problems; (2) use of trained bilingual/ bicultural staff to conduct the intervention; and (3) careful translation of all materials into written Chinese (as well as English). Focus group members also selected topics and activities from the coping classes that they thought would be most helpful to Chinese CGs. The resulting intervention program was called the In-Home Behavioral Management Program (IHBMP) and it consisted of six components: (a) education about dementia and caregiving stress, appraisal, and coping; (b) techniques for managing troublesome and disruptive behaviors of the CR; (c) how to deal with ones own negative feelings and thoughts associated with caregiving; (d) developing skills to improve communication with other family members and professionals in the health care system; (e) behavioral activation techniques to increase pleasurable events in the daily lives of the CG and the CR; and (f) end-of-life issues. The program was presented in twelve 90-minute individual sessions extending over a 12- to 16-week interval, depending on scheduling issues. Homework assignments were made at each session, and a workbook was provided to encourage active practice of techniques between sessions. The IHBMP was found to be more effective than a Telephone Support (comparison) Condition (TSC). CGs in the IHBMP program were significantly less stressed and bothered by the CRs memory and behavior problems, and their level of depressive symptoms was also significantly less, than CGs in the TSC condition (Gallagher-Thompson et al., 2007). They reported significantly greater satisfaction towards their caregiving role and greater improvement in their quality of life than did CGs in the TSC condition. Further, they believed that the program would enable them to care for their loved one longer at home. These results indicate that Chinese CGs would participate in, and benefit from, intervention programs designed to help them deal with the stress of caregiving, if the program was culturally appropriate. However, the intervention used here was labor intensive, requiring considerably more staff time than our customary small group Coping with Caregiving classes. Thus, the IHBMP does not provide a costeffective program for helping a large number of Chinese CGs, who often are reluctant to discuss with others outside the family the problems and stress they encounter in caregiving. Study II Recent work with rural CGs using a comparable psycho-educational program delivered in a series of videotapes demonstrated that this can be an effective

Traditional views of the Chinese family hold that caring for the elderly in the home is a moral and social responsibility (Lan, 2002; Wang et al., 2006) and that placement of the elder in a residential environment may be construed as a failure of filial duty (Lan, 2001). According to filial piety adult children must assume the responsibility of caring for their aging parents, but in the present sociocultural climate of rapid change in China (including more out-migration and greater exposure to Western values and beliefs), this obligation may be associated with greater caregiver burden and distress than what would have been experienced in earlier times (Chou et al., 1999; Fuh, Wang, Liu, Liu, & Wang, 1999). Cultural expectations for the care of dependent older adults influence the way family caregivers perceive caregiving duties (Yu et al., 1993), yet conflict between traditional cultural expectations and what the adult child CG of today can or is willing to provide can lead to disappointment and distress in both the CG and dependent elder. In one of the few studies to ask directly about caregiver stress, a high frequency of disruptive behaviors on the part of the care recipient (CR) was reported to be extremely stressful for the CG (Fuh et al., 1999). Negative feelings about the caregiving role may be present. Compared to other ethnic groups Chinese CGs reported more intense feelings of guilt that their care is insufficient (American Association of Retired Persons, 2001). To date a number of studies have researched the broad range of interventions used to reduce distress in Caucasian CGs and certain other ethnic and racial groups (cf. Hispanic/Latino and African American/ Black CGs in the REACH II project; REACH II investigators, 2006), and comprehensive reviews as to their effectiveness as empirically supported have been published (cf. Gallagher-Thompson & Coon, 2007; Pinquart & Sorensen, 2005). However, few intervention studies have been reported with Asian CGs generally, and Chinese American CGs specifically. Recently our laboratory completed a randomized clinical trial with Chinese American CGs, using a psychoeducational intervention derived from cognitive and behavioral theories and therapies (Beck, Rush, Shaw, & Emery, 1979; Lewinsohn, Youngren, Munoz, & Zeiss, 1986); results are found in GallagherThompson et al. (2007), which is briefly described below as Study I. This provided the foundation for Study II, which is reported in detail in this article. Study I Results of focus groups with Chinese American health care professionals indicated that our Coping with Caregiving psychoeducational model (with its emphasis on learning to manage stress and build coping skills) would be a suitable intervention strategy for Chinese American CGs if it were tailored in order to be culturally appropriate (see Gallagher-Thompson et al., 2003b for discussion of the tailoring process and for

Aging & Mental Health intervention tool (Gant, Steffen, Silberbogen, & Gallagher-Thompson, 2001; Steffen, 2000; Steffen, Mahoney, & Kelly, 2003). Encouraged by these studies, we decided to explore the feasibility of using video technology for providing a modified IHBMP. Key points for CG skill training were illustrated on a DVD for home use. A workbook to accompany the new DVD was also created to amplify the material with practice assignments, and to encourage more indepth learning of coping strategies. The DVD vignettes illustrating coping strategies (and the workbook) were created under the supervision of the first author, over a 15 month period. Vignettes were performed by the volunteer Hua Yi Performing Troupe of San Francisco. The Mandarin dialect was used for audio and either English or Chinese character subtitles could be selected for participants who did not speak Mandarin. Technical production of the DVD, including filming, editing, adding English and Chinese sub-titles, and creating master discs for distribution, was done by Photozig, Inc. of Mountain View, CA.

265

Hypotheses of current study We hypothesized that this intervention designed to increase skills to deal with caregiver stress (Skill Training DVD; SKDVD) would be more effective in reducing caregiver stress than a commercial DVD designed to educate Chinese individuals about dementia (Educational DVD; EDDVD). More specifically, we thought that CG negative reactions to problematic behaviors observed in Care Recipients (CRs) would decrease, and that positive affect would increase more, in the SKDVD group than in the EDDVD group. Although this intervention is not intended to be a treatment for depression, in view of the results observed in study 1, we also expected that the level of depressive symptoms (reflected in certain factor scores of the CES-D) would decrease more in the SKDVD than in the EDDVD condition. Finally, participants in the SKDVD group were expected to report greater satisfaction with the program, and rate their understanding of caregiver-specific coping skills at a higher level, than those who received the EDDVD.

Methods Participants CGs were recruited through media advertisements, professional referrals, and informal communications within the Chinese community. Inclusion criteria were: (a) older than 21 years, (b) caring for a person with significant memory loss or deterioration in cognitive abilities for at least six months, (c) involved at least eight hours a week in caregiving activities, (d) plans to remain in the area for the duration of the study (6 months), and (e) has access to a phone. Both family members and non-family members were recruited. Additionally, CGs had to be willing to be randomly

assigned to one of two intervention conditions. They were excluded if unable to understand and follow instructions for completing questionnaires, were unable to use a DVD player, reported that their health was so poor that they could not engage in the project without assistance, or reported evidence of severe depression or suicidal ideation/intention. In the 6-month recruitment period, 107 caregivers inquired about the study. Following initial phone screening, 100 participants who met criteria for inclusion and agreed to continue were mailed a packet including a consent form and baseline measures. Of these, 17 failed to return the consent forms and other materials and another 7 returned incomplete materials and declined further involvement in the study. We were able to contact 19 of these individuals to ask a few questions about why they had not returned the materials. Approximately half (53%) felt that some questions were too personal; 84% felt that the program would be too time consuming; and 74% thought that it would not meet their caregiving needs. The remaining 76 CGs were randomly assigned to either the skill training (SKDVD) intervention (N 40) or the education DVD (EDDVD; N 36) which was the control condition. After receiving the DVD for their intervention condition, six (8%) additional participants withdrew from the study because of time constraints. Two were in the EDDVD group and four were in the SKDVD group. Demographic information of CGs who completed the intervention (N 70) was compared to data from the six CGs who withdrew, and no significant differences were found on any of these measures (data not shown; available on request). As soon as consent forms and baseline questionnaires were returned (in a self-addressed postage-paid envelope), they were checked for completeness and follow up phone calls made to clarify responses and/or obtain missing data. In addition, a check for $25 was mailed as a token of appreciation for their time. Treatment assignment was then determined and CGs were mailed a packet containing the SKDVD or EDDVD, workbook, and instructions for how to implement the program. Randomization was carried out by a trained administrative assistant who did not see the baseline measures and was not briefed on the intervention procedures. Within one to two weeks, participants were called to see if the materials arrived and to address any problems or questions about the program. Sixteen weeks after baseline, CGs were sent the packet of post measures and asked to return them in the enclosed self-addressed postage-paid envelope. Returned questionnaires were checked for completeness. CGs were mailed a thank you note and a second check for $25 for completing the post measurement package. Upon request, CGs in the EDDVD condition were mailed a copy of the SKDVD package following their post intervention. No further follow-up has been done with these CGs at this time.

266 Intervention conditions

D. Gallagher-Thompson et al. allowed to select which language they preferred; only one caregiver selected the English version, and that caregiver did not complete the program (her discontinuing was not, however, related to the DVD but to her care-recipients declining health). Both DVDs were approximately 90 min in length and could be viewed in segments or in their entirety. Written educational materials about dementia (in English and Chinese) were included to supplement the DVD. Periodic phone calls were also included in the protocol for both conditions. CGs were called at 34 week intervals to inquire how often and under what circumstances they were viewing the DVD and whether or not it was helping them. All CGs were called an average of three times. CGs reported that calls were helpful to keep them on track though no advice or problem solving was done. Staff did enquire about level of current stress and general functioning on each call. At the programs conclusion, referrals for ongoing care were provided to those who expressed such a need during the calls. In addition, referrals for treatment of depression were also made. We found that 15 of 34 (44%) participants in the control group and 12 of 36 (33%) in the treatment group had a score of 16 or greater on the CES-D at the end of the intervention. This difference was not statistically significant (Fishers exact test, p 0.462). Of this group, 22 (12 in the control and 10 in the treatment condition) selfreported having problems with depression. All were given a referral to an appropriate Chinese-oriented mental health service at the conclusion of their participation.

The SKDVD was designed to illustrate, through roleplaying and through prepared comments by a narrator, preferred ways to handle difficult behaviors of the CR and stressful family situations. Content selection was based on two sources: (1) qualitative review of comments given at exit interviews conducted with CGs in the first study; and (2) input from an Advisory Committee comprised of Chinese professionals working with dementia patients and their families in the Chinese community who volunteered their time to work with key concepts and decide how they could best be illustrated in the DVD format. A group of Chinese actors and performers volunteered their time to do the role plays; the project coordinator (PCW) served as the DVD narrator; and filming occurred in several settings (home, office, outdoors). The CRs portrayed in the vignettes were a mother, father or spouse with dementia; various actors played the caregiver role, so that CGs of different relationships could identify with the characters. Grandchildren were included in several scenes, as well as spouses and adult daughters and sons. Filming took about eight sessions to complete, after storyboards of the content had been created in Chinese. Editing took an additional 4 months. Chinese and English sub-titles were added to the final version and the workbook was created to amplify content of the DVD. This DVD contains several tracks including: (a) information about dementia and caregiver stress; (b) how to recognize and change stressful behaviors of the CR (and/or learning to change CG reactions to these behaviors); (c) how to communicate more effectively with other family members about caregiving issues and how to obtain assistance if needed; (d) how to talk with health care providers about the CRs health issues and problematic behaviors; (e) how to access community resources along the continuum of care to obtain appropriate services when needed; and (f) how to deal with legal issues and prepare for placement or other problems near end-of-life. The narrator explains what is happening in each scene and why certain responses are better or not so good to facilitate understanding and CG effectiveness in dealing with problems that produce stress. The entire DVD plays for about 2.5 h. CGs were encouraged to watch one segment at a time, use the workbook to strengthen their learning, and do the home practice exercises outlined in the workbook. The EDDVD condition had two DVDs with comprehensive information about dementia: how to recognize it, how it is diagnosed, what current treatments are, and how CGs become stressed. One was made by the Alzheimers Association in Taiwan and was recorded in Mandarin Chinese; the other came from the Eastern North Carolina chapter of the Alzheimers Association and was recorded in English. Contents were compared by bilingual staff and were determined to be very similar. CGs therefore were

Measures Sociodemographic characteristics A demographic questionnaire used in the REACH I project (Wisniewski et al., 2003) was modified and translated to obtain information on age, gender, education, occupation, country of birth, years in the US, preferred language usage, living conditions, and health status for the CG and CR, as well as their relationship to one another. Depressive symptoms The Center for Epidemiological Studies Depression scale (CES-D; Radloff, 1977) in Chinese was used to assess level of depressive symptoms before and after intervention. This 20-item scale inquires about the frequency, within the past week, of affective, cognitive, and somatic symptoms associated with depression. Reliability as a measure of change with older adults has been demonstrated (Hertzog, Van Alstine, Usala, & Hultsch, 1990) and it has been used in cross-cultural research (Gupta & Yick, 2001). Estimates of internal consistency using Cronbachs alpha are high across a variety of populations typically between 0.85 and 0.90. In the present study alpha was 0.93.

Aging & Mental Health Positive affect subscale of the CES-D We also report results on the positive affect subscale to index change in this dimension. The four items ask about how often in the past week CGS enjoyed life, felt just as good as other people, hopeful about the future and happy. Coefficient alpha for these items was 0.87 which is in an acceptable range for research purposes. Stress associated with memory and behavior problems The Chinese translation of the Revised Memory and Behavior Problems Checklist (RMBPC; Teri et al., 1992) was used to assess how much CGs were upset or bothered by the occurrence of everyday problem behaviors before and after intervention. The scale contains 24 items (7 for memory problems, 8 for depressive symptoms of the CR, and 9 for disruptive behaviors) frequently observed in patients with dementia. For each item CGs are asked to indicate if it occurred within the past week, and if so, how much did it upset or bother them, on a 5-point scale ranging from not at all 0 through extremely 4. A conditional bother score is calculated, which is the average upset or bother rating for only the problematic behavior items that occurred. Internal consistency is high in this and our previous study with Chinese Americans (Cronbachs  0.927 and 0.930 respectively). Internal consistency for the subscales of the RMBPC, were also in the acceptable range (Cronbachs  0.866 for memory items;  0.820 for disruptive behaviors; and  0.876 for depressive symptoms). Program evaluation questionnaire This is a modified version of a similar questionnaire designed for Study I. The current measure contains 14 items: 5 are general (I feel more confident in my role as CG) and the rest inquire about what the program helped them to understand (e.g., memory loss and its effect on people or skills for monitoring and changing behavior). All but one item is rated on 4-point scales (from strongly disagree to strongly agree); the overall program satisfaction item is rated on a 5-point scale (from not at all satisfied to completely satisfied).

267

used to examine the pattern of change for variables with slopes indicating a significant main effect for treatment. Treatment groups were contrast coded so that EDDVD 0.5 and SKDVD 0.5. Baseline values for the CES-D and RMBPC measures included in the regression analyses were centered at the mean as suggested by Kraemer and Blasey (2004). Separate regression analyses were run with the relevant centered variable and experimental condition entered as independent variables and the follow-up score for the relevant outcome measure as the dependent variable. Effect size for each measure was obtained by calculating the mean and SD of the individual slopes.

Results Table 1 contains the demographic data for CG/CRs grouped according to which condition they were randomly assigned. Both groups were comparable on all measures except the number of years CRs had lived in the US. Mean years in the US for CRs in the SKDVD group was 31, whereas CRs in the EDDVD group had lived in the US roughly a decade less. On average, CGs in the two groups were in their late fifties; their mean educational level was one year beyond high school. They were caring for CRs with a mean age in the low eighties and mean educational level at the high school level or slightly less. Roughly 25% of the CGs were spouses and about 30% were not immediate family relatives. Virtually all CGs preferred to speak in Mandarin (75%) or Cantonese (20%), with only 5% preferring to communicate with project staff in English. Table 2 shows pre and post treatment means and SDs for the overall CES-D, its Positive Affect subscale, and the RMBPC CB for CGs grouped by intervention condition. The mean depression score for both groups at baseline was slightly below the screen cutoff of 16 for depression, but roughly 40% of the CGs in both conditions had scores in this elevated range, and over 20% scored above 24 (the severe range). Both groups showed a decrease in the level of depressive symptoms from pre to post treatment. The mean for the EDDVD group showed less than a one point decrease, while the SKDVD group decreased by more than two points from pre to post treatment. However, linear regression analysis for this measure did not yield a significant effect for treatment on change in level of depressive symptoms (see Table 3). A comparison of the CGs scoring in the severe range was also made. Fifty percent of the CGs in the EDDVD and 67% of the CGs in SKDVD were no longer in the severe range at post. Again this difference was not statistically significant (Fisher Exact p 0.415). Thus, the interventions did not differentially affect level of depressive symptoms, as reflected in the total CES-D. The four factor scores on the CES-D were also examined. Scores on the three factors that contain

Statistical methods Demographic variables were compared for CGs who completed the intervention program and those who withdrew using t and Fishers exact probability tests. Similar comparisons were made for CGs in the treatment and control conditions. Linear regression analyses were completed to test hypotheses that the SKDVD treatment would be more effective than the EDDVD in reducing conditional bother, increasing positive affect and reducing other negative symptoms of depression. Paired t tests were

268

D. Gallagher-Thompson et al.
Table 1. Sociodemographic measures for participants according to treatment group. SKDVD Mean N 36 Caregivers Age Education Years in USA Health ratinga Household incomeb Care recipients Age Education Years in USA Health ratinga Male/female Spouse/non-spouse Family/other 59.7 13.5 24.3 2.4 5.9 83.9 12.0 31.0 2.0 6/30 9/27 25/11 SD 13.2 3.3 16.4 0.9 3.7 6.6 4.5 22.9 0.7 EDDVD Mean N 34 57.8 13.3 17.5 2.3 4.7 81.6 9.5 21.3 1.9 3/31 9/25 23/10 SD 11.6 3.4 13.1 0.8 3.7 8.2 6.2 15.4 0.8 df 68 0.631 0.263 1.90 1.02 1.47 1.33 1.94 2.06* 0.488 p 0.479c p 1.00c p 1.00c

Notes: aHealth rating: 1 poor; 2 fair; 3 good; 4 very good. bHousehold income: 4 $20,000 29,000; 5 $30,00039.000. cFisher exact probability. *p 5 0.05.

Table 2. Means and SDs for total CES-D, CES-D Positive Affect Scale and for the RMBPC-CB Scale grouped according to treatment condition. SKDVD (N 36)a Mean CES-Db Pre-treatment 13.03 Post-treatment 10.78 Positive affect Pre-treatment 8.08 Post-treatment 9.86 RMBPC CBc number items endorsed Total items pre 10.11 Total items post 11.61 RMBPC CB mean reactiond Total items pre 1.60 Total items post 1.23 SD 11.85 8.41 3.76 2.38 5.43 5.22 0.62 0.41 EDDVD (N 34)a Mean 14.74 13.94 8.21 8.44 9.12 11.00 1.56 1.54 SD 12.51 10.31 3.77 3.10 5.05 4.19 0.65 0.61

Notes: aSKDVD Skills Training; EDDVD Educational DVD; bCenter for Epidemiologic Studies Depression Scale; cRevised Memory and Behavior Problem Checklist Conditional Bother Scale. d 0 Not at all; 1 A little; 2 Moderately; 3 Very much; 4 Extremely.

Table 3. Linear regression analysis summary of primary outcome variables. B CES-D Total 2.219 Positive affect 1.470 RMBPC Total items endorsed Number 0.335 Reaction 0.308 SE 1.578 0.552 1.093 0.120 B .118 0.262 0.036 .286 p 0.164 0.010 0.760 0.012 Effect Sizea 0.15 0.49 0.10 0.51

Notes: B Unstandardized Coefficient for estimated difference in slope of change between the SKDVD and EDDVD groups. SE Standard Error of Measurement; B Standardized Beta; CES-D Center for Epidemiological Studies Depression Scale; RBMPC CB Revised Memory and Behavior Problem Checklist (Conditional Bother Scale). a Mean of individual participant slopes for the two groups divided by the SD of the slopes.

Aging & Mental Health negative items also did not show a differential treatment effect. Descriptive statistics and results of analyses are not reported on these three factors. As predicted, however, linear regression analyses indicated that the Positive Affect subscale score increased significantly more for CGs in the SKDVD group than for those in the EDDVD group (see Table 2). There was a 22% increase in the mean score for the SKDVD group, and less than a 3% increase for the EDDVD group. Paired t tests for the two groups showed the change in the SKDVD group to be highly significant (t(35) 3.549, p 0.001), while there was no significant change in the EDDVD group (t(33) 0.403, p 0.683). These results indicate that the larger change in the CES-D total score for the SKDVD group, though not statistically significant, was most likely due to change in the Positive Affect subscale score. A post-hoc comparison of the two groups using the sum of the remaining negative items revealed negligible differences between the two groups. Changes in the RMBPC CB also occurred as hypothesized. Looking first at the total number of problems endorsed, both treatment groups increased slightly but not significantly over time (Table 2). However, in looking at the mean reaction to the total number of problems, the linear regression analysis indicated that change from pre to post was significantly different for the two groups (see Table 3). This interaction between the time of assessment and group membership is illustrated in Figure 1. The mean reaction for the SKDVD group clearly decreased (t(35) 3.425, p 0.002), while it remained essentially the same for the EDDVD group (t(33) 0.051, p 0.960), indicating a differential change in the stress appraisal level reported by CGs in these two conditions. A similar pattern was apparent for the subscales of the RMBPC CB. Paired t tests showed a significant decrease in the stress reaction score for expression of depressive symptoms by the CRs in the SKDVD group (t(22) 2.538, p 0.019) but not for the EDDVD group (t(16) 0.123, p 0.903). Similarly, average stress ratings for memory problems declined significantly (t(35) 2.762, p 0.009) for the SKDVD group, while it remained essentially the same for the EDDVD (t(33) 0.319, p 0.752). Finally, for the CGs reaction to behavioral disruptions, the regression analysis also suggested a differential change across time for the two groups. Paired t tests showed a marginally significant decrease in stress reactions for behavioral disruptions in the SKDVD group (t(24) 1.994, p 0.058), but no evidence of change in the EDDVD group (t(17) 0.637, p 0.533). (Note the df is smaller and variable across the subscales, because not all the CGs reported the occurrence of problems within each subscale.) Table 4 provides means and SDs of the items included in the overall program evaluation for both interventions. CGs were asked to evaluate their
1.7 1.6 Mean reaction 1.5 1.4 1.3 1.2 1.1 Pre Post

269

SKDVD

EDDVD

Figure 1. Effect of treatment on change in mean reaction (bother or stress) to the total number of memory and behavior problems (N 70; t 2.58; p 0.010). SKDVD CBT Skill Training DVD (N 36); EDDVD Educational DVD on Alzheimers Disease (N 34).

respective treatment using a 4-point Likert scale for most of the items and a 5-point Likert scale for overall program satisfaction. The latter was significantly higher (p 5 0.001) for the SKDVD group (Very Satisfied range) than for the EDDVD group (Satisfied range). The SKDVD group rated all except four of the items more highly than the EDDVD group. Both groups agreed that the intervention program did not require too much work. Both also rated their confidence in their role as caregiver about the same. Similar ratings were also made by both groups on the statements that the intervention helped me understand memory loss and its effect on people, and helped me understand stress and its relation to caregiving and health. On the other hand CGs in SKDVD group stated that participation in the intervention enhanced my ability to care for my relative (p 5 0.01), and helped me improve my relatives life (p 5 0.05) to a greater extent than CGs in the EDDVD group. Understanding of skills that were specifically taught in the SKDVD program were all rated substantially higher by this group (p 5 0.01 or better), but this was expected, given that little emphasis was placed on these in the EDDVD intervention. In summary, positive affect increased, and level of specific stress or bother decreased, to a greater degree for CGs in the SKDVD group when compared to the EDDVD group, and the overall project evaluation indicated that the SKDVD group was more satisfied and felt that they had benefited more than those in the EDDVD group. Of interest here is the fact that roughly 30% of our participants were non-family members. Since little has been written about non-family caregivers in the Chinese community, we decided to include those who indicated an interest in being participants. Post-hoc exploratory analyses were done comparing family to non-family caregivers, in order to generate hypotheses for future research. There were

270

D. Gallagher-Thompson et al.

Table 4. Response to the overall project evaluation by CGs grouped according to intervention. SKDVD (N 36) Mean Feel more confident in my role as caregivera Enhanced my ability to care for my relative Helped me improve my relatives life Program required too much work Overall satisfaction with programb Helped me understand:a Memory loss and its effect on people Stress and its relation to caregiving and health Skills for monitoring and changing behavior How to reduce unhelpful thoughts Assertive communication End-of-life issues and advance directives When and how to use relaxation skills How and why to increase pleasant events
b

EDDVD (N 34) Mean 2.25 2.06 2.0 1.21 3.19 2.24 2.12 1.38 1.39 1.38 1.26 1.35 1.32 SD 0.55 0.49 0.49 0.68 0.83 0.43 0.59 0.65 0.75 0.74 0.71 0.65 0.68 t 0.11 2.58** 2.38* 0.86 3.56*** 0.13 1.99 5.49*** 4.33*** 4.34*** 3.34** 6.01*** 4.70***

SD 0.55 0.60 0.58 0.50 0.79 0.50 0.55 0.75 0.68 0.67 0.74 0.60 0.76

2.25 2.40 2.31 1.08 3.91 2.25 2.40 2.31 2.14 2.11 1.85 2.25 2.14

Notes: a4-point scale: 1 Strongly disagree; 2 Disagree; 3 Agree; 4 Strongly agree. 5-point scale: 1 Not at all satisfied; 2 Not very satisfied; 3 Satisfied; 4 Very satisfied; 5 Completely satisfied. *p 5 0.05; **p 5 0.01; ***p 5 0.001.

several demographic differences- for example, only one of the 20 non-family CGs lived with a CR; the remainder took care of Chinese elders who lived alone or in an assisted living facility. The non-family participants were slightly younger (mean age 53 vs. 61; t(68) 2.52, p 0.014) than the family caregivers but were comparably educated (mean number of years 12 for the non-family members and 13 for the family members; t(68) 1.67, p 0.10). They were substantially lower in level of depressive symptoms (mean CESD 6.7 for the non-family caregivers vs. 16 for the family member caregivers; t(68)3.570, p 0.001) and they were less stressed by CR problematic behaviors (mean bother 1.72 for the family members and 1.18 for the non-family members; t(68) 3.19, p 0.002). Despite these differences, trends in the outcome variables obtained for the nonfamily caregivers were similar to those obtained for family caregivers, although some values did not reach statistical significance because of high variability and small Ns. In addition, there were no significant effects for the family member versus non-family variable, and no significant interaction between family member and treatment condition, for any of the program evaluation measures listed in Table 4. Thus, at this time, it appears that non-family caregivers may derive some benefit from a psychoeducational intervention such as this, although more research is needed to test this hypothesis in a rigorous manner.

Discussion Taken together, these results indicate that Chinese American CGs, who were given a DVD and workbook combination (in their own language) that taught a variety of cognitive and behavioral skills for dealing with the stress of caregiving, were less bothered, and

reported less perceived stress, when taking care of everyday memory and behavior problems of their CR compared to CGs who received an education only DVD that emphasized information about AD but did not focus on helping caregivers learn better ways of coping. This group also reported increased positive affect compared to the CGs in the education only group. Further, the results of the overall project evaluation made it clear that the SKDVD group was more satisfied, felt that they had benefited more, and that their relative likewise improved to some extent, compared to those in the in the EDDVD condition. These findings are encouraging, and suggest greater development and testing of technology-based and/or multimedia approaches to caregiver education and training are warranted. For some caregivers, at least, this may be adequate to help them manage their stress effectively. It is noteworthy, however, that contrary to results in our first CBT-based skill training study with Chinese Americans, the current study did not find a significant decrease in overall depressive symptoms from pre to post intervention as indexed by the CES-D scale. However, the mean baseline CES-D scores for both groups were below the usual cut-off level of 16 for classification as depressed. Perhaps if the study had included only depressed CGs, the SKDVD would have had a greater impact on depressive symptoms. Alternatively, Chinese American caregivers who score above generally accepted cut-off scores for depression may require a more intensive intervention for example, referral to a mental health professional, in addition to participating in a program focused on teaching them better ways to manage caregivingrelated stress. These results also point to the need for a comprehensive assessment at the outset, before caregivers are assigned or referred to intervention

Aging & Mental Health programs, so that good matching can occur. Future research is needed that targets depression reduction in Chinese CGs perhaps a variant of the psychoeducational approach described here can be developed and evaluated, in order to shed more light on this issue. It is not surprising that this particular program was not very effective in reducing symptoms of depression over the 4-month interval of the study, considering that the DVD and workbook present information in a general way. The program is not targeted to each caregivers specific problems or concerns, the way a more individualized intervention program would be. On the plus side, this study does suggest that technologically based interventions can be helpful for reducing caregiving-related stress. Thus, although these results may be viewed as negative from one perspective, they can also be viewed as enlightening and informative from another point of view. We now have data to support the clinical observation that Chinese American caregivers who are depressed (in addition to being stressed from caregiving) should be referred for treatment and not simply placed in a program designed primarily for stress management. This is in contrast to results with other racial and ethnic groups of caregivers, where studies have found that psychoeducational programs can and do reduce depressive symptoms (see Gallagher-Thompson et al., 2003a for example). The difference in response may be due to the use of technology here, which is an impersonal medium to convey complex information, compared to individual or small group modalities generally used in prior research. It is also likely that the DVD approach may be better suited to the needs of some caregivers versus others. When it comes to the efficacy of caregiving interventions, clearly one size does not fit all (Gallagher-Thompson et al., 2009; Zhan, 2004). At the same time, there was both a significant increase in the positive affect sub-scale of the CES-D, suggesting an improved quality of life, and a reduction in everyday stress reported by caregivers who watched the SKDVD. These results are consistent with those reported in the comprehensive review of evidencebased interventions for CGs by Gallagher-Thompson and Coon (2007) in which psychoeducational skill building programs were the most effective interventions of all those that were evaluated. The current study has some limitations to consider when interpreting the findings. First, the sample size is relatively small for a randomized trial; replication with larger numbers of Chinese American caregivers both male and female would add to the generalizability of the findings. In addition, further work is needed to determine the impact of this kind of program on nonfamily members (who were a relatively small proportion of the CGs in this study) since they provide an increasing proportion of in-home care to demented older adults (Lan, 2001, 2002).

271

Second, although on each phone call the research assistants inquired about the number of times the DVDs were watched, which segments, and for how long, we cannot be certain that these data are reliable. Some caregivers watched different segments of the skill training DVD, for example, more than three or four times, and used the workbook to amplify the material and deepen their understanding, whereas the majority (about 55%) of those in the skill training condition said they watched each segment only once or twice (despite encouragement from the research assistants to watch regularly over the 4 month interval from pre to post assessment). Some said they never watched the entire DVD (about 10%). Future studies might consider including a more precise method of counting frequency of use, so that this factor could be explored for its relationship to outcome. Finally, the exploration of internet-based learning is recommended for future research. The widespread use of the internet at the present time, even among older adults, would appear to present a valuable opportunity to provide caregiver education and training in the privacy of ones home or office, which can be done in a self-paced manner. It will be very important to study whether or not such web-based presentation of material can be effective in reducing caregivers stress and burden among Chinese Americans.

Acknowledgements
Substantial support for this project was provided by funding from the national office of the Alzheimers Association (grant # IIRG-04-1109) to Dolores Gallagher-Thompson, Principal Investigator. The authors first wish to thank the Alzheimers Association chapter of Eastern North Carolina, located in Raleigh, NC - specifically, Mrs. Alice Watkins, Executive Director, who graciously provided the English language DVD that was planned for use in the comparison condition, and Dr. Frank Longo, MD, of Stanfords Neurology Department, who facilitated this exchange. That DVD is entitled: Accepting the Challenge: Providing the Best Care for People with Dementia. It is a four-module DVD containing segments describing what dementia is, improving communication skills with persons with dementia, managing problem behaviors, and scheduling meaningful activities. However, the DVD that was used virtually by all caregivers in the control condition was one produced in Mandarin Chinese, which was provided to the project by Taiwanese professionals through the efforts of Dr. PengChih Wang, PhD, of our staff. The authors express grateful appreciation to the members of the Hua Yi Performing Troupe of San Francisco, who volunteered their time to enact the role plays in this DVD. Specifically, Joanna Zhang, DVD director, spent countless hours working with the script and the actors to achieve excellent results. These are the actors from that group: Luona Wang, Angel Tu, Frank Huang, Amy Dung, Jason Tu, Jessica Chang and Jason Chang. Additional actors were Tenny Tsai-Eng, Weiling Liu, Hank Wang, Hui-Qi Tong, M.D., and Peng-Chih Wang, PhD, who served as the DVD narrator. Many thanks are also extended to Shao Chun, Sunny Liao, Yung Hui, and Vivien Lin for their tireless efforts in creating and translating the DVD workbook. Thanks are also due to Chunyu Pu, PhD, Laurie Leung, PhD, Florence

272

D. Gallagher-Thompson et al.
Gallagher-Thompson, D., & Coon, D.W. (2007). Evidencebased treatments to reduce psychological distress in family caregivers of older adults. Psychology and Aging, 22, 3751. Gallagher-Thompson, D., Coon, D.W., Solano, N., Ambler, C., Rabinowitz, Y., & Thompson, L.W. (2003a). Change in indices of distress among Latina and Anglo female caregivers of elderly relatives with dementia: Site specific results from the REACH national collaborative study. The Gerontologist, 43, 580591. Gallagher-Thompson, D., Gray, H., Tang, P., Pu, C.-Y., Tse, C., Hsu, S., et al. (2007). Impact of in-home intervention versus telephone support in reducing depression and stress of Chinese caregivers: Results of a pilot study. American Journal of Geriatric Psychiatry, 15, 425434. Gallagher-Thompson, D., Haley, W., Guy, D., Rupert, M., Arguelles, T., Zeiss, L., et al. (2003b). Tailoring psychological interventions for ethnically diverse caregivers. Clinical Psychology: Science and Practice, 10, 423428. Gallagher-Thompson, D., Lonergan, K., Holland, J., China, D., & Ashford, A.J.W. (2009). Supporting family caregivers. In M.F. Weiner, & A.M. Lipton (Eds.), Textbook of Alzheimers disease and other dementias (pp. 353366). Washington, DC: American Psychiatric Publishing. Gant, J.R., Steffen, A.M., Silberbogen, A.K., & GallagherThompson, D. (2001). The dementia caregiving skills program: Reducing distress and enjoying time with your family member (coachs manual and accompanying video). St. Louis, MO: University of Missouri at St. Louis. Guo, Z., Levy, B.R., Hinton, W.L., Weitzman, P.F., & Levkoff, S. (2000). The power of labels: Recruiting dementia affected Chinese-American elders and their caregivers. Journal of Mental Health and Aging, 6, 103112. Gupta, R., & Yick, A. (2001). Validation of CES-D Scale for older Chinese immigrants. Journal of Mental Health and Aging, 7, 257272. Haley, W.E., Roth, D.L., Coleton, M.I., Ford, G.R., & West, C.A. (1996). Appraisal, coping, and social support as mediators of well-being in Black and White family caregivers of patients with Alzheimers disease. Journal of Consulting and Clinical Psychology, 64, 121129. Haley, W.E., Roth, D.L., Coleton, M.I., Ford, G.R., & West, C.A. (2001). Racial, ethnic, and cultural differences in the dementia caregiving experience: Recent findings. The Gerontologist, 41, 334347. Hertzog, C., Van Alstine, J., Usala, P.D., & Hultsch, D.F. (1990). Measurement properties of the Center for Epidemiological Studies Depression Scale (CES-D) in older populations. Psychological Assessment, 2, 6472. Hinton, L., Franz, C., & Friend, J. (2004). Pathways to dementia diagnosis: Evidence for cross-ethnic differences. Alzheimers Disease & Associated Disorders, 18, 134144. Hinton, L., Guo, Z., Hillygus., J., & Levkoff, S.E. (2000). Working with culture: A qualitative analysis of barriers to recruitment of Chinese-American family caregivers for dementia research. Journal of Cross-Cultural Gerontology, 15, 119137. Kraemer, H.C., & Blasey, C.M. (2004). Centring in regression analyses: A strategy to prevent errors in statistical inference. International Journal of Methods in Psychiatric Research, 13, 141151.

Kwo, Susan Long, Catherine Sunderland, and most importantly, Mr. John Di Mario, for technical support, and finally, to Ms. Danielle China for her skillful database management. The authors also wish to acknowledge the contribution of individuals from the following agencies who volunteered their time and input throughout the process of creating, and evaluating, this new DVD: Edie Yau, Diversity Program Manager, and Herching Ku, Helpline Volunteer, from the Northern CA/ Northern NV chapter of the Alzheimers Assn; Traci Wei of the Asian Network Pacific Home Care, Oakland, CA; May Au, LCSW and Diane Diep Ngo of SelfHelp for the Elderly, San Francisco, CA; Lina Chen Pan and Virginia Chan from Self-Help for the Elderly in Santa Clara, CA; Emily Poon, Executive Director of Aegis Gardens Assisted Living and Memory Care (specializing in serving the Chinese community) in Fremont, CA; Marsha Fong from the Office of Aging and Adult Services of San Mateo County; Regina Kwo from On-Lok, Inc., in San Francisco, CA; Teresa Mo, Cupertino Senior Center, Cupertino, CA; and last but not least, Tenny Tsai-Eng, Area and National Alzheimers Association Board Member. Finally, this entire project would never have been completed without the extraordinary talent and skills of Mr. Bruno Kajiyama, Executive Director of Photozig, Inc., of Mountain View, CA, which is the company that produced and edited this DVD and formatted it suitably for wider distribution by the national office of the Alzheimers Association. Copies of the DVD and workbook can be ordered for a nominal fee through the website of the Stanford Geriatric Education Center (http://sgec.stanford.edu) and from the Diversity Resources section of the national Alzheimers Association website (www.alz.org).

References
Adams, B., Aranda, M.P., Kemp, B., & Takagi, K. (2002). Ethnic and gender differences in distress among Anglo American, African American, Japanese American, and Mexican American spousal caregivers of persons with dementia. Journal of Clinical Geropsychology, 8, 279301. American Association of Retired Persons (2001). In the Middle: A Report on Multicultural Boomers Coping with Family and Aging Issues. Retrieved January 14, 2004 from http://research.aarp.org/il/in_the_middle.html Ayalon, L., & Arean, P.A. (2004). Knowledge of Alzheimers disease in four ethnic groups of older adults. International Journal of Geriatric Psychiatry, 19, 5157. Beck, A.T., Rush, J., Shaw, B., & Emery, G. (1979). Cognitive therapy of depression. NY: Guilford Press. Chou, K.-R., LaMontagne, L., & Hepworth, J.T. (1999). Burden experienced by caregivers of relatives with dementia in Taiwan. Nursing Research, 48, 206214. Chow, T.W., Liu, C.K., Fuh, J.L., Leung, V.P., Tai, C.T., Chen, L., et al. (2002). Neuropsychiatric symptoms of Alzheimers disease differ in Chinese and American patients. International Journal of Geriatric Psychiatry, 17, 2228. Dilworth-Anderson, P., Williams, I.C., & Gibson, B.E. (2002). Issues of race, ethnicity, and culture in caregiving research: A 20-year review (19802000). The Gerontologist, 42, 237272. Fuh, J.L., Wang, S.J., Liu, H.C., Liu, C.Y., & Wang, H.C. (1999). Predictors of depression among Chinese family caregivers of Alzheimer disease patients. Alzheimer Disease & Associated Disorders, 13(3), 171175.

Aging & Mental Health

Lan, P.C. (2001). Subcontracting filial piety: Elder care in dual-earner Chinese immigrant households in the Bay Area. Berkeley, CA: Center for Working Families. Lan, P.C. (2002). Subcontracting filial piety: Elder care in ethnic Chinese immigrant families in California. Journal of Transcultural Nursing, 13(3), 202209. Lewinsohn, P.M., Youngren, M.A., Munoz, R., & Zeiss, A.M. (1986). Control your depression. New York: PrenticeHall. National Alliance for Caregiving and AARP (2004). Caregiving in the US. Patterson, T.L., Semple, S.J., Shaw, W.S., Yu, E., He, Y., Zhang, M.Y., Wu, W., & Grant, I. (1998). The cultural context of caregiving: A comparison of Alzheimers caregivers in Shanghai, China and San Diego, California. Psychological Medicine, 28, 10711084. Pinquart, M., & Sorensen, S. (2005). Ethnic differences in stressors, resources, and psychological outcomes of family caregiving: A meta-analysis. The Gerontologist, 45, 90106. Radloff, L.S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement, 1, 385401. REACH II investigators (in alphabetical order: Belle, S.H., Burgio, L., Burns, R., Coon, D., Czaja, S.J., GallagherThompson, D., et al.). (2006). Enhancing the quality of life of dementia caregivers from different ethnic or racial groups: A randomized, controlled trial. Annals of Internal Medicine, 145(10), 727738. Schulz, R., & Martire, L. (2004). Family caregiving of persons with dementia: Prevalence, health effects, and support strategies. American Journal of Geriatric Psychiatry, 12, 240249. Steffen, A.M. (2000). Anger management for dementia caregivers: A preliminary study using video and telephone interventions. Behavior Therapy, 31, 281299. Steffen, A.M., Mahoney, D., & Kelly, K. (2003). Capitalizing on technological advances to help distressed caregivers. In D.W. Coon, D. Gallagher-Thompson, &

273

L.W. Thompson (Eds.), Innovative interventions to reduce dementia caregivers distress: A sourcebook and clinical guide (pp. 189209). NY: Springer. Teri, L., Truax, P., Logsdon, R., Uomoto, J., Zarit, S., & Vitaliano, P.P. (1992). Assessment of behavioral problems in dementia: The revised memory and behavior problems checklist (RMBPC). Psychology and Aging, 7, 622631. US Government: Fact Finder Census Information (February 23, 2005). http://factfinder.census.gov/servlet/ DatasetMainPageServlet?_programDEC&_lang en&_ ts.2005. Vitaliano, P.P., Zhang, J., & Scanlan, J.M. (2003). Is caregiving hazardous to ones physical health? A metaanalysis. Psychological Bulletin, 129, 946972. Wang, P.-C., Tong, H.-Q., Liu, W., Long, S., Leung, L.Y.L., Yau, E., & Gallagher-Thompson, D. (2006). Working with Chinese American families. In G. Yeo, & D. Gallagher-Thompson (Eds.), Ethnicity and the dementias (2nd ed., pp. 173188). NY: Routledge/Taylor & Francis Group. Wisniewski, S.R., Belle, S.H., Coon, D.W., Marcus, S.M., Ory, M.G., Burgio, L.D., et al. (2003). The Resources for Enhancing Alzheimers Caregiver Health (REACH): Project design and baseline characteristics. Psychology and Aging, 18, 375384. Yates, M., Tennstedt, S., & Chang, B. (1999). Contributors to and mediators of psychological well-being for informal caregivers. Journals of Gerontology Series B, Psychological Sciences and Social Sciences, 54, P12P22. Yu, E.S., Liu, W., Wang, Z., Levy, P.S., Katzman, R., & Zhang, M. (1993). Caregivers of the cognitively impaired and the disabled in Shanghai, China. In S.H. Zarit, L.I. Pearlin, & K.W. Schaie (Eds.), Caregiving Systems: Formal and Informal Helpers (pp. 530). New Jersey: Lawrence Erlbaum Assoc. Zhan, L. (2004). Caring for family members with Alzheimers disease: Perspectives from Chinese American caregivers. Journal of Gerontological Nursing, 13, 1929.

Copyright of Aging & Mental Health is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.