Development and Validation of the Brief Assessment Scale for Caregivers in Chinese

Abstract

Background:

Chinese family caregivers in the United States experience high rates of burden. Studies in this population require culturally appropriate and valid instruments.

Objective:

To adapt a Chinese version of the Brief Assessment Scale for Caregivers (BASC), a tool measuring positive and negative effects of caregiving.

Methods:

New items for the Chinese adaptation were generated from the literature, an interdisciplinary professional group, and caregiver focus groups. The new items were translated into Chinese, added to the original 14-item BASC, then administered to caregivers in two surveys (n=205); participants also completed Chinese versions of the Caregiver Burden Inventory-Chinese (CBI-C), the Hospital Anxiety and Depression Scale (HADS), and the Katz Index for Activities of Daily Living. Item reduction and exploration of internal consistency, reliability and validity using correlational, exploratory, and confirmatory factor analysis were performed.

Results:

Factor analysis and item-total correlations supported reduction in the number of new items. The final BASC-Chinese (BASC-C) included 10 new items plus the 14 items from the original tool with an overall α coefficient of 0.79 for the 24-item scale. Three Chinese-specific factors—cultural strain, decision-making strain, and emotional strain—showed α coefficients of 0.69 to 0.79. Significant first order correlations were found between the BASC-C and previously validated measures (total BASC-C with HADS Anxiety [r=0.64, p<0.001], HADS Depression [r=0.60, p<0.001], and CBI-C [r=0.60, p<0.001]). Partial correlations indicated that the new Chinese items provided information supplemental to the original BASC.

Conclusion.

This study suggests that the BASC-C is a valid measure of caregiver burden among Chinese family caregivers.

Introduction

Chinese Americans comprise a large and rapidly changing population. According to the 2012 census, there are 17.3 million U.S. residents of Chinese descent. This represents a 46% growth, above any other racial group in the country. The largest concentrations of Chinese Americans reside in California and New York.¹ In an analysis of trends in New York City, the Asian American Foundation reported that most Chinese immigrants are older (72.4 years); female (56%); and widowed, single or divorced (50%).² Poverty (18.5%) and near-poverty (40.4%) rates are high, and most first-generation immigrants have less than a high school education (70%) and limited English fluency (50%).³ Recent immigrants are therefore characterized by poor acculturation and high rates of economic and social vulnerability.

For Chinese Americans, family responsibility for a sick relative resonates with a norm of reciprocal intergenerational assistance. Chinese Americans count on their adult children for assistance, while parents provide high levels of instrumental and financial support for children.² Providing care to a family member who is seriously ill is expected.

The challenges associated with family caregiving are multifactorial.⁴ Responsibilities include assistance with activities of daily living, decision making, and emotional support.⁵ The capabilities of family caregivers and the potential for distress vary, and may be influenced by cultural factors. A systematic review of 116 studies found Chinese American caregivers providing significantly higher numbers of caregiving tasks and hours while using lower levels of support services compared to white, African American, and Hispanic caregivers.⁶ In other studies, Chinese caregivers refused help because they felt “too proud,” “didn't want outsiders coming in,” or found “bureaucracy too complex.”^7,8 These findings imply Chinese American caregivers may be at high risk of distress or burden.

Few studies have characterized these outcomes, and research is needed to profile the needs of this population, develop culturally appropriate instruments and interventions.^9,10 The present study was conducted to adapt and validate a Chinese version of the Brief Assessment Scale for Caregivers (BASC), a tool with minimal administration burden measuring both positive and negative effects of caregiving.¹¹ Although other validated caregiver burden tools exist for Chinese caregivers, this tool goes beyond existing tools in measuring cultural and caregiver strain for a variety of medical illnesses.

Methods

The study was approved by the Institutional Review Board of Beth Israel Medical Center. All participants signed informed consent prior to participation.

Item generation

New items for the Chinese adaptation of the BASC (BASC-C) were developed from a literature review; input from an interdisciplinary group of physicians, nurses, social workers, and geriatricians with years of experience; and data from two focus groups. The focus groups, convened at community centers in New York City's Chinatown, included 11 Cantonese-speaking caregivers and 12 Mandarin-speaking caregivers. Each group responded to open-ended questions from a professional group facilitator. Responses were recorded and transcribed in Chinese. Transcripts were translated into English by two translators fluent in Chinese and English using a Translation, Review, Adjudication, Pre-testing and Documentation (TRAPD) model.¹² Each translator created a separate English version of the transcript, followed by a joint review of both versions. Discrepancies were reconciled, a final version of the transcript was reviewed by a native Chinese speaker for coherence and meaning, and further corrections were made.

Focus group transcripts were independently and jointly reviewed by two of the investigators (M.G. and C.T.), who are English- and Cantonese-speaking, respectively. Using an iterative process, they reviewed the original transcripts looking for subthemes and concepts related to caregiving in general and specific concerns related to language or culture. The process was repeated until saturation occurred and data analysis returned no new themes or information. Theme definition and naming completed the process. Concerns included alienation within the Western medical system, intimidation when placed in the role of translator, stress as decision maker between patient and staff, and conflicts between Chinese traditions and Western medicine. These themes were used to generate 24 new questions. Each item was developed on a four-point Likert Scale ranging from “not at all” to “a lot,” “ not at all distressed” to “a lot of distress,” and “agree a lot” to “disagree a lot.”

The new questions were translated back to Chinese by the same translators using the TRAPD model. Concurrently, these procedures were used to translate the original 14 BASC items into Chinese. The BASC was previously validated in English with an overall Cronbach α of 0.70 and a range of 0.58 to 0.80. One factor, the five-item Negative Personal Impact (NPI) factor had sufficiently high reliability (Cronbach α=0.80) to be considered as a stand-alone measure. The initial pool of test items called the BASC-C comprised 38 questions in total.

Measures

In addition to the BASC-C, the questionnaire included a demographic survey, the Hospital Anxiety and Depression Scale (HADS), the Caregiver Burden Inventory-Chinese (CBI-C), and the Katz Index of Independence in Activities of Daily Living (ADL). These instruments were included as comparison measures for establishing construct validity of the BASC-C. The HADS is a widely used 14-item scale with anxiety and depression subscores. The Chinese version, tested in 154 patients with coronary heart disease,¹³ yielded a Cronbach α coefficient of 0.82 (depression) and 0.77 (anxiety). The CBI-C is a 14-item scale that measures caregiver burden in five domains. It was validated in 150 Chinese caregivers,¹⁴ yielding a Cronbach α of 0.95. In contrast to the BASC-C, the CBI-C was validated on nonimmigrant Chinese caregivers of dementia patients in Hong Kong and Taiwan, with no measure of cultural strain. The Katz Index of Independence in ADL measures functional status in six areas: bathing, dressing, toileting, transferring, continence, and feeding. The Chinese version has a Cronbach α of 0.85.¹⁵ The Marin Acculturation Scale's five-item language use subscale was used to assess proficiency in and preference for speaking and reading language. This subscale has good internal consistency (Cronbach α is 0.92) and correlates significantly with generational level, time in the United States, and ethnic identity. Finally, the packet included a questionnaire that covered living arrangements and demographics.

Surveys

Two convenience samples of Chinese caregivers were recruited from inpatient, ambulatory, and community settings (n=102 and n=103). The first survey was undertaken for the purposes of exploratory analysis, the second, for confirmatory analysis. All respondents were Chinese-speaking adults who self-identified as the primary caregiver of a patient with chronic or advanced medical illness; those with psychiatric or cognitive impairments preventing them from giving consent were excluded. The procedures for the two surveys were identical. A trained research associate obtained signed informed consent, recorded relevant sociodemographic information, and administered the study questionnaires in person. Caregivers were given the option of participating in Mandarin, Cantonese, or English.

Data analysis

Responses to the test items from the first survey were submitted to exploratory factor analysis (EFA) with principal components extraction and varimax rotation. Factors were retained if they had two or more items and an internal consistency of 0.60 or above. Data from the second survey were subjected to confirmatory factor analysis (CFA) to test whether the factors from the initial EFA could be replicated. The goodness of fit Index (GFI) and the adjusted goodness of fit index (AGFI) were calculated for the overall model to determine whether the results indicated an adequate fit of the model to the original correlational data. Items were retained or dropped from the model based on whether (1) the parameter estimates yielded by the model were valid (non-zero error terms) and (2) the intercorrelations between item score and total score were greater than 0.

Pearson correlations were calculated to assess the validity of the BASC-C and its underlying factors, including the original and newly developed Chinese items. These were compared with the HADS, CBI-C, and Katz ADL scales and with demographic and clinical characteristics thought to be associated with higher burden (female gender, adult child, depression, gastrointestinal disorders, hypertension). Finally, to determine whether the new Chinese-specific items contributed information about caregiver burden that was independent of the existing BASC items, multiple regression was used to test whether the new items still correlated significantly with the validation measures after controlling for the original 14 items.

Results

Participants in the first (n=102) and second (n=103) surveys had similar characteristics (Table 1). The mean age was 50 (range, 22–85). Approximately 13% ranged in age from 18 to 30. The majority were married women living with a spouse. Approximately 40% were employed full-time. A large majority spoke Chinese at home (99%) and chose the Chinese version of the interview (80%). Approximately one-third was caring for a spouse; more than 40% were caring for a parent. Most of the patients had cancer or heart disease. Approximately one-third of the caregivers reported depression or high blood pressure, and one-fifth had gastrointestinal problems.

Table 1.

Characteristics of Caregivers in the First and Second Surveys

Characteristic	First survey (n=102)	Second survey (n=103)
Setting of survey
Community center	—	6 (6%)
Inpatient	77 (75%	57 (55%)
Outpatient	25 (25%)	43 (42%)
Caregiver age	51.5±17.6^a	53.7±16.3
Patient age	71.2±16.7	72.7±16.3
Relationship to patient
Spouse	37 (36%)^b	36 (35%)
Daughter/son	46 (45%)	52 (50%)
Parent	2 (2%)	3 (3%)
Sibling	4 (4%)	4 (4%)
Other	15 (15%)	8 (8%)
Female	59 (58%)	62 (60%)
Years in United States	23.0±12.0	24.6±13.5
Interview form administered in Chinese	82 (80%)	80 (78%)
Home language
Cantonese	72 (71%)	65 (64%)
Fukinese	9 (9%)	24 (24%)
Mandarin	4 (4%)	11 (11%)
Other Chinese	15 (15%)	1 (1%)
English	6 (6%)	1 (1%)
Marital status
Married	81 (79%)	74 (73%)
Single	15 (15%)	21 (21%)
Other	6 (6%)	7 (7%)
Caregiver living arrangement
Patient	63 (62%)	57 (55%)
Spouse	58 (57%)	39 (38%)
Children	50 (49%)	41 (40%)
Parents	16 (16%)	9 (9%)
Education
None	5 (5%)	1 (1%)
<High school	45 (44%)	53 (54%)
High school/tech school	23 (23%)	16 (16%)
Junior/some college	13 (13%)	13 (13%)
College	12 (12%)	13 (13%)
Some post-college	1 (1%)	—
Advanced degree	3 (3%)	3 (3%)
Employment
Full-time	40 (40%)	43 (42%)
Unemployed	13 (13%)	12 (12%)
Part-time	10 (10%)	2 (2%)
Homemaker	8 (8%)	10 (0%)
Retired	28 (26%)	27 (26%)
Other	3 (3%)	1 (1%)
Weekly hours of care	25 (5, 119.5)	28 (1, 168)
Primary diagnosis of patient
Cancer	39 (38%)	23 (22%)
Heart disease	28 (27%)	32 (31%)
Arthritis	5 (5%)	9 (9%)
Pulmonary	4 (4%)	9 (9%)
Gastrointestinal	4 (4%)	10 (10%)
Mental health	3 (3%)	—
Other^c	19 (18%)	20 (19%)

Mean±standard deviation.

Frequency (%).

Parkinson's disease, renal failure, embolism, pain.

Item reduction

The EFA applied to the 24 new items yielded 12 factors. Four factors were discarded because they consisted of only one item each; two were discarded because of Cronbach α less than 0.60. The remaining six-factor solution accounted for 90% of the overall variance of items. The α coefficient for all 16 items was 0.77 with α coefficients for the group of six factors ranging from 0.63 to 0.82.

Review of the individual items loading on each of the six factors revealed meaningful constructs: Factor 1 captured the conflict between traditional Chinese culture and the culture embodied in the health system; Factor 2 described distress associated with medical decision-making; Factor 3 reflected personal stress in dealing with the loved one; Factor 4 described shame for not taking care of the family member; Factor 5 depicted traditional beliefs in taking care of the loved one; and Factor 6 reflected concerns about medical bills and dealing with physicians.

Based on the results from the first survey, eight items were eliminated due to lack of internal consistency; two items were retranslated to improve clarity. Following these changes, a second version of the survey was administered for final item selection and validation. CFA was then performed on data from the 30 items used in the second survey, to determine whether the factor structure from the EFA could be confirmed or whether further item reduction was necessary. The GFI and AGFI were 0.49 and 0.32, respectively, indicating a good fit to the data. The item “worried about how to pay for medical care” was found to have a negative error variance indicating misspecification; because only one item remained, the factor was eliminated. The GFI and AGFI of the resulting five-factor model were 0.51 and 0.34, with no negative error variances for any items.

Further exploration of the item-total correlations and alpha coefficients indicated additional item reduction was possible. Item-total correlations revealed values that were negative or close to zero. Elimination of these items resulted in a three-factor model consisting of 11 items. An additional item was eliminated based on the α coefficient for Factor 3, which showed a substantial increase in internal consistency when the item “felt hurt when ____ gets angry at me” was eliminated.

The final three-factor model comprising 10 items had acceptable levels of fit to the data (GFI=0.56; AGI=0.32; Table 2). The α coefficient for the total BASC-C consisting of the original 14 BASC items and the 10 Chinese-specific items was 0.79. The coefficients for the three factors derived from the 10 Chinese-specific items ranged from 0.69 to 0.79. The α coefficient for the original BASC was 0.76; the α for the NPI factor was 0.80. The α coefficients for the remaining factors were in the adequate to fair range of reliability, similar to those for the original BASC. The level of reliability of the NPI factor was high enough for it to be considered as a stand-alone measure of caregiver burden, while the remaining factors were more appropriate for administration as part of an integral measure.

Table 2.

Final Ten-Item, Three-Factor Model for the Chinese-Specific Scale

Factor	Question
1	It is very stressful when hospital staff don't understand what ______ is saying or asking.	0.79
	I feel intimidated when I talk with non-Chinese– speaking health care providers.
	It is very stressful when _______ and family are afraid to talk about death/dying.
	It can be lonely to be a Chinese caregiver in a Western medical system.
2	Distress when you have to make decisions about ________'s care and treatment	0.69
	Distress about who is responsible for making medical decisions on behalf of _______
	Distress if there is disagreement and I have to decide between what the doctor says should be done and what ______says.
	Having to translate and explain medical terms to _______ makes me anxious
3	Felt ________ doesn't appreciate all the things you do as a caregiver	0.70
	Found myself getting angry with ______
Overall coefficient α for ten Chinese-specific items of the BASC-C		0.79

Factor 1, cultural strain; Factor 2, decision-making strain; Factor 3, emotional strain.

BASC-C, Brief Assessment Scale for Caregivers-Chinese (BASC-C).

Construct and criterion validity

Table 3 shows the first-order correlations and associated p values between total and factor scores of the BASC-C and scores on the other questionnaires. In the case of statistically significant first order correlations, p values are included for the partial correlations after controlling for the original 14-item BASC. The first-order correlations show that the BASC-C was significantly correlated with HADS anxiety (r=0.64, p<0.001), HADS depression (r=0.60, p<0.001), CBI total (r=0.60, p<0.001), caregiving hours per week (r=0.25, p=0.01), ADL bathing (r=0.31, p=0.002), ADL dressing (r=0.34, p<0.001), ADL toileting (r=0.32, p=0.001), ADL transfer (r=0.38, p<0.001), and ADL feeding (r=0.21, p=0.04). In line with construct validity, there were strong correlations with other measures of caregiver burden (HADS, CBI). Correlations with measures of constructs related, but not equivalent, to caregiver burden (caregiving hours, ADL's) showed lower, but still significant correlations.

Table 3.

Correlations Obtained in the Second Survey

Variable	BASC-C (24-items)	BASC (14-items)	NPI (5-items)	All New Chinese items (10-items)	Chinese scale Factor 1(4- items)	Chinese scale Factor 2 (4- items)	Chinese scale Factor 3 (2- items)
HADS Anxiety	0.64 (<0.001; 0.002)^a	0.56 (<0.001)^b	0.60 (<0.001)	0.59 (<0.001;<0.001)	0.53 (<0.001;<0.001)	0.50 (<0.001; 0.01)	0.18 (0.08)
HADS Depression	0.60 (<0.001; 0.005)	0.52 (<0.001)	0.60 (<0.001)	0.55 (<0.001;<0.001)	0.50 (<0.001;<0.001)	0.48 (<0.001; 0.01)	0.12 (0.23)
CBI Tot	0.60 (<0.001; 0.18)	0.61 (<0.001)	0.67 (<0.001)	0.46 (<0.001; 0.18)	0.34 (<0.001; 0.28)	0.42 (<0.001; 0.34)	0.27 (0.006; 0.34)
Patient's age	0.12 (0.21)	0.11 (0.27)	0.05 (0.65)	0.11 (0.26)	0.13 (0.19)	0.06 (0.57)	0.04 (0.66)
Caregiver's age	0.09 (0.35)	0.03 (0.78)	0.03 (0.78)	0.14 (0.14)	0.22 (0.03; 0.02)	−0.02 (0.82)	0.14 (0.16)
Single marital status	0.05 (0.61)	−0.07 (0.47)	−0.03) (0.80)	−0.18 (0.06)	−0.30 (0.002;<0.001)	−0.05 (0.59)	0.07 (0.45)
Marin Acculturation	−0.03 (0.74)	0.08 (0.45)	0.05 (0.61)	−0.14 (0.15)	−0.34 (<0.001;<0.001)	0.01 (0.95)	0.25 (0.01; 0.02)
Education level	−0.07 (0.51)	0.03 (0.73)	0.05 (0.67)	−0.15 (0.13)	−0.27 (0.006; 0.002)	−0.05 (0.62)	0.13 (0.21)
Caregiving hours per week	0.25 (0.01; 0.27)	0.23 (0.02)	0.23 (0.02)	0.23 (0.02; 0.24)	0.21 (0.04; 0.19)	0.15 (0.14)	0.16 (0.11)
ADL Bathing	0.31 (0.002; 0.14)	0.28 (0.004)	0.35 (<0.001)	0.28 (0.005; 0.18)	0.21 (0.03; 0.29)	0.27 (0.006; 0.19)	0.09 (0.36)
ADL Dressing	0.34 (<0.001; 0.004)	0.36 (<0.001)	0.44 (<0.001)	0.26 (0.008; 0.59)	0.21 (0.03; 0.46)	0.23 (0.02; 0.81)	0.11 (0.27)
ADL Toileting	0.32 (0.001; 0.12)	0.29 (0.003)	0.36 (<0.001)	0.28 (0.004; 0.18)	0.29 (0.003; 0.05)	0.23 (0.02; 0.47)	0.004 (0.97)
ADL Transfer	0.38 (<0.001; 0.01)	0.37 (<0.001)	0.43 (<0.001)	0.31 (0.002; 0.29)	0.24 (0.01; 0.31)	0.28 (0.004; 0.40)	0.13 (0.20)
ADL Continence	0.18 (0.06)	0.26 (0.009)	0.27 (0.007)	0.07 (0.48)	0.03 (0.78)	0.09 (0.37)	0.05 (0.63)
ADL Feeding	0.21 (0.04; 0.06)	0.23 (0.01)	0.26 (0.008)	0.14 (0.16)	0.10 (0.33)	0.14 (0.16)	0.06 (0.53)

First-order correlation (p value for first order correlation; p value for partial correlation controlling for 14 item BASC score).

First-order correlation (p value for first-order correlation).

Factor 1, cultural strain; Factor 2, decision-making strain; Factor 3, emotional strain.

BASC-C, Brief Assessment Scale for Caregivers in Chinese (BASC-C); NPI, Negative Personal Impact; HADS, Hospital Anxiety and Depression Scale; CBI Tot, Caregiver Burden Inventory total; ADL, assistance with daily living.

Like the BASC-C total score, the score from the 10 new Chinese-specific items significantly correlated with other items. These included HADS anxiety (r=0.59, p<0.001), HADS depression (r=0.55, p<0.001); burden (CBI r=0.46, p<0.001); caregiving hours (r=0.23, p=0.02); time spent helping with ADL's (bathing, r=0.28, p=0.005; dressing, r=0.26, p=0.008; toileting, r=0.28, p=0.004; transferring, r=0.31, p=0.002).

The three factors derived from the Chinese-specific items also yielded significant correlations with other outcomes. In addition to the HADS anxiety, HADS depression, and CBI, Factor 1, cultural strain, was significantly related to caregiver age (r=0.22, p=0.03), marital status (r=−0.30, p=0.002), acculturation (r=−0.34, p<0.001), education level (r=−0.27, p=0.006), caregiving hours (r=0.21, p=0.04), bathing, dressing, toileting, and transfer. Factor 2, decision-making strain, was significantly correlated with HADS Anxiety and Depression, CBI Tot, bathing, dressing, toileting, and transfer. Factor 3, emotional strain, was significantly correlated with CBI Tot, and acculturation.

To determine whether the Chinese-specific items yielded information above that provided by the original BASC, separate regression analyses controlling for the original BASC items were done for items that showed significant zero correlations with either the full BASC-C, the 10 new Chinese-specific items, or Factors 1, 2, and 3 (Table 3). HADS anxiety and depression remained significantly correlated with the BASC-C, the 10 new Chinese-specific items, and Factors 1 and 2. In contrast, the CBI was not significantly correlated to the BASC-C (or Chinese specific factors) after controlling for the original BASC items. The BASC-C was significantly correlated with ADL dressing, transfer and continence after controlling for the original BASC. In addition, caregiver's age, marital status, acculturation, and education level remained significantly correlated with Factor 1 after controlling for the original items. Factor 3 was significantly correlated with acculturation after controlling for the original BASC items. Examination of the partial correlations therefore supports the conclusion that the Chinese-specific items provide additional information about caregiver burden.

Discussion

Caregiving for a family member with serious illness is challenging.¹⁶ The original BASC was developed to measure caregiver distress, burden, and associated positive emotions. Early validation in a largely Caucasian population established the tool's validity.¹¹ This study was undertaken to adapt the BASC to the Chinese population.

Numerous tools exist for caregiver assessment in Chinese. A recent systematic review identified 74 burden tools with adequate psychometric properties.¹⁷ The two most widely used tools are the Zarit Caregiver Burden Interview¹⁸ and the Caregiver Burden Interview,¹⁴ both of which exist in Chinese. However, closer inspection of existing tools suggests limited generalizability for U.S. immigrant populations. Many tools were validated on nonimmigrant Chinese caregivers in Hong Kong and Taiwan, and intended for caregivers of patients with specific conditions, notably dementia (Zarit and CBI) and cancer. No available tool successfully captures the dimension of cultural strain so prevalent for Chinese caregivers within the U.S. health system. For all these reasons, we decided to validate the BASC for use among this culturally vulnerable caregiver subgroup. Findings presented here support the validity of the new Chinese-language BASC-C comprised of 24 items: original BASC (14 items) and a Chinese-specific scale (10 items) (Appendices A and B).

The 24-item BASC-C takes approximately 15 minutes to complete. The BASC-C may be useful in identifying and characterizing Chinese caregivers burdened by a range of difficulties including linguistic and cultural concerns. The tool can be self- or interviewer-administered in the inpatient, ambulatory, primary, and community settings, and used as a vehicle for communication about the sensitive topic of caregiver burden. Further studies are needed to confirm the value of the total score and proposed factor scales, determine whether the items measuring positive emotions are meaningful in this population, and assess test–retest reliability and responsiveness to change.

Limitations in the present study must be acknowledged. The population in New York City's Chinatown may not be representative of the Chinese American population overall. The focus group process was limited to two meetings and may not have ensured comprehensive saturation of caregiver burden themes. It is possible that other salient concerns were missed. As a rule, Chinese immigrants tend to be insular and reluctant to participate in research, making random sampling challenging. Testing the reliability and validity of the new instrument is a process and the data collected in the present surveys may provide only a limited perspective. Further work is needed to assess the adequacy of the BASC-C as a culturally-appropriate caregiver measure.

The BASC-C may prove useful in evaluating the complex experience of family caregiving in a vulnerable subculture. Future studies could be designed to determine the extent to which caregiver distress contributes to compromised informed consent, psychological or physical disorders, or problems accessing community services. With the development of valid measures for large U.S. subpopulations, the possibility of meaningful comparative studies exists. These studies can lay the groundwork for the design of culturally appropriate interventions to reduce caregiver burden. These cross-cultural distinctions have clinical, research and educational utility.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Appendix A.

Brief Assessment Scale for Caregivers of the Medically Ill–Chinese (BASC-C)

Please check off to what extent you have had any of the following feelings in the past month because of taking care of ________________________
Because of _______________'s illness, how much have you:	Not at all	A little	Some	A lot
1. Worried about _________ even when you are not with him/her.
2. Been depressed because of _____'s illness.
3. Been upset about not having enough time for yourself because of caring for _______.
4. Been overwhelmed by the responsibility of caring for _________.
5. Felt ________ doesn't appreciate all the things you do as a caregiver.
6. Found myself getting angry with ______.

Please rate your distress during this past month:	Not at all distressed	A little distress	Some distress	A lot of distress	Does not apply
7. Distress over seeing _____ in so much pain or discomfort.
8. Distress at not having enough time to do your job, other responsibilities and chores.
9. Distress about making decisions about hospitalizing ______.
10. Distress over talking about what medical procedures _____ would want or would not want if his/her heart or breathing were to stop.
11. Distress of seeing how much _____'s illness has changed your relationship.
12. Distress at having strained relationships with other family members over taking care of _____.
13. Distress about who is responsible for making medical decisions on behalf of _______.
14. Distress when you have to make decisions about ________'s care and treatment.
15. Distress if there is disagreement and I have to decide between what the doctor says should be done and what ______says.

Please rate whether you agree or disagree with each of these statements as it applies to you in your care of __________ this month:	Agree a lot	Agree a little	Disagree a little	Disagree a lot	Does not apply
16. Taking care of _____ has drawn the two of us closer together.
17. Taking care of _____ has brought meaning to my life.
18. Taking care of _____ has drawn other members of our family closer together.
19. Taking care of _____ makes me feel good about myself.

Please think about your experiences as a Chinese caregiver:
20. It is very stressful when hospital staff don't understand what ______ is saying or asking.
21. Having to translate and explain medical terms to _______ makes me anxious.
22. It is very stressful when _______ and family are afraid to talk about death/dying.
23. I feel intimidated when I talk with non-Chinese speaking health care providers.
24. It can be lonely to be a Chinese caregiver in a Western medical system.

Appendix B.

References

US Census Bureau. www.census.gov/newsroom/releases/archives/facts_for_features_special_editions/cb12-ff09.html May 2012 2013 October 1.

The Asian American Foundation of New York. Asian American Elders in New York City: A Study of Health, Social Needs, Quality of Life and Quality of Care. New York: The Asian American Foundation of New York, February 2003.

The Asian American Foundation of New York. Working but Poor: Asian American Poverty in New York City. Data derived from analysis performed by the Asian American Federation Census Information Center. The Asian American Foundation of New York, October 2008.

Given

, Given

, Kozachik

. Family support in advanced cancer. CA Cancer J Clin, 2001; 51:213–231.

Glajchen

. Physical well-being of oncology caregivers: An important quality of life domain. Semin Oncol Nurs, 2012; 28:226–235.

Pinquart

, Sorensen

. Ethnic differences in stressors, resources, and psychological outcomes of family caregiving; A meta-analysis. Gerontologist, 2005; 45:90–106.

Schlarlach

, Kellam

, Ong

, Baskin

, Goldstein

, Fox

. Cultural attitudes and caregiver service use: Lessons and focus groups with racially and ethnically diverse family caregivers. J Gerontol Soc Work, 2006; 47:133–156.

. Barriers to and unmet needs for supportive services: Experiences of Asian-American caregivers. J Cross Cult Gerontol, 2004; 19:241–260.

Gallagher-Thompson

, Coon

. Evidence-based psychological treatments for distress in family caregivers of older adults. Psychol Aging, 2007; 22:37–51.

10.

Sheridan

. Treating the world without leaving your ED: Opportunities to deliver culturally competent care. Acad Emerg Med, 2006; 13:896–903.

11.

Glajchen

, Kornblith

, Homel

, Fraidin

, Mauskop

, Portenoy

. Development of a brief assessment scale for caregivers of the medically ill. J Pain Symptom Manage, 2005; 29:245–254.

12.

Guillemin

, Bombardier

, Beaton

. Cross-cultural adaptation of health-related quality of life measures: Literature review and proposed guidelines. J Clin Epidemiol, 1993; 46:1417–1432.

13.

Leung

, Wing

, Kwong

, Lo

, Shum

. Validation of the Chinese-Cantonese version of the Hospital Anxiety and Depression Scale and comparison with the Hamilton Rating Scale of Depression. Acta Psychiatrica Scand, 1999; 100:456–461.

14.

Chou

K-R

, Lin

J-C

, Chu

. The reliability and validity of the Chinese version of the Caregiver Burden Inventory. Nurs Res, 2002; 51:324–331.

15.

Dong

, Simon

. Is impairment in physical function associated with increased risk of elder mistreatment? Findings from a community-dwelling Chinese population. Public Health Rep, 2010; 125:743–753.

16.

Lee

, Farran

. Depression among Korean, Korean American, and Caucasian American family caregivers. J Transcult Nurs, 2004; 15:18–25.

17.

Whalen

, Buchholz

. The reliability, validity, feasibility of tools used to screen for caregiver burden: A systematic review. JBI Library of Systematic Reviews 2009. www.joannabriggs.edu.au/pubs/systematic_reviews.php. 2013 October 2.

18.

, Wang

, Yang

, Feng

. Zarit Caregiver burden interview: Development, reliability and validity of the Chinese version. Psychiatry Clin Neurosci, 2009; 63:730–734.