The Caregiver–Provider Relationship Assessment

Abstract

This article summarizes the development of the caregiver–provider relationship assessment (CPRA) designed to measure family caregivers’ perceptions of relationship quality with health care providers. Using an online sample of family caregivers (n = 156), the patient reactions assessment (PRA) was adapted for use with family caregivers and subjected to principal component and reliability analyses. Analyses indicate that the CPRA factor structure is analogous to the original PRA scale, and Cronbach’s α for the three CPRA subscales range from .85 to .91. The tool can be used by clinicians and researchers in efforts to help family caregivers become more effective communicators in health care contexts.

Keywords

family caregivers health care communication instrument development

Increasingly, the role of family caregivers in helping patients manage chronic illness is a focus of research. Patients are often accompanied to medical visits by a family member or friend (Prohaska & Glasser, 1996; Wolff & Roter, 2008), and the communication in the health care triad (medical provider, family caregiver, and patient) is a significant dynamic in ongoing patient care and treatment adherence (Fortinsky, 2001; Glasser, Prohaska, & Gravdal, 2001; Kahana & Kahana, 2003; Miller & Brown, 2005; Prohaska & Glasser, 1996; Silliman, Bhatti, Khan, Dukes, & Sullivan, 1996; Weinberg, Lusenhop, Gittell, & Kautz, 2007; Wolff & Roter, 2008). The quality of communication between family caregivers and providers appears to be an important factor in promoting patient health but there is a lack of standardized evaluation tools designed to assess this variable. To fill this void, a new instrument was developed, the caregiver–provider relationship assessment (CPRA), an adaptation of a validated scale for patients, the patient reactions assessment (PRA; Galassi, Schanberg, & Ware, 1992).

The PRA measures “patient-perceived quality of patient-provider medical relationships” (Galassi et al., 1992, p. 350) is visit-specific and comprised of three subscales: patient information index (perceived quality of information provided by the professional), patient affective index (perceptions of the provider’s affective behaviors), and patient communication index (self-perceived ability to initiate communication with the provider). The PRA items were subjected to exploratory and confirmatory factor analyses, and “the PRA appears to be a brief, factorially valid and promising measure” (Galassi et al., 1992, p. 350). To develop the CPRA, the 15 PRA items were reworded to focus on family caregiver reactions to provider interaction quality. This study’s aim is to examine the factor structure of the CPRA and test its reliability.

Method

The CPRA face and content validity was explored prior to statistical analyses. A total of 10 family caregivers were interviewed using the CPRA, and two experts (a family practice physician and the executive director of a national caregiving advocacy organization) reviewed the instrument. All agreed that the measure adequately captured the central domains of caregiver–provider communication. The present sample was recruited through the National Family Caregivers Association (NFCA, n.d.), a nonprofit advocacy organization for family caregivers. A total of 898 self-identified family caregivers participated in a free online webinar for family caregivers offered by NFCA that addressed strategies to improve communication with health care providers, and 156 (17%) volunteered to complete the CPRA online. The factor structure of the CPRA was examined through principal component analysis. Cronbach’s αs were calculated for the three subscales and total scale score, and bivariate correlations ran to examine the relationships among the three subscales.

Results

The mean age of the sample was 54 years (SD = 11). In all 12% (n = 19) were male and 88% (n = 134) were female (three left the item blank). Participants had been caring for their family member for an average of 8 years (SD = 8 years). In all 34% (n = 52) were caring for their spouse/significant other, 24% (n = 36) for a parent, 28% (n = 43) for an adult child, 2% (n = 3) for a child under 18, and 12% (n = 18) for another family member or friend.

The values of the Kaiser-Meyer-Olkin measure of sampling adequacy (.91), Bartlett’s test of sphericity, χ² (105) = 1570.37, p < .001, and examination of the correlation matrix (singularity was not an issue) indicated that factor analysis was appropriate for these data. Principal component analysis with an oblique rotation produced the best-fitting and most theoretically sound solution (SPSS 19). The scree plot and eigenvalues revealed three latent variables. The eigenvalues and variance uniquely explained by each factor were: Factor 1, 5.76 (52%); Factor 2, 5.62 (13%); and Factor 3, 5.10 (7%); for a total of 72%. The factor loading pattern was analogous to the pattern for the PRA. The PRA factor names were adapted for the CPRA: Factor 1, caregiver affect (CA); Factor 2, caregiver communication (CC); and Factor 3, caregiver information (CI). See Table 1 for the factor loading matrix and communalities. Cronbach’s α for the total scale score was .93 and for the CA, CC, and CI were .91, .85, .91, respectively. See Table 2 for descriptive statistics. A correlation matrix examining the relationships among the three subscales revealed that the subscales were significantly related (p < .01) to one another (CA and CC, r = .54; CA and CI, r = .64; and CC and CI, r = .63).

Table 1.

CPRA Factor Loading Matrix and Communalities

Item Number	Item	Factor Loading			Communalities
Item Number	Text	CA	CC	CI	Communalities
3	This medical professional is warm and caring toward me	.78			.65
6	This medical professional makes me feel comfortable about discussing personal or sensitive issues	.79			.82
9	This medical professional really respects me	.78			.70
12	When I talk to this medical professional, I sometimes end up feeling insulted	.82			.71
15	This medical professional does not seem interested in me as a person	.87			.79
2	If this medical professional tells me something that is different from what I was told before, it is difficult for me to ask about it in order to get it straightened out		.93		.81
4	If I do not understand something the medical professional says, I have difficulty asking for more information		.92		.81
7	It is hard for me to tell the medical professional about my family member’s new symptoms		.67		.61
8	It is hard for me to ask about how my family member’s treatment is going		.79		.78
13	I have difficulty asking this medical professional questions		.75		.70
1	I understand the possible side effects of my family member’s treatment			.75	.67
5	The medical professional told me what he/she hopes the treatment will do for my family member			.76	.59
10	I understand pretty well the medical plan for helping my family member			.60	.68
11	After talking to this medical professional, I have a good idea of what changes to expect in my family member’s health over the next weeks and months	.41		.60	.68
14	My family member’s treatment procedure was clearly explained to me			.68	.72

Note. CA = caregiver affect; CC = caregiver communication; CI = caregiver information; CPRA = caregiver–provider relationship assessment. Items are scored on a 7-point Likert scale that ranges from 1 (very strongly disagree) to 7 (very strongly agree). Directions for the instrument read: “Think about your contact with the medical professional who primarily examined your family member most recently. Then answer the following questions by checking the answer that best describes how you feel about your recent contact with that medical professional.” The CPRA is scored by first reverse scoring items 2, 4, 7, 8, 12, 13, and 15 and then summing responses for subscale and total scale scores. The CA subscale consists of the sum of items 3, 6, 9, 12, and 15; the CC subscale is comprised of the sum of items 2, 4, 7, 8, and 13; and the CI subscale consists of the sum of items 1, 5, 10, 11, and 14.

Table 2.

CPRA Descriptive Statistics

Item	Mean	Std Dev
1	5.41	1.10
2	5.21	1.52
3	5.17	1.31
4	5.32	1.40
5	5.22	1.19
6	5.15	1.23
7	5.56	1.17
8	5.42	1.28
9	4.98	1.36
10	5.32	1.23
11	4.79	1.24
12	5.21	1.53
13	5.52	1.23
14	5.01	1.28
15	4.99	1.48
CA^a	25.53	5.91
CC^a	27.03	5.67
CI^a	25.65	4.78

Note. CA = caregiver affect; CC = caregiver communication; CI = caregiver information; CPRA = caregiver–provider relationship assessment.

^aIndicates subscale.

Discussion

The CPRA helps to fill a void in the family caregiving literature by providing a measure of communication quality in the family caregiver–health care provider relationship. The CA subscale measures caregivers’ perceptions of the provider as valuing and respecting them. The CI assesses perceptions of how effective the professional was in providing medical information and the extent to which the caregiver comprehended that information. The CC measures the ease or difficulty caregivers experience in communicating information to the professional. The CI and CA subscales focus on the caregiver’s perceptions of the provider side of the relationship, and CC addresses the family caregiver’s contribution. All three tap important dimensions of the health care communication process (Hall, Roter, & Katz, 1988).

Although more testing is needed, this preliminary analysis suggests that the CPRA has promise to be a useful tool for those interested in the caregiver–provider communication interface. This analysis revealed the same factor structure as the original tool developed for patients (PRA) suggesting that both scales are likely assessing the same latent variables. Parallel to Galassi, Schanberg, and Ware’s (1992) analysis of the PRA, the CPRA subscales are interrelated but the size of the correlations and underlying factor structure indicate that the subscales are measuring related, yet distinct constructs. Although far from rigorous evidence, interviews with family caregivers and experts suggest that the content of the items tap integral components of the caregiver–health care provider relationship. On the other hand, the current online sample may not be representative, and confirmatory factor analysis should be conducted with larger samples to validate the CPRA factor structure and establish criterion-related and construct validity. Family caregivers who have Internet access and participate in online education may have different health-related communication behavior patterns than others. The ethnic breakdown of the sample, type and severity of conditions that necessitated caregiving, and time spent on caregiving tasks are not known. These variables may impact the instrument’s factor structure and should be considered in future validation studies. Due to the selectivity of the sample and its inherent bias, researchers are strongly encouraged to validate the instrument using their own samples of family caregivers prior to drawing any conclusions based on CPRA scores.

Analogous instruments that measure relationship quality with health care providers are unique and potentially useful. Upon further validation of the CPRA, clinicians and researchers could use it and the PRA together to assess the dynamics of communication in the health care triad (for PRA items, see Corcoran & Fischer, 2000). In conclusion, more testing is needed before the CPRA can be confidently utilized with other samples, but it is a good beginning to a promising line of research.

Footnotes

Acknowledgment

The author would like to thank the National Family Caregivers Association for their continued collaboration.

The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The author received financial support from the National Family Caregivers Association to collect the research data. No financial support was received for authorship or publication of this article.

References

Corcoran

Fischer

(2000). Measures for clinical practice: Volume 2. New York, NY: The Free Press.

Fortinsky

(2001). Health care triads and dementia care: Integrative framework and future directions. Aging and Mental Health, 5, S35–S48.

Galassi

Schanberg

Ware

(1992). Patient reactions assessment: A brief measure of the quality of the patient-provider medical relationship. Psychological Assessment, 4, 346–351.

Glasser

Prohaska

T. R.

Gravdal

(2001). Elderly patients and their accompanying caregivers on medical visits. Research on Aging, 23, 326–348.

Hall

J. A.

Roter

D. L.

Katz

N. R.

(1988). Meta-analysis of correlates of provider behavior in medical encounters. Medical Care, 26, 657–672.

Kahana

(2003). Patient proactivity enhancing doctor-patient-family communication in cancer prevention and care among the aged. Patient Education and Counseling, 50, 67–73.

Miller

Brown

J. L.

(2005). Marital interactions in the process of dietary change for Type 2 diabetes. Journal of Nutrition Education and Behavior, 37, 226–234.

National Family Caregivers Association. (n.d.). About NFCA. Retrieved September 1, 2010, from http://www.thefamilycaregiver.org/about_nfca/

Prohaska

T. R.

Glasser

(1996). Patients’ views of family involvement in medical care decisions and encounters. Research on Aging, 18, 52–69.

10.

Silliman

R. A.

Bhatti

Khan

Dukes

K. A.

Sullivan

L. M.

(1996). The care of older persons with diabetes mellitus: Families and primary care physicians. Journal of the American Geriatrics Society, 44, 1314–1321.

11.

Weinberg

D. B.

Lusenhop

R. W.

Gittell

J. H.

Kautz

C. M.

(2007). Coordination between formal providers and informal caregivers. Health Care Management Review, 32, 1–10.

12.

Wolff

J. L.

Roter

D. L.

(2008). Hidden in plain sight: Medical visit companions as a resource for vulnerable older adults. Archives of Internal Medicine, 168, 1409–1415.