Caregiver Reaction Assessment

1. Abstract

The Caregiver Reaction Assessment (CRA) is a multidimensional psychometric instrument specifically designed to evaluate both the negative and positive consequences of providing informal care to individuals suffering from chronic physical impairments, terminal illnesses, or cognitive disorders. Developed by Charles W. Given and colleagues in 1992, the CRA fundamentally challenged the prevailing unidimensional conceptualizations of caregiver burden by operationalizing caregiving as an intricate, bi-directional experience comprising both subjective strain and psychological enrichment. The instrument consists of 24 self-report items distributed across five distinct, conceptually independent subscales: Caregiver Esteem (7 items), Impact on Finances (3 items), Impact on Schedule (5 items), Impact on Health (4 items), and Lack of Family Support (5 items). Respondents rate each statement using an authentic 5-point Likert scale ranging from 1 (“Strongly Disagree”) to 5 (“Strongly Agree”). Unlike aggregate-score instruments such as the Zarit Burden Interview (ZBI), the CRA generates separate domain-specific scores rather than a composite global score, thereby preventing the dilution of critical stressors by positive psychological adaptations. Extensive psychometric evaluations across international populations—including validation studies in cancer, dementia, stroke, and palliative cohorts—demonstrate robust internal consistency (Cronbach’s alpha coefficients typically ranging from .65 to .90 across subscales), confirmed structural validity via exploratory and confirmatory factor analyses, and excellent concurrent and predictive validity against measures of depression, functional dependence, and family functioning. As both an epidemiological assessment tool and an evaluative outcome measure, the CRA provides clinicians and researchers with granular insights necessary to tailor psychosocial interventions to the nuanced, multi-faceted realities of family caregiving.

2. Keywords

Caregiver Reaction Assessment, caregiver burden, informal caregiving, caregiver esteem, family support, psychometrics, chronic illness, oncology caregiving, palliative care, health-related quality of life

3. Authors

The Caregiver Reaction Assessment was originally developed and psychometrically validated by an interdisciplinary team of researchers led by Charles W. Given, Ph.D., Professor Emeritus in the Department of Family Medicine and the Institute for Health Policy at Michigan State University (East Lansing, Michigan, USA). Dr. Given’s pioneering investigative work focuses on health services research, symptom management, cancer care delivery, and the longitudinal trajectory of informal caregiving systems.

Co-developers of the original 1992 instrument include Barbara Given, Ph.D., RN, FAAN, University Distinguished Professor and Associate Dean Emerita in the College of Nursing at Michigan State University, whose sustained scholarship has profoundly shaped contemporary family caregiving research; Marlon Stommel, Ph.D., sociologist and psychometrician specializing in longitudinal data analysis, health status measurement, and applied biostatistics; Victoria Collins, M.S.; Sharon King, Ph.D.; and Suzan Franklin, M.S.W.

Significant cross-cultural adaptations and European validations were subsequently spearheaded by Dutch health services researcher Ida J. M. (Curie) Nijboer, Ph.D., along with Marlies Tempelaar, Ph.D., Robbert Sanderman, Ph.D., and colleagues at the Northern Centre for Healthcare Research and the Department of Health Sciences, University of Groningen, Netherlands. Inquiries regarding original scale instrumentation are traditionally directed to the College of Nursing and Institute for Health Care Studies at Michigan State University.

4. Purpose

The primary objective of the Caregiver Reaction Assessment is to systematically capture, disentangle, and quantify the specific dimensions through which informal caregiving reverberates across an individual’s life. Throughout the late 20th century, behavioral medicine and gerontological psychometrics frequently treated caregiving as a monolithic, universally deleterious stressor. Instruments constructed during that era predominantly aggregated physical fatigue, emotional distress, financial depletion, and relational resentment into a solitary index of “burden.” Such total-score methodologies systematically obscured the reality that family members routinely experience profound subjective fulfillment, meaning, and elevated self-worth concurrent with objective physical exhaustion or financial strain.

The CRA was intentionally constructed to circumvent these theoretical and methodological limitations. Rather than forcing positive and negative perceptions onto opposite ends of a single continuum, the CRA recognizes that self-esteem and burden operate as orthogonal or distinct constructs. The clinical rationale for this architecture is paramount: an intervention that mitigates schedule disruption or financial toxicity will not necessarily augment feelings of caregiver esteem, nor does high caregiver esteem insulate an informal provider from physical health degradation or lack of social support.

In clinical practice, the CRA functions as a diagnostic and triaging instrument. Multidisciplinary healthcare teams within oncology, neurology, physical rehabilitation, and hospice settings utilize the profile to identify specific vulnerabilities requiring targeted support. For instance, an elevated score on Impact on Schedule paired with low Lack of Family Support suggests the need for respite care or day-center referral rather than family therapy. Conversely, elevated scores on Lack of Family Support indicate intra-familial conflict or isolation, directing clinical social workers toward structured family meetings and mediation.

In research domains, the CRA serves as an evaluative metric in randomized controlled trials (RCTs) testing psychoeducational workshops, mobile health coaching apps, and palliative care consultation models. By providing independent domain metrics, the scale permits researchers to detect subtle, domain-specific treatment effects that would otherwise remain masked within a composite score.

5. Psychological Construct

The Caregiver Reaction Assessment conceptualizes the reaction to caregiving as a five-dimensional, multi-determined psychological construct. Each dimension represents a discrete domain of subjective appraisal, reflecting how care demands interact with the caregiver’s personal resources, interpersonal systems, and daily life routines.

Caregiver Esteem (7 Items)

The Caregiver Esteem subscale measures the subjective positive appraisal and self-worth derived from fulfilling the caregiving role. Rooted in theories of self-efficacy and existential meaning-making, this dimension quantifies the extent to which caring for the care recipient enhances the caregiver’s personal fulfillment, validates reciprocal obligations, and fosters feelings of value, dignity, and competence. It assesses whether caregiving makes the individual feel good, important, and privileged, rather than merely burdened. Sample expressions reflect pride in one’s capacity to give back to a spouse or family member and intrinsic enjoyment in care tasks.

Lack of Family Support (5 Items)

The Lack of Family Support subscale appraises the caregiver’s perception of familial cohesion, relational abandonment, and equity in the distribution of care responsibilities. Rather than measuring objective network size, this dimension taps into subjective alienation and resentment stemming from the perceived dumping of responsibilities onto the primary caregiver. It captures relational friction, feelings of isolation from other relatives, and the perceived inability to count on family members during crisis periods.

Impact on Finances (3 Items)

The Impact on Finances subscale gauges the economic strain and financial toxicity directly attributable to managing the patient’s illness and care needs. It reflects the caregiver’s perception of difficulty meeting daily living expenses, the prohibitive costs of health services and medical supplies, and the overarching drain on family financial reserves. Because informal caregiving frequently necessitates reduced employment hours or early retirement, this domain quantifies the economic vulnerability imposed by the disease course.

Impact on Schedule (5 Items)

The Impact on Schedule subscale captures the objective and subjective disruption of the caregiver’s daily routine, personal time, occupational commitments, and social interactions. Items in this domain evaluate the extent to which the caregiver’s entire life centers around the patient’s immediate needs, the cessation or curtailment of recreational activities and social visitations, and the chronic intrusion of unpredictable care demands that interrupt work and leisure.

Impact on Health (4 Items)

The Impact on Health subscale assesses the caregiver’s subjective physical deterioration, somatic depletion, and somatic vitality since assuming the caregiving role. It operationalizes somatic burden through self-reported chronic fatigue, subjective declines in baseline physical health, and the physical stamina or strength required to perform demanding personal care tasks (such as transfers, hygiene, and positioning).

6. Theoretical Framework

The CRA is grounded theoretically in the Cognitive Appraisal Theory of Coping and Adaptation advanced by Richard S. Lazarus and Susan Folkman (1984), alongside the Sociological Stress Process Model developed by Leonard Pearlin and colleagues (1990).

The Transactional Model of Stress and Coping

Lazarus and Folkman posited that psychological stress is not an inherent property of an objective event, but rather emerges from a dynamic transaction between the individual and the environment. This transaction is governed by two sequential appraisal processes: primary appraisal, in which an individual determines whether an event represents a threat, harm/loss, or challenge; and secondary appraisal, in which the individual evaluates their available coping resources to mitigate that threat. Given et al. (1992) mapped the CRA directly onto this framework. The primary objective demands of care—such as managing physical symptoms, medication administration, and monitoring functional decline—are filtered through the caregiver’s cognitive appraisals. The negative subscales (Impact on Finances, Schedule, Health, and Lack of Family Support) represent appraised threats or resource deficits, whereas Caregiver Esteem embodies the appraisal of caregiving as a challenge that fosters personal growth, resilience, and reciprocal reward.

Pearlin’s Caregiver Stress Process Model

Pearlin’s framework conceptualizes caregiving through an ecological paradigm encompassing background context, primary stressors, secondary role strains, and psychological outcomes. Primary stressors arise directly from the patient’s illness severity and cognitive/physical dependence. These primary stressors proliferate into secondary stressors across non-care domains, including workplace conflict, social isolation, and financial strain. The CRA specifically measures these secondary role strains. By assessing how primary care duties spill over into economic stress (Impact on Finances), relational strain (Lack of Family Support), and personal life restructuring (Impact on Schedule), the CRA operationalizes the proliferation pathways theorized by Pearlin. Crucially, the model accounts for intra-psychic resources—such as self-esteem and perceived mastery—as vital mediating buffers against clinical burnout and clinical depression.

7. Validity

The psychometric validity of the Caregiver Reaction Assessment has been extensively verified across diverse clinical populations, health delivery contexts, and international cultural settings.

Construct and Structural Validity

Construct validity was initially established by Given et al. (1992) in a baseline cohort of 377 family caregivers assisting chronically ill older adults. Confirmatory factor analyses demonstrated that a five-factor oblique structure provided superior fit over competing unidimensional or three-factor models. In the cross-cultural validation of the Dutch version (CRA-D) conducted by Nijboer et al. (1999) among partners of newly diagnosed cancer patients, structural equation modeling affirmed the stability of the five-factor solution across longitudinal assessment intervals (at 3, 6, and 12 months post-diagnosis), exhibiting Root Mean Square Errors of Approximation (RMSEA) below .06 and Goodness-of-Fit Indices (GFI) exceeding .90.

Convergent and Discriminant Validity

Convergent validity is robustly demonstrated through consistent, statistically significant correlations between specific CRA subscales and validated external psychometric instruments:

• Depressive Symptoms: The negative subscales—particularly Impact on Health (r = .45 to .58) and Impact on Schedule (r = .38 to .52)—correlate strongly with caregiver depressive symptomatology as measured by the Center for Epidemiologic Studies Depression Scale (CES-D) and the Beck Depression Inventory (BDI).
• Physical Well-being: The Impact on Health subscale exhibits strong negative correlations with the Physical Component Summary (PCS) score of the Short Form-36 (SF-36) (r = -.52 to -.64).
• Social and Family Functioning: The Lack of Family Support subscale demonstrates strong inverse associations with the Family Assessment Device (FAD) and perceived social support scales (r = -.40 to -.55).
• Divergent Autonomy of Esteem: Caregiver Esteem displays weak or near-zero correlations with patient functional dependence (Activities of Daily Living / Instrumental Activities of Daily Living), confirming that subjective pride and meaning are psychologically independent of the patient’s objective physical impairment.

Predictive and Criterion Validity

Longitudinal investigations demonstrate that elevated baseline scores on Impact on Health and Impact on Schedule independently predict subsequent caregiver clinical burnout, elevated healthcare utilization, and earlier institutionalization or nursing home placement of care recipients, even after controlling for baseline patient cognitive and physical functional status.

8. Reliability

The reliability of the Caregiver Reaction Assessment has been demonstrated across dozens of independent empirical investigations worldwide. Internal consistency, measured primarily via Cronbach’s coefficient alpha ($\alpha$), consistently meets or exceeds accepted psychometric standards for behavioral and health sciences research.

Internal Consistency across Subscales

In the seminal psychometric development study by Given et al. (1992), internal consistency estimates were as follows:

• Caregiver Esteem: $\alpha = .80$ to $.84$ (demonstrating high homogeneity among the positive appraisal items)
• Impact on Schedule: $\alpha = .82$ to $.86$
• Impact on Finances: $\alpha = .81$ to $.83$
• Impact on Health: $\alpha = .78$ to $.80$
• Lack of Family Support: $\alpha = .79$ to $.82$

In the European Dutch validation study among oncology caregivers (Nijboer et al., 1999), coefficient alphas remained highly consistent over repeated assessment waves: Caregiver Esteem ($\alpha = .76-.81$), Impact on Schedule ($\alpha = .79-.82$), Impact on Finances ($\alpha = .77-.84$), Impact on Health ($\alpha = .71-.78$), and Lack of Family Support ($\alpha = .65-.72$). Similar reliability parameters have been documented in Asian adaptations (e.g., Chinese, Japanese, and Korean translations), with overall subscale alphas rarely dipping below .70, confirming cross-cultural stability.

Test-Retest Stability

In clinically stable cohorts evaluated across short-term intervals (2 to 4 weeks), the intraclass correlation coefficients (ICC) and Pearson test-retest correlation coefficients consistently range between .75 and .88 across the five subscales, confirming excellent stability in the absence of acute clinical deterioration in the care recipient.

9. Factor Analysis

The factorial validity of the CRA was established through successive stages of Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA).

Exploratory Factor Structure

During initial scale development, Given et al. (1992) subjected an initial pool of candidate items to principal axis factoring with oblique (Promax) rotation, accommodating the theoretical expectation that dimensions of caregiving strain would be intercorrelated. The empirical analysis extracted five distinct factors with eigenvalues exceeding 1.0, explaining more than 58% of the total variance. Factor loadings were clean and decisive: all items retained in the final 24-item instrument exhibited primary factor loadings $\ge .45$ on their designated construct, with cross-loadings on alternative factors consistently remaining below .25.

Confirmatory Factor Analysis and Goodness-of-Fit

Subsequent confirmatory factor analyses in independent validation samples (e.g., Stommel et al., 1999; Nijboer et al., 1999; Grov et al., 2006) examined multiple competing structural models:

1. A single general “caregiver burden” factor model;
2. A two-factor model separating positive (esteem) from negative (general strain) dimensions;
3. A five-factor orthogonal model; and
4. The proposed five-factor oblique (correlated) model.

The five-factor oblique model demonstrated substantially superior fit compared to all alternative configurations. Representative goodness-of-fit indices reported in literature include $\chi^2/\text{df}$ ratios between 1.8 and 2.4, Comparative Fit Index (CFI) values ranging from .92 to .96, Tucker-Lewis Index (TLI) values $\ge .91$, and RMSEA values between .045 and .062. These findings substantiate the hypothesis that caregiving reactions cannot be collapsed into a global score without substantial loss of structural and theoretical fidelity.

10. Instrument / Measurement Tool

• Instrument Name: Caregiver Reaction Assessment (CRA)
• Dutch Adaptation Title: Caregiver Reaction Assessment – Nederlandse versie (CRA-D)
• Target Respondent: Informal, family, or partner caregivers assisting individuals with chronic illnesses, functional limitations, cognitive impairments, or advanced cancer
• Target Demographics: Adults, older adults, and children/adolescents functioning in informal care roles
• Administration Format: Self-administered paper-and-pencil questionnaire, clinician-administered structured interview, or secure digital web-based survey
• Time Required: Approximately 8 to 12 minutes to complete all 24 items
• Total Item Count: 24 statements
• Authentic Response Scale: 5-point Likert scale:
  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Neither Agree nor Disagree
  • 4 = Agree
  • 5 = Strongly Agree
• Subscale Composition:
  • Caregiver Esteem (7 items): Items 1, 5, 9, 11, 15, 17, 21
  • Lack of Family Support (5 items): Items 2, 6, 12, 18, 22
  • Impact on Finances (3 items): Items 7, 14, 24
  • Impact on Schedule (5 items): Items 3, 8, 13, 19, 23
  • Impact on Health (4 items): Items 4, 10, 16, 20
• Scoring Architecture and Rules:
  • No Total Score: The CRA has no cumulative total score. Calculating a single composite index violates the validated psychometric structure. Each subscale is scored and interpreted independently.
  • Subscale Scoring Method: Subscale scores are calculated either as the arithmetic sum of the items or, more commonly in research, as the mean item score (sum divided by the number of items in that subscale), producing a continuous subscale metric ranging from 1.0 to 5.0.
  • Directionality & Reverse Scoring: In standard orientation, higher scores on the negative dimensions (Impact on Finances, Impact on Schedule, Impact on Health, Lack of Family Support) indicate greater perceived burden, strain, or lack of support, whereas a higher score on Caregiver Esteem denotes greater personal reward and positive appraisal. Positively worded support items (Items 6 and 18) are reverse-scored when computing Lack of Family Support. Negatively worded health items (such as Item 4 and Item 20) are reverse-scored when computing Impact on Health. In the Caregiver Esteem dimension, negatively phrased items (such as Item 9) are reverse-scored so that elevated scores consistently reflect positive esteem.

11. Permissions & Fee and Test Year

The Caregiver Reaction Assessment was officially published in 1992 by Charles W. Given and his research team at Michigan State University. The European-Dutch adaptation was published in 1999 by Ida J. M. Nijboer and colleagues at the University of Groningen.

The instrument was developed under federal research funding through the National Institutes of Health (NIH) / National Institute on Aging (NIA) and the National Cancer Institute (NCI). Consequently, the scale is widely regarded as being in the public domain for academic, non-commercial research, and clinical practice purposes, provided appropriate bibliographic citation is extended to the original authors. No licensing fee is typically required for standard non-profit investigative or healthcare applications. Commercial developers, pharmaceutical clinical trials, or electronic health record (EHR) vendors seeking proprietary integration are advised to consult the Office of Technology Licensing at Michigan State University and original copyright holders.

12. References

• Given, C. W., Given, B., Stommel, M., Collins, C., King, S., & Franklin, S. (1992). The Caregiver Reaction Assessment (CRA) for caregivers to persons with chronic physical and mental impairments. Research in Nursing & Health, 15(4), 271–283. https://doi.org/10.1002/nur.4770150406
• Grov, E. K., Dahl, A. A., Moum, T., & Fosså, S. D. (2006). Anxiety, depression, and quality of life in caregivers of patients with cancer in late palliative phase. Annals of Oncology, 17(7), 1175–1181. https://doi.org/10.1093/annonc/mdl099
• Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. Springer Publishing Company.
• Nijboer, C., Triemstra, M., Tempelaar, M., Sanderman, R., & van den Bos, G. A. (1999). Measuring both negative and positive reactions to giving care to cancer patients: Psychometric qualities of the Caregiver Reaction Assessment (CRA). Social Science & Medicine, 48(9), 1259–1269. https://doi.org/10.1016/s0277-9536(98)00426-3
• Pearlin, L. I., Mullan, J. T., Semple, S. J., & Skaff, M. M. (1990). Caregiving and the stress process: An overview of concepts and their measures. The Gerontologist, 30(5), 583–594. https://doi.org/10.1093/geront/30.5.583
• Stommel, M., Given, C. W., & Given, B. A. (1990). Depression as an overriding variable in measuring caregiver burdens. Journal of Aging and Health, 2(1), 81–102. https://doi.org/10.1177/089826439000200105
• Stommel, M., Wang, S., Given, C. W., & Given, B. (1999). Confirmatory factor analysis of the Caregiver Reaction Assessment (CRA) in cancer caregiving. Nursing Research, 48(5), 268–276. https://doi.org/10.1097/00006199-199909000-00007

13. Items of the Scale