Participants

Participants were 130 primary caregivers of an individual with dementia recruited from the Instituto de Neurociencias de San Lucas, Rosario Argentina and the CETYS University in Baja California, Mexico. The sample was composed of caregivers from Argentina (n = 110) and Mexico (n = 20). In this study, caregivers were defined as individuals providing active daily care for a person with dementia. Inclusion and exclusion criteria designated that the participant must be the primary caregiver of the person with dementia; must have been providing care to the person with dementia for at least three months; must be at least 18 years old; and must have no personal history of physical, psychological, or neurological problems. All caregivers provided care to a patient who had had a diagnosis of dementia, which was confirmed by a medical records review.

The mean age of caregivers was 56.84 years old (SD = 13.18), and most of the sample (77.7%) was female. The majority of the participants were married (76.9%), 12.3% were never married, and 4.6% were divorced or separated. Regarding relationship to the patient, 43.8% of caregivers were spouses, 43.1% were children, 7.7% were aunts or uncles, 2.3% were other, 1.5% were professional caregivers, 0.8% were friends, and 0.8% were parents. The average weekly time spent caregiving was 68.48 hours (SD = 30.64), and participants had been providing care for an average of 46.94 months (SD = 26.66). In regard to employment, 36.9% of the sample worked part-time outside the home, 7.7% worked full-time outside the home, and 10.8% were retired. In terms of highest level of education completed, 1.5% of the sample did not complete elementary school, 14.6% completed elementary school, 3.1% did not complete high school, 37.7% completed high school, 1.5% did not complete technical school, 3.8% completed technical school, 3.1% did not complete college, 30.8% completed college, and 38% completed post-graduate work. For income level, 0.8% of the sample earned under minimum wage, 13.1% earned one to two times minimum wage, 37.7% earned three to four times minimum wage, 24.6% four to five times, and 23.8% earned over five times minimum wage.

Measures

Caregivers completed a series of questionnaires assessing resilience and related constructs. A Spanish version of the 13-Item Sense of Coherence Scale (SOC-13)51 was readily available.52 The Resilience Scale for Adults (RSA)22, Brief Resilience Scale,53 and Life Orientation Test-Revised (LOT-R)54 did not have Spanish versions available and were translated into Spanish. Chapman and Carter’s55 methodology was used by which a researcher who was bilingual and bicultural translated the original measure into Spanish. Another bilingual and bicultural researcher translated the items back into English. If any discrepancies arose between the original English and re-translated English versions of the measure, they were resolved mutually.

Resilience Scale for Adults (RSA). The RSA22 was administered to assess the presence of protective resources that promote healthy adjustment to psychosocial adversity. As noted above, this 36-item scale consists of five dimensions of resilience: personal competence, social competence, family coherence, social support, and personal structure. On a 7-point Likert scale, respondents indicate their degree of agreement with each item (1 being “not at all true” to 7 being “very true”). Although the scale authors originally developed 45 items, there were nine items that did not load distinctly onto any factor or achieve simple structure22 and thus were removed from the current study, bringing the RSA to 36 items. Total scores range from 36 to 252, with higher scores indicating greater intrapersonal and interpersonal protective factors that are presumed to facilitate adaptation to life stresses.22 Reliability of the RSA is adequate, with subscale αs ranging from .67 to .90, as are all subscale test-retest correlations (.69 to .84).22 The scale has high construct validity, as evidenced by strong convergent and discriminant validity and its ability to differentiate between a patient sample and randomly chosen control sample.22 Windle and colleagues,50 in their review of resilience measurement scales, identified the RSA as a scale of high quality across a number of criteria.

Brief Resilience Scale (BRS). In order to provide an index of convergent validity for the RSA, the BRS53 was administered which measures the ability to recover from stress or “bounce back.” The scale is comprised of six items measuring a unitary construct of resilience. Caregivers were asked to rate the extent to which they agreed with each item, following a Likert-style response format ranging from 1 (“Strongly disagree”) to 5 (“Strongly agree.”) Scores range from 6 to 30, and higher scores suggest a greater ability to “bounce back.” The BRS has high internal consistency ranging from .80 to .91. Test-retest reliability is high, with a one-month intraclass correlation (ICC) .69 for one sample and a three-month ICC of .62 for a second sample.53

Life Orientation Test-Revised (LOT-R). To measure caregivers’ degree of dispositional optimism, the current study utilized the 10-item LOT-R.54 Participants respond on a 5-point Likert- style scale ranging from 0 (“Strongly Disagree”) to 4 (“Strongly Agree”). Total scores range from 0 to 40, with greater scores implying higher levels of optimism and a more positive overall outlook.54 Scheier and colleagues54 found a Cronbach’s α of .78, suggesting an adequate level of internal consistency, as well as test-retest reliability correlations ranging from .56 to .79. The LOT-R has demonstrated appropriate convergent and divergent validity; for instance, it is positively correlated with measures of self-esteem and self-mastery and negative correlated with measures of trait anxiety and neuroticism.54

Sense of Coherence Scale (SOC). In the current study, the 13-item SOC was used to assess the tendency to view the world as comprehensible (5 items), manageable (4 items), and meaningful (4 items).51 “Comprehensibility” refers to the feeling that the world makes sense, and that information about the environment is ordered, consistent and can be explained. “Manageability” is the feeling that sufficient resources are available for meeting internal and external demands, while “meaningfulness” refers to the feeling that these demands are worthy of investment and engagement.51,56 There are at least 15 different versions of the questionnaire available, with the SOC-13 having been used in at least 33 languages in 32 countries.57 While response anchors differ, each item is answered with two opposite anchoring phrases on a 7-point scale. Total scores range from 13 to 91 with higher total scores indicating a greater sense of coherence or meaningfulness.51 Eriksson and Lindstrom57 found that in 127 studies using the SOC-13, α values ranged from .70 to .92. Test-retest correlations ranged from .69 to .78 (1 year), .64 (3 years), .42 to .45 (4 years), .59 to .67 (5 years) to .54 (10 years). There is strong evidence for criterion validity, as evidenced by a high negative correlation with anxiety and depression and a high positive correlation with optimism and self-esteem57. The SOC-13 has been previously validated on eight Spanish samples of varying age, education, gender, levels of disability, and level of functioning52; thus, this version of the scale was used in the present study. Virues-Ortega and colleagues52 found adequate internal consistency (Cronbach’s α = .80) as well as moderate correlations with self-reported measures of physical health, quality of life, and depression.52

Procedure

Participants were recruited from the Instituto de Neurociencias de San Lucas, Argentina and CETYS University in Baja California, Mexico. Caregivers in Argentina were recruited from neurologist appointments attended by the family member with dementia. The caregiver was interviewed at the hospital while the patient was meeting with the physician. Participants in Mexico were recruited through flyer and word-of-mouth within the local community and via telephone using a contact list provided by the Alzheimer’s Foundation of Baja California. The majority (85%) of these caregiver interviews took place at the individual’s home; three interviews occurred at the Foundation. A psychologist met with each caregiver for approximately one hour during which socio-demographic information was collected and questionnaires were administered. All enrolled participants provided informed consent and were told that they could withdraw from the study at any time. Approval for this study was obtained by the appropriate ethics committees.

Statistical Analyses

Data were analyzed using SPSS v.23 and AMOS v.23. First, a confirmatory factor analysis (CFA) was conducted on the RSA to determine whether the original version’s previously identified factor structure emerged in the same manner for dementia caregivers from Latin America in the Spanish version. A CFA tests the fit of a hypothesized pattern of relationships among manifest (observed) variables and latent (hidden) variables. Although traditionally CFAs tend to use samples greater than 200 participants,58 models with high degrees of freedom can achieve higher power with smaller samples.59 MacCallum et al.59 found that models with degrees of freedom of 100 applied to samples with 100 participants have power between .65 and .43. Because the degrees of freedom for the CFA in this study is 584, the current sample size (n = 130) has substantial power, though the results of the CFA should be interpreted with some caution. Then, to determine whether a better factor structure could emerge and whether any items needed to be removed, an exploratory factor analysis (EFA) with principal axis factoring and a Promax rotation was performed on the items. After a final item composition and subscale structure were identified, Cronbach’s αs were calculated for the total score and subscale scores. The final subscales were then correlated with the BRS, LOT-R, and SOC to examine the Spanish RSA’s convergent validity in this sample of dementia caregivers.