The significance of the work lies in a stubborn measurement gap. Dementia currently affects approximately 55 million people worldwide, with nearly ten million new diagnoses each year according to the World Health Organization, and most of these individuals are cared for at home by family members before eventually moving into long-term care. Caregiving for a relative with dementia is well established as a risk factor for depression and perceived burden, and the demands typically intensify over time, often becoming overwhelming enough to precipitate a move to a care home. Yet while decades of research have shown that a caregiver’s belief in their own capacity to cope, a concept rooted in Albert Bandura’s classic 1977 theory of self-efficacy, helps explain why some caregivers develop psychological problems while others do not, no existing Dutch-language instrument could measure this construct across the full trajectory of the disease, regardless of whether the person with dementia lived at home or in institutional care.

Self-efficacy, in this context, refers to a caregiver’s belief in their capacity to act adequately and confidently in various situations involving their family member with dementia. The evidence base behind its importance is substantial. Caregivers with higher self-efficacy report fewer depressive symptoms and lower feelings of burden than those with lower self-efficacy. Self-efficacy also appears to buffer the impact of the behavioral and psychological symptoms of dementia, which otherwise drive caregiver burden upward. Beyond protection against distress, high self-efficacy is associated with positive dimensions of caregiving, including feelings of usefulness, appreciation and meaning-making even in difficult care situations, while caregivers with low self-efficacy tend to focus disproportionately on the negative aspects of their role. This dual relevance, both as a shield against burden and as a gateway to positive experience, has made self-efficacy an increasingly popular outcome measure in intervention research, particularly for psychoeducational programs designed to strengthen caregivers’ coping skills.

The problem the Dutch team confronted was one of fit. A systematic appraisal of eleven existing self-efficacy measures by Pione and colleagues in 2021 revealed that most instruments were too narrow for widespread use. Four focused exclusively on caregivers of people with dementia living in the community, one targeted only self-efficacy for managing behavioral and psychological symptoms, one concerned use of community support services, and one measured the related but distinct concept of sense of competence. Three more had no connection to dementia caregiving at all. Only the RIS Eldercare Self-efficacy Scale, developed by Benjamin Gottlieb and Joanne Rooney in 2003, was designed to capture caregiving self-efficacy regardless of the care recipient’s living situation, making it the natural candidate for translation and validation in Dutch.

The RIS is deceptively compact. Its ten items are rated on a five-point scale ranging from “I’m certain I can’t do this” to “I’m certain I can do this,” and it can be completed in under five minutes, either by self-report or in an interview. The items cluster into three theoretically distinct subscales. Instrumental self-efficacy reflects a caregiver’s confidence in accomplishing the essential tasks of personal care. Relational self-efficacy captures confidence in preserving the reciprocal relationship with the relative, maintaining connection and mutual recognition as the disease progresses. Self-soothing self-efficacy addresses the ability to protect one’s own well-being despite the emotional demands and stressors of caregiving. Subscale scores are calculated as means, with higher values indicating greater confidence in each domain. The translation into Dutch followed the standard forward and back-translation procedure to ensure conceptual equivalence across languages.

The validation data came from the fifth measurement cycle of the Living Arrangements for People with Dementia study, a long-running cross-sectional research program first conducted in 2008 and repeated every two to three years, funded in part by the Dutch Ministry of Public Health, Welfare and Sports. During the 2019 to 2020 data collection wave, 58 long-term care homes participated. Family caregivers of randomly selected residents were invited by mail to complete a questionnaire covering demographics, role perception, perceived pressure, involvement in care, and self-efficacy, available in both paper and online formats, a dual format chosen deliberately because some caregivers, particularly older individuals, may lack digital skills. In total, 479 caregivers completed the survey.

The statistical evaluation followed established psychometric practice. The sample size comfortably exceeded conventional requirements, with a subject-to-item ratio of nearly 48 to 1, well above the 10 to 1 benchmark typically deemed adequate for factor analysis. A principal component analysis on all ten items produced a Kaiser-Meyer-Olkin measure of sampling adequacy of 0.86, indicating the data were highly suitable for factor analysis, and Bartlett’s test of sphericity was strongly significant. The analysis recovered a clean three-factor solution mirroring the original Canadian validation, with instrumental, self-soothing and relational self-efficacy emerging as distinct dimensions. Together the three factors explained 69.7 percent of the variance, slightly exceeding the 66 percent reported in the original study, and factor loadings ranged from 0.70 to 0.86, indicating that each item mapped strongly onto its intended factor.

Reliability also held up. Internal consistency, assessed by Cronbach’s alpha, was adequate for all three subscales, at 0.83 for instrumental, 0.81 for self-soothing and 0.78 for relational self-efficacy, all within the accepted range of 0.70 to 0.95. No item pairs showed correlations so high as to suggest redundancy, no items correlated below 0.20 with the rest of their subscale, and all item-total correlations exceeded the conventional 0.30 threshold. Construct validity was tested by examining expected patterns of correlation with two other instruments from the LAD study battery. As predicted, the Dutch RIS showed small but significant positive relationships with the Family Perceptions of Caregiving Role instrument, which measures how involved caregivers feel in their relative’s care, and small negative relationships with the Self-Perceived Pressure from Informal Care scale, which measures the felt strain of caregiving. Because ceiling effects emerged on all items, with more than 15 percent of respondents scoring in the top 10 percent of the range, the team applied Tobit regression, a statistical approach designed for data compressed against a boundary, to analyze these relationships appropriately.

The group comparisons revealed a pattern with real-world implications. Women scored significantly higher than men on instrumental self-efficacy, possibly because female caregivers tend to be more familiar with providing care and carry more helpful cognitions about their role. Spouses and partners reported lower self-efficacy than adult children across all three subscales, a finding the authors attribute to reduced mastery and weaker social support among spousal caregivers, who are often older and more isolated. Caregivers aged 60 or younger scored higher than older caregivers on every subscale. Given that roughly two-thirds of respondents were children or children-in-law of the person with dementia and a similar proportion were female, the authors suggest that daughters and daughters-in-law likely experience the highest caregiving self-efficacy, consistent with earlier findings and possibly explained by the multiple life roles such women juggle, which may generalize into confidence in the caregiving role itself. Notably, the original Canadian validation study found no gender differences, making this a point of divergence between the two populations.

The study is not without limitations, and the authors are candid about them. The ceiling effects on every item are the most significant, since scores compressed at the top of the scale can mask genuine differences between caregivers and may prevent the detection of improvement in intervention studies, precisely the setting where the instrument is most needed. The team suggests that alternative response formats, such as visual analogue scales or positively packed rating scales, might improve differentiation and responsiveness in future work. The sample also included only caregivers of people with dementia living in long-term care homes, so validation in home-based caregiver populations, where most people with dementia actually live, remains a priority, even though the original RIS was validated in exactly that community setting. The sensitivity of the Dutch RIS to change over time has yet to be tested.

Even with these caveats, the practical payoff is clear. A brief, valid Dutch measure of caregiving self-efficacy enables researchers to monitor how caregivers’ confidence evolves across the disease trajectory and across care settings, and gives evaluators a proper yardstick for psychoeducational programs, an area where the evidence is genuinely mixed. Systematic reviews have found strong evidence that multicomponent psychoeducational interventions enhance caregiver self-efficacy, while recent meta-analyses of online psychoeducation programs have found no significant impact. Resolving that contradiction requires reliable outcome measures, and the Dutch RIS, requiring less than five minutes of a caregiver’s time, now fills a long-standing hole in the Dutch-language toolkit. The authors also speculate that, with appropriate modification and revalidation, the instrument could prove useful for family caregivers of people with other chronic diseases, extending its reach well beyond dementia care.