Multicomponent School of Caring Program to Support Caregivers of Patients With Dementia: A Quasi-Experimental Effectiveness Study in Primary Care

Abstract

PURPOSE

We aimed to evaluate the multicomponent School of Caring program supporting caregivers of patients with dementia with respect to caregiver burden, quality of life, emotional well-being, self-efficacy, and social support.

METHODS

We conducted a quasi-experimental pre-/postintervention study with a 6-month follow-up conducted in primary care in Spain. Participants were caregivers of dependent people with dementia. Sociodemographic characteristics and information regarding care tasks were collected using an ad hoc questionnaire. We evaluated the program’s effectiveness using the Zarit Burden Interview, EuroQol 5 dimensions 3 levels scale, General Self-Efficacy Scale, Warwik-Edinburgh Mental Wellbeing Scale, and Oslo Social Support Scale, validated questionnaires to assess caregiver burden, quality of life, self-efficacy, emotional well-being, and social support. The analysis compared the results of the outcomes at 3 time points (baseline, postintervention, and follow-up). We used multivariate regression to identify characteristics related to improvements produced by the program.

RESULTS

The School of Caring program was delivered to 8 groups, comprising 14-19 participants/group, from 2019 to 2023. A total of 132 caregivers completed the program and were evaluated pre-/postintervention, 33 (25%) of whom were lost to follow-up. The program produced significant improvements in all outcomes, which was maintained up to 6 months postintervention. At this stage, inexperienced caregivers showed the greatest improvement compared with former caregivers relative to burden levels. Overall, the amount of time dedicated to caregiving each day and experiencing a health issue were the elements that most significantly affected caregiver burden and quality of life.

CONCLUSIONS

The School of Caring program produced notable improvements in the burden and well-being of informal caregivers of dependent people with dementia. The findings provide evidence of the effectiveness of group-based, community primary care programs that support caregivers of people with high care demands such as dependent people with dementia.

INTRODUCTION

Alzheimer disease and other dementias are chronic conditions that cause long-term dependency in those affected. As such, they have significant health and social impacts, but above all, they seriously affect the family environment.^1,2 A patient with a diagnosis of dementia might need an average of 70 hours of care per week, placing a heavy burden on caregivers and significantly affecting their daily lives.³ The role of the informal caregiver takes on social relevance as they become the main person responsible for caring for others. Approximately 60% of a caregiver’s burden arises from the subjective perception of the disease, and 40% includes factors such as social isolation, stress, physical exertion, feelings of guilt, and emotional pressure.^4,5 In addition to the burden, caregiver quality of life is affected by the progression of the illness of the person being cared for, other family and work responsibilities, and whether or not they live with the patient.⁶ Social class, migration, and gender roles in informal care are other factors that must be considered when analyzing the effect of caregiving on the daily lives of caregivers.^7,8 The profile of informal caregivers is highly gendered; in Spain, 73.8% are women, in line with similar figures for the rest of Europe.⁹

The complexity of dementia and its chronic progression over time requires the coordinated use of all available social, health, and community resources to care for patients and their caregivers. These include primary health care, general hospitals, social and health services, and family associations.¹⁰ Multicomponent interventions designed to support caregivers include a combination of education, problem solving, and skills training. According to several studies, these types of interventions can decrease the caregiver’s burden, alleviate depressive symptoms, decrease the caregiver’s discomfort with the symptoms associated with dementia, and improve their quality of life.^11,12 In general, psychosocial, psychoeducational, and tailored interventions are associated with substantial decreases in caregiver burden.^13-17 However, contrary to the vast scientific literature, there are very few consolidated programs implemented in primary care settings. In fact, according to the latest Organization for Economic Co-operation and Development report on policies supporting informal caregivers of older people, support programs are typically provided by social services, and the health sector is not very involved.¹⁸

The School of Caring project is a mixed-methods program¹⁹ completely deployed in the primary care setting. Grounded on validated interventions,^20,21 the program was codesigned with the participation of primary care health professionals and caregivers of people with dementia. A recent study identified the following 3 key components that should be incorporated into a support program for caregivers of people with dementia: education and care skills, professional psychological support, and social and community resources.²² The main objective of the present study was to evaluate the effectiveness of the multicomponent and multidisciplinary School of Caring program in alleviating caregiver overload and improving quality of life, emotional well-being, self-efficacy, and social support.

METHODS

Trial Design

We conducted a quasi-experimental study based on a multicenter, single-arm, pragmatic, community-based intervention deployed as part of primary care. Intervention effectiveness with respect to caregivers’ outcomes was assessed by repeated measures at the following 3 time points: preintervention, postintervention, and at a 6-month postintervention follow-up.

The School of Caring project was approved by the Research Ethics Committee of the Institut d’Investigació en Atenció Primària Jordi Gol (code 18/072). All participants were given an information sheet about the study and provided signed informed consent. Consensus Reporting Items for Studies in Primary Care was used as a reporting guideline (Supplemental Appendix 1).²³

Participants and Setting

Clinics

Clinical participants comprised primary care professionals from the primary health care management area, including family doctors, nurses, physiotherapists, social workers, neurologists from the neurology department of the referral hospital, and psychologists from the local Association of Relatives of People with Alzheimer’s (https://afamontsia.org).

Caregivers

Caregiver participants comprised those adults caring for noninstitutionalized people with cognitive impairment/dementia, with a degree of dependency of 1 or 2 or awaiting their degree to be determined (dependence degree 1: patient requiring assistance recurrently, or ≥ 1 time a day to perform basic activities of daily living; dependence degree 2: patient requiring assistance 2-3 times a day but not needing a permanent caregiver to perform activities of daily living), in accordance with the Spanish National Law on the Promotion of Personal Autonomy and Care for Dependent Persons.²⁴ Caregivers of institutionalized people or those with difficulties completing the questionnaires were not eligible.

Setting

The study was conducted in the primary care service of the Catalan Health Institute in a rural area of northeastern Spain, with an aging population (23% of the region’s population is > 65 years of age) during the period 2019 to 2023.

Recruitment

The research team presented the School of Caring project to the aforementioned professionals, and those who agreed to participate voluntarily were included in the study. The participating professionals were responsible for reviewing the census of their assigned population and listing the caregivers who met the eligibility criteria. The research team contacted the listed caregivers by telephone and set about creating groups of 14-19 participants as soon as they agreed to participate in the study, until the calculated sample size was reached.

Sample Size Calculation

We calculated the sample size to fulfil the requirement to detect a minimum difference between the pre- and postintervention means of 5 points in the main variable of caregiver burden, as measured with the Zarit Burden Interview questionnaire. A 95% confidence level and 90% power were established in a 2-tailed test, assuming an SD of 16. A sample size of 132 subjects was obtained, allowing for a 30% loss rate.

Intervention

The School of Caring program is a multicomponent intervention based on the following 3 main components: (1) training, (2) psychological support, and (3) social and community resources. The first component addresses theoretical aspects of dementia and practical aspects of caregiving tasks. The second component focuses on helping to identify and cope with overload and to recognize the spectrum of feelings that emerge at different stages of their family member’s illness. The third component provides information on social assistance and community resources. These components are integrated into a group intervention carried out by a multidisciplinary team in a community space via nine 90-minute sessions (Supplemental Appendix 2). Sessions were delivered once per week in the afternoon (15:30-17:00), and home-based assistance for patients was offered to facilitate caregivers’ attendance at the sessions. Assistance allocation was based on social workers’ availability from local social services or the local Association of Relatives of People with Alzheimer’s.

Data Collection and Outcomes

Sociodemographic, health, and caring-related data were collected from caregivers via an ad hoc questionnaire administered at baseline. The primary outcome was caregiver burden, assessed by the Zarit Burden Interview.^25,26 Secondary outcomes were health-related quality of life, emotional well-being, self-efficacy, and social support, as measured by the validated questionnaires EuroQol 5 dimensions 3 levels scale (EQ-5D-3L),²⁷ Warwik-Edinburgh Mental Wellbeing Scale (WEMWBS),²⁸ General Self-Efficacy Scale (GSES),²⁹ and Oslo Social Support Scale (OSSS-3),³⁰ respectively. All questionnaires were self-completed by caregivers on 3 occasions (baseline, postintervention, and at 6 months after the intervention [medium-term follow-up]). In addition, the following process outcomes were recorded: recruitment, adherence and retention, and program satisfaction (Supplemental Appendix 3).

Data Analysis

The following descriptive statistics of sociodemographic variables of the caregivers in the sample were calculated: mean and SD for quantitative variables, and absolute (N) and relative (%) numbers for qualitative variables. To evaluate the effect of the intervention on the study outcomes before, immediately after, and in the medium-term postintervention period, 1-way analysis of variance and Dunnett multiple comparison tests were used to examine effects relative to baseline values. We assessed Hedges’ g effect sizes measurement, outlier identification, and false-discovery rate control according to the Benjamini, Krieger, and Yekutieli method (Supplemental Data Set).³¹ We used the χ² test to compare the proportions of caregivers with each of the different degrees of overload before and immediately after the intervention and in the medium-term postintervention period. To study the relations between independent variables and each different study outcome, a mixed multiple regression model was used, for which the independent variables were introduced as fixed effects, and the participants and 3 time points were considered random effects. Results present the data for the complete sample unless otherwise stated. To address potential attrition bias, 2 sensitivity analyses were performed: complete case analysis and worst-case scenario imputation (Supplemental Data Set). We used a significance level of P < .05. We performed statistical analysis and data presentation using GraphPad Prism v8 (Dotmatics) and R v4.5.0 (R Project for Statistical Computing).

RESULTS

Reach

The School of Caring program involved 28 professionals from 7 primary care centers and 21 professionals from hospital and community services. In accordance with the eligibility criteria, 154 caregivers were invited to participate in the study, 132 of whom accepted and were assigned to an intervention group (Figure 1). None dropped out of the intervention. Thirty-three dropped out of follow-up, owing to institutionalization or death of the patient. We then analyzed the data for the 132 caregivers who completed the intervention. In total, the program was delivered to 8 groups comprising 14-19 participants each, and it was carried out in 7 community spaces.

Figure 1.

Participant Flow Diagram

Sociodemographic Characteristics of Caregivers

Caregivers were predominantly middle-aged and elderly (mean [SD] age 66.93 [9.24] years) women and married (Table 1). Most were homemakers or retired. More than one-half had some health problem and needed daily medication.

Table 1.

Caregiver Sociodemographic Characteristics

Characteristic	Total sample (N = 132)
Sex, No. (%)
Female	117 (88.6)
Age, y, mean (SD)	66.93 (9.24)
Relationship status, No. (%)
Single	6 (4.5)
Married	109 (82.6)
Separated	10 (7.6)
Widowed	7 (5.3)
Living area, No. (%)
Urban	45 (34.1)
Rural	87 (65.9)
Education level, No. (%)
Illiterate	2 (1.5)
Primary education	97 (73.5)
Vocational education and training	18 (13.6)
General Certificate of Education	10 (7.6)
University degree	5 (3.8)
Employment status, No. (%)
Homemaker	45 (34.1)
Retired	54 (40.9)
Employed	29 (22.0)
Unemployed	4 (3.0)
Health problem, No. (%)
Yes	83 (62.9)
Medication, No. (%)
Daily	73 (55.3)
Occasionally	48 (36.4)
Never	8 (6.1)
Medication type, No. (%)
Analgesic	71 (53.8)
Anxiolytic	31 (23.5)
Sleep aid	29 (22.0)
Diabetes treatment	14 (10.6)
Hypertension treatment	38 (28.8)
Perceived health status
1-10 (1 = poor, 10 = excellent), mean (SD)	6.52 (1.8)

Caregiving Characteristics

We assessed characteristics and context of care using an ad hoc questionnaire administered at the start of the study, before the intervention. The caregivers in the sample mainly cared for their partners or parents (Table 2). A total of 85% of caregivers spent ≥ 7 hours a day caring for their relative; in fact, the majority spent 13-24 hours a day on caregiving. In addition, 1 in 4 caregivers in the sample was the primary caregiver for another person, in addition to the dependent person with dementia. A total of 30% reported being completely alone in caring for their family member, whereas the rest reported receiving help from family members or hired nonprofessional staff, although more than one-half considered this help to be insufficient. Consequently, overall, 80% felt a loss of their social life.

Table 2.

Caregiving Characteristics

Characteristic	Total sample (N =132)
Patient sex, No. (%)
Male	82 (62.1)
Relationship to patient, No. (%)
Spouse	50 (37.9)
Offspring	69 (52.3)
Relative	11 (8.3)
Unrelated	2 (1.5)
Daily time spent on caregiving, h, No. (%)
< 1	1 (0.8)
1-6	19 (14.4)
7-12	28 (21.2)
13-24	84 (63.6)
Caregiving duration, y, No. (%)
< 1	19 (14.4)
1-5	45 (34.1)
6-10	40 (30.3)
> 10	27 (20.5)
Cohabitation, No. (%)
Yes	110 (83.3)
Number of people cared for, No. (%)
1	83 (62.9)
2	35 (26.5)
3	12 (9.1)
4	1 (0.8)
Quality of relationship with patient, No. (%)
Easy	59 (44.7)
Difficult	57 (43.2)
Indifferent	16 (12.1)
Work dropout owing to caregiving, No. (%)
No	112 (84.8)
Assistance in caregiving tasks, No. (%)
Yes	90 (68.2)
Assistance type, No. (%)
Relatives	54 (40.9)
Hired nonprofessional	18 (13.6)
Hired professional	5 (3.8)
Health or social services	17 (12.9)
Assistance sufficiency, No. (%)
No	75 (56.8)
Home social assistance, No. (%)
No	98 (74.2)
Telecare program, No. (%)
No	127 (96.2)
Social assistance for dependents, No. (%)
Degree 1	43 (32.6)
Degree 2	69 (52.3)
Pending	20 (15.2)
Need to share emotional state, No. (%)
Always	32 (24.2)
Sometimes	91 (68.9)
Never	9 (6.8)
Opportunities to share emotional state, No. (%)
Always	28 (21.2)
Sometimes	56 (42.4)
Less than I would like	43 (32.6)
Never	5 (3.8)
Feeling of having little time for self-care, No. (%)
Always	28 (21.2)
Often	38 (28.8)
Sometimes	57 (43.2)
Never	9 (6.8)
Feeling of loss of social life, No. (%)
Yes	105 (79.5)
No	17 (12.9)
Do not know	10 (7.6)

Primary Outcome: Caregiver Burden Assessed by Zarit Burden Interview

The main variable, caregiver burden, was assessed using the Zarit Burden Interview questionnaire. A scale of 22-110 points was used. No burden was considered to exist for scores of < 47 points. A mild burden was concluded for 47-55 points and a severe burden for > 55 points.²⁶ First, we evaluated the distribution of the levels of burden among the sample. Before the intervention, 17% of caregivers had no burden, and 65% had severe burden (Figure 2A). After the intervention, the percentage of caregivers without overload had increased by 11 points to 28%, whereas the percentage of caregivers with severe overload had decreased to 53%. This decrease was maintained at the 6-month follow-up. In line with this, the mean overall score for the sample with regard to overload dropped from 61.77 points before to 57.62 after the intervention (Figure 2B). This significant decrease (P < .0001) was maintained up to 6 months postintervention.

Figure 2.

Effect of School of Caring Program on Caregiver Burden

A. Stacked bar chart showing distribution of the sample by burden level at baseline, postintervention, and at 6-month follow-up. B. Bar chart with error bars showing Zarit Burden Interview scores for the entire sample at the different times. C. Line chart with error bars showing Zarit Burden Interview scores by caregiving duration (y) at the different times.

ANOVA = analysis of variance; Post = postintervention.

Note: Data are expressed as mean ± SE. Statistical analysis involved the χ² test for (A), 1-way ANOVA with Dunnett multiple comparison test for (B), and 2-way ANOVA with Dunnett multiple comparison test for (C).

^a P < .0001.

Analyzing the data with respect to the number of years spent caring showed that by the end of the intervention, the School of Caring program appeared to have had a positive effect of improving the burden by a similar magnitude in all groups, and this was maintained during follow-up in the intermediate groups (1-5 and 6-10 years) (Figure 2C). However, caregivers with > 10 years of caregiving experience went back to having levels of overload similar to those at the start; they were the only group to be worse off after intervention. It should be noted that caregivers who had been caring for < 1 year were those who started with lower levels of burden and, in turn, were those who benefited most from the program and who continued to improve during the follow-up period.

Secondary Outcomes: Quality of Life, Self-Efficacy, Mental Well-Being, and Social Support

The School of Caring program produced an improvement in the 4 secondary outcomes studied: health-related quality of life assessed by the EQ-3D-5L questionnaire; self-efficacy assessed by the GSES; mental well-being assessed by the WEMWBS, and social support assessed by the OSSS-3 (Figure 3A-D). A small to moderate effect size according to Hedges’ g measurement was obtained after the intervention (P < .001), which was maintained during follow-up (P < .001). Concretely, the effect sizes obtained after the intervention were 0.30 (EQ-3D-5L), 0.48 (GSES), 0.44 (WEMWBS), and 0.25 (OSSS-3).

Figure 3.

Bar Charts With Error Bars Showing Effects of School of Caring: (A) Caregiver Health-Related Quality of Life, (B) Self-Efficacy, (C) Mental Well-Being, and (D) Social Support

ANOVA = analysis of variance; EQ-5D-3L = EuroQol 5 dimensions 3 levels scale; GSES = General Self-Efficacy Scale; OSSS-3 = Oslo Social Support Scale; Post = postintervention; WEMWBS = Warwik-Edinburgh Mental Wellbeing Scale.

Note: Data are expressed as median (95% CI). Statistical analysis involved 1-way ANOVA with Dunnett multiple comparison test for (A-D).

^a P < .0001.

Adjusted Analysis: Mixed Regression Models

We performed a multivariate analysis to evaluate the moderating effect of sociodemographic characteristics and the caregiving context. We generated a regression model that included the most relevant independent variables and evaluated their relations with the outcomes obtained during the study period (Table 3). The variables that influenced both overload and quality of life were the presence of a health problem and daily time providing care. Having a health problem increased overload by 64.51 points and decreased quality of life by 94%, on average. In addition, every additional hour devoted to caregiving increased the burden by an average of 3.97 points and decreased quality of life by 5%. Adequacy of the help received by caregivers was the variable that provided significant values for all outcomes except quality of life (P = .057). Specifically, the improvement in overload scores was 13.2 (2.67) points higher on average among caregivers who considered that they received sufficient help compared with caregivers who thought that they did not receive sufficient help. Likewise, on average the former group of caregivers scored approximately 2 points higher on the emotional well-being and self-efficacy scales (2.32 and 1.83, respectively) than caregivers who did not receive sufficient help. In terms of quality of life, receiving sufficient help was associated with an 8% improvement compared with those who did not receive sufficient help (0.08 points [0.04]; P = .057). In addition, we observed that for every point of improvement in self-perceived health, there was a 2.78-point decrease in the burden on caregivers participating in the program.

Table 3.

Model Estimates for Primary and Secondary Outcomes

Outcome	Zarit Burden Interview		EQ-5D-3L		WEMWBS		GSES
	Estimate (SE)	P value	Estimate (SE)	P value	Estimate (SE)	P value	Estimate (SE)	P value
Health problem, yes	64.51 (10.86)	<.001a	−0.94 (0.16)	<.001a	29.18 (3.74)	<.001a	27.49 (3.71)	<.001a
Daily time providing care	3.97 (1.65)	.018a	−0.05 (0.02)	.05a	0.15 (1.82)	.933	−0.47 (1.81)	.795
Number of people cared for	1.96 (1.78)	.275	−0.02 (0.03)	.477	0.17 (0.6)	.782	0.37 (0.59)	.534
Medication, sometimes	5.2 (3.46)	.137	0.01 (0.01)	.573	0.37 (0.3)	.227	0.26 (0.32)	.423
Medication, daily	4.01 (4.99)	.425	−0.09 (0.07)	.227	0.91 (1.7)	.593	0 (1.66)	.999
Perceived health status	−2.78 (0.9)	.003a	−0.06 (0.05)	.287	−1.42 (1.18)	.231	−0.91 (1.18)	.441
Assistance sufficiency, yes	−13.2 (2.67)	<.001a	0.08 (0.04)	.057	2.32 (0.91)	.013a	1.83 (0.92)	.05a
Home social assistance	−5.2 (2.94)	.081	0.04 (0.04)	.376	0.04 (1)	.966	0.16 (0.99)	.873
Dependency assistance 1	2.61 (3.91)	.507	−0.07 (0.06)	.217	−0.95 (1.33)	.48	− 1.33 (1.33)	.323
Dependency assistance 2	6.76 (3.94)	.09	−0.1 (0.06)	.103	−1.73 (1.33)	.198	−1.81 (1.32)	.175
Caregiving duration < 1 y	−2.82 (5.36)	.6	−0.11 (0.08)	.158	0.09 (0.56)	.875	−0.7 (0.56)	.217
Caregiving duration 1-5 y	−2.24 (3.75)	.553	0.06 (0.06)	.315	−0.01 (1.28)	.995	−0.37 (1.3)	.779
Caregiving duration 6-10 y	−3.4 (3.96)	.392	0 (0.06)	.983	− 1.03 (1.34)	.446	−0.57 (1.36)	.675
Caregiving duration > 10 y	−7.77 (5)	.124	0 (0.07)	.953	0.23 (1.69)	.891	0.79 (1.7)	.641

EQ-5D-3L = EuroQol 5 dimensions, 3 levels scale; GSES = General Self-Efficacy Scale; WEMWBS = Warwik-Edinburgh Mental Wellbeing Scale.

Note: Data from complete case analysis.

^aP value < .05.

Process Outcomes: Recruitment, Adherence and Retention, and Participant Satisfaction

The recruitment rate was defined as the proportion of eligible professionals and caregivers contacted who started the study. Adherence and retention rates were defined as the proportions of caregivers who attended all program sessions and who completed the study, including follow-up, respectively. Satisfaction was assessed using a locally validated questionnaire for primary care intervention programs.

The study obtained 85% recruitment, 100% adherence, and 75% retention rates. The mean (SD) degree of satisfaction was 56.59 (3.67) out of a possible total of 60 points (Supplemental Appendix 3).

DISCUSSION

This project showed the effects of a group intervention in the field of community primary care, based on theoretical and practical training on dementia and its care, psychological support, and the provision of information on social and community resources. The project examined effects on the burden, quality of life, and emotional well-being of informal caregivers of people with dementia. This is highly pertinent given that a recent review concluded that there is insufficient evidence showing the effectiveness of group-based interventions in improving the well-being of caregivers of patients with dementia.³²

The School of Caring program decreased burden by an average of 4 points on the Zarit Burden Interview scale, consistent with results of previous studies on the effectiveness of multicomponent interventions in decreasing the burden on caregivers of people with dementia.^33,34 Likewise, the School of Caring program improved the overall quality of life score, although depression was the only dimension for which a significant improvement was observed, reinforcing the findings of earlier studies showing that caregivers of people with dementia experience more depression and anxiety and that psychoeducational interventions can improve these symptoms and emotional well-being^15,35 The results obtained regarding the program’s effect on caregivers’ self-efficacy were more modest but nevertheless significant. Self-efficacy plays an important role in caregiver outcomes, and studies had suggested that strategies aimed at improving caregiver self-efficacy should be incorporated into interventions targeting informal caregivers.^36,37

The design of the intervention, carried out in advance with primary care professionals and with the participation of local social and community agents,²² was crucial for its optimal development. In addition, the School of Caring program is in line with the national strategic health plan³⁸ and the Spanish health system transition³⁹ in terms of sociodemographic demands, intersectionality, multidisciplinarity, and provision of evidence-based therapies that would favor its adoption and scalability. We argue that community interventions aimed at informal caregivers should be a basic primary care service that is complementary to other services and public policies aimed at this group, and whose purpose is to care for caregivers’ physical and mental health.^40,41 Therefore, this approach is based on person-centered care for caregivers. This population group is growing in parallel with the increase in dementia cases predicted over the coming years.^42,43 The chronicity and dependence of people with dementia place great demands on care, which is why informal caregivers are of exceptional importance in this context. In fact, there is ample evidence of the cost-effectiveness of interventions aimed at informal caregivers and of the positive health effects on people with dementia and their caregivers.^42,43

Strengths and Limitations

The major strength of this study lies in its pragmatic nature. The School of Caring program was designed and deployed by primary care professionals with the involvement of main stakeholders. The quasi-experimental single-arm trial design limited the strength of evidence compared with randomized controlled trials; nevertheless, it has many advantages such as practical feasibility, direct applicability into routine practice, low cost, and being beneficial for all study participants. It also provides evidence of the relevance and effect of community-based primary care programs on the health of informal caregivers. However, the study had a local focus; therefore, the results cannot be generalized to other community and primary care services, although it is capable of adaptation to other primary care services. Another factor that should be considered is that caregivers faced difficulties in attending the program, owing to lack of time or the impossibility of leaving the person being cared for alone; therefore, the results of this study might have been subjected to selection bias.

CONCLUSION

The multicomponent community program School of Caring, designed with the participation of caregivers and primary and community care professionals, is effective in alleviating the burden on informal caregivers of people with dementia. In addition, the program can improve the quality of life, emotional well-being, self-efficacy, and social support of participants.