Testing the effects of a text message intervention on depression and distress among Latino dementia caregivers: a randomized controlled trial protocol

Abstract

Background:

Latino caregivers have poor mental health and access to caregiver support services. Here, we describe the protocol for a randomized controlled trial (RCT) to evaluate the effect of a text message intervention on depression and distress among informal Latino dementia caregivers. We will also assess mechanisms of action.

Methods:

We are enrolling 288 Latino dementia informal caregivers 18 or older into a parallel group RCT. Participants randomized to the intervention group receive a remote, asynchronous, bilingual, bi-directional, 6-month texting program focused on dementia education, skill-building and community resources. The control group enters a 7-month waitlist after which they are offered the same intervention as the intervention group. Randomization is stratified by each of the four recruitment sites at a 1:1 ratio. Outcomes (e.g., caregiver depressive symptomatology, distress) are measured via surveys at baseline, 3, 6 and 7 months.

Conclusions:

This RCT addresses two priority areas: eliminating dementia disparities and optimizing caregiver support. Findings have the potential to make clinical and policy-relevant contributions by providing appropriate caregiver support to Latinos in a highly scalable way.

1. Introduction

Family caregiving for people with Alzheimer’s disease and related dementias (ADRD) takes a serious mental health toll [1–5]. Latino family caregivers of people with ADRD experience substantial disparities and are often excluded from advances in research [6, 7]. The number of Latinos with ADRD is growing more than any other ethnic or racial group in the US [8]. Compared to non-Latino Whites, Latinos are more likely to become family caregivers [9], provide longer caregiving, and experience more caregiver distress and depression [4, 9–16]. Caregiver support interventions can improve mental health outcomes [17, 18]. However, most interventions have been designed for or tested among non-Latino Whites [6, 7], and evidence shows extant caregiver support interventions’ generalizability is questionable among Latinos [19–21]. Latinos experience unique barriers that might reduce traditional caregiver support interventions’ feasibility and efficacy, including transportation, financial, language, and cultural barriers [22, 23]. There is a critical need for efficacious caregiver support interventions for Latinos.

To address this need, we developed CuidaTEXT (Spanish for self-care and texting) using a set of user-centered principles [24]. CuidaTEXT is a multidomain, bilingual 6-month text message intervention that includes delivery and content features tailored to Latino caregivers. All caregivers receive one core daily automatic message on five domains: ADRD education, social support, self-care, ADRD care management, and behavioral symptoms. Caregivers can access on-demand automatic messages to expand on the five domains via sending keywords and interact with a coach via a live chat for more personalized help.

CuidaTEXT is the first ADRD caregiver support intervention to capitalize on text messaging. Newer and more fashionable technologies for caregivers exist [25]. These technologies include websites such as “Webnovela Mirela” focused on teaching caregiving coping skills [26], apps that deliver caregiver skills or mindfulness [27], and videocalls with multiple components (e.g., community resources, counseling, ADRD education) [28]. Some of have been tested among Latinos, usually at the pilot stages, having shown some level of improvement pre- to post-intervention in mental health-related outcomes [6]. However, relying only on these newer technologies may perpetuate caregiver support disparities due to low access by Latinos [29]. CuidaTEXT was designed to be highly accessible among Latinos, as nearly 100% of Latino adults own a cellphone and 87% use texting (10% more than non-Latino Whites) [29, 30]. CuidaTEXT’s remote, asynchronous, and largely-automated features address important Latino caregiver barriers, including transportation, time constraints, and implementation costs of caregiver support interventions [31–34]. Its simple bilingual language and tailored content (e.g., resources for the uninsured, immigration-stress tips) address language, cultural, and contextual barriers.

Our one-arm feasibility study (n=24) showed CuidaTEXT was highly usable and acceptable, reduced caregiver stressors (e.g., behavioral symptom distress) and their negative consequences (e.g., depressive symptoms), and improved ameliorating factors of stress (e.g., approach-coping, preparedness for caregiving) at 6 months [24, 35]. In the current manuscript, we aim to describe the protocol for a randomized clinical trial (RCT) we are currently conducting to test the efficacy of CuidaTEXT among Latinos. To optimize intervention potency [36, 37], we will test mechanisms of change informed by the Stress Process Framework [38], the Science of Behavior Change, and our quantitative feasibility data [35].

This project has three specific aims informed by the Stress Process Framework: determine the effects of CuidaTEXT on Latino caregiver stressors and their negative consequences compared to a waitlist group (outcomes; Aim 1), explore CuidaTEXT’s effects on ameliorating factors of stress (targets; Aim 2) and examine the mediation of ameliorating factors on CuidaTEXT’s outcomes (Aim 3; Figure 1). Hypotheses include: Reductions in depressive symptoms (H1) and behavioral symptom distress (H2) will be higher in CuidaTEXT than control at Month 6 (Aim 1); Baseline to Month 6 increases in approach-coping (H3) and preparedness for caregiving (H4) will be higher in CuidaTEXT than control (Aim 2); and CuidaTEXT’s effects on change in depressive symptoms and behavioral symptom distress from Month 3 to 6 will be mediated by change in approach-coping (H5-H6) and preparedness for caregiving (H7–8) from baseline to Month 3 (Aim 3).

Figure 1.

CuidaTEXT’s hypothesized mechanism of change

2. Methods

2.1. Trial design and setting

This is a two-arm superiority RCT that uses parallel design with surveys administered at baseline, 3, 6 months (end of treatment), and 7 months (post-treatment; Figure 2). After completing the baseline assessment, the research team randomizes participants using a computer-generated randomization list implementing random permuted blocks that allocate participants to either CuidaTEXT or control at a ratio of 1:1, stratified by each of the four participating recruitment sites. Given the psychosocial nature of the study, intervention masking is only conducted at the data analyst level via color coding on REDCap, the data collection tool used in the project, so that they are not able to know which group participants were randomized to [39]. Enrollment started in February 2025 and is expected to end in September 2028. Four sites are conducting the recruitment of participants: One in Kansas (coordinating site), two in Florida, and one in Puerto Rico. All sites prescreen potential participants and refer candidates to the Kansas team, which completes the final screening, consents, conducts assessments, randomizes and provides coaching services within the CuidaTEXT intervention. Sites recruit from multiple sources including neurology and geriatric clinic visits, Electronic Health Records, community events, scientific and community presentations, social and ordinary media, research registries, community referrals, flyers, and word of mouth. A nation-wide advisory board with caregivers, researchers, healthcare staff and community-serving professionals meets quarterly via videocalls to guide recruitment efforts. For example, this advisory board has informed the addition of a texting option to hear more about the study and simplify the project’s contact survey.

Figure 2.

Flow diagram for recruitment and follow up in the CuidaTEXT Study

Inclusion criteria includes Spanish or English-speaking people who are over the age of 18, identify as Latino, and report providing hands-on care for a relative who has been given a clinical or research ADRD diagnosis and has an Ascertaining Dementia-8 (AD-8) screening score ≥2 [40, 41], indicating cognitive impairment. Participants also need to have a score of 7 or higher on the 10-item Center for Epidemiologic Studies Depression Scale (CESD-10) [42] to avoid floor effects in our primary outcome (depressive symptomatology). In our feasibility study, effect sizes of improvements in depressive symptomatology increased as we limited the sample to those with higher baseline CESD-10 levels [35]. Participants are not eligible if they participate in another wellbeing-related clinical trial, cohabit with another participant, provide care for a person with ADRD already cared for by another CuidaTEXT participant, or plan to move to another country within seven months. Participants must self-report being able to read and write and use a cell phone at least weekly for texting. All types of cellphones are eligible. In the rare event that we encounter a potential participant who does not own a cell phone, we provide one for their use in the study which will be returned to the study team at the end of their participation.

All study procedures have been approved by the Institutional Review Board of the [INSTITUTION] (STUDY00150585), and recruitment sites rely of that IRB approval. All caregiver participants give written digital or in-person informed consent, unless when participants are unable to access and complete these consents, in which case their consent signature is waived, and a verbal consent is obtained and documented. This study was registered in ClinicalTrials.gov under the ID NCT06728436.

2.2. Eligibility, screening, consent, assessment, randomization, enrollment, and retention.

The site recruitment teams provide potential participants with study information in their preferred language adapted for people with low literacy levels, either in person, at any of the recruitment sites, via phone, or videocall. After providing study information, they obtain a verbal consent and assess participants’ preliminary eligibility. Those preliminarily eligible for continuing screening are referred to the study operations team, which assigns the staff member who will complete the full eligibility assessment, signed consent, randomization, and baseline and follow-up assessments with that participant. Sites let potential participants know that they will receive a phone call from a Kansas phone number to reduce the risk of mistaking the call for spam. In line with methods to optimize retention [43], the team sets clear expectations for potential participants (e.g., length of assessments), explains the scientific principles behind trial methods (e.g., importance of retention irrespective of the group allocated to), helps diffuse ambivalence about participating (e.g., encourage to weigh pros and cons of participation). Randomized individuals are considered enrolled.

With regards to retention, study operations staff contact participants in both study arms to schedule an appointment two weeks before each assessment timepoint. They make multiple text, call, and email attempts as needed to schedule participants, and send reminder text messages two weeks before the timepoint and one day before the assessment day. Assessments are mostly remote but allow an in-person alternative as needed. During the eligibility screening, we collect participants’ and one to three relatives’ addresses, emails, and two telephone numbers to facilitate follow-up. Those who cannot be reached at follow-up are sent a certified letter to update the contact information. The team keeps flexible hours and allow assessments to be remote or in-person at the participants’ preferred location. We compensate participants with prepaid cards for baseline ($30), 3-month ($30), 6-month ($60), and 7-month assessments ($30). We waive the collection of social security numbers for compensation purposes to reduce the risk of stress.

2.3. Interventions

The intervention group receives CuidaTEXT immediately. This 6-month intervention has been described earlier [24, 35]. CuidaTEXT entails sending 1–3 automated daily messages covering various aspects such as logistical instructions, ADRD education, social support, self-care, ADRD care management, and behavioral symptoms. Additionally, participants can text keyword-based queries for immediate assistance on the above-mentioned topics and engage in live chat sessions with the coach for further guidance upon request. Coaches are bachelor’s degree level bilingual professionals with expertise in engaging the Latino community, with no or some background in ADRD research. Their responses are protocolized in a brief and simple manual to address these needs, which is also used for a 1-hour training before implementing the trial. For example, coaches will remind participants requesting more messages on a specific domain to use the keywords available as many times as necessary and visit specific websites or call the number of a national entity that can provide further help. Coaches will assist participants requesting help finding a specialist in their region using an online provider finder. Figures 3 and 4 show examples of CuidaTEXT’s interaction modalities and the domains, functions, size of automatic and keyword libraries, and available keywords for those functions.

Figure 3.

Examples of CuidaTEXT’s interaction modalities

Figure 4.

Domain, function, and size of daily automatic and keyword libraries in number of messages (n), and available keywords

The intervention is guided by the Stress Process Framework [38, 44], which poses that caregivers’ background, context, and stressors have negative health consequences. Ameliorating factors of stress can reduce those stressors and its negative consequences. CuidaTEXT’s goal is to reduce stress resulting from caregiving and its negative consequences by targeting known ameliorating factors of stress. CuidaTEXT’s delivery and content features were designed to engage stress ameliorating factors (e.g., coping strategies and preparedness for caregiving) while addressing common Latino context, background, and stress-related barriers as follows:

Delivery features:

These features were integrated to avoid adding stressors (e.g., commitment to in-person live interventions) and to reduce context and background barriers (e.g., transportation, time constraints).

• Time constraints. CuidaTEXT is asynchronous. Participants can use CuidaTEXT at any time. Caregivers receive only one message per day with core information to reduce burden. Caregivers can request additional on-demand messages as the program progresses based on their needs. Messages sent after office hours are either responded automatically if they are keywords or within the next two business days if these require a response from a coach.

• Transportation barriers. CuidaTEXT is remote. Participants can use the intervention anywhere.

• Technological divide. CuidaTEXT works via Short Message Service (SMS), which is integrated in all cellphones by default, is easy to use, and does not require a home broadband. Nearly 100% of Latino adults own a cellphone with texting capabilities [25, 26, 29, 45]. Focusing solely on apps, computers, websites, or videocalls, which are underused by Latinos, could perpetuate caregiver support access disparities [25, 26, 29, 45].

• Financial difficulties. Most of CuidaTEXT is automatic, relying little on workforce to reduce costs. Text message interventions for other conditions are more cost-effective than standard-of-care [31, 32].

• Language barriers. CuidaTEXT is available in English and US Spanish and does not use local slang or specialized jargon.

• Literacy limitations. CuidaTEXT was co-developed with the community to ensure text messages are easy to understand among Latinos [24]. During the development of the messages, we followed the Seven Principles of Communication: completeness, concreteness, courtesy, correctness, clarity, consideration, and conciseness [46].

• Cultural aspects. Receiving text messages on a private cellphone may address ADRD stigma [5, 33].

Content features:

CuidaTEXT’s five domains described earlier target ameliorating factors of stress (e.g., approach-coping and preparedness for caregiving). Content considers key context and background factors as follows:

• Cultural aspects: All domains include testimonials from our Latino caregiver advisory board using mock Latino names. Video demonstrations for skill-building include Latino individuals. These content features address cultural congruence preferences [47, 48]. Tips for self-care or ADRD care management include cultural values (e.g., Latin dancing and healthy Latin American recipes). Tips for coping with behavioral symptoms address common Latino stressors, including the care recipient missing their home country or traveling to their home country for holidays. Given the centrality of family among Latinos [49], CuidaTEXT includes substantial content on family support and encourage forwarding text messages to their relatives.

• Transportation, language, financial, and digital barriers. Messages include links to community resources that are bilingual, remote, free, or that provide financial aid (e.g., remote caregiver support groups, respite grants, discounts for medications).

We ensure fidelity in delivering CuidaTEXT by integrating research process phases proposed by the Treatment Fidelity Workgroup of the NIH Behavior Change Consortium.[50] These include: 1) standardizing most of the intervention delivery via daily automatic and keyword-driven text messages, 2) standardizing coach protocols, 3) the coach supervisor completing a checklist weekly to track the history of the coach’s text message interactions (response times, legibility, and adequacy to caregiver needs), and providing feedback , 4) preventing contamination by excluding participants from the same household, and 5) planning for implementation setbacks by following up with participants to identify and address technical issues. To explore the impact of differential text message interaction, we track the frequency of all interactions between CuidaTEXT and participants and the time it takes the coach to respond to each live chat message. We will assess correlations between these frequency/time variables and change in outcomes.

Participants allocated to the control group are offered the opportunity to start CuidaTEXT after their 7 Month follow-up assessment, to increase participation interest. A waitlist control group serves as a benchmark to assess the effects of the intervention against not receiving treatment during that same time-period. This solution is more ethical than no treatment and may increase recruitment and retention, as all participants will ultimately receive CuidaTEXT. Waitlist control groups are common in caregiver support research, allowing comparability [17] and control for non-specific treatment effects [51]. Therefore, control group participants participate in the waitlist for the first 7 months and, if interested, they join CuidaTEXT for an additional 6 months after they complete their 7 Month follow-up assessment (13 months total) without additional assessments besides the tracking of text interactions with CuidaTEXT. Conversely, participants in the CuidaTEXT treatment group only participate the 6 months of the intervention plus 1-month post-intervention (7 months total). We considered continuing to assess outcomes in both groups at 13 months, but this was beyond the scope of the study and would have added logistical complications. Secondary analyses comparing efficacy between groups at Month 7 (1-month post-CuidaTEXT treatment vs waitlist control) will elucidate whether the effects of CuidaTEXT are maintained after the intervention has ended, or whether the attention provided by the intervention drives improvements in outcomes.

2.4. Assessments

All surveys include scales for outcomes and targets. See Table 1 for a schedule of assessments and interventions. Outcomes include:

Table 1.

Timeline of assessments and interventions in the CuidaTEXT Study

	Screening	Baseline	Month 3	Month 6	Month 7	Post-7 months
Main efficacy outcomes a (AD-8, CESD-10)	X		X	X	X
Main efficacy outcomes b (FAQ, NPI-Q, COPE-28, PCS)		X	X	X	X
Secondary efficacy outcomes (financial inadequacy, positive affect, satisfaction with life, unmet needs)		X		X
Acceptability outcomes				X
Demographic information	X	X
Verification of eligibility	X
Adverse Events and Unanticipated Problems Collection	X	X	X	X	X
CuidaTEXT intervention (intervention group)		X	X	X
CuidaTEXT intervention discontinuation (intervention group)					X
Wait list (control group)		X	X	X	X
CuidaTEXT intervention (control group)						X

Caregiver depressive symptomatology.

The primary outcome is the change in caregiver depressive symptomatology. We use the CESD-10 [52]. The CESD-10 is a self-report rating scale that measures characteristic symptoms of depression in the past week (e.g. depression, loneliness). Each item is rated on a 4-point scale (Rarely or None of the Time-Most or All of the Time). The summary score ranges from 0 to 30 with higher scores indicating higher depressive symptom severity. This scale has shown strong psychometric properties among English and Spanish-speaking Latinos [42, 53].

Behavioral symptom distress.

The secondary outcome is the change in behavioral symptom distress. We use the Neuropsychiatric Inventory–Questionnaire (NPI-Q) [54, 55]. If any of 12 care recipient behavioral symptoms is present in the last month (e.g., repeating, delusions), caregivers rate their own distress on a 6-point scale (Not at All-Extreme). A distress summary score is calculated by adding the scores of all items. Higher scores indicate higher distress. The NPI-Q distress is responsive to change [56], and has shown strong psychometric properties among English and Spanish-speaking Latinos [53].

Targets include:

Approach-coping.

The primary target is change in approach-coping. We use the Coping Orientation to Problems Experienced Inventory (COPE-28) [57, 58]. The COPE-28 is a self-report questionnaire designed to measure ways to cope with caregiving. The scale is rated using a 4-point Likert scale (Not at All-a Lot). The approach-coping subscale uses 12 items corresponding with active coping, positive reframing, planning, acceptance, seeking emotional support, and seeking informational support [59]. High scores indicate caregiver participants use approach-coping techniques more often. The COPE-28 has an English [57] and a Spanish version [58], and the approach-coping subscale achieved an alpha coefficient of 0.8 in our feasibility study [35].

Preparedness for caregiving.

The secondary target is change in preparedness for caregiving. We use the Preparedness for Caregiving Scale (PCS) [60, 61]. The PCS is a self-rated instrument that consists of eight items that ask caregivers how well prepared they believe they are for multiple domains of caregiving. Responses are rated on a 5-point Likert scale (Not at All Prepared-Very Well Prepared). The total score is the mean of all items and ranges from 0 to 4, with higher scores indicating higher preparedness. The scale has an English [60] and a Spanish version [61] and had an alpha coefficient of 0.8 in our feasibility study [35].

We also collect background, context and stressor variables, adverse events and unanticipated problems.

Background, context, and stressor variables.

We collect background (age, sex, educational level, English proficiency, Latino subgroup [e.g., Mexican, Puerto Rican], ADRD subtype) and context variables (insurance status, rural setting, site) at baseline using English and Spanish items from reputable epidemiological surveys to assess background and context variables [62, 63]. We collect stressor variables at all timepoints using English and Spanish versions of the Functional Activities Questionnaire (FAQ; functional dependency) and the AD8 (cognitive status) [40, 41, 55, 64]. These variables are relevant for Latino health and ADRD research [65, 66].

Adverse events and Unanticipated problems.

We collect AEs and Serious AEs (SAEs) in this trial and monitor for unanticipated problems through participant complaints. In line with successful caregiver support interventions, we have developed Standard Operating Procedures (SOPs) for tracking and reporting AEs, SAEs and unanticipated problems [67]. We have also developed SOPs for emergencies, suicidality, homicidality, and abuse. Our study staff is trained in these SOPs and notifies the PI of any suspicious information. The PI classifies reported events as AEs or unanticipated problems.

2.5. Sample size and data analysis

Aim 1 evaluates change in caregiver depressive symptomatology (CESD-10) from baseline to Month 6. Six months is the end of treatment of the CuidaTEXT group and is a comparable timeframe to many caregiver support trials [18]. In our feasibility study, caregivers with CESD-10 ≥7 had a mean 5.3-point reduction in CESD-10 scores (SD=5.2) [35], compared to <0.7-point reduction in controls in other studies [28, 68, 69]. A 3.5-point reduction (30%) is considered clinically meaningful [70], and thus our target effect. A sample of 114 caregivers per group will provide 99% power (α=0.05) to detect a difference between groups using a two-sided independent samples t-test. Assuming 25% attrition, we will recruit 144 per group. Secondary analyses will compare changes in caregiver distress (NPI-Q). Our feasibility data showed a 7.4-point mean reduction (SD=8.3), compared to 0.2 in controls [71]; this exceeds the 3.1-point minimal clinically important difference [75], yielding ≥80% power with 114 per group. Additional analyses will explore change in these outcomes at different timepoints (e.g., baseline to three or seven months).

Primary analyses will use intent-to-treat, imputing missing change scores (including death) as 0. We will examine missing data patterns and apply multiple imputation if needed. Completer-only and transformed variable analyses will be conducted as secondary checks.

Aim 2 compares CuidaTEXT vs control in changes in COPE-18 approach-coping and PCS preparedness scores using Bonferroni-corrected t-tests (α=0.025). Feasibility data showed 3.54-point (SD=6.79) and 0.53-point (SD=0.45) increases, respectively [35]. Compared to literature-reported changes [72, 73], our sample size will provide 80% power to detect an effect size of 0.41 for coping and 99% power for preparedness (effect size=1.04).

Aim 3 assesses whether increases in coping and preparedness mediate intervention effects on depressive symptomatology and distress via joint significance tests (Bonferroni corrected α=0.013) [74]. Four mediation tests will be conducted: (1) coping change from baseline to Month 3 as mediator between treatment and depressive symptoms change from Month 3 to 6; (2) coping change as mediator between treatment and distress change; (3) preparedness change as mediator between treatment and depression changes; and (4) preparedness change as mediator between treatment and distress change. Each test was powered at ≥80% based on effect sizes from our feasibility data and those used in Aim 2 [35]. If both treatment effects on the mediator and mediator effects on outcomes are significant, we will consider the target a potential mediator. Structural equation models will follow Diaz et al. (2011) [75], modeling causal pathways based on time-sequenced changes. Power for the final structural models will also be ≥80%, supporting their ability to detect meaningful mediated effects.

3. Summary

3.1. Overview of the CuidaTEXT study

This RCT examines the efficacy of the first SMS texting intervention for ADRD informal caregiver support in reducing depressive symptomatology and caregiver distress among Latinos. Individuals are recruited in four US sites. The intervention is based on the Stress Process Framework and was co-created with Latino ADRD caregivers. The primary outcome includes depressive symptomatology change from baseline to 6 months. Secondary outcomes include changes in caregiver distress, preparedness for caregiving, approach coping. Mediation analyses explore these four outcomes with preparedness for caregiving and approach coping as targets.

3.2. Novel features of CuidaTEXT

This study will be the first to test the efficacy of a text message ADRD caregiver support intervention in any population and among Latinos in particular. The innovation is not the text message modality itself but its application to address Latinos’ multiple barriers to caregiver support. Existing intervention modalities (i.e. face-to-face, web/app-based) have low reach among Latinos and older age groups [1, 76]. Our key innovation will be a new model that delivers scalable, wider reaching, and more impactful interventions to all Latino caregivers of people with ADRD. Indeed, CuidaTEXT can deliver caregiver support readily available to nearly 100% of Latino adults via SMS text messages. Most caregiver support interventions have been developed for and tested with non-Latino Whites [7]. CuidaTEXT tailors its delivery and content features to the Latino community [6, 76]. Most caregiver interventions are hard to implement in real-world settings [1, 76]. CuidaTEXT’s heavy reliance on automated text messages and the use of simple protocols for coaches who do not need to be experts in ADRD makes it an innovative and sustainable modality that will increase intervention fidelity and will reduce workforce costs.

3.3. Limitations

While rare, a person might care for multiple relatives with ADRD. In these cases, we will record the number of care recipients and ask about the one with the most severe symptoms in measures such as the NPI-Q. We considered allowing family caregiver clusters, as Latino family members tend to share decision-making [49, 77]. However, we will not allow this option given Latinos’ low willingness to participate in clusters in our feasibility study [35]. We do not include long-distance caregivers, who comprise 15% of caregivers [78].

Newer technologies such as apps might be more popular among Latinos in the future. However, their use might also fade and never reach the level of accessibility of SMS text messages. Neglecting text messaging and focusing solely on newer technologies, all of which are less accessible among Latinos than SMS texting, can perpetuate already-existing disparities among Latino caregivers [29]. This project’s experience will inform future interventions using popular newer technologies such as Facebook and WhatsApp [79], and language processing tools that may further reduce reliance on coaches.

4. Conclusions.

This manuscript outlines a protocol for testing the efficacy of a text message intervention for informal Latino ADRD caregivers. This work is a continuation of more than five years of co-creation and feasibility testing of this SMS texting intervention with and for Latino caregivers of people with ADRD. Besides testing its efficacy, we will test mechanisms of action to inform how to optimize the intervention’s potency. This work is the first of its kind to capitalize on the high accessibility of text messaging among Latinos to deliver culturally-tailored ADRD caregiver support to this underserved and under-resourced population. Findings have the potential to make clinical and policy-relevant contributions by providing appropriate caregiver support to Latinos in a highly-scalable way [80].