“Caminando y Socializando con HOLA:” Results of a Randomized Clinical Trial to Promote Health and Prevent Depression and Anxiety in Older Latinos

Abstract

Objective:

This study sought to evaluate the feasibility, acceptability, and preliminary efficacy of the Happy Older Latinos are Active (HOLA) health promotion intervention in a group of older Latinos who were at-risk for developing Major Depressive Disorder or Generalized Anxiety Disorder.

Method:

Sixty older Latinos age 60+ with subthreshold depression or anxiety were randomized to receive either: HOLA (n=30) or enhanced psychoeducation through fotonovela control (n=30). The primary outcomes of interest were feasibility, acceptability, and reduction in depression and anxiety symptom severity. Outcome measures were administered at baseline and at the end of the intervention.

Results:

Within a year, the enrollment target was met with <5% of eligible participants refusing randomization. The randomization scheme produced equal numbers of participants randomized to each condition. Four participants (6.7%; HOLA = 1; Control = 3) were lost to follow-up; 69% of the HOLA sessions were attended; and participants reported high satisfaction with the intervention. Finally, compared to control, a significant proportion of participants in HOLA experienced a clinically significant reduction in their anxiety symptoms (60% vs. 26.7%).

Conclusions:

Findings highlight the feasibility, acceptability, and significant impact the HOLA intervention can have in reducing psychological distress since it is responsive, respectful, and specific to the needs of older Latinos. Furthermore, using a community health worker to deliver a health promotion intervention to prevent common mental disorders in older Latinos is an innovative approach for reducing disease burden in a population living with high disparities in accessing and engaging in mental health services.

Introduction

Latinos are the fastest growing segment of the older adult population in the United States (U.S.). In 2020, 4.6 million adults in the U.S. age 65 or older identified as Hispanic or Latino, comprising approximately 9% of the older adult population (U.S. Census, 2020). This number is projected to increase nearly five-fold by 2060 when an estimated 21% of older adults will identify as Latino (U.S. Census, 2020). Historically, national surveys have rarely differentiated Latino adults by age, creating a misleading narrative that Latino adults, in aggregate, are at lower risk of all lifetime psychiatric disorders compared to non-Latino Whites (Alegria et al., 2008; Breslau et al., 2006; Kessler et al., 2003). However, when age differences are considered, significant nuances in the prevalence of common mental disorders (depression and anxiety) begin to emerge between older and younger Latinos (Jimenez et al., 2020). A comprehensive understanding of the prevalence of common mental disorders, the cultural and contextual realities of older Latinos, and the pre-existing social inequities and stigma that impact help seeking is crucial for creating effective interventions to improve the mental health of this population.

Older Latinos experience psychological distress at comparable or higher rates than both their age-matched peers and younger Latino counterparts. Jimenez et al. (Jimenez et al., 2010) found that the prevalence of depression and anxiety in older Latinos (aged 60 years and older) was equal to, and in some cases greater, compared to older non-Latino Whites. Although Latinos and non-Latino Whites had similar lifetime prevalence rates of any depressive disorder (16.4% vs. 12.2%) and of any anxiety disorder (15.3% vs. 13.5%; older Latinos demonstrated significantly higher 12-month prevalence rates of any depressive disorder (8.0% vs. 3.2%) and major depressive episodes (7.3% vs. 2.9%) compared to their non-White peers (Jimenez et al., 2010).

Despite the need for mental health services, older Latinos are not seeking these services at the same rate as their non-Latino White counterparts. In a study measuring mental health care utilization among older Latinos and non-Latino Whites across the U.S., fewer Latinos (24%) with elevated psychological distress initiated mental health care compared with non-Latinos (35%) (Jimenez et al., 2013). Furthermore, even when older Latinos seek mental health treatment; they are less likely to receive adequate care and tend to drop out of treatment two to three times more frequently than non-Latino Whites (Arnow et al., 2007; Jimenez et al., 2013).

Common mental disorders remain as one of the highest health burdens for older Latinos due to persistent disparities in mental health care services compared to other health care service domains (Dankwa-Mullan et al., 2010; Jimenez et al., 2010, 2013). Although challenges to equitable mental health care access affect older adults regardless of race or ethnicity, they may be even more pronounced for older Latinos who tend to have more stigmatized attitudes towards mental health disorders compared to White communities, often stemming from historical and culturally based representations of mental health disorders as a sign of personal weakness. (Jimenez et al., 2013). Moreover, systemic factors such as redlining, transportation challenges, language barriers, and financial obstacles, contribute to persistent disparities in the ability of older Latinos to access high-quality mental health care services (Jimenez, Park, et al., 2022).

Mental health treatments may not match the experiences, values, and beliefs of older Latinos which can further complicate older Latinos’ willingness to seek and engage with psychological care (Jimenez et al., 2012). Older Latinos often express psychological distress as physical ailments (malaise, pain, headaches, fatigue, gastrointestinal distress) rather than by more traditional symptoms, such as sadness or loss of interest (Ruiz, 1997). This somatization of symptoms leads many older Latinos to seek treatment from their primary medical provider rather than from a psychotherapist out of hesitancy that therapy would not address the symptoms they are experiencing. Furthermore, cultural values influence beliefs on the causes of mental illness that ultimately shape the type of care pursued. Older Latinos often conceptualize mental illness as being directly caused by the loss of family and friends, family issues, and even moving to a different place. Migration and relocation involve a series of stressful experiences that can shape individuals' perceptions of reality and influence the family system and generations after, particularly as it relates to psychological well-being (Jimenez et al., 2012). The disruption of the social support network by immigration to the U.S. is traumatic in that it is socially isolating, facilitating a detrimental loss of belonging, and often contributes to poorer health outcomes (Jimenez et al., 2012).

In addition to psychological morbidity, older Latinos experience a disproportionate burden of physical health risk factors compared to their age-matched peers. They are 1.2 times more likely to be obese and 1.5 times more likely to have diabetes than non-Latino Whites (Chow et al., 2012; McGrath et al., 2019). Similarly, older Latinos are at increased vulnerability for developing metabolic syndrome (MetS) which affects nearly 50% of adults 60 years and older and occurs earlier and more commonly among Latinos compared with other racial and ethnic groups (Aguilar et al., 2015). Furthermore, older Latinos are also more likely to be sedentary compared to their non-Latino White counterparts (August & Sorkin, 2011). Taken together, disparities in mental and physical health put older Latinos at greater risk of having poorer overall health than non-Latino Whites.

Preventing psychological distress is an essential component of improving quality of life among older Latinos. Indicated prevention – the process of focusing on individuals exhibiting subthreshold symptoms of a given disorder – is the most efficient way to address prevention of common mental disorders (Smit et al., 2006). Individuals with depressive or anxiety symptoms that do not meet criteria for major depressive disorder (MDD) or generalized anxiety disorder (GAD; i.e., subclinical conditions) are at a greater risk of developing the respective disorders within two years (Karsten et al., 2011). Although not as severe as the clinical disorders themselves, subclinical depression and anxiety are debilitating and are associated with psychological suffering, impairment in social, family, and occupational functioning, poor perceived emotional and physical health, and diminished quality of life (Judd et al., 2002; Porensky et al., 2009). Milder or subclinical states increase the risk of developing the full clinical disorder; however, these subclinical states are more reversible than their advanced counterparts and may be associated with less neurobiological changes at an earlier (and more modifiable) stage of development (Karsten et al., 2011).

The U.S. mental healthcare infrastructure is currently ill-equipped to support the unique needs of older Latinos; therefore, the need for effective and scalable prevention models is imperative (Institute of Medicine (US) Committee on the Future Health Care Workforce for Older Americans, 2008). The ability of community health workers (CHWs) to deliver simple behavioral interventions to prevent and treat common mental disorders may increase the scalability and impact of feasible, culturally competent interventions among older Latinos. CHWs are lay community members who work almost exclusively in community settings to promote health and prevent diseases among groups that have traditionally lacked consistent access to quality care (Barnett et al., 2018). CHWs are effective, in part, because they are ethnically, sociodemographically, and experientially connected to the communities from which they serve. They possess an intimate understanding of community social networks and health needs; communicate in a similar language; recognize and incorporate culture to promote health; and empower community members to identify their own needs and implement their own solutions in ways that improve personal and community self-efficacy (Perez et al., 2006; Waitzkin et al., 2011). Interventions that employ a CHW model have been successfully implemented globally to support marginalized groups with complex medical conditions such as HIV, palliative care, and chronic disease management (Dias et al., 2019; Perez et al., 2006; Waitzkin et al., 2011). CHWs have the potential to reduce or eliminate health disparities and are an important tool to facilitate task shifting to enable efficient utilization of scarce mental health resources (Barnett et al., 2018).

Older Latinos’ conceptualizations of mental illness as physical illness – combined with low rates of mental health service use – highlight the need for mental health services to include strategies that enhance general well-being and bring mental health benefits. Although antidepressant medications are the most widely used modality for treating common mental disorders, their use in older adults is more likely to be associated with adverse effects (Fournier et al., 2010). Older Latinos are less likely to find antidepressant medication acceptable than non-Latino Whites (Givens et al., 2007); thus making psychosocial strategies a potentially more suitable intervention for this group. Additionally, there is growing evidence to suggest that health promotion interventions – defined as behavioral interventions that use counseling strategies to equip participants with the necessary knowledge and skills to modify and sustain a healthy diet, increased physical activity, and/or healthy weight – can lead to improved mental and physical health outcomes (Jimenez, Begley, et al., 2015; Jimenez et al., 2018). Increased physical activity effectively reduces symptoms of depression and anxiety in older adults (Belvederi Murri et al., 2019; Blumenthal et al., 2007) and increasing pleasant events has been shown to be an effective intervention for geriatric depression (Scogin et al., 2016). Thus, there may be a synergy between increased physical activity and pleasant events to target psychological distress in older adults.

In this context, Happy Older Latinos are Active (HOLA), a CHW-led, multicomponent, health promotion intervention seems particularly promising. HOLA incorporates the principles of Social Learning Theory (SLT; Bandura, Albert, 1977) and Behavioral Activation (BA; Lewinsohn, Peter M., 1974) to engage older Latinos in a group-based, physical activity routine and pleasant events scheduling to reduce the risk for major depression and generalized anxiety disorder, prevent the incidence and recurrence of these disorders, and improve health-related outcomes. The current study reports the outcomes from a pilot randomized clinical trial in which HOLA was compared to an enhanced psychoeducation through fotonovela control condition to determine the feasibility, acceptability, and preliminary efficacy in a sample of older Latinos with subclinical depression or anxiety.

Methods

The study was approved by the Internal Review Board (IRB) of the University of Miami (IRB ID: 20140607). All participants provided informed consent for participation in the study. A total of 60 older Latinos age 60+ with subthreshold depression or anxiety – defined as a score ≥ 5 on the PHQ-9 (Kroenke et al., 2001) OR GAD-7 (Spitzer et al., 2006) – were randomized to either: HOLA (n=30) or the fotonovela control condition (n=30). No upper limits were selected since we were administering the Mini International Neuropsychiatric Inventory (MINI) (Sheehan et al., 1998) to rule out Major Depressive Disorder (MDD) and Generalized Anxiety Disorder (GAD) and did not want to narrowly restrict our sample; hinder our recruitment efforts; and limit generalizability. In addition, other prevention trials (Dias et al., 2019; Reynolds et al., 2014) have had minimum cutoffs for their depression screens but not upper limits. Recruitment efforts were led by the CHW who recruited participants from the community via advertising, outreach to agencies serving older adults, snowballing techniques, and targeting high volume areas (e.g., churches and health fairs) via active recruitment. A CONSORT diagram (Schulz et al., 2011) illustrating participant flow throughout the study is presented in Figure 1. A detailed description of the study methods is provided elsewhere (Jimenez, Reynolds, et al., 2015).

Figure 1:

HOLA CONSORT Diagram

Participants

A two-step screening process was used to identify potential participants. First, participants were screened for depression, anxiety, and cognitive impairment using the PHQ-9 (Kroenke et al., 2001), GAD-7 (Spitzer et al., 2006), and Mini Mental Status Exam (MMSE; Folstein et al., 1975). Second, participants were screened for psychiatric diagnosis and walking ability using the MINI (Sheehan et al., 1998), and the 10-meter walk test, which has been used in lieu of a stress test in physical activity interventions with older adults (Peters et al., 2013). Participants were included if they self-identified as Latino; were 60 years or older; screened positive for anxiety or depression; did not meet criteria for current MDD or GAD; received medical clearance from a physician to participate in the intervention; expected to stay in South Florida for the next 12 months; and volunteered informed consent. Participants were excluded if they were currently receiving antidepressant medication or participating in other mental health treatment such as psychotherapy; had a diagnosis of any neurodegenerative disorder or dementia (e.g., Parkinson’s disease, Alzheimer’s, vascular, frontotemporal dementia, etc.) or significant cognitive impairment as indicated by a MMSE (Folstein et al., 1975) score <24; were unable to complete the 10-meter walk test; or had an acute or severe medical illness that precluded them from safely participating in a health promotion intervention.

Procedures

Baseline data were collected by interview following the second screening. Trained research assistants, who were blind to group assignment conducted all assessments in private offices at the University of Miami. Upon completion of the baseline assessment, participants were randomly assigned to one of the two treatment conditions. A biostatistician who was not present at the time of assessment conducted randomization using an automated program embodying the strategies of permuted block randomization. Participants completed one follow-up assessment after the interventions ended (post-intervention follow-up) and were paid $25 upon completion of assessments at baseline and the end of the intervention for a total of $50. consisting of the same measures that were obtained at baseline.

Treatment Conditions

Happy Older Latinos are Active (HOLA)

HOLA employs pleasant event scheduling via a BA (Lewinsohn, Peter M., 1974) approach to engage participants in a physical activity routine. Constructs of observational learning, reinforcement, and enhanced self-efficacy were adopted to further complement this process (Bandura, Albert, 1977). See Figure 2. The relationship between the participants and the CHW capitalizes on the personal relationship to motivate, model, and maintain health behavior change. The CHW fosters a collaborative and supportive environment where group members feel a shared responsibility for success. This sense of community motivates individuals to actively participate and invest in their own growth through the intervention.

Figure 2.

Conceptual Framework

HOLA is a 16-week, multicomponent health promotion intervention. The first component consists of two manualized social and physical activation sessions. Prior to beginning the group walk phase of the program, each participant meets individually with the CHW (DP-J) to: a) receive printed psychoeducational material regarding depression and anxiety from the National Institute of Mental Health; b) collaboratively engage in goal setting; and c) proactively problem solve to identify and overcome potential obstacles that may interfere with meeting the demands of the intervention. Half-way through the program during week 8, participants meet individually with the CHW (DP-J) again for 30 minutes to discuss progress of physical and social activity goals. HOLA’s second component consists of a moderately intense group walks led by the CHW (DP-J). These group walks are held 3 times a week, for 16 weeks, and last 45 minutes. Each walk begins with 10 minutes of stretching and warm up, 25 minutes of active walking, and 10 minutes of cool down. During the cool down phase of each walking session, the CHW (DP-J) asks each participant to identify a pleasant event that they intended to do with another person before the next meeting. This provides a means to generalize the intervention into participants’ everyday lives and relationships. For this study, walks were conducted with a group of 5 bilingual and monolingual Spanish speaking participants at a centrally located public park, which is owned and operated by Miami-Dade County.

Enhanced psychoeducation through fotonovela

The selection of the fotonovela as a control condition, as opposed to usual care, was deliberate due to its inherent health-related nature and relevance to older Latinos with poorer psychological functioning and potentially low literacy levels. Fotonovelas are informational booklets utilizing staged photographs and succinct text bubbles to depict soap opera-style narratives conveying educational messages. Distinguishing themselves from conventional educational materials, fotonovelas uniquely incorporate popular imagery, cultural norms, straightforward text, compelling stories, and vivid visuals to raise awareness, promote health, and challenge stigma by fostering patient awareness. They have been shown to be more effective in reducing stigma and increasing knowledge concerning symptom identification and treatment options when compared to NIMH educational materials among depressed Latinos (Cabassa et al., 2015). The fotonovela has not demonstrated efficacy in increasing mental health service utilization nor has it been utilized as a treatment or preventive tool making it an excellent control condtion (Cabassa et al., 2015).

Participants assigned to the enhanced psychoeducation control condition received a copy of the theoretically informed and empirically grounded fotonovela, crafted by Cabassa and colleagues (2015). Additionally, those in the control group convened one month after receiving the fotonovela for an open forum. During this forum, participants expressed their opinions regarding the fotonovela and shared insights gained from reading the material. These hour-long discussion groups were facilitated by a clinical psychologist and consisted of six control participants each.

Cultural tailoring

Culturally tailoring of the intervention has occurred in several ways. First, all intervention and assessment materials are available in English and Spanish. Second, all recruitment and intervention materials are tailored in terms of graphics, language, and content to meet the needs of older Latinos. Since many older Latinos have low health literacy (Soto Mas et al., 2017), the social and physical and activation manual includes visual aids and includes Latino relevant metaphors, values, and cultural proverbs (dichos). Third, HOLA’s recruitment strategies focus on relationship building. Among Latinos, there is fear and mistrust in participating in clinical research, which can lead to stigmatizing attitudes (Arredondo et al., 1996; Calzada et al., 2010; Marin & Marin, 1991). Therefore, in order to build this trust, we establish collaborative relationships with community agencies when recruiting for HOLA. The Latino cultural value of personalismo (the cultural value that emphasizes social relationships and the mutual respect that those in the relationship have for one another) plays a key role in Latinos' decision to enroll in clinical research (Marin & Marin, 1991). Moreover, establishing relationships with agencies that serve Latinos may inspire confianza, or trust and familiarity, with Latino families, who traditionally have tended to be wary of professionals whom they do not know (Calzada et al., 2010). Community agencies often spend considerable amounts of time building confianza with those they serve, and when researchers establish a working relationship with these agencies, they may also establish indirect confianza with those in the Latino community (i.e., “I trust those whom my friends trust”). Fourth, is the use of CHWs. Since CHWs possess an intimate understanding of the community, they understand cultural values and norms and know how to incorporate them to promote health and health outcomes. For example, the CHW understands the importance of face-to-face contact, thus being a visible presence was essential in our recruitment efforts. This allowed us to shed the “ivory tower” reputation of academia; overcome the stigma associated with a potentially sensitive topic; and foster trust and enthusiasm for the project.

Outcomes

The primary study outcomes were feasibility and acceptability as measured by recruitment, retention, and attendance. Successful recruitment was defined as meeting 100% of targeted randomization (N = 60), with 20% or less of eligible participants refusing randomization. Adequate retention was defined as 85% or more of randomized participants completing the post-intervention assessment. Acceptability was defined as 80% or more of sessions attended by participants. This approach to analysis of feasibility and acceptability data mirrors the structure of previous pilot prevention trials (Reynolds et al., 2018; Stahl et al., 2020).

Consistent with recommendations from biostatistical workgroups funded by NIH (Kraemer et al., 2006), this pilot study was not powered to test a hypothesis. However, we sought to determine the preliminary efficacy of HOLA by identifying the effect, if any, HOLA had on depression and anxiety severity.

Measures

Participant satisfaction was measured using the 8-item Client Satisfaction Questionnaire-8 (Larsen et al., 1979)(CSQ-8; alpha = .91 ). The measure uses four response choices, where “1” indicates the lowest degree of satisfaction and “4,” the highest. Scores for each item are summed. Total scores range from 8 to 32, with the higher number indicating greater satisfaction. In addition, there are two open-ended questions that allow participants to state in their own words what they liked most about the intervention and what they liked the least.

The Quick Inventory of Depressive Symptoms – Self Report (QIDS-SR; Rush et al., 2003) was used to assess depression symptom severity. It is a comprehensive measure of depressive symptomatology that includes core, melancholic, and atypical depressive items, has high validity and reliability, is sensitive to change, and allows for the ascertainment of self-report measures of depression severity. It has high internal consistency and is highly correlated with the Hamilton Rating Scale for Depression-17 (Hamilton, 1967). It has been used in adults with minor depression (Judd et al., 2002) and translated for use with Latinos (Bernstein et al., 2007; González et al., 2010).

The Beck Anxiety Inventory (BAI; Beck et al., 1988) was used to assess anxiety symptom severity. The 21-item instrument measures primarily physiological symptoms of anxiety and is an effective assessment in measuring clinically relevant, subthreshold anxiety (Karsten et al., 2011). It has been validated on older adults (Wetherell & Gatz, 2005) and has been translated for use with Latinos (Osman et al., 1997).

Statistical Analyses

All analyses were performed with the intent-to-treat principle so that comparisons were made according to the assigned intervention groups. Group differences on sociodemographic variables were tested using t-tests for continuous variables and chi-square analyses for categorical variables. Although the pilot study was not powered to test a hypothesis, we sought to identify to what effect, if any, HOLA had on depression and anxiety severity. A two- (group) by two- (time) way ANOVA with repeated measures was used to evaluate main treatment effects and interactions. In addition to statistical significance, we were interested in examining if participants had shown a clinically significant decrease in their depression and anxiety symptoms. To do this, we analyzed the data using a distribution-based approach whereby clinically important differences on the QIDS-SR and BAI was determined by setting a threshold as a reduction of one-half the standard deviation of the scale’s score at baseline (Harvey et al., 2017; Norman et al., 2003). Based on this classification, we used chi-square analyses to compare the proportion of participants in HOLA who reported a clinically significant decrease in their symptoms compared to those in the fotonovela condition.

Responses to the two open-ended questions on the CSQ-8 (Larsen et al., 1979) that allow participants to state in their own words what they liked most about the intervention and what they liked the least were transcribed and analyzed thematically (Whitley & Crawford, 2005). Researchers independently identified themes in the transcripts before collaboratively developing a coding system with descriptive labels (Miles & Huberman, 1994). To minimize bias, two coders independently reviewed the responses before reconciling their coding schemes and reaching consensus on the prevalence of themes (Whitley & Crawford, 2005).

Results

Descriptive characteristics

Table 1 shows baseline sociodemographic characteristics of the participants who were randomized to either HOLA or the fotonovela control. We did not find baseline differences in any of the outcome measures.

Table 1.

Sociodemographic Characteristics

		HOLA (n=30)	Fotonovela (n=30)
		M (SD)	M (SD)	F
Age		72.6 (6.9)	70.6 (7.2)	.05
Years in the US		39.2 (22.2)	31.2 (21.6)	.01
Years of Education		12.1 (3.3)	12.8 (2.9)	.29
PHQ-9		9.5 (5.7)	8.9 (4.3)	.11
GAD-7		6.4 (4.5)	7.3 (4.6)	.27
		% (n)	% (n)	χ2
Sex				.11
	Men	17 (5)	20 (6)
	Women	83 (25)	80 (24)
Marital Status				6.6
	Single/never married	3.3 (1)	20 (6)
	Married	23.3 (7)	33.3 (10)
	Widowed	26.7 (8)	20 (6)
	Divorced	43.3 (13)	26.7 (8)
	Separated	3.3 (1)	0 (0)
Country of Origin				11.8
	Cuba	36.7 (11)	63.3 (19)
	Colombia	16.7 (5)	13.3 (4)
	Puerto Rico	13.3 (4)	6.7 (2)
	United Staes (mainland)	3.3 (1)	0 (0)
	Venezuela	0 (0)	6.7 (2)
	Peru	10 (3)	3.3 (1)
	Nicaragua	3.3 (1)	3.3 (1)
	Mexico	3.3 (1)	0 (0)
	Guatemala	3.3 (1)	0 (0)
	Ecuador	6.7 (2)	0 (0)
	Dominican Republic	3.3 (1)	3.3 (1)

Note: M: mean; SD: standard deviation

Feasibility and acceptability

In a year, we met our enrollment target with 5% (n=3) of eligible participants refusing randomization. The randomization scheme produced equal numbers of participants randomized to each condition. Only 4 participants (6.7%; HOLA = 1; Control = 3) were lost to follow-up. Participants in the HOLA condition attended over 69% of the walking groups (M=33; SD = 12.9) and had a mean score 29.6 (SD=3.3) on the CSQ-8, indicating a high degree of satisfaction with the intervention.

From the participants’ responses to the two short-answer questions on the CSQ-8 (Larsen et al., 1979), we have learned several key lessons. First, the CHW and the group format are essential. Second, participants really liked the focus on health promotion and recognized that physical and mental health are linked. Third, splitting up HOLA into its multiple components would not work suggesting that the whole is greater than the sum of its parts. Participants greatly appreciated the individual physical and social activation sessions. They liked the information that was presented, the ground rules, goal setting, and problem solving. They enjoyed the walking component, and the pleasant events scheduling done during the cool down phase of each walk. Fourth, participants wanted a maintenance phase added to the intervention. They stated that by building in regularly scheduled booster sessions, it would help keep them motivated.

Preliminary intervention effects on depression and anxiety symptoms

Table 2 includes results of the repeated measures ANOVA illustrating the effect of treatment on depression and anxiety symptom measures. There was no significant main effect of treatment on depression symptoms F(1,58) = .62, p = .44. Regarding anxiety symptoms, a significant main effect was found F(1,58) = 12.2, p < .01. There was also a significant interaction effect between time and treatment condition on anxiety symptoms, F(1,58) = 5.0, p = .03.

Table 2.

Intervention Effects

	HOLA (n=30)		Fotonovela (n=30)		F	p	η2
	Baseline M (SD)	Post- intervention M (SD)	Baseline M (SD)	Post- intervention M (SD)
QIDS-SR	6.6 (4.3)	5.8 (4.2)	6.3 (4.2)	6.1 (3.9)	.62	.44	.01
Interactiona					.30	.59	.01
BAI	14.1 (10.5)	8.7 (10.4	12.0 (12.3)	10.8 (10.8)	12.2	<.01	.17
Interactiona					5.0	.03	.08

^aInteraction is group x time.

We examined the data using a distribution-based approach whereby clinically important differences in each measure were determined by setting a threshold as a reduction of one-half the standard deviation of the scale’s score at baseline. This approach has been commonly used to identify clinically important differences and analyze the distribution of treatment effects (Harvey et al., 2017; Leifker et al., 2010; Norman et al., 2003). Results indicated that a significant proportion of participants in HOLA (53.3%) experienced a clinically significant reduction in their anxiety symptoms compared to participants in the fotonovela condition (20%). See Table 3.

Table 3.

Proportion of Participants Who Reported a Clinically Significant Decrease in Depression and Anxiety Symptoms

	HOLA (n=30)	Fotonovela (n=30)	χ2	p
	% (n)	% (n)
QIDS-SR	60 (18)	26.7 (8)	2.9	.09
BAI	53.3 (16)	20 (6)	3.7	.05

Discussion

The HOLA intervention is a highly innovative health promotion intervention that is uniquely tailored to prevent incidence and recurrence of depression and anxiety among older Latinos at increased risk for poor psychological functioning. Our pilot study showed that a health promotion intervention that matches the experiences, values, and beliefs of older Latinos was feasible and acceptable. Notable strengths include high retention of study participants and the utilization of CHWs to improve mental health and facilitate sustained engagement in the program. These positive feasibility and acceptability metrics further suggest that HOLA is well-positioned to be scaled up and out to older Latinos across South Florida.

Prevention of common mental disorders (i.e., depression and anxiety) in older Latinos is a key factor in addressing mental health disparities in a group that views traditional mental health services as highly stigmatizing. Therefore, the fact that 53.3% and 60% of participants in HOLA exhibited clinically significant reductions in anxiety and depression respectively is quite encouraging. Specifically, improvements in anxiety symptoms were significantly greater for participants in the HOLA condition compared to their peers in the fotonovela control condition while improvements in depression symptoms trended towards significance for older Latinos in the HOLA condition compared to those in the fotonovela group. These findings are consistent with prior pilot data suggesting that older Latinos who fully engage with the HOLA intervention experience improved psychological functioning (Jimenez et al., 2018; Jimenez, Weinstein, et al., 2022). HOLA centers on behaviorally activating older Latinos in ways that bolster self-efficacy and foster connection between peers, which in turn, help to ameliorate symptoms consistent with depression and anxiety. Engaging in this minimally resourced intervention allows older Latinos to strengthen and expand their social networks in ways that get them active both in the moment (i.e., communal walks) and in the future (e.g., activity scheduling) and gain confidence in building trust within a new community on which they can rely. It is these new meaningful connections that most likely contribute to older Latinos reporting reduced anxiety and depression. Future studies should continue to develop integrated prevention efforts that center on building connections, increasing physical activity, and behaviorally activating older Latinos, a multiple marginalized population with limited resources and experience significant mental and physical health disparities.

In addition to objectively improving psychological functioning, the HOLA intervention was also highly feasible and acceptable in a small sample of older Latinos living in South Florida. Older Latinos were excited about participating in HOLA from the start, as evidenced by authors meeting 100% of their targeted recruitment (N = 60). Additionally, overall retention in the pilot study was excellent with 93.8% of enrolled participants completing post-intervention assessments (96.7% of participants randomized to HOLA and 90.0% of participants randomized to HOLA). When considering these feasibility metrics together – smooth recruitment, minimal attrition, and positive participant feedback – it becomes increasingly clear that older Latinos were not only interested in the intervention, but that the intervention has clear potential for future scale up to other older Latinos.

Several factors may have contributed to older Latinos’ acceptability and interest in the HOLA program. First, the intervention’s facilitation of social connectedness may have increased interest in and acceptability of the program by giving participants a group from which they were accountable and from whom they could draw support. In offering advice to their peers on activity scheduling and health promotion, participants developed their own self-efficacy in actively applying the intervention’s skills (Jimenez et al., 2018). As one participant noted, it was the support she was able to lend and at times receive that made the difference: “To me, it was the sharing. If I was having a bad day or week, I knew that I could come to the group and tell them what was going on and get their support. Other times, I could just listen to others and offer my advice from my own experience.”

Similarly, the program’s emphasis on health promotion helped older Latinos gain insight into the links between physical and mental health, which, in turn, may have increased participant’s interest in engaging fully with the intervention. As one participant stated, “I have noticed that when I don’t feel well physically, then my mind is not right, but in this program, we walk, we sweat, and it makes me feel good. It makes me feel like I have accomplished something.”

Finally, the CHW (DP-J) motivated participants and facilitated a warm and affirming environment that was uniquely tailored to the cultural needs of older Latinos. Knowing that they had to meet with the CHW (DP-J) at certain intervals, participants were motivated to maintain the gains that they had achieved. One participant stated, “My doctor told me that I needed to walk more, but I never did it. The support from Doris [the CHW] and the interacting with others is what keeps me going.” The CHW was an important conduit of health information because of her ability to deliver the intervention not only in the participants’ preferred language, but also with the cultural intricacies that only a member of the community could truly possess (Waitzkin et al., 2011). Participants may have felt more comfortable divulging personal information to someone from their own community, allowing them to communicate with study personnel with more comfort. This fostered trust and enthusiasm for the project which may have facilitated the low dropout rates and increased acceptability of the program.

Although older Latinos expressed significant interest in the intervention, identifying strategies that counteract potential issues to participation is a necessary step to successfully scaling HOLA “up and out.” Structural barriers related to changes in employment and transportation hurdles affected participants’ consistent participation throughout the study period. Similarly, health challenges (e.g., sickness, doctors’ appointments, etc.) sometimes prevented older Latinos enrolled in the pilot study from attending weekly walk sessions or engaging fully with behavioral activation in between sessions. Authors are currently exploring ways to implement these lessons learned in current iterations of HOLA and encourage other researchers engaged with intervention development for older adults to further consider how structural barriers to care may affect older adults’ willingness and ability to fully participate in health promotion interventions.

This randomized prevention trial has some limitations. First, it targeted older Latinos living in South Florida so findings may not be generalizable to older Latinos living elsewhere. However, older Latinos enrolled in this study possessed significant diversity as it relates to race, country of origin, immigration status, and years living in the U.S. which may increase the generalizability of findings to other diverse groups of older Latinos across the nation. Second, due to the small sample size, this study was underpowered to detect the presence of large intervention effects. Despite this, participants in the HOLA condition experienced clinically significant reductions in anxiety symptoms and improvements in depression symptoms that trended towards significance. Third, the pre-post study design limited our ability to observe if improvements in depression and anxiety symptoms were sustained over time. This is important given the preventive nature of the work. Lastly, this study did not collect data on perceived or actual social support and participants’ physical activity levels which might have influenced the findings presented here. To improve upon these limitations, authors are currently conducting an adequately powered, randomized controlled trial examining the effectiveness of HOLA (NCT03870360). In this larger study, we are assessing physical activity levels and social support as potential mediators of the HOLA intervention, and we are following participants for two years to evaluate the sustainability of the intervention's effects.

Despite these limitations, this study provides preliminary evidence for this intervention to be scaled towards other high-risk older people of color that experience significant exposure to disease burden and mental health stigma. Other underserved and racially and ethnically minoritized groups may be likely to benefit from this health promotion intervention in the service of preventing disabling mental health disorders. Integrating HOLA with a selective prevention approach may also show promise in more medically complex groups such as older Latinos with HIV (OLWH). Guided by a selective prevention model, a pilot trial was recently conducted to test the feasibility and acceptability of HOLA to prevent cognitive decline with a group of OLWH (Jimenez et al., 2024). This pilot study yielded promising results indicating that OLWH who engaged with the intervention demonstrated significant reductions in depression and anxiety symptoms as well as improvements in several neuropsychological domains (e.g., delayed recall, executive functioning, verbal fluency) from baseline to post-intervention (Weinstein et al., under review). Future research should prioritize conducting randomized control trials to better ascertain the extent to which HOLA may stave off psychological, neurocognitive, and cardiometabolic morbidity among older Latinos with complex medical conditions.

Conclusions

HOLA is a highly innovative, culturally tailored health promotion intervention that used a CHW to prevent anxiety and depression in older Latinos via increasing physical activity and social support. Results of this randomized prevention trial demonstrated the feasibility and acceptability of HOLA and provided preliminary evidence of the intervention’s efficacy in a sample of older Latinos with subthreshold anxiety and depression. These results warrant further exploration of HOLA as a scalable, sustainable, and effective prevention intervention and highlight the considerable public health potential of applying a health promotion intervention to address significant disparities in mental healthcare access and utilization within the older Latino community.

Public Health Significance.

Given the prevalence and morbidity of depression and anxiety in older Latinos, the inadequacies of current treatment approaches for averting years living with disability, the disparities in access to the mental health care delivery system, and the workforce shortages to meet the mental health needs of older Latinos, development and testing of innovative strategies to prevent depression and anxiety are of great public health significance and have the potential to change practice. Effective approaches to this challenge are likely to involve using nontraditional means (i.e., community health workers [CHWs], health promotion, etc.) that are acceptable and scalable. Results of this study indicate that the Happy Older Latinos are Active (HOLA) intervention is a feasible and culturally acceptable alternative that could potentially impact mental illness prevention among at-risk older Latinos.

Appendix

Findings from the data collection have been reported in a separate manuscript. The published article is a case study describing the impact the HOLA intervention had on one participant. In contrast, the current manuscript examines the feasibility, acceptability, and preliminary efficacy of the HOLA intervention using data collected from the entire sample, not just one participant.

Data Availability Statement:

Data will be made available to users under a data-sharing agreement that provides for a commitment to: (1) not identify any individual participant; (2) securing the data using appropriate computer technology; and (3) destroying or returning the data after analyses are completed.