Rural- and Non-rural-specific Themes in Personalized Intervention Messages for Adults with At-Risk Alcohol Use Living in Rural Areas

Abstract

Objective:

There is a lack of personalized interventions designed specifically to meet the healthcare needs of rural populations. The goal of this qualitative work is to identify themes in intervention content developed by people residing in rural areas as part of a remotely delivered intervention for at-risk alcohol use.

Methods:

This is formative, exploratory work from data obtained in a pilot clinical trial that employed episodic future thinking (EFT, visualizing future events) and volitional choice (VC, in-the-moment alternatives to drinking) interventions aimed at reducing alcohol use. Using participant-created messages, we double-coded messages into thematic categories (e.g., recreational activities, protective behavioral strategies) and subcategories (e.g., family/kinship, purchases). Subcategories were further classified as rural-aligned or non-rural-specific. Each message had additional participant ratings of condition-specific variables (e.g., importance, enjoyment).

Results:

Participants in active intervention conditions (N=57) (i.e., EFT, VC, EFT/VC), created a total of 434 personalized messages. Of the 212 EFT messages, 17.0% were categorized as goal events, 81.1% as recreation events, and 1.9% as both. Of the 222 VC messages, 51.4% were categorized as alcohol protective behavioral strategies, 45.9% as alternative alcohol-free activities, and 2.7% as both. Nearly half (46.8%) of EFT messages were rural-aligned subcategories, whereas 17.1% of VC messages were rural-aligned.

Conclusions:

The messages displayed common themes, including both rural-aligned and non-rural-specific features. Consideration of cultural-centeredness in early intervention content may enhance its appeal and alignment with the populations served.

1. Introduction

At-risk alcohol use is associated with a variety of adverse health consequences such as liver and heart diseases, injury, certain types of cancers, impaired driving, and developing an alcohol use disorder (Kranzler, 2023; World Health Organization, 2019; Kaner et al., 2018). The 2023 National Survey on Drug Use and Health (Substance Abuse and Mental Health Services Administration, 2024) found that an estimated 134.7 million people (47.5 percent of those ages 12 and older) drank alcohol in the past month. Among them, 61.4 million people, or 21.7 percent, were binge drinkers, a form of at-risk drinking (Substance Abuse and Mental Health Services Administration, 2024). Despite its widespread use in the United States and recognition as a major public health issue, alcohol use is undertreated in healthcare settings (DiMartini et al., 2022).

Alcohol use in rural areas.

Geographical area and population density plays a role in alcohol use and alcohol-related harms (Friesen et al., 2022). Research to understand how alcohol use and alcohol-related harms vary between rural and urban communities has largely indicated that hazardous alcohol use and alcohol-related harms are more prevalent in rural and remote communities relative to urban communities even as risk factors are present among both populations (Borders & Booth, 2007; Dixon & Chartier, 2016; Friesen et al., 2022), though some other reports have found conflicting evidence (Dixon & Chartier, 2016). Individuals living in rural communities have a notable range of complex risk factors and harms relating to alcohol use; some risk factors include smoking cigarettes, having friends or family members who drink alcohol, being single, unmarried, divorced, or widowed, and having no religious involvement (Friesen et al., 2022). A scoping review by Freisen et al. (2022) found that earlier onset of alcohol use and high volumes and/or frequency of alcohol use were associated with subsequent alcohol-related harms, particularly in rural and remote communities, though many of these risk factors may be prevalent both in rural and urban settings. Despite these known risk factors and harms of alcohol use in rural and remote communities, rural health centers have fewer services and less diverse options for specialized treatment (Bond Edmond et al., 2015). Improved access to prevention and treatment services are a need within rural communities, however, very few interventions are tailored to rural-residing individuals.

The role of mobile health.

Mobile interventions offer feasible, low-cost, and scalable means for reaching rural-residing individuals and have potential to fill the gap between preventive and higher levels of care (Bayrakdarian et al., 2024; Kaner et al., 2018; Laditka & Laditka, 2009; Weinhold & Gurtner, 2014). Access to mobile interventions may reduce barriers such as transportation (Benavides-Vaello et al., 2013), as 91% of rural Americans own cell phones and rates of home broadband and smartphone ownership are increasing (Mobile Fact Sheet, 2024; Vogels, 2021). Additionally, remotely-delivered care may minimize stigma-related concerns by supporting enhanced privacy, which is especially important given rural communities hold greater concern for anonymity. In small rural towns, residents often know one another personally, increasing the likelihood that someone seeking help for substance use will be recognized when engaging with a treatment facility. This visibility can intensify fears of judgment or professional and social consequences, especially in communities where stigma surrounding alcohol and other substance use remains high (Browne et al., 2016; Nelson et al., 2023). Approaches that protect privacy, such as digital and telehealth services, anonymous support groups, or discreet entry points, can reduce perceived risks and empower more people in rural areas to seek care without fear of exposure or judgment.

Personalizing intervention content to reduce alcohol use.

Episodic Future Thinking (EFT) interventions involve guiding individuals to vividly imagine personal, positive future events to increase the salience of delayed rewards, thereby enhancing decision-making and reducing impulsive behaviors associated with substance use (Snider et al., 2016). EFT has demonstrated promise in addressing at-risk alcohol and substance use by reducing impulsivity and immediate reward preference, critical factors underpinning addictive behaviors (Bulley & Gullo, 2017; Patel & Amlung, 2020). Specifically, EFT interventions have been associated with decreased alcohol consumption, reduced craving, and improved delay discounting, highlighting their effectiveness as a brief, accessible tool for mitigating risky substance use (Stein et al., 2016; Voss et al., 2022).

Volitional choice (VC) interventions, also known as implementation intention interventions, involve helping individuals form specific, detailed plans that link anticipated challenging situations to predefined behavioral responses (Gollwitzer & Sheeran, 2006). Volitional choice interventions have been associated with reduced alcohol consumption, fewer binge drinking episodes, and increased ability to resist situational triggers by creating automatic associations between high-risk contexts and healthier behavioral alternatives (Hagger et al., 2012; Armitage, 2009). The ease of integration into existing interventions make volitional choice strategies a promising approach for addressing substance use (Norman et al., 2018).

The current study.

Although rural communities encompass a complex mixture of priorities and interests, prior work suggests the presence of several shared values common to rural regions (Deggs & Miller, 2011). Individuals living in rural areas tend to place a high value on kinship and family relationships, religion, and close ties to the outdoors and their land (Belanche et al., 2021). Recent work has emphasized the importance of interventions that reflect and center on the values of the target population, such as rural Americans, to increase relatability and efficacy (Staton et al., 2019). Thus, to make treatment options more person-centered for those residing in rural communities, the aim of this work is to better understand the integration of rural values into personalized interventions for at-risk alcohol use. The goal of this work is to inform ways in which rural values may be integrated into alcohol use interventions to increase receptivity and impact. The current report is a secondary analysis of a mobile intervention for at-risk alcohol use in adults living in rural areas (Bayrakdarian et al., 2024). This clinical trial included two distinct behavioral economic intervention components: episodic future thinking (EFT), focusing on future, positive personal events (Athamneh et al., 2022; Moody et al., 2018), and volitional choice (VC), focusing on alternative activities to engage in instead of drinking (Brown & Stein, 2022). Participants in active intervention conditions were randomly assigned to EFT, VC, or a combined EFT/VC condition with an even allocation by sex and Alcohol Use Disorders Identification Test (AUDIT-C) score.

As described in Bayrakdarian et al. (2024), during the telephone-delivered intervention induction session, collaboratively with research staff, participants created highly personalized intervention messages. Intervention messages were later delivered via text message as ecological momentary interventions. Research staff followed a semi-structured induction session guide. During EFT induction sessions, staff prompted participants to separately envision six personal events or experiences at future timepoints (one month, two months, three months, six months, one year, five years). After the event or experience was named, participants were prompted with follow-up questions to elicit more details about the event, and then they were guided to create a message summarizing the event. Table 1 includes examples of messages generated by participants from each of the six future timepoints included in the EFT condition. Based on prior EFT literature, for each message, participants (N=17) were then asked to rate: (1) how important the event is to them; (2) how enjoyable they expect the event to be; (3) how excited they are for the event to happen; and (4) how vivid the thought of the future event is, each on a 0–100 scale, where 0 equals not at all (e.g., vague or hard to think about in much detail) and 100 equals extremely (e.g., vivid and almost like it is happening right now) (Athamneh et al., 2021).

Table 1.

Select examples of personalized messages created by participants during induction sessions.

Condition	Participant-created Personalized Messages
Episodic Future Thinking (EFT)	“In 1 month, I will take my grandson to the fair.”
	“In 2 months, I will be getting ready to start my garden.”
	“In 3 months from now, we’ll be celebrating Easter with the kids and enjoying a beautiful spring day.”
	“In 6 months from now, we will be trout fishing and riding quads up north.”
	“In 1 year, I will reach my health goal.”
	“In 5 years, I’ll be retired!”
Volitional Choice (VC)	“If I am having a hard day, then I will call a friend or family member.”
	“If I am having a bonfire, then I will drink apple cider.”
	“If I’m feeling stressed when coming home from work, then I will work in the garden.”
	“If I am on vacation, then I will order a lemonade instead.”
	“If it’s a Friday after work, then I will take the dog for a walk.”
	“If I am experiencing back pain, then I will try stretching.”

During VC induction sessions, staff prompted participants (N=21) to identify six situations or times where they typically drink alcohol and feasible, in-the-moment alternatives to drinking for each scenario. After identifying a situation and alternative, participants were asked follow-up questions to obtain details about the alternative, and then they were guided to create “if/then” messages to sum up the situation and alternative. Table 1 includes participant-generated examples of the VC “if/then” messages created for various alternatives. For each message, participants were asked to rate: (1) how likely they are to engage in that alternative and (2) how enjoyable they expect the alternative to be, each on a 0–100 scale (0 equals not at all, 100 equals extremely). Participants in the combined EFT/VC condition (N=19) completed both intervention components, resulting in a total of 12 personalized messages (6 EFT and 6 VC). Messages developed during the induction session were subsequently texted twice per day over a 14-day intervention period, with the order of messages randomized without duplication of messages within the same day.

Herein, we examine the brief, personalized intervention messages developed collaboratively by participants and research staff for the EFT and VC conditions to identify key themes of personalized intervention content across participants, as well as to assess alignment with rural-specific features, based on rural values identified in prior literature (Staton et al., 2019; NORC at the University of Chicago, 2018).

2. Methods

This project focuses on the intervention content developed with participants in a longitudinal intervention study aiming to reduce at-risk alcohol use among rural-residing primary care patients. The project received Institutional Review Board approval (HUM00181940) and is registered in ClinicalTrials.gov (NCT05235971). Intervention outcomes are reported in Bayrakdarian et al. (2024); herein, we focus on the components of the personalized intervention. Participants resided in rural Michigan and were patients at a large health system. Participants were eligible to enroll in the study if the following criteria were met: (1) an AUDIT-C score of ≥3 in females and ≥4 in males (Bradley et al., 2007; Bush et al., 1998), (2) age 18 or older with a primary care appointment within the past two years, (3) a home address zip code in a rural-designated area, and (4) regular access to an internet-enabled device (e.g., smartphone, computer, tablet). Rural-designated areas include those identified as Rural-Urban Commuting Area (RUCA) (Category E; RUCA, n.d.) or CMS Rural Health Areas (Rural Health Clinics Center, n.d.). Study exclusion criteria were: (1) does not understand English, (2) currently pregnant, (3) unable to provide informed consent due to medical/psychiatric reasons, and/or (4) current treatment for a substance use disorder (Bayrakdarian et al., 2024).

The goal of this work is to identify, categorize, and describe the themes of personalized messages created collaboratively with participants (see Figure 1). Then, we present the association between categories of themes, including both rural-aligned and more non-rural-specific themes, with participant ratings of each message (e.g., vividness, enjoyment, importance, etc.) on features hypothesized to drive intervention effectiveness.

Figure 1. Process schematic.

2.1). Development of the Message Categorization Process

Prior to analysis, research staff created a list of topical theme areas (e.g., family, religion, outdoors) grounded in prior work on embedding rural values into rural behavioral health interventions (Oser et al., 2022; Staton et al., 2019). Next, research staff reviewed messages to refine themes based on observed message content. Using this set of themes, we assessed samples (n~15 messages), refining the categories and subcategories over the course of three iterations to improve fit with the heterogeneity of messages (see Table 2). The fourth and final iteration of message categories (See Table 3) split EFT messages into non-exclusive goal and recreational events categories, with subcategories for each category. VC messages were organized into non-exclusive categories of protective behavioral strategies to reduce alcohol use and alternative alcohol-free activities, with multiple subcategories thereafter. Every message was grouped into at least one of the categories; however, due to the wide variety of personalized messages, it was possible for a message to be labeled as both condition-specific categories. For example, the EFT message “In 5 years from now, I will retire and travel” was categorized as both a goal and a recreational event.

Table 2.

Refinements to Personalized Message Categorization Process.

	# of Personalized Messages Tested	Main Discrepancies	Main Refinements
Iteration1	12	Wide variety of outdoor activities. Lack of specific rural-aligned language and subcategories.	Dividing of outdoor activities into nested subcategories. Changing “family” to “kinship” subcategory. The EFT subcategories were reorganized to add “Volunteering” and “Indoor physical activity”. “Social” was removed.
Iteration2	36	“Milestone event” not the best fit/description for the range of subcategories. Repeated mention of live performance and socializing events was not well captured in subcategories.	Removing the EFT “Milestone event” category and integrating the subcategories into new subcategories of “Health”, “Vocation”, “Purchase”, and “Moving” in the “Goal” category. The subcategories of “Birthday”, “Live performance”, “Socializing” were added into the “Recreation event” category.
Iteration3	76	Need for explicit definitions and examples as staff completed categorizations. Lack of context from VC “if” statement.	A categorization codebook was added for staff to use as a reference that defined each category and subcategory. “If” statement categorization process was added for VC categorization to capture context for the first half of the personalized message statement
Iteration4	425	N/A	No refinements. Consensus followed the double entry categorization of all 425 personalized messages for staff to address any discrepancies in the categorization process.

Table 3.

Relative frequencies of all categories and subcategories for EFT and VC messages

	Number of Messages Endorsed
VC Personalized Messages
Alcohol Protective Behavioral Strategy	114 (51.4%)
Substitute Foods or Drink	99 (44.6%)
Limiting Accessibility	11 (5.0%)
Alternative Pain Management	6 (2.7%)
Cost Reminders	3 (1.4%)
Alternative Alcohol-free Activity	102 (45.9%)
Selfcare / Relaxation	36 (16.2%)
Outdoor – Physical Activity	22 (9.9%)
Socializing	14 (6.3%)
Physical Activity – Typically Indoor	8 (3.6%)
Family / Kinship	5 (2.3%)
General Outdoor Activity	4 (1.8%)
Outdoor – Home Cultivation	4 (1.8%)
Pets / Animals	4 (1.8%)
Outdoor – Selfcare / Relaxation	2 (0.9%)
Other	22 (9.9%)
Both Alcohol Protective Behavioral Strategy and Alternative Alcohol-free Activity	6 (2.7%)
EFT Personalized Messages
Goal Event	36 (17.0%)
Vocation	15 (7.1%)
Health	12 (5.7%)
Purchase	6 (2.8%)
Moving	5 (2.4%)
Recreation Event	172 (81.1%)
Family / Kinship	67 (31.6%)
Travel / Vacation	44 (20.8%)
Birthday	22 (10.4%)
Seasonal / Weather	22 (10.4%)
Holiday	18 (8.5%)
Socializing	17 (8.0%)
Outdoor – Physical Activity	16 (7.5%)
General Outdoor Activity	12 (5.7%)
Festival / Fair	12 (5.7%)
Live Performance	11 (5.2%)
Outdoor – Home Cultivation	6 (2.8%)
Outdoor – Selfcare / Relaxation	6 (2.8%)
Religion	3 (1.4%)
Physical Activity – Typically Indoor	0 (0.0%)
Volunteering	0 (0.0%)
Other	14 (6.6%)
Both Goal Event and Recreation Event Categories	4 (1.9%)

Key refinements included reformatting the categorization of VC messages to separately label information relating to the situation and alternative activity (the “if” and “then” portions of the message), allowing for more specific assessment of the alternative rather than the situation in which a participant was likely to use alcohol. Another key refinement was to define broader message category labels based on condition (EFT or VC) and inclusion of rural-aligned language and subcategories. Adapted from prior literature (Staton et al., 2019; NORC at the University of Chicago, 2018), subcategories that align with rural living and values were specified within the categorization schema prior to final classifications as a result of the iterative refinement process.

2.1.1. Message Categories

The final set of categories and subcategories for messages from each condition reflects the feedback from refinement processes (See Figure 1).

Rural-aligned subcategories.

For EFT messages, rural-aligned subcategories were kinship and family, outdoor activity (including home cultivation, physical activity, and relaxation), religion, festival/fair, volunteering, and seasonal/weather. For VC messages, rural-aligned subcategories included kinship and family, outdoor activity (i.e., home cultivation, physical activity, and relaxation), and pets or animals.

Non-rural-specific subcategories.

For EFT messages, non-rural-specific subcategories were health, moving, purchase, vocation, holiday, socializing, birthday, live performance, indoor physical activity, and travel/vacation. For VC messages, non-rural-specific subcategories were substitute food/drink, limiting accessibility, pain management, cost reminders, socializing, and indoor physical activity.

2.2). Message Coding

Research staff involved with coding messages were trained on categorization procedures to enhance fidelity and consistency. Training included review of instructions to read individual messages and assign one or both categories, and up to 4 subcategories that best fit the content of the message. Coding staff completed training, including supervised categorization practice prior to independently coding messages. A codebook with additional examples of events or activities in each subcategory was created as a resource for staff when coding messages to enhance reliability (see Supplementary Material A for further details).

Each personalized message was double-coded by independent coders. After sorting and organizing all personalized messages, staff examined discrepancies between the first and second coders. Messages were flagged if there were inter-coder discrepancies. Discrepancies in coding were resolved by a lead staff member (L.H.) in collaboration with the staff coders. First, consensus was reached for the overarching categories (e.g., EFT: goal or recreational events), followed by subcategory discrepancies. The coders achieved consensus through discussion of the discrepant categorizations, with the lead staff member (L.H.) resolving discrepancies.

2.3. Analysis Plan

Analyses presented herein are descriptive and aimed at informing future work on rural populations who use alcohol. First, we describe the sociodemographics of the participants in this study. Next, we present descriptives on the personalized intervention messages. Then, we present the coding outcomes, including message categorizations that were rural- and non-rural-specific. Finally, we look at the intersection between message categorizations and participant-rated features of each message (e.g., importance, vividness, etc.).

3. Results

Among participants in active conditions (N=57), 61.4% identified as female sex, 100% identified as Caucasian, and the mean age was 55.6 (Bayrakdarian et al., 2024). Participants in active conditions (EFT N=17, VC N=21, Combined N=19) created a total of 434 personalized messages during their telephone-delivered intervention induction sessions. Participants in the EFT or VC conditions were guided to create six total messages, with participants in the combined condition guided to create 12 total messages. This would give a possible total of 456 messages, however ten participants chose not to create the full set of messages, resulting in 22 total (4 EFT and 18 VC) missing messages.

3.1). Personalized Message Categories

Of the 212 EFT messages created, 36 (17.0%) were categorized as a goal event and 172 (81.1%) were categorized as recreation events. Four (1.9%) EFT messages endorsed both categories. Of the 222 VC messages created, 114 (51.4%) were categorized as alcohol protective behavioral strategies, and 102 (45.9%) were categorized as alternative alcohol-free activities. Six (2.7%) VC messages endorsed both categories. Table 3 includes relative frequencies of all categories for EFT and VC messages.

3.2). Personalized Message Subcategories

The most common subcategories for EFT were family/kinship (67 total; 21.8%) and travel/vacation (44 total; 14.3%). For VC, the most common subcategories were substitute food/drink (99 total; 41.3%) and self-care/relaxation (36 total; 15%). Table 3 includes relative frequencies of all subcategories for EFT and VC messages.

3.3). Rural-aligned Categorization

The distribution of subcategories used for EFT messages is shown in Figure 2, with rural-aligned subcategories highlighted in dark gray. Rural-aligned subcategories were endorsed 46.8% of the time for EFT messages. The distribution of subcategories used for VC messages is shown in Figure 3, with rural-aligned subcategories highlighted in dark gray. Rural-aligned subcategories were endorsed 17.1% of the time for VC messages.

Figure 2: EFT subcategory distribution.

Figure 3: VC subcategory distribution.

3.4). Message ratings.

In the EFT condition, goal categories had higher importance ratings, while recreational categories had higher enjoyment, excitement and vividness ratings. The EFT subcategory with the highest importance rating was religion (M=95.0, SD=5.0). The EFT subcategory with the highest enjoyment rating was also religion (M=98.3, SD=2.9). The EFT subcategory with the highest excitement and vividness rating was holiday (excitement: M=91.3, SD=13.0; vividness: M=89.4, SD=9.7; see Supplementary Table 1).

In the VC condition, alternative alcohol-free categories received higher ratings for enjoyment and likelihood of engagement than protective behavioral strategies. The VC subcategory with the highest enjoyment and likelihood of engagement rating was outdoor home cultivation (enjoyment: M=95.0, SD=7.1; likelihood: M=100, SD=0; see Supplementary Table 2).

Figure 4 shows mean ratings for rural-aligned EFT subcategories compared to non-rural-specific subcategories, with rural-aligned subcategories highlighted in dark gray. Figure 5 shows mean ratings for rural-aligned VC subcategories compared to non-rural-specific subcategories. Rural-aligned subcategories for EFT messages had higher ratings for importance and enjoyment (Figure 4).

Figure 4: Mean ratings for EFT subcategories.

Figure 5: Mean ratings for VC subcategories.

4. Discussion

The current report looks at personalized intervention message content developed by adults residing in rural communities with at-risk alcohol use who are non-treatment-seeking, but enrolled in an early intervention trial. These individuals typically can face barriers to accessing effective care that are perpetuated by intervention options that may not fit within their unique culture and lifestyle (Browne et al., 2016). Aligned with personalized medicine approaches, one direction to improve treatment options in rural areas is the incorporation of person-centered, individualized intervention content that aligns with rural lifestyle and values. The increased use of personalized treatment options may help increase the appeal and engagement of treatment options in rural areas. The presence of rural-aligned themes and values in the personalized messages, which were guided by participant input during the induction session, highlights an opportunity to better incorporate content that aligns with the lives of the populations served.

One finding of note is that nearly half of personalized message content in the EFT condition was rural-aligned, whereas only a fraction (17.1%) of message content in the VC condition was rural-aligned. This may be tied to the structure of these interventions, given that EFT involves envisioning oneself in the future, which may be closely tied to a person’s rural lifestyle, whereas VC involves identifying alternative in-the-moment activities instead of alcohol use, which may lend itself to more global strategies for drinking reduction. The case may be that personalized intervention messages developed in non-rural communities may show fewer themes related to items aligned with rural-living (e.g. outdoor activities, religion, etc.) and, thus, these items may be less common for adults residing in other geographic localities.

This work is formative and exploratory, and we did not do inferential statistical testing; however, descriptively, it appears that messages within the goal category were rated as highly important to participants, while upcoming recreational events were rated as highly enjoyable, exciting, and vivid. Our findings fit with prior EFT research (Athamneh et al., 2021) showing that individual perceptions differ on types of future events and may have varying impacts on effectiveness of behavioral interventions. With regard to the VC condition, alternative alcohol-free activities were rated as having high importance and likelihood of engagement. This is consistent with prior behavioral economic intervention work showing that engaging in alternative alcohol-free activities reduced alcohol use and related harms (Murphy et al., 2012).

Rural-aligned message content stood out as particularly salient for both the EFT and VC messages. In the EFT condition, rural-aligned messages generally showed high participant-endorsed ratings in importance and enjoyability. In the VC condition, rural-aligned messages were rated highly by participants in terms of likelihood of engagement and enjoyment. Together, these findings highlight that participant-driven personalized intervention content incorporates rural values that the participants perceive as particularly salient. Future research may examine the difference in other behavioral economics outcomes and alcohol use between these two intervention options.

The current report should be considered in light of several limitations. First, this is formative, hypothesis-generating work based on a relatively modest sample size of adults living in rural Michigan. This may limit the generalizability of findings and limit the ability to compare salience and other key variables of rural-aligned vs. non-rural-specific messages. For example, there was a low quantity of religious messages created by participants but the ones created were rated as being highly important and enjoyable, impacting analysis of those messages in the context of subcategory salience. In addition, participants resided across a wide geographic region of rural Michigan, spanning several counties. We acknowledge that rural areas are highly diverse and that rural values vary across different rural areas. Future work may focus more specifically on particular rural communities to extend this research, noting that intervention messages are highly personalized and that different types of messages could emerge as salient in different communities. A next step for this line of research is to evaluate the association between the delivery of messages from particular categories and changes in drinking to examine how message content may impact alcohol use. Another direction for future work may be to evaluate the presence of rural- and non-rural aligned message categories in rural, urban, and other geographically diverse settings to better understand the unique values of rural communities.

In summary, this study offers a novel contribution by systematically identifying how rural-residing individuals naturally embed culturally aligned values into personalized intervention content. These findings extend beyond descriptive insights to provide a foundation for developing message libraries that reflect the priorities of rural populations—an often-overlooked aspect in the design of remotely delivered health interventions. These findings have the potential to inform scalable, person-centered approaches that resonate with the lived experiences of rural individuals and increase their engagement with care. By demonstrating how behavioral economic interventions, such as EFT and VC, can be adapted to reflect rural-specific themes, this work can inform future studies seeking to enhance cultural congruence, engagement, and effectiveness in remotely delivered interventions for alcohol and other substance use. Furthermore, the thematic coding framework developed here can be applied or adapted to tailor content in other geographic or cultural contexts, supporting broader efforts toward personalized, scalable interventions.

Public Significance Statement:

Individuals living in rural areas face a greater risk of alcohol-related harms, yet lack accessible care options that are tailored to meet their needs. This work highlights the benefit of creating rural-focused interventions to support meaningful treatment for substance use among this population.