David Akopian
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
David Akopian has a total of 214 co-authors. Here are the top co-authors by number of shared publications:
- Jaakko Astola
Department of Signal Processing, Tampere University of Technology, Finland
12 shared publications Sos Agaian
Electrical and Computer Engineering Department University of Texas San Antonio TX 78249 USA
12 shared publications- Grant Huang
University of Texas at San Antonio, San Antonio, TX, USA
11 shared publications Arsen Melkonyan
University of Texas, San Antonio
10 shared publications- Devasena Inupakutika
University of Texas at San Antonio, USA
10 shared publications - Chun Lung Philip Chen
South China University of Technology
9 shared publications
Latest Publications
- 31 March 2026
- Published by Oxford University Press (OUP) in Health Education Research
- Vol. 41 (4)
Dataset development and performance analysis of a specialized conversational artificial intelligence model for targeted smoking cessation dialogues
Conversational artificial intelligence (AI) is increasingly being used in healthcare to encourage behavioral modification, information dissemination, and patient engagement. However, its effectiveness depends on high-quality, domain-specific training data. General-purpose datasets are often inadequate in terms of clinical specificity and dialog coherence, particularly in sensitive domains like smoking cessation. This study introduces a curated question-and-answer dataset comprising 3000 Q&A pairs from reputable health resources, including expert-authored guidelines and cessation programs. Each entry is annotated with metadata reflecting informational intent, action-oriented content, and user engagement signals, supporting both single-turn and multi-turn conversational modeling. A compact generative language model, Chat Generative Pre-trained Transformer 4o-mini (ChatGPT-4o-mini), was fine-tuned using this dataset and compared against the pretrained base model using BLEU, ROUGE-L, and BERTScore metrics, assessing syntactic accuracy, lexical coverage, and semantic alignment. The fine-tuned model demonstrated improved response clarity, conversational coherence, and adherence to evidence-based content. This study emphasizes methodological benchmarking and model performance rather than behavioral or clinical outcomes, isolating the model-optimization layer to assess how curated, domain-specific training data influence response quality under controlled conditions. By contributing a reproducible pipeline for dataset construction and evaluation, this work supports safe and effective conversational agents in healthcare contexts and promotes responsible AI practices in digital health applications. Statement of significance The success of large language model (LLM)-based conversational agents depends on access to high-quality, domain-specific training data. However, such data is typically absent from models trained on general web corpora. Smoking cessation, as a behavior-change domain with substantial clinical burden and health disparities, requires emotionally attuned and evidence-based messaging to be effective. Rather than introducing a new smoking cessation intervention, this study contributes a reproducible methodological framework for evaluating how curated, evidence-based datasets influence LLM behavior in health education contexts. By isolating the dataset-to-model performance relationship, this work provides a technical foundation that complements prior user-centered intervention studies and supports responsible development of conversational AI systems prior to behavioral deployment. This study contributes a rigorously curated dataset drawn from evidence-based sources, annotated with metadata for motivational framing, tone, and dialogue intent. In parallel, we present a methodology for fine-tuning LLMs and evaluating performance in both single-turn and multi-turn conversational settings.
- 31 March 2026
- Published by Oxford University Press (OUP) in Health Education Research
- Vol. 41 (4)
Dataset development and performance analysis of a specialized conversational artificial intelligence model for targeted smoking cessation dialogues
Conversational artificial intelligence (AI) is increasingly being used in healthcare to encourage behavioral modification, information dissemination, and patient engagement. However, its effectiveness depends on high-quality, domain-specific training data. General-purpose datasets are often inadequate in terms of clinical specificity and dialog coherence, particularly in sensitive domains like smoking cessation. This study introduces a curated question-and-answer dataset comprising 3000 Q&A pairs from reputable health resources, including expert-authored guidelines and cessation programs. Each entry is annotated with metadata reflecting informational intent, action-oriented content, and user engagement signals, supporting both single-turn and multi-turn conversational modeling. A compact generative language model, Chat Generative Pre-trained Transformer 4o-mini (ChatGPT-4o-mini), was fine-tuned using this dataset and compared against the pretrained base model using BLEU, ROUGE-L, and BERTScore metrics, assessing syntactic accuracy, lexical coverage, and semantic alignment. The fine-tuned model demonstrated improved response clarity, conversational coherence, and adherence to evidence-based content. This study emphasizes methodological benchmarking and model performance rather than behavioral or clinical outcomes, isolating the model-optimization layer to assess how curated, domain-specific training data influence response quality under controlled conditions. By contributing a reproducible pipeline for dataset construction and evaluation, this work supports safe and effective conversational agents in healthcare contexts and promotes responsible AI practices in digital health applications. Statement of significance The success of large language model (LLM)-based conversational agents depends on access to high-quality, domain-specific training data. However, such data is typically absent from models trained on general web corpora. Smoking cessation, as a behavior-change domain with substantial clinical burden and health disparities, requires emotionally attuned and evidence-based messaging to be effective. Rather than introducing a new smoking cessation intervention, this study contributes a reproducible methodological framework for evaluating how curated, evidence-based datasets influence LLM behavior in health education contexts. By isolating the dataset-to-model performance relationship, this work provides a technical foundation that complements prior user-centered intervention studies and supports responsible development of conversational AI systems prior to behavioral deployment. This study contributes a rigorously curated dataset drawn from evidence-based sources, annotated with metadata for motivational framing, tone, and dialogue intent. In parallel, we present a methodology for fine-tuning LLMs and evaluating performance in both single-turn and multi-turn conversational settings.
- 15 March 2026
- Published by SAGE Publications in Hispanic Journal of Behavioral Sciences
- Vol. 48 (3) , 263-284
Longitudinal Relationships Between Parental Support and Physical Activity Behavior Among Hispanic Children With Overweight/Obesity
Childhood obesity disproportionately affects Hispanic children in the United States, with physical inactivity as a major contributor. Hispanic culture emphasizes family ties, positioning parents as key influencers of children’s physical activity (PA). Emerging evidence suggest that children’s PA may shape parental behaviors, such as parents’ provision of support for PA and engagement in PA alongside their child. This study examined bi-directional associations between parental PA support and child PA among Hispanic children with overweight/obesity. Secondary analyses were conducted using three waves of data from the Health4Kids randomized controlled trial, a family-based intervention in Texas with rolling enrollment between June 2015 and August 2018. Participants included 253 Hispanic children ( M age = 8.75 ± 1.68 years; 50.6% male) with overweight/obesity and one parent (94% mothers). Children wore an ActiGraph GT3X accelerometer for 7 days to measure moderate-to-vigorous PA (MVPA) at baseline, 6 months, and 12 months. Parents completed the Comprehensive Home Environment Survey to assess PA modeling and support at baseline, 6 months, and 12 months. Separate cross-lagged panel models were computed for child MVPA with parental modeling and support. Results revealed a bi-directional relationship between child MVPA and parental support from baseline to 6 months, however, this relationship did not persist from 6 to 12 months. No significant cross-lagged associations were observed between parental modeling and child MVPA across any time points. Findings highlight a bi-directional relationship between parental support and child PA, suggesting that while parental support facilitates child PA, active children may also prompt parents to provide greater support. Parental modeling showed no significant influence on child PA, indicating that direct support may be more impactful than modeling. Overall, results highlight the importance of examining specific parental influences and suggest that parental support may be a critical target for interventions aiming to promote PA among Hispanic children with overweight/obesity.
- 15 March 2026
- Published by SAGE Publications in Hispanic Journal of Behavioral Sciences
- Vol. 48 (3) , 263-284
Longitudinal Relationships Between Parental Support and Physical Activity Behavior Among Hispanic Children With Overweight/Obesity
Childhood obesity disproportionately affects Hispanic children in the United States, with physical inactivity as a major contributor. Hispanic culture emphasizes family ties, positioning parents as key influencers of children’s physical activity (PA). Emerging evidence suggest that children’s PA may shape parental behaviors, such as parents’ provision of support for PA and engagement in PA alongside their child. This study examined bi-directional associations between parental PA support and child PA among Hispanic children with overweight/obesity. Secondary analyses were conducted using three waves of data from the Health4Kids randomized controlled trial, a family-based intervention in Texas with rolling enrollment between June 2015 and August 2018. Participants included 253 Hispanic children ( M age = 8.75 ± 1.68 years; 50.6% male) with overweight/obesity and one parent (94% mothers). Children wore an ActiGraph GT3X accelerometer for 7 days to measure moderate-to-vigorous PA (MVPA) at baseline, 6 months, and 12 months. Parents completed the Comprehensive Home Environment Survey to assess PA modeling and support at baseline, 6 months, and 12 months. Separate cross-lagged panel models were computed for child MVPA with parental modeling and support. Results revealed a bi-directional relationship between child MVPA and parental support from baseline to 6 months, however, this relationship did not persist from 6 to 12 months. No significant cross-lagged associations were observed between parental modeling and child MVPA across any time points. Findings highlight a bi-directional relationship between parental support and child PA, suggesting that while parental support facilitates child PA, active children may also prompt parents to provide greater support. Parental modeling showed no significant influence on child PA, indicating that direct support may be more impactful than modeling. Overall, results highlight the importance of examining specific parental influences and suggest that parental support may be a critical target for interventions aiming to promote PA among Hispanic children with overweight/obesity.
- Open Access
- 2 February 2025
- Published by Society for Imaging Science & Technology in Electronic Imaging
- Vol. 37 (3) , 323-1
Overcoming Chatbot Platform Limitations: A Standalone Client for Task-oriented Dialog System Support
This paper presents the design and development of a standalone, cross-platform client application that can connect and support various chatbot development platforms, thereby avoiding the limitations imposed by mainstream messaging channels that are typically used for user access. While third-party systems like Facebook Messenger and WhatsApp facilitate chatbot communication, they impose various restrictions on automated messaging, including message timing, scheduling, chatbot lifecycle, and content. These limitations disrupt continuous, so-called deep-logic or long-term interactions with users, hindering the effectiveness of extended engagement in customer service and support. As advanced chatbots gain enormous popularity, this paper envisions a growing need for standalone and perhaps open-source solutions that can connect to chatbot development platforms to support full ownership of the channel, thereby eliminating communication limitations. The proposed generic cross-platform client app is demonstrated utilizing Google Dialogflow chatbot development platform. Dialogflow supports natural language processing (NLP), and it seamlessly integrates with Google Firebase for backend data storage and real-time messaging. The client app supports flexible message scheduling continuity, multimedia features, quick-reply buttons, and scalability for up to 10,000 users. Through a flexible, independent design, this prototype showcases a scalable, unrestricted chatbot solution that enhances user satisfaction and engagement across various devices.
- 2 February 2025
- Published by Society for Imaging Science & Technology in Electronic Imaging
- Vol. 37 (3) , 323-1
Open AccessOvercoming Chatbot Platform Limitations: A Standalone Client for Task-oriented Dialog System Support
This paper presents the design and development of a standalone, cross-platform client application that can connect and support various chatbot development platforms, thereby avoiding the limitations imposed by mainstream messaging channels that are typically used for user access. While third-party systems like Facebook Messenger and WhatsApp facilitate chatbot communication, they impose various restrictions on automated messaging, including message timing, scheduling, chatbot lifecycle, and content. These limitations disrupt continuous, so-called deep-logic or long-term interactions with users, hindering the effectiveness of extended engagement in customer service and support. As advanced chatbots gain enormous popularity, this paper envisions a growing need for standalone and perhaps open-source solutions that can connect to chatbot development platforms to support full ownership of the channel, thereby eliminating communication limitations. The proposed generic cross-platform client app is demonstrated utilizing Google Dialogflow chatbot development platform. Dialogflow supports natural language processing (NLP), and it seamlessly integrates with Google Firebase for backend data storage and real-time messaging. The client app supports flexible message scheduling continuity, multimedia features, quick-reply buttons, and scalability for up to 10,000 users. Through a flexible, independent design, this prototype showcases a scalable, unrestricted chatbot solution that enhances user satisfaction and engagement across various devices.
- Open Access
SDR-Fi-Z: A Wireless Local Area Network-Fingerprinting-Based Indoor Positioning Method for E911 Vertical Accuracy Mandate
The Enhanced 911 (E911) mandate of the Federal Communications Commission (FCC) drives the evolution of indoor three-dimensional (3D) location/positioning services for emergency calls. Many indoor localization systems exploit location-dependent wireless signaling signatures, often called fingerprints, and machine learning techniques for position estimation. In particular, received signal strength indicators (RSSIs) and Channel State Information (CSI) in Wireless Local Area Networks (WLANs or Wi-Fi) have gained popularity and have been addressed in the literature. While RSSI signatures are easy to collect, the fluctuation of wireless signals resulting from environmental uncertainties leads to considerable variations in RSSIs, which poses a challenge to accurate localization on a single floor, not to mention multi-floor or even three-dimensional (3D) indoor localization. Considering recent E911 mandate attention to vertical location accuracy, this study aimed to investigate CSI from Wi-Fi signals to produce baseline Z-axis location data, which has not been thoroughly addressed. To that end, we utilized CSI measurements and two representative machine learning methods, an artificial neural network (ANN) and convolutional neural network (CNN), to estimate both 3D and vertical-axis positioning feasibility to achieve E911 accuracy compliance.
Open AccessSDR-Fi-Z: A Wireless Local Area Network-Fingerprinting-Based Indoor Positioning Method for E911 Vertical Accuracy Mandate
The Enhanced 911 (E911) mandate of the Federal Communications Commission (FCC) drives the evolution of indoor three-dimensional (3D) location/positioning services for emergency calls. Many indoor localization systems exploit location-dependent wireless signaling signatures, often called fingerprints, and machine learning techniques for position estimation. In particular, received signal strength indicators (RSSIs) and Channel State Information (CSI) in Wireless Local Area Networks (WLANs or Wi-Fi) have gained popularity and have been addressed in the literature. While RSSI signatures are easy to collect, the fluctuation of wireless signals resulting from environmental uncertainties leads to considerable variations in RSSIs, which poses a challenge to accurate localization on a single floor, not to mention multi-floor or even three-dimensional (3D) indoor localization. Considering recent E911 mandate attention to vertical location accuracy, this study aimed to investigate CSI from Wi-Fi signals to produce baseline Z-axis location data, which has not been thoroughly addressed. To that end, we utilized CSI measurements and two representative machine learning methods, an artificial neural network (ANN) and convolutional neural network (CNN), to estimate both 3D and vertical-axis positioning feasibility to achieve E911 accuracy compliance.
- Open Access
- 6 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 81631-81645
Assessing Performance of Cloud-Based Heterogeneous Chatbot Systems and A Case Study
Recently, human-machine digital assistants gained popularity and are commonly used in question-and-answer applications and similar consumer-supporting domains. A class of more sophisticated digital assistants (chatbots) employing more extended dialogs follows the trend. Chatbots have become increasingly popular in recent years. Nowadays, chatbot deployments in the cloud have become a common practice because of their benefits, including flexibility, scalability, reliability, security, remote working, cost, and power outages. However, measuring the cloud-based chatbot systems’ performance is challenging as human-machine information exchanges are performed in heterogeneous environments such as cloud hosting platforms, information processing units, and several machine-to-machine and human-machine communication channels. This paper investigates different methodologies for assessing the performance of such heterogeneous deployments and identifies performance metrics for evaluating the performance of cloud-based chatbot deployment. The study employs chatbot performance measurements with both real users (human) and automated (simulated) users. The experimental results discuss communication metrics such as response time, throughput, and load testing (connection loss) through the performance assessment of a case study deployment that utilizes an automated protocol chatbot development framework. The findings presented in this paper can further help practitioners to better understand the performance characteristics of a cloud-based chatbot and assist in making informed decisions related to the chatbot development and deployment options.
- 6 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 81631-81645
Open AccessAssessing Performance of Cloud-Based Heterogeneous Chatbot Systems and A Case Study
Recently, human-machine digital assistants gained popularity and are commonly used in question-and-answer applications and similar consumer-supporting domains. A class of more sophisticated digital assistants (chatbots) employing more extended dialogs follows the trend. Chatbots have become increasingly popular in recent years. Nowadays, chatbot deployments in the cloud have become a common practice because of their benefits, including flexibility, scalability, reliability, security, remote working, cost, and power outages. However, measuring the cloud-based chatbot systems’ performance is challenging as human-machine information exchanges are performed in heterogeneous environments such as cloud hosting platforms, information processing units, and several machine-to-machine and human-machine communication channels. This paper investigates different methodologies for assessing the performance of such heterogeneous deployments and identifies performance metrics for evaluating the performance of cloud-based chatbot deployment. The study employs chatbot performance measurements with both real users (human) and automated (simulated) users. The experimental results discuss communication metrics such as response time, throughput, and load testing (connection loss) through the performance assessment of a case study deployment that utilizes an automated protocol chatbot development framework. The findings presented in this paper can further help practitioners to better understand the performance characteristics of a cloud-based chatbot and assist in making informed decisions related to the chatbot development and deployment options.
- Open Access
- 18 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 144778-144798
A Comparative Assessment of Unsupervised Keyword Extraction Tools
The extraction of keywords is a critical task in natural language processing and information retrieval. It has become increasingly important in a wide range of applications, from search engines and e-commerce platforms to news and social media analysis. However, the evaluation of keyword extraction methods remains a challenging task due to the diverse range of data types and contexts used, as well as the complexity of the methodologies and techniques involved. In this regard, this study reviews the prior surveys on keyword extraction methods to comprehend the fundamental principles, difficulties in keyword extraction, and benchmark datasets. The reliability of evaluation techniques and an examination of their flaws remain two of the largest problems. Hence, this study includes the literature that performed a comparative analysis of popular keyword extraction methods. Furthermore, in this paper, we present a comparative evaluation of open-source unsupervised keyword extraction tools, analyzing their performance across a range of data types and under different testing conditions. The experimental analysis shows that, in terms of f-score, KPminer performs most consistently for different text lengths while KeyBERT(mmr) outperforms other tools. Considering the execution time, RAKE and YAKE are the fastest tools. Though graph-based tools tend not to perform well on long text, TopicRank and MultipartiteRank perform very well on long text as they use topics as nodes of the graph, which is another finding of this study. By highlighting the key factors that influence the performance of keyword extraction tools, our analysis contributes to directing the reader in selecting suitable keyword extraction tools for different applications.
- 18 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 144778-144798
Open AccessA Comparative Assessment of Unsupervised Keyword Extraction Tools
The extraction of keywords is a critical task in natural language processing and information retrieval. It has become increasingly important in a wide range of applications, from search engines and e-commerce platforms to news and social media analysis. However, the evaluation of keyword extraction methods remains a challenging task due to the diverse range of data types and contexts used, as well as the complexity of the methodologies and techniques involved. In this regard, this study reviews the prior surveys on keyword extraction methods to comprehend the fundamental principles, difficulties in keyword extraction, and benchmark datasets. The reliability of evaluation techniques and an examination of their flaws remain two of the largest problems. Hence, this study includes the literature that performed a comparative analysis of popular keyword extraction methods. Furthermore, in this paper, we present a comparative evaluation of open-source unsupervised keyword extraction tools, analyzing their performance across a range of data types and under different testing conditions. The experimental analysis shows that, in terms of f-score, KPminer performs most consistently for different text lengths while KeyBERT(mmr) outperforms other tools. Considering the execution time, RAKE and YAKE are the fastest tools. Though graph-based tools tend not to perform well on long text, TopicRank and MultipartiteRank perform very well on long text as they use topics as nodes of the graph, which is another finding of this study. By highlighting the key factors that influence the performance of keyword extraction tools, our analysis contributes to directing the reader in selecting suitable keyword extraction tools for different applications.
- Open Access
- 7 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 138986-139003
Detection and Mitigation of Spoofing Attacks Against Time Synchronization and Positioning
Global Navigation Satellite System (GNSS) receivers are vulnerable to intentional spoofing attacks which can manipulate position, velocity, and time (PVT) measurements. Previous work has demonstrated that Time Synchronization Attacks (TSAs) can be detected and mitigated using sparse optimization techniques which reveal spoofers’ presence in inflicted signals’ derivative domains. Aiming to provide an efficient protection algorithm against spoofing, this paper expands the scope of the sparse signal processing framework to address more complicated attacks for stationary and low-dynamic receivers. In particular, TSAs against stationary receivers should be addressed differently if the position and velocity are manipulated by the spoofer at the same time. A new sparse processing method is presented employing a novel linearization of the measurement equation that includes attacks against time, position, and velocity. The method is assessed against both authentically and synthetically spoofed signals to verify its robustness in two test beds: 1) a lab-based software defined GPS receiver, and 2) a commercial hand-held device.
- 7 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 138986-139003
Open AccessDetection and Mitigation of Spoofing Attacks Against Time Synchronization and Positioning
Global Navigation Satellite System (GNSS) receivers are vulnerable to intentional spoofing attacks which can manipulate position, velocity, and time (PVT) measurements. Previous work has demonstrated that Time Synchronization Attacks (TSAs) can be detected and mitigated using sparse optimization techniques which reveal spoofers’ presence in inflicted signals’ derivative domains. Aiming to provide an efficient protection algorithm against spoofing, this paper expands the scope of the sparse signal processing framework to address more complicated attacks for stationary and low-dynamic receivers. In particular, TSAs against stationary receivers should be addressed differently if the position and velocity are manipulated by the spoofer at the same time. A new sparse processing method is presented employing a novel linearization of the measurement equation that includes attacks against time, position, and velocity. The method is assessed against both authentically and synthetically spoofed signals to verify its robustness in two test beds: 1) a lab-based software defined GPS receiver, and 2) a commercial hand-held device.
- 16 January 2023
- Published by Society for Imaging Science & Technology in Electronic Imaging
- Vol. 35 (3) , 366-5
An RF modulation recognition method using machine learning
In recent few years, deep learning has been successfully applied in many fields to optimize decision making including self-driving cars, health care, machine translation, image recognition, etc. In wireless communication, deep learning has been used in channel estimation, signal classification, massive MIMOs, heterogeneous networks, energy harvesting, device-to-device (D2D) communications, and so on. In this paper, we applied machine learning (ML) and deep learning (DL) neural networks to RF signal recognition. Specifically, we built, trained, and tested two ML models, SVM and XGBoost, and two DL models, ConvNet and ResNet. We utilized the online dataset at radioml.com. Our goals are to learn how to scientifically apply ML/DL in terms of dataset processing, deep neural network constructing, training, testing, fine-tuning, and results analyzing and reporting.
- 16 January 2023
- Published by Society for Imaging Science & Technology in Electronic Imaging
- Vol. 35 (3) , 366-5
An RF modulation recognition method using machine learning
In recent few years, deep learning has been successfully applied in many fields to optimize decision making including self-driving cars, health care, machine translation, image recognition, etc. In wireless communication, deep learning has been used in channel estimation, signal classification, massive MIMOs, heterogeneous networks, energy harvesting, device-to-device (D2D) communications, and so on. In this paper, we applied machine learning (ML) and deep learning (DL) neural networks to RF signal recognition. Specifically, we built, trained, and tested two ML models, SVM and XGBoost, and two DL models, ConvNet and ResNet. We utilized the online dataset at radioml.com. Our goals are to learn how to scientifically apply ML/DL in terms of dataset processing, deep neural network constructing, training, testing, fine-tuning, and results analyzing and reporting.
- Open Access
- 2 June 2022
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 6 (6) , e28059
Direct Outreach in Bars and Clubs to Enroll Cigarette Smokers in Mobile Cessation Services: Exploratory Study
Background: Cigarette smoking and alcohol use are well known to be concomitant behaviors, but there is a lack of studies related to recruitment of smokers for mobile cessation services at places where alcohol is consumed, such as bars and clubs. Adapting recruitment strategies to expand the reach of cessation programs to where tobacco users are located may help decrease the health-equity gap in tobacco control by improving reach and enrollment of underserved smokers residing in low-income and rural areas who are not reached by traditional cessation services. Objective: The purpose of this exploratory study was to assess the feasibility of direct outreach in bars, clubs, and restaurants to recruit smokers to Quitxt, our mobile smoking cessation service. Quitxt is delivered through SMS text messaging or Facebook Messenger. Methods: We collaborated with an advertising agency to conduct in-person recruitment of young adult smokers aged 18-29 years, focusing on urban and rural Spanish-speaking Latino participants, as well as English-speaking rural White and African American participants. Street team members were recruited and trained in a 4-hour session, including a brief introduction to the public health impacts of cigarette smoking and the aims of the project. The street teams made direct, face-to-face contact with smokers in and near smoking areas at 25 bars, clubs, and other venues frequented by young smokers in urban San Antonio and nearby rural areas. Results: The 3923 interactions by the street teams produced 335 (8.5%) program enrollments. Most participants were English speakers with a mean age of 29.2 (SD 10.6) years and smoked a mean of 8.5 (SD 6.2) cigarettes per day. Among users who responded to questions on gender and ethnicity, 66% (70/106) were women and 56% (60/107) were Hispanic/Latino. Among users ready to make a quit attempt, 22% (17/77) reported 1 tobacco-free day and 16% (10/62) reported maintaining cessation to achieve 1 week without smoking. The response rate to later follow-up questions was low. Conclusions: Direct outreach in bars and clubs is a useful method for connecting young adult cigarette smokers with mobile cessation services. However, further research is needed to learn more about how mobile services can influence long-term smoking cessation among those recruited through direct outreach, as well as to test the use of incentives in obtaining more useful response rates.
- 2 June 2022
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 6 (6) , e28059
Open AccessDirect Outreach in Bars and Clubs to Enroll Cigarette Smokers in Mobile Cessation Services: Exploratory Study
Background: Cigarette smoking and alcohol use are well known to be concomitant behaviors, but there is a lack of studies related to recruitment of smokers for mobile cessation services at places where alcohol is consumed, such as bars and clubs. Adapting recruitment strategies to expand the reach of cessation programs to where tobacco users are located may help decrease the health-equity gap in tobacco control by improving reach and enrollment of underserved smokers residing in low-income and rural areas who are not reached by traditional cessation services. Objective: The purpose of this exploratory study was to assess the feasibility of direct outreach in bars, clubs, and restaurants to recruit smokers to Quitxt, our mobile smoking cessation service. Quitxt is delivered through SMS text messaging or Facebook Messenger. Methods: We collaborated with an advertising agency to conduct in-person recruitment of young adult smokers aged 18-29 years, focusing on urban and rural Spanish-speaking Latino participants, as well as English-speaking rural White and African American participants. Street team members were recruited and trained in a 4-hour session, including a brief introduction to the public health impacts of cigarette smoking and the aims of the project. The street teams made direct, face-to-face contact with smokers in and near smoking areas at 25 bars, clubs, and other venues frequented by young smokers in urban San Antonio and nearby rural areas. Results: The 3923 interactions by the street teams produced 335 (8.5%) program enrollments. Most participants were English speakers with a mean age of 29.2 (SD 10.6) years and smoked a mean of 8.5 (SD 6.2) cigarettes per day. Among users who responded to questions on gender and ethnicity, 66% (70/106) were women and 56% (60/107) were Hispanic/Latino. Among users ready to make a quit attempt, 22% (17/77) reported 1 tobacco-free day and 16% (10/62) reported maintaining cessation to achieve 1 week without smoking. The response rate to later follow-up questions was low. Conclusions: Direct outreach in bars and clubs is a useful method for connecting young adult cigarette smokers with mobile cessation services. However, further research is needed to learn more about how mobile services can influence long-term smoking cessation among those recruited through direct outreach, as well as to test the use of incentives in obtaining more useful response rates.