A Digital Real-Time Support for Stress-Induced Smoking and Unhealthy Eating Among Gender and Sexual Minorities in Dhaka
试验速览
- 阶段
- 不适用
- 状态
- 尚未招募
- 入组人数
- 115
- 试验地点
- 1
- 主要终点
- Target-behavior momentary urge intensity
研究概览
简要总结
Gender and sexual minority adults in Bangladesh experience high levels of stigma-related stress. Acute stress can trigger strong, affect-driven urges to smoke, vape, or engage in stress-attributed eating. These urges can rise and fall within minutes to hours, operating outside the reach of conventional clinic-based interventions delivered days or weeks later.
This pilot micro-randomized trial (MRT) evaluates whether delivering a brief, culturally adapted support message via smartphone during moments of high self-reported stress reduces the intensity of the urge for the participant's pre-specified target behaviour, relative to no message. Over a 24-day period, 115 participants will be scheduled for up to 5 ecological momentary assessments (EMA) per day. Whenever a participant reports an acute stress level of 5 or higher (0 to 10 scale) and satisfies pre-specified privacy, safety, behavioural, and spacing criteria, the app randomizes with probability 0.5 to deliver either a brief support message or no message. By contributing multiple randomized decision points over time, each participant serves as their own micro-control.
The primary outcome is the intensity of the urge for the participant's pre-specified target behaviour, measured 45 to 60 minutes after randomization. An exploratory urge assessment is also collected 10 minutes after randomization; comparing the two time points will help establish whether the support message's effect is immediate, delayed, or both, and will inform the choice of follow-up window for a future, fully powered trial. Prior to trial launch, all EMA items, message content, notification logic, and digital safety protocols will be developed and tested with community members, reviewed by a Community Advisory Board, and locked.
As a pilot optimization trial, its objectives are to estimate the short-term causal effect of the prompt, to identify the contexts in which that effect is larger or smaller, and to establish whether momentary assessment and just-in-time intervention research can be delivered safely, privately, and acceptably in this high-stigma setting. The study does not evaluate long-term behavioural maintenance and does not claim to address the underlying structural stigma and discrimination that produce minority stress.
详细描述
Registered Study and Formative Phase: The registered interventional study is a micro-randomized trial (MRT) involving 115 participants over a 24-day assessment period. It is preceded by a formative cultural adaptation and technical refinement phase conducted to develop, optimize, and lock the momentary assessment items, support message library, notification wording, and digital safety features. This preliminary phase comprises cognitive walkthroughs and think-aloud interviews with approximately 10 community members, a short structured momentary assessment run to characterize the naturalistic distribution of momentary stress ratings, and a technology validation period. The formative phase involves no randomization and no assignment to any intervention. It is not part of the registered interventional cohort, is not included in the registered enrollment figure, and does not determine the study start date.
Micro-Randomization Framework: Randomization occurs repeatedly within each participant rather than once between participants. Each participant is randomized at every decision point during the 24-day period that satisfies all predefined eligibility criteria, thereby contributing both intervention and control observations over time. The primary unit of randomization and proximal observation is the eligible decision point; repeated observations are nested within participants.
Decision Point Definition: Randomization occurs at available high-stress decision points, defined as a completed momentary assessment at which all five of the following conditions are satisfied simultaneously: the assessment is completed within the allowed 30-minute response window; the participant reports a current stress level of 5 or higher on a 0 to 10 scale; the participant confirms that it is currently safe and private enough to receive a message; the participant has not already reported performing their designated primary target behavior since the preceding completed assessment; and no randomization has occurred in the preceding 90 minutes. Moments that fail any criterion are logged by the platform for system audit and feasibility monitoring but are not randomized.
Randomization Procedure: At each available decision point, the digital platform randomizes prompt assignment independently, with a fixed probability of p = 0.50, to either delivery of a supportive coping prompt or no prompt (active assessment control). The randomization outcome and timestamp are securely logged.
Assessment Schedule and Conflict Resolution: Participants receive five scheduled baseline momentary assessments per day, delivered at randomized times within five pre-specified daily time blocks, with one assessment per block across the 24-day period. Following every randomization, in both the prompt and no-prompt conditions, participants complete a brief micro-check assessment at 10 minutes, with a 5-minute response window, and a main proximal follow-up assessment at 45 minutes, with a 15-minute response window. To prevent assessment overload and survey collision, if a regularly scheduled baseline assessment falls within 60 minutes after a randomization, that scheduled assessment is automatically suppressed by the platform and recorded as protocol-suppressed rather than missing.
研究设计
- 研究类型
- Interventional
- 分配方式
- Na
- 干预模型
- Single Group
- 主要目的
- Treatment
- 盲法
- None
入排标准
- 年龄范围
- 18 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 否
入选标准
- •Age 18 years or older
- •Self-identifies as a gender and/or sexual minority person
- •Resides in Dhaka at enrolment and expects to remain in Dhaka throughout the 24-day assessment period
- •Reports at least one qualifying behaviour on average at least 3 days per week during the past month: (a) smoking or vaping, or (b) eating primarily in response to stress, anxiety, or emotional distress
- •Owns a personal Android or iOS smartphone compatible with the study platform that is used solely by the participant and is not shared with others
- •Able to read, understand, and respond to simple Bangla text
- •Able to operate the study app independently after a structured onboarding session
- •Willing to complete five scheduled app-based momentary assessments per day for 24 days and additional brief follow-up assessments after randomized decision points
- •Provides written informed consent
排除标准
- •Currently participating in another behavioural or pharmacological intervention study targeting smoking, vaping, or eating behaviour
- •Has a severe or unstable psychiatric condition that, based on screening and review by the study clinician, would make participation in the repeated daily assessment protocol unsafe
- •Is unable to operate the study app independently after structured onboarding
研究组 & 干预措施
Micro-randomized JITAI cohort
All participants enter a single 24-day micro-randomized cohort. At each eligible high-stress decision point, the platform independently randomizes with probability 0.50 between two momentary options.
Support prompt: The app immediately delivers one brief, culturally adapted coping-support message from a locked library. The message is matched to the participant's prespecified target behavior, can be read in under one minute, and does not disclose identity or community membership.
No prompt: No immediate coping-support message is delivered. The 10-minute micro-check and 45-minute follow-up are delivered on the same schedule in both options. No existing care is withheld, and participants retain access to study contact and distress-referral information.
The contrast is immediate support prompt versus no immediate support prompt. Thus, the estimated proximal effect combines receiving the message with its coping content; these components are not estimated separately.
干预措施: Just-in-Time Adaptive Support Message (Behavioral)
结局指标
主要结局
Target-behavior momentary urge intensity
时间窗: At the primary follow-up delivered 45 minutes after each randomized decision point, with responses accepted through 60 minutes after randomization; assessed repeatedly throughout the 24-day assessment period.
Self-reported current urge intensity for the participant's locked target behavior, rated on an 11-point numeric rating scale from 0 to 10. For participants targeting smoking or vaping, the outcome is the urge to smoke or vape. For participants targeting stress-attributed eating, the outcome is the urge to eat in response to stress or emotional distress. Only the rating corresponding to the participant's locked target behavior contributes to the primary outcome. Scores are not combined into a within-assessment composite. A score of 0 indicates no urge at all and 10 indicates the strongest urge imaginable; higher scores indicate a worse outcome.
次要结局
- Smoking or vaping episode following randomization(From each randomized decision point through the primary follow-up delivered 45 minutes later, with responses accepted through 60 minutes after randomization; assessed repeatedly throughout the 24-day assessment period.)
- Stress-attributed eating episode following randomization(From each randomized decision point through the primary follow-up delivered 45 minutes later, with responses accepted through 60 minutes after randomization; assessed repeatedly throughout the 24-day assessment period.)
- Momentary assessment completion rate(Across the 24-day assessment period.)
- Participant retention through day 24(Across the 24-day assessment period.)