Trial Outcomes & Findings for A Study of Hookah Café Customers in the Atlanta Area (NCT NCT05916170)

Results Overview

Change in eCO from baseline (before entry into the hookah café) to follow-up (upon exit from the hookah café) - eCO will be measured with the Bedfont Micro+ Smokerlyzer. Higher scores represent higher concentrations of Carbon Monoxide (CO), which is correlated with greater waterpipe use.

Recruitment status

COMPLETED

Study phase

NA

Target enrollment

977 participants

Primary outcome timeframe

Baseline and after exiting the café up to 10 hours

Results posted on

2025-12-23

Participant Flow

This is a cluster randomized crossover trial in which hookah cafes started in the No Warning Control condition and then crossed over to the Health Warnings Intervention, in which they were randomized to display one of four health warning messages (Milestones). Participants were cafe patrons on the days/nights that data collection was being conducted. Participants were only permitted to participate in one instance of data collection at one hookah cafe. Some cafes only had Control period.

Unit of analysis: Hookah Cafe

Participant milestones

Participant milestones
Measure	Hookah Cafes in Atlanta, GA Each hookah café will serve as its own control and will be used for data collection for both arms of the study. Cafes will start in the no-warning control arm and then will cross over to the intervention arm, where they will display one of four text-only warning signs. Participants will only participate in one of the arms.
No Warning Control STARTED	507 27
No Warning Control COMPLETED	479 27
No Warning Control NOT COMPLETED	28 0
Health Warning Intervention STARTED	470 23
Health Warning Intervention Lung Damage Warning	148 6
Health Warning Intervention Heart Damage Warning	104 5
Health Warning Intervention CO Poison Warning	112 6
Health Warning Intervention Fetal Damage Warning	106 6
Health Warning Intervention COMPLETED	431 19
Health Warning Intervention NOT COMPLETED	39 4

Reasons for withdrawal

Reasons for withdrawal
Measure	Hookah Cafes in Atlanta, GA Each hookah café will serve as its own control and will be used for data collection for both arms of the study. Cafes will start in the no-warning control arm and then will cross over to the intervention arm, where they will display one of four text-only warning signs. Participants will only participate in one of the arms.
No Warning Control Lost to Follow-up	12
No Warning Control Withdrawal by Subject	15
No Warning Control Noncompliance with protocol	1
Health Warning Intervention Lost to Follow-up	15
Health Warning Intervention Withdrawal by Subject	10
Health Warning Intervention Noncompliance with protocol	14

Baseline Characteristics

A Study of Hookah Café Customers in the Atlanta Area

Baseline characteristics by cohort

Baseline characteristics by cohort
Measure	No Warning Control n=507 Participants No warnings posted in the hookah café No Warnings: No warnings posted in the hookah café	Health Warning Intervention n=470 Participants Cafes will display text-only warning signs Health Warnings: Text-only warnings about one health harm will be displayed in the hookah cafés where data collection is being conducted.	Total n=977 Participants Total of all reporting groups
Age, Continuous	30.9 Years STANDARD_DEVIATION 6.9 • n=68 Participants	31.5 Years STANDARD_DEVIATION 7.3 • n=4 Participants	31.2 Years STANDARD_DEVIATION 7.1 • n=219 Participants
Sex/Gender, Customized Female	295 Participants n=68 Participants	289 Participants n=4 Participants	584 Participants n=219 Participants
Sex/Gender, Customized Male	203 Participants n=68 Participants	177 Participants n=4 Participants	380 Participants n=219 Participants
Sex/Gender, Customized Unknown	9 Participants n=68 Participants	4 Participants n=4 Participants	13 Participants n=219 Participants
Ethnicity (NIH/OMB) Hispanic or Latino	35 Participants n=68 Participants	26 Participants n=4 Participants	61 Participants n=219 Participants
Ethnicity (NIH/OMB) Not Hispanic or Latino	458 Participants n=68 Participants	427 Participants n=4 Participants	885 Participants n=219 Participants
Ethnicity (NIH/OMB) Unknown or Not Reported	14 Participants n=68 Participants	17 Participants n=4 Participants	31 Participants n=219 Participants
Race (NIH/OMB) American Indian or Alaska Native	4 Participants n=68 Participants	6 Participants n=4 Participants	10 Participants n=219 Participants
Race (NIH/OMB) Asian	38 Participants n=68 Participants	28 Participants n=4 Participants	66 Participants n=219 Participants
Race (NIH/OMB) Native Hawaiian or Other Pacific Islander	4 Participants n=68 Participants	0 Participants n=4 Participants	4 Participants n=219 Participants
Race (NIH/OMB) Black or African American	389 Participants n=68 Participants	380 Participants n=4 Participants	769 Participants n=219 Participants
Race (NIH/OMB) White	12 Participants n=68 Participants	14 Participants n=4 Participants	26 Participants n=219 Participants
Race (NIH/OMB) More than one race	23 Participants n=68 Participants	19 Participants n=4 Participants	42 Participants n=219 Participants
Race (NIH/OMB) Unknown or Not Reported	37 Participants n=68 Participants	23 Participants n=4 Participants	60 Participants n=219 Participants
Region of Enrollment United States	507 participants n=68 Participants	470 participants n=4 Participants	977 participants n=219 Participants

PRIMARY outcome

Timeframe: Baseline and after exiting the café up to 10 hours

Population: Participants were included in analysis if they completed both pre- and post-cafe eCO collection in cafes that completed both the control and intervention arms of this crossover trial. Participants in each arm are unique.

Change in eCO from baseline (before entry into the hookah café) to follow-up (upon exit from the hookah café) - eCO will be measured with the Bedfont Micro+ Smokerlyzer. Higher scores represent higher concentrations of Carbon Monoxide (CO), which is correlated with greater waterpipe use.