Computer-Assisted Intervention for Smoking During Pregnancy

Participants \& Procedure

Women will be approached while in the waiting room of various clinics at the Detroit Medical Center. Clinic staff will inform patients of the study and its goals via a flyer that will be distributed at the reception window as women arrive for their appointments. If the women are interested, they will answer some questions and provide their contact information on the back of the flyer. Then they will place the flyer in a ballot box that is locked and only can be accessed by the Research Staff. Research staff may also be at the clinic to recruit women from the waiting room. When the research staff meets with potential participants. The study will be briefly described, and those indicating tentative interest will be screened for age and language ability (informally). Those indicating tentative interest, reporting age of at least 18 years, gestation 27 weeks or less, and demonstrating adequate ability to speak and understand spoken English will be brought to a nearby private room.

First, informed consent for screening will be obtained using an information sheet; this sheet will describe the purpose of the screening, and will emphasize that participation in the initial screening is anonymous. Those who agree will be given a very brief demonstration of the use of the computer, after which they will complete the brief (10-item) screener. All interactions with the computer will involve a touch-screen and headphones for private presentation of all information aurally as well as visually; in many previous studies, including pilot testing at this clinic, the software has been proven extremely easy to use, even for women lacking any prior computer experience. Those who endorse smoking cigarettes (even a puff) in the past week will be told of the full study in detail, and written informed consent will be sought (with a brief quiz to assure understanding). Those who provide consent will again interact directly with the computer.

Intervention

The computer itself will perform randomization and will gather all research data except for baseline breath CO (which is higher among persons who smoke), which will be obtained (via a Micro Smokerlyzer® from Bedfont Scientific Ltd.) and entered into the computer by the research staff. The entire process, including consent, will take approximately 40 minutes regardless of experimental condition (to which the research staff will be blind). Research staff will tell clinic staff of the patient's whereabouts so that the patient is not passed over when staff call her to an exam room; unfinished study procedures will be completed in the exam room prior to the physician's entrance, and/or after the appointment is completed.

Women (N = 110) who report smoking cigarettes (even a puff) during pregnancy will be randomly assigned to computer-based brief intervention alone, CM alone, brief intervention plus CM, or attention control conditions. Women randomly assigned to the computer-based intervention condition will receive an approximately 20-minute intervention modeled after motivational interviewing (MI).

Women randomly assigned to the CM condition will be told that, if they choose, they can ask to be tested (via provision of a urine sample) for smoking, up to five times, at a regularly scheduled prenatal visit. Research staff will conduct these tests.

8-Week Follow-Up

Women who enroll in the study will be contacted via mail and telephone, as necessary, to schedule a 8-week follow-up at the PI's lab or at the prenatal clinic, depending upon the participant's preference. Research staff completing follow-up testing will be kept blind to intervention condition. The primary outcome measure will be urinary cotinine-verified abstinence at follow-up; birth outcome data will also be gathered from hospital records. At follow-up, participants will (a) complete all measures on the computer again, (b) provide breath CO; (c) provide a final urine sample for testing of urinary cotinine (evidence of smoking).

• Computer-delivered brief intervention (5As) (Behavioral), also known as Brief intervention, 5As, Brief motivational intervention: After completing the brief assessment battery, participants will interact with the computer for approximately 20 minutes, with structure being based on the Five A model and Motivational Interviewing.
• Contingency Management (Behavioral), also known as Voucher-based reinforcement: Participants randomized by the computer into this condition will view a 20-minute music and tv videos after completing the brief assessment. The research assistant will then briefly describe the CM process, with a period for questions to assure understanding. The CM condition will involve participant-initiated submission of urine samples at prenatal visits. Clinic staff will have no responsibility for the CM component other than calling research staff when a participant wishes to submit a sample. Clinic staff will not schedule any new, additional, or unnecessary prenatal visits.

Primary outcomes

• Smoking Behavior (Self-report Confirmed by Expired Breath CO) [8 week follow up]: All participants were analyzed as assigned. Total N was determined primarily by resource availability in this Stage Ib trial. Number represents number of abstinent participants in each condition.
• Urinary Cotinine [8 week follow-up]: Cotinine, as measured by urine sample taken and follow-up. Measured as number of participants abstinent by cotinine analysis.

Sex: Female, ages: 18 Years to any

Inclusion Criteria:

* at least 18 years
* pregnant, gestation <= 27 weeks
* cigarette smoking in the past week (even if only one puff)

Exclusion Criteria:

* inability to speak or understand spoken English
* do not intend to carry pregnancy to full term
* frank psychosis

Participant flow

Outcome results

• Smoking Behavior (Self-report Confirmed by Expired Breath CO) (Primary), Number, participants: Control 2; Computerized Brief Intervention (5As) 7; Contingency Management Alone 2; Combined Brief Intervention and CM 5
• Urinary Cotinine (Primary), Number, participants: Control 4; Computerized Brief Intervention (5As) 10; Contingency Management Alone 3; Combined Brief Intervention and CM 4

Adverse events

Limitations and caveats: Although all effects favored brief computer session, the small sample size limited our ability to detect small effects. The sample was almost exclusively low-income African-American women. Follow-up period was also, of necessity, relatively brief.

• Results publication: Ondersma SJ, Svikis DS, Lam PK, Connors-Burge VS, Ledgerwood DM, Hopper JA. A randomized trial of computer-delivered brief intervention and low-intensity contingency management for smoking during pregnancy. Nicotine Tob Res. 2012 Mar;14(3):351-60. doi: 10.1093/ntr/ntr221. Epub 2011 Dec 9. PMID 22157229

• Steven J Ondersma, PhD, Principal investigator, Wayne State University

NIH grants: R21DA021668