Update: Smoking Cessation in African American Populations

By Sydney Batchelder, 2020 Innovative Student Research Grants (Dissertation) Grantee

Progress Report

COVID-19 restrictions required a modification of some of our methods, as our original design included monitoring of participants’ carbon monoxide which requires removing a mask and close proximity. My mentor (Wendy Donlin Washington) and I made the decision to pivot the research from a treatment-based approach to an assessment-based approach. She sent an email to ABAI about this in July 2020.  

Below I outline the updated method. At this point, I have completed data collection and data analysis for the study after recruiting a total of 361 participants. I am currently writing up the discussion of the findings and plan to present some of the data as a poster at ABAI’s Annual Conference in May, 2021. This is consistent with the timeline I originally proposed, albeit a little delayed starting data collection.

The Current Study

For several reasons, Black Americans are at higher risk for smoking related illness than other ethnicities (Berg et al., 2017; Huxley et al., 2012; Inoue-Chol et al., 2017; Lortet-Tieulent et al., 2016; Manley et al., 1997; McEvoy et al., 2015). Etiology of ethnic differences may include knowledge of smoking risks (Reimer et al., 2010; Webb et al., 2007), trauma (Berg et al., 2017), socioeconomic status (Martinez et al., 2018), discrimination (Parker et al., 2016; Parker et al., 2017), metabolic differences (Ahijevych et al., 2002; Benowitz et al., 2016; Benowitz et al., 1999), cigarette preference (Gubner et al., 2018; Stahre et al., 2010), and ability to quit (Soulakova et al., 2017; Stahre et al., 2010; Trinidad et al., 2011). One way to determine effects of these variables on cigarette smoking is to analyze them within the Experimental Tobacco Marketplace.

The purpose of the present study is to examine the effects of several variables including race, menthol status, socioeconomic status, discrimination, and past quit attempts on the demand for cigarettes using the ETM. Finally, the ETM and CPT will also be used to determine whether multiple alternative nicotine products serve as substitutes for cigarettes in Black and White smokers.


Three-hundred sixty-one adult smokers were recruited via Amazon Mechanical Turk (mTurk) and paid $1.50-$3.00 for taking the survey. Participants were older than 18, lived in the United States, had a task approval rate of > 80%, and were verified smokers on mTurk.


An informed consent form detailed the procedures, risks, benefits, purpose, length, costs, and researcher and IRB contact information. An intake questionnaire also provided the participants’ demographic information, daily experience of discrimination (Everyday Discrimination Scale; Williams et al., 1997), knowledge of smoking risks (Reimer et al., 2010), cigarette preference, whether they have used NRT or nicotine alternatives in the past, the Fagerstrom Test for Nicotine Dependence (FTND; Heatherton et al., 1991), intentions to quit, and Timeline Followback (Bernstein et al., 2016; Sobell & Sobell, 1992). The Experimental Tobacco Marketplace (Bickel et al., 2018) and Cigarette Purchase Task (MacKillop et al., 2008) were then given to participants. Finally, the study employed the use of Qualtrics, Microsoft Excel, and R Statistical Software Version 3.3.2 for data collection and analysis.


Participants were recruited via mTurk to take several measures. After consenting, they answered the intake questionnaire. Finally, participants took the ETM and CPT. The order of ETM and CPT were counterbalanced across participants so that half of participants took the ETM first and half of participants took the CPT first. For the cigarette purchase task, first cigarettes were evaluated alone at all prices. Then, pairs of cigarette and one alternative commodity were assessed for all prices of cigarettes until all pairs were exhausted (i.e., cigarette/electronic cigarettes, cigarette/snus, cigarette/dip, cigarette/nicotine gum, cigarette/nicotine lozenge, cigarette/cigarillo). They were then shown an exit screen where they retrieved their code to receive payment from mTurk.

Experimental Tobacco Marketplace and Cigarette Purchase Task

The following products were available to participants in the units and at the following prices (determined by online cost): their usual brand of cigarettes, disposable electronic cigarettes ($8.00, 1 e-cig, 24mg of nicotine), snus ($6.09, 0.32oz can, 12.5mg of nicotine), dip ($6.69, 1.2oz can, 88mg of nicotine), nicotine gum ($0.57, 1 piece, 4mg of nicotine), nicotine lozenges ($0.57, 1 piece, 4mg of nicotine), and cigarillos ($0.50, 1 cigarillo, 10.8mg of nicotine). Participants made one week’s worth of nicotine-product purchases for the ETM for each of eight price conditions ($0.06, $0.12, $0.25, $0.50, $1.00, $2.00, $4.00, and $8.00 per cigarette; Bickel et al., 2018). Participants made one day’s worth of cigarette (and alternative, if appropriate) purchases for the CPT for each of 14 price conditions ($0, $0.01, $0.05, $0.13, $0.25, $0.50, $1.00, $2.00, $3.00, $4.00, $5.00, $6.00, $11.00, $35.00 per cigarette; MacKillop et al., 2008).

Use of Funds to Date

The grant money has been used to pay participants for their responses on mTurk, and for a GraphPad Prism License to analyze and graph data. Participants were recruited in four waves. The first wave consisted of 95 participants, paid $1.50. The second wave consisted of 108 participants, paid $3.00. The third wave consisted of 90 participants, paid $3.00. Finally, the fourth wave recruited 68 participants who self-identified as African American and were paid $3.00. With fees from mTurk ($0.70 per participant), total amount of money spent paying participants is $1,193.20. To meet criteria for power for my multilevel model regressions, I plan to recruit a minimum of 120 more participants on mTurk to be paid $3.00 each. This would be a minimal additional cost of $444.

Learn more about Sydney Batchelder and her project.

About Author