Kim 2018.

Methods	Design: 2‐arm, parallel group, randomised pilot trial Setting: US communities – most participants recruited from Massachusetts and New York. Recruitment: professional networks of healthcare providers who were working with people living with HIV and study adverts placed on the free website, Craigslist. Callers were screened via a brief telephone interview and those who were eligible were informed of the study purpose and procedures. Of the 94 callers screened for eligibility, 56 were eligible and 49 were randomised (88% of eligible). Inclusion criteria: women; English‐speaking ability; HIV‐positive serostatus and CD4 cell count and viral load; 18–75 years old; smokers reporting ≥ 5 cigarettes a day for the past 6 months; access to video calling, via smartphone; willingness to set a quit date within 4 weeks from the first counselling session and usage of an approved form of contraception during the study period. Exclusion criteria: unable to speak English; pregnant or lactating; self‐reported and active skin disease; serious alcohol use (≥ 26 on the Alcohol Use Disorders Identification Test); history of serious mental illness (psychotic and bipolar disorders) or illegal substance use, excluding marijuana.
Participants	Total number randomised: 49 randomised; 25 to video counselling arm, 24 to telephone counselling arm Withdrawals and exclusions: in video counselling arm, 4 excluded (3 could not install a video call app, 1 had already quit). In telephone counselling arm, 3 excluded (2 had already quit, 1 died). 42 included in analysis; 21 in video counselling arm, 21 in telephone counselling arm Sociodemographic characteristics: 100% women, mean age 51.12 years (SD 7.65), 30.95% married or living with partner, 28.57% Hispanic, 73.81% black, 57.14% had 12 years of education, 80.5% employed Smoking status and history: mean age at smoking onset 18.08 years (SD 6.82), mean years of smoking 33.14 years (SD 10.00), mean number of cigarettes per day 14.23 (SD 6.73), mean nicotine dependence score 5.57 (SD 1.84), mean self‐efficacy in quitting smoking score 23.90 (SD 8.29)
Interventions	Video counselling: 8 weekly individualised counselling sessions with duration of 10–30 minutes each delivered via IMO or other video communication software. Quit date set during the first counselling session (participant encouraged to choose a quit date between the third and fifth sessions). Counselling content based on a cognitive behavioural therapy foundation, guided by Bandura's Social Cognitive Theory. Telephone counselling: 8 weekly individualised counselling sessions with duration of 10–30 minutes delivered via telephone. Quit date set during the first counselling session (participant encouraged to choose a quit date between the third and fifth sessions). Counselling content based on a cognitive behavioural therapy foundation, guided by Bandura's Social Cognitive Theory. All participants: 8‐week supply of nicotine patches Providers: tobacco treatment specialist and a trained graduate student
Outcomes	Cessation: prolonged abstinence at 6 months postquit (self‐report at each follow‐up and biochemical validation at 3 and 6 months postquit), point prevalence abstinence at end of intervention (self‐report), 3 months (biochemically validated) and 6 months postquit (biochemically validated) Intervention adherence: 8 counselling sessions completed Satisfaction: Client Satisfaction Questionnaire
Notes	Funding: partially supported by a Joseph P. Healey Research Grant awarded by the University of Massachusetts (UMass) Boston and the UMass Boston – Dana Farber Harvard Cancer Center U54 Partnership grant. Conflicts of interest: none declared
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Random sequence generated using a computer‐generated random number.
Allocation concealment (selection bias)	Unclear risk	Random number along with the corresponding group was enclosed in a sealed envelope; however, it was unclear if the envelope was opaque.
Blinding of outcome assessment (detection bias) Smoking cessation	Low risk	Smoking cessation was biochemically verified and so low risk of detection bias.
Incomplete outcome data (attrition bias) All outcomes	High risk	The attrition rate at the 6‐month postquit follow‐up was significantly higher in the telephone counselling arm (52.4%) than in the video counselling arm (19%).
Selective reporting (reporting bias)	Low risk	Primary and secondary outcomes reported in the paper aligned to those prespecified in the protocol paper.