Public submissions on proposed amendments to the Poisons Standard

Subdivision 3D.2 of the Therapeutic Goods Regulations 1990 (the Regulations) sets out the procedure to be followed where the Secretary receives an application under section 52EAA of the Therapeutic Goods Act 1989 (the Act) to amend the current Poisons Standard and decides to refer the proposed amendment to an expert advisory committee. These include, under regulation 42ZCZK, that the Secretary publish (in a manner the Secretary considers appropriate) the proposed amendment to be referred to an expert advisory committee, the committee to which the proposed amendment will be referred, and the date of the committee meeting. The Secretary must also invite public submissions to be made to the expert advisory committee by a date mentioned in the notice as the closing date, allowing at least 20 business days after publication of the notice. Such a notice relating to the scheduling proposals initially referred to the November 2016 meeting of the Joint Advisory Committee on Medicines and Chemicals Scheduling (Joint ACMS-ACCS #14) was made available on the TGA website on 4 August 2016 and 22 September, closing on 1 September 2016 and 20 October 2016 respectively. Public submissions received on or before these closing dates were published on 2 February 2017 on the TGA website in accordance with regulation 42ZCZL.

Public submissions relating to the rescheduling of nicotine are published again here in accordance with regulation 42ZCZL of the Regulations. Also in accordance with regulation 42ZCZL, the Secretary has removed information that the Secretary considers confidential.

Under regulation 42ZCZN of the Regulations, the Secretary, after considering the advice or recommendation of the expert advisory committee, must (subject to regulation 42ZCZO) make an interim decision in relation to the proposed amendment. If the interim decision is to amend the current Poisons Standard, the Secretary must, in doing so, take into account the matters mentioned in subsection 52E(1) of the Act (including, for example, the risks and benefits of the use of a substance, and the potential for abuse of a substance) and the scheduling guidelines as set out in the Scheduling Policy Framework for Chemicals and Medicines (SPF, 2015), available on the TGA website.

Under regulation 42ZCZP of the Regulations, the Secretary must, among other things, publish (in a manner the Secretary considers appropriate) the scheduling interim decision, the reasons for that decision and the proposed date of effect (for decisions to amend the current Poisons Standard, this will be the date when it is expected that the current Poisons Standard will be amended to give effect to the decision).

Also in accordance with regulation 42ZCZP of the Regulations, the Secretary must also invite the applicants and persons who made a submission in response to the original invitation under paragraph 42ZCZK(1)(d), to make further submissions to the Secretary in relation to the interim decisions by a date mentioned in the notice as the closing date, allowing at least 10 business days after publication of the notice. Such a notice relating to the interim decisions of substances initially referred to the November 2016 meeting of the Joint Advisory Committee on Medicines and Chemicals Scheduling (Joint ACMS-ACCS #14) was made available on the TGA website on 2 February 2017 and closed on 16 February 2017.

Public submissions received on or before 16 February 2017 are published here in accordancewith regulation 42ZCZQ of the Regulations. Also in accordance with regulation 42ZCZQ, theSecretary has removed information that the Secretary considers confidential.

Part 4

Privacy statement

The Therapeutic Goods Administration (TGA) will not publish information it considers confidential, including yours/other individuals’ personal information (unless you/they have consented to publication) or commercially sensitive information. Also, the TGA will not publish information that could be considered advertising or marketing (e.g. logos or slogans associated with products), information about any alleged unlawful activity or that may be defamatory or offensive.

For general privacy information, go to https://www.tga.gov.au/privacy. The TGA is part of the Department of Health and the link includes a link to the Department’s privacy policy and contact information if you have a query or concerns about a privacy matter.

The TGA may receive submissions from the public on a proposed amendment to the Poisons Standard where there has been an invitation to the public for submissions on the proposal in accordance with the Therapeutic Goods Regulations 1990. These submissions may contain personal information of the individual making the submissions and others.

The TGA collects this information as part of its regulatory functions and may use the information to contact the individual who made the submissions if the TGA has any queries.

As set out above, the TGA is required to publish these submissions unless they contain confidential information.

If you request for your submission to be published in full, including your name and any other information about you, then the TGA will publish your personal information on its website. However, if at any point in time, you change your mind and wish for your personal information to be redacted then please contact the Scheduling Secretariat at medicines.scheduling@health.gov.au so that the pubic submissions can be updated accordingly.

Please note that the TGA cannot guarantee that updating the submissions on the TGA website will result in the removal of your personal information from the internet.

Please note that the TGA will not publish personal information about you/others without your/their consent unless authorised or required by law.

m 0 a :x,

Table 1 S1udy characteristics rela1ed 10 design of study, setting, nurmer of participants, mean age, gender, inclusion and exclusion criteria, and follow up.

Design of No.• No. male Author, year study Location participants Mean age (%) I ncluaion criteria

Randomized con110iled 1rials Adriaens, Parallel L81Nen, Belgium 50 ENDS1: 44.7 21 (43.7) Being a smoker for at least three years; 201430 RCT ENDS2: 46.0 smoking a minimum of 10 factory made

ENDS and e liquid-: 40.3 cigarettes per day and not having the intention 10 quit smoking in the near fu1ure, but willing 10 try ou1 a less unheal1hy alternative

Bullen, Parallel New Zealand 657 16 mg ENDS: 43.6 252 (38.3) Aged 18 years or older; had smoked ten or 201~ 1-36 RCT 21 mg pa1ches NRT: 40.4 more cigarettes per day for the past year;

ENNDS:43.2 wanted 1o stop smoking; and could provide consen1

Caponnetto, Parallel Catania, 11aiy 300 7.2 mg ENDS: 45.9 190 (63.3) Smoke 10 fac1ory made cigarettes per day 201~ 5 RCT 7.2 mg ENDS-+6.4 mg (cig/day) for a1 leas1 the past five years; age

ENDS: 43.9 18 70 years; in good general health; not ENNDS:42.2 currently attempting 1o quit smoking or wishing

to do so in the nex1 30 days; commit1ed 1o follow the 1rial procedures

Cohort studies Ai Delaimy, Cohort California, US 828 No1 reported 478 (47.8) Residents of California; aged 18 to 59 years 201537 who had smoked at leas1100 cigarettes during

their lifetime and are cumtnt smokers

Biener, 201s38 Cohort Dallas and 1374 Not reported 383 (55.2) Adults smokers residing in the Dallas and Indianapolis Indianapolis me1ropoiilan areas, who had been areas, US interviewed by 1eiephone and gave permission

to be re contacted Brose, Cohort Wet, based, 3991•·· ENDS 2,015 Members were invited by e mail 1o participate 2015'0-42 United Kingdom Among daily users: 45.7 (49.6) in an ontine study about smoking and who

Among non daily users: 45.2 answered a screening question about their No ENDS": 45.7 past year smoking stmus

Hajek201 ~ Cohort Europe 100 ENDS: 41.8 57 (57) All smokers joining the UK Stop Smoking No ENDS:39 Services in addition 10 the standard treatment

(weekly support and stop smoking medications includi!!Q NRT and varenicline l·

Follow-up Exclusion criteria (mon1hs}

Self reported diabe es; severe 8 allergies; asthma Or other respira1ory diseaS8J~ psychiatric problems; depend1 eon chemicals other thEm nicotine, pregnancy; breast feeding; high blood pressure; ca i:liovascuiar disease; currently !Sing any kind of smoking cessation r,erapy and prior use of an e ciga"'IJ~ Pregnant and breai!lfeeding women; 6 people using cess1F drugs or in an existing cessat~,n programme; 1hose reporting he~\rt attack, stroke, or severe angina i the previous two weeks; and ~ e with poorly con110iled medical ~orders, allergies, or other chemical dependence Symp1ornatic card~r,ascular 12 disease; symp1oma ic respiratory disease; regular ~ r chotropic medication use; currrent or past history of alcohol al~ se; use of smokeless 1obacoci or nicotine replacement therall~, and pregnancy or bre eeding

Participants who reF,rted that 1hey 12 "might use e cicj' oj~ hanged their reporting at follow ! , as they did not represent a def 'nilive g~ of useis or never us~ rs e cig and might overlap with jl'°th Anyone over 65 ye,;irs old 36

Baseline pipe or c~rir smokers, and 12 follow up pipe or ci ;iar smokers or unsure about sm

1ing status

No exclusion critena 4 weeksP

Continued

m 0 er ::i:,

c.n

Table 1 Continued

Author, year

Harrington 2015~6

Manzoli, 2015'3

Borderud, 201439

Prochaska 2014«

Vickerman 2013'5

Design of study

Cohon

Cohon

Cohon

Cohon

Cohon

Location

us

Abruzzo and Lazio region, Italy

New York, US

us

us

No.• participants Mean age

979 46.o-·

1355 ENDS only: 45.2 Tobacco cigarettes only: 44.2 Dual smoking: 44.3

1074 ENDS use+ behavioral and pharmacological treatment: 56.3

956

No ENDS+behavioral and pharmacological treatment: 55.6 39.o····

No. male (%) Inclusion criteria

525 (53.6) Hospitalized cigarette smokers at a teniary care medical center; self identified smoker who smoked at least one puff in previous 30 days; Engtish speaking and reading; over age 18 and; cognitively and physicaRy able to participate in study

75 7 (55.9) Aged between 30 and 75 yea is; smoker of e cig (inhating at least 50 puffs per week) containing nicotine since six or more months (E cig only group); smoker of at least one traditional cigarette per day si nee six or more months (traditional cigarettes only group); smoker of both electronic and traditional cigarettes (at least one per day) since six or more months (mixed Gr~)

467 (43.5) Patients with cancer referred to a tobacco cessation program who piovided data on their recent ( past 30 days) e cig use

478 (50.0)

Adult daily smokers (at least 5 cigarettes/day with serious mental mness at four psychiatric hospitals in the San Francisco Bay Area

2,758f: Used ENDS one month or 913 (36.9) Participants from six state quitlines who more:48.1 Used ENDS less than one month:45.3 No ENDS: 49.6

registered for tobacco cessation services. Adult tobacco users, consented to evaluation follow up, spoke English, pl'l)lided a vatid phone number, and completed at least one intervention call

Exclusion criteria

Pregnant

llticit drug use, bre~f.stfeeding or pregnancy, major d~ression or other psychiatric oci nditions, severe allergies, active an ihypertensive medication, angi~ pectoris, past episodes of major j ~n:liovascular diseases (myocardiial infarction, strokefTIA, congas IV8 heart failure, COPD, cancer of tliie lung, esophagus, larynx oral cavity, bladder, pancreas, idney, stomach, cervix, and myek>11 leukemia No exclusion criten J

Non English s~jng; medical contraindications NAT use (pregnancy, race myocardial infarction); and lac~i of _capacity to consent as determireo by a 3 item screener of study ~urpose, risks, and benefits No exclusion crite~ i

Follow-up (months)

6

12

6 to 12

18

7

no.: number; e-dg: ~ lgarettes; ENDS: Bectronlc nicotine delivery system; ENNDS: electronic non-nicotine delivery systems; ACT: randomized controAed trial; US: I ,nited States; ENDS1 and ENDS2: the ~lg gm ~ received the &-clg and four bottles of e-liquid al session 1 (group e-cig1 received the "Joyetech eGo-C" and group ~ ig2 received the "Kan!Jer T2-CC"); al session 2, parti~nts' empty bottles were replenished up to again four bottles and at session 3, they were aUowed to keep the remaining bottles. ·Randomized or at baseline .. For the f irst two m onths control group consisted of no a-cigarettes use. After that period, the participants of control group received the e-cig and a-liquid. ENDS 1 •J pyetech eGo-c" e-cig and ENDS2 "Kanger T2-CC" ~lg. I __ .. "The 41 17 were reported In a publication that focused on baseHne characteristics, not on the use of a-cigarettes and changes In smoking behavior, so the remainin!I 53 parti~nts are Irrelevant t o this review . .... Mean age of the o.,erall population. "The comparator comprises cl current non-users of ~lg. which included never-users and those who had previously tried but were not using at the moment. ~Hajek 2015 was the only study that entered In the review due to meet the criteria for adverse events. eBut only 2,476 asnwered the question "Have ~u ever used a-cigarettes, electronic, or vapor cigarettes?•

Open Access g non-electronic smoking ces.sation aid including NRT and alternative electronic smoking cessation aid (ENDS or ENNDS) ) . For meta-analyses, we used 6 months data or the nearest follow-up to 6 months available.

For dealing with missing data, we used complete case as our primary analysis; that is, we excluded participants with missing data. If result~ of the primary analysL~ achieved or approached statistical significance, we con­ducted sensitivity analyses to test the robustness of those results. Specifically, we conducted a plausible worst-case sensitivity analysis in which all participants with missing data from the arm of the study with the lower quit rates were assumed to have three times the quit rate as those with complete data, and those with missing data from the other arm were assumed to have the same quit rate as participants with complete data.30 31

We assessed variability in results across studies by using the 12 statistic and the p-value for the x2 test of hetero­geneity provided by Review Manager. We used Review Manager (RevMan) (V.5.3; Nordic Cochrane Centre, Cochrane) for all analyses. 32

RESULTS Study selection Figure 1 presents the process of identifying eligible studies, including publications in the last systematic

review, 8 citations identified through search in elec­tronic databases and studies identified through contact with experts in the field. Based on title and abstract screening, we assessed 69 full texts of which we included 19 publications describing three RCTs involv­ing 1007 participants25 3

3-39 and nine cohort studies

with a total of 13 115 participants.26-29

4-0-46 The inter­

observer agreement for the full-text screening was sub­stantial (kappa 0.73).

We contacted the authors of the 12 included studies, 9 of whom2

6-29 33 4 1 43 44 46 supplied us with aU requested

data; authors of further 3 studies25 42 46 did not supply the requested information (see online supplementary appendix table S2).

Study characteristics Table 1 describes study characteristics related to design of study, setting, number of participants, mean age, gender, inclusion and exclusion criteria and follow-up.

Five studies25-

28 33 42 46 were conducted largely in Europe, six in the USA, 29 40 41

43-45 and one in New Zealand.34-3

9 Randomised trials sample size ranged from 5033 to 657,34-39 and observational studies from 10046 to 3891.26-

28 Typical participants were women in their 40 s and 50 s. Studies followed participants from 4 weeks46 to 36 months.29

13 sludles retrieved rrom a Coehrane 4101cttations identified through database

I 9 studies identified I

revieW' (19 pubi cations} search limlted to years 2014 and 2015 lhrough contact Wl1h

• 2 RCT (5 pu/J/icB/lons) MEDLINE 870 CINAHL 143 experts in tile field • 2 case series (2 publjcalio,)s) EMBASE 990 CENTRAL 55 • 9 Obst,rtStioM/ study (12 publications) Psycinlo 497 Web of Science 486

Cllntca1trla1S.gov 159 PubMe0901 1 ovenapping a

i-reference retrieved from

I 1450 I both Coehrane revielY 1 dupncates and database search

I 2651 ti!les/abstracts I screened I 8 full-text assessed I

I 19 lull·text assessed I 1

2606 not relevant I ror eligibility

for eligibility

5 overlapping the

I references retrieved from 7 excluded studies

1 O excluded sludies the Cochrane review • 4no<anRCTo,a (13 publications) cohodswdy

• 5 not"" RCT « a cohorl

I 45 full-rext assessed

I ~ • 3 d'id l'IOl Inc/Ilda one - stvd)I (6 puM<ations) for eligibill\Y'" enn w,rl> ENDSIENNOS

• 5 did not inc/tide one onn ~ 10 et.,.,,./we MlhENOSIENNOS 33 exeluded sludies str,i,:egies <X>m()aredtoa1emative • 13 not an RCT or a oohort study

stmtegies (1 p,,r,t;cations} • 2did not include one sm, w;u, ENOSIENNDS com,wed to .5/temative. strategi,t.s

;, , ............. ' ... ~ .............. • 4 did not reporl outcomes o( In/ores! (6 publlcations) • 14 ongo/!IQ stud'-'$ No addlllonal

• 2 RCT (5 pulJ/icl,tionS) studies-• 1 COhOt1 &fully ( 1 9 ,neluded srudies (1 publication) ..... it._., __ ,

(12 publications) -1,..... RCT (31wblications"} •8 cohon studies (!I p11b/ication$)

l 1 ~ 1ne1uoeo STU01es (19 publications)

•3 RCT (8 publics/ions) •9 cohott studies (11 nubJicationsJ

Figure 1 PRISMA diagram of included studies. ·McRobbie, 2014.8 *•Further two publications from one RCT included by the Cochrane review were identified only in our search strategy. ••*Further one publication from one cohort study identified by our search strategy was identified throughout the expert search.

6 El Dib R, et al. BMJ Open 2017;7:e012680. doi:1 0.1136/bmjopen-2016-012680

m 0 er ;Q

Table2 Study characteristics r&lated to population, intervention or exposure groups, comparator, and assessed outcomes.

No.• of participants No! of participants in Description of intend to quit intervention or exposunt intervention or

Author, year Population smoking groups and comparator exposunt groups

Randomized controlled trials Adriaens, Participants unwilling to quit YesO ENDS1: 16 ENDS("Joyetech 201433 smoking (participants f10m the No SO ENDS2: 17 eGo C")

control gl'Ol4) kept on smoking Control/ENDS: 17 ENDS regular tobacco cigarettes E cigarettes ('1<anger during the first eight weeks of T2 CC") the study)

Bullen, Hsd smoked ten or mor& Yes 657 ENDS:289 16 mg nicotine ENDS 201:334-39 cigarettes per day for the past NoO NRT:295

year, interested in quitting ENNDS: 73

Caponnetto, Smokers not intending to quit YesO ENDS 1: 100 7 .2 mg nicotine ENDS 201~ 5 No300 ENDS2: 100 7.2 mg nicotine ENDS

ENNDS: 100 +5.4 mg nicotine ENDS

Cohort studies Al Delaimy, Current smokers; regardless of Yes 415 ENDS:236"' ENDS 2015' 0 whether the users were using No542 No ENDS: 392"'

ENDS as part of a quit attempt

Biener, All respondents had r&ported Yes 354P 1374$ ENDS£ intemnittent use 201529 being cigarette smokers at No 331 £ ENDS£ intensive use

baseline; regardless of whether the users were using ENDS as part of a quit attempt

Brose, Current smokers; regardless of Not r&ported ENDS: 1507 ENDS daily 20152 6-28 whether the users were using No ENDS: 2610 ENDS non daily

ENDS as part of a quit attempt

Hajek, 201546 69% (n=69) accepted e cigs as Not r&ported ENDS:69 ENDS was offer&d to aD part of their smoking cessation No ENDS: 31 smokers in addition to treatment the standard treatment

(weekly 514)port and stop smoking medications induding NAT and varenicline) ENDS

Description of Definition of quitlers or comparators Measuntd outcomes abstinence

ENDS and Quitting, defined as eCO of No mor& cigarette e liquid•• 5 ppm or smaller; smoking

questionnaire self repor1 bf reduction in cigarettes of>60% or complete qui Ing

2 1 mg patches Continuous smoking Abstinence allowing !>5 NAT abstinence, biochemical r cigar&ttes in total, and ENNDS verified (eCO rneasur&m pnt p10portion reporting no

<10 ppm); seven day poij,t smoking of tobacco pr&valence abstinence; cigar&ttes, not a puff, in reduction; and adverse the past 7 days events

ENNDS Self r&port of reduction i , Complete self reported cigar&ttes of>S0%; abstinence f10m tobacco abstinence f10m smokin", smoking not even a def1ned as complete puff self reported abstinence

from tobacco smoking r t even a puff, biochemical verified (eCO rneasur&n mt !>7 ppm); and adverse events

No ENDS Quit attempts; 20% redu ::lion Duration of abstinence of in monthly no. of cigarett~s; one month or longer to and current abstinence from be currently abstinent cigar&tte use

No ENDS (used Smoking cessation; and Smoking cessation was once or twice reduction in motivation tr defined as abstinence ENDS) quit smoking among tho ie flt>ffl cigarettes for at

who hsd not quit, not least one month otherwise specified

No ENDS£ Quit attempts•; cessatior "'; Change from being a and substantial reductio1 smoker at baseline to defined as a r&duction b\f at being an ex smoker al least 50% flt>ffl baseline follow 14> was coded as CPD to follow 14> CPD cessation

No ENDS Self r&ported abstinencE Self r&ported abstinence was biochemically validated flt>ffl cigarettes at 4 by exhaled CO levels in weeks end expired breath usin! a cut off point on 9ppm, adveise events

No ENDS

Continued

CD

m 0 C"

:r,

Table 2 Continued

Author, year

Harrington, 2015' 5

Manzoli, 2015'2

Borderud, 2014"

Prochaska, 2014' 3

Vickennan, 2013"

No.• of participants No.• of participants in intend to quit intervention or exposu111

Population smoking groups and comparator

Hospitalized cigarette smokers. All were cigarette smokers initially; regardless of whether the users were using ENDS as part of a quit attempt Smokers of ;?:.1 tobacco cigarette/day (tobacco smokers), users of any type of e cig, inhaling ;?:.50 puffs weekly (e smokers), or smokers of both tobacco and e cig (dual smokers) Patients who presented for cancer treatment and identified as current smokers (any tobacco use within the past 30 dll'/s); regardless of whether the users were using ENDS as part of a quit attempt

Yes: 220 ...

No: not reported

Not reported

Yes 633"' No 42"'

Adult daily smokers with serious At basefine, 24% mental illness; regardless of intended to quit whether the users were using smoking in the next ENDS as part of a quit att8111)t month

ENDS: 171 No ENDS: 759

ENDS:343 Tobacco and ENDS: 319 Tobacco only: 693

ENDS:285 No ENDS: 789

ENDS: 101 No ENDS: 855

Description of intervention or exposu111 groups

ENDS Tobacco and ENDS

Description of comparators

Tobacco cigarettes only

ENDS£+Evidence based No ENDS behavioral and pharmacologic treatment

ENDS

+Evidence based behavioral and pharmacologic treatment

No ENDS

Adult tobacco current or past Not reported ENDS: 765 ENDS used for 1 month No ENDS (never users; regaldless of whether the No ENDS: 1,711 or more tried) users were using ENDS as part ENDS used for less than of a quit attempt 1 month

Meesu111d outcom es Definition of quitlers or abstinence

Quitting smoking based )11 Only self reported 30 day point prevalence at 6 quitting smoking months

Abstinence, proportion o· quitters, biochemicaDy verified ( eOO measurement> ?ppm), reduce tobacco smoking , and serious adverse events

Smoking cessation by self report

Smoking cessation by self report and, biochemically verified (C 1J and cotinine) Tobacco abstinence

Percentage of subjects reporting sustained (30 days) smoking abstinence from tobacco smoking

Patients were asked if they had smoked even a puff of a (traditional) cigarette within the last 7 days

Past 7 day tobacco abstinence

Self reported 30 day tobacco abstinence at 7 month follow up

no.: number; C: comparator group; CPD: cigarettes smoked per day; &-eig: &-Cigarettes; ENDS: Bectronlc nicotine deUvary system; ENNDS: electronic non-nicotine c ~ivery systems; eCO: exhaled breath carbon monoxkle; NE: non-exposure group; NAT: Nicotine replacement therapy. "Numbers randomized or al baseHne. **For the first two months control group consisted of no &-Cigarettes use. After that period, the participants of control group received the &-eig and e-liquid ENDS1 "J pyetech eGo-C" &-elg and ENDS2 "Kanger T2-CC" &-eig . ... Only among those who reported any previous use of e~igs. alnformalion retrieved through contact with author. €The comparator comprises of current non-users of &-eig, which included never-users and those who had prelliously tried but were not using al the moment "'Participants who wBI never use &-eig plus those who never heard of &-eig 392; partlc~nts who have used &-eig 236 (numbers taken from the Cslifomla Smokers Ei:ohort, a longitudinal survey). ~Intentions to quit smoking, those who tried &-Cigarettes only once or twice are grouped with never users ("non-usersMers;. elntermlttent use (i.e., used regularly, but not dally for more than 1 month) plus intensive use (I.e., used &-eig daHy for at least 1 month). 5No. of the whole sample Including comparator. £All ENDS. "'The other participants either quit more than a month ago but less than six months, less than a month ago, or more than six months ago. •smokers and recent ex-smokers were asked about the number cl attempts to stop they had made in the previous year. Those reporting at least one attempt and 37 respondents who did not report an attempt but had stopped smoking be- tween baseline and foDON-up were coded as having made an attempt mChange from being a smoker at baseline to being an ex-smoker al foHON-up was coded as cessation.

m 0 C"

::0

~ ~

Table 3 Mean number of conventional cigarettes used per day at baseline and the end d study*

~ Mean no. of conventional Mean no. of conventional Biochemically quitters Self- r,ported quitters cigarettes used per day at cigarettes used per day at (no. of events per no. of (no. of events per no. of

-& Author, year Groups baseline the end of study total participants) total participants) gi

"" Adriaens, 201433 ENDS 1 20.1 7.0t 3/13 4/13 =1 .... ENDS2 20.6 8.1t 3/12 3/12 .... § ControVENDS:t:,§ 16.7 7.7t 4/13 4/13

"" Bullen, 2013~ 9 ENDS 18.4 o.ni 21/241 Note ~ailable "' a, ENNDS 17.7 0.7 3/57 Note 1ailable 0

C. NAT 17.6 0.8,t 17/215 Note ~ailable 0

~ Capon nett>, 7.2mg ENDS 19.0 (14.~25.0)"* 12 (5.8-20)"*,tt Combined ENDS groups: 22/128 Note 1aHable 0

~ 201325 7.2mg ENDS 21.0 (15.~26.0)"* 14 (6-20)**,tt Note ~ailable c.,

plus5.4mg ~ 2 . ENDS 0

ENNDS 22.0 (15.~27.0)"* 12 (9-20)**,tt 4/55 Not ~~aHable "C Ct) ::::, r(, AI-Delairny, ENDS 14.1:t::t: 13.8§§ Not available

12/11~ =1 201540 ENNDS Not available 32/14 ~ Biener, 201529 ENDS 16.n),t Not available Not available Com~ ined ENDS groups: 42/ "" intermittent use 331 "' a, 0 ENDS intensive 11.1,i,i Not available Not available

use No ENDS 15.4,t,t Not available Not available 82/3E 4

Brose, 20152&-28 ENDS daily users 14.3 13.0··· Not available 7/86 ENDS non-daily 13.5 13.9*** Not available 25/263 users NoENDSttt 13.3 13.5 Not avaUable 168/1007

Hajek, 201546 ENDS Not available Not available Not applicable:t:;:t: Not e ~licable:t:;:t: No ENDS Not available Not available Not applicable;tt Not 8i licable:t:;:t:

Harringt>n, ENDS 14.1§§§ 10.3§§§ Not available 21/~~ 201546 No ENDS 11.9§§§ 9.8§§§ Not available Manzoli, 201542 ENDS only Not available 12 Not available :y,~· Tobacco 14.1 12.8 101/491 Not ~ailable

cigarettes only Dual smoking 14.9 9.3 51/232 Not ~ailable

Borderud, 201441 ENDS 13.7 12.3 Not available 25/ ba No ENDS 12.4 10.1 Not available 158/~,

Continued

...... C)

m 0 C"

:x,

Table 3 Continued

Mean no. of conventional Mean no. of conventional cigarettes used per day at cigarettes used per day at

Author, year Groups baseline the end of study

Prochaska, ENOS 17.0 10.0 201443 No ENOS 17.0 10.1 Vickennan, ENOS used for 19.4 13.5 201344 1 month or more

ENOS used for 18.9 14.0 less than 1 month No ENOS (never 18.1 12.9 tried)

*When authors p~ded data for different time points, we presented the data for the longest follow-up. ta months from start of Intervention.

Biochemically quitters II

Self- eported quitters (no. of events per no. of (no. J~f events per no. of total participants) total participants)

21/101 Not f ailable 162/855 Not ~~ailable Not available 59/2

Not available 73/~9

Not available 535/1 711

:tControl group consisted of received the ~lg and e-llQuld (six bottles) for 2 mooths at the end a session 3 (eight of the 16 participants of the control gro14> received the 'Joyatech eGo-C' and the remaining eight participants received the 'Kanger T2-CC'). §For the first 2 months control group cooslsted of no e~lgarettes use. After that period, the participants of control group received the ~lg and e-llQuld. ENDS1 'JoyEitech eGo-C' ~lg and ENDS2 'Kanger 12-CC' e-clg. 1]For those reporting smoking al least one cigarette In past 7 days. **Data shown as median and lnterquartHe. ttAt 6 months after the last laboratory session. :t:tOf the 1000 subjects, 993 responded to the question "How many conventional cigarettes smoked per day during the past 30 days". §§Of the 1000 subjects, 881 responded to the question "How many cigarettes smoked per day during the past 30 days.• 1]1]Nurnber of conventional cigarettes used In the prior month al baseUne. ***No. of cigarette per week divided by 7 days. tttThe comparator comprises of current non-users a e-dg which Included never-users and those who had previously tried but were not using al the moment tt:tNot applicable because they followed participants only for 4 weeks, but the study reported adverse events at 1 week or longer. §§§Data for baseUne current e-dg users. ~lg. eletronlc cigarettes; ENDS, electronic nicotine delivery system; ENDS1 and ENDS 2, the ~lg grol4)S received the e-clg and four bot11es of e-llquld al sessloo · (group ~lg1 received the 'Joyatech eGo-C' and group ~lg2 received the 'Kanger T2-CC'); al session 2; ENNDS, electronic non-nicotine delivery systems; No., number; AYO, roll your C1Nn (k fose tobacco) cigarettes.

m C C"

:x,

~ ~

~ -& gi N

=1 ..... ..... § N 0) 0, 0

C. 0

~

0

~

w ~ 2 . 0 -0 ct) ::::, ,(, =1 0)

~ N 0) 0, 0

.... ....

0 ,, g. - · CD (Q .... ~ ~ (D

ii! w

I :xi

)> C.

CIJ ::::,. 5. Q)

(t) m ::, ::, (/)

"' "' ~ =1 CD (/) w ~

~ " 9. • • C' 0) (/) • • 0 .... :xi • • () -I (/) • • 8 3 • • -0 ~ • • ::i (0

m • • z 0 (/) • • < CD ci! C • •

Was the randomization sequence adequately generated?

Was allocation adequately concealed?

Was there blinding of participants?

Was there blinding of caregivers?

Was there blinding of data collectors?

Was there blinding of statistician?

Was there blinding of outcome assessors?

Was loss to follow-up (missing outcome data) infrequent?

Are reports of the studyrree or suggestion or selective outcome reporting?

m ,, Z ea' z c 0 .. (/) (D • I\)

:xi ()

~ 8 3 j .... ::i (0

m z 0 (/)

< CD ci!

(') ., "O 0 ::, IJJ ::, 5. (t) ro "' 0 ::,

"' "' =1 =1 w w

• • • • • • • • • • • • • • • • • • (/) • • Was the study apparentlyrree or other problems that could put it al a risk of bias? C (/) • •

Table 4 Risk of bias assessment for the randomised controAed trials

Was the randomisation Was Was there sequence allocation Was there Was there blinding of Was there adequately adequately bllndlng of bllndlng of data blinding of

Author, year generated? concealed? participants? caregivers? collectors? statistician?

Randomised controlled trials- assessing ENDS vs ENNDS Bul an, 201 ~ 9

Definitely yes Definitely yes Definitely yes Definitely yes Definitely yes Defiritely yes

Caponnetto, 201:325

Definitely yes Definitely yes Definitely yes Definitely yes Definitely yes Definitely yes

Randomised controlled trials- assesslng ENDS vs other qtitting mechanisms Adriaens, 201433

Definitely yes Probably no Probably no Probably no Probably no Probably no

Bunen, Definitely yes Definitely yes Definitely no Definitely no Probably yes Probably yes 201 ~

AU answers as: definitely yes (I~ risk of bias), prooably yes, probably no, definitely no (high risk of bias).

was the randomization sequence adequately g,enerated?

Was allocation adequately concealed?

Was there blinding of participants?

was there blinding or caregivers?

Was there blinding of data collectors?

Was there blinding of statistician?

was there blinding of outcome assessors?

Was loss to follow-up (missing outcome data) infrequent?

Are reports of the study free of suggestion of sellective outcome reporting?

Was the study apparently free of other problems that could put it at a risk of bias?

Are reptrof Was the study Was loss to the st~ free of apparently free

Was there follow-up suggest n of of other bllndlng of (missing aelectlw problems that outcome outcome data) outconJ could put It al a assessors? Infrequent?* reportln ~? risk of bias?

Definitely yes Definitely no Deflnltel• ' yas Definitely yes

Definitely yes Definitely no Definite!• yes Definitely yes

Probably no Definitely no ProbablY, yes Probably yes

Definitely yes Definitely no Deflnltel•

yas Definitely yes

*Defined as less than 10% loss to outcome data or difference between grol4)S less than 5% and those excluded are na likely to have made a material difference In I 1e effect observed. ENDS, electronic nicotine deHYery systems; ENNDS, electronic non-nicotine delivery systems .

Open Access &

AI- Delaimy 20 15

Biener 201S

Borderud 2014

Brose 20 1S

Hajek 201 S

Harrington 2015

Manzoli 2015

Prochaska 20 14

Vickerman 20 13

,.., C .Q ] ::, a. 0 a. ., E Al

"' ., -5 E 0 ,e C ~ Al ...

"O

"' t:: 0

J:: 0 V

"O ., "' 0 C. X ., I

C 0 C

"O C n!

"O ., "' 0 C. X ., .... 0 C: 0 z; V ., ~ "' "' n! 3:

• • • • • • • • •

,... ~ ::, "' 0 C. X ., 0 c ., E "' "' ., "' "' .. ., -s .: c .,

"O .;: C 0 u ..

..0 ., :: C:

"' u

• • • • • • • • •

,.., "' ., li .~ ~

.!,/

Cl C 01 0

,.., a ::'.!

g '- u ,.. . 0 ,f! ,. ... ~ "O "' u

3 ::, ., tr "' ... 0 0 "' C

"' C,

~ ·;;; 0 .:?:: C.

"' .,

0 ;;; la i: -;;;

., u ., u C .,

"' "' "' ., ..0

C. § "' 0 "' 0 ., C

-5 .,

"' u

"' C ~ "O .,

'ii "' :;; ., .. C. ~ 0 ., ~ "O

.s ., -5 "' 0 0 ... C. 0 ~ ., c

E C ., ::, E 0

u "O "' ;; C "' ., 0 "' "' ., "O "' ., "' -5 "' ., :;; 0 -s C.

-5 X .!: ., i: .s:: i: u ., .; Ill

~ :g E C C 0 ~ 0 u u ., ::, ..

..0 ;;; ..0 .. ., .. :: -5 :: la "O la u cS u

• • • • • • • • • • • • • • • • • • • • • • • • • • • Figure 4 Risk of bias for cohort studies.

..... ., E 0 u ,.., ;; "' C. 0 ::,

0 0 15, ,.. . c ., C ., :;,; OJ

E OJ ::, ! "' CT

"' ., ., ., "O

"' "' ..0

"' .. "' "' ~ .. t:: .E -s 0

.s::. ~ C 0

u "' c ... C 0 .g .. C. "O C

.;: ::, .. C: :: i! 0 ..Q OJ u i: ., :§

..0 ' .. .. 0

:: -5 u ., C: :Q ... "' OJ u 3: 3:

• • • • • • • • • • • • • • • • • • • • • • • • • • • Table 2 describes study characteristics related to popu­

lation, intervention or exposure groups, comparator and assessed outcomes.

Of the three RCTs, one compared ENDS with NRT and ENNDS,34-39 another compared different concen­trations of ENDS with ENNDS25 and the third compared different types of ENDS.33 Only the Borderud study41

included participants who were also currently receiving

12

other behavioural and pharmacologic treatment. The participants from Vickerman 201344 study were all enrolled in a state quitline programme that provided behavioural treatment and in some cases NRT. All nine cohort studies26-29 34-46 compared ENDS with no use of ENDS2

6-29 40 41 or tobacco cigarettes only;42 in one,4 1

exposure and non-exposure groups received behavioural and other pharmacologic treatment.

Table 3 describes the mean number of conventional cigarettes used per day at baseline and the end of study.

The mean number at baseline ranged from 11.9 in the no ENDS group45 to 20.6 in the ENDS group.33 In only two studies2

6-28 45 the mean number of conven­

tional cigarettes used per day presented a reduction from the baseline to the end of study in the ENDS group compared with the no ENDS groups, mainly in the daily users.26-

28 No included study addressed users of combustible tobacco products other than cigarettes.

Online supplementary appendix table S3 presents the types of e-cigarettes used in the included studies. The three RCTs25 33 3

6-39 evaluated only ENDS-type cigalikes.

A total of 23.7% of the participants from Brose 201526-

28

study used tank and in the Hajek 201546 study partici­pants used either cigalike or tank. The remaining studies did not report the type of ENDS used.

Risk of bias Figures 2 and 3, and table 4, describe the risk of bias assessment for the RCTs.

The major issue regarding risk of bias in the RCTs of ENDS versus ENNDS was the extent of missing outcome data. 25 34-39 RCTs comparing ENDS with other nicotine replacement therapies had additional problems of con­cealment of randomisation33 and blinding.33-

39

Figure 4 and table 5 describe the risk of bias assess­ment of the cohort studies .

Seven26-29 40-42 44 45 of nine cohort studies were rated as high risk of bias for limitations in matching exposed and unexposed groups or adjusting analysis for progno­sis variables; confidence in the assessment of the pres­ence or absence of prognostic factors; confidence in the assessment of outcome and similarity of cointerventions between groups; all studies suffered from high risk of bias for missing outcome data .

Outcomes The mean number of conventional cigarettes/ tobacco products used per day at the end of the studies ranged from 0.734-39 in ENDS and ENNDS groups to 13.926-28

among non-daily users of ENDS (table 3). The three RCTs25 3s...39 and one cohort study42 biochemically con­firmed nicotine abstinence while the others presented only self-reported data2

6-29 40 41 4345 (table 3).

Tobacco cessation smoking Synthesised results from RCTs Results from two RCTs25 34-39 suggest a possible increase in smoking cessation with ENDS in comparison with

El Dib R, et al. BMJ Open 2017;7:e012680. doi:10.1136/bmjopen-2016-012680

m 0 C"

::0

~ ~

~ -& gi

"" =1 ..... ..... § "" a, a, 0

C. 0

~

0

~

w

~ 2 . 0 -0 Ct) ::::, r(, =1

~ "" a, a, 0

..... w

Table 5 Risk of bias assessment of the cohort studies

Did the study match exposed and

Can we be unexposed for all confident that variables that are Can we be

Was selection of the outcome associated with the confident In the exposed and Can we be of Interest outcome of Interest assessment of Can we be non~xposed confident In was not or did the statistical the presence or confident In Was the cohorts drawn the present at analysis adjust for absence of the follow-up of

Author, from the same assessment of start of these prognostic prognostic assessment of cohorts year population?* exposure?t study?* variables?§ factors?11 outcome?** adequate?~it

AI-Delaimy Definitely yes Probably yes Definitely yes Definitely no Definitely no Definitely no Definitely nc I 201 540

Biener Definitely yes Probably yes Definitely yes Definitely no Definitely no Definitely no Definitely nc 201 529

Brose 201 526-28

Definitely yes Probably yes Probably no Definitely no Definitely no Definitely no Definitely nc

Hajek Probably yes Probably yes Probably yes Definitely no Probably yes Probably yes Probably ye· 201546

Harrington Definitely yes Definitely no Definitely no Definitely no Definitely no Definitely no Definitely nc 201 546

Manzoli 201542

Definitely yes Probably yes Definitely no Definitely no Definitely no Probably no Definitely nc

Borderud Definitely yes Probably yes Definitely yes Definitely no Definitely no Definitely no Definitely nc 201441

Prochaska Definitely yes Probably yes Definitely yes Definitely yes Probably yes Definitely no Definitely yE ~ 201443

Vickerman Probably yes Definitely no Definitely no Definitely no Definitely no Definitely no Definitely nc 201344

AU answers as: definitely yes (lc,,y risk of bias), pn:t>ably yes, probably no, definitely no (high risk of bias). *Examples of low risk of bias: Exposed and unexposed drawn from same administrative data base of patients presenting at same points of cam rNer the same time flame. t Thls means that investigators accurately assess the use cJ ENDS at baseline. :!:This means that smoking cessation was not present at the start cJ the study. §Examples of lc,,y risk cJ bias: compiehensive matching or adjustment for an plausible prognostic variables.

Were colnterventlons similar between groups?**

Probably no

Probably no

Probably no

Probably no

Definitely no

Probably no

Definitely yes

Probably No

Definitely no

,iExamples of low risk of bias: Interview of an participants; self-<:0mpleted survey from all participants; review of charts with reprO<iJclblDty demonstrated; from data bel$8 with documentation cJ accuracy of abstraction of progiostlc data. I **Outcome self-reported was consldel8d as definitely no for adequate assessment. Smoking abstinence, biochemically verified was consldel8d as definitely yes for a~ate assessment ttDefined as less than 10% loss to outcome data or subjects lost to foRc,,y-up unHkely to Introduce bias. I ttExamples of lc,,y risk of bias: Most or all relevant colnterventlons that might Influence the outcome of Interest are documented to be similar In the exposed and unej~·