Methods

Country:USA
Recruitment: telephone company employees
Design: RCT, no details of randomization method

Participants

36 employees and spouses (25 women and 11 men)
69% female. Av. age: 37
Smoked: average of 18 yrs and on average 30 cpd
Participation rate: not reported

Interventions

Group therapy
Three groups: 1. abrupt reduction 2. gradual reduction 3. gradual reduction with feedback
pre‐ and two post‐tests; 7 weekly meetings with goals of 50% reduction per week in abrupt group; 25% per week in gradual group; 25% per week with graphs of daily nicotine intake for gradual/feedback group.

Outcomes

Self report of smoking status and consumption at 6m, with CO validation and cigarette butt weight.

Type of intervention

1. Intensive behavioural: GROUPS

Notes

Analyses were conducted on non‐abstinent subjects at end of treatment, to assess reduction efficacy.
Outcomes included changes in nicotine content (brand smoked), amount of cigarette smoked, and number of cigarettes smoked.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: University of Missouri employees
Evaluation: determine the effects of various amounts of hypnosis and hypnosis plus behavioural sessions
Design: RCT, no details of randomization method

Participants

63 smokers
Female: 62%
Median education: 16 yrs
Median income: US$27,000
Participation rate: not reported

Interventions

In the initial study, 48 subjects of the total (N = 63) used, were assigned to one of three treatments:
1. four hypnotherapy (HYP) sessions + booster
2. 2 HYP sessions
3. 2 HYP + 2 behavioural sessions + booster.
A follow‐up group was later recruited composed of 15 subjects who received 4 HYP + booster with less time between sessions.

Outcomes

Self‐reported cessation at 3m and 6m, with saliva thiocyanate confirmation at 3m only.

Type of intervention

1. Intensive behavioural: GROUPS

Notes

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: Employees from 4 worksites in Fargo, North Dakota and 4 in Eugene, Oregon
Design: Cluster (worksite) randomization but individuals the unit of analysis. Two (competition/nocompetition) by two (relapse prevention training/no relapse prevention training) factorial design

Participants

Participants: 136 smokers from 8 worksites. Site size ranged from 50 ‐ 380
Av. age: 38. av cpd: 28
Smoked: average 19 years
Participation rate: not reported ‐ estimated 28% across all sites

Interventions

Evaluates the incremental effectiveness of competition and relapse prevention training in the context of a multicomponent cessation programme
Multicomponent cognitive behavioural programme for 6 weekly sessions; within‐site competition with weekly feedback on a visible barometer and monetary prizes at programme completion and at 6m; relapse prevention booster sessions were held at 1m and 2m intervals following the programme.

Outcomes

Cessation at 6m
Validation: CO and saliva thiocyanate

Type of intervention

1. Intensive behavioural: GROUPS

Notes

The competition incentive was conducted within each intervention worksite, rather than between the worksites.
Other outcomes included relapse prevention, smoking reduction, nicotine levels (brands), % of cigarette smoked.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: Employees at 43 worksites, recruited prior to a 3w television smoking cessation programme.
Design: Cluster randomization by worksite, matched for size

Participants

233 smokers in 21 group discussion worksites, 192 in 22 non‐group work sites.
Groups led by trained employees
Participation rate: not reported

Interventions

All participants were given self‐help manuals by company co‐ordinators and instructed to view the televized segments
1. Twice weekly group meetings
2. Self help alone

Outcomes

Abstinence at 12m (multiple PP)
Partial validation by salivary cotinine or family/colleague report

Type of intervention

1. Intensive behavioural: GROUPS

Notes

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: Single worksite (13,000 workers, 9 employers)
Randomization: by nurses at aid stations using randomized assignment lists generated by research centre, within preference for format.

Participants

159 smokers (av. age 43, 66% male, av.cpd 25) with preference for group programme or no preference.
243 smokers with a preference for self help randomized to 3 different S‐H programmes
Groups lead by instructors trained in both programmes.
Participation rate: 11%

Interventions

Group therapy preference:
1. Multiple Component programme. 3 sessions over 3w
2. Relapse Prevention programme. 6 sessions over 6w
3. Minimal Treatment programme. Self‐help materials only. American Cancer Society's 22 page 'Quitter's Guide' 7‐day plan.
S‐H preference:
Same 3 programmes, all in manual form, with no group meetings.

Outcomes

Abstinence at 12m (single PP)
Validation: saliva cotinine <= 35ng/ml

Type of intervention

1. Intensive behavioural: GROUPS

Notes

Group programmes were held away from worksite in non‐work hours.
50% random sample of continuing smokers supplied salivary samples

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Low risk

A ‐ Adequate

Methods

Country: USA
Recruitment: Employees at 38 worksites, recruited prior to a 3w television smoking cessation programme.
Design: Cluster randomization by worksite

Participants

419 smokers who participated in the worksite programmes, 206 Group, 213 No Group conditions.
Participation rate: not reported

Interventions

1. 6 x twice‐weekly group meetings to coincide with the 3w television series, then monthly meetings for a year. Abstinent smokers and 5 of their family and 5 co‐workers entered for a lottery at the final group meeting and 12m follow up.
2. Self‐help manuals only

Outcomes

Abstinence from end of programme to 24m
Validation by saliva cotinine and co‐worker or relative confirmation.

Type of intervention

1. Intensive behavioural: GROUPS

Notes

This study featured monthly booster sessions and monetary incentives for abstainers, as a development of the design of the first De Paul study

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 8 worksites in Bloomington, Minnesota
Design: Cluster randomized trial.

Participants

Intervention worksites (I): 1885 workers,
Comparison sites (C): 1479 workers. 39% smoked at baseline in I and 31% in C worksites
Participation rate: 12% of smokers (range 8‐29% by site); 3.7% of nonsmokers participated in classes to assist quitters.

Interventions

The 3m intervention included consultation for employers on the adoption of a non‐smoking policy, training for nonsmokers to provide assistance to smokers attempting to quit, and cessation classes for smokers

Outcomes

Quit rate, self‐reported ( an attempt was made to collect saliva samples for analysis for cotinine). Baseline survey of all employees was conducted 9m before intervention, companies then randomized, then 3m intervention period, 1m and 6m after the completion of intervention.
Evaluation period: 6m

Type of intervention

1. Intensive behavioural: GROUPS

Notes

Analyses were by individuals for some outcomes, although randomization was by worksite.
The study area had been an intervention site for the Minnesota Heart Health Program, and outcomes may not be generalizable.
Other outcomes included nonsmokers' support for quit attempts, co‐worker requests not to smoke, co‐workers non‐smoking, number of quit attempts.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 61 worksites
Design: Cluster randomization by worksite

Participants

844 smokers recruited; 289 Self Help (SH), 281 Incentives (I), 283 Group (G).
Av. age 38, Av cpd 21
72% female in SH, 58% female in I, 59% female in G
Participation rate: 58% in SH, 59% in I, 55% in G

Interventions

Worksite interventions timed to coincide with a mass media intervention consisting of a week‐long smoking cessation series on TV, and a complementary newspaper supplement.
SH: Self‐help manual (ALA Freedom from Smoking in 20 days)
I: Self‐help manual and incentive payment of US$1 for each day abstinent up to US$175
G: 6 group meetings over 3w followed by 14 booster meetings over 6m. Incentive payments. Handouts from same S‐H manual. Maintenance manual (ALA A Lifetime of Freedom from Smoking)

Outcomes

Sustained abstinence at 12m
Validation: CO < 9ppm. Saliva cotinine at 6m only

Type of intervention

1. Intensive behavioural: GROUPS

Notes

Discussion section includes some cost‐benefit analysis.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: Belgium
Recruitment: workplace volunteers in 32 companies
Design: Open‐label cessation phase, then RCT for relapse prevention, using random numbers and blinded list

Participants

344 quitters, abstinent for at least 1m at end of 3m X 7 cessation programme including group therapy and NRT.
38% female, av age 39.

Interventions

1. Relapse Prevention (RP). 10 sessions inc group discussion and role play led by professional counsellor
2. RP. 10 sessions of group discussion led by former smokers.
3. No RP

Outcomes

Abstinence for 9m from start of RP programme.
Validation by expired CO < 10ppm and urinary cotinine <= 317ug/ml.(Rates for CO and self report alone also reported; higher than for doubly validated rates)

Type of intervention

1. Intensive behavioural: GROUPS

Notes

All participants for this study had achieved abstinence after a 3m group and NRT programme. This is a relapse prevention study, rather than cessation.
Other outcomes include predictors of sustained abstinence, weight gain.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Low risk

A ‐ Adequate

Methods

Country: Japan
Setting: Omihachiman city office
Design: RCT

Participants

53 volunteer smokers

Interventions

1. Intervention group received intensive education (i.e. the effect of smoking on health, the beneficial aspects of quitting smoking, how to stop smoking and how to deal with the withdrawal symptoms) for 5m, group lectures (twice) and individual counselling (three times).
2. Control group had no special treatment for 1st 5m

Outcomes

Self‐reported and validated using expired air CO concentration.

Type of intervention

1. Intensive behavioural: GROUPS

Notes

Other outcomes included predictors of cessation success.
Data were derived from abstract only

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: Germany
Setting: 4 worksites
Design: RCT, sessions allocated randomly to standard behavioural (SB) or relapse prevention (RP) programmes

Participants

79 workers, mean age 40, 58% male, mean cpd 24, mean FTND score 5.

Interventions

6x90min sessions over 8 wks, group counselling + NRT if wanted. First 2 sessions same, then:

1. SB: psycho‐educational, self‐monitoring, environmental cue control, problem‐solving, behavioural control strategies, operant conditioning, social support.

2. RP: functional analysis of high‐risk situations, planning for them, coping strategies, self‐monitoring, noting triggers.

Outcomes

Self‐reported CA and PPA at 1m and 12m.

Type of intervention

1. Intensive behavioural: GROUPS

Notes

12m non‐responders were phoned for smoking status.
SB participants were given 'rescue' RP training when necessary, so separation of interventions not guaranteed.
New for 2008 update.

Methods

Country: Turkey
Setting: Textile factory in Malatya
Design: matched controlled study, followed up at 6m

Participants

200 workers (425 smokers completed baseline questionnaire); 100 in each group, matched on age, education, working periods and amount smoked. Intervention and control groups worked different shifts. All male workforce, mean age 29.3, 81.7% married, 44.5% attended high school. Prevalence 65.9% smokers, 6.8% ex‐smokers.

Interventions

3‐wk 7‐step programme, based on stages of change model, and ALA programme.

Outcomes

Primary outcome was movement through stages of change, but 6m cessation rate also reported. PPA self‐report, no verification.

Type of intervention

1. Intensive behavioural: GROUPS

Notes

Reported as no attrition or losses to follow up.
New for 2008 update.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

High risk

N/A

Methods

Country: Belgium
Setting: 30 factories
Design: Cluster‐randomized matched pair design RCT. Randomization method not described.

Participants

Participants: 16,230 men aged 40‐59 (83.7% of eligible men)

Interventions

1. Intervention: All screened for height, weight, cholesterol, smoking, BP, ECG, personality and psychological testing. Top 20% at risk counted as the 'high risk' group, who received 6‐monthly individual physician counselling. Complete cessation was encouraged, but pipes or cigars allowed if necessary. Advice booklet also supplied. All smokers of 5 or more cpd received written advice to quit.. Environmental components included anti‐smoking posters and a factory conference on dangers of tobacco.
2. Control: a 10% sample screened at baseline were followed up; the 20% of this sample with the highest risk score were also identified as the control 'high risk' subset, to be analyzed separately. The 'Design and Methodology' paper reports that all eligible men in the control factories all received an ECG, but this is not mentioned in later reports.

Outcomes

7‐day PP at 2 yrs follow up.
5% sample of intervention group (327 men) were tested, + all of the original high‐risk group (1268). The 10% random sample control subjects were reviewed after 2 yrs, including the 20% high risk subgroup (202 men). Self report only, without biochemical verification

Type of intervention

1. Intensive behavioural: INDIVIDUAL COUNSELLING

Notes

This trial was added to the 2005 update
This is the Belgian Heart Disease Prevention Project

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: France
Recruitment: Worksite volunteers in 160 sections of an administrative organization.
Design Cluster‐randomized controlled trial. Randomization method not described

Participants

3336 men aged 25‐35 at baseline. 424 classified as at high risk of coronary disease, 868 at low risk.
Mean cpd 8.9 intervention, 10.0 control

Interventions

1. High risk intervention subjects recalled at 6m, 12m, 24m, low risk at 12m, 24m. All intervention subjects measured blood sample, weight, BP, no. of cpd. Given tailored advice on diet, alcohol and smoking at each visit.
2. Controls received no counselling or measurement between baseline and follow up

Outcomes

Abstinence/reduction at 2 yrs.
At 2yrs 568 (86%) of intervention group returned, and 529 (84%) of control group.
Validation: Blood CO

Type of intervention

2. Intensive behavioural: INDIVIDUAL COUNSELLING

Notes

This trial was added to the 2005 update
95 intervention subjects lost to follow up were heavier smokers (+4.4 cpd) vs 100 control subects lost to follow up (+0.4 cpd).

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: naval shipyard
Recruitment: Smokers identified at worksite screening (unselected)
Design: RCT, no details of method

Participants

871 male asbestos‐exposed smokers
Av cpd: 24‐26

Interventions

1.Advice from occupational physician; minimal warning, results of pulmonary function tests, leaflets
2. As group 1 plus behavioural counselling

Outcomes

Sustained abstinence at 11m
Validation: expired CO

Type of intervention

2. Intensive behavioural: INDIVIDUAL COUNSELLING

Notes

Other outcomes included stratification by lung function, reduction by continuing smokers, predictors of successful quitting and characteristics of smokers refusing to participate in the study.
Randomization ratio (method not explained) changed halfway through the study from 3:1 to 1:1.

The study found wide variation in implementation of the study procedure by physicians

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

High risk

C ‐ Inadequate

Methods

Country: Australia
Setting: 28 Sydney ambulance stations
Design: Cluster‐randomized RCT. method of randomization not described.

Participants

431 participants (88%) in 28 stations. av age 32 yrs. 128 smokers, mean cpd 17.9.

Interventions

1. Health Risk Assessment (HRA): (10 stations, 40 smokers): Measurement of BMI, % body fat, BP, cholesterol, smoking status, aerobic capacity. Feedback given, with high risk people referred to family GP. This minimal 30 minute intervention was the control group.
2. Risk Factor Education (RFE): (8 stations, 28 smokers): Same measures as HRA, + advice through manual and videos in a 50 minute session.
3. Behavioural Counselling (BC): (6 stations, 30 smokers). Same as RFE group, + up to 6 counselling sessions (averaged 3) over 10w, + staged change manual.
4. Behavioural Counselling + Incentives (BCI): (4 stations, 30 smokers).
As RFE, + manual and goal‐setting and follow‐up counselling (average 2 hrs). Also lottery draw for A$40 voucher if interim targets achieved, and final prize of A$1000 for highest achieving station at 6m.

Outcomes

Baseline, 3, 6 and 12m assessments.
PP abstinence at 12m, validated by serum cotinine.

Type of intervention

2. Intensive behavioural: INDIVIDUAL COUNSELLING/ INCENTIVES

Notes

This trial was added to the 2005 update
Fewer stations and participants were allocated to the more intensive interventions (BC and BCI) because of cost. Some contamination between conditions reported.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: Japan
Setting single factory, 542 employees
Design: RCT, allocation by random number

Participants

263 male smokers
Av. age 34, av cpd 19

Interventions

1. Physician advice, CO feedback, cessation contract, self‐help materials. follow up over 5m. Smoking Cessation Marathon during month 4
2. Delayed intervention control

Outcomes

Abstinence for > 1m at 5m
(also 12m follow up but by then control group also treated)
Validation: CO < 9ppm, plus urine test at 12m

Type of intervention

2. Intensive behavioural: INDIVIDUAL COUNSELLING

Notes

All male smokers (62.9%) were entered compulsorily into the trial. Female smokers (3.4%) were not included.
Other outcomes included smoking reduction, willingness to quit and predictors of success.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Low risk

A ‐ Adequate

Methods

Country: France
Setting: Annual health check in one large gas and electric company
Design: Cluster randomization by site physician, physician as unit of analysis

Participants

28 site physicians covering 1269 smokers and 2614 nonsmokers
Av. age: 38, 82% male
Av cpd: 14

Interventions

1. Low intensity intervention: Physician advice 5‐10 mins incl. leaflets
2. High intensity: as 1. plus quit date, moral contract, follow‐up phone call, and 2nd visit

Outcomes

Abstinence (self‐reported) for at least 6m at 1 yr follow up
Validation: CO measurement in subgroup

Type of intervention

2. Intensive behavioural: INDIVIDUAL COUNSELLING

Notes

Other outcomes included BMI and depression score

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: Japan
Setting: Occupational health clinic
Design: RCT; details of randomization not described

Participants

228 smokers, randomized to intervention (117) or control (111). Average age 39, av cpd 23; 50% had made previous quit attempts.

Interventions

Baseline questionnaire during routine health check up, with CO and urinary metabolites measured and reported back.
1. Intervention: Stage‐matched counselling (15‐20 mins) by trained nurses, + 4 follow‐up phone calls for those prepared to set a quit date.
2. Control: baseline questionnaire and usual care.

Outcomes

Continuous abstinence at 6m and 12m. Validated by CO ?

Type of intervention

2. Intensive behavioural: INDIVIDUAL COUNSELLING

Notes

This trial was added to the 2005 update
25 smokers in the intervention group set a quit date and received the follow‐up calls.
Data were derived from abstract only

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: Japan
Setting: 12 companies (500‐1000 employees each)
Design: Matched pairs controlled study (randomization intended but not feasible).

Participants

Int/Cont: 1382/1736 current smokers, 94%/97.4% M, 92.9%/95.7% blue‐collar workers, 66.1%/61.5% smoke >20 cpd. Significant differences between groups on age, gender, occupation type, cpd, controlled for in analysis.
71.6%/73.2% in precontemplation; 35.1%/34.5% had made at least one previous quit attempt.

Interventions

Intervention: 1. Posters, newsletters, website, advertising the cessation campaign and stages of change model.
2. Worksite smoking cessation programme, conducted by worksite nurse using research team materials, giving (a) 5 brochures on stages of change, (b) 4 counselling sessions + NRT if requested, (c) award to winner among abstainers. 6‐wk programme, run 5 times over 36m.
3. Advice re secondhand smoke, and designation of smoking areas.
4. Regular site visits by member of research time.
Control: Standard annual health checks.
Participation rate was 9% across all sites.

Outcomes

Sustained 6m abstinence at 12m, 24m, 36m, not biochemically verified.
Movement through stages of change, cost‐benefit analysis.

Type of intervention

2. INDIVIDUAL COUNSELLING/ ENVIRONMENTAL/ INCENTIVES

Notes

New for 2008 update (previously an excluded study)

Methods

Country: USA
Setting: Phone‐based and mailed info, targeted at members of LIUNA (construction workers union).
Design: RCT, no details given

Participants

674 workers (354 intervention and 320 control*) completed baseline survey, and 582 (188 smokers [= current or quit within last 6m]) at 6m follow up. 94% male, mean age 40, smoking prevalence I:45%, C:40%.

Interventions

3m programme to increase fruit & veg consumption and quit smoking.
Intervention: (i) Phone‐based counselling, up to 4 calls in 3m.
(ii) Mailed tailored feedback. (iii) 6 mailings of targeted info. (iv) NRT if requested.
Control: Nothing during programme, but all targeted written info at study end.

Outcomes

Self‐reported 7‐day PPA at 6m, no verification.
Increased fruit & veg consumption.

Type of intervention

2. Intensive behavioural: INDIVIDUAL COUNSELLING

Notes

Tools for Health programme; specifically targeted blue‐collar workers.
* data supplied by authors
New for 2008 update.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

'survey respondents agreeing to participate were randomly assigned to one of two conditions'

Methods

Country: USA
Setting: faculty and staff of the University of Minnesota
Design: RCT

Participants

59 volunteer smokers. Av age 36.8, female 64.5%
Participation rate: 2%

Interventions

Self‐help manual; optional education/counselling; financial contracts of US$5 to US$25 bi‐weekly. One group aimed at cessation, the other at reduction or cessation.

Outcomes

Self‐reported cessation rate immediately post‐treatment and at 6m, biochemically validated at both points (CO, SCN)

Type of intervention

3. SELF‐HELP

Notes

15,000 staff members were approached to join the study. Of 137 smokers expressing an interest in the programme, only 59 actually signed up to it.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: UK
Setting: Company A with occupational health program near London
Design: RCT: cessation motivation vs seat belt video groups

Participants

77 in videotape conditions (33 for smoking video, 44 for seatbelts video), 55 non‐participant smokers (no‐treatment control group).

Interventions

Trial was described to company as a 'health information programme', and was open to all employees, whether or not they smoked.
1. 25‐minute video 'Dying for a Fag' (DFF) plus a cessation booklet, the Heatth Education Council's 'The smoker's guide to non‐smoking'
2. 25‐minute video on seatbelt use, + a leaflet about seatbelts
3. Smokers who chose not to participate ‐ no videos or information

Outcomes

Self‐reported PP smoking cessation at 3m and 1yr with CO validation < 10 ppm

Type of intervention

3. SELF‐HELP

Notes

Although all 4 trials (a‐d) are of similar design, and are reported in a single paper, we have treated them here as four separate RCTs.
Cash incentives were offered at baseline and at 12m follow up to boost questionnaire response rates.
The authors also present a 4‐study pooled analysis, which failed to detect signficant differences in cessation rates.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: UK
Setting: Company B with occupational health program near London
Design: RCT: cessation motivation vs cessation motivation plus confidence boosting vs. political aspects of tobacco video groups

Participants

150 in videotape conditions (46, 50 and 54 in the 3 groups), + 374 non‐participant smokers

Interventions

Trial was described to company as a 'smoking education programme', and was open only to smokers.
1. 25‐minute video 'Dying for a Fag' (DFF) plus a cessation booklet, the Heatth Education Council's 'The smoker's guide to non‐smoking'
2. DFF with additional sequence to boost the confidence of those making a quit attempt (DFF+C)
3. 'Licence to Kill', on the political aspects of smoking (LTK).
4. Smokers who chose not to participate ‐ no videos or information

Outcomes

Self‐reported PP smoking cessation at 3m and 1yr with CO validation < 10 ppm

Type of intervention

3. SELF‐HELP

Notes

Cash incentives were offered at baseline and at 12m follow up to boost questionnaire response rates.
The authors also present a 4 study pooled analysis, which failed to detect significant differences in cessation rates.
Although the cessation rates appear to be significantly better in this study than in the other 3, the authors point out that follow up was around New Year, when many people try and stop anyway, and may also have been influenced by the concurrent BBC series 'So you want to stop smoking'

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: UK
Setting: Company C with occupational health program near London
Design: RCT : cessation motivation vs cessation motivation minus a gory sequence vs. advertising aspects of tobacco videotapes groups

Participants

197 in videotape conditions (56, 67 and 74 in the 3 groups) + 226 non‐participant smokers

Interventions

Trial was described to company as a 'smoking education programme', and was open only to smokers.
1. 25‐minute video 'Dying for a Fag' (DFF) plus a cessation booklet, the Heatth Education Council's 'The smoker's guide to non‐smoking'
2. DFF with graphic 'shock' sequence about diseased lungs edited out, to lower fear element (DFF‐G)
3. 'The Tobacco War', on the advertising aspects of smoking (TW).
4. Smokers who chose not to participate ‐ no videos or information

Outcomes

Self‐reported PP smoking cessation at 3m and 1yr with CO validation < 10 ppm

Type of intervention

3. SELF‐HELP

Notes

Cash incentives were offered at 12m follow up to boost questionnaire response rate.
There were no differences between the video and non‐participant groups in long‐term abstinence .
The authors also present a 4 study pooled analysis, which failed to detect signficant differences in cessation rates.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: UK
Setting: Company D with occupational health program near London
Design: RCT: cessation motivation vs another cessation motivation vs. advertising aspects of tobacco videotapes groups

Participants

179 in videotape conditions (62, 59 and 58 in 3 groups) + 360 non‐participant smokers

Interventions

Trial was described to company as a 'smoking education programme', and was open only to smokers.
1. 25‐minute video 'Dying for a Fag' (DFF) plus a cessation booklet, the Heatth Education Council's 'The smoker's guide to non‐smoking'
2. "Smoker's Luck", on a continuing smoker suffering from advanced smoking‐related disease (SL)
3. 'The Tobacco War', on the advertising aspects of smoking (TW).
4. Smokers who chose not to participate ‐ no videos or information

Outcomes

Self‐reported PP smoking cessation at 3m and 1yr with CO validation < 10 ppm

Type of intervention

3. SELF‐HELP

Notes

There were no differences between the video and non‐participants groups in long term abstinence.
Cash incentives were offered at baseline and at 12‐month follow‐up to boost questionnaire response rates.
The authors also present a 4 study pooled analysis, which failed to detect signficant differences in cessation rates.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: Veterans Administration Medical Centre employee volunteers
Design: RCT, no details of randomization method

Participants

58 smokers
57% female, Av. age 44
All participants smoked for at least 6m (validated with CO measurement)
Participation rate: not reported

Interventions

1. American Cancer Society and ALA pamphlets about smoking, a telephone hotline, and a stop‐smoking contest which gave vouchers for a draw, for each day when expired CO < 8ppm.
2. As 1, plus use of a computer to enter data on smoking behaviour and to smoke a cigarette through a filter attached to the computer; this produced an individualized nicotine fading programme

Outcomes

Abstinence at 6m
Validation: CO < 8ppm

Type of intervention

3. SELF‐HELP

Notes

Participants in the computer group had lower self efficacy scores than the contest‐only group

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: Worksite volunteers
Design: RCT, no details of randomization method

Participants

87 smokers
36% female, Av. age 38, av. cigs/day 15
Participation rate: not reported

Interventions

1. The Last Draw, an internet‐based interactive programme to aid preparation, quitting and relapse prevention, plus FadeAid, an aid to nicotine fading
2. ALA Freedom from Smoking booklet, 2 manuals and an audio relaxation tape

Outcomes

Abstinence at 6m (7day PP)
Validation: CO

Type of intervention

3. SELF‐HELP

Notes

73% of Group 1 participants used the interactive programme, compared with 90% of the comparison group who used the ALA programme

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: 10 small manufacturing companies in NC.
Design: Cluster RCT, no details of randomization

Participants

859 blue‐collar women at baseline (73% of eligible). 538 completed programme to 18m. 53% aged 40 or younger, 58% African American. Mean BMI 29. 30% I group, 22% C group smoked.

Interventions

1. Intervention: computer‐tailored 'magazine' with dietary, exercise, smoking advice, at baseline and 6m, plus social support at work from trained helpers in participants' chosen activity. N.B. No lay helpers offered smoking support.
2. Delayed intervention (control): One computer‐tailored 'magazine at 6m, no social support.

Outcomes

Abstinence at 18m: self‐reported, no biochemical validation.

Type of intervention

3. SELF‐HELP

Notes

This trial was added to the 2005 update
Natural (lay) helpers declined training in smoking cessation, so this arm of the intervention was not available to participants trying to quit

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: Belgium
Recruitment: industrial worksite primary care clinic
Design: RCT, no details of randomization method

Participants

199 adult male smokers (av cpd 24‐5)

Interventions

1. Nicotine gum (4 mg) for at least 3m
2. Nicotine gum (2 mg) for same time period.
Minimal physician support

Outcomes

PP abstinence at 12m
Validation: cotinine and carboxyhaemaglobin in a sub‐sample

Type of intervention

4. PHARMACOLOGICAL

Notes

Blinding was broken at 3m, and participants were free to choose their dosage of nicotine gum.
Results were stratified by Fagerstrom score.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

High risk

C ‐ Inadequate

Methods

Country: UK
Recruitment: Worksite primary care clinic in UK retail company (employees 3,253)
Design: RCT, no details of method

Participants

270 participants invited out of 334 who expressed an interest
Av age:34, 70% F
av cpd 15.5

Interventions

1. Nicotine gum (2 mg) at least 4 boxes, duration not stated. (172 people)
2. Non‐intervention control group (no placebo) of 64 continuing smokers
Low level of support

Outcomes

Sustained abstinence at 12m; Validation: expired CO

Type of intervention

4. PHARMACOLOGICAL

Notes

Slight contamination of intervention group, as 4 control group members were moved at their own request into the intervention group.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

High risk

C ‐ Inadequate

Methods

Country: UK
Recruitment: Worksite primary care clinic (employees 3,253)
Design: RCT, no details of method

Participants

161 adult smokers who were still smoking after 3m of a videotape smoking cessation programme. Av cpd 15‐19

Interventions

1.Nicotine gum (2 mg) for up to 12w
2.Non‐intervention control group (no placebo).
Low level of support

Outcomes

Validated long‐term abstinence at 12m
Validation: expired CO

Type of intervention

4. PHARMACOLOGICAL

Notes

Participants are the non‐quitters at 3m from Sutton 1988d
5/82 control subjects asked for and received treatment. One was a long‐term abstainer, and is classed as a control group success.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

High risk

C ‐ Inadequate

Methods

Country: Belgium
Recruitment: Worksite
Design: RCT, computer‐generated list

Participants

374 volunteers
male and female, age > 20 yrs. No. of cigarettes: > 10 day for > 3 years

Interventions

1.Active patch and active gum (2mg as required)
2.Active patch and placebo gum
3.Placebo patch and placebo gum
High level of adjunct support.

Outcomes

Sustained abstinence at 12m
Validation: baseline salivary cotinine, and expired CO < 10 ppm at each follow up

Type of intervention

4. PHARMACOLOGICAL

Notes

Other outcomes included dermatological and systemic adverse effects, and time to relapse.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Low risk

A ‐ Adequate

Methods

Country: Spain
Setting: 1 transport company (mostly bus drivers) and 2 electrical utility worksites (mostly clerical) in Bilbao.
Design: Open RCT, with randomization by sealed opaque envelopes and computer‐generated random lists

Participants

218 participants randomized to intervention (115) and control (103). All had physical check up, Fagerstrom NTQ, lab tests and ECG at baseline

Interventions

1. Intervention: 5‐8 mins structured individual counselling on smoking cessation at baseline by occupational physician, + further contacts at 2 days, 15 days and 3m. Grade I (Fagerstrom score < 5) counselling only. Grade II (Fagerstrom score 5‐7) 8 wks x14 mg nicotine patches. Grade III (Fagerstrom score > 7) 4 wks x 21 mg, 4wks x 14mg, 4wks x 7mg. Lower grade interventions could be upgraded if necessary. Participants kept records of progress, withdrawal symptoms, adverse events; weight and tobacco consumption were checked at specified intervals.
2. Control: minimal (30‐60 secs) sporadic unstructured advice, usually at annual medical check up

Outcomes

Continuous abstinence (7 day PP at each assessment) at 12m.
Validated at each assessment by expired CO <= 10ppm

Type of intervention

4. PHARMACOLOGICAL

Notes

This trial was added to the 2005 update
Secondary outcomes were: change in tobacco withdrawal symptoms, and weight changes.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Low risk

A ‐ Adequate

Methods

Country: USA
Setting: volunteers from telephone company (8) and a medical clinic (16)
Design: RCT, no details of randomization method

Participants

24 participants av age 34, had smoked for an average of 16 years, and av cpd 24.
Average score on the Fagerstrom NTQ 6.0
Participation rate: not reported

Interventions

Group therapy
1. controlled smoking
2. controlled smoking plus partner support

Outcomes

Self‐monitoring records, laboratory analyses of spent cigarette butts, and CO at 6m

Type of intervention

5. SOCIAL SUPPORT

Notes

Other outcomes included nicotine levels (brand smoked), smoking reduction, CO levels in continuing smokers and % of cigarette smoked.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country:USA
Recruitment: VA hospital, savings and loan association, and a health insurance agency employee volunteers
Design: RCT, no details of randomization procedure

Participants

29 adult cigarette smokers
69% female. Av. age 33.5
Average 25 cpd
Fagerstrom score 5.7, indicating moderate levels of tobacco dependence.
Participation rate: not reported

Interventions

1. Basic program (BP): subjects participated in 6 weekly group meetings‐ focused on making reductions in the no. of cpd and reductions in nicotine content. Midway through the programme subjects given the option of either complete cessation or reducing the percentage of each cigarette smoked.
2. BP and social support (SS): the same treatment as subjects in the BP group; in addition, each BP plus SS subject selected a partner who provided support and encouragement during non‐work hours.

Outcomes

Self reports, examination and weighing of saved cigarette. Butts and 2 biochemical measures of smoking exposure, CO and saliva thiocyanate.

Type of intervention

5. SOCIAL SUPPORT

Notes

Outcomes included changes in nicotine content (brand smoked), amount of cigarette smoked, and number of cigarettes smoked. The influence of social support, or lack of it, was also assessed.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: worksite volunteers in 2 comparable oil refineries in Southern Louisiana
Design: RCT, no details of randomization method

Participants

30 smokers (14 at intervention site and 16 at comparison site)
76% male
Av. age: 39, av. cpd 21
Participation rate: not reported

Interventions

1. Intervention: comprehensive programme of smoking control, discouragement, cinnamon sticks as cigarette substitutes, and smoking cessation
2. Control: smoking cessation alone

Outcomes

Self‐reported smoking cessation with urinary cotinine validation

Type of intervention

6. ENVIRONMENTAL SUPPORT

Notes

Introduction includes lengthy discussion of economic and health costs of smoking

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 4 General Motors worksites, Michigan
Design: Cluster randomization by worksite

Participants

Random sample of 400‐500 employees screened at baseline and followed up 3 yrs later.
Predominantly male, white, blue collar.
41‐45% smoked at baseline, but in the rescreened sample only 41% in site 3 and 36% in site 4 smoked at baseline

Interventions

Smoking, high blood pressure & obesity targetted.
1 worksite was allocated to each of 4 conditions:
1. Wellness screening; identify risks & referral
2. As 1. + media, programme sign‐up campaigns and classes
3. As 1. + media, program sign‐up campaigns, menu of interventions including guided self‐help, group or individual counselling + follow up
4. As 3 + follow‐up counselling + Plant Organization including peer support, aimed at reducing relapse.
All sites initiated no smoking areas during the period.

Outcomes

Self‐reported smoking status

Type of intervention

6. ENVIRONMENTAL SUPPORT

Notes

Quit rates were calculated by combining 1985 smokers and ex‐smokers (i.e. at risk of relapse)as the denominator. If the calculation is based only on current smokers at 1985 compared with 1988 quitters, the results do not reach statistical significance.
Reduced prevalence at all 4 sites coincided with the setting‐up of restrictive policies in each site.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 6 white‐collar worksites. No worksite had a formal no‐smoking policy or ongoing smoking cessation activities.
Design: Cluster randomized trial

Participants

6 worksites ranging in size from 950 to 3,300 employees. 25% smoking prevalence.
252 employees aged 21 and older participated, representing only a small portion of the total number of smokers at each worksite.
62% female. Av. age 42.3
>60% White

Interventions

1. Full programme (I): volunteers participated in a 5w training programme for quit‐smoking group leaders, and received additional training ,support, and how‐to manuals to carry out a protocol for health education and sitewide intervention activities, as well as for the implementation of worksite smoking policies.
2. Group‐only (C): volunteers participated in the training programme for group leaders, but did not carry out the protocols for health education and smoking policies

Outcomes

Self‐reported cessation at 12m
Validation: expired air CO

Type of intervention

6. ENVIRONMENTAL SUPPORT

Notes

Unit of randomization was worksite but unit of analysis was the individual.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: Smoking volunteers employed at Francis Scott Key Medical Center, Baltimore.
Design: RCT, no randomization detail

Participants

47 subjects who completed 5 days verified abstinence.

Interventions

1.Contingent payment for continued abstinence + frequent monitoring (n = 17)
2. Non‐contingent payment for abstinence + frequent monitoring (n = 16)
3. Non‐contingent payment, infrequent monitoring (n = 14)

Outcomes

Quit rate at 6m, confirmed by CO validation

Type of intervention

7. INCENTIVES

Notes

Subjects had received a minimal cessation programme, i.e. a 15‐minute talk and a booklet, with no skills training in cessation or relapse prevention.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Recruitment: University of Alabama employees volunteering for a quit smoking programme
Design: randomized trial, using sealed numbered envelopes containing computer‐generated assignment prior to baseline interview.

Participants

378 smokers
Av. age 37, av cpd 23‐27
Therapist: health visitor

Interventions

All groups received a 10 minute session of brief advice
1.+ S‐H manuals
2. +S‐H and another session of counselling (20‐30 mins) with skills training, buddy selection and a contract
3.as 1. With monetary awards for cessation
4.as 2 with monetary rewards for cessation

Outcomes

Abstinence at 1 yr (sustained at 6w, 6m & 1 yr)
Validation: saliva thiocyanate < 100 ng/ml at all follow ups

Type of intervention

6. ENVIRONMENTAL SUPPORT/ INCENTIVES

Notes

Other outcomes included some cost‐benefit analysis, including efficacy of incentives..

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Low risk

A ‐ Adequate

Methods

Country: USA
Recruitment: 19 worksites in Oregon.
Design: Cluster randomized RCT

Participants

Worksites from 140‐600 employees. Smoking prevalence of 21‐22%; Av age 40‐41. 63% female. 474 in Incentives (I) Group, 623 in No incentives (NI) Group

Interventions

Company steering groups ran the programmes
1. I Group members were paid US$10 for each verified abstinent month, up to 10m, + monthly and end‐of‐programme lotteries. There was also a buddy scheme, with cash prizes to helpers.
2. NI Group operated their normal company policy, which usually restricted but didn't ban smoking

Outcomes

Cessation rates at 12m and 2 yrs, verified by CO and salivary cotinine

Type of intervention

7. INCENTIVES

Notes

Analysis was at both worksite and individual level.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 24 worksites in and around St Paul. No overlap with the Healthy Worker Project.
Design: Randomized 2 x 3 factorial design, with smokers followed up at 12m and 24m.
85.5% responded to 12m survey, and 81.7% to 24m survey

Participants

2402 smokers on 24 sites, four sites randomized to each of the 6 conditions. There were significant differences in demographic characteristics between sites.
Smoking prevalence ranged from 10.7% to 37.2%

Interventions

The three programme formats were group counselling, telephone counselling or a choice of group or phone.
The programmes were then offered with and without incentives (=6).
The incentive site smokers received US$10 for signing up to a programme, and US$20 for near or full completion. They were also offered US$20 for 30 days cessation, and were then entered into a prize draw for a US$500 cash prize.

Outcomes

Rates of recruitment to the programmes, and 7‐day smoking PP at 12m and 24m follow up.
Validation was by self report, confirmed by family member or friend.
A sample of 188 quitters at 24m were asked to supply a saliva sample (128 complied).
Winners of the prize draw could only claim their prizes by verifying abstinence with salivary cotinine.

Type of intervention

7. INCENTIVES

Notes

This is the SUCCESS Project.
Significant differences between worksites meant that several covariates had to be controlled for in the analyses.
Other outcomes included comparing quit rates of registrants for the programmes with non‐registrants

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country:USA
Setting: 9 Pacific Gas and Electric worksites, allocated to 4 levels of intervention.
Design: Quasi‐experimental, random assignment of worksites. Sites were blinded to other intervention conditions. 

Participants

2887 workers across 9 sites at baseline HRA survey (69% of eligibles). At 2 yr follow up 1998 (48%) were surveyed. Cross‐sectional, not cohort surveys.
> 40% of participants were manual workers, 25‐31% clerical, 15‐21% managerial and 12‐16% technical staff.. 74‐79% male, > 70% aged 30‐49. 

Interventions

1. (3 sites, 1372 participants): HRA (height, weight, smoking, BP, cholesterol, HDL levels) at start and end of programme, + a bi‐monthly health newsletter (counts as control group).
2. (2 sites, 1083 participants): As 1, + health resources centre and free self‐care booklets.
3. (2 sites, 1016 participants) As 2, + behaviour change workshops and a divisional HealthWise social support team.
4. (2 sites, 693 participants): As 3, + case management programme for high‐risk participants (the 15% with the highest risk scores) and an environmental policy (space, smoking policies, incentives, health fairs) 

Outcomes

Smoking prevalence at 2 yr follow up in all four intervention groups. Self‐report 'current smoker' at HRA; no biochemical confirmation.

Type of intervention

8. COMPREHENSIVE

Notes

This trial was added to the 2005 update.
This is the HealthWise Stepped Intervention Study (HSIS). Level 4 sites were pre‐selected by PG&E management (non‐random) and were significantly smaller than the other levels, reflecting the expense of the Level 4 interventions 

Methods

Country: USA
Setting: 26 worksites in Oregon
Design: Cluster randomized trial

Participants

26 heterogeneous worksites in Oregon with between 125 and 750 employees ‐ an average of 247.
Participation rate: at baseline, early intervention rate was 38% and delayed intervention 58%.
At 2 yr follow up, early intervention rate was 40% and delayed intervention was 57%

Interventions

Take Heart Project, focusing on diet and smoking
Early intervention (multifaceted programme consisting of employee steering committee and a menu approach to conducting key intervention activities tailored to each site) vs. delayed but similar intervention

Outcomes

Self‐reported smoking cessation

Type of intervention

8. COMPREHENSIVE

Notes

This is the Take Heart worksite wellness program. Other outcomes included dietary intake and cholesterol levels

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 108 worksites in 16 US states
Design: Randomized matched‐pair trial, using cross‐sectional surveys at baseline and 2 yr follow up

Participants

108 worksites with over 28,000 employees ( 49 ‐ 1700 workers per site). Participation rate 72%, Av age 41, 77% male, 92% white.
Only 3 of the 4 study centres (84 sites) measured changes in smoking, as the 4th centre sites (Florida) had smoking bans already in place.

Interventions

Each workplace had an employee as co‐ordinator, and an employee advisory board.
1) Individual core interventions: Process included a kickoff event, interactive activities, posters and brochures, self assessments, self‐help materials, campaigns and contests, and direct education through classes and groups.
2) Environmental core interventions:
Consultation on smoking policy, changes in cafeteria and vending machine food, and additional nutritional education.
Control sites had results of employee survey, and in some cases an optional minimal intervention of posters and newsletters.

Outcomes

Self‐reported smoking cessation, without biochemical validation. 6m abstinence at follow up, smoking prevalence.

Type of intervention

8. COMPREHENSIVE

Notes

This is the Working Well Trial.
Randomization and analysis were both based on worksite.
Other outcomes were dietary fat reduction, fibre intake and fruit and vegetable consumption.
Some control sites had minimal interventions such as posters and brochures.

The Working Well trial generated a nested cohort study, the WellWorks Trial, which examined dietary and smoking changes stratified by job type at the Massachusets worksites.
See Sorensen 1998 reference.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 24 mainly manufacturing worksites in Massachusets, randomized into 12 pairs, and all thought to be using known or suspected carcinogens.
Randomization was by worksite, but analysis was by individual. Analysis in this paper was cohort‐based

Participants

5914 (61%) of sampled employees responded at baseline, and 5406 (62%) at 2 yr follow up.
The cohort who responded to both surveys was 2658 employees.

Interventions

3 elements of intervention:
1) Joint worker‐management programme planning and implementation
2) Consultation by project staff with management on environmental changes, inc tobacco control policies, healthy foods, occupational hazard reduction
3) Health education programs targeting individual behaviours in the risk factor areas.

Outcomes

Self‐reported abstinence for 6m before final survey.
No biochemical validation

Type of intervention

8. COMPREHENSIVE

Notes

The WellWorks Study is a nested component of the Working Well trial, but, unlike that trial, attempted to integrate health promotion and health protection interventions, and is therefore assessed separately.
Other outcomes included fat, fibre and fruit and vegetable consumption, and differences between blue‐ and white‐collar workers in all outcomes.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: Holland
Setting: 4 work sites (chemical, telecommunication, public transport and local government) and 4 other similar worksites
Design: cluster randomized trial

Participants

279 employees at intervention sites and 234 employees at comparison sites
Average age: 41 years
75% male

Interventions

1. Comprehensive program (self‐help manuals, group courses, a mass media campaign, smoking policies and a 2nd yr programme)
2. Minimal intervention (self‐help manuals only).

Outcomes

Self‐reported smoking cessation and saliva cotinine estimation

Type of intervention

8. COMPREHENSIVE

Notes

Analysis of light vs heavy smokers suggests greater efficacy among heavy smokers (P values not given).
Other outcomes included relapse rates, the effectiveness of a 2nd yr programme.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 26 worksites in RI and SE Mass (Brown University based).
Only 22 sites completed the trial.
Design: randomized matched pair, following a cohort over 3 yrs. Randomization process not described

Participants

22 worksites, and 2055 participants who completed all surveys. No demographic differences between intervention and control groups. Smoking prevalence 28% across both groups.

Interventions

1. Intervention sites: As with Working Well Trial (Sorensen 1996), but including physical activity; a combination of individual and environmental programmes, including space, showers, equipment and discounted membership of fitness facilities.
2. Control sites: Minimal care: Could offer 2 S‐H smoking cessation programmes and 1 each on nutrition and physical activity.

Outcomes

Self‐reported abstinence at 3 yrs for 6m prior to assessment, and 7‐day PP
No biochemical validation used.
Secondary outcome: movement through stages of change

Type of intervention

8. COMPREHENSIVE

Notes

This trial was added to the 2005 update
This is the Working Healthy Projected, nested within the Working Well trial

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: Sweden
Recruitment: 4 public sector worksites (568 employees) in Helsinborg.
Design: RCT
Randomization: method of allocation not stated.

Participants

Of 128 at‐risk workers invited, 60/65 randomized to the intervention group attended for baseline assessment, and 53/63 from the control group.
Mean age was 49.7, 61% female.

Interventions

1. Intervention group received 16 group sessions a year, as well as individual counselling by a nurse. Sessions included lectures, discussions, video sessions and outdoor activities.
2. Control group received standard written and oral advice about cardiovascular risk factors at the start of the intervention, and nothing thereafter.

Outcomes

PP at 12m and 18m. No biochemical validation.

Type of intervention

8. COMPREHENSIVE

Notes

Smoking was only one of several risk factors targeted, including BMI, BP,heart rate, low‐density lipoprotein and cholesterol.
Group sessions were held in working hours but away from the worksites.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear

Methods

Country: USA
Setting: 15 manufacturing sites, probably handling hazardous chemicals, in Massachusetts.
Design: RCT, randomized by worksite, but analysed by individual employee.

Participants

9019 employees (80%) across 15 sites. Mean workforce size 741 employees. Responders in the control groups were younger, more likely to be female, less educated, less likely to be white, and less likely to be hourly‐paid rather than salaried.

Interventions

1. Control [8 sites] had Health Promotion (HP) intervention, i.e. consultation to management on tobacco control policies, catering and cafeteria policies, and programmes aimed at individuals, including self assessment with feedback, self‐help activities, contests, demonstrations and displays, opportunities to try behaviours and goals, and group discussions.
2. Experimental Group [7 sites] (HP/OHS= health promotion with occupational health and safety) had the same elements as the Control sites, plus management recommendations to reduce occupational hazard exposure. For individuals, occupational health and safety training was added to the tobacco and nutritional elements of the control programme.

Outcomes

Quit rates (PP) at 6m, reported by cross‐sectional survey and for the smoking cohort.
Self report only, no biochemical validation

Type of intervention

8. COMPREHENSIVE

Notes

This is the Wellworks‐2 Trial, targeting particularly blue collar workers. Analyses were cross‐sectional and cohort
Other primary outcomes were nutrition and perceived exposure to occupational hazards.

Risk of bias

Bias

Authors' judgement

Support for judgement

Allocation concealment?

Unclear risk

B ‐ Unclear