Table 1.

Characteristics of included reviews

Review ID (Search dates) Risk of bias	Review title	Indigenous population	Interventions	No. and type of included studies	Synthesis	Main outcomes reported (summary)	Summary of reviewer conclusions
Minichiello 2016 [32] (1980 to 2014) Moderate	‘Effective strategies to reduce commercial tobacco use in Indigenous communities globally: A systematic review’	All	Any	93 (73 interventions) Quantitative [56] Mixed method [25] Qualitative [12]	Mainly statements about statistical significance	Smoking cessation: Mostly increased quit rates (4 studies) Prevention of initiation: 2/4 studies reported sig. effect Prevalence reduction: No sig. change (3 studies) Tobacco sales reduction: Unclear/no sig. change (1 study). Smoke-free homes/workplaces: no sig. effect (8 studies) Knowledge: Mostly positive impact (8 studies) Engagement: Increased community interest	Increasing priority and readiness to tackle high rates of commercial tobacco use employing comprehensive (multiple activities, centring of Aboriginal leadership, long-term community investments) and tailored interventions (provision of culturally appropriate health materials and activities).
Carson 2014 [33] (Up to 15 Aug 2014) Low	‘Smoking cessation and tobacco prevention in Indigenous populations’	All	Any	91 Randomised controlled trials [10], controlled clinical trials [5], pre-post studies [10], government reports [53], and protocols [4]	Mainly qualitative statements	Smoking cessation: Reduced smoking levels at follow-up in 12/15 controlled trials. Prevention of initiation: Results for youth not clear (9 studies).	Recommend multifaceted programmes that concurrently address behavioural, psychological and biochemical sides of addiction, using culturally tailored resources for individual Indigenous population needs. Interventions with more components, and greater intensity, were more likely to be effective than those of shorter duration and lower intensity.
Johnston 2013 [21] 1980 to May 2012 Moderate	‘Reducing smoking among indigenous populations: new evidence from a review of trials’	All	Any (if reporting Indigenous and non-Indigenous outcomes to assess effect of cultural tailoring)	5 Randomised controlled trials and cluster RCT’s [5]	Mainly statements about statistical significance	Smoking cessation: No sig. effect for either Indigenous or non-Indigenous participants in 3/5 studies.	No significant difference between Indigenous and non-Indigenous populations for smoking cessation and suggest not all tobacco control interventions can/need to be culturally adapted. Promising evidence on effectiveness of behavioural interventions using mobile phone technology.
Carson 2012a [27] (Up to April 2011) Low	‘Interventions for smoking cessation in Indigenous populations’	All	Any	4 Randomised controlled trials and cluster RCT’s [2] Non-randomised [2]	Mainly pooled effect estimates from meta-analysis	Smoking cessation: Sig. effect (risk ratio 1.43, 95% CI 1.03 to 1.98, p = 0.032). Adverse effects: Insomnia, rash and other minor complications reported from NRT treatment (26% versus 9%), compared to placebo. Knowledge: No sig. difference in ‘readiness to quit’. Costs and mortality reported.	Review highlights lack of available evidence to assess effectiveness of smoking cessation interventions, despite recognised success in non-Indigenous populations. Limited but available evidence does show smoking cessation interventions specifically targeted at Indigenous populations can result in smoking abstinence.
Carson 2015 [36] (Unclear) High	‘Culturally tailored interventions for smoking cessation in indigenous populations: A Cochrane systematic review and meta-analysis’	All	Any (focus on cultural tailoring)	9 Randomised or non-randomised controlled trials [9]	Mainly pooled effect estimates from meta-analysis	Smoking cessation: Non-sig effect (risk ratio 1.43 (95% CI 0.96 to 2.14); p = 0.08, 7 studies).	Some evidence supports using culturally tailored smoking cessation interventions for Indigenous populations. Most effective interventions were multifaceted cognitive and behavioural, mixing several initiatives simultaneously with health professional participation
DiGiacomo 2011 [22] (1990-2010) Moderate	‘Smoking cessation in indigenous populations of Australia, New Zealand, Canada, and the United States: elements of effective interventions’	All	Smoking cessation	9 Randomised controlled trials and cluster RCT’s [1] Non-randomised [8]	Mainly qualitative and descriptive statements	Quit rates: Higher quit rates reported for bupropion vs. placebo. Prevalence reduction: Mixed results from 5 studies. Cultural considerations: Describes cultural tailoring and levels of community engagement. Workforce/organisation: Describes Indigenous workforce involvement, organisational support, and financial/transport assistance for clients. Self-determination/flexibility: Describes programme flexibility and availability. Partnerships and engagement: Discusses strategies to promote engagement and principles for establishing partnerships.	Few identified interventions tailored for Indigenous populations. Successful interventions featured integrated, flexible, community-based approaches that addressed known barriers/facilitators to quit smoking.
CADTH 2013 [38] (Jan 1 2003 to Jun 26 2013) Moderate	‘Indigenous Knowledge for Smoking Cessation: Benefits and Effectiveness	All	Indigenous knowledge for smoking cessation	1 Systematic review [1]	Mainly qualitative statements	No studies found in systematic review.	No evidence regarding Indigenous knowledge for smoking cessation was identified.
Gould 2013a [25] (Up to Oct 2011) Low	Should anti-tobacco media messages be culturally targeted for Indigenous populations? A systematic review and narrative synthesis’	All	Culturally tailored mass media campaigns	21 Randomised controlled trials [4] Non-randomised [4] Database analysis [1] Qualitative [6] Mixed methods [6]	Mainly qualitative and descriptive statements	Smoking cessation: Higher quit rates reported among intervention groups. Intention to smoke: Significant decrease in future intention to smoke. Knowledge: Variable impact on recall, knowledge and intentions to quit reported. Cultural respect: 12 studies measured cultural suitability and/or relevance and qualitative studies showed preference for culturally targeted messages. Believability and usability also reported (3 studies).	Indigenous people had good recall of generic anti-tobacco messages, but preferred culturally targeted messages. Maori possibly less responsive to holistic targeted campaigns than generic fear campaigns. Culturally targeted internet/mobile phone messages just as effective in American Indians/Maori as generic general population messages. Where culturally targeted messages trialled, campaigns shown to be effective regarding change of knowledge, attitudes and behaviour.
Passey 2013 [26] (Up to Dec 2012) High	‘How will we close the gap in smoking rates for pregnant Indigenous women’	All (pregnant women only)	Any	2 Randomised controlled trials and Cluster RCT’s [1] Non-randomised [1]	Mainly qualitative statements	Smoking cessation: No sig. effect. Relapse prevention: No sig. effect.	No evidence for effective interventions that support pregnant Indigenous Australian women to quit smoking.
Carson 2012b [34] (Up to Nov 2011) Low	‘Interventions for tobacco use prevention in Indigenous youth’	All (adolescents only)	Any (controlled trials only)	2 Randomised controlled trials [2] Controlled clinical trial [0]	Mainly effect estimates for single studies	Tobacco use: No sig. changes between intervention/control groups at final follow-up. Changes in attitudes towards drugs and self-esteem: No sig. differences. Changes in knowledge: Sig. increases in knowledge in intervention groups.	Conclusion cannot be derived about efficacy of tailored tobacco prevention initiatives for Indigenous youth. This review highlights lack of data and need for more research in this area.
Carson 2013 [28] (Unclear) High (abstract only)	‘Interventions for tobacco prevention in Indigenous youth: A Cochrane review and a narrative synthesis’	All (adolescents only)	Any (controlled trials only)	6 Randomized or non-randomized controlled clinical trials [6]	Mainly qualitative statements	Tobacco use: No evidence of change.	Review highlights lack of data for tobacco prevention initiatives tailored to Indigenous youth.
Ivers 2003 [23] (1980 to March 2001) Moderate	‘A review of tobacco interventions for Indigenous Australians’	Australian (includes reflection on evidence from other populations)	Any	4 Qualitative [3] Other [1]	Mainly qualitative statements	Prevention of initiation: Reduced consumption reported. Knowledge: Knowledge about tobacco increased (1 study). Practice change/human capability: Some practice changed after training health professionals in brief interventions (1 study).	Major lack of research/evaluation on tobacco interventions for Indigenous people.
Ivers 2011 [18] (Unclear) High	‘Anti-tobacco programmes for Aboriginal and Torres Strait Islander people’	Australian	Any	Unclear Randomised controlled trials and Cluster RCT’s, Non-randomised, and qualitative - unclear how many	Mainly qualitative statements	Smoking cessation: Reports ‘successful approaches’ as: health professionals providing brief quit advice with pharmacotherapy; training health professionals; Quit groups; and multicomponent anti-tobacco programmes. Prevention of initiation/prevalence reduction: Sig. increases in readiness to quit and knowledge of tobacco from multicomponent interventions. Consumption declined in community with most tobacco control activity. Smoke-free homes/workplaces: A workplace quit smoking programme was acceptable. No other programmes aimed at decreasing environmental smoke sufficiently evaluated. Self-efficacy: Many smokers quit ‘by themselves’, emphasising importance of self-efficacy. Equality: Presents differential effects of interventions. Partnership and engagement: Notes importance of partnerships with community health organisations, and that programme delivery is enhanced by community involvement, ongoing funding and coordination. Cultural respect: Culturally appropriate, non-coercive counselling approaches likely to be appropriate. Aboriginal people believed tobacco programmes must be locally based, include local content, involve Elders and significant community members in design/delivery, and have a broad community focus.	Suggest successful approaches include: health professionals providing brief quit advice and pharmacotherapy; cessation advice training for health professionals; Quit groups; and well-delivered multicomponent anti-tobacco programmes. Community health organisations play key role in tobacco control, mainly in delivery of brief interventions and prescribing nicotine replacement therapy/pharmacotherapies, promoting smoke-free environments in antenatal/early childhood programmes, and in quit groups’ coordination.
Ivers 2014 [19] (Unclear) High	‘Attachment Two: The NSW Strategic Framework for Aboriginal Tobacco Resistance and Control – Supporting evidence’	Australian	Any	Unclear Randomised controlled trials and Cluster RCT’s, Non-randomised, and qualitative - unclear how many	Mainly qualitative statements	Smoking cessation: Effect seen from brief advice combined with pharmacotherapy; a locally developed intensive tobacco intervention; free nicotine patches/brief advice; and a quit group. No effect seen in an intervention for pregnant Aboriginal women; or National Tobacco Campaign evaluation. Prevalence reduction: Tobacco price reduced prevalence and cigarette costs identified as one of the reasons for quitting smoking. Knowledge: High levels of awareness, increases in knowledge and recall seen several campaigns; with high proportions finding them believable/relevant and considering quitting or cutting down but few report accessing Quitline. Workforce/practice change: Few health workers/practitioners recommended Quitline, despite increases in health workers confidence to talk about smoking. Increased numbers of health workers reported giving advice about NRT, environmental tobacco smoke, and reducing tobacco use. Self-efficacy: Evidence suggests quitting unaided improves self- efficacy in quitting. Partnership and engagement: Suggests critical to success. Cultural respect: Preferred campaigns are specifically designed for Aboriginal people, locally based, include local content, involve elders and significant community members in design/delivery, and have a broad community focus. Brief advice preferred in culturally appropriate, supportive and non-coercive way.	Factors that are vital to tobacco resistance and control programmes success include: Aboriginal communities develop, deliver and evaluate programmes; comprehensive and multi-component; funding for sustainable programmes over the long term; prevent duplication of effort between communities, non-government organisations and government agencies by coordination and partnerships. Types of effective interventions in decreasing Aboriginal smoking include: health professionals providing brief quitting advice and pharmacotherapy; cessation advice training for Aboriginal health workers and health professionals; multi-component tobacco resistance and control programmes; Quit groups; and intensive advice on smoking cessation.
Power 2009 [35] (2001 to 2007 (update of Ivers 2001 [18])) Moderate	‘Tobacco interventions for Indigenous Australians: a review of current evidence’	Australian	Any	12 Non-randomised [10] Qualitative [2]	Mainly qualitative statements	Smoking cessation: Increased quit rates reported in several studies. Prevalence and tobacco sales reduction: Changes in prevalence and store compliance with sales restriction legislation reported, except where vendor machines available (1 study). Knowledge: knowledge changes reported (1 study). Human capability and practice change: Increased health worker confidence in brief intervention (1 study). Cultural respect: 1 study attributes success to creating a culturally safe space.	Individually targeted smoking cessation approaches (e.g. NRT and/or counselling) may be effective for Indigenous Australians. No evidence about interventions likely to be effective in encouraging more Indigenous Australians to access quit support strategies. Limited evidence about possible effective approaches in surmounting major social/cultural barriers to Indigenous smoking cessation.
Upton 2014 [20] (2004–2014) High	‘Tackling Indigenous Smoking and Healthy Lifestyle Programme Review: A rapid review of the literature’	Australian	Any	36 (27 interventions) Randomised controlled trials and cluster RCT’s, non-randomised studies, qualitative studies, systematic reviews, policy documents, and unpublished reports	Mainly qualitative statements	Smoking cessation: Some of the 7 studies showed increased quit rates, including from intensive counselling and NRT. Prevention of initiation: Increased student self-esteem; positive impact on students’ knowledge/attitudes and self-efficacy (3 studies). Prevalence reduction: Reduction in self-reported smoking prevalence, but only statistically significant in one site (1 study). Tobacco sales reduction: Compliance with legislation around selling tobacco to minors in Indigenous Australian communities more difficult to achieve and problems with ongoing monitoring in remote areas, especially if tobacco access via vending machines/independent traders (2 studies). Quit attempts: Smoke-free workplace policies encouraged Quit attempts. Knowledge: Changes in knowledge reported from several interventions. Two studies demonstrated a clear link between health messages/negative attitudes to smoking/increased promotion/maintenance of smoke-free areas at home and in broader Indigenous Australian community. Also part of multi-component interventions. Adverse effects: Concerns anti-tobacco campaigns may have led smokers to feel persecuted/more defensive; and reports smokers feel more knowledgeable about smoking impacts, but possible barrier to ongoing engagement with messaging as ‘know everything’. Human capability and practice change: Increased confidence (1 study) and reviewer recommends developing local capacity/local workforce; expanding the Indigenous Australian workforce and increasing its capacity to deliver effective care. Self-efficacy: Discussion of importance of holistic approach.	Smoking environment changed significantly over recent years, with mixed evidence about if this has led smokers to feel persecuted/more defensive. Clear link seen in two studies between health messages/negative attitudes to smoking, and greater promotion/maintenance of smoke-free areas at home and in broader Indigenous Australian community. Many motivations to quit, but no particular reason encouraged Indigenous Australian smokers to ‘choose’ to quit. Evidence shows multilevel tobacco control approaches likely more effective for smoking prevalence decrease in Indigenous Australian communities. Formal/informal policies to ensure smoke-free environments in local organisations/businesses can also be effective, but require active participation of community members to ensure local ownership. Evidence supports high intensity counselling and brief interventions and use of NRT. Limited evidence around: school based interventions, Quitlines and pricing increases.
Clifford 2011 [31] (Jan 1990 to Aug 2007) Moderate	‘Smoking, nutrition, alcohol and physical activity interventions targeting Indigenous Australians: rigorous evaluations and new directions needed’	Australian	Any	5 Non-randomised [5]	Mainly qualitative statements	Smoking cessation: Increased quit rates reported in 3/4 studies. Costs: reported for 1 study.	Reviewer suggests it is comparatively rare for evaluations to be methodologically rigorous. Findings consistent with previous reviews showing intervention studies seldom done in Indigenous health and tend to have small effects. Recommend development and implementation of evaluation designs be informed by building capacity of local Indigenous communities and their healthcare services to engage as equal partners in research process.
Brusse 2014 [24] (Unclear) Moderate	‘Social media and mobile apps for health promotion in Australian Indigenous populations: scoping review’	Australian	Social media and mobile applications	4 Randomised controlled trials [1] App/social media programmes [3]	Mainly qualitative statements	Smoking cessation: Increased cessation in intervention (28%) compared to control (13%) group: Intervention as effective in Maori as non-Maori. Knowledge: Reports information on web downloads.	Current evidence for effectiveness/health benefit of social media and mobile software interventions especially for Indigenous/other traditionally underserved populations is scant and mixed.
Gould 2013b [37] (Up to March 2011) Low	‘Knowledge and views about maternal tobacco smoking and barriers for cessation in Aboriginal and Torres Strait Islanders: A systematic review and meta-ethnography’	Australian (women only)	Knowledge, attitudes, beliefs and barriers around smoking and cessation.	7 Non-randomised [1] Qualitative [5] Mixed methods [1]	Mainly qualitative statements	Smoking cessation: ‘Quitting is hard’ (1 study). Attitudes, beliefs and knowledge detailed. Smoke free homes/workplaces: Importance of reducing harm, being a protector (2 studies). Human capability/workforce: Role of IHW’s and other health professionals is challenging. Cultural respect: Cultural appropriateness and ethics is an important consideration in Indigenous studies.	Reviewer suggests comprehensive approaches, considering environmental context, increase knowledge of smoking harms/cessation methods, and provide culturally targeted support. Long-term, broad approaches are needed to de-normalise smoking in Indigenous communities as social norms and stressors perpetuate tobacco use in pregnancy. There is lack of knowledge of smoking harms and inadequate salience of current antismoking messages for maternal smokers, as well as poor knowledge of, access to, and use of evidence-based treatments for smoking cessation in pregnancy.
Thompson 2011 [29] (Unclear) Moderate	‘A review of the barriers preventing Indigenous Health Workers delivering tobacco interventions to their communities’	Australian (health workers only)	Impact of smoking status on provision of tobacco information.	14 Non-randomised [3] Qualitative [8] Reviews [3]	Mainly qualitative statements	Smoking cessation: Reports 9% quit rate; Relapse related to stressful times in clients lives. Knowledge/practice/workforce capability: Degree smoking information delivered may depend on IHWs’ tobacco use. Non-smoking IHWs more likely than smokers to talk to community about smoking (1 study) and smoking was barrier to giving support and/or information to community (1 study). Overall outcome showed IHWs own smoking was a barrier to service provision, but was not conclusive in one study. Need for workforce development recommended in 8 publications. Specific recommendations included training, mostly of health staff in brief interventions.	Overall, literature suggests IHWs’ smoking status is a barrier, but poor quality of most studies weakens evidence for this conclusion. Literature review has shown a need for practical quit support to help IHWs who want to quit. Training may also help increase IHWs knowledge in supporting community members wanting to alter smoking behaviour.
Clifford 2009 [30] (Jan 1990 to Aug 2007) Moderate	‘Disseminating best-evidence healthcare to Indigenous healthcare settings and programmes in Australia: identifying the gaps’	Australian	Dissemination of ‘smoking, nutrition, alcohol and physical activity’ interventions.	2 Non-randomised [2]	Mainly qualitative statements.	No smoking-related outcomes reported.	Review shows dissemination strategies targeting uptake of evidence-based SNAP interventions by healthcare providers working in Indigenous healthcare settings are not widely implemented, and evaluation outcomes often not published in peer-review literature. Recommend need for effective partnerships between government and research agencies, health-care providers and Indigenous healthcare services to improve likelihood of dissemination strategies implemented in Indigenous healthcare settings are feasible, acceptable and effective.

See Additional file 6 for detailed AMSTAR ratings for each review