Examining smoke-free coalitions in Armenia and Georgia: baseline community capacity

Abstract

Local coalitions can advance public health initiative but have not been widely used or well-studied in low- and middle-income countries. This paper provides (i) an overview of an ongoing matched-pairs community-randomized controlled trial in 28 communities in Armenia and Georgia (N = 14/country) testing local coalitions to promote smoke-free policies/enforcement and (ii) characteristics of the communities involved. In July–August 2018, key informants (e.g. local public health center directors) were surveyed to compare their non-communicable disease (NCD) and tobacco-related activities across countries and across condition (intervention/control). More than half of the informants (50.0–57.1%) reported their communities had programs addressing hypertension, diabetes, cancer and human papilloma virus, with 85.7% involving community education and 32.1% patient education programs. Eleven communities (39.3%) addressed tobacco control, all of which were in Georgia. Of those, all included public/community education and the majority (72.7–81.8%) provided cessation counseling/classes, school/youth prevention programs, healthcare provider training or activities addressing smoke-free environments. Informants in Georgia versus Armenia perceived greater support for tobacco control from various sectors (e.g. government, community). No differences were found by condition assignment. This paper provides a foundation for presenting subsequent analyses of this ongoing trial. These analyses indicate wide variability regarding NCD-related activities and support across communities and countries.

Introduction

Multi-sectoral local coalitions are effective policy change strategies [1–6]. Among the best-known and largest examples of such approaches is the World Health Organization’s (WHO) Healthy Cities initiative [7], an initiative that began in 1986 in high-income countries (i.e. Canada, USA, Australia, many European nations) and now involves thousands of cities worldwide. This program is successful in promoting health initiatives, community participation and empowerment and intersectoral partnerships committed to health initiatives [8, 9].

Community coalitions have been particularly well-documented as effective in shifting social norms and creating community readiness for tobacco control policies [10, 11]. The US CDC’s ‘Best Practices for Comprehensive Tobacco Control Programs’ states that ‘state and community coalitions are essential partnerships…they can keep tobacco issues before the public, combat the tobacco industry, enhance community involvement and promote community buy-in and support, educate policy makers and help to inform policy change’ (p. 18) [10]. Community Coalition Action Theory (CCAT) was developed as a framework for articulating factors influencing coalition effectiveness [12, 13] and has been highlighted as a useful model for studying community-based collaborative efforts to create policy change [14].

Unfortunately, research has found that insufficient resources can inhibit the formation and effectiveness of coalitions [15, 16]. This is particularly relevant for low- and middle-income countries (LMICs) where resources may be scarce. This is also particularly concerning given that LMICs are disproportionately affected by a range of public health problems, such as tobacco-related diseases and deaths [17, 18] including those attributed to secondhand smoke exposure (SHSe) [19]. In many LMICs, much policy progress is initiated at the national level, with data indicating that compliance with such smoke-free policies may be poor [20, 21]. Thus, parallel local work may be required shift social norms and build support for and compliance with such policies [1, 6]. Local coalitions are well-positioned to play this role.

One high-risk region for tobacco use is the area of the former Soviet Union [22]. Armenia and Georgia are two LMICs in this region that represent among the highest smoking prevalence among men (11th and 6th highest in the world; 52.3% and 57.7%, respectively), with lower smoking prevalence among women (1.5% and 5.7%, respectively) [23]. Moreover, recent data indicate high SHSe in these countries [20, 24], even in places where smoking is banned [20]. For example, a 2014 national survey indicated that 42.2% of Georgian adults reported daily SHSe, and 38.8% reported daily SHSe at home, with past-week SHSe being 54.2% in the home, 29.9% in indoor public places and 33.0% in outdoor public places [24]. In Armenia, the 2016–17 STEPS survey indicated that, among adults, past 30-day SHSe in the home was 26.6% and 56.4% at workplace [25]. These data reflect an urgent need to address the impact of SHSe in Armenia and Georgia.

The WHO Framework Convention on Tobacco Control (FCTC) mandates that nations that ratify the FCTC implement specific evidence-based tobacco control policies, including comprehensive public smoke-free legislation. Public smoke-free policies have been shown to be effective in reducing SHSe, youth tobacco use initiation, overall use prevalence and tobacco-related morbidity and mortality [26], as well as promoting cessation and reducing cigarette consumption in smokers [26].

The FCTC was ratified in 2004 and 2006 in Armenia and Georgia, respectively; however, few FCTC-recommended policies were implemented until recently. In Armenia, the current legislation, accepted in 2004, bans the consumption of tobacco in educational, cultural, healthcare, public transportation settings and in all other public buildings except dining facilities (e.g. cafes, bars, restaurants). In 2017, Armenia accepted the Tobacco Control Strategy for 2017–20 and is circulating a new draft tobacco control law, which would advance various tobacco control policies, including smoke-free air to apply to all public indoor places. In Georgia, the 2013 Tobacco Control National Strategy and 2013–18 Action Plan were adopted, and new progressive tobacco control laws were implemented in 2017–18, including a comprehensive smoke-free air policy that covers alternative tobacco products (e.g. e-cigarettes, hookah) across a broad range of indoor and outdoor areas.

In summary, given the promising effects of coalitions on advancing public health and the specific public health needs of LMICs, examining coalition effectiveness and CCAT in communities with limited resources such as LMICs is crucial in informing best practice recommendations to support local collaborative tobacco control action in such contexts [27]. Armenia and Georgia represent two LMICs with high smoking prevalence and SHSe, limited history and reach of smoke-free policies, likely policy enforcement challenges, and little history of community mobilization to promote public health initiatives [23]. Thus, the goals of the current paper are 2-fold. First, we provide a brief description of an ongoing matched-pairs community-randomized controlled trial aimed at testing the effectiveness of local coalitions in reducing SHSe in Armenia and Georgia. Second, we provide baseline characteristics of the communities selected to participate in the study, particularly in regard to their non-communicable disease (NCD) activities, and compare communities across countries and across conditions (i.e. intervention versus control) to assess and account for any baseline differences in order to inform future investigation of processes and outcomes related to this study.

Materials and methods

Ongoing study overview

This study was approved by the Institutional Review Boards of Emory University (#IRB00097093), the National Academy of Sciences of the Republic of Armenia (#IRB00004079), the American University of Armenia (#AUA-2017-013) and the National Center for Disease Control and Public Health of Georgia (#2017-026).

This study uses a matched-pairs community-randomized controlled trial and the CCAT as a framework for examining the effectiveness of local coalitions in promoting smoke-free air in Armenia and Georgia. This study defines a ‘community’ as a distinct municipality. We purposively selected 14 communities per country with (i) small to medium populations and (ii) local public health coordinating centers with sufficient capacity to engage in the proposed research. These criteria were determined for both pragmatic and scientific reasons. In particular, each country only has a couple of large cities, so working with small to medium countries was necessary to get a sufficient sample size of municipalities. Additionally, in the communities randomly assigned to be in the intervention condition, lead agencies are: (i) regional offices of Armenia’s National Center for Disease Control (NCDC) under the Ministry of Health and (ii) local municipal public health centers that collaborate with the National Center for Disease Control and Public Health (NCDC) of Georgia. Because we planned to have local public health branches/centers play key roles in coordinating local coalition activity, selected communities needed stable center leadership and sufficient resources to conduct and coordinate community-based work. Relevant to this, we attempted to match communities within each region based roughly on population size and public health branch/center budget. Scientifically, our justification for focusing on municipalities with small to medium populations is based on research that suggests that coalitions serving small to medium catchment areas are most effective in adopting evidence-based programs [1]. As indicated above, communities were paired in each country based on region (and distance from Yerevan or Tbilisi), population size and local public health branch/center budget and then randomly assigned to be in the intervention versus control conditions.

In the intervention communities, the lead agencies are charged with forming coalitions by recruiting partner organizations from civil society and other government sectors (e.g. health care, education), conducting situational assessments, and developing and implementing action plans to promote the adoption and/or enforcement of smoke-free policies. At study launch (January–February and June 2019), trainings were provided to the key members of the lead agencies, specifically to conduct situational assessments, form a coalition and execute action plans to promote smoke-free policy adoption/enforcement. Throughout the 3-year period of coalition activity, annual meetings with grantee communities will be held to present on activities, progress and lessons learned. To execute coalition activity, each intervention community receives grant funding of approximately $17 500 over the study period.

Among all 28 intervention and control communities, baseline and end-of-intervention assessments of the communities and activities in NCD prevention and control are conducted with key informants, (i.e. heads of the municipal public health centers in Georgia and regional offices in Armenia). In addition, assessments of coalition members/activities in the 14 intervention communities will be conducted throughout the study period. Among all communities (intervention and control), population-level surveys (i.e. of community member) were conducted before the launch of the coalition member trainings (October–November 2018) and then will be conducted at the end of the 3-year period (in October–November 2021).

Community capacity survey

As part of understanding the communities involved in the study, we conducted a baseline community capacity survey in July–August 2018. Heads of regional offices in Armenia and heads of municipal public health centers in Georgia were asked to complete a survey assessing their organizational characteristics, their NCD-related activities with particular questions focused on tobacco-related activities and some personal characteristics (described below).

Participant and organizational characteristics

Participants were asked to report selected sociodemographics, their role in the organization, length of employment and smoking status. They also were asked to describe the type of organization they represented (e.g. public health center; NCDC local branch) and the number of full-time staff employed by the organization.

NCD-related activities

Participants were asked: (i) whether their organization had any programs, services or activities that address various NCDs (see Table II); (ii) what programs or services they have addressing NCDs; (iii) whether any other organizations in their communities implement programs on NCD prevention and control, and if so, to describe them and (iv) whether their organization interacts with other agencies or organizations to work on NCD prevention or control, and if so, to list their partners and projects.

Table II.

Cross-country and cross-condition comparisons of organizational NCD- and tobacco-related activities

	All	Armenia	Georgia	P	Control	Intervention	P
NCD activities
Have programs, services or activities that address: (N, %)
Hypertension	14 (50.0)	0 (0.0)	14 (100.0)	<0.001	7 (50.0)	7 (50.0)	0.999
Diabetes	15 (53.6)	12 (85.7)	3 (21.4)	0.001	6 (42.9)	9 (64.3)	0.256
Cancer	16 (57.1)	12 (85.7)	4 (28.6)	0.002	8 (57.1)	8 (57.1)	0.999
Asthma	3 (10.7)	0 (0.0)	3 (21.4)	0.067	1 (7.1)	2 (14.3)	0.541
Chronic respiratory disease	6 (21.4)	0 (0.0)	6 (42.9)	0.006	2 (14.3)	4 (28.6)	0.357
Other	13 (46.4)	12 (85.7)	1 (7.1)	<0.001	6 (42.9)	7 (50.0)	0.705
None	2 (7.1)	2 (7.1)	0 (0.0)	0.142	1 (7.1)	1 (7.1)	0.999
Have programs/services provided for NCDs: (N, %)
Primary care/clinical	6 (21.4)	0 (0.0)	6 (42.9)	0.006	3 (21.4)	3 (21.4)	0.999
Cancer screening	4 (14.3)	0 (0.0)	4 (28.6)	0.031	1 (7.1)	3 (21.4)	0.280
HPV vaccination	14 (50.0)	12 (85.7)	2 (14.3)	<0.001	7 (50.0)	7 (50.0)	0.999
Tobacco cessation (NRT)	5 (17.9)	0 (0.0)	5 (35.7)	0.014	2 (14.3)	3 (21.4)	0.622
Patient education	9 (32.1)	0 (0.0)	9 (64.3)	<0.001	4 (28.6)	5 (35.7)	0.686
Community education	24 (85.7)	11 (78.6)	13 (92.9)	0.280	13 (92.9)	11 (78.6)	0.280
Other	10 (35.7)	10 (71.4)	0 (0.0)	<0.001	5 (35.7)	5 (35.7)	0.999
None	2 (7.1)	2 (14.3)	0 (0.0)	0.142	1 (7.1)	1 (7.1)	0.999
Other organizations in community work on NCDs (N, %)	12 (42.9)	8 (57.1)	4 (28.6)	0.127	5 (35.7)	7 (50.0)	0.445
Works with others on NCD prevention/control (N, %)	20 (71.4)	14 (100.0)	6 (42.9)	0.001	12 (85.7)	8 (57.1)	0.094
Tobacco-related activities
Have NCD programs addressing tobacco control (N, %)	11 (39.3)	0 (0.0)	11 (78.6)	<0.001	5 (35.7)	6 (42.9)	0.699
NCD program addresses:a	N = 11 a	N = 0 a	N = 11 a		N = 14 b	N = 14 b
Cessation counseling or classes	9 (81.8)	—	9 (81.8)	—	4 (28.6)	5 (35.7)	0.686
Referrals to Quitlines/cessation services	5 (45.5)	—	5 (45.5)	—	2 (14.3)	3 (21.4)	0.622
School/youth prevention	9 (81.8)	—	9 (81.8)	—	5 (35.7)	4 (28.6)	0.686
Healthcare provider training	8 (72.7)	—	8 (72.7)	—	4 (28.6)	4 (28.6)	0.999
Public/community education	11 (100.0)	—	11 (100.0)	—	5 (35.7)	6 (42.9)	0.705
Smoke-free environments	8 (72.7)	—	8 (72.7)	—	5 (35.7)	3 (21.4)	0.064
Involved in influencing/developing policy on: (N, %)
Restricting tobacco ads and displays	12 (42.9)	0 (0.0)	12 (85.7)	<0.001	6 (42.9)	6 (42.9)	0.999
Restricting sales to youth	13 (46.4)	0 (0.0)	13 (92.9)	<0.001	7 (50.0)	6 (42.9)	0.705
Increasing tobacco taxes	1 (3.6)	0 (0.0)	1 (7.1)	0.309	1 (7.1)	0 (0.0)	0.309
Increasing youth possession penalties	1 (3.6)	0 (0.0)	1 (7.1)	0.309	1 (7.1)	0 (0.0)	0.309
Advancing smoke-free policies	8 (28.6)	0 (0.0)	8 (57.1)	0.001	4 (28.6)	4 (28.6)	0.999
None of the above	15 (53.6)	14 (50.0)	1 (7.1)	<0.001	7 (50.0)	8 (57.1)	0.705
Involved in enforcing any tobacco-related policies (N, %)	13 (46.4)	0 (0.0)	13 (92.9)	<0.001	6 (42.9)	7 (50.0)	0.705
Perceived support for tobacco control from: (M, SD)c
Regional government officials	1.1 (0.6)	0.8 (0.4)	1.3 (0.6)	0.030	1.0 (0.6)	1.1 (0.5)	0.510
Local government officials	1.2 (0.6)	0.9 (0.4)	1.5 (0.7)	0.003	1.1 (0.6)	1.3 (0.6)	0.364
Staff within your organization	1.1 (1.0)	0.1 (0.4)	2.0 (0.0)	<0.001	1.1 (1.0)	1.1 (1.0)	0.999
Community members	0.6 (0.7)	0.1 (0.4)	1.2 (0.6)	<0.001	0.6 (0.8)	0.6 (0.7)	0.999
NGOs or international organizations	0.9 (0.9)	0.3 (0.7)	1.6 (0.7)	<0.001	0.9 (1.0)	0.9 (1.0)	0.861
Religious leaders	0.5 (0.8)	0.0 (0.0)	1.2 (0.8)	<0.001	0.5 (0.8)	0.5 (0.8)	0.799

^aAmong the N = 11 with NCD programs addressing tobacco control.

^bCalculated among the N = 14 in each condition.

^cOn a scale of 0 = None to 2 = A lot. The number of participants indicating ‘Prefer not to answer’ were as follows: regional government officials: 1; community members: 2; non-governmental organizations or international organizations: 3 and religious leaders; 4.

Tobacco-related activities

Participants were asked: (i) whether their organization has any NCD programs that include tobacco control, and if so, to indicate the types of programs (see Table II); (ii) whether their organization is involved in influencing or developing policy or enforcement of policy in various areas; (iii) if they are involved in advancing smoke-free policies, to describe that work and (iv) whether their organization is involved in enforcing any tobacco-related policies, and if so, to describe their involvement. Participants were also asked to indicate their perceived support for tobacco control across various groups (e.g. local government, religious officials) on a scale of 0 = None to 2 = A lot, with a response option of ‘Prefer not to answer’ (see Table II).

Data analyses

Descriptive analyses were conducted, and bivariate analyses (i.e. Chi-squared for categorical variables; analysis of variance (ANOVA) tests for continuous variables) were used to compare communities across countries and communities across the intervention and control conditions. All analyses were conducted using SPSS v25.0, and alpha was set at 0.05.

Results

Participant and organizational characteristics

The majority of participants were 51–64 years old (N = 19, 67.9%), were female (N = 21, 75.0%), were employed by the organization for over five years (N = 26, 93.9%), had medical degrees (N = 18, 64.3%) and/or degrees in epidemiology (N = 17, 60.7%), and were never smokers (N = 22, 78.6%). Eleven (39.3%) respondents reported on public health centers, 14 (50.0%) were local branches of the National Center for Disease Control, with others representing municipalities, non-governmental organizations or other entities, with an average of 2.2 [standard deviation (SD) = 1.4] full-time staff (Table I). The average population of the 28 communities was 24, 114.3 (SD = 11745.7).

Table I.

Cross-country and cross-condition comparisons of participant and organizational characteristics

	All	Armenia	Georgia	P	Control	Intervention	P
Participant characteristics
Age (N, %)				0.591			0.501
36–50 years	8 (28.6)	4 (28.6)	4 (28.6)		4 (28.6)	4 (28.6)
51–65 years	19 (67.9)	10 (71.4)	9 (64.3)		9 (64.3)	10 (71.4)
Older than 65 years	1 (3.6)	0 (0.0)	1 (7.1)		1 (7.1)	0 (0.0)
Sex (N, %)				0.663			0.663
Male	7 (25.0)	4 (28.6)	3 (21.4)		4 (28.6)	3 (21.4)
Female	21 (75.0)	10 (71.4)	11 (78.6)		10 (71.4)	11 (78.6)
Length of employment (N, %)				<0.001			0.926
<1 year	—	—	—		—	—
1–2 years	2 (7.1)	2 (14.3)	0 (0.0)		1 (7.1)	1 (7.1)
2–5 years	—	—	—		—	—
5–10 years	13 (46.4)	12 (85.7)	1 (7.1)		7 (50.0)	6 (42.9)
>10 years	13 (46.4)	0 (0.0)	13 (92.9)		6 (42.9)	7 (50.0)
Educational background (N, %)
Medical degree	18 (64.3)	4 (28.6)	14 (100.0)	<0.001	9 (64.3)	9 (64.3)	0.999
MPH	3 (10.7)	0 (0.0)	3 (21.4)	0.067	1 (7.1)	2 (14.3)	0.541
Epidemiology degree	17 (60.7)	8 (57.1)	9 (64.3)	0.699	9 (64.3)	8 (57.1)	0.699
Other master’s degree	1 (3.6)	0 (0.0)	1 (7.1)	0.309	0 (0.0)	1 (7.1)	0.309
Other	6 (21.4)	4 (28.6)	2 (14.3)	0.357	3 (21.4)	3 (21.4)	0.999
Smoking status (N, %)				0.654			0.717
Current smoker	3 (10.7)	2 (14.3)	1 (7.1)		1 (7.1)	2 (14.3)
Former smoker	3 (10.7)	2 (14.3)	1 (7.1)		2 (14.3)	1 (7.1)
Never smoker	22 (78.6)	10 (71.4)	12 (85.7)		11 (78.6)	11 (78.6)
Organization characteristics
Type of organization (N, %)				<0.001			0.543
Public health center	11 (39.3)	0 (0.0)	11 (78.6)		6 (42.9)	5 (35.7)
National Center for Disease Control local branch	14 (50.0)	13 (92.9)	1 (7.1)		7 (50.0)	7 (50.0)
Municipality	1 (3.6)	0 (0.0)	1 (7.1)		1 (7.1)	0 (0.0)
Non-governmental organization	1 (3.6)	0 (0.0)	1 (7.1)		0 (0.0)	1 (7.1)
Other	1 (3.6)	1 (7.1)	0 (0.0)		0 (0.0)	1 (7.1)
Full-time staff (M, SD)	2.2 (1.4)	2.7 (1.4)	2.0 (1.3)	0.323	2.3 (1.4)	2.1 (1.3)	0.730
Community population (M, SD)	24114.3 (11735.7)	19835.7 (10873.7)	28392.9 (11330.5)	0.052	24971.4 (12768.7)	23257.1 (11019.2)	0.707

NCD-related activities

Half or more of the communities had programs, services or activities addressing hypertension (50.0%), diabetes (53.6%) or cancer (57.1%; Table II). Georgian communities were more likely to have NCD programs addressing hypertension (P < 0.001) and chronic respiratory diseases (P = 0.006), while those in Armenia were more likely to have NCD programs addressing diabetes (P = 0.001), cancer (P = 0.002) and other programs (e.g. diet, alcohol use, P < 0.001).

The vast majority (85.7%) had NCD programs involving community education, with 50.0% having human papillomavirus (HPV) vaccination promotion programs and 32.1% having patient education programs. Georgian communities were more likely to have NCD programs that provided primary care or clinical services (P = 0.006), cancer screening (P = 0.031), tobacco cessation (P = 0.014) and patient education (P < 0.001). Armenian communities were more likely to have NCD programs providing HPV vaccination (P < 0.001) and other services (e.g. NCD epidemiological surveillance, prevention related to risk factors, population awareness programs, P < 0.001).

Almost half (42.9%) of the communities had other organizations in their communities working on NCD prevention and control. Of the eight Armenian communities indicating such organizations, these largely represented local non-governmental organizations (NGOs) and international organizations working on health care, youth, children and women’s issues (e.g. ‘Young Families Support Center’ NGO, ‘Young Generation’ NGO, World Vision’s regional branches and Red Cross regional offices). Of the four Georgian communities indicating such organizations, these included other public healthcare centers and involved activities around cancer screenings.

Three quarters (71.4%) worked with other groups on NCD prevention and control. Armenian organizations were more likely than those in Georgia to work with other organizations/partners on NCD prevention and control (P = 0.001). In Armenia, two collaborations universally reported included: (i) regional administration and the mayor’s office on NCD risk factors and (ii) medical facilities on epidemiological surveillance of chemical poisoning and injuries. In Georgia, these collaborations varied and included working with: (i) the NCDC on a campaign to increase blood pressure screening and reduce the impact of hypertension and (ii) local governmental and educational institutions (pre-schools and schools) for Enterobiasis and iodine deficiency screening.

Tobacco-related activities

Eleven communities (39.3%) addressed tobacco control, all of which were located in Georgia (Table II). Of those, all included public/community education, and nearly all provided cessation counseling or classes (81.8%), school/youth prevention programs (81.8%), healthcare provider training (72.7%) or activities addressing smoke-free environments (72.7%).

Of the 28 communities, 53.6% reported no involvement in any of the tobacco control policy efforts assessed; however, 46.4% were involved in influencing, developing, or enforcing policy on restricting sales to youth, 42.9% in restricting tobacco ads and displays and 28.6% in smoke-free policies. Thirteen (46.4%) of the communities reported being involved in enforcing any tobacco-related policy. Those communities involved in these activities were all located in Georgia. These activities included a broad range of initiatives related to the implementation and monitoring of the tobacco control policies implemented in 2018, specifically: educating the local community and small businesses regarding the tobacco control legislation that went into effect in May 2018 via public meetings; informative-consulting visits with stakeholders and target institutions; tobacco product retail monitoring; educational meetings with youth and prospective mothers on tobacco-relate harms; and monitoring the enforcement of and compliance with the smoke-free policy in various locations (e.g. retail stores, restaurants, public transportation, and public, administrative and educational facilities).

Average scores regarding perceived support for tobacco control were highest regarding support from local government officials and lowest from religious leaders and community members. The key informants in Georgia versus Armenia rated the support they perceived from the various groups as greater (P’s < 0.05, respectively).

Discussion

This paper provided baseline data comparing communities involved in an ongoing trial examining coalition processes and activities, particularly in relation to organizational activities and existing capacity to mobilize their communities toward social norms and policy change around tobacco and NCDs more broadly. No differences were documented between the intervention and control communities, a critical finding that establishes that the randomization was effective in establishing conditions with similar baseline characteristics deemed important for this trial. Moreover, these communities reported considerable local activity across a range of NCD-related public health issues, particularly cardiovascular disease, cancer, diabetes and HPV prevention, with the most prominent programs involving community and patient education. Also of note is that almost half had other organizations in their communities working on NCD prevention and control, and nearly three quarters worked with other groups on NCD prevention and control. Regardless of some differences across countries, findings from the current study indicate promising existing capacity to promote NCD-related activities across communities, which CCAT suggests may facilitate effectiveness [12, 13].

While no differences were found across conditions, some pronounced differences were found across the two countries, particularly regarding tobacco control activity. This is likely due to the more aggressive timeline and recent action related to tobacco control in Georgia, with its implementation of several progressive tobacco control measures in 2017–18. In addition, in 2018, Georgia was one of the 15 FCTC parties (and the only European country) selected to receive direct support under the FCTC 2030 project. In brief, FCTC 2030 started in February 23018 and involved technical and logistical support the new tobacco legislation, communication strategy development and implementation, fostering mobilization of stakeholders and dialog among them, and awareness raising about upcoming legislative changes among civil society, state structures, retailers, the hospitality industry, etc., among other supports and activities. This additional support in Georgia facilitated local public health agencies’ work related to the implementation of various tobacco control measures, including those related to the implementation of the smoke-free air policy.

While this opportunity has supported and catalyzed tobacco control efforts in Georgia, it has created differences in tobacco control activity between the two countries. Moreover, additional measures in Georgia regarding tobacco control more generally (e.g. plain cigarette packaging) and smoke-free air specifically (e.g. tobacco-free stadiums and hotels) are on the horizon. However, it is also critical to note pending legislative actions in Armenia to advance tobacco control legislation (e.g. pictorial health warnings) including smoke-free air policies (i.e. a comprehensive public policy).

Indeed, community-based research often faces challenges in terms of unexpected events that might occur and ‘noise’ that impacts the research design and interpretation of findings [28, 29]. Specific to the current study, irrespective of the coalition intervention being tested in this trial, these policy changes and external support for tobacco control will likely impact local community capacity, community mobilization, social norms related to tobacco and SHSe. Another concern that is more difficult to address is the possibility of a ceiling effect that might hinder interpretation of study results. More specifically, these external change agents could result in drastic reductions in SHSe, thus limiting the extent to which we will be able to detect differences in changes in SHSe from baseline to follow-up between the intervention and control conditions, particularly given the modest power of involving 28 communities and roughly 1500 community survey participants at each time point.

From this perspective, there are several ways in which the research design will yield important findings regardless. First, the nature of the randomized control trial is critical, as it allows isolation of the effects of the intervention. Second, examining baseline capacities allows us to account for the differences in community capacity across countries in subsequent analyses. Third, we will use CCAT as a framework for studying coalition processes and the types of advantages such external changes or support related to tobacco control might entail. Examining coalition processes within this context during dynamic changes will call for thoughtful ways in which to leverage this opportunity by integrating these evolving factors into study measures and contextualization of the results. Indeed, this multifaceted research design will yield findings that will contribute to the literature and inform CCAT [12] and future public health practice.

Conclusions

In conclusion, multi-sectoral local coalitions aligning civil society and local public health agencies is a well-documented strategy for effecting policy change [1–6]. It is critical to understanding coalition characteristics and processes, as well as sociocontextual factors (e.g. political context), that foster success in local coalitions in communities with limited resources such as LMICs [27]. This paper serves as a foundation to provide context for subsequent analyses related to an ongoing, long-term community-randomized control trial testing the impact of local coalitions in LMICs to effect policy change and enforcement, specifically focused on smoke-free air. This initial set of analyses indicated that the randomization of the 28 cities yielded no differences in important community and organizational factors relevant to the implementation of the trial. It also highlights the types of NCD- and tobacco-related activities taking place in Armenia and Georgia, which might provide some indication of such activities in the region more broadly. Ultimately, the parent study will serve as a catalyst for future research and evidence-based practice to support tobacco control progress in LMICs and more broadly.