Recent analyses of the global burden of disease attributable to mental disorders have called attention to the magnitude of this burden and the urgent need for improved systems of prevention and treatment of these diseases [1]. Yet health systems globally are failing to meet this need, [2] especially in low-and middle-income countries (LMICs) [3, 4]. Potential effects of untreated mental disorders include: personal suffering and distress, poor social functioning, lower productivity, and increased likelihood of physical illnesses – particularly chronic conditions for which treatment adherence may be impacted [5, 6]. While there is an increasing evidence base of cost-effective interventions for mental disorders, the majority of people with mental disorders in LMICs do not receive treatment [7]. Effective treatment of mental disorders may be especially challenging in LMICs currently or recently engaged in armed conflict, where the likelihood of mental disorders such as post-traumatic stress disorder, depression and anxiety is increased and health systems are destabilized [8–10]. Despite the well-documented high-levels of mental health needs among conflict-affected populations, [8, 9] there is much less evidence on access to services for mental health care services for conflict-affected persons in LMICs (which is where the vast majority of conflict-affected persons live), with only a few studies explicitly investigating this [11–15]. What research does exist from conflict-affected settings in the former Soviet Union region has revealed large gaps in treatment of mental disorders. For example, one study in Ukraine found that 74% of IDPs who likely required mental health and psycho-social support did not receive it [15]. Another study, from Kosovo, among female civilians 10 years after the war found that more than half used health care services during the previous three months but only a small minority used specialist mental health services [16]. There is also little evidence on the role of health systems in responding to the mental health care needs of conflict-affected populations, [17] despite its importance.

One conflict-affected population where research on mental health care access is needed is in the Republic of Georgia. Georgia has experienced two main phases of conflict, each of which involved secessionist movements [18, 19]. The first phase occurred in the early 1990s, in the regions of Abkhazia and Shida Qartli (South Ossetia), and led to the internal displacement of approximately 300,000 people, of whom approximately 200,000 remain as internally displaced persons (IDPs). The second phase occurred in August 2008, due to conflict between Georgia and the Russian Federation over Shida Qartli. This phase led to at least 128,000 ethnic Georgians being internally displaced, of whom up to 100,000 have now returned to their home areas. Some 200,000 Georgians therefore, remain as IDPs [20]. Approximately 40% of these live in collective centres, with the remaining IDPs in private accommodation. Collective centres are usually made up of former public or administrative buildings, or purpose-built villages constructed by the government after the 2008 conflict. These IDP communities are characterised by poor living conditions, high unemployment, poverty, limited integration with local communities and financial barriers to access health care and medicines. Recent estimates of the prevalence of mental health disorders among the IDP population found a persistent high burden of psychiatric symptoms and disability. For example, the estimated prevalence of post-traumatic stress disorder (PTSD) among IDPs in Georgia is 27.1% (1990s IDPs) and 22.9% (2008 IDPs); for depression these figures are 18.7 and 9.9% and for anxiety 13.0% and 9.2% [19] Despite this evident need for services however, research on utilization of mental health care among IDPs in Georgia found that only just over a third of those with a current mental disorder sought any assistance from health services. Approximately 27% did not seek care because they did not self-report having problems, while roughly 33% did report problems but cited barriers to accessing care, most commonly not being able to afford it [13].

Given the large treatment gap uncovered in Georgia among IDPs with symptoms of mental disorders, it is clear that barriers remain in access to treatment of mental disorders among IDPs, and that care users with mental disorders may not be accessing adequate care. The aim of this study is to explore how mental health care for IDPs is organized in Georgia, and to identify both user and provider-side barriers that may impede access to primary and secondary mental health care services for this population. As our data collection occurs five years after the most recent phase of conflict, we focus on access to routine mental health care services (i.e. those that are available to the general population) rather than designated mental health and psycho-social support (MHPSS) services that are usually provided in the immediate aftermath of a humanitarian emergency.

This study used rapid appraisal (RA) methods to explore barriers to access to mental health care among adult IDPs in Georgia. The RA approach involves collection and triangulation of multiple data sources to provide an understanding of a situation in a more timely and cost-effective manner than standard social research methods, and to seek a diverse range of perspectives, without aiming for statistical precision [21]. Triangulation provides internal validity and reliability of the data collected, [22] and these data can then be used to develop specific health care interventions [21, 22]. Rapid appraisals of health systems have the advantage of requiring only limited time and resources while still providing reliable findings and have previously been successfully used for understanding health system performance for non-communicable diseases (NCDs) such as diabetes, [22, 23], including in Georgia [24, 25]. First, we undertook an initial review of the system for mental health care in Georgia using available documents. The document review involved examination of available policy documents, studies and evaluation reports [13, 18–20, 24, 26–31] with the aim of obtaining a comprehensive understanding of the organization and implementation of mental health care in the context of the national health system, and in particular the existing mental health service structure for IDPs in Georgia.

Then, to better understand how the mental health care system for IDPs works in practice and where the barriers may lie, we conducted semi-structured interviews with stakeholders from a range of perspectives - governmental and non-governmental actors, mental health care providers and mental health care users themselves. Participants from the first two groups were identified by snowball sampling i.e. identifying initial key stakeholders to invite as participants, who then provides the name of a subsequent participants, and so on.

Mental health care users were sampled from the Mtskheta-Mtianeti region, about 30 k north-west of Tbilisi and home to the Tserovani IDP settlement, the largest settlement of 2008 IDPs, as well as from the city of Gori, which was heavily affected by the 2008 conflict and is home to many IDPs living in collective centres. IDPs were sampled using convenience sampling (sampling based on accessibility and ease). Specifically, users were recruited by asking health professionals involved in the study to approach users who came in for mental health care and/or to call existing users by phone and invite them to participate in the study. This approach was chosen as it was more practical and likely to result in participation than attempting to identify participants with mental disorders in the community. Providers were asked to recruit users with a diverse set of mental disorders in order to capture various perspectives.

The final number of key informants was n = 28. All interviews were conducted in Georgian using a semi-structured interview guide. Semi-structured topic guides allow for the interview to be participant-driven and are thus useful for exploration of an individual’s perception of a particular phenomenon [32]. Topic guides covered the burden of mental health care in Georgia, the organization of provision and financing of mental health care services for IDPs in Georgia, the user pathway, including diagnosis, treatment and follow-up, pharmaceuticals and communication of knowledge regarding mental health care to IDPs. The guides for each of the interview participant categories can be found in Additional files 1, 2 and 3.

Interviews were conducted in October 2013 in Tbilisi (the capital city of Georgia) and in Mtskheta-Mtianeti region. They were conducted in participants’ offices (in the case of government or NGO stakeholders and health professionals), or in a private room in the health facility (in the case of care users). Prior to inclusion in the study, all participants were given an information sheet and consent form in their native language. Each interview lasted approximately 60 min. Interviews were conducted by trained researchers from Curatio International Foundation (Tbilisi), who have experience conducting research with IDPs about sensitive issues. Interviews with care users were conducted in their homes, those with providers were conducted at their place of work. All interviews were audio-recorded, transcribed and translated into English by Curatio researchers. Interview data were then analysed using inductive thematic analysis methods, Inductive thematic analysis involves coding data for emerging themes and concepts without trying to accommodate a hypothetical framework, and is thus ‘data driven’ and reflexive [33, 34] Analysis followed the steps to thematic analysis outlined by Braun and Clarke [33] and emerging themes were triangulated with data from the above mentioned document review. All analyses were conducted using NVivo 10™.

The rate of mental health care utilisation among conflict-affected IDPs in Georgia is low. This paper is the first to use qualitative methods to provide insight into some of the barriers that may affect access to treatment and thus utilization rates in this population. The barriers that emerged as important appear consistent between patients and providers and other stakeholders, but experienced in different ways at different levels of the health system. Lack of comprehensive insurance coverage, case identification and referral of patients with mental disorders (resulting in out-of-pocket payments), and poor funding for mental health services emerged as two key system barriers to effective access to treatment for mental disorders in Georgia. While some of these barriers, in particular out-of-pocket payments for mental health care and medicines, have also been observed among the general population of Georgia, [29] it is likely that these barriers disproportionately affect IDPs given the high prevalence of mental disorders in the Georgian IDP population, and the generally low socio-economic status of this group. Other research suggests that stigma as a barrier to accessing care may be particularly relevant in IDP or refugee communities due to a history of fear and repression [40].

These barriers also reflect those observed with non-conflict-affected populations globally and in the region, related to under-investment in mental health financing, staffing, and facilities, and the way in which services are organized [5, 41–49]. The WHO has developed guidelines for policy-makers to support adequate financing of mental health care, stating that where possible, coverage of mental health services should be mandatory, through national, tax-based or social insurance [50]. These guidelines also identify factors that can result in government underfunding of mental health care, including inadequate recognition of the importance of mental disorders and their consequences for those who suffer from them and their families, and the failure of policy-makers to understand the potential effectiveness of interventions for mental disorders, resulting in an assumption that other services are more beneficial to the population [50]. The extent to which these factors are relevant in the Georgian context requires further investigation.

While, coverage, funding, human resources and other structural (or macro) barriers appear important, attitudinal barriers, specifically stigma associated with mental disorders, also plays a key role in preventing access to appropriate treatment. Our results are not dissimilar to results of World Mental Health Survey from 17 countries which showed that although structural barriers to accessing effective treatment were more prevalent among severe cases of mental disorders, attitudinal barriers were more important to initiating and continuing treatment for mild-moderate cases of mental ill health [51]. They are also not dissimilar from recent research on IDPs in Ukraine, where in addition to not being able to afford care and low awareness, trust or geographic access to high-quality services, stigma and embarrassment were found to be important barriers to using MHPSS [15].

One possible strategy for addressing these barriers is the integration of mental health care for a wider range of disorders, including anxiety and depression, into PHC services, with capacity-building of PHC personnel. There is evidence that integration of mental health care into PHC is cost-effective, improves access to care and reduces stigma and discrimination [52]. Integration is recommended by WHO [53] and a programme of integration of a basic package of mental heatlh care services into PHC is currently being evaluated in five low- and middle-income countries by the PRIME study [7]. Another approach for which there is some evidence of effectiveness in conflict or post-conflict settings creating new health worker cadres to fill service gaps. Health workers recruited from local communities can ensure long term retention and encourage culturally appropriate and acceptable interventions, [17] thus addressing human resource and stigma barriers. This approach could also address our finding of a low level of knowledge and skills required to diagnose and manage people with mild mental health illnesses, especially in primary care settings. Opportunities for reducing stigma might also include community-based approaches such as those advocated by UNHCR, that engage mental health care users and communities in promoting their own health and well-being, and use participatory assessment (dialogues with affected persons to understand needs and resources, and develop strategies to better address needs), to identify groups that need targeted attention [54]. The WHO and its partner agencies have also produced guidelines for low-intensity interventions aimed reducing symptoms of common mental disorders through behavioural strategies, and that can be delivered by laypeople from the affected communities, without formal mental health training [55]. Whether or not such approaches would be appropriate in the Georgian context requires investigation.

Our study has also contributed to the growing body of work demonstrating the potential value of RA as a tool understanding barriers to mental health treatment among conflict-affected populations [56–60]. Importantly, the study has taken a health systems approach, investigating potential barriers at all levels of the system, from government to providers and service users. Guidelines for mental health care assessment from the Inter-Agency Standing Committee recommend collection of health systems information for a comprehensive overview of mental health care, [39] RAs of barriers at various levels of the health system are rare. We have shown that with few resources and within a short time frame, RA methodology can provide a comprehensive overview of health system barriers to effective treatment of mental disorders among conflict-affected populations. In the early stages of conflict, RA might provide a useful tool for understanding mental health needs and health system barriers to care sooner, allowing for these to be addressed in a more timely manner than usual.

Inadequate insurance coverage, case identification and referral to treatment for mental disorders, underfunding, shortage of human resources, poor information systems, patient out-of-pocket payments and stigmatization all appear to be important factors affecting effective provision of mental health care services to IDPs in Georgia. The appropriateness and potential effectiveness of policy interventions such as insurance coverage of a wider range of mental disorders, integration of services for these at the primary health care level, and community-based approaches in this context should be explored.