Approaches to deliver depression care and understanding barriers and facilitators to implementation in people with tuberculosis in LMICs: a systematic review

Article history

The contractual start date for this research was in April 2018. This article began editorial review in September 2022 and was accepted for publication in July 2023. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The Global Health Research editors and publisher have tried to ensure the accuracy of the authors’ article and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this article.

Copyright statement

Copyright © 2024 Nava-Ruelas et al. This work was produced by Nava-Ruelas et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

2024 Nava-Ruelas et al.

Eligibility criteria

The inclusion criteria are described in Table 1. We included studies conducted in LMICs, according to the World Bank classification at the time of the searches, in primary, secondary or tertiary care, with the characteristics listed in Table 1.

Search strategy

We designed and carried out the searches with advice from the librarian of the Health Sciences Department at the University of York. One reviewer (RNR) carried out the searches during June–July 2020 and covered all databases from inception. We searched 10 electronic databases for peer-reviewed and grey literature: MEDLINE, Scopus, PsycINFO, Web of Science, PubMed, EMBASE, Cochrane Library, SciELO, LILACS and Health Management Information Consortium. We searched the websites of relevant organisations (World Bank, Global Fund to Fight AIDS, Tuberculosis and Malaria and World Health Organization) using the Google Advanced Search feature. We considered these organisations important as they are major stakeholders in the delivery and funding of TB services in LMICs. 14 We also hand-searched the reference lists of the included studies to identify other relevant studies. An example of the search strategy used across databases is included in Appendix 1.

Quality assessment

We used the Mixed-Methods Appraisal Tool (MMAT) to appraise the quality of eligible studies. 15 We used this tool anticipating that the included studies would either be qualitative, quantitative or mixed methods studies. The advantage of the MMAT tool is that it provides a framework to assess qualitative, quantitative and mixed methods studies, following a minimum set of quality criteria for the methodological design. 16 This tool was developed in response to the inclusion of different types of evidence into systematic reviews for health services research. 17 This facilitates their integration into systematic reviews as it allows a comparison of overall quality between different methodological stances according to the quality criteria they have met.

Data extraction

Two reviewers (RNR, OT) participated in the screening and data extraction process. We used Rayyan software for the title and abstract screening phases. 18 We resolved disagreements through consensus and discussion before proceeding to the next stage. Two reviewers (RNR, OT) independently performed the full-text screening. One reviewer (RNR) extracted the data of the included studies, and a second reviewer (OT) cross-checked the data extracted against the original study to ensure consistency and reliability for two studies.

Data were extracted from the results and discussion sections of the included studies (see Appendix 2). We extracted the description of the study design, the type of intervention and the implementation process from the background and methods sections of the included studies (see Table 2; further descriptive information on the interventions and their implementation is described in Appendix 5).

To describe the interventions implemented, we used the description as it was stated in the included studies. After this, we classified the studies according to the main categories of interventions used for depression, according to the NICE guidelines19 and based on the descriptions provided in the studies. We created four intervention categories based on the descriptions provided: (1) psychological-only, (2) combined, (3) combined, with socioeconomic support and (4) screening-only. We defined psychological-only interventions as any type of group or individual therapy used to treat mental health conditions, such as depression, anxiety or psychosis (e.g. behavioural therapy, interpersonal therapy, cognitive–behavioural therapy, psychodynamic therapies, family therapy). We defined combined interventions as psychological and pharmacological interventions delivered simultaneously to the patient. We defined combined interventions with socioeconomic support as combined interventions provided alongside any form of support that aimed to address socioeconomic needs in people with TB. We defined screening-only interventions as screening for mental health conditions, followed by referral to mental health services outside of TB services, if needed.

In addition to the descriptive characteristics of the studies, we extracted qualitative and quantitative data, as appropriate, from the studies that described the interventions implemented for data analysis and synthesis. From the qualitative studies, or studies with qualitative components, we extracted the narrative descriptions of the intervention, such as the number and type of study participants, therapeutic strategies, the thematic content of sessions, narrative descriptions of participants’ responses to the interventions, interview responses and other excerpts verbatim from the articles. From the quantitative studies, or studies with quantitative components, we extracted narrative descriptions of the intervention and participants’ experiences (if available); numerical data on the type and number of participants, number of participants who received (or not) the intervention, number of participants who received training (if applicable), number of participants who adhered (or not) to the intervention, depression scores at baseline, depressions scores at the end of the intervention, number of death and descriptive statistics about the characteristics of participants, process outcomes (e.g. the number and type of interventions delivered, enrolment rate) and clinical outcomes (e.g. depression scores, TB treatment outcomes, lost-to-follow-up rates). From the mixed methods studies, we extracted both quantitative and qualitative data. A detailed description on the qualitative and quantitative data we extracted from each study is given in Appendix 2, where the qualitative data are described in Appendix 2, Table 5 and the quantitative data are described in Appendix 2, Table 6.

Data synthesis

We coded the quantitative and qualitative data from all studies according to the Consolidated Framework for Implementation Research (CFIR). This framework was chosen for analysis given its appropriateness to explore the facilitators and barriers to implementing diverse types of interventions. 20 This framework encompasses five main constructs (intervention characteristics, inner setting, outer setting, characteristics of individuals involved and process), divided into 26 subconstructs.

After extracting data from the eligible studies, we converted quantitative data [e.g. frequency tables, forest plots, ratios, number of participants who received (or not) the intervention] to qualitative data by creating narrative descriptions of the quantitative data and results. This process of qualitising data is used in mixed methods literature reviews to synthesise diverse types of evidence. 16 We coded the narrative descriptions of quantitative data along with the qualitative data. We used an analytical approach that was based on the framework method:21 we treated all data as primary qualitative data and coded this information in each study as barriers or facilitators. To create framework matrices, we used the CFIR constructs and subconstructs as an a priori coding framework, organised in columns while the different intervention categories were organised in rows. The data in each cell were verbatim extracts of qualitative data or narrative descriptions of quantitative results from the included studies. We defined the inner setting as the TB national services and/or where healthcare delivery for TB was routinely provided. We defined the outer setting as other governmental or private healthcare services provided within the country, or any other national and international institutions collaborating with TB national services.

The CFIR rating rules state that the coding process should consider the positive or negative influence that coded data have on implementation, and how weak or strong that influence is on implementation. 22 To determine valence, a rating of ‘X, ‘0’, ‘+’ or ‘− is assigned to the coded data. These symbols mean that the coded data had a mixed (positive and negative), null, positive or negative influence on implementation, respectively. To determine strength, a rating of ‘1’ or ‘2’ is assigned to the coded data. A notation of ‘1’ means that the coded data have a weak influence on implementation while ‘2’ means its influence is stronger. To determine the valence of the influence, we considered how many times the narrative data were mentioned in the study, the adjectives used to describe the influence of a construct (e.g. ‘very strong’, ‘very important’, ‘not strong’, ‘not important’), and whether the influence it had on implementation was negative, positive or mixed.

We adapted the CFIR construct definitions to the research objectives of the study. We used the definitions provided in the handbook and the website, and then re-interpreted them according to their relevance to TB services and mental health interventions (see Appendix 4).

Search results

The total number of records retrieved from electronic databases was 10,982. After deduplication,23 two reviewers (RNR, OT) independently screened the titles and abstracts for 8858 records. We recorded our screening process using PRISMA guidelines (see Figure 1). 24

FIGURE 1.

PRISMA figure. 24

Thirty-nine articles were included in the full-text screening phase. At this stage, 11 articles were excluded because they were based in high-income countries, mostly the USA. One study was excluded because it did not include people with pulmonary TB. Two studies were excluded because they were prevalence studies. Five studies were excluded because they were discussion papers or reports. Eight studies were excluded because they did not describe interventions to treat depression – they focused on adherence or psychosis without depression as a comorbidity. Two articles were excluded because it was not possible to establish contact with the authors or national libraries in Brazil and Bolivia to access the full text. We give the full list of references for the excluded studies in Appendix 6.

Characteristics of the included studies

Ten articles met the inclusion criteria and were included in the analysis. They varied in terms of study design. One was a qualitative study, three were retrospective cohort studies, one was a controlled-cohort study, two were cluster randomised trials and three were mixed methods studies. 25–34 More details of the included studies are given in Table 2.

Four studies described a psychological intervention. 25,30,31,34 Three studies described interventions that combined psychological and pharmacological approaches. 27,28,33 Two studies described a pharmacological and psychological intervention that also included socioeconomic support (combined, with socioeconomic support). 26,29 One intervention consisted of screening only followed by referral to mental health services, when needed. 32

Quality assessment

Table 3 describes the results of a quality assessment using the MMAT tool. We assessed the included studies against the appropriate methodological quality criteria according to the study design. Given the research objectives of this review, we did not exclude any of the studies based on their methodological quality. The purpose of the quality assessment was to explore whether the findings from the included studies were of sufficient methodological quality. All the studies had appropriate study designs for their stated research questions, except for one. 29 In this study, we considered that the study design (quantitative non-randomised) was not appropriate for the research questions that the authors described in the study, which were to examine the effectiveness of a combined intervention, with socioeconomic support, delivered by pharmacists. 29

Description of approaches to deliver depression care as part of TB services

The 10 studies included in this analysis describe four main types of interventions that aim to provide depression diagnosis and/or treatment or referral to mental health services to people with TB in LMICs at the point of care for TB services. The four major approaches used to deliver depression care for people with TB in LMICs were psychological interventions, combined (pharmacological and psychological) interventions, combined interventions with socioeconomic support and screening-only interventions. See Appendix 5 for a detailed description of the interventions, components and resources required for implementation.

The most commonly implemented intervention was psychological. Nine out of 10 studies included in this review described the implementation of psychological interventions, or interventions with a psychological component, alone or in combination with other approaches; one study did not describe the implementation of a psychological intervention. 32 Psychological interventions were described as counselling or psychotherapy. The most common intervention for depression was counselling. 26–30,33,34 Counselling was delivered in interventions that implemented psychological interventions only, combined interventions or combined interventions with socioeconomic support. Counselling was defined as motivational and emotional counselling, patient counselling or psychological counselling. 27,30,34 It was also defined as TB counselling. 30 Most counselling sessions were individual, except when also conducted with family members and/or patient groups. 30,34 The content of counselling sessions was not described in most interventions. Only one study described the content of the counselling sessions as based on behavioural activation. 34

None of the interventions used pharmacological treatment alone; all pharmacological interventions were part of a combined strategy, incorporating psychological and socioeconomic support components.

Two interventions included socioeconomic support components combined with psychological and pharmacological interventions. 26,29 Socioeconomic support consisted of transport subsidies, food, fees for medical tests, school fees for family members and covered diagnostic tests’ fees for other comorbidities. Both studies justified the inclusion of socioeconomic support given the socioeconomic needs of people with TB. Both studies mentioned that the socioeconomic support was provided by non-governmental organisations (NGOs). 26,29

The interventions were described inconsistently. Information on treatments’ length, dose and end points was often missing from the published data in all the included studies. The studies that reported on psychological interventions, combined interventions and combined interventions with socioeconomic support did not report on treatment length or the number of psychological sessions received by participants. The frequency of psychological treatment sessions was mentioned in four studies. 25,29–31 Of the three combined interventions, one study provided detail about the combination of pharmacological therapy with psychotherapy given to patients, which were selective serotonin reuptake inhibitor, tricyclic antidepressants or benzodiazepines;28 the dosages were not reported.

There was also variability in the instruments used to measure and diagnose depression; some studies described the use of more than one tool. 26,34 All studies reported measurements of depression scores at baseline. Baseline referred to the beginning of the intervention for depression in some studies, while in others it referred to the beginning of TB treatment. We found the same patterns in end point measurements. End-point measurements could refer to the end of the depression intervention or the end of the TB treatment. We extracted these data as reported in the studies (see Appendix 2). The follow-up measurements of depression scores were taken at different points in all the interventions; one study had no follow-up measurement. 32

All the interventions were implemented as part of routine public TB services and were often given at the same location and time as TB-service delivery, except in two interventions where patients received mental health services elsewhere. 32,33 Where applicable, the interventions used routine TB care as a control or comparator. 29–31 All the interventions relied on existing TB care pathways to incorporate screening and treatment for depression. All the interventions were provided free of charge for participants at the point of service delivery.

All the interventions relied on multidisciplinary teams to deliver the depression intervention. The teams comprised TB healthcare workers, primary healthcare workers with training in mental health or mental health specialists (i.e. psychologists or psychiatrists). In the studies that described the number and type of providers involved, there was a higher number of community or lay healthcare workers in comparison to the mental health specialists within the team. 25–27,33,34 When community or lay healthcare workers were involved in the intervention delivery, this was due to their role in the existing TB service-delivery structure, as they were often responsible for diagnosis and follow-up for people with TB and were often their first and only contact with TB services.

Most studies described how they required additional human or physical resources to those routinely used or available in the existing TB services. For example, most interventions described the requirement of additional human resources to deliver the intervention; the exceptions to this were combined and screening-only interventions where they used routinely available human resources. 27,28,32,33 Studies describing psychological interventions described the need for mental health specialists (psychologists, psychiatrists or both), social workers or additional community healthcare workers to deliver the intervention, working alongside routinely available TB healthcare workers. 25,30,31,34

All the interventions required community or lay healthcare workers to receive training for screening for depression and delivering the described intervention. The training components were described unevenly in all the studies. In some studies, it was not specified whether the training that the healthcare workers underwent was delivered as part of the intervention or before its implementation. 26–28,30 In some studies, training was only provided for the use of diagnostic tools. 31 In two studies, the training content or length was not described at all. 25,29 In the studies that described the training components, these usually covered the use of diagnostic tools and the provision of mental healthcare and were targeted to TB healthcare workers. 28,33,34 One study mentioned training the TB healthcare workers for a total period of 6 days. 34

All the interventions required the use of additional physical spaces to those routinely available in TB services in which to deliver the diagnosis or follow-up sessions for the intervention, for groups or individuals. Other requirements that were specific to the types of intervention were the use of IT infrastructure or mobile data plans to facilitate communication between patients and healthcare workers or between healthcare workers and the implementation team. 30,32 In the case of interventions that delivered antidepressants as part of the treatment, these were offered to patients at no cost. 27,28 Only one study provided in-kind incentives for community healthcare workers to participate in the intervention delivery. 32

Barriers and facilitators to delivering depression care as part of routine TB services

There was a lot of variability in terms of what represented barriers or facilitators to implementation across the intervention types and CFIR domains.

In Table 4, we describe an overview of the barriers and facilitators to the implementation of the interventions according to the CFIR framework, along with information on the constructs where there were no data available. The table includes a definition of the CFIR subconstructs and a summary of their role as barriers or facilitators to the implementation of each type of intervention for depression.

Approaches to deliver depression care as part of TB services

There were different approaches to deliver depression care for people with TB in LMICs. The most common approach was the implementation of psychological components, alone or in combination with pharmacological components and with socioeconomic support. This is consistent with the findings of other systematic reviews for interventions that integrate mental health into primary care in LMICs. 36 Psychological and pharmacological interventions for depression have the potential to be delivered in resource-constrained settings by providing training to non-specialist mental healthcare providers and using task-shifting approaches, achieving highly effective results when delivered by non-specialists alone or supervised by mental health specialists. 37 However, the implementation process will inevitably require additional physical and training-related resources.

Barriers and facilitators to implementation

The provision of mental health training to lay or community healthcare workers was an important facilitator for implementation. This finding is in line with the results of other systematic reviews: it is essential to provide training to lay healthcare workers or community healthcare workers to provide mental health care. 38 The type of training and the number of sessions required will vary according to the healthcare workers’ baseline characteristics and their previous knowledge of mental health and mental health interventions. There is some preliminary research on the activities within training that are necessary to implement sustained change, such as the formal appointment of supervisors and self-reflection or evaluation of fidelity to the intervention, to name a few. 39 Training content should cover basic information about mental health conditions, interventions, terminology and diagnostic and referral criteria. Also, the provision of mental health and TB knowledge to patients, families and communities facilitates implementation. Other studies have similar findings: when mental health interventions facilitate interaction between providers, beneficiaries and their families, in culturally appropriate ways, it facilitates the implementation process and creates a positive experience for those involved. 40 When healthcare workers felt they lacked the skills to address mental health in their patients, they regarded this as a barrier to integrating mental health assessments or related activities in their practice. 41

All interventions and their implementation processes incurred costs, mainly related to the use of human resources, training and physical spaces. This finding is consistent with the results of a systematic review that analysed the effectiveness and cost-effectiveness of integrating mental health care in LMICs, where the findings of the study show that all intervention arms increased the effectiveness of depression treatment and costs. 36 The cost of the intervention and its implementation can be attributed to the legacy of insufficient funds allocated to mental health care. 11 ‘Given the low levels of spending on mental healthcare … the finding that increasing the availability of mental health services increases direct costs should not surprise’ (p. 48). 36 Based on evidence from 42 studies, the integration of mental health services for depression in primary care can result in an immediate increase in health costs at a service level, mainly due to the increased use of human resources and the cost of treatment but is cost-effective from a healthcare perspective or reduces costs from a societal perspective. 36

To date, there is not a systematic review that analyses the costs of implementation of depression interventions in LMICs, which makes it hard to assess what resources are needed for implementation. In all the included studies, it was mentioned that there were not enough resources available for TB services, which will impact the feasibility of the implementation of mental health interventions in routine TB services. This tension overlaps with the availability of human resources for the delivery of mental health services. It is often not enough to train existing healthcare workers to deliver mental health care in the context of a poorly resourced workforce. 40 Further resources should be created and earmarked to facilitate the training of healthcare professionals in mental health care, add dedicated healthcare workers to provide mental health care, add specific mental healthcare tasks to existing care pathways or use a care co-ordinator. 36 The findings of this review suggest that the additional workload related to mental health care for existing healthcare workers is likely to be unfeasible and unsustainable within existing TB programmes. TB services in LMICs are well-established and have dedicated resources apart from wider health systems’ structures. 42 Thus, in this context, it would be more effective to earmark existing resources to train TB healthcare workers on mental health care delivery or use care co-ordinators to integrate mental health care into people with TB’s care. Mental health interventions in LMICs can be more effectively implemented if they adhere to existing care pathways and service delivery networks. 40

The provision of training alone is not sufficient if this is not accompanied by earmarked resources that create the space to deliver these interventions in existing service delivery structures, including considerations to increase the number of providers. TB services, as part of their aim to provide patient-centred care,10 could frame that as a window of opportunity to create said space and use existing resources to deliver mental health interventions for people with TB as part of their routine care.

Greater understanding and information on the organisational characteristics of TB services, such as internal incentives and rewards, the role of peer pressure, the learning climate or individuals’ identification with the organisation, can help to inform how to integrate care for mental health into TB services so that they can be feasibly and sustainably delivered. Understanding organisational characteristics can help with planning adequate interventions and implementation strategies. For example, interventions for depression that include financial incentives can help to alleviate the economic vulnerability of people with TB in LMICs. 43,44 This is important to explore because of the way that TB care is financed and provided in LMICs, including a range of organisations and collaborations. 14 Published peer-reviewed systematic reviews hint that health system characteristics, such as funding structures and available human resources, can become important barriers to implementation if they are not addressed before implementation stages, or if there are not discrete resources allocated to the integration of mental health care in existing health services. 41

Future research

This review includes studies that were published or carried out before the COVID-19 pandemic. Thus, the results presented here are likely to not be reflective of the current status of mental health and TB services in the countries reflected in the studies included in this review. A feasible way to improve this review would be to include an analysis of how the TB and mental health services of the countries represented have been affected by the pandemic.

Lessons learnt

The availability and reliability of the information on implementation efforts depend more on the study design (e.g. feasibility and acceptability) rather than on the intervention type being implemented (e.g. psychological, combined). Because the nature of the majority of the studies was not focused on implementation, that meant that a lot of cells within the framework matrix were empty (no data) for most intervention types and studies, except for two which had implementation-focused study designs. 32,45 This is consistent with other implementation reviews that focus on depression interventions in LMICs. 46,47 Future studies should be based on implementation-based questions such as the feasibility, acceptability and sustainability of the intervention. Future research should focus on assessing the interventions’ effectiveness and their implementation in resource-constrained settings, which would focus on the later implementation stages’ outcomes, such as cost, penetration and sustainability. 46

Community engagement involvement

There was no patient and public involvement in this study. We considered this to be inappropriate given the objective of the study, which was to synthesise the available evidence.

Limitations

The CFIR framework may not be able to capture the full complexity of TB services and their impact on mental healthcare delivery in LMICs. Although this framework was useful in identifying barriers and facilitators, there were characteristics from local health systems that could not be coded to the existing constructs in the framework. For example, national and international donors for TB and mental health programmes can influence the interventions that are implemented, based on their preferences and agendas. 40 Adaptations that make the CFIR framework more relevant to LMICs’ contexts are possible and should be explored as they can help assess barriers and facilitators that are prevalent in resource-constrained settings. 47 In line with the findings from other reviews,46,47 the next research priorities should move away from trying to test the effectiveness of interventions and rather focus on the implementation strategies and outcomes that can lead to successful implementation efforts for depression interventions in LMICs.

Due to the objectives of this study, the information analysed corresponds only to that reported in the published studies and is, therefore, subjected to reporting bias in the individual studies. Although a quality assessment was made according to each study type, this did not influence the weighting of how findings were represented in this review, as this study was concerned with providing a narrative synthesis of the approaches to depression care in TB services. Thus, this study should not be interpreted as evaluating the effectiveness of the interventions included in the analysis. Nonetheless, we are confident that we were able to identify and describe the various approaches that have been implemented to provide depression care in TB services in LMIC.

What this study adds

We identified psychological interventions as the most common approach used to deliver depression care for people with TB in LMICs. The findings from this review suggest that the implementation of interventions for depression is facilitated by the belief that the intervention is advantageous to the status quo (i.e. no provision of mental health care) (relative advantage), the compatibility of the intervention with existing workflows and the positive attitude that patients, families and communities display towards the intervention (innovation participants). All the interventions faced barriers to implementation related to the costs of the intervention and/or its implementation, which were expressed in terms of costs associated with insufficient human resources, lack of physical resources and lack of financial resources altogether for the provision of mental health care to people with TB. The choice of the best intervention to implement will depend on the available resources and the attitude of stakeholders to the intervention. From the evidence reported in the included evidence in this study, it appears that psychological interventions with some form of community engagement are the most compatible with local attitudes towards mental health interventions and can be delivered with existing or few additional resources, if training is provided. These are important factors that will facilitate the implementation of these types of interventions.

Key learning points

• There are barriers and facilitators faced by all types of interventions for depression implemented into existing TB services. However, different barriers and facilitators can arise according to the types of intervention implemented.

• Interventions to treat depression among TB patients are unlikely to be feasible without investment in human resources to deliver care.

• The implementation of any mental health intervention in TB services will require some degree of training. However, for the sustainable delivery of the intervention, mental health tasks and activities could be included in routine TB care or there could be a designated workforce for the delivery of this intervention.

Deviations from protocol

Outcomes of interest as stated in the protocol were defined as

those described in eligible studies, which could be process or clinical outcomes, for example: process outcomes (e.g. number of diagnostic tests performed, number of patients with access to the intervention to treat depression) or clinical outcomes (e.g. patients with negative sputum smear test, mortality, patients with reduced depression severity).

In practice, none of these data were extracted as they were not consistently reported across studies, and they were not required for the main analysis of barriers and facilit

Contributions of authors

Rocio Nava-Ruelas (https://orcid.org/0000-0001-5890-7519) (PhD Candidate, Health Sciences, University of York) contributed to the design of the study and conducted the literature searches, screening, analysis and writing of the report.

Olamide Todowede (https://orcid.org/0000-0002-5785-1426) (Research Fellow, Health Sciences, University of Nottingham) conducted the screening and collaborated with writing the report.

Najma Siddiqi (https://orcid.org/0000-0003-1794-2152) (Professor of Psychiatry, Health Sciences, University of York) contributed to the design of the study and the writing of the report.

Helen Elsey (https://orcid.org/0000-0003-4724-0581) (Senior Lecturer in Global Health, Health Sciences, University of York) contributed to the design of the study and the writing of the report.

Disclosure of interests

Full disclosure of interests: Completed ICMJE forms for all authors, including all related interests, are available in the toolkit on the NIHR Journals Library report publication page at https://doi.org/10.3310/GRWH1425.

Primary conflicts of interest: Rocio Nava-Ruelas received grant support from Consejo Nacional de Ciencia y Tecnologia for her doctoral studies. Olamide Todowede declares no competing interest. Najma Siddiqi was part of Grant 17/63/130 NIHR Global Health Research Group: Improving Outcomes in Mental and Physical Multimorbidity and Developing Research Capacity (IMPACT) in South Asia at the University of York. Helen Elsey declares no competing interest.

Data-sharing statement

Data can be obtained from the corresponding author upon request.

Ethics statement

This study did not require ethical approval as it did not involve human participants as data subjects.

Equality, diversity and inclusion

We designed our study to be of relevance to the population of interest (people with TB), regardless of their socioeconomic status, gender, ethnicity, religion or age. We believe that the study population from the included studies is reflective of the population of interest. We believe we have addressed EDI concerns pertaining to the study design, which is an evidence synthesis, in that we did not exclude any study based on the type of population as long as they met the principal inclusion criteria.

Department of Health and Social Care disclaimer

This report presents independent research commissioned by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, MRC, CCF, NETSCC, the GHR programme or the Department of Health and Social Care.

Study registration

The protocol of this review is registered with PROSPERO under registration number CRD42020201095.

Funding

This article presents independent research funded by the National Institute for Health and Care Research (NIHR) Global Health Research programme as award number 17/63/130 using UK aid from the UK Government to support global health research. The views expressed in this publication are those of the author(s) and not necessarily those of the NIHR or the UK government. Research is published in the NIHR Global Health Research Journal. See the NIHR Funding and Awards website for further award information.

About this article

The contractual start date for this research was in April 2018. This article began editorial review in September 2022 and was accepted for publication in July 2023. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The Global Health Research editors and publisher have tried to ensure the accuracy of the authors’ article and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this article.

This article was published based on current knowledge at the time and date of publication. NIHR is committed to being inclusive and will continually monitor best practice and guidance in relation to terminology and language to ensure that we remain relevant to our stakeholders.

Copyright

Copyright © 2024 Nava-Ruelas et al. This work was produced by Nava-Ruelas et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

Other barriers across interventions

Complexity is part of the intervention characteristics construct. This was a barrier for psychological and combined interventions, but it had a mixed impact in interventions combined with socioeconomic support and screening-only interventions. In psychological and combined interventions, the participants’ perspective was that the initial screening component of the intervention was complicated to implement due to the additional time and paperwork required, or the number of steps required to implement some components of the intervention. 34 In one intervention, the feedback from healthcare workers made the implementing team change the original screening tool for one more acceptable to TB healthcare workers and their practices. TB healthcare workers also found it challenging to deliver the psychological treatment component of the intervention. 25 In interventions where complexity had a mixed impact, the studies report that participants expressed positive and negative opinions regarding the implementation of the intervention. In the screening-only intervention there was some flexibility to allow healthcare workers to implement the intervention as they saw fit but they still considered that the intervention was hard to implement, even though the technological elements of the intervention facilitated their paperwork activities. 32 In some interventions, the processes of delivering the intervention components had been designed with the input of the healthcare workers, and were adapted during implementation according to their feedback. 32,34

Culture, part of the inner setting, was referred to as a barrier when organisational or patient culture clashed with the goals of the intervention. Culture as a barrier is best exemplified in the quote by Lovero et al. ,33 where they compare the goals of the intervention with the goals of patients and the organisation (i.e. public TB services):

The goals of the intervention did not align with wider organizational goals, as these often conflicted with each other, causing the clinic staff to feel stressed by the competing directives and interests … most DPMs [district programme managers] highlighted a need for increased mental health literacy and awareness among district-level administrators to better facilitate mental healthcare integration.

In the screening-only intervention, the cultural attitude towards mental health was reflected in patients’ and healthcare workers’ discomfort when asking diagnostic questions regarding patients’ mental health. 32 This challenge was echoed in the implementation of two psychological interventions, which affected the screening and treatment components of the intervention. 25,34 Culture was a barrier for all types of interventions, except for combined interventions with socioeconomic support, where there were no data available.

Other facilitators across interventions

Evidence strength and quality was defined as the belief, sustained by evidence, that the intervention would have the desired outcome. This was a facilitator in psychological, combined and combined with socioeconomic support interventions. In studies that described these types of interventions, the interventions were justified based on previous empirical research. In studies that reported quantitative outcome assessments, the interventions improved depression symptoms, decreased stigma scores and improved TB knowledge in patients and family members,30,31 which were the interventions’ desired outcomes. In qualitative studies, patients and participants expressed their belief in the beneficial impact of the intervention, particularly in interventions where current and cured patients interacted. 25,34

Cosmopolitanism, part of the outer setting, seemed to be a facilitator for all types of interventions, except for the screening-only intervention where there were no data to assess this. Cosmopolitanism, defined as the linkages between TB services and other public or private services, was an obvious facilitator in interventions that were implemented as part of collaborations between public TB services and private NGOs. 26–29 In three combined interventions, NGOs were the main point of delivery of TB services, which acted as the main point of delivery of the intervention for depression. 26–28 The role of NGOs as the main community healthcare service delivery and their role as facilitators of the implementation for the depression intervention were more prominent in the studies that described interventions implemented in Peru. 25,26,28

External policy and incentives, part of the outer setting, seemed to be facilitators for all types of interventions, except for the screening-only intervention for which there were no data to assess this. We considered the influence of local, national and international policies. In the implementation of psychological interventions, the existence of policies that supported the provision of mental health care to people with TB, or supported patient-centred care, acted as a facilitator for implementation. For example, the description of one psychological intervention mentioned that the provision of mental health care to people with TB was ‘… a strategy that is well aligned with and supported by the people-centred model of care advocated by the WHO’ (Khachadourian et al. ). 30 In one combined intervention, the implementation of training, diagnosis, treatment and referral components was enshrined in a national mental health policy which supported the integration of mental health services into primary care services.

One of the main objectives of the Strategic Plan is to scale up decentralized integrated primary mental health services, with the goal of including universal, routine screening and a stepped approach to management of mental disorders in primary care clinics.

Lovero et al. 33

In another combined intervention, implemented as part of NGO-provided TB services, the availability of external policies and incentives supported implementation and facilitated the delivery of screening and treatment.

Patients were also eligible for services provided by the municipality such as quarterly food baskets for patients with TB … In Lima, Peru, all these interventions are available and supported by local or national policy. Thus, it is feasible to implement a comprehensive approach to managing the health issues …

Contreras et al. 26

National and international policies acted as facilitators for implementation. International policies and guidelines, like those of the WHO, acted as facilitators at the local level.

Access to knowledge and information, part of the inner setting, acted as a facilitator for all types of interventions, except for combined interventions with socioeconomic support, where there were no data available to assess this. In all studies that described psychological, combined or screening-only information, the access to knowledge and information (e.g. training, provision of educational materials) for healthcare workers was a facilitator for implementation. This was particularly relevant for the implementation of the interventions’ screening components, as it was reported that some interventions provided training to healthcare workers to use mental health screening tools in their day-to-day practice in TB services. Only one study described the number of training sessions, along with the content of the training. 34 One combined intervention recognised the importance of providing training to healthcare workers but also acknowledged that training was provided unevenly and it was hard to assess its impact as there were no indicators to measure its effect. 33 In the screening-only intervention, it was stated that healthcare workers valued the mental health training they received to implement the intervention. ‘[Healthcare workers] believed the training made them more aware of the importance of caring for people’s mental health … and offered them necessary skills to help their patients’ (Diez-Canseco et al.). 32

Individual stage of change, part of the characteristics of individuals domain, seemed to be a facilitator for all types of interventions, except for combined interventions with socioeconomic support, where there were no data to assess this. This construct refers to the stages that healthcare workers go through towards the sustained and skilled delivery of the intervention, in terms of the change in their confidence and knowledge to implement the intervention. In one psychological intervention, the progression of healthcare workers in their mental health treatment skills was considered a facilitator. ‘Primary healthcare workers showed willingness to learn to deliver the intervention either as co-facilitators alongside a mental health specialist or in the absence of one’ (Acha et al. ). 25 In combined interventions, healthcare workers felt more comfortable providing screening and treatment components of the intervention after receiving training in mental health. ‘though the majority [of healthcare workers] … responded they felt “very comfortable” or “comfortable” with the management of mental illness in patients’ (Lovero et al. ). 33 A similar opinion was echoed in the screening-only intervention, in which healthcare workers felt more confident in their mental health screening abilities and decisions after receiving training (Diez-Canseco et al.). 32

The influence of external change agents as facilitators of implementation was present in all interventions, except for the screening-only intervention where there were no data available to assess this. External change agents were defined as individuals affiliated to other organisations that are not governmental or public TB services. The role of external change agents as a facilitator was more evident in the studies describing interventions implemented in partnership with NGOs, as these had collaborations with local public health authorities. For example, in one combined intervention with socioeconomic support that was implemented in the context of a pre-existing collaboration between NGOs and governmental health services, external change agents influenced the recruitment of participants for the intervention and its delivery, including the delivery of psychological treatment components and training for healthcare workers. 26 This was also the case in two combined interventions (without socioeconomic support) where the role of external change agents was also to provide training to healthcare workers and provide psychological and pharmacological treatment to participants in the intervention. 27,28,31 In interventions that were implemented solely by governmental services, the influence of external change agents in the implementation of the intervention was less clear or not mentioned at all. 30,32,33

Discrepancies in barriers and facilitators across intervention types

The construct, needs of those served by the organisation (i.e. people with TB needs), part of the outer setting domain, was a facilitator in psychological, combined and combined interventions with socioeconomic support. However, it was a barrier in the screening-only intervention. In the studies that described interventions where this construct was a facilitator, it was mentioned that the interventions were based on the care needs expressed or manifested by people with TB and their families. In one psychological intervention where depression treatment was delivered in a group setting, the intervention addressed the non-clinical needs of patients, like emotional support and the reduction of stigma. ‘… for many patients the support groups appeared to be one of the only places in which patients could speak freely about their experiences and feelings and receive empathetic support from others’ (Acha et al. ). 25 Two studies mentioned how the delivery of combined interventions was rooted in the mental health needs of people with TB that had been observed in pre-implementation years. 33 Combined interventions with socioeconomic support were underpinned by the economic needs often expressed by people with TB and their families. One study described how socioeconomic support for the person with TB often involved helping them to provide for their families, which the intervention supported by means of covering school fees and transport and food costs or linking them to existing social assistance services. 26 However, in screening-only interventions, this construct was defined as a barrier to the implementation as the patients’ needs were only partially met or not met at all. The study describing this intervention mentioned that, after screening, not all patients were referred to mental health services even if they needed to be and screening was not done consistently for all patients; this was possibly because the depression-screening activities were not included in performance assessments at the time the intervention was implemented. 32

The role of available resources, a construct part of the inner setting domain, was a facilitator for implementation in combined interventions with socioeconomic support. However, this construct was described in the context of barriers to implementation in psychological and combined interventions. Additionally, this construct was one of the few to have a mixed impact, as a barrier and facilitator, in the implementation of screening-only interventions. In combined interventions with socioeconomic support, the role of NGOs was a facilitator to improving the available resources for the implementation of screening, treatment and socioeconomic support components of the intervention. In these interventions, NGOs would provide the socioeconomic support, provide mental health training to TB healthcare workers and offer support to link patients with socioeconomic assistance available in their communities. 26 This finding seems to overlap with the role of costs of intervention and implementation: in the case of combined interventions with socioeconomic support, it was stated that the increased costs were mainly absorbed by the NGOs working locally, as they supplied sufficient human resources, training, socioeconomic support components and other relevant components for the delivery of the intervention. In terms of psychological interventions, one of the studies reported how physical and human resources were constantly hard to obtain during the 5 years that the intervention lasted. 25 For combined interventions, the negative impact of the lack of available resources was echoed and it conflated the reasons behind the lack of human and physical resources as a general lack of funding for mental health and primary care services. ‘Most [district programme managers] cited shortages in personnel and training opportunities, both of which often corresponded with a lack of funding’ (Lovero et al. ). 33

Self-efficacy, part of the characteristics of individuals domain, was a facilitator for the implementation of psychological-only and screening-only interventions, but it was a barrier to implementation for combined interventions. One study describing the implementation of a psychological-only intervention mentioned that the self-efficacy of the healthcare workers delivering the intervention was a facilitator. Healthcare workers were able to implement the intervention by themselves, with initial guidance from psychiatrists. 25 This was described as a progressive process in the study, meaning that the healthcare workers were more confident in implementing the intervention with time. This echoes the findings described in the screening-only intervention study, where the study discussed how healthcare workers were able to overcome challenges during the delivery of the intervention by applying their learning from the training provided to them. ‘Furthermore, some PHCPs reported referring patients who, despite having negative screening results, showed symptoms that they deemed relevant’ (Diez-Canseco et al. ). 32