Integration of mental health counselling into chronic disease services at the primary health care level: Formative research on dedicated versus designated strategies in the Western Cape, South Africa

Abstract

Objective

To explore health care providers’ views on the acceptability and feasibility of two models for integrating facility-based counsellor delivered mental health counselling into chronic disease care, and how such an approach could be improved in South Africa.

Methods

Fourteen focus group discussions and 25 in-depth individual interviews were conducted with 109 health care workers and facility managers from 24 primary health clinics in the Western Cape, South Africa.

Results

Findings suggested that despite recent efforts to integrate mental health counselling into chronic disease services for common mental disorders, there remains limited availability of psychosocial and psychological counselling. Feedback on the two models of integration suggested equipoise and the potential of a hybrid approach where these approaches may be tailored to the specific needs and available resources of each facility. Participants identified constraints within the health system and broader social context that require consideration for integrating mental health counselling into chronic disease care.

Conclusion

Although study participants unanimously agreed that counselling for common mental disorders should be integrated into chronic disease services, they had differing views on the type of model that should be adopted. There is a need for further testing of the two models and aspects of the health service that may require strengthening to implement any such model.

Introduction

Like other low- and middle- income countries (LMICs),¹ the prevalence and associated burden of mental disorders is high in South Africa. According to a nationally representative study, approximately 30% of the adult population in South Africa meet diagnostic criteria for a mental disorder at some point in their life². Despite this high prevalence, around 75% of people living with a common mental disorder (CMD) such as a mood, anxiety and alcohol use disorder have limited access to treatment³. Not only are untreated mental disorders the third largest direct contributor to the South African burden of disease; they are also associated with chronic communicable diseases (such as HIV) and non-communicable diseases, such as diabetes⁴. Untreated depression and alcohol use disorders contribute to sub-optimal adherence to chronic disease treatment, resulting in an increased risk of treatment failure and negative treatment outcomes^{5, 6} as well as increased health service utilisation and costs ⁷. Given this relationship, the integration of counselling for CMDs into chronic disease services at the primary healthcare (PHC) level has been recommended by the World Health Organization ⁸ and adopted by many LMICs as a strategy for improving the physical and mental health outcomes of chronic disease patients with co-occurring CMDs, including South Africa ^9–11.

Severe shortages of mental health specialists in South Africa have impacted on the successful implementation of this strategy¹². Recent data from South Africa reported an average of 0.31 public sector psychiatrists and 0.97 psychologists per 100 000 uninsured population, with a slightly higher average found in the Western Cape (0.89 psychiatrists and 1.22 psychologists per 100 000 population)¹³. To overcome this challenge, and in keeping with the World Health Organisation’s (WHO) recommendations for increasing mental health care access⁸, South Africa has endorsed task sharing of basic mental health counselling to non-specialist providers, including facility-based counsellors (FBCs) who are often referred to as lay counsellors in the available literature ¹¹. Systematic reviews highlight the feasibility, acceptability and effectiveness of using trained FBCs to deliver counselling in LMICs^14–16. There is also emerging evidence that FBC-delivered interventions may reduce depression and hazardous alcohol use¹⁷. Despite this promising evidence, unanswered questions remain about the optimal model for the routine delivery of task shared interventions in low resource settings¹⁸.

The WHO has identified two approaches to ensuring sufficient human resources for mental health within PHC where chronic disease services are provided. These involve mobilizing human resources currently available within PHC services to provide mental health counselling (designated approach) or re-distributing funding from mental health services provided at tertiary and secondary levels of care to primary care to allow for the employment of additional counsellors for the delivery of mental health counselling¹⁹ (dedicated approach). In South Africa, it is not yet certain which of these approaches to configuring human resources for FBC-delivered mental health counselling within chronic disease care is feasible, acceptable and therefore preferable for successful integration.

Some argue that FBCs already working in chronic disease teams have spare capacity to deliver mental health counselling provided that their skill set is enlarged and job description broadened. Currently they are providing adherence support for people living with HIV. These individuals can then be designated to deliver counselling in addition to their current responsibilities²⁰. Others contend that these FBCs are already overloaded so it is impossible for them to deliver additional counselling. In this view, additional FBCs dedicated to the delivery of mental health counselling must be employed to ensure the feasibility of this service. To guide health planners in how to proceed with integrating FBC-delivered mental health counselling into chronic disease care, more information on primary care providers’ views of the feasibility of the proposed dedicated and designated approaches to counselling provision within chronic disease services is needed.

Although previous studies have examined systemic barriers to integrating mental health services into PHC more generally,^21–23 these studies did not focus on the provision of task-shared mental health counselling and the perspectives of primary care providers who would need to change the way they organise and deliver chronic disease services to accommodate this new service. As the successful implementation of mental health counselling will require the support of these primary care providers, a better understanding of their concerns and the constraints within which they work is needed to inform the development of counselling integration models that are feasible to deliver ^{24, 25}. As such, this study aims to explore primary care providers’ views on FBC-delivered mental health counselling, the feasibility of the proposed dedicated and designated approaches to counselling integration, and potential constraints to counselling provision within chronic disease services with a focus on HIV and diabetes care.

Methods

Study design

We conducted focus group discussions and in-depth individual interviews with health workers responsible for delivering mental health, chronic disease and HIV services in PHC clinics in the Western Cape Province, South Africa. This was part of the formative work for Project MIND,²⁶ a trial comparing dedicated and designated approaches to integrating mental health counselling into HIV and diabetes chronic disease care.

Procedures

Between September 2015 and March 2016, the research team conducted 14 focus group discussions (FGD) and 25 in-depth individual interviews with health workers from 26 primary health facilities that provided both HIV and diabetes treatment services. The Western Cape Department of Health (WCDOH) purposively selected these facilities as potential sites for Project MIND. These clinics were selected to reflect the geographic distribution and variability in size, structure and organisation of clinics in the province. Prior to the research, two researchers conducted facility-wide meetings where they provided an overview of Project MIND. After these meetings, research assistants approached staff working in the HIV, chronic disease and mental health services (where these were available) to explain the purpose of the research and assess their interest in participation. We used maximum variation sampling to ensure the sample represented the range of provider roles and levels of specialisation available within the PHC system. We aimed to ensure our sample was representative of the staffing profile of PHC clinics in South Africa. These clinics are headed by a facility manager and are comprised predominantly of female nursing staff with a few medical doctors on site. Depending on size and services offered, a health promotor who is responsible for providing health education within chronic disease services is available as well as a psychiatric nurse. All chronic disease services at these clinics are staffed with FBCs who currently provide HIV adherence support or testing and counselling services. Although several health care workers chose not to be interviewed; we did not record the number who refused consent or their reasons.

Research assistants with postgraduate qualifications who had training and experience in qualitative research methods obtained health workers’ written informed consent before beginning the focus groups and interviews. They interviewed 109 health workers, of whom 88 were female (81%) and 21 (19%) were male. The final sample included 12 facility managers; six family physicians; three pharmacists; nine psychiatric nurses (professional nurses with an additional qualification in psychiatric nursing); 29 professional (four year diploma) and nine enrolled (two year diploma) nurses working in HIV or chronic disease care; two administrators; seven health promoters within chronic disease services; and 32 FBCs providing HIV adherence counselling (n= 29) or HIV testing and counselling services (n= 14).

FGDs and interviews followed a semi-structured guide that comprised two parts (See Supplementary file). The first part included questions about the availability of mental health services to patients with HIV or other chronic diseases, perceived acceptability of integrating FBC-delivered mental health counselling into chronic disease care, and possible barriers to counselling integration. Next, we sought participants’ opinions about the feasibility of two different approaches to integrating mental health counselling into chronic disease care, as well as recommendations for how to adapt these approaches for better fit. The first approach (known as the designated model) involves identifying and training a facility-based FBC, currently part of the chronic disease team, to deliver this counselling in addition to their other work responsibilities. The second approach (known as the dedicated model) involves employing an additional FBC with the sole responsibility of delivering a structured mental health counselling programme to chronic disease patients and no other formal chronic disease care duties.

All interviews and FGDs took place in a private room at the health facility in the participant’s choice of English, isiXhosa or Afrikaans (the three official languages of the region). Twenty-one interviews were conducted in Afrikaans and six in isiXhosa, with the remainder conducted in English. Interviews were digitally audio-recorded and lasted up to 55 minutes. All interviews were transcribed verbatim. Interviews conducted in Afrikaans or isiXhosa were first transcribed before being translated into English using standard forward-back translation techniques. The quality of the translations and transcriptions was checked by researchers proficient in the local languages.

The South African Medical Research Council (EC 004-02/2015), the University of Cape Town (089/2015), and University of Oxford (OxTREC 567-15) in the United Kingdom approved the study. The Western Cape Department of Health approved all procedures and granted permission to access the health facilities (WC 2015_RP 28-480).

Analysis

We used the framework approach (familiarization, identifying a thematic framework, indexing, charting, mapping, and interpretation) for data analysis ²⁷. This approach allows themes to be explored in relation to the research questions and for new themes to emerge from the data. Two of the researchers (KS and BM) read and summarized each transcript, identified key themes and categories, and developed a coding framework. Working independently, they used NVivo version 12 to code the transcripts using the coding framework (involving the indexing and charting stages of the Framework Approach). They met regularly to compare notes and resolve coding discrepancies through discussion. A third person was not needed to break coding ties. Inter-coder reliability was high, with a Kappa score of 0.92. When interpreting the data, the researchers remained aware of their assumptions about available resources for mental health care within public health facilities. These assumptions were discussed and reflected upon when the researchers met. Researchers presented the findings to health workers from participating facilities and at a stakeholder advisory group meeting where stakeholders had opportunities to provide feedback.

Results

Three broad categories emerged from the data including: 1) Structure and availability of human resources for provision of mental health services; 2) Possible approaches to counselling integration; and 3) Perceived factors that may affect patient uptake of mental health counselling.

Structure and availability of human resources for provision of mental health services

All participants reported that their facility had a mental health service delivered largely by psychiatric nurses. According to participants, this service was predominantly used to manage severe mental illness (SMI), with services largely limited in scope to medication provision and treatment support for patients who had been discharged from specialised psychiatric hospitals. Given this focus, persons with CMDs were often overlooked. As this provider described:

I follow them up. I see if they are still stable … see if they have relapsed. So, my role is basically to keep them out of hospital. [Psychiatric nurse; Facility 24]

According to participants, intern psychologists and psychiatry registrars assisted these psychiatric nurses by providing services on a rotational basis across several facilities. Where these trainees were not available, professional nurses without mental health training would sometimes step in to provide additional support to patients. A few participants reflected that the availability of mental health services seemed to rely on the presence of providers who had a passion for mental health. Participants in senior positions raised questions about the sustainability of a mental health service that was chronically under-staffed and relied heavily on a temporary or voluntary workforce. They were particularly concerned about the impact of staff resignations and absences on service delivery. As one participant reflected when describing the psychiatric nurse in charge of mental health care:

Sister X is amazing with those patients. They come to her and seek her out specifically. We don’t know what is going to happen when she retires next year. I don’t think anyone has really thought about it.” [Professional nurse, Facility 5].

Most psychiatric nurses expressed concerns about expanding this service to include patients with CMDs as they were struggling to cope with their current caseload and because SMIs and CMDs require different interventional approaches. Some highlighted that they were concerned about the impact an increased caseload would have on the quality of care provided to patients with SMIs and therefore their treatment outcomes. Other health workers were aware of these resource constraints which they attempted to alleviate through only referring patients with suspected SMIs to the mental health service. These patients were either identified through symptom disclosure or through health workers observing severe disturbances in behaviour.

There is a lot of schizophrenia, lots of substance use psychosis, there is many bipolar, and severe depression. Many people with depression do not get identified…. its only with those extreme cases, those people who need to be put on medication to function that gets sent to the mental health nurse. So many people in the ‘in between’ do not get help.” [Psychiatric nurse; Facility 3]

Most participants identified the lack of counselling to address patients’ psychosocial needs as a major shortcoming of current mental health service provision. They described how many of their patients with HIV or other chronic diseases drank heavily and presented with psychological distress related to an HIV or other chronic disease diagnosis, trauma, relationship difficulties, and life stressors. Providers were acutely aware of the adverse effect that these psycho-social difficulties had on patients’ adherence to their medication and self-care regimens, engagement in care, and health. As one provider highlighted:

In our clinic we have very high levels of psychosocial dysfunction and distress, and a big part of that is unemployment and poverty. There’s also lot of patients that are not adhering to medication. These issues all interact. [Family Physician, Facility 4]

Several providers expressed frustration with being unable to assist patients with these psychosocial concerns and feeling “stuck with that because we don’t know where to refer them to”. Although a sub-set of facilities seemed to have psychologists and social workers available, access to these providers was limited. For example, in one facility where a psychologist was available once a week, it was reported that “she’s fully booked until next year”. Mindful of their patients’ unmet needs and the difficulties of accessing mental health services, all participants supported the integration of mental health counselling for CMDs into chronic disease services. For health managers, this support was contingent on a clear plan for integration being available that outlined each health worker’s role and responsibilities in relation to mental health screening and counselling provision.

Possible approaches to counselling integration

When presented with two possible approaches for the integration of FBC-delivered mental health counselling, participants held divergent views about which integration approach they thought feasible to implement in their setting. While some clearly favoured the designated or dedicated approach, others proposed a hybrid approach.

A sizeable proportion of participants thought that only the dedicated approach was feasible, describing the designated approach as “doomed to failure.” Underpinning this view were perceptions that staff lacked capacity to take on additional tasks due to high patient caseloads and staff shortages. These participants described staff as being “already overloaded with work” and “exhausted and burned out already.” Additionally, managers specifically expressed concern that this limited capacity would result in lengthy waiting periods for patients and impact health service efficiency:

I chose the dedicated model because we are already short-staffed. Maybe we will give this designated person another task and she will be the only one doing the counselling - who will relieve her? We will have a backlog of patients with problems to be attended to, but with the dedicated person, she will always be available. [Manager; Facility 22]

These participants also thought that asking overburdened staff to perform additional duties would impact on counselling quality. Participants reflected how designated counsellors (with other tasks) would have less time to spend with their patients than dedicated counsellors which they reasoned would impact on counselling effectiveness. As one participant commented:

If a person is doing one specific job, they get better at it. They understand how it goes. Whereas if you get another person where they do this and they do that … there’s just too much to focus on for one person. If you have one [dedicated] person, you get quality of care. [Professional nurse, Facility 11]

Others thought that current providers lacked the appropriate skills and competencies for delivering mental health counselling, raising concerns that this might compromise the quality of counselling provided in the designated approach. These participants were mainly nurses or other health professionals who thought that an additional person “with specific training and educational background in mental health” was needed to ensure an effective counselling service. In addition, some providers thought that patients would prefer to receive counselling from a new, dedicated counsellor as they had negative views about the staff currently providing chronic disease services which could impact on their willingness to disclose mental health concerns. As one provider reflected:

If there is somebody from outside they believe this person will listen and give them attention because sometimes they look at us and conclude, no this sister doesn’t look like someone who can assist me with my problem … and a patient will have fears about disclosing because she will be dealing with a person she is familiar with and will worry her problems are going to be spreading. [Auxiliary nurse; Facility 22]

However, an equal number of providers did not share these views and favoured the designated approach. While recognising that this approach posed challenges, they thought they could “make it work.” Several participants (mainly FBCs) described how they were already working with patients who had mental health concerns even though this was outside their scope of practice. As a result, they did not think that their caseload would increase considerably. Instead they seemed eager for training to develop the skills that they needed to work more effectively with these patients:

This can only be an assistance to us because we deal with these things, but we are not trained so we don’t know what to do really. [Adherence counsellor, Facility 4].

Other providers in more senior positions favoured the designated approach as they considered it more sustainable than a dedicated approach that would require an additional resource investment. They thought it unlikely that the government would provide funding for additional counsellors, with one manager stating that “we must find ways and means of working with what we have.” These providers thought that upskilling existing staff to provide the counselling service would ensure programme sustainability:

Another positive thing is that, the person [designated counsellor] is already here. It’s sustainable. [Manager; Facility 24]

Most participants who supported the designated approach also thought that patients would be more open to the offer of counselling from a staff member with whom they already had a relationship than a dedicated counsellor who patients did not know. They thought this was particularly the case where staff had already established a good therapeutic relationship with the patient. As this participant reflected:

This designated approach, it might just be that you are going to use less time [to build a relationship] … and they might open quicker to you, easier to you than to the total stranger. [FBC; Facility 5]

Rather than endorsing either of these approaches to counselling integration, some participants proposed a hybrid approach that “fused the two.” These providers suggested adding another FBC dedicated to counselling delivery to the chronic disease team and designating counsellors already present in the facility to provide mental health counselling. Providers thought this hybrid approach would alleviate their concerns about there being no “back up” to provide counselling when the dedicated or designated person was busy with other activities or not present at work.

You must bear in mind the sick leave and annual leave, so we can’t employ one person only, we need to make provision for someone to relieve him or her. [Professional nurse, Facility 5].

They also suggested that if the dedicated person was skilled and experienced in mental health counselling, the hybrid approach would offer an opportunity for the designated counsellors to learn from the dedicated person and develop their skills.

This hybrid … We will need people from outside to come and assist and show us the way to operate on this project. The person we bring from outside is obviously well-versed (in mental health). That’s what I am thinking. Then the one from within (designated) is getting trained. [Professional nurse, Facility 20]

Many of the providers who proposed this hybrid approach were aware of the cost implications associated with having both dedicated and designated counsellors. These providers suggested that a hybrid approach could be used as a temporary strategy to support implementation of the counselling programme through building designated counsellors’ skills, competencies and confidence to deliver mental health counselling. As one nurse reflected:

My suggestions is that one should perhaps first start with this hybrid approach … but then with time move on to the designated approach. [Facility 13]

Factors that may affect the uptake of mental health counselling

Irrespective of their preferred approach to counselling integration, participants identified several factors within the health system and broader social context that they thought could impact on patients’ uptake of counselling. Many providers reflected that the way in which chronic disease services were organised impacted on patient flow, creating bottlenecks and lengthy waiting times. Providers noted that staff shortages exacerbated these waiting times. Because of the time burden associated with obtaining chronic disease care, some providers expressed concern about patients’ willingness to return to the facility for additional counselling appointments.

Some participants thought that patient waiting times for mental health counselling might be worse at facilities where mental health care was not prioritised as there would be little effort to shorten waiting times for this service. Participants highlighted the importance of limiting patient waiting times, particularly for patients who would need to take a day off work to receive counselling. Providers noted how this could place patients’ jobs in jeopardy, particularly where employers had low levels of mental health literacy and limited appreciation for the potential value of counselling. This was particularly salient among providers working in rural facilities where many patients were farm workers. Despite high levels of problem alcohol use in these communities, they described how farmers often limited farm workers access to health services, particularly in harvest season, as the “farmer wants them there, and that’s final”.

Participants also reflected on factors within the community that could impact on the uptake of mental health counselling. They described high levels of stigma towards people with mental health difficulties that could impede symptom disclosure and help-seeking due to patients’ fears of being labelled as mentally ill. Participants were concerned that patients may not access services due to fears that people may think “there’s something wrong upstairs” and perceptions that “only mad people need a psychologist or a psychiatrist or any counselling.” A few participants described how they perceived HIV to have less stigma than mental health problems, including depression:

They’re not willing to say openly to other people that ‘I suffer from depression’ or open up. Many people living with HIV do this. They often disclose and say their status out loud. You never see a person saying ‘I’m depressed’ or ‘I’m having depression so long’ out loud. But you will know this person is struggling. [Professional nurse, Facility 3]

Discussion

This study contributes to the growing evidence base of task-shared mental health interventions in Africa and other LMICs^21–23, by exploring primary care providers’ views on the feasibility and acceptability of two different models for integrating FBC-delivered mental health counselling. Findings suggest that 1) the limited availability of psychosocial and psychological counselling is a major gap in the current mental health service system; 2) providers were split in their preference for a dedicated or designated approach to integration, with some providers suggesting a hybrid approach tailored to the needs and available resources of each facility; and 3) several factors within the health system and broader social context require consideration when integrating mental health into chronic disease care.

Findings reflect the fragile state of the mental health services provided in PHC facilities in the Western Cape. At present, the service is only able to treat patients with SMI due to the limited availability of human resources for mental health. This is consistent with expenditure on mental health care in South Africa where only 4.6% of the total health budget is attributed to mental health services. Further, most mental health spending in the Western Cape is directed toward inpatient services with only 9.5% for outpatient care¹³.

Integrating counselling into chronic disease care may improve health system responsiveness to population needs for psychological counselling and psychosocial support while assisting South Africa to achieve universal health coverage for mental disorders. Despite South Africa’s endorsement of task-shared mental health services¹¹, this study found little evidence of the implementation of this policy, with mental health services continuing to be provided by specialist providers for SMI, and CMDs mostly undetected and untreated in primary care facilities.

Providers were acutely aware of the high levels of unmet needs for counselling among their patients and the negative impact these unmet needs had on the management of their chronic disease. While broadly supportive of plans to integrate mental health counselling into chronic disease services, they did not know how to proceed. Similar to previous research conducted in LMICs, many felt out of their depth and ill-equipped to provide therapeutic counselling ^{21, 28}. While providers in senior positions generally lacked the time to provide these services due to high patient caseloads,^{21, 23} many FBCs felt that, with training, their role could be expanded to include mental health counselling as they were already interacting with patients struggling with these problems and had more time to spend with patients than other cadres of healthcare workers. From this account, task sharing counselling to FBCs may be a plausible option when considering integrating mental health services into chronic disease care. Together these findings confirm that integration is unlikely to occur without significant efforts to build capacity among non-specialist providers to deliver mental health counselling and to formulate plans for integration that outline the roles and responsibilities of health workers in relation to counselling provision.

Engaging with health care providers and facility managers on their views of the two models of integration proved highly valuable. First, findings suggest equipoise regarding the feasibility of the models, with participants being divided in their opinions about which model was most feasible to implement. Given current health systems reforms potentially preferring the designated approach given the shift towards horizontal integrated approaches from vertical, those in favour of the designated approach argued that this approach was more sustainable. On the hand, those in support of the dedicated approach reasoned that this approach was more feasible to implement and could result in better quality of care. This feedback provides support for comparing the relative effectiveness and cost-effectiveness in a larger trial and highlights the importance of exploring counselling quality and concerns about sustainability in a future process evaluation.

Second, the designated and dedicated approaches to counselling integration seem more complex to implement in PHC facilities than we originally hypothesized. For both approaches, we learned the importance of (1) training more than one FBC to ensure counselling coverage during periods of absence, (2) including stigma-reduction components, (3) developing protocols for the identification and referral of patients who require additional mental health care, and (4) engaging with facilities to prepare them for the introduction of counselling and help create an environment conducive to counselling provision. For the designated approach, findings also highlighted the need to expand FBCs’ job descriptions and work schedule to incorporate mental health counselling. This may require negotiations with non-governmental organisations (NGOs) who are, contracted by the WCDOH to provide FBC-related services. We have subsequently expanded our trial preparatory activities to include facility-wide workshops building readiness for counselling implementation and engaged with NGOs and managers to obtain their support and help designated FBCs schedule these additional activities. Third, engaging with providers allowed for the emergence of a third, hybrid approach to integration that included the addition of a dedicated counsellor for a brief time during initiation of the service who could train and mentor the designated FBCs in counselling delivery until they had developed the required competencies. Depending on the findings of the project MIND trial, this approach may be worth considering in the future.

Finally, our findings suggest that simply integrating a counselling programme into chronic disease care without addressing contextual and service delivery constraints may impact on programme success. These findings highlight other elements that need to be in place for integration to be successful. This builds on previous work in South Africa that highlights the importance of creating an enabling environment in the facility^{23, 29} and the community³⁰ to support integration. At the facility level, system factors such as facility management, degree of facility organization, organizational readiness to change need to be considered ^{23, 29}. In the community, like other studies, low levels of mental health literacy and mental health prioritisation among providers, patients and community were identified as a significant barrier to counselling access and uptake. To support the scale up of mental health services in PHC, it is critical that any myths and misinformation regarding mental health are addressed through awareness-raising activities and provider training.

This study has some limitations. Participants were likely not representative of all healthcare providers and facility managers, despite the large sample size. We do not have the response rate for individuals who were approached participate and those that refused.

Conclusion

This formative study provides valuable information that informs our understanding of how integration might work in practice and the preparatory activities that need to occur before either of these approaches to resourcing FBC-delivered counselling models are introduced to facilities and tested in a trial.