The study aimed to understand healthcare providers’ perceptions of barriers to mental health services, including early identification of mental ill-health, and the reported challenges their patients faced. Based on the multilevel model , we organised the findings into the following levels: Structural (unclear policy); Sociocultural (family support, social support networks and cultural attitudes); Organisational (organisational characteristics, service access and inadequacy of resources) and Individual (knowledge, attitudes and individual characteristics of women, their families and healthcare providers) [36, 44]. Our results indicate that barriers to early identification of mental illness (and accessing mental health services more broadly) exist at many levels. In the next section, we present the findings, employing illustrative quotes by level and subtheme. To reflect the key informant narratives and tell the story of the data, we begin with organisational-level factors.
Training does not prepare providers for addressing mental healthcare in public health practice
Respondents described a mismatch between didactic, in-service training contrasted with the ‘grounded’ realities of clinical public health practice. A psychiatrist highlighted the lack of prioritisation of mental health in medical school rotations. Psychiatric rotations in South Africa often occur in the final year of medical school and are typically briefer than other rotations. She, like others, considered this preparation insufficient since most doctors will encounter patients experiencing mental disorders frequently in practice:
[I]n your casualty or emergency department [rotation], you'd find that 50% of patients have some sort of psychiatric issue, emergency psychiatric issue…the training doesn't speak to what is happening on the ground. (Key informant 21)
In addition to the lack of emphasis on psychiatry in medical school, there is currently no specialisation in perinatal psychiatry. Psychiatrists identified this as a gap in their preparation and training, noting,
“there's no specific qualification [perinatal mental health qualification for psychiatrists] here yet, so everyone just does their own thing, and then we meet up and say we all do it. There are a couple of us in the country who are interested in it, but it's not as organised as it is in the private sector. (Key informant 20)
Equally, a Director of Nursing Education highlighted a gap in the nursing curriculum and preparation. She emphasised the need for critical thinking approaches that also incorporate social determinants of health into clinical practice:
How can you have empowered students that are critical thinkers when your lectures aren’t critical thinking? How do you get that critical engagement with those issues [underlying social determinants of health] and realisation that you can't just send people back to where they came from, and give them a nice little disconnected lecture on various things that they're going to do, and then send them back to where all the problems are? It is a challenge to translate that in a meaningful way in a curriculum, especially when they also work in the practical environments where people come from very traditional training... (Key informant 9)
The Director just cited underscored a lack of training to address the ‘real world’ needs of women, including social determinants of health that affect them within their family and social contexts.
Key informants reported that the dearth of preparation in medical and nursing school and lack of specialization sets a poor foundation for healthcare providers entering clinical practice: i.e., once in practice, training to identify, refer and treat women with CPMDs remains inadequate. A medical doctor confirmed the need for more effective training, but emphasised challenges in the feasibility of implementing and sustaining such training:
I think there is a massive need for training providers to engage in mental health during ANC visits...my hesitancy comes from how you practically do that? Taking people out for training means that there's less people to provide services, which is a constant challenge that we find with anything that we try to implement. It takes training and also follow up mentorship and support. (Key informant 22)
Overburdened public health clinics create missed opportunities for identification, screening and referrals for mental health treatment
Most key informants commented on the busy clinic environment not being conducive to engaging in discussions about mental ill-health. As one key informant commented, nurses have so many patients, they will rush through an hour-long visit in 15–20 min—neglecting the opportunity to screen for CPMDs or discuss current experience of distress with women, thus posing missed opportunities for addressing mental ill-health during the visit. Further, the key informant suggested that a woman in distress that is experiencing mental ill-health might not return, even if a subsequent visit is scheduled for her.
…if it's one nurse and she has 30 or 50 new first visit ladies that's announcing their pregnancy for the first time and because of our HIV burden the first visit should take you an hour. But that's not the only burden, there’s another 60 [patients] coming for their second and third visits. And she will try and rush through this first visit in about 15 or 20 minutes. And the lady will get a second visit date, but she's mentally not well so she will never pitch up again, you see, and then that's a missed opportunity and that’s exactly what we have. (Key informant 10)
Psychiatrists and other mental health specialists typically only see women when they are in late stages of pregnancy and have more severe mental illness, with little time to intervene prior to childbirth. A psychiatrist noted that this exposes the gaps in the system:
…a lot of these women, you can see where the gaps are coming in. By the time they reach us, a lot of them get to us at late gestations and we don't have much time to sort them out and you can see where they [have] fallen through the gaps along the way because of the overburdened clinics that they're coming from. (Key informant 22)
One key informant suggested that a simple question, “How are you?” might begin to engage women in conversations about their mental health. Yet clinics are usually too burdened to allow for much time with the patient, so providers often avoid the conversation entirely.
Some of these women they feel quite isolated so just asking them how they are and just the brief discussion would alleviate [mental health issues], especially in the milder forms. …but for the midwives, it’s so pressured. They will see 35 women in the waiting room…not having the time to spend with the women. I think they avoid that [discussion of mental health]. (Key informant 5)
Long queues and overwhelmed staff in many antenatal clinics further impede women from asking questions of providers during visits. Due to long queues, one provider shared an example where a woman struggling with depression would not be comfortable disclosing concerns about her mental health, even if she desired it.
I'm not going to go to the doctor and start talking about my emotions and how I'm feeling, because we've been waiting since five in the morning. And there are 100 other people…outside. So they're not going to easily say “Hey Doc, you know, things aren't going well.” …and then the doctors aren’t asking the questions on their end, everything just kind of gets messed up. (Key informant 21)
Lack of mental health prioritization and parity compared with physical health and need for better integration of mental health into antenatal and primary care
Respondents discussed the lack of streamlined care and integration of mental health into ANC as part of the challenge of meaningfully addressing women’s mental health.
It’s very difficult for these women to have a separate health visit for mental health…because you've got to take time off work. You’ve got to organise childcare. You've got to pay for traffic transport, get to the clinic, see the doctor and then do all your antenatal visits…it really needs to be done all at the same time. It takes a lot of motivation to take the two taxis with your toddler to sit in on the line at another clinic to see someone that may or may not help you. (Key informant 23)
Deploying community health workers to address mental health, as part of a task-sharing approach, is one health system innovation to reach more women and better integrate mental healthcare. And yet a psychiatrist stressed that the overwhelming and unrealistic caseloads faced by community health workers lessens their ability to care for women’s mental health.
… you read their job description and you just think how on earth are they supposed to do that … for 250 households…it's insane. Maybe a community healthcare worker could manage 50 households, if 40 of them are stable. It’s all very nice, in theory, on paper, but when you try and practically implement it, you realise that these poor community healthcare workers. There's no way that they can handle all of this. (Key informant 22)
Scarce resources devoted to mental health contribute to highly reactive and fragmented system of care
Competition for public health resources occurs at the highest levels and this reduces the attention that mental health receives, as one key informant related.
At the macro level, I think in South Africa we have a situation of enormous competing for the public health crises…in terms of not only the obvious things like funding and programmatic attention, but they compete psychologically in people's mind. [There are] challenges which are enormous, like gender-based violence and corruption and food insecurity and a lack of access to basic resources, and it just feels that mental health is not as tangible. The average high-level minister or politician or person who allocates budgets… I think it just must be very difficult for them to keep all of these things in mind and to have a concurrence. (Key informant 3)
As noted by many key informants, South Africa’s mental health service delivery system, like many other settings, is highly reactive, not anticipatory. One key informant described this culture in the following way.
…part of the issue in South Africa is the translation of all these wonderful ideas into reality is where it falls down. And what you find…on the ground is very different to what we might be on paper, and I just think that the services and the service delivery that we have in South Africa is highly reactive and there is no probing to and educational work being done. Mental health services are reactive and when there's a problem. (Key informant 19)
Key informants expressed frustration, indicating that it is not enough to provide mental health education for women in the community when the health system is weak. Some key informants called for a better system for all types of healthcare providers to work together to improve women’s mental health.
It can't just be one person, it needs midwives, it needs psychologists, it needs lactation specialists. It needs the paediatricians, obs and gynae. All of us need to work together…I decided it's no use me going on all these platforms and educating the community and then when they get to the antenatal clinic there's absolutely no services available or no one knows what they're [the patients] talking about. (Key informant 21)
Inadequate systems in place for identifying mental illness during pregnancy and postpartum
There are emerging systems, including a brief, validated screening tool, for early identification of common perinatal mental disorders, as one provider noted, “It is however in the primary healthcare guidelines…it’s not routinely done yet, but it should be because the primary healthcare guidelines advocate for mental health to be incorporated into the antenatal clinic. (Key informant 23).” However, rather than being delivered routinely, many key informants portrayed an environment where early identification of mental disorders depended on the ability or willingness of providers.
How much it practically filters down into the clinics [mental health screening], I think it depends on the sort of clinician who maybe has an interest who sort of pick up the cues on a patient who has got issues. The size of our clinics and probably just the interest of people. I don't think it's screened for enough. (Key informant 18)
A psychiatrist noted a failure of early screening, which leads to seeing patients with more complex issues later:
Specifically, a lot of the postpartum stuff we don't pick up…it only presents when you take a history a couple of years later, like what happened then… that initial screening, that postpartum screening. I don't think it's being done accurately by gynaes, you know they're supposed to do that, at the 10 day visit, six week visit. I don't think it's being picked up, but I also don't think women are aware. And these are kind of first onset stuff. It also kind of depends on which gynae you have, I think. If a gynae is more psychologically minded, they will ask more. (Key informant 20)
Different cultural values and translation barriers affect the identification of mental illness
One key informant mentioned that nurses will record in a patient’s file that a patient is “uncooperative” if they don’t speak the same language or understand directions they are given in the clinical setting. She indicates this notation in the file follows the patient, who might be treated poorly throughout her time in care.
Instead of somebody sitting down and thinking maybe she's just scared or maybe she doesn't actually understand what's going on because the language barrier is just as bad… First of all, she's really intimidated by what's going on, and in the black cultures going in and having somebody A) that’s very educated and B) that’s a senior member of your community then coming in and treating you so badly, it just breaks down your entire self-esteem. So you think to yourself….I am a terrible, terrible person. (Key informant 12)
One respondent described challenges directly translating depression and anxiety, also observing that some patients’ experiences of distress manifested as physical symptoms.
If you say you're depressed, there isn't a direct translation for example for depression or anxiety …patients will often somatise and say… in Xhosa “I have a headache or it’s my back”…and you’ll treat that headache to death. Because it’s culturally more appropriate to go to the doctor or nurse and say, I have a headache [than to admit to depression]. (Key informant 21)
A different key informant discussed how local understandings of depression and idioms of distress were not readily translated to conform with commonly understood definitions of depression. The respondent suggested that the lack of understanding of local idioms of distress, combined with low community level awareness of symptoms of mental ill-health, limited healthcare provider identification of CPMDs, like depression. Ultimately, these expressions of distress might be ‘normalised’ as simply part of the experience of a difficult life and not understood as mental ill-health.
If I try and ask women directly about depression, very few of them even know what that means. So then I ask a nursing sister to come and translate it into the vernacular language so that they can have a sense. Listening to those translations, I get the sense that there isn't even a word for depression in many of the South African vernacular languages so I think it's not something that's widely acknowledged in the community. And I think most women would probably think it's pretty normal to be miserable about life because life is tough. I'm not sure that women would realise that low mood or suicidal thinking or finding it hard to bond with your baby or not feeling excited about the pregnancy is unusual or abnormal. (Key informant 22)
Lack of cultural humility and need for centring women’s health within the cultural context
Respondents continued to stress that women might use less specific terms or clues that indicate they are not doing well and are sad, and the respondents reinforced that clinicians need to shift to focus on the holistic health of women, to understand different manifestations or expressions of mental ill-health in perinatal women, while keeping women at the centre.
… some people will come in and they'll talk about poor sleep and poor appetite and poor memory and concentration and anhedonia but they don't come in and say I'm feeling actually really sad. (Key informant 23)
A doula who supports women during childbirth observed that women are routinely discouraged from asking providers questions about their health. She suggested that if providers were reminded of the importance of putting patients’ needs first, in accordance with Batho Pele principles (‘People First’ in Sesotho), rather than solely focusing on routine requirements in a ‘robotic’ manner, this could be an important step in opening the dialogue about mental health . The intent of Batho Pele includes transformation service delivery in the public sector; increase access and generally ensuring good customer service.
Providers need to be reminded about the importance of individual patients. It’s crazy because it’s your body. We feel like it’s so hard to ask that question. You are not questioning their knowledge, in fact you want them to share more with you. Often you will find when you ask the provider, they are humans too. Sometimes they have to do something in a robot way…they need a little reminder that they need to take a moment with someone. (Key informant 7)
Healthcare provider stigma of mental illness shapes approaches to engaging in identification
Key informants observed the mismatch between the content of the current training and that required to address the mental health needs of perinatal women, mentioning that the insufficient attention to mental health in training serves to stigmatise and undermine the importance of mental health:
That in itself [insufficient training on mental health] already cements mindsets and stigma with regards to the importance of mental health. (Key informant 21)
Key informant 21 elaborated her view of the cultural construct and stigma of mental illness and the implications of poor mental health in communities, suggesting that this affects how mental illness is viewed by individuals and providers alike:
South Africa is 80% black population and culturally, if we look at it from…an African point of view, there is no such thing as mental illness. So we already have that stigma. So it’s stigma within medicine itself and the perceptions of psychiatry, and then you have a cultural stigma as well. (Key informant 21)
A doula shared her observations of stigma at the individual-level in South Africa, as shaping perinatal women’s interactions with their providers, “It’s hard because there’s such a stigma especially when you’re a new mother, it’s so hard to say “I do need the help.” It’s a great tragedy of our time that women have the burden. The mental load of everything the mother needs to think. (Key informant 7).
Similarly, a perinatal psychiatrist emphasised that stigma is present at multiple levels.
…we've got stigma at all different levels. We’ve got stigma at…the midwives, who have called the patients crazy literally. We've got [self] stigma at the patients themselves, not wanting to say that they're struggling. And then you know, there's this whole “go to psych, go see psych” and it's a very derogatory statement and even like in the clinics where I used to work…the psych section was the bad section. (Key informant 20)
Key informant 20 indicates that stigma acts in multiple ways to compromise providers’ ability to detect and treat mental ill-health during ANC visits.
Providers own unaddressed mental health reduces their ability to engage in mental healthcare for perinatal women
A doctor with extensive clinical and research experience in maternal health depicted a context where nurses typically come from the same communities as their patients, and thus face similar challenges.
If you look at the sort of predictors of the patient and …social circumstances of the pregnant women…nurses have the same predictors. And then you match the two, and you have a woman that's down and you have a nurse that's down and there’s no care happening. They miss the biggest things that you even can think about. (Key informant 10)
Key informant 10, and others, indicated that providers’ unaddressed mental health needs appeared to limit their capacity to engage women in identification, referral and treatment of CPMDs. Indeed, many key informants emphasised that the frequency of unaddressed mental health issues among health professionals rebounds to undercut their ability to engage their patients in mental healthcare. A psychiatrist explained how providers with experiences of unresolved trauma and mental illness are less likely to engage in mental health conversations with women because these might be a triggering experience.
I think a lot of it comes down to your own personal experience and…past exposure to people who've had mental health issues. I think one of the big reasons it's not engaged in is because our staff have a high rate of mental health issues that are not addressed. We are always encouraging them, particularly in the antenatal clinics, to screen for gender-based violence. But if you're a victim of that yourself, then the triggers involved in screening other people for it are just so big…I'm just very aware of how much vicarious trauma and sort of personal trauma exists within the healthcare workforce that can make it quite difficult for them to then open up a can of worms in someone else. (Key informant 22)
Similarly, another respondent noted that the absence of psychosocial support (referral services) reduced the emotional capacity of providers. This, together with the high burden of patients’ overall needs, signified that providers might be less likely to address the mental health needs of women.
And then what do you do when you're sitting with an awareness that a patient has a problem, but you've got nowhere to divert them to and you don't have the skills or the time or the emotional capacity to deal with it. It's much better to just not go there. Safer for you to not go there. And I think there's quite a lot of sort of hardness amongst some healthcare workers because that's their survival option…just shut off and don't connect with the humanity in the person that you're seeing, because if you do, you're just never going to make it through the day, there's so many women with so many challenges that you would…only get through five people in a day, and you've got 73. (Key informant 22)
This ‘hardness’ seemed to serve as a sort of coping strategy for providers, given the absence of resources, training and emotional support for themselves.
Intersection of social determinants of mental health poorly understood
A doctor with many years of experience in perinatal mental health stated that the multiple, intersecting social determinants of mental health are poorly understood, jeopardise overall well-being of patients, and seem to reduce expectations of what providers offer women.
We are finding that they're not facing one or two risk factors, they are facing five or six, in their histories or in their current life circumstances. So we've had to kind of adjust expectations and not necessarily expect that women will end up being completely well and completely flourishing. It won't solve the fact that they've got no running water, or…that their education means that they will only earn 3000 rand a month [approximately 163 Euros] if they're lucky. So we've had to get comfortable with our limitations and try to work within that. (Key informant 3)
In response, one doctor discussed working within these limitations to provide a brief mental health screen, which she observed could make a meaningful difference in the face of these intersecting risk factors and determinants of mental health.
It's almost like people perceive there to be bigger problems [than mental health]. But I don't think people understand that a woman who is depressed is more likely to be in an abusive relationship, is more likely to be exposed to HIV. There's that triad. So all those are all interlinked. …you sort out her mental health, you might actually put her at less risk of HIV, or you might make them healthier so that she can actually comply with her ARVs and be healthy. But I think,…we're so used to putting out fires. (Key informant 23)
Several key informants mentioned that these complex, interlinked factors contributed to an assumption among providers that depression and anxiety are, again, ‘normal’ responses to life’s challenges in South Africa. They explained that this normalisation contributes to women suffering in silence.
…you're looking at almost 40% of women having postnatal or antenatal depression, anxiety, it’s not recognised because it’s so common. …one of the gynaes [gynaecologists] was saying to me: “Well, obviously, everyone's depressed. No one's got partners, no one's got money, you know, everybody's HIV positive...no wonder they’re feeling depressed.” (Key informant 23)
Respondents highlighted, per previous comments, that doctors, nurses and midwives often race through checklists and miss important disclosures, such as intimate partner violence or food insecurity, emphasising that centring the conversation around the holistic needs of the individual women is needed. As one provider elucidated,
…you must ask a patient what matters to them. They have a certain reason why they want to come to you. And that is one of the things that you need to address because the patients will actually tell you when they're not well, but if you don't give them the opportunity, you will never hear it. A pregnant woman will tell you that she's very worried about the baby, that it’s not going well, but actually, the husband is being violent or there's not enough food at home, and now the baby's not growing well…and if you go through a checklist, like a robot, you will not hear this. (Key informant 10)
Key informants were asked during the interviews if they had any specific suggestions concerning how perinatal mental health policies could be improved, and they eagerly shared potential interventions and recommendations to overcome barriers by level (Table 1).