Study design

Formative research included qualitative and observational data collected in preparation for establishing a prospective TBM cohort study.

Setting

The Western Cape province comprises one metropolitan municipality known as City of Cape Town (COCT). Within this area, there are more than 100 PHC facilities, including mobile and satellite facilities. The COCT is divided into eight sub-districts. Two tertiary referral hospitals serve the entire paediatric population in the province, i.e., Tygerberg Hospital, and Red Cross War Memorial Children’s Hospital. Brooklyn Chest Hospital is a specialised TB hospital where the majority of paediatric TB patients requiring prolonged hospitalization are admitted.

Sampling

We purposively sampled six PHC facilities from two sub-districts in Cape Town and included two tertiary hospitals, one specialized TB hospital and one district hospital to be qualitatively representative of the health system in Cape Town. PHC facilities were conveniently sampled from two sub-districts with the support from the district managers, based on paediatric TB burden and trends, using TB data from 2019 and 2020. We worked with these two sub-districts due to limited resources and feasibility. We consulted with Department of Health stakeholders to select facilities using routine TB data. We reviewed the following performance indicators; percentage children of the total TB case load for 2019 and 2020; ratio of 0–4 vs 5–14-year-olds; change in adult TB case load– 2020 vs 2019; change in paediatric TB case load– 2020 vs 2019; change in percentage of children 2020 vs 2019 (i.e., relative change in paediatric TB compared to adult TB caseload). At each health facility we purposively sampled 2 to 3 HCWs for diversity in role (e.g., nurse, doctors, and specialized medical staff) and for ‘rich cases’.

This is a purposive sampling technique where participants are selected because they have in-depth or expert knowledge with a particular program or patient group for example [15]. An example would be inviting nurses with multiple years of experience delivering a particular service to participate in a study, rather than including any nurse involved in delivering that service in a sample. The sampling logic is about maximizing collection of data with participants with extensive experience to make good use of limited spots in a ‘small’ qualitative sample.

Data collection

We collected data between 1 November 2021 and 28 February 2022, conducting in-depth interviews with nurses and doctors (n = 14), semi-structured observations (one day of observations per facility), including informal conversations with staff to help understand patient flow. We used a study-specific discussion guide to facilitate interviews, which included questions on the existing TBM care pathway, barriers that impact patients’ progression along the care pathway, and the impact of COVID-19 on access to health services. Interviews were conducted in participants’ preferred language–English or Afrikaans–and lasted approximately 30–45 minutes.

Analysis

We summarised the interview and observation data into case descriptions of the TBM care pathway at each facility. We followed an iterative process to inductively analyse these case descriptions in a Microsoft Excel Spreadsheet, to describe the health systems for the paediatric TBM care pathway at hospitals and at PHC facilities, as well as barriers identified across the TBM care continuum.

Ethics considerations

We obtained ethics approval for the study from the Health Research Ethics Committee at Stellenbosch University (N21/04/041), and approval for the study to be conducted from the Western Cape Provincial Department of Health (WC_202107_021) and Cape Town City Health (Project ID: 9434). All participants voluntarily consented to be interviewed and completed written informed consent.

Findings

We charted the health systems for paediatric TBM care at both PHC and hospital level in Cape Town, South Africa and present HCWs’ perspectives on barriers that impact patients’ progression along this care pathway. Fig 1 provides an overview of the existing care pathways as well as barriers identified at each step. We also explored HCWs’ suggestions on optimizing the TBM care pathway.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Paediatric tuberculous meningitis care pathways and barriers identified by healthcare workers (HCWs).

Children with probable TBM enter care mainly through walk-in into Primary Health Care (PHC). Grey arrows indicate movement between various levels of care. Both-sided arrows indicate that children could have more than one encounter at the facility and move back and forth between facilities. After first encounter at PHC, children require a referral from PHC to district, tertiary and specialized TB hospitals. Some children are discharged from tertiary hospital for home-based care treatment. The bottom of the page indicates several identified barriers that hinder and prevent children from progressing into appropriate care. CT = computed tomography; MRI = magnetic resonance imaging; TBM = tuberculous meningitis; TB = Tuberculosis.

https://doi.org/10.1371/journal.pgph.0003518.g001

Overview of paediatric TBM care pathways

We divided the pathway into pre-admission and primary care, hospital admission and inpatient care, and post-discharge follow-up care.

Pre-admission and primary care

This pathway begins with caregivers recognizing symptoms and seeking care at a PHC facility, the initial point of contact with HCWs. Most PHC facilities in COCT are nurse-run with doctor support from the larger facilities. After receiving a patient folder, the child is referred to the weighing room for observations and then to child-health services for evaluation of symptoms of pulmonary TB. Nurses screen for TB and manage symptoms based on Integrated Management of Childhood Illness (IMCI) guidelines [16]. Children presenting with early non-specific TBM (or other extrapulmonary TB) symptoms like appetite loss, reduced playfulness, and vomiting are often treated symptomatically, resulting in numerous visits before referral or correct diagnosis. However, those with more severe symptoms including seizures, decreased consciousness, and high fever are rapidly triaged for emergency assessment by a nurse or doctor, if available, to determine the need for urgent hospital referral. Thereafter, caregiver-child dyads will be referred to the next level healthcare facility by ambulance. HCWs emphasized that definitive diagnosis rarely occurs at the PHC level given limited resources for lumbar puncture, computed tomography (CT) scans, and other essential investigations. Thus, most possible TBM cases at PHC level are referred to district, regional or provincial hospitals based on proximity and symptom severity.

Hospital admission and inpatient care

Most children enter hospital admission and inpatient care through a referral mechanism pathway. However, some children do access care without a referral from PHC, especially those with advanced TBM symptoms such as seizures or decreased level of consciousness. Children who are severely ill or unstable are transferred through inter-facility transport. Those considered not seriously ill but referred from PHC to hospital for further investigations must use private or public transport. These next level of care facilities have specialised medical staff (paediatric doctors and specialists) as well as neuroimaging necessary to make a diagnosis of TBM. HCWs will start by stabilizing children according to their presenting symptoms, assess vital signs and conduct basic clinical examination, laboratory and radiological investigations including chest radiography and, if safe, do a lumbar puncture. At this point, prior to any microbiological confirmation, children are often treated empirically for both bacterial meningitis and TBM. Based on the results of investigations, children are usually referred for neuroimaging to support the diagnosis and establish the need for additional interventions such as neurosurgery. At the tertiary healthcare level, HCWs reflected that they receive referrals from both the PHC and district level to conduct additional diagnostic investigations and consult with sub-specialists. Children with hydrocephalus will be investigated by paediatric neurologists and/or neurosurgeons and referred for neurosurgical intervention if indicated. Thereafter children will be managed in a paediatric ward or intensive care unit. HCWs reflected that children are hospitalised until they are stable on TBM treatment and will be discharged to continue treatment at home or as an in-patient at a specialised TB hospital, depending on the level of available support at home and the child’s specific treatment needs.

Post-discharge follow-up care

At the tertiary level of care caregiver-child dyads will be evaluated by a multi-disciplinary team prior to discharge to determine their suitability to enter a home-based care programme [5, 17, 18]. The team typically includes a paediatrician (or paediatric neurologist), social worker, occupational therapist, and physiotherapist. If children require additional supportive care during treatment, often due to severe impairment or disability, or if they do not have adequate social support to receive home-based care, they will be transferred to a TB hospital for in-patient hospital treatment [11]. The duration of hospitalization depends on multiple factors, but the child’s health takes priority. HCWs described that children continuing treatment on the home-based care programme are followed up monthly at the tertiary hospital.

Barriers identified across the TBM care continuum

We categorized identified barriers into pre-admission and primary care, hospital admission and inpatient care, and post-discharge follow-up care but also included barriers associated with COVID-19 restrictions.

Pre-admission and primary care barriers

Barriers prior to accessing care at PHC level included delayed recognition of TBM symptoms by caregivers and poor socioeconomic circumstances. Limited post-training mentorship to recognize the early symptoms of TBM at PHC level, overburdened facilities and time constraints were identified as barriers at PHC level.

Most HCWs reflected and recognized that PHC facilities often missed children presenting with early TBM symptoms, even when there was a history of TB exposure. HCWs attributed these missed opportunities to recognize TBM symptoms to a lack of post-training mentorship and limited exposure to diagnosing TBM in children, which mostly occurred in hospital. However, these facilities were the first point of care for children with probable TBM symptoms, and early recognition and referral are critical. A doctor from a tertiary hospital explained:

“Looking back at some of the families caring for children diagnosed with TBM, they reported going to a clinic [PHC]. Their children were misdiagnosed with teething, flu, or gastro. TBM is very hard to detect especially at PHC level. They will only consider TBM if the child presents with a seizure.”

HCWs recognised that PHC facilities were overburdened and were constrained to see all patients seeking care on a given day. This led HCWs to rush through consultations and to take incomplete patient histories, which could help to identify possible TB exposure; children presenting with potential TBM symptoms were not always examined thoroughly. A district manager reflected:

“They [healthcare workers] are trained to identify these [TBM] symptoms, they experience challenges when clinics are full of patients and screening just becomes completing a tick box. Healthcare workers are not attentive to the danger or risks signs of disease, for instance TB contact in the household of a sick child.”

HCWs described that caregivers often struggled to recognize that their children were unwell, due to the non-specific nature of TBM symptoms, which delayed seeking health care. For instance, a HCW emphasized:

“Caregivers cannot pick up the growth faltering between ages two- to six-year old, until the child presents with severe symptoms such as seizures and acutely ill at the hospital.”

“A child with flu-like symptoms, parents (are) less likely to seek help. However, for the extreme cases where the child with possible TBM cannot move, parents do seek help.”

HCWs explained that patients’ challenging socio-economic circumstances often prevented them seeking care at PHC facilities especially if they needed to travel out of their communities. Many communities in Cape Town experience regular periods of gang violence, community unrest and transport strikes which at times made accessing PHCs unsafe and impossible. A nurse explained further:

“Most patients come from poor circumstances, and they want their children to be healthy and well. They often struggle with money to come to the facility for follow-ups.”

Hospital admission barriers

Barriers to care at district and tertiary levels included delay in referral pathways between PHC and hospitals, and limited sensitive and specific diagnostic tests for TBM.

HCWs described that further investigations at hospital level were often delayed due to poor communication between health worker cadres. A HCW reflected:

“There are often difficulties between healthcare workers during the referral process. For instance, doctors at secondary level will question the nurses at PHC when they refer patients. In such cases it delays patients receiving care.”

HCWs recognized that a diagnosis of TBM requires several investigations due to the lack of sensitive diagnostic tests. HCWs emphasized:

“The diagnostic tools that diagnose childhood TBM have low sensitivity. We require better diagnostic tools due to low sensitivity of the existing tools. The diagnosis for TBM is based on clinical features, scans, and tests. It is challenging to identify TB in brain fluid.”

Post-admission and post-discharge barriers

Post-admission and post-discharge (for treatment) barriers include lack of palatable and easy-to-administer medication, as well as limited post-TBM rehabilitation services and poor socio-economic circumstances of the families. HCWs reflected that current TBM treatment is not palatable for children, making it difficult to administer to children especially when they are discharged to continue treatment from home. For instance, a HCWs described:

“If we [healthcare workers] had palatable drugs, this would make our jobs easier. For instance, a prescribed drug [ethionamide] has an ugly taste, children usually spit it out and it makes the children feel nauseas.”

“They [caregivers] struggle to break up tablets as there are no child-friendly medication. Some caregivers prefer their children to take their treatment from home.”

In Cape Town, some children are not considered for post-TBM rehabilitation services, partly because access to and availability of these services are very limited. The availability of neurodevelopmental, psychological or educational assessments are also very limited in South Africa’s public health sector, as is the availability of special schooling for children with special needs, including children suffering with post-TBM disability or behaviour challenges.

COVID-19 related barriers

Data collection for this study coincided with the COVID-19 pandemic and response, and HCWs shared how the pandemic compounded existing health systems challenges for the prevention, diagnosis, treatment, and rehabilitation of children with TBM. HCWs identified barriers that included limited access to PHC services due to COVID-19 related lock down and restrictions, missed opportunities for prevention (i.e., TPT and BCG) and to screen for and diagnose TB and TB-related disease (including TBM), and reduced post-hospitalization follow-up and care options.

During data collection, some PHC facilities implemented an electronic appointment system whereby patients could access care through a scheduled appointment only and this system was adapted to reduce transmission of the COVID-19 virus. This service adaptation possibly increased delays in diagnosis and referral of children presenting with probable TBM symptoms. HCWs reflected:

“We (facility) did not do any walk-in’s, and patients would have to go to the day hospital to be assisted. Patients were given appointment dates to come to the facility, however if they do not show up, they return for another appointment date.”

HCWs highlighted that there was more emphasis on the screening and testing of COVID-19 symptoms. Other HCWs reflected that they struggled to recognise the difference in COVID-19 and TB symptoms, resulting in missed opportunities of diagnosing TB disease. HCWs reflected:

“COVID-19 pandemic had an impact on the diagnosing of TB, due to the similarities of the symptoms. There was more emphasis on testing for COVID-19 as opposed to TB in general.”

HCWs reported that prior to the COVID-19 pandemic uptake on TPT was sub-optimal, with many eligible candidates not initiated. For those who were initiated on TPT, completion rates remained poor. HCWs believed these numbers were underreported during COVID-19as fewer patients sought care at the facilities and fewer were diagnosed with TB during this time. HCWs described:

“There were less patients seeking health care at facilities, and less patients diagnosed with TB. This implied that more children were not initiated on TPT.”

“There is still low coverage of TPT. About 70% of children that are initiated, do not complete TPT.”

HCWs highlighted that BCG vaccinations were available in facilities; however, they still encountered children who were unvaccinated. From a logistics perspective facilities usually batched children in groups before opening a BCG vial to maximise use and avoid wasting. However, this process also posed challenges for caregivers and children. For instance, a HCW shared:

“Caregivers can go to the facility for catch-up vaccinations and they might not have stock or an opened BCG vial on the day of return appointment.”

HCWs explained that efforts and resources were redirected towards the COVID-19 pandemic. For instance, a TB hospital was converted to a COVID-19 specialising hospital which affected the TBM follow-up pathway. One HCWs shared:

“There was a massive impact on follow-up processes. Prior to COVID, children diagnosed with TBM were managed at a local TB hospital. However, during the COVID-19 pandemic the one TB hospital was changed to a COVID hospital and closed for TB services. Patients were discharged to another hospital further away from home and other patients were treated through a home-based care programme.”