Barriers to the integration of palliative care in Ghana: evidence from a tertiary health facility

Abstract

Background:

Integration of palliative care (PC) in cancer management is critical to improving the overall quality of life of cancer patients and their families. Nevertheless, only a few people in need of PC services actually receive them.

Objective:

The study explored the barriers to the successful integration of PC in cancer management in Ghana.

Design:

The design was an exploratory descriptive qualitative research design.

Methods:

In all, we conducted 13 interviews with service providers (7), patients (4) and caregivers (2). An inductive thematic analysis was carried out. Data were managed using QSR NVivo 12.

Results:

Our study reveals the different levels of barriers that adversely affect the successful integration of PC and cancer management. Emerging from the findings are patient- and family-level barriers (denial of the primary diagnosis and understanding of PC and financial constraints), service provider–level barriers (healthcare providers’ misunderstanding of PC and late referrals), and institutional and policy-level barriers (infrastructural and logistical constraints, non-inclusion of PC in the National Health Insurance Scheme, low staff strength).

Conclusion:

We conclude that different levels of barriers are encountered in the integration of PC in cancer management. There is a need for policymakers to develop comprehensive guidelines and protocols for the integration of PC into cancer management. These guidelines should address the various levels of factors that serve as barriers to PC integration. The guidelines should also emphasise the importance of early referral for PC and educate service providers on the benefits of PC for patients with life-limiting illnesses. Our findings underscore a need to include PC services and medication in the benefits package of the health insurance scheme to reduce the financial burden on patients and their families. In addition, continuous professional training of all cadre of service providers is needed to facilitate PC integration.

Background

Palliative care (PC), which refers to an approach centred on improving the quality of life of persons with life-limiting conditions and their families, has become essential to the continuum of comprehensive cancer care. ¹ Evidence suggests that incorporating PC in routine cancer management significantly improves their quality of life, ² promotes better symptoms management, ³ ensures cost-effectiveness in care, ⁴ and enhances the survival chances of patients. ⁵ Notwithstanding, there is high unmet need for PC.

Globally, approximately 56.8 million are in need of PC. ⁶ Of this number, 54.8% (representing 31.1 million) are in early stage need of PC while the remaining 45.2% (representing 25.7 million) are near the end-of-life (EoL). ⁶ The World Health Organisation (WHO) reports that only 14% of persons in need of PC receive care. ⁷ Low- and middle-income countries (LMICs) account for nearly 70% of the global adult population in need of PC. A report from the WHO indicates that compared with the other regions, Africa has the greatest need for PC, with its adult need for PC being at 2163 per 100,000 adults. ⁶

Some studies have reported that to address the unmet need for PC services, particularly in oncology, there is the need for an integrated system.^6–8 Here, integration of PC refers to bringing together administrative, organisational, clinical and service components of care in a collaborative way to improve the health outcomes of patients and their families. ⁸ This involves an integration of care dimensions (i.e. physical, psychosocial and spiritual care), service delivery models, (e.g. home-based, in-patient, out-patient models, or a combination of all three), and referral systems.

In Ghana, the call for PC integration into the national healthcare system culminated in the formulation of the National Strategy for Cancer Control. ⁹ Korle Bu Teaching Hospital (KBTH) was the first tertiary hospital in Ghana to operate specialist PC services and has since been the main health facility providing PC services to the country’s most populous city, Accra, and its peripheral regions. ⁹ However, in the quest to integrate PC services into cancer management, several barriers are encountered. Previous studies have shown that poor perceptions about PC, late referrals and financial constraints are recurring barriers that impede the successful integration of PC in cancer management.^10–12 Given that Ghana is a resource-constrained country that began PC service provision less than two decades ago, the barriers that constrain the integration of PC services may not necessarily be the same as in highly resourced nations. Yet, there is little empirical evidence that documents and assesses these barriers in resource-constrained settings like Ghana. Knowing the barriers to PC integration is critical for developing interventions to promote early PC integration. The study aimed to explore barriers to the integration of PC in cancer management in Ghana’s leading tertiary health facility.

Methods

Design

We employed a qualitative exploratory study design in this study. This study design was used because it offers investigators the chance to explore issues with ‘limited coverage within the literature and allows the participants of the study to contribute to the development of new knowledge in that area’. ¹³ The study was conducted between October and December 2021.

Setting

This study was carried out at the KBTH. Specifically, the study was conducted at the Palliative Care Unit which is located within the Department of Family Medicine at the Korle Bu Polyclinic. KBTH is the first tertiary hospital in Ghana to operate specialist PC services and has since been the main health facility providing PC services to the country’s most populous city, Accra, and its peripheral regions. ⁹ The PC unit has a team that offers both out-patient and in-patient consultation to clients as well as home visits. In-patient consultations are organised three times a week whereas home visits are done every Wednesday.

Sample

KBTH has a total of nine professionals in its PC unit; however, at the time of the study, one of the members of the PC team was on sabbatical. The remaining eight service providers were approached; however, one person declined to participate without giving any reason. Healthcare professionals were recruited based on the inclusion criteria that (a) must be a member of the PC unit and (b) must have been working with the PC unit for at least 6 months. Patients were purposively sampled based on the following inclusion criteria: (a) must have been diagnosed with any form of cancer at any stage (0–IV), (b) be an adult (i.e. they should be above 18 years) and (c) be receiving PC for at least 1 month and be able to verbally communicate. The patients were identified at the PC unit. As the PC unit provided services not only to cancer patients, we asked the service providers to alert us when patients who came for their appointments were cancer patients. The first author then proceeded to contact the patient after they had completed their day appointment with the service provider and screen their eligibility to participate in the study. The first author introduced himself and the purpose of the study and then obtained written consent from them. Most of the patients who were screened for eligibility to participate in the study were accompanied by caregivers; often a relative. Caregivers of patients who could not verbally communicate effectively were recruited.

The participants included a geriatric nurse, community health nurse (CHN), PC nurse specialist, pharmacist, family physician, general nurse, clinical psychologist and a social worker (see Table 1). Each member of the PC team had received a 6-week intensive training from the Institute of Hospice and PC in Africa, based in Kampala, Uganda. This training programme is a certificate course that has several modules. During this programme, participants are provided training on symptoms and pain management, spiritual assessments, and dealing with patients or caregivers’ psychological distress. By the fourth interview with the patients, no new analytical data were emerging, indicating a point of saturation. To confirm that we had reached data saturation, we included the caregivers of two patients. These interviews also revealed similar themes and patterns. In all, we conducted 13 interviews with service providers (7), patients (4) and caregivers (2).

Table 1.

Participants’ socio-demographic characteristics.

Participant ID	Age	Specialty	Years of experience	Marital status	Type of cancer
SP1	30–34	Geriatric nurse	2	N/A	N/A
SP2	30–34	CHN	8	N/A	N/A
SP3	55–59	PC nurse	9	N/A	N/A
SP4	50–54	Pharmacist	9	N/A	N/A
SP5	35–39	Family physician	9	N/A	N/A
SP6	35–39	General nurse	9	N/A	N/A
SP7	30–34	PC nurse	2.5	N/A	N/A
P01_F	40–44	N/A	N/A	Married	Colon cancer
P02_F	65–69	N/A	N/A	Widowed	Uterine cancer
P03_M	45–49	N/A	N/A	Not married	Mandibular cancer
P04_F	55–59	N/A	N/A	Married	Breast cancer
CG1_F	35–39	N/A	N/A	Not married	N/A
CG2_F	25–29	N/A	N/A	Not married	N/A

CHN, community health nurse; N/A, not applicable; PC, palliative care.

Data collection

Data were collected through face-to-face (10) and phone interviews (3). We conducted three phone interviews because these participants requested that the interviews be conducted on the phone as they were either busy (2 service providers) or not ready for a face-to-face interview (1 caregiver). The interviews were conducted using a semi-structured interview guide (see Supplementary file) that was prepared based on findings from previous literature.^12,14,15 For instance, Hawley ¹⁴ reports that some barriers to PC integration include strict eligibility criteria to receive PC services, lack of resources, reluctance to refer, patients’ reluctance to be referred and ignorance about what constitutes PC. These studies informed the probing questions in the interview guide. The first section of the interview guide asked questions about the participants’ socio-demographic characteristics. This was followed by a question that solicited responses about different levels of barriers to the integration of PC at KBTH. The first author contacted members of the PC team to schedule the date and time of the interview. Prior to each interview, participants were briefed about the objective of the study. Also, participants were informed about the voluntary nature of the study and the absence of any compensation whatsoever. Each face-to-face interview was recorded using a tape recorder. For the telephone interviews, the principal investigator put the call on loud speaker and recorded it using a tape recorder. This was done in the office of the principal investigator. Hence, no other person could eavesdrop on the interview even though it was on loud speaker. In addition to that, field notes were taken to record significant non-verbal gestures that would be essential in the interpretation of the text data. The interviews were all conducted in English Language; they lasted between 25 and 54 min.

Analysis

Verbatim transcription of the recorded interviews was done. QSR NVivo 12 was used to manage and analyse the data by adopting the five-staged thematic analytical framework for qualitative research. ¹⁶ The transcripts were imported into the software to follow Lacey and Luff’s five-staged thematic analyses. The decision to use this analytical framework is premised on the fact that it facilitates a more comprehensive analysis and creates an enabling environment for the inclusion of emergent concepts. ¹⁶ The first stage involved getting familiar with the data. We achieved this by reading through all the transcripts. This was followed by organising concepts and ideas that emanated from the study to determine similarities and variations in the barriers to PC integration. Using the ‘nodes’ function of the QSR NVivo 12, key narratives and phrases were identified; codes were assigned accordingly. After assigning the codes, recurring issues and patterns were identified and categorised as themes and sub-themes. The authors deliberated on the emerging themes to arrive at the final themes and sub-themes. The final stage of the analysis involved interpretation of the findings in corroboration with existing literature.

Rigour

Rigour in qualitative research is necessary to achieve trustworthiness. Key to achieving this is the issue of credibility, transferability and dependability.^17,18 Credibility refers to the degree of trust that can be placed in the accuracy and truthfulness of the research findings. ¹⁷ It is determined by assessing whether the findings are a credible reflection of the original data collected from the participants and whether the interpretation of their views is accurate and reasonable. To ensure credibility, only verbatim quotes from the participants were used. Also, the authors strictly adhered to the research methodology. Member checking (i.e. a technique used to validate the findings of a study by seeking feedback from the participants who were involved in the research) was done with two of the respondents a week after the data collection for them to verify the results. ¹⁹ Moreover, credibility was also ensured by designing the interview guide to reflect the objectives.

Transferability in qualitative research pertains to the extent to which the research findings can be applicable to other contexts or settings with different participants. ¹⁷ Researchers often facilitate the evaluation of transferability by providing detailed and comprehensive descriptions of the study methods, participants and findings. In the context of this study, this was achieved by providing detailed descriptions of the study setting and the methods for selecting the participants.

Dependability refers to the consistency and reliability of the research findings over time. It entails that the study’s interpretations, recommendations and conclusions are well-supported by the data collected from the study’s participants. ¹⁷ To achieve this, maintaining an audit trail has been advocated. This implies documenting the research journey from the initiation of a project to the presentation of its outcomes in a clear and comprehensive manner. ¹⁸ The authors ensured this by leaving an audit trail of audio-records, transcripts, interview questions and consent forms for any future confirmatory audits. ¹⁹

Results

Table 2 provides a summary of the emerging themes from the analysis. Our study reveals the different levels of barriers that adversely affect the successful integration of PC and cancer management.

Table 2.

Barriers to PC integration at KBTH.

Level of barrier	Emerging themes
Patient and family level	• Denial of the primary diagnosis and poor understanding of PC • Financial constraints
Service provider level	• Healthcare providers’ misunderstanding of PC • Late referrals
Institutional and policy level	• Infrastructural and logistical constraints • Non-inclusion of PC in the NHIS • Low management support and funding constraints • Low staff strength

KBTH, Korle Bu Teaching Hospital; NHIS, National Health Insurance Scheme; PC, palliative care.

Patient- and family-level barriers

At the patient and family level, two main barriers emerged: denial of the primary diagnosis and poor understanding of PC, and financial constraints.

Denial of the primary diagnosis and poor understanding of PC

Participants expressed that often, patients and their families refuse to accept cancer as a disease that requires biomedical help. Rather, they perceive cancer as a spiritual condition. Such attributions towards the disease impede health seeking and slow early referrals which is central to PC integration. Participants also asserted that some patients associate PC with death. Misconceptions about PC led to non-compliance when patients are referred for PC. This affects referral and early initiation of PC services which are the benchmark for successful PC integration.

Some of them [patients] think cancer is a spiritual thing and that slows the integration process. The reason is that, if patients think that cancer is a spiritual disease, they will report late to the hospital and that will affect referral for palliative care. In the end, instead of us to provide palliative care, we end up providing end-of-life care. (SP1_IDI_30–34 years)

Another participant also shared this view:

Some patients don’t have an in-depth understanding of what palliative care is, and what it seeks to do. Based on what they hear from others; they associate palliative care with end of life. So, they have a phobia of the care. (SP5_IDI_35–39 years)

Financial constraints

Service providers asserted that patients often faced financial difficulties. The patients themselves also affirmed this assertion. These financial constraints tend to limit patients’ capacity to access vital PC services such as home visits, thus creating a loophole in the integration of PC. Here is what some participants had to say:

Some of the patients cannot afford services like home visits. In such cases, we do not do home visits for those who are financially challenged. The cost depends on the distance. For example, from here to Kasoa, we charge 600 cedis [$41]. The distance will tell us how to charge. So, if in a day you are paying 600 cedis [$41], I am sure there is no way you will opt for a home visit. So, that affects how we are able to integrate home visits too. (SP7_IDI_30–34 years)

When you come for appointments, you will have to pay a consultation fee. But when they refer you to go for palliative radiation at radiotherapy, you will pay more for the treatment. That is a challenge because it is expensive and is not covered by national health insurance. Sometimes, I am unable to afford the cost of the morphine because it is very expensive. So, you can just imagine. (P04_Female_55–59 years)

Service provider–related barriers

Under this level of barrier, two themes emerged: healthcare providers’ misunderstanding of PC and late referrals.

Healthcare providers’ misunderstanding of PC

Participants reported that primary healthcare providers equated PC care to EoL care. As such, they are known as the angels of death. Service providers expressed concerns about how the wrong perceptions about PC were a challenge to the integration process. According to the participants, this perception of PC as being equal to EoL care gives the impression that PC services must be provided only when the patient has a poor prognosis.

Some of our colleagues in the other departments call us the ‘Angels of Death’. They think that palliative care is the same as end-of-life care. When we get to the wards, they will start saying that the Angels of Death are here. So, those are some of the misconceptions that they have. (SP2_IDI_30–34 years)

Another service provider stated that

The primary service providers in other departments who refer cases to us don’t understand palliative care. So, we even got the name ‘angel of death’. So, everywhere we go, they will be like the angels of death are coming. Such perceptions do not help us because it delays referral and worsens the patients’ health outcomes. (SP3_IDI_55–59 years)

Late referrals

Corollary to the findings about equating PC to EoL care was the challenge of late referral of patients for PC services. According to the participants, primary healthcare providers are supposed to refer cancer patients to the PC unit at the beginning of the healthcare journey. However, this standard is not adhered to. One participant opined that primary healthcare providers attempt to exhaust all the curative options before they are convinced to refer to the PC unit for palliation to begin.

As I said from the beginning, they don’t refer early. Ideally, the departments and general health care providers should refer cancer patients to the palliative care unit early but they don’t do that. They wait till the person is in the advanced stage where nothing much can be done for the person. I quite remember that for some referrals that we get here, they refer today then the person is gone [dead] the next day. Some even are referred and by the time you get to the ward, they have passed away. That is really how bad the referral system is over here. (SP2_IDI_30–34 years)

One of the caregivers narrated that late referrals worsened their relative’s condition:

[sighs] . . . for me, I think that is where I have a problem with the system here. When we came at first, we were at radiotherapy so that is where we were receiving treatment. My mother was going through cycles of chemotherapy to shrink the cancer but it was working and they kept doing it until one day, they [primary providers] told us that they are done with their part so they are referring us to polyclinic to come for palliative care. At that time, the condition had worsened. So, I think that if we had been referred here early, perhaps her condition would have been better than now. (CG2_Female_25–29 years)

Institutional and policy-related barriers

The main theme that emerged at this level of barriers included infrastructural and logistical constraints, non-inclusion of PC in the National Health Insurance Scheme (NHIS), and low staff strength.

Infrastructural and logistical constraints

Participants indicated that one of the major challenges was infrastructural and logistical inadequacies. They asserted that the space available for practicing PC was not in the best shape to facilitate PC integration. Meanwhile, infrastructural space is needed to promote in-patient care, which is one of the three models of care used for integrating PC at KBTH. Also, home visit, which is another model for the integration of PC at KBTH, was impeded by the lack of vehicles for transportation. Others viewed logistical constraints from the perspective of inaccessibility to opioids.

Unavailability of medications is another challenge to providing integrated care to our patients. Sometimes, we don’t have morphine because there is a quota that has been allocated to Ghana. There is a quota for the opioids and so we cannot exceed that quota. I think for Korle Bu, we cannot get more than 2 kg of morphine powder a year, which is used for the preparation of the syrup. Komfo Anokye is also doing the same. That is why sometimes we have shortages of morphine. (SP4_IDI_50–54 years)

Another participant had this to say:

My sister’s condition requires hospitalisation but when we come here and I raise that issue, they [PC team] will tell us that they don’t have beds for that. So, what they can do is the appointments that we come every two weeks or they will come for home visit. So, the hospital should do something about it. (CG1_Female_35–39 years)

Non-inclusion of PC in the NHIS

Participants expressed that in integrating PC, the cost of PC services and medications should be absorbed by the NHIS. However, the situation is different. The NHIS does not include PC within its benefits package for its subscribers. From the participants’ perspective, patients spent a lot of their money on radiotherapy even before they are referred for PC. As such, by the time they report to the PC unit for care, they would have exhausted their financial resources.

The cost of palliative care for cancer patients is not covered by the health insurance. That is one major challenge to achieving full integration with cancer care. We have been trying to get it included in the health insurance but it is still pending. This year we had two meetings and we are hoping that next year, we would be able to have it on the NHIS. (SP3_IDI_55–59 years)

One of the patients reported that, ‘the insurance does not cover palliative care. So, it is very difficult for me. Coming here for medications has really drained me financially. I am sure things would have been better if it was covered by the health insurance’. (P02_Female_65–69 years).

Low management support and funding constraints

The results from the thematic analysis showed that service providers felt management provided little support for PC. In the view of the participants, management did not prioritise PC as they cared little about the psychological well-being of service providers. This is what one of the participants had to say:

Management does not care about how we debrief or how we retreat. We get emotionally drained as service providers in this field but no one seems to care about that. But this is a kind of job that requires consistent follow up on the team to find out how we are doing. So, that is also a big challenge because if the team members are not fine, then the work will not go on well. (SP6_IDI_35–39 years)

Another participant expressed their sentiment in the quote below:

The challenge with trying to make use of the opportunities around to integrate palliative care better for patients and their families is that there is no budget allocated for organising workshops, seminars and training sessions for service providers. (SP3_IDI_55–59 years)

Low staff strength

It was evident from the study that staff strength at the PC unit was low in relation to the task that service providers had to perform. This challenge emanated from the lack of sufficient healthcare providers who specialise in PC. The participants also shared concerns about the hospital’s inability to retain some key staff for the PC unit.

Like I mentioned, we don’t have adequate human resources. Palliative care is not an individual thing; it is delivered through a team of health professionals. Yet, we don’t have sufficient healthcare providers in palliative care. (SP6_IDI_35–39 years)

No, I think our staff strength is challenging because as at now, the physician can only come on Mondays, and for the rest of the days, she is part of the family resident programme. (SP4_IDI_50–54 years)

Discussion

The present study used a qualitative exploratory study design to explore barriers to the integration of cancer PC in Ghana. From the findings, there are different levels of barriers that impede PC integration in cancer management. These levels of barriers include patient- and family-level barriers, service provider–level barriers, and institutional and policy-level barriers. Overall, the findings are a reflection of the socio-ecological model that posits that there are different levels of factors that affect individuals and healthcare systems. ²⁰ Our study shows that at the patient and family level, there is the challenge of the denial of the primary diagnosis and poor understanding of PC. This finding mirrors that of several studies.^21–23 The present finding reveals that patients who ascribed their disease to spiritual causation often reported late for palliation. A plausible explanation for this could be that patients with poor knowledge and wrong perceptions about PC would want to assume a state of denial and explore all other options available to them before considering PC. This could significantly delay the initiation of PC.

Previous studies on PC in LMICs have established that shortage or lack of funding at the institutional level is a major barrier to the advancement of PC.^20,24,25 Our findings support this. Financial constraints on the part of patients and their caregivers were what contributed to the challenges encountered in integrating PC. A plausible justification for this finding could be that patients’ financial constraints significantly delay referral for specialist PC which is a critical pathway in PC integration. ⁹ Another possible explanation for this finding could be that, often patients spent a significant proportion of their financial resources in search of curative treatment. Hence, by the time they resort to PC, their financial resources would have been significantly depleted, thereby, making it difficult for them to receive the requisite care.

Ideally, patients are expected to be referred for PC immediately after their diagnosis.^9–11 However, the study found that there were substantial delays with referrals. It is possible to explain this observation from the point that there is often a disconnect with the understanding that PC is relevant to all patients facing a life-limiting disease, rather than being prognosis-dependent. ²⁶ Hence, service providers mainly refer to the EoL. This assertion was reflected in our findings that service providers’ misunderstanding of and equating PC to EoL care was a barrier to integrating care. The result corroborates Aldridge et al.’s ²⁷ study that showed that clinicians often equated PC to EoL, thereby exacerbating the risk of delays in referring patients. This may also be due to the fact the hospital operates a specialist PC system. Therefore, primary health service providers may feel that they can offer some primary and intermediary PC services and only bring in the specialists when the patient’s case is complex or at the EoL.

It is evident from the study findings that there are infrastructural and logistical constraints that impede the successful integration of PC in cancer management. The absence of a standard PC centre coupled with the lack of beds at the PC unit seriously affects the capacity of the health facility to provide in-patient care to patients. Also, the constrained infrastructure does not allow service providers to ensure the patients’ privacy. These findings are corroborated by a related study that found infrastructural constraints to be among the key barriers to the effective integration of PC. ²⁸ The result suggests that apart from infrastructural inadequacies, there were challenges with respect to the availability of intravenous opioid medications and other pain-relief medications. This is consistent with findings from previous studies.^29,30 Thus, highlighting a need to prioritise the supply of essential medications to facilitate PC integration as described in the public health model of developing PC services. ³¹

Our findings indicate that the non-inclusion of PC services and medication in the benefits package of the NHIS was a major challenge to achieving integrated PC. This result is corroborated by Sarfo-Walters and Boateng’s ³² findings that Ghana’s NHIS does not cover PC services. Without coverage from the NHIS, patients are more likely to suffer catastrophic health expenditures. Moreover, in situations where the patient is in the EoL, financial constraints that arise as a result of the non-inclusion of PC services in the NHIS may be a disincentive for patients to seek PC services, thus, suggesting that inclusion of PC services into the NHIS would facilitate PC integration. The need to include PC in the NHIS is reflective of the International PC Initiative (IPCI) roadmap for a public health approach to providing PC services. ³¹

Overall, the findings suggest that PC integration fits into the WHO health system building blocks which identify seven core components (i.e. leadership and governance, service delivery, health system financing, health workforce, medical products, vaccines and technologies, and health information systems). ³³ Service provider–level barriers, such as misunderstanding of PC and equating it with EoL care, delays in referrals, and lack of availability of essential medications, highlight lapses in the service delivery of integrated PC services. Institutional and policy-level barriers, such as the absence of standard PC centres, lack of beds at the PC unit and non-inclusion of PC services in the NHIS benefit package, indicate the need for strong leadership and governance to create policies and systems that support the integration of PC services into cancer management.

Strengths and limitations

The use of qualitative research methods does not allow for generalisation of the findings. Including primary healthcare providers would have provided a deeper understanding of the delays in referrals. However, this was not possible given the short time of the study. Also, because the study was limited to only patients who visited the PC unit, we could not explore the perspectives of other patients such as those who relied solely on home-based PC services and those who were admitted at the radiotherapy or oncology department. It is possible that we may have introduced some bias during the data collection process, as the authors were responsible for screening patients for eligibility to participate in the study. This could have influenced the selection of patients and caregivers and may have resulted in the exclusion of some eligible participants.

Conclusion

We conclude that different levels of barriers are encountered in the integration of PC in cancer management in Ghana’s tertiary health facility. There is a need for policymakers to develop comprehensive guidelines and protocols for the integration of PC into cancer management. These guidelines should address the various levels of factors that serve as barriers to PC integration. The guidelines should also emphasise the importance of early referral for PC and educate service providers on the benefits of PC for patients with life-limiting illnesses. Our findings underscore a need to include PC services and medication in the benefits package of the health insurance scheme to reduce the financial burden on patients and their families. In addition, continuous professional training of all cadre of service providers is needed to facilitate PC integration.