Background

Globally in 2020, only 7 million people - approximately 12% of the 56.8 million people in need of palliative care - received it [1]. Low- and middle-income countries account for 76% of the global need in palliative care [1]. For countries to meet this demand for essential care, it is recommended to integrate palliative care services into existing healthcare systems and track them with proper data and indicators to ensure quality of services [2,3,4]. Quality data directly affects the provision of quality palliative care as it is a major factor in financial decision-making by governments, third-party payers, and insurers [5]. Capturing patient-level data is a necessary condition in identifying and addressing gaps in palliative care [6]. One method of ensuring quality data collection is using mobile phones and tablets as tools to further advance both data collection and service delivery in healthcare [7, 8]. This is especially relevant in Africa where, on average, there are 76 mobile-cellular subscriptions per 100 people [9]. Leveraging this technology, mobile health (mHealth) interventions in the form of short message service (SMS) notifications have been used to increase appointment adherence and rapport with patients in oncologic settings of rural Uganda [10]. To facilitate collaboration between current mHealth activities, the African Palliative Care Association (APCA) has initiated a mHealth Research Network [11].

Palliative care development in Uganda

A systematic approach to palliative care in Uganda was established in 1993 by Dr. Anne Merriman and Fazal Mbaraka as a model for African palliative care [12]. The high prevalence of HIV/AIDS and cancer drove the development of palliative care in Uganda [12]. The majority (85%) of the palliative care services in Uganda are provided by public hospitals, and there are also 77 private hospices or palliative care services in private hospitals in the country [13]. The overwhelming amount of funding for palliative care, 94%, comes from external donors [14]. In addition to this external funding, there are many champions of palliative care development in the government and private sector in Uganda who have worked to increase education and awareness of the need for palliative care as well as ensure access to essential medicines. Medicines, such as opioids, are critical to relieve pain and other symptoms in palliative care provision [4, 12]. Oral morphine, produced from imported morphine powder, was introduced to Uganda in 1993 as a cheap and effective option for pain relief [12].

Advocacy by the Palliative Care Association of Uganda (PCAU) and Hospice Africa Uganda (HAU) contributed to official registration of morphine, allocation of state funds and free provision of oral morphine to patients [14]. Imported powdered morphine is used to produce oral morphine of two concentrations - 5 mg/ml green and 50 mg/ml red morphine. Morphine is distributed in Uganda via two suppliers - National Medical Stores (NMS) which supplies public facilities and Joint Medical Stores (JMS) which supplies private facilities [15, 16]. The International Narcotics Control Board (INCB) recognizes oral morphine as a controlled substance and this means governments must license, supervise, and report to INCB the production and distribution of all morphine [17]. In Uganda, this issue is addressed by tracking morphine distribution by suppliers and requiring receiving facilities to be accredited by PCAU. Accreditation status depends on three criteria that must be present at a facility: 1) properly trained palliative care staff, 2) double-locked morphine storage cabinet, 3) proper morphine tracking records [3]. Uganda was the first country to allow nurses and clinical officers with specialized training to prescribe opioid analgesics to increase accessibility and uptake of morphine [18]. Because nurses provide a significant amount of palliative care across the globe, providers in many other countries aspire to implement similar legislation [19].

Palliative care data in Uganda

In Uganda, it is estimated that only 10% of the need for palliative care is met. Through their national health management information system (HMIS), the Ministry of Health (MoH) collects facility-level data on number of patients seen in pain and amount of morphine prescribed [20]. There are national and regional efforts to gather better data on palliative care to fully describe the current situation of palliative care services in the country [21]. On the national level, PCAU collects data on facility accreditation status and morphine distribution. In 2015, PCAU, the Center for Hospice Care/Hospice Foundation and the University of Notre Dame piloted an mHealth surveillance project to collect key palliative care indicators via mobile phones and address the lack of quality data at a subset of healthcare facilities across the country [15]. The project collects palliative care data at 20 facilities geographically dispersed throughout the country. Building on this project, PCAU is now working in partnership with the MoH to integrate palliative care data collection into the national health information system so Uganda can have a consistent and reliable source of palliative care data. In March of 2021, the MoH issued a directive for hospitals to allocate space for palliative care services, re-affirming the need in quality palliative care collection nationwide [22].

Research aims

Given political, social, and technological developments since the last audit was conducted in 2009, this study offers a timely and comprehensive assessment of the current situation of palliative care in Uganda [21]. The aim of this study is to describe the palliative care situation in Uganda in terms of availability of morphine and services by district, availability of personnel at facilities, patient demographics and diagnoses contributing to the palliative care burden. By assessing aggregate data from the mHealth surveillance project facilities, along with national morphine distribution and accreditation data, this study provides an in-depth analysis of key palliative care indicators on the regional and national levels and offers maps visualizing geographic distribution of palliative care services.

Methods

We conducted a descriptive quantitative study of data collected by PCAU. The analyzed data were obtained from three sources: the list of accredited facilities updated as of January 2020, morphine distribution data for 77 facilities for 2019, and mHealth surveillance survey (afterwards referred to as survey) responses from 20 facilities from January 2018 to February 2020.

National-level data

The indicators analyzed at national level were: number of accredited palliative care facilities, number and type of services that received morphine, volume of morphine distribution, and population coverage. The most recent national census data is from 2014 and were used for the population coverage assessment. The districts in Uganda are grouped into ten subregions [23]. Due to frequent changes in the number of districts in Uganda, these larger subregional administrative divisions were used as a more permanent and comparable option.

mHealth surveillance survey data

The survey consists of 75 questions and collects data from 20 facilities on: number of medical personnel involved in palliative care provision, morphine availability by type, palliative care patient diagnosis and mortality, palliative care services offered at the facility and challenges in offering these services. The 20 participating facilities represent both urban and rural locations, private and public facilities and are spread across ten of Uganda’s subregions. To make measurements comparable between facilities, monthly averages of numerical indicators (i.e., number of patients, volume of morphine used) were used.

Statistical tests

The geographic distribution of accredited facilities was mapped using ArcGIS Desktop 10.7. The palliative care situation in the country was described in terms of availability of morphine and accredited services by district, availability of personnel at facilities, patient demographics, and diagnoses. The relationship between the facility type and palliative care indicators collected was examined with Wilcoxon-Rank Sum test, Spearman’s Rank correlation, and Kruskal-Wallis test. P-values below 0.05 were considered statistically significant. One survey facility was excluded from the analysis because of an absence of data.

Ethical considerations

The study did not involve recruitment of vulnerable populations, collection of personal medical records, or any other sensitive information. No personal identifiers were collected with any data. All data is anonymized and were aggregated at the facility level for analysis. The data were stored on a password-protected personal computer of the principal investigators. Authorized members of the research team from PCAU, the University of Notre Dame, and Uganda Martyrs University had access to the data. The study was approved by the University of Notre Dame Institutional Review Board (NDIRB), protocol number is 20-05-6066 and Hospice Africa Uganda Research Ethics Committee (HAUREC) reference number HAUREC-083-20. The study adheres to all human data institutional guidelines of NDIRB and HAUREC.

Results

National level data

A total of 226 accredited palliative care facilitiesFootnote 1 were identified in Uganda, and of them, 154 (68.1%) were public facilities and 72 (31.9%) were private facilities. Only 77 (34.1%) of accredited facilities reported receiving pain-relieving morphine in 2019 (Table 1). A full list of those accredited facilities who received no morphine in 2019 can be found in Additional file 1.

Table 1 Palliative care accreditation and morphine use in Uganda by subregions, 2019 (n = 226)
Full size table

In addition to the 77 accredited facilities receiving morphine in 2019, four other facilities who were not accredited by PCAU received morphine. JMS distributed a slightly larger amount of morphine (100,627,500 ml) and more frequent deliveries of morphine than NMS that delivered a total of 9,120,000 ml. There was an inverse relationship in the volumes of morphine supplied by JMS and NMS that was statistically significant in May (p < 0.001), June (p = 0.003), September (p = 0.043), and December (p = 0.045) of 2019. The detailed morphine volumes by month provided to the mHealth survey facilities are listed in Additional file 2. In the months when the difference in the volume of green morphine delivered was statistically significant, NMS orders were on the rise when JMS orders decreased, and vice versa (Fig. 1).

Fig. 1
figure 1

Seasonal distribution of the morphine volume by facility type

Full size image

Geographical distribution of palliative care services

Accredited palliative care facilities were present in all ten of Uganda’s subregions, but 30 of the country’s 135 (at the time of analysis) districts had no palliative care facilities (Fig. 2). Almost a third of the districts in Elgon, a subregion in Eastern Uganda, had no palliative care facilities. The list of districts with no accredited palliative care facilities is provided in Additional file 3.

Fig. 2
figure 2

Map of accredited palliative care facilities in Uganda as of January 2020

Full size image

Population coverage with palliative care services

To evaluate the population coverage of palliative care services, we mapped the 2014 Census data available at the subcounty (administrative level 5) level on a 2020 administrative districts map [24]. These data were used to allocate the population of each 2014 subcounty to a 2020 district. Since the internal boundaries did not change, it is reasonable to use these data for comparisons over time. Based on the 2014 census data, 88.5% of people in Uganda live in a current district with at least one palliative care service available, whereas 18.0% of people live in a district with five or more palliative care services (Fig. 3, Additional file 4). Districts with the most palliative care services were Kampala [18], Wakiso [12], Tororo [6]; Jinja, Kabarole, Masaka, Mbarara, Ntungamo and Rukungiri each had five accredited facilities.

Fig. 3
figure 3

Population coverage of palliative care services

Full size image

Data collected from mHealth surveillance survey

Twenty facilities participated in the mHealth surveillance survey. One is excluded from analysis because of insufficient data, leaving a sample size of 19. These facilities represent 16 districts and 10 subregions of Uganda (Table 2).

Table 2 Type and location of the mHealth surveillance survey participating facilities
Full size table

Distribution of morphine

Based on the survey data from 19 facilities, private facilities on average received higher volumes of green morphine (5 mg/ml); although this finding was not statistically significant (Table 3).

Table 3 Morphine distribution in mHealth survey participating facilities, 2019
Full size table

Challenges in accessing morphine

All 19 facilities reported experiencing issues with receiving morphine. The most cited challenges among public facilities had to do with a supplier, while private facilities cited a non-specified ‘other’ category as the most common one (Table 4). It is worth noting that NMS delivers morphine to facilities, whereas JMS requires facilities to pick up morphine from them. The challenge “morphine not requisitioned” may mean a facility did not place an order in time, and this was more common for public facilities.

Table 4 Number of times a challenge in accessing morphine was reported during 2019
Full size table

Patient demographics and diagnoses

On average private facilities treated a larger number of palliative care patients. Public facilities treated a larger proportion (27.6%) of child patients than private facilities (10.0%). The difference in the number of patients across demographic groups and the number of deaths (Table 5) was not statistically significant.

Table 5 Palliative care patient characteristics seen at 19 facilities (January 2018 to February 2020)
Full size table

Cancer was the top diagnosis of patients seen (38.5%) followed by HIV/AIDS (17.0%). Cancer types are provided in Additional file 5. Palliative care teams are caring for patients with a wide variety of diagnoses; 38.6% of visits to palliative care facilities were attributed to more than 50 other conditions with hypertension, arthritis, and pains reported the most (Table 5, Additional file 6). Although the total number of visits was approximately the same in private and public facilities, percentage-wise public facilities saw more patients with other conditions (33.4%, compared to 5.2% at private facilities) while private facilities saw more cancer patients (31.3% vs 7.2% at public facilities).

Availability of palliative care staff and services

As shown in Table 5, public facilities had a larger total number of personnel directly involved in palliative care per patient than private ones (p = 0.003). Similarly, the number of involved doctors was also larger in public facilities (p = 0.042). On average, both types of facilities offer at least one out of six possible types of palliative care services throughout the year. Psychosocial, bereavement, spiritual support and home visits appeared to be most likely to be available in public and private facilities (Table 6).

Table 6 Average number of months palliative care services offered at mHealth facilities by type per year
Full size table

The average number of personnel per patient by health care worker type and facility were correlated to the average number of months a palliative care service was offered at a facility using the Spearman’s Rank correlation. Home visits, outreach, bereavement, legal, and spiritual services did not have a statistically significant correlation with the average number of any type of HCWs. There was a weak, negative correlation between the frequency of psychosocial services offered and the average number of other HCWs involved, which was statistically significant (p = 0.04).

Discussion

This study provides a description of palliative care availability throughout Uganda with a deeper look at providers, services and patient demographics and diagnoses at a subset of facilities prior to the COVID-19 pandemic. This is the first study to comprehensively map palliative care services and morphine use across Uganda.

A seemingly widespread population coverage with palliative care facilities of 88.5% may not capture the whole picture of access to such facilities by patients. Living in a district with an accredited palliative care facility does not necessarily equal accessibility and availability of the service, as it may be hindered by transportation challenges and high medical costs. In Uganda, cancer patients cited lack of money and transportation as major reasons for delaying treatment or missing appointments; some patients had to travel between 13 and 212 km to a medical facility [25]. In another study, 46% of HIV/AIDS patients reported difficulty traveling to a hospital as a main barrier to accessing care, while 23% reported that they cannot afford medical care [19]. Even though more than 80% of people in Uganda lived within 5 km of a health facility, many preferred to travel further distances to access private health centers due to their perceived better quality [26]. There are no studies evaluating geographical access to palliative care services in Uganda. Geographical access has been estimated for malaria patients using the cumulative case ratio method of defining the hospital’s catchment area [27]. More complex geographical analysis was beyond the scope of this study but is recommended as an important follow up in determining accessibility of palliative care services. A similar approach could be taken to analyze access to palliative care facilities and combined with data on outreach and home visit services from these facilities. All mHealth survey facilities in our study provided outreach and home visit services, though these were not offered consistently throughout the year, with transport and other costs noted as barriers to providing the services. It is important to track the extent of such services for a better understanding of palliative care coverage, especially in rural areas.

In 2013, Nabudere et al. stated cancer and HIV/AIDS patients contribute 80% of all palliative care patients in Uganda [20]. Our data from the mHealth survey showed that overall, 61.4% of patients were diagnosed with cancer, HIV/AIDS, or both, while the remaining 38.6% was attributed to other conditions including arthritis, hypertension, and cardiac disease. This supports other evidence that sub-Saharan Africa is undergoing an epidemiological transition, as the proportion of disability-adjusted life years contributed by non-communicable disease (NCDs) increased from 18.6% in 1990 to 29.8% in 2017 [28]. The top three contributors to the NCD burden across sub-Saharan Africa were other NCDs, such as congenital disorders, cardiovascular diseases, and neoplasms [28]. Palliative care is recognized by the World Health Organization as an essential component of a comprehensive response to NCDs [29]. Despite this, only 4% of African countries offer palliative care services to NCD patients as a part of primary care [30].

Due to the lack of epidemiologic data on NCDs burden among palliative care patients, governments and donors have invested in HIV/AIDS care, while other conditions requiring palliative care are prioritized lower [30]. This lack of funding may manifest as the lack of resources, including transport and staff. Our findings showed both private and public facilities reported lack of transport and facilitation as the challenges across the six types of services (bereavement, spiritual, psychosocial, legal, outreach, home visits) for both private and public facilities. Hence, it is important to increase awareness and advocacy of the universality of palliative care services to increase funding opportunities and address patient needs outside of HIV/AIDS care.

Although access to physical pain relief is only part of palliative care services, the availability of morphine remains an important indicator of palliative care services. The inverse relationship between the two national suppliers of morphine in Uganda may suggest NMS and JMS compete for morphine produced and available. There are more than double the number of public accredited facilities than private, but private facilities were supplied with more morphine. Past studies suggest this difference is due to higher uptake of private medical services by richer people [26, 31]. Our data did not allow for the analysis of patients’ socioeconomic status and distance to the private or public facilities. Private and public facilities in our mHealth survey cited the supplier stockouts and challenges with delivery or pick-up from NMS and JMS, respectively, as a major challenge in accessing the necessary volume of morphine. This again may indicate that transport challenges are experienced by suppliers and facilities and are hindering availability of palliative care. In the months of lower volume of morphine received, public facilities also reported a challenge of not ordering morphine on time which suggests the need for a more systematic approach to plan, order and manage morphine at the facility level.

Limitations

The study analyzed data from a self-reported survey that introduces the possibility of human errors. The number of months reported in the survey varied from 5 to 26 over a 26-month period; therefore, the presented analysis may not be representative of all 19 facilities to the same extent. Challenges in accessing morphine were only assessed for facilities that received morphine at least once in 2019; however, the accredited facilities that did not receive morphine were not surveyed as part of this study and may indicate even more challenges. Accreditation data was last updated in January 2020, however the period for which accreditation status was granted for was not recorded. Hence, some facilities on the list may no longer meet the accreditation criteria nor be actively providing palliative care services. PCAU typically monitors facilities quarterly, usually by phone call and occasionally by site visit. But especially in public facilities, staff are frequently moved to other facilities and sometimes hard to track. Using morphine ordering as one means of continued monitoring of accredited facilities is an important tool for initial analysis of the status of service provision at these facilities.

Geographical coverage with palliative care services was implied from a presence of at least one palliative care service in a district. Since the straight-line distance or road network distance to a palliative care service was not considered, the actual accessibility may differ from that presented here.

Conclusion

Palliative care was introduced to Uganda in 1993, and since then palliative care services provision has improved in quality, volume, and geographical coverage. Although accredited facilities are present in most administrative districts, more research is needed to evaluate the actual accessibility of these services, especially to rural residents. The existing services, both private and public, are limited by the amount of pain-relieving morphine, financial and transport resources. By aggregating morphine distribution data with the list of accredited facilities we were able to show there are many accredited facilities that do not regularly receive morphine; although the underlying reasons remain unclear, these facilities present an opportunity to better understand and address reasons that morphine may not be used when medically needed. The facilities primarily treat cancer, HIV/AIDS patients, and patients with other conditions such as hypertension, arthritis, and congestive cardiac failure. This indicates an important epidemiological transition, and the need for palliative care services to partner with organizations focused on treatment of NCD’s to increase access to those in need and ensure sustainability of palliative care services.