Cancer care workforce in Africa: perspectives from a global survey

The burden of cancer in Africa is growing with recent estimates reporting 1.06 million new cancer cases per year. This figure is expected to increase by 102% to 2.12 million by 2040 [1]. Moreover, the mortality-incidence ratio in Africa is substantially higher than high-income countries (Africa 0.66 vs Northern Europe 0.40 vs United States 0.29) [1]. This disparity in outcome is largely explained by the fact that at least 80% of patients in Africa are diagnosed with cancer at an advanced stage [2]. Other factors that likely contribute to poor cancer outcomes in Africa include limited health system infrastructure, a scarcity of specialized health personnel (including oncologists and pathologists) and patients’ inability to afford cancer treatment [3].

Given the late stages of disease at diagnosis, the vast majority of the patients in Africa are treated with palliative intent. The number of oncologists in Africa ranges from zero in countries like Lesotho, Benin, Gambia, South Sudan and Sierra Leone to single digits in Malawi, Burkina Faso, Rwanda and Togo, and up to 1500 in Egypt [4]. This translates to a very high case-load per physician. Moreover, with the scarcity of medical oncologists (MOs) in Africa and other low-middle income regions, a large proportion of cancer patients may not even have the opportunity to see a medical oncologist in consultation. These patients therefore do not have access to early interventions to reduce suffering and improve outcomes.

A recent review included data from 21 countries in Africa, and showed a high burden of incident cases per oncologist while also highlighting the large numbers of oncologist in Africa who practice both radiation and medical oncology (i.e. clinical oncologists) further compounding the workload [4]. Four studies have explored medical oncology workload in high-income countries (HICs) [5–8]. In these studies from New Zealand, The United States, Australia, and Canada the MO case-load was 175–280 new patients/year. We have recently reported the results of a global analysis of MO workload in which we observed striking difference in case volumes between HICs and low-middle income countries (LMICs) [9].

To address gaps in knowledge, we report a sub-set analysis of a global study in which we describe: 1) the clinical workload of African oncologists compared to those of other countries; 2) available infrastructure and supports; and 3) delivery of clinical care in the African context. Data from this study will inform cancer policy and human resource planning in Africa.

In this study we describe the clinical workload, infrastructure and delivery of care among African clinical oncologists. For comparative purposes we also present data from 47 other (predominantly high-income) countries. Several important findings emerged. First, the clinical workload is substantially higher among African oncologists compared to oncologists in other countries. Second, African oncologists are substantially older than oncologists in other countries which suggests that without new models of care and an increase in capacity, clinical workload volumes might worsen in the coming years. Third, job satisfaction among African oncologists is lower than oncologists in other countries. This raises the concern of physician burn-out which could further exacerbate human resource challenges. Finally, the most important barriers to oncologic clinical care reported by African oncologists are the inability of patients to pay for care, high clinical volumes, limited access to new therapies, standard chemotherapy and radiotherapy.

We have recently completed the first global analysis of medical oncology workload and available infrastructure [9]. In our global analysis we found striking differences in workload and delivery of care among countries in different World Bank income categories. Annual case volume in LMICs (median consults of 425, 40% respondents seeing > 500 consults) was substantially higher than UMICs (median consults of 175, 14% seeing > 500 consults) and HICs (median consults of 175, 7% > 500 consults) (p < 0.001). The results presented in this study confirm these disparities in oncology workload, given that most of the African respondents are from LMICs and the other respondents in our comparative group are predominantly from HICs.

Our results are consistent with previous reports highlighting the scarcity of cancer treating physicians in Africa [10], and are also similar to findings from a recent study published by Mathew [4] which evaluated the global clinical oncology workforce by estimating the oncologist-to-cancer burden ratio in 93 countries, including 21 from countries in Africa. The new consult volumes per oncologist in Mathew’s report are substantially higher than reported in our analysis. Whereas we report a median of 325 new consults per oncologist per year, Mathew reports several countries in Africa with workloads greater than 1000 new cancer cases per oncologist per year. While Mathew’s analysis are based on the International Agency for Research on Cancer estimates, our results are based on survey responses which reflect the actual clinical burden, as opposed to projected estimates of new cancer diagnoses and our results may therefore be more representative of the clinical oncology workforce burden. A significant proportion of patients diagnosed in Africa do not come into contact with the cancer health system. Therefore not all of the estimated incident cases are seen by an oncologist. However we do anticipate that with increase in awareness the clinical volume and cancer workload will increase substantially.

Our results also highlight not only the magnitude of workload but also the complexity of cancer care delivery in Africa; most cancer physicians prescribe chemotherapy and radiotherapy and treat patients with all cancer sites. In Africa this model may be driven by limited resources, however, the joint practice and accreditation of clinical oncology to cover both chemotherapy and radiotherapy is standard for many countries including the United Kingdom. Thus this may also reflect clinical practice and training based on Commonwealth and other structures. A 2014 survey by the European Society for Radiotherapy and Oncology revealed that 75% of radiation oncologists also administered systemic cancer therapies [11]. The United Kingdom still continues to train dual-prescribing clinical oncologists alongside medical oncologists. However, realizing the evolving complexities of systemic cancer therapy and radiation therapy in the modern era, the European Union issued a directive in 2011, recognizing medical oncology as a separate entity from radiation oncology [12].

Another important finding from our analysis is job satisfaction is lower among African oncologists compared to other countries. This may relate to high clinical workload, the complexity of delivering care in low resource settings, and poor compensation. These trends may precipitate oncology burnout in Africa, contribute to brain drain among African oncologists [13], and lead to fewer African physicians willing to specialize in oncology [14].

With the expected future increase in cancer incidence and a disproportionate burden of disease in LMICs, this survey highlights an important healthcare personnel gap in scaling up cancer control programs in Africa. The combination of increased workload, complexity of cancer care, lower job satisfaction rates and projected increases in cancer incidence calls for continued growth in the oncology workforce in addition to developing new models of care that make use of service extenders. The recent 2017 World Health Organization cancer resolution to improve access to quality cancer care, cannot be realized in LMICs without pragmatic strategies to increase the cancer care delivery workforce [15, 16]. Realistically the stagnant growth in the clinical oncology workforce will not be sufficient to manage the projected increase in the burden of disease in the region. Therefore in these resource-limited settings, predominantly in sub-Saharan Africa, mechanisms for using service extenders and task shifting is paramount, but should be done within an implementation science framework to ensure these adaptive models do not compromise delivery of high quality care. Countries like Rwanda have experimented with task shifting and re-training of general practitioners with reported success on a small scale [17].

While a significant proportion of African respondents reported training outside of their country compared to respondents from other countries, 67% reported having training programs in their center. For many years the International Atomic Energy Agency, supported the training of clinical oncologists from many LMICs and UMICs (including physicians from African centers with existing RT facilities) to be trained in HICs [18]. At the time, this model was thought to be a cost-effective approach to improving the cancer workforce. However, this meant that centers without planned radiotherapy facilities were without a clinical oncologist to manage solid tumors. Our data may therefore reflect a scaling up of oncology training centers across Africa. This shift in training African oncologists on the continent has been accompanied by significant reductions in the cost of training and is expected to curb the rates of brain drain [19]. Currently there is no database of oncology training centers on the continent, however the survey results and our clinical experience in Africa, are consistent with an increase in new and improved local training opportunities in medical oncology and the formation of regional training centers of excellence across the continent [20, 21].

Cancer care is complex and requires a multidisciplinary approach. The barriers highlighted here suggest that in addition to resource allocation and healthcare planning to expand the cancer care workforce, there should be a parallel focus to improve access to high-quality radiotherapy, surgery, and palliative care. Special focus is needed to identify funding for cancer medicines that offer substantial and clinically meaningful benefit. Moreover, availability of diagnostic platform systems such as radiology and pathology are severely limited across Africa [22] and require increased capacity to meet the population needs. Recent global data also describe the association between high clinical volumes and low job satisfaction [23]. Further work is required to understand the risk of burn-out among cancer care providers in the African context.

Our study results should be considered in light of important methodologic limitations. Due to the potential for selection bias, it is possible that our results are not generalizable to all African countries; this is particularly true given the small sample size. Our survey was delivered within the AORTIC network and therefore we are unable to assess a denominator to estimate the response rate to our survey. There is the possibility of a selection bias as countries with allocated resources for oncology may be disproportionately represented in AORTIC. This is however likely to result in a bias of our results towards the null, and the oncology workforce burden in Africa might be much larger than our survey results show. Finally, the workload metric (new consults per year) is a crude marker of clinical workload as the work involved across cancer sub-types and patient populations can vary substantially.

In summary, this study offers some insight into oncology workload and delivery of cancer care in Africa. Our data identify a significant gap in healthcare personnel. AORTIC is currently mapping out regional oncology training facilities, centers of excellence and oncology workforce in Africa with the aim of establishing an accurate database for future cancer care planning. Future work is needed to explore innovative models of care in order to mitigate the clinical workload and ensure high-quality care for patients in Africa.