3.1 Differences in Healthcare and Outcomes Among Countries
Inequity in access to cancer treatment is known to have a measurable impact on patient outcomes [
13,
14]. In Europe, the recent EUROCARE studies have investigated differences in cancer treatment and 5-year survival rates across European countries [
13,
14]. All-cancer 5-year survival ranged from 37 % in Slovakia to 61 % in Sweden, and correlated linearly with spending (
R = 0.8; spending was adjusted to take account of the varying purchasing power of national currencies). In general, countries with high spending had high numbers of diagnostic and radiotherapy units, and the number of magnetic resonance imaging (MRI) units per capita directly correlated with survival (
R = 0.7), reflecting the importance of early diagnosis. Surprisingly, given the political focus on cancer drugs, 5-year survival was more closely linked to the availability of radiotherapy units than to overall healthcare spending, highlighting the importance of efficient resource allocation [
13]. A similar analysis used data from the EUROCARE-5 study of 107 cancer registries in Europe to investigate 5-year survival rates for ten common cancers [
14]. Countries with high national expenditure on cancer care generally had higher survival rates than those with low expenditure. However, differences among countries with high levels of healthcare spending demonstrated that other factors influence survival rates. Some of these differences may reflect an uneven distribution of different cancer types; for example, the incidence of melanoma in Sweden is approximately three times that in Spain [
15]. For patients with similar cancers, comparisons between countries are complicated by a number of factors influencing survival. Explanations for differences in survival between countries may include differences in rates of early diagnosis [
16‐
18], accessibility of medical care other than cancer drugs, different diagnostic intensity and screening approaches, and differences in cancer biology [
14]. However, variation in survival may also be a result of differences in socioeconomic status, general health and lifestyle factors (for example, prevalence of smoking) [
19].
In addition to the absolute level of expenditure, the organization of care within a healthcare system can improve both efficiency and outcomes. For example, the German certification process for cancer centres is intended to promote collaboration of the various disciplines involved in cancer care, including diagnosis, therapy and aftercare. Such an integrated approach can improve the quality of healthcare processes, as well as treatment satisfaction and outcomes for patients [
20]. In addition, there are differences among countries in the availability and funding of cancer screening programmes (for example, Germany is the only country in which skin cancer screening is a standard benefit of public health insurance).
3.2 Differences in Healthcare and Outcomes Within Countries
As well as access at an international level, local socioeconomic and personal factors also influence access to cancer care, and this potential source of inequity is a significant concern for payers within healthcare systems. Across Europe, socioeconomic status, primarily education level and employment status, has been found to have a significant effect on the uptake of cancer screening, even among individuals otherwise using the healthcare system [
21‐
27].
Where screening programmes are available, there is significant variation in the uptake of screening for prostate cancer, breast cancer and cervical cancer [
21,
25]. In each case, education levels and/or employment type appear to be significant determinants of screening rates when individuals were required to initiate screening. At a local level, several studies have identified differences in cancer screening rates according to socioeconomic factors, particularly education [
22‐
24,
26,
27]. We note that in addition to taking steps to improve access to screening, it is important for payers to consider which types of cancer screening are beneficial. For example, population-wide prostate cancer screening is associated with only small reductions in mortality, but over-diagnosis and over-treatment are common, and are associated with treatment-related harms [
28].
Socioeconomic factors also affect the incidence of particular cancer types and the likely stage at diagnosis [
29‐
32]. In addition, both the speed of treatment initiation and the type of treatment received by patients appear to be negatively influenced by low education and deprivation [
33,
34]. For example, a retrospective study in the UK found that patients in affluent areas are more likely to receive treatment for colorectal cancer within 6 months of their first contact with the National Health Service (NHS) than those in deprived areas [
34]. Among those receiving treatment, the most deprived patients were less likely to receive treatment within 1 month than the most affluent group, and more likely to receive treatment only after 4–6 months. The most deprived patients were also less likely to survive for 3 years after diagnosis; this difference appears to reflect the delay in treatment initiation [
34]. It is possible that patients in the most affluent groups are also likely to be better educated than other groups, and may be more able to seek appropriate and high-quality treatment [
34]. The impact of education level on treatment outcomes has been seen in many studies, including a Swedish analysis that identified significantly higher rates of 5-year colorectal cancer survival for highly educated patients than for those with low education [
33]. Similarly, in France a mortality gradient has been described from north to south along the RER B train line which links wealthy central Paris with less affluent suburbs [
35].
3.4 Healthcare Funding Priorities in Europe
Finite resources mean that governments and payers have to decide on the priorities for their individual healthcare systems. Although access to early diagnosis and radiotherapy are the key factors associated with good outcomes in oncology [
13,
16‐
18], public and political attention often focusses on the availability of cancer drugs [
8]. In the EU, cancer drug costs represented 27 % of direct healthcare costs due to cancer in 2009, with considerable variation among countries—in France drug costs were 43 % of the total, compared with 17 % in the Netherlands. This variation may reflect differences in the focus of cancer care spending, although caution is needed when comparing non-drug expenditure due to differences in funding systems and healthcare labour costs [
7]. In 2009, Germany spent slightly less than France on cancer drugs, but had double the cancer care budget overall [
7], suggesting greater expenditure on other aspects of care, such as access to diagnostic and radiotherapy facilities, which have a significant impact on cancer outcomes [
14]. The use of cancer drugs can also vary within a single healthcare system, as shown in a recent analysis in the Netherlands, which found a large variation in the use of particular (expensive) oncology drugs between hospitals [
38].
Reimbursement decisions reached through HTA processes [e.g. cost-effectiveness analysis (CEA)] can sometimes be bypassed by alternative mechanisms. For example, from 2011 to 2016 cancer drugs not available through the NHS in England owing to their high cost could be funded by the CDF. However, while the CDF has benefited some patients with cancer, there was a significant level of criticism of this approach, from the National Institute for Health and Care Excellence (NICE) among others [
39]. In 2015, an analysis of the costs of a quality-adjusted life-year (QALY) in the healthcare system as a whole suggested that the policy decision to fund expensive therapies through the CDF may have had a detrimental effect on care for patients with non-cancer diseases, and that making the CDF’s budget available to the wider NHS would result in much greater improvements in survival and quality of life overall [
40]. In response to the financial pressure on the CDF, a new framework was adopted in July 2016 [
41]. In this framework, new therapies considered by NICE to be potentially cost effective, but around which there is considerable clinical uncertainly, can be recommended for interim funding through the CDF. The CDF will then operate as a managed access fund, with the level of reimbursement for each therapy agreed between NHS England and the manufacturer, and monitoring of patient and population outcomes. After a data collection period of up to 2 years, each therapy will be reappraised by NICE—it is intended that the reappraisal outcome will be a final positive or negative recommendation.
In addition to the criticism of the 2011–2016 CDF described above, it has been suggested that even when new drugs are deemed cost effective by NICE, the threshold used may be too high [
40]. New therapies are typically considered to be cost effective if their incremental cost-effectiveness ratio (ICER), the cost of gaining a fixed improvement in health outcomes compared with existing therapies, is below £30,000 per QALY, with this threshold extended to approximately £50,000 for therapies that can extend life by at least 3 months (typically cancer therapies) [
42]. However, NHS expenditure overall was estimated to generate 1 QALY for every £13,000 spent, suggesting that displacement of existing treatments to fund new, expensive therapies might lead to a net loss of QALYs [
40].
These studies raise important issues, and highlight the need for payers to consider the impact of reimbursement for new therapies beyond the patient population in question. The approach taken by NICE in the UK, and by payers across Europe, is more complex than an attempt to maximise QALYs across the healthcare system, and in many cases includes a willingness to encourage the development of innovative new medicines and devices [
43]. However, while all payers aim to provide patients with access to the most effective treatments, ultimately with curative intent, it is also important that funding can be provided for other aspects of cancer care, including screening, early diagnosis, radiotherapy and palliative care. At a national level, individual countries’ cancer plans take into account all aspects of care. However, it is possible that initiatives such as the 2011–2016 CDF may divert funding towards cancer drugs, leading to reduced expenditure on early diagnosis and radiotherapy, and thereby potentially doing harm to care outcomes overall. As well as allocating resources appropriately within cancer care, payers must also ensure that other disease areas are treated fairly, both in terms of managing major costs to the healthcare system in other areas and of avoiding cuts to budgets in areas that may be less costly and less high profile than oncology, but represent good value for money [
40].
In some countries, for example Germany, the healthcare system does not give priority to particular disease areas. However, funding arrangements such as the CDF exist in a number of other countries, demonstrating a political willingness to overturn technical decisions on reimbursement. For example, the Italian government has established a fund to reimburse the costs of new innovative medicines, as well as a separate fund for new hepatitis C drugs, although with restrictions on the number of patients to be treated. A fund for new hepatitis C treatments has also been set up in the UK. A number of therapies for orphan diseases have also been made available through special earmarked funds. In each case, funding of one disease area over another is potentially problematic—in the UK at least this may undermine the fundamental NHS principle that all patients should be treated equitably. If funding is to be allocated according to the priority assigned to different disease areas, this raises a fundamental problem: how does one decide whether it is worse to have cancer than another disease such as heart failure? In systems that make use of the QALY in order to compare health outcomes across different disease areas (“a QALY is a QALY is a QALY”) [
44] it is not obvious that there should be a willingness to pay more for a QALY in one disease area than in another.
It is theoretically possible that in a particular country, society as a whole may decide that increased funding should be given to cancer therapies, even at the expense of other disease areas. If this is the case, rather than overriding the decision-making processes of HTA bodies, it may be more effective to capture such societal preferences through evaluation approaches such as value-based assessment. In principle, such processes could allow prioritization of the use of healthcare budgets without leading to inequity of access from the perspective of society. More general attempts have also been made to take the views of patients and of society as a whole into account in reimbursement decision making. In the Netherlands, the National Healthcare Institute has recently announced that willingness-to-pay levels for new therapies will be weighted according to disease burden [
45]. In Belgium, a recent “citizens’ lab” initiative has revealed that society seems to favour quality of life and longer-term benefits as major criteria for reimbursement of novel therapeutics [
46]. In addition, the Scottish Medicines Consortium, will accept more uncertainty in the economic case for medicines licensed for the treatment of orphan diseases, and will accept a higher ICER for treatments providing a substantial improvement in life expectancy [
47]. NICE also consider life extension at the end of life, as well as innovation and the non-health objectives of the NHS [
48]. Recently, further proposed ‘modifiers’—including burden of illness and wider societal impact have been investigated in public preference surveys [
49‐
51]. Although proposals to prioritize innovative treatments and those for severe diseases were broadly supported [
49,
51], results were inconsistent with regard to an end-of-life premium [
49‐
51]. Participants in one study reported a substantial preference for health-related quality of life improvement over life extension [
50]—in oncology, this may suggest that greater attention should be paid to the side-effect profiles of therapies, and to the role of palliative care.