Impact of COVID-19 on cancer diagnosis and management in Slovenia – preliminary results

The first confirmed COVID-19 patient in Slovenia was registered on the 4^th of March 2020. The first cases were imported, though soon, secondary, tertiary and quaternary transmissions of the novel virus were detected and on the 12^th of March, the Health Minister following the advice of the National Institute of Public Health (NIPH) declared an epidemic, which meant the activation of the Slovenian Pandemic Plan. Control measures implemented thereafter were strict and introduced rapidly with the aim of mitigating the spread of COVID-19. On the 16^th of March, all schools and educational institutions were closed, all public transport services stopped and all non-essential services shut. Soon after, all gatherings of people were prohibited, with the exception of members of the same household, working from home was encouraged and restrictions on movement of people were put in place limiting movement to within their municipality (lock-down).

Measures concerning the provision of health care services were enacted through the Ordinance on temporary measures in health care to contain and control the COVID-19 epidemic⁴ from the 20^th of March, which stipulated that all non-essential ambulatory visits (those not referred as needing urgent or very fast management) and elective surgery appointments be put on hold. Oncological services were listed as an exception, though all preventive care activities were also put on hold by decree, meaning all three cancer screening programmes (cervical, breast and colorectal cancer) were temporarily stopped. Screening was stopped also in other countries.5, 6 At the Institute of Oncology Ljubljana, the only tertiary comprehensive cancer centre, COVID-19 preventive measures were being continually introduced and adapted starting on 26^th of February. A triage, at first only physical and later also via telephone, was set up to screen patients for COVID-19 symptoms, relative escorts of patients to the hospital and visits of hospitalised patients were not allowed, except for dying patients, while non-essential follow-up visits and surgeries were postponed through May. Despite this, work at the Institute continued almost uninterrupted. Similar measures were taken by oncology departments across Europe7, 8 and many highlighted the need for stricter measures and more testing with the aim of keeping cancer clinics COVID-free given reports of the higher risk COVID-19 poses to people with cancer9 and in order to maintain the provision of oncological services.10, 11

Towards the end of March, the epidemic peaked with daily cases starting to decrease. In the second half of April, easing of control measures in the country started and on the 9^th of May, the government lifted restrictions on provision of healthcare services. Following this, on the 15^th of May Slovenia declared an end to the epidemic. During this time, the Institute of Oncology Ljubljana continued with normal follow-up and surgeries, also introducing working Saturdays to make up for the delay in these services. Furthermore, cancer screening programmes gradually began sending invitations again and were operating close to or at full capacity in June 2020.

In light of severe restrictions in movement of individuals, cancellation of non-essential health care services and ensuing behavioural responses among the population, there might be collateral consequences of COVID-19 related measures for cancer control, despite the Institute of Oncology Ljubljana having retained almost normal functioning. In order to gain a quick and timely understanding of how cancer care in Slovenia has been affected by the COVID-19 epidemic, we carried out an analysis on readily available, up-to-date and reliable data sources.

Figure 1 shows the time trend of monthly referrals during November 2019 – May 2020 where a significant reduction in the number of referrals can be seen in April, with a somewhat smaller reduction in March. The reduction was seen for all types of referrals, though significantly more pronounced for control referrals compared to first referrals, whereas referring for oncological genetic testing and counselling stopped almost completely. Compared to the November – February average, the decrease in April was about 33%, 46% and 85% for first, control and genetic counselling referrals, respectively. In May, the number of all types of referrals started rising again.

Figure 1

The drop in control referrals can most likely be explained as a consequence of the cancer institutes’ policies to defer non-essential control visits. All patients were notified about their deferral and thus there was probably lower demand for control referrals from patients though other reasons could also play a role. Oncological genetic testing and counselling is a preventive service, meaning that doctors were probably less likely to refer patients for this type of care, since the decree on health care stipulated these services are temporarily dis-continued, even though oncological services were clearly listed as an exception. Patients themselves were also less likely to seek services for non-urgent care during lock-down. The reasons behind the drop in first referrals are difficult to determine. It is possible that, compared to pre-epidemic period, during lock-down, people were less likely to seek medical care even if they experienced symptoms of disease. On the other hand, access to primary and secondary level care could have been so disrupted that some patients could not get through to their doctors in order to complain about their issues. Another factor could be that primary level doctors were less likely to refer symptomatic patients for secondary and tertiary diagnostics because these services were not freely available and most of the first symptoms of cancer are rather unspecific. A combination of these factors was likely at play. As a result, we fear that fewer cancers were diagnosed in early stages.

Other countries have reported similar findings. In the UK, a dermatology service found a reduction in referrals for skin cancer of more than 50% in April 2020 compared to April 2019. Additionally, they analysed referrals for other types of cancer in their hospital and found similar reductions for a wide range of cancers, most pronounced for colorectal cancer.12 Also in the UK, others have shown that in the whole country, urgent referrals for cancer from GPs fell by 60% in April.13 In Italy, the referrals for BRCA testing to a genetic laboratory had decreased by about 60%.14

Figure 2 shows the trend in the number of cancer notifications from pathohistological and clinical departments sent to the Cancer Registry from the two main cancer centres, Institute of Oncology Ljubljana and University Medical Centre Maribor. Again, the same pattern can be observed with the largest decrease observed for April and an upward trend in May. Compared to the November 2019 – February 2020 average, the decrease in April was about 43% and 29% for pathohistological and clinical cancer notifications, respectively.

Figure 2

The absolute number of new notifications is not equivalent to the number of newly diagnosed cancers because a cancer case can be reported to the Cancer Registry more than once from different healthcare providers that come into contact with a patient with a cancer, while on the other hand, a small part of notifications turn out not to be malignant cases after additional investigations by the Cancer Registry. Despite this, the relative decline in new notifications can be interpreted as a decrease in newly diagnosed cancers. Roughly, this means in April there were about a third fewer cancers diagnosed in Slovenia compared to the average pre-epidemic period. The reasons for the lower number of cancer notifications are likely related to the drop in referrals. It is not surprising therefore, that the maximum drop in referrals and the maximum drop in newly diagnosed cancers are concurrent. Perhaps the time shift might have been visible if we stratified the data into weeks instead of months, because it takes a week or two for patients who are referred for oncological exam to be diagnosed with cancer. No doubt, another factor for the drop in notifications was the temporary two-month long complete cessation of all three cancer screening programmes, though at the moment it is not possible to quantify what proportion of delayed cancer diagnoses could be attributed to lack of access to cancer screening.

Our results are in line with a study from the Netherlands, where the Netherlands Cancer Registry recorded a decrease in weekly pathological cancer notifications between the end of February and start of April 2020. The decrease was observed for all age groups and all cancer groups but was largest (max. 60%) for skin cancer (excluding basal cell carcinoma), followed by breast cancer (max. 50%). The largest weekly decrease for all cancers excluding skin cancer was approximately a quarter.15 Fewer cancer diagnoses were reported also in Italy. A Pathologic Anatomy Unit in the province of Macerata recorded a decrease in pathohistological diagnoses of cancer during weeks 11-20 (March and April) of 2020 compared to the same period in 2018 and 2019. Unlike in the Netherlands, they did not observe a decrease for malignant melanoma but observed the highest drops for prostate (75%), bladder (66%) and colorectal cancer (62%). Clinically relevant delay was considered only for colorectal cancer. Interestingly, screening for colorectal cancer was disrupted in Italy but was more preserved for breast cancer, which saw a reduction of (only) 25%.16

Administrative data from the Institute of Oncology Ljubljana are presented in Figures 3–5. Compared to the November 2019 – February 2020 average, the decrease in first outpatient visits in April 2020 was 19%, 43% and 20% at the radiotherapy, surgery and medical oncology sectors of the Institute, respectively, whereas for control outpatient visits, these numbers were 53%, 72% and 21%.

Figure 3

Figure 4

Figure 5

Visits to the medical oncology department, where patients receive active chemotherapy treatment, were least affected. The largest drop in both first and control visits can be observed for the surgical department. These results are expected, as all truly elective surgeries at the Institute were postponed, though we cannot say if a part of the decrease in the first visits could also to a minor degree reflect less patients having been diagnosed with cancer and planned for surgery as part of their primary treatment. The decline in first visits to radiotherapy and medical oncology departments was small but could also point to fewer (newly diagnosed) patients being treated. In general, reductions in control outpatient visits were expected due to either postponing non-essential follow-up visits or carrying them out as telehealth visits. For radiotherapy, it might be also indicative of the rationalisation in radiotherapy regimes (such as fewer fractions of radiotherapy).

Outpatient visits to oncological centres must have declined across Europe, though we could not find any already published European study which reported on the number of cancer outpatient visits. A report from the US shows that oncology outpatient visits had fallen by as much as 47% in April 2020.17

Regarding diagnostic imaging, in April 2020 compared to the November 2019 – February 2020 average, there were also significant reductions in X-rays (48%), mammograms (76%) and ultrasounds (42%) performed at the Institute. This could again point to fewer patients being in the diagnostic process though there were changes in the functioning of the Institute that could also have contributed to this result. The numbers of CT and MRI scans were not affected. The reduction in diagnostic imaging was thus most pronounced for mammography, which is only in part linked to the suspension of breast cancer screening, as mammograms within the screening programme are tallied separately.