Health-care Development

Radiotherapy capacity in European countries: an analysis of

the Directory of Radiotherapy Centres (DIRAC) database
Eduardo Rosenblatt, Joanna Izewska, Yavuz Anacak, Yaroslav Pynda, Pierre Scalliet, Mathieu Boniol, Philippe Autier

Radiotherapy is used for cure or palliation in around half of patients with cancer. We analysed data on radiotherapy Lancet Oncol 2013; 14: e79–86
equipment in 33 European countries registered in the Directory of Radiotherapy Centres (DIRAC) database, managed Published Online
by the International Atomic Energy Agency. As of July, 2012, Europe had 1286 active radiotherapy centres. The average January 24, 2013
http://dx.doi.org/10.1016/
number of teletherapy machines per radiotherapy centre ranged from 1·2 to 7·0 in different countries. Nordic
S1470-2045(12)70556-9
countries, the UK, the Netherlands, and Slovenia all have large centres with four to ten teletherapy machines. Most
See Health-care Research
western and southern European countries have several small centres with one or two machines, with few larger Lancet Oncol 2006; 7: 584–95
centres. The fragmentation in radiotherapy services that prevails in many European countries might affect the International Atomic Energy
economic burden of radiotherapy and its quality. Eastern and southeastern European countries need to expand and Agency, Vienna, Austria
modernise their radiotherapy equipment. (E Rosenblatt MD, J Izewska PhD,
Y Anacak MD, Y Pynda MSc);
Department of Radiotherapy,
Introduction This Health-care Development article provides an Catholic University of Louvain,
Cancer is a major public health problem in Europe. Every inventory of radiotherapy capacity in Europe, based on Brussels, Belgium
year, 3∙2 million Europeans are diagnosed with the an analysis of the European section of the Directory of (P Scalliet MD); and
Radiotherapy Centres (DIRAC).8 The European Network International Prevention
disease;1 this number is expected to increase, mainly
Research Institute, Lyon,
because of population ageing and, particularly for some for Information on Cancer (EUNICE) project, which France (M Boniol PhD,
cancers (eg, breast and colorectal cancer), because of the analysed the DIRAC database, was done from 2008–12 by P Autier MD)
increasing prevalence of obesity.2 Radiotherapy is an the International Atomic Energy Agency (IAEA), the Correspondence to:
effective and widespread method for treating cancer with International Agency for Research on Cancer (IARC), Dr Eduardo Rosenblatt, Applied
curative intent. It is also an effective method of palliation and the International Prevention Research Institute Radiation Biology and
Radiotherapy Section,
in patients with advanced disease. A steady rise in the (iPRI). This report summarises the radiotherapy International Atomic Energy
number of patients with cancer is boosting demand for component of the EUNICE project, and aims to provide Agency, Wagramer Strasse 5,
radiotherapy services in Europe. Roughly 45–55% of basic indicators for planning radiotherapy infrastructure Vienna A-1400, Austria
patients with cancer require radiotherapy at some point, and manpower at national and regional levels. e.rosenblatt@iaea.org

and about 20–25% will have more than one course of

treatment.3,4 An assessment of radiotherapy in cancer
care in Sweden estimated that, of patients with cancer
who are cured, 49% are cured by surgery, 40% by
radiotherapy alone or in combination with other
treatment methods, and 11% by chemotherapy alone or
in combination with other methods.5,6
In view of the importance and cost of radiotherapy and
the infrastructure required, the European Society for
Radiotherapy and Oncology (ESTRO) organised the
QUAntification Of  Radiation  Therapy Infrastructure
And Staffing Needs (QUARTS) project in 2003, which
aimed to establish objective and quantifiable criteria for
long-term capacity planning of radiotherapy services in
Europe.7 QUARTS gave an overview of national
guidelines and proposed new evidence-based recom-
mendations for planning radiotherapy infrastructure and
staff requirements. The project showed that the
availability and need for radiotherapy services varies
greatly from one European country to another; however,
the data assembled for comparing existing and required
equipment included only 13 countries and were of
inconsistent quality, many being out-of-date or
incomplete. The project’s final report underlined the
need for a prospective and continuously updated
inventory of radiotherapy facilities and cost in Europe, in
parallel with a continuous assessment of how treatment
needs are being fulfilled.
Data source and collection
For the EUNICE project, Europe encompassed
33 countries: the 27 members of the European Union,
three candidate countries for membership (Croatia,
Macedonia, and Turkey), and three countries that are part
of the European Free Trade Association (Iceland, Norway,
and Switzerland). To categorise these countries into
broader regions, we used the geographical subregions
proposed by the UN Statistics Division, namely western,
northern, and eastern Europe.9 We grouped together
Croatia, Macedonia, Greece, Cyprus, and Turkey as a For more on the QUARTS
fourth region—southeastern Europe. project see http://www.estro.
org/projects/Pages/QUARTS.
Radiotherapy data were collected using DIRAC,8 a aspx
registry of worldwide radiotherapy facilities maintained
by the Division of Human Health of the IAEA since 1959.
The current electronic version of DIRAC contains data
collected since 1995 pertaining to radiotherapy machines,
sources and devices used in brachytherapy, equipment
for dosimetry, staffing, dose calculation, and quality
assurance instruments. Worldwide data stored in DIRAC
are collected from 137 countries, representing
7620 radiotherapy centres and 12 249 teletherapy
machines.
DIRAC is continuously updated by direct online access
by centres, and with data from the IAEA and WHO
thermoluminescence dosimetry postal audit programme,
the National Secondary Standards Dosimetry Laboratories,

www.thelancet.com/oncology Vol 14 February 2013 e79

 Health-care Development

Total Number of Number of Mean number of Megavoltage Number of Number of cobalt-60

population10 radiotherapy megavoltage megavoltage teletherapy linear megavoltage
centres teletherapy teletherapy machines machines per accelerators teletherapy machines
machines per centre million people
Austria 8 443 018 16 43 2·7 5·1 42 1
Belgium 11 041 266 36 96 2·7 8·7 92 4
Bulgaria 7 327 224 13 15 1·2 2·0 5 10
Croatia 4 412 137 7 18 2·6 4·1 16 2
Cyprus 862 011 1 3 3·0 3·5 3 0
Czech Republic 10 504 203 34 59 1·7 5·6 45 14
Denmark 5 580 516 10 54 5·4 9·7 53 1
Estonia 1 339 662 2 4 2·0 3·0 4 0
Finland 5 401 267 13 45 3·5 8·3 45 0
France 65 397 912 177 426 2·4 6·5 416 10
Germany 81 843 809 289 529 1·8 6·5 508 21
Greece 11 290 785 27 48 1·8 4·3 37 11
Hungary 9 962 000 13 38 2·9 3·8 27 11
Iceland 319 575 1 2 2·0 6·3 2 0
Ireland 4 495 351 10 26 2·6 5·8 25 1
Italy 60 850 782 172 396 2·3 6·5 381 15
Latvia 2 042 371 4 11 2·8 5·4 9 2
Lithuania 3 199 771 5 11 2·2 3·4 9 2
Luxembourg 524 853 1 3 3·0 5·7 3 0
Macedonia 2 059 794 1 3 3·0 1·5 2 1
Malta 420 085 1 2 2·0 4·8 1 1
Norway 4 985 870 10 38 3·8 7·6 37 1
Poland 38 208 618 29 107 3·7 2·8 101 6
Portugal 10 541 840 21 48 2·3 4·6 46 2
Romania 21 355 849 19 28 1·5 1·3 12 16
Slovakia 5 404 322 14 26 1·9 4·8 16 10
Slovenia 2 055 496 1 7 7·0 3·4 6 1
Spain 46 196 277 117 250 2·1 5·4 216 34
Sweden 9 482 855 19 78 4·1 8·2 78 0
Switzerland 7 952 555 30 75 2·5 9·4 70 5
Netherlands 16 730 348 22 127 5·8 7·6 127 0
Turkey 74 724 269 95 201 2·1 2·7 143 58
United Kingdom 62 435 709 76 340 4·5 5·4 335 5
Total 597 392 400 1286 3157 2·5 5·3 2912 245

Table 1: Teletherapy centres and equipment in 33 European countries (July, 2012)

and data gathered during expert visits and planned update
efforts focused on specific countries or regions. From
2008 to 2012, the European component of the DIRAC
database underwent a systematic update and analysis. The
results presented here reflect the situation as of July, 2012.
Definition of a radiotherapy centre
DIRAC considers a radiotherapy centre to be a health
facility with radiotherapy equipment (radionuclide
teletherapy unit, clinical accelerator, radiosurgery unit,
heavy particle accelerator, or brachytherapy afterloader)
used for treating patients with cancer. Facilities are not
registered if they meet any of the following criteria: those
set up to treat non-malignant diseases only, those
operating an orthovoltage x-ray machine only, radiation
research centres where no active treatment is given to
patients, veterinary care centres, or urology and
gynaecology centres that provide manual brachytherapy
alone. When a hospital, centre, institute, or a health-care
group has more than one facility in separate postal
addresses, each is registered as an individual radiotherapy
centre. Centres recorded in DIRAC as inactive due to
discontinuation of radiotherapy services were excluded
from our analysis.
Equipment
In DIRAC, radiotherapy equipment is recorded in
separate sections for external-beam therapy (teletherapy)
machines, brachytherapy equipment, imaging devices,
computerised planning systems, and dosimetry

e80 www.thelancet.com/oncology Vol 14 February 2013


instruments. Teletherapy and brachytherapy machines
were the focus of this study.
Teletherapy machines are classified by DIRAC as
radionuclide units (cobalt-60 or caesium-137 machines),
clinical accelerators and linear accelerators that generate
high energy x-rays or electron beams, or both, and heavy-
particle accelerators that generate proton and heavy-ion
beams. New generation machines such as helical tomo-
therapy, robotic radiotherapy, high-energy stereotactic
x-ray radiosurgery machines, and mobile electron
accelerators for intraoperative radiotherapy are counted
as clinical accelerators. Heavy-particle accelerators are
counted only if routinely used for patient care.
Radionuclide stereotactic machines are counted as radio-
nuclide therapy units. Kilovoltage x-ray generators
(superficial and orthovoltage) are counted separately.
Brachytherapy equipment is classified as using manual
delivery (wires, needles, or seeds) or afterloaders (low
dose rate [LDR], high dose rate [HDR], or pulsed dose
rate [PDR]), and the type of isotope is recorded (eg,
iridium-192, cobalt-60, caesium-137, or iodine-125). The
DIRAC database does not have an accurate record of the
number of LDR sources for manual loading; therefore,
only data on LDR afterloaders are presented here.
Devices for radiotherapy imaging are classified as
conventional simulators and CT simulators. The data on
CT simulators are less reliable than for conventional
simulators, because CT machines in radiology depart-
ments are also used for general imaging and no
information is available on time-sharing between
radiotherapy planning and other activities. Furthermore,
analysis of data related to radiotherapy planning systems
proved difficult because of the rapid turnover of computer
hardware and software, and is therefore not presented
here.
Computation of indicators
We used population, epidemiological, and economic data
to compute basic indicators that allowed comparisons of
radiotherapy equipment between countries.
The number of teletherapy machines, linear
accelerators, and radionuclide units per radiotherapy
centre is an indicator of the degree of fragmentation of
radiotherapy services in a country. A high number
represents low fragmentation, with most teletherapy
machines located in a few centres. A low number
represents a high level of fragmentation, with many
centres operating only one or two teletherapy machines.
The number of teletherapy machines, linear accelerators,
and radionuclide units per million people is another key
indicator. To calculate this number, we used 2012
population data extracted from the statistical office of the
European Commission (Eurostat) database.10 To interpret
this indicator, we used two benchmarks from the
QUARTS project;6 we considered the European average of
5·9 machines per million inhabitants to be the desirable
density for radiotherapy equipment, and a density of four
www.thelancet.com/oncology Vol 14 February 2013
Health-care Development
Number
Medical accelerators 3072
Linear accelerators 2912
Electron-beam units for intraoperative radiotherapy 46
X-ray radiosurgery units 40
Helical tomotherapy units 61
Charged-particle accelerators 13
Radionuclide units 277
Cobalt-60 units 245
Caesium-137 units 12
Radiosurgery units 20
Kilovoltage machines 399
Brachytherapy equipment 945
Afterloaders 712
Manual systems 233
Table 2: Radiotherapy equipment in the 33 European countries included
in this study
machines per million people to be the minimum
requirement—this being the estimated density of
radiotherapy machines in Cyprus, the European country
that QUARTS estimated as having the lowest requirements
according to its cancer epidemiological profile.
A third indicator calculated was the number of patients
who received cancer treatment per teletherapy machine,
using cancer incidence in 2008 for each of the 33 European
countries.1 The incidence was probably underestimated
for Turkey because of suboptimum quality of data on
cancer cases and death;11 we therefore estimated an
incidence based on figures for neighbouring countries
(Malta, Greece, Bulgaria, Macedonia, and Romania). We
used as a benchmark the estimate that 50% of patients
with cancer will need radiotherapy treatment, which we
increased by a factor of 25% to account for retreatment.7
Therefore, we multiplied the estimated number of new
cancer cases in 2008 by a correction factor of 62·5% to
obtain the country-specific number of patients with cancer
who would have required radiotherapy in 2008. To
estimate the number of machines needed, we used the
QUARTS benchmark of 450 patients per year per machine
(whether for cancer treatment or retreatment).7
Finally, we used World Bank data on country-specific
gross national income (GNI) per head12 to correlate the
wealth of nations against the country-specific number of
patients with cancer who received radiotherapy in 2008.
To explore the relationship between macroeconomic data
and supply in radiotherapy machines, we fitted a log-
linear regression between the number of patients with
cancer per teletherapy machine and the GNI per head in
US$. We then computed the corresponding Pearson
correlation coefficient.
Infrastructure
In July, 2012, Europe had 1286 active radiotherapy centres
registered in the DIRAC database (table 1). Germany had
e81
 Health-care Development

Number of Number of Number of Number of Expected Expected— % unmet

megavoltage patients with treatments and treatments and number of observed number needs‡
teletherapy cancer1 retreatments retreatments per machines† of machines
machines needed* machine, per year
Austria 43 38 790 24 244 564 54 11 20%
Belgium 96 59 780 37 363 389 83 –13 –16%
Bulgaria 15 30 110 18 819 1255 42 27 64%
Croatia 18 22 530 14 081 782 31 13 42%
Cyprus 3 2210 1381 460 3 0 0%
Czech Republic 59 53 470 33 419 566 74 15 20%
Denmark 54 32 110 20 069 372 45 –9 –20%
Estonia 4 5460 3413 853 8 4 50%
Finland 45 25 580 15 988 355 36 –9 –25%
France 426 332 690 207 931 488 462 36 8%
Germany 529 479 110 299 444 566 665 136 20%
Greece 48 38 230 23 894 498 53 5 9%
Hungary 38 50 250 31 406 826 70 32 46%
Iceland 2 1260 788 394 2 0 0%
Ireland 26 19 300 12 063 464 27 1 4%
Italy 396 339 800 212 375 536 472 76 16%
Latvia 11 9400 5875 534 13 2 15%
Lithuania 11 14 090 8806 801 20 9 45%
Luxembourg 3 2340 1463 488 3 0 0%
Macedonia 3 6230 3894 1298 9 6 67%
Malta 2 1460 913 456 2 0 0%
Norway 38 24 630 15 394 405 34 –4 –12%
Poland 107 141 200 88 250 825 196 89 45%
Portugal 48 42 440 26 525 553 59 11 19%
Romania 28 70 920 44 325 1583 99 71 72%
Slovakia 26 21 090 13 181 507 29 3 10%
Slovenia 7 9600 6000 857 13 6 46%
Spain 250 199 050 124 406 498 276 26 9%
Sweden 78 46 060 28 788 369 64 –14 –22%
Switzerland 75 36 780 22 988 307 51 –24 –47%
Netherlands 127 82 790 51 744 407 115 –12 –10%
Turkey§ 201 257 293 160 808 800 357 156 44%
United Kingdom 340 310 280 193 925 570 431 91 21%
Total 3157 2 806 333 1 753 958 556¶ 3898 741 19%

*Based on an estimate that 62·5% of patients with cancer will need radiotherapy treatment or retreatment. †Based on a maximum of 450 patients treated or
retreated per machine per year.5 ‡(Expected − observed number of machines) / expected number. §Number of patients with cancer estimated from numbers in
Macedonia, Malta, Greece, Romania, and Bulgaria. ¶Average number per country.

Table 3: Existing and required megavoltage radiotherapy machines in European countries

the highest number (n=289), followed by France (n=177),
Italy (n=172), Spain (n=117), Turkey (n=95), and the UK
(n=76); the remaining countries had between one and
36 centres each. These centres operated a total of
3157 high-energy linear accelators or cobalt-60 teletherapy
machines. In addition, they operated electron-beam
units for intraoperative radiotherapy, x-ray radiosurgery
units, helical tomotherapy machines, proton or heavy-
particle accelerators, and other radionuclide teletherapy
units, such as caesium-137 machines and radiosurgery
units using cobalt-60 sources (table 2). There were also
399 registered kilovoltage x-ray machines in the database.
The mean number of teletherapy machines per
radiotherapy centre was 2·5 (table 1). Slovenia had the
highest average number of machines per centre (7·0),
and Bulgaria had the lowest (1·2). Five countries had an
average of four or more machines per centre, whereas
six countries had an average of less than two, meaning
that in these countries, many radiotherapy centres
operate with only one teletherapy machine. In general,
the Nordic countries, the UK, the Netherlands, and
Slovenia had a more centralised radiotherapy
infrastructure with a high number of machines
per centre.

e82 www.thelancet.com/oncology Vol 14 February 2013


There was an average of 5·3 teletherapy machines per
million people in Europe, with large variations across
countries (table 1). More than eight machines per million
people were available in Belgium, the Nordic countries
(except Iceland), and Switzerland, whereas Macedonia
and Romania had fewer than two machines per million
people. 17 countries had more than 5·1 machines per
million people, and ten had fewer than four machines
per million people.
In the 33 countries studied, linear accelerators
accounted for 92% of all teletherapy machines. Although
radionuclide teletherapy units are rare or non-existent in
most countries in northern and western Europe, they
represent about 31% of all teletherapy machines in the
group including Bulgaria, Hungary, Croatia, Czech
Republic, Malta, Slovakia, Romania, Macedonia, and
Turkey. However, in general, radionuclide-based tele-
therapy units are being replaced with linear accelerators.
DIRAC registered nine teletherapy machines that use
caesium-137: eight in the Czech Republic and one in
Hungary. Of these machines, only four in the Czech
Republic are operational and are mostly used for
treatment of inflammatory musculoskeletal disorders.
12 centres were providing treatment to patients using
proton or heavy-ion particle beams. These centres are
located in Italy, France, UK, Germany, Poland, Russia,
Sweden, and Switzerland. 14 particle-therapy facilities
were in the planning stage or under construction in
Austria, the Czech Republic, Germany, Italy, Switzerland,
Sweden, and Slovakia.13
Population, patients with cancer, and economic
indicators
In 2008, about 2∙806 million people living in the
33 countries that were part of this study were diagnosed
with cancer,1 about half of whom required radiotherapy.
The actual and required availability of teletherapy
machines in 2012, according to the estimated cancer
incidence in 2008 and the QUARTS benchmark for
supply of teletherapy machines, is shown in table 3. We
estimated that the number of patients with cancer per
radiotherapy machine per year ranged from 307 in
Switzerland to 1583 in Romania, with an average of
556 patients per machine. Unmet needs were defined as
the percentage of patients with cancer in each country
who would not have access to radiotherapy, based on the
QUARTS criteria of 450 patients per machine per year.
A negative percentage for unmet needs means that
more machines are available than the number defined
by the QUARTS benchmark. Our results show that the
Nordic countries, Belgium, the Netherlands, and
Switzerland are well-equipped with teletherapy
machines. There is an apparent shortage of machines in
Germany, Italy, Austria, Portugal, and the UK, and most
countries in eastern and southeastern Europe (in
particular Bulgaria, Macedonia, and Romania) are
largely underequipped.
www.thelancet.com/oncology Vol 14 February 2013
Health-care Development
1400
Number of patients with cancer per teletherapy machine
1200
1000
800
600
400
200
0
0 10 000 20 000 30 000 40 000 50 000 60 000 70 000 80 000 90 000
GNI per head
Figure: Log-linear regression between the number of patients per teletherapy machine and gross national
income (GNI) per head (r=–0·80; p<0·0001)
We found a negative correlation between GNI per
head and the number of the patients per radiotherapy
machine per year (figure). High-income countries (GNI
per head of US$12 476 or more in 2011) have fewer
patients per year served by each machine than countries
whose GNI per head is lower. At the higher-end income
of $25 000 GNI per head or more, the number of
patients treated achieves the level of 400–450 patients
per machine per year.
Brachytherapy
945 brachytherapy systems were identified; these
included 712 remote afterloading machines and
233 manual brachytherapy systems (table 2).
546 machines used HDR sources, 328 used LDR, 31 used
medium dose rate, and 40 used PDR.
Overall, 657 centres in Europe had brachytherapy
facilities, representing about 52% of all radiotherapy
centres. 562 centres had remote afterloading systems
and the remaining 125 centres used manual delivery of
brachytherapy. Thus, only 43·7% of all European centres
had remote afterloading capability. For individual
countries, the percentage of centres with brachytherapy
systems varied from less than 40% of centres in France,
Italy, and Spain, to 60% or more in northern, eastern,
and southeastern European countries (data not shown).
Countries with smaller populations typically used only
remote afterloading systems (Croatia, Cyprus, Estonia,
Macedonia, Iceland, Latvia, Luxembourg, Malta, Slovakia,
and Slovenia); some countries with a strong tradition of
LDR brachytherapy had a large proportion of manual
brachytherapy systems (eg, France, Belgium, Bulgaria,
and Spain).
Iridium-192 was the most common radioactive source
used for brachytherapy in Europe (table 4). This was
followed by caesium-137, iodine-125, and cobalt-60.
e83
 Health-care Development
High dose Low dose Medium Pulsed Total
rate rate dose rate dose rate
Iridium-192 376 79 10 37
Caesium-137 - 115 18 - 133
Iodine-125 - 77 - -
Cobalt-60 12 - 2 - 14
No data 158 57 1 3
Total 546 328 31 40
Table 4: Number of brachytherapy machines, according to isotope and
operating mode, in the 33 European countries included in this study
In three countries (Belgium, Czech Republic, and Spain),
a few institutions kept a stock of radium-226 brachytherapy
sources, but these were no longer used for patient care.
There was not a marked east–west difference for quality of
brachytherapy equipment, possibly due to the relatively
lower cost of this type of equipment. For 219 systems,
entries in DIRAC did not provide details of the
radionuclide source.
Conclusions
This first comprehensive analysis of DIRAC data for
Europe shows a substantial disparity in the availability
and organisation of radiotherapy services between
countries.
For countries with a well-developed radiotherapy
infrastructure, we identified two main approaches to
organisation of radiotherapy services. In Nordic
countries, the UK, and the Netherlands, radiotherapy
services are usually centralised in a few large cancer-care
centres that provide all types of radiotherapy techniques,
with four to ten machines per centre, and all necessary
equipment and personnel. In most other European
countries, radiotherapy facilities vary in size and capacity,
with many small facilities fitted with one or two machines
and treatment focused on specific cancer types. If we use
a benchmark of four machines per radiotherapy centre
as the threshold below which fragmentation of
radiotherapy facilities occurs, then fragmentation exists
in 28 of the 33 countries included in this analysis, and
hyperfragmentation (ie, fewer than two machines per
centre) exists in six countries. A study from Belgium
calculated a lower threshold of 1000 patients per
radiotherapy centre (ie, a minimum of two machines per
centre) to optimise cost-effectiveness.14 Fragmentation of
radiotherapy services might have an effect on the
economic burden of radiotherapy and on its quality.
Some European countries spend more on resources and
have more equipment and personnel than others.
However, because of fragmentation, greater resource
allocation does not necessarily translate into benefits for
patients; countries that devote fewer resources to
radiotherapy might do so in a way that fulfils needs better
than suggested by the statistics in table 3. The question
of whether differences in equipment and organisation
e84
have an effect on cancer outcomes was beyond the scope
of this study, but further research is needed into how to
optimise the efficiency of radiotherapy services.
502
The type of radiotherapy equipment used differs
between regions of Europe. Apart from radionuclide
77
units specific to radiosurgery, radionuclide teletherapy is
almost non-existent in northern and western Europe. In
219
eastern and southeastern Europe, cobalt-60 machines
945
still represent a sizeable part of the radiotherapy
armamentarium. In some eastern European countries,
teletherapy machines are out-dated and need to be
replaced by modern equipment; these countries have the
greatest need to broaden the coverage of their
radiotherapy services. Eastern and southeastern
European countries need to expand and modernise their
radiotherapy equipment. How these needs will be met is
yet to be determined, since our study found that, in
Europe, less wealthy countries allocate fewer resources to
radiotherapy. Similar relationships between GNI per
head and level of radiotherapy infrastructure have been
reported,15 in particular for African,16 Asian,17 and Latin
American18 countries.
Half of all radiotherapy centres in Europe provide
brachytherapy services, and around three-quarters of
brachytherapy systems use afterloading delivery. DIRAC
data probably underestimate the use of brachytherapy,
because many gynaecology and urology centres provide
only manual brachytherapy for cervical and prostate
cancers, and these centres are not recorded in the
database. HDR is the most common method of delivering
brachytherapy, although LDR is still widely used. Many
centres have capacity for both techniques, since their use
depends on cancer site. A recent survey on brachytherapy
in Europe (which did not include Germany, Turkey,
Greece, Ireland, and Slovakia)19 reported increased use of
HDR and PDR techniques between 2002 and 2007, and a
decline in LDR brachytherapy. Radium-226 sources have
been almost completely replaced in Europe by newer
iridium-192, caesium-137, coblalt-60, and iodine-125
brachytherapy sources.
This study has several limitations. The benchmarks
we used were crude, but proposing new benchmarks
was beyond the scope of this report. The benchmarks
we used do not take into account the epidemiological
cancer profiles specific to each of the 33 countries, or
the possible existence of new national guidelines for
radiotherapy. In this respect, estimating radiotherapy
equipment requirements on the basis of population
(table 1) or on cancer cases (table 3) might not be
consistent, since incidence rates are higher in older
populations. Although the QUARTS project7 was a start,
there is still a need in most European countries for
benchmarking on the basis of incidence and treatment
methods specific to different types of cancer. Since
many European countries do not have nationwide
cancer registration, benchmarks based on cancer
burden depend on estimates of incident cancer cases.
www.thelancet.com/oncology Vol 14 February 2013

The demand for radiotherapy depends heavily on the
incidence of breast and prostate cancers, which is itself
affected by the extent of screening programmes. Also,
the proportion of patients with breast or prostate cancer
who are receiving radiotherapy can vary from country to
country. Another approach would be to set benchmarks
based on the proportion of the population older than
50 years, since most cancers occur in older individuals.
Although our report provides a quantitative overview of
radiotherapy infrastructure in Europe, it does not address
quality issues. How radiotherapy capacity is being used
needs to be explored. The quality of the infrastructure
(age and condition of equipment, radiotherapy
techniques in routine use, and quality assurance
programmes) and manpower (training and certification)
should be assessed separately. Although DIRAC contains
some equipment quality data (eg, on dosimetry and
quality assurance equipment), information on quality
assurance programmes falls outside the scope of the
DIRAC datasets; therefore, it would not be appropriate to
draw conclusions about the quality of radiotherapy
services based on DIRAC data.
Data from DIRAC on manpower were not at present
considered sufficiently accurate for our purposes. The
radiation oncology community in Europe includes about
6000 radiation oncologists, more than 3000 medical
physicists, and 10 000 radiotherapy technologists.
Collecting and interpreting manpower data is difficult
because the definitions of radiation oncologist, medical
physicist, and radiation technicians differ between
European countries, and sometimes also within
countries. For example, in the UK and Macedonia,
radiation oncologists are not differentiated from medical
oncologists (also known as clinical oncologists).
Furthermore, technologists often work part-time, mainly
in small radiotherapy centres, which complicates
manpower statistics. Tasks done by medical physicists
differ from country to country and from centre to centre,
with varying levels of specialisation. In some areas,
medical physicists perform all tasks related to radiation
medicine (eg, radiotherapy, nuclear medicine, and
diagnostic radiology), whereas in other countries or
centres, they are dedicated to one field of radiation
medicine.
Despite radiotherapy being more cost effective than
surgery and chemotherapy for treating cancer,5,6 the
building and running of a radiotherapy centre requires
substantial financial and technical investment. This
includes construction of treatment rooms and purchase
of treatment machines, conventional or CT simulators,
computer hardware and software, dosimetry equipment,
and patient positioning devices. Training for radiation
oncologists, medical physicists, and radiotherapy
technicians is necessary. The initial cost of a basic centre
equipped with a single radiotherapy machine is at least
€5 million, a figure which can climb to tens of millions
depending on the size of the centre and the radiotherapy
www.thelancet.com/oncology Vol 14 February 2013
Health-care Development
Search strategy and selection criteria
We searched PubMed, ISI Web of Science, Scopus, and Google
Scholar, with the terms “radiotherapy” or “brachytherapy”
combined with “economics” or “health care organization” or
“staffing” for papers published in English between January,
1995, and July, 2012. We searched references cited in reports
and websites dedicated to projects on radiotherapy having a
European agenda. We also searched documents and technical
reports stored or produced by the IAEA. We compiled the
literature and selected reports relevant to the objective and
scope of the DIRAC project, with special attention to Europe.
technology used. Running costs, maintenance contracts,
replacement of radioactive sources, and upgrades of
software and hardware are all recurring costs. Because of
the capital investment required for radiotherapy
equipment and its long depreciation period (a linear
accelerator might work for 10 years or more), changes in
radiotherapy needs must be predicted far in advance.
In this respect, the DIRAC database is very useful for
planning radiotherapy infrastructure and manpower.
However, to take full advantage of its potential, the
DIRAC database needs complementary data, such as
epidemiological trends in number and types of cancers,
including time projections for demographic and risk-
factor changes. Other factors need to be taken into
account, such as variation in cancer treatment and
differences in treatment guidelines between and within
nations. Finally, the service life of existing radiotherapy
equipment is a crucial parameter, since the cost of
replacement might divert resources for building new
infrastructure.
Some countries have developed standards for running
radiotherapy services;20,21 however, there is no common
policy within Europe regarding installation and operation
of radiotherapy centres.3,7 The IAEA published guidelines
and requirements for planning national radiotherapy
services for its member states,22,23 including European
countries. Several reports and recommendations on the
minimum requirements for radiotherapy infrastructure
and staffing are available, but few guidelines exist on the
optimum way to organise and run radiotherapy services.
The IAEA has initiated a process for defining staffing
requirements at radiation medicine centres based on the
number of patients served, and on an activity-based
model that takes into account the complexity of modern
equipment and techniques.
In conclusion, we have documented the current
radiotherapy capacity in Europe and identified shortfalls
in infrastructure and manpower, particularly in some
European countries. We hope that governments,
European Union bodies, and international organisations
will take advantage of the DIRAC initiative and use it as a
guideline for setting standards and planning investment
into radiotherapy services.
e85
 Health-care Development
Contributors
ER and JI drafted the article. YA and YP managed the DIRAC database.
MB did statistical analyses. PS provided clinical expertise. PA finalised the
article. ER, JI, PS, and PA contributed to the interpretation of findings.
Conflicts of interest
PA and MB are former staff members of the International Agency for
Research on Cancer. All other authors declared no conflicts of interest.
References
1 Ferlay J, Parkin DM, Steliarova-Foucher E. Estimates of cancer
incidence and mortality in Europe in 2008. Eur J Cancer 2010;
46: 765–81.
2 Wang YC, McPherson K, Marsh T, Gortmaker SL, Brown M. Health
and economic burden of the projected obesity trends in the USA
and the UK. Lancet 2011; 378: 815–25.
3 Slotman BJ, Cottier B, Bentzen SM, Heeren G, Lievens Y,
van den Bogaert W. Overview of national guidelines for
infrastructure and staffing of radiotherapy. ESTRO-QUARTS: work
package 1. Radiother Oncol 2005; 75: 349–54.
4 Delaney G, Jacob S, Featherstone C, Barton M. The role of
radiotherapy in cancer treatment: estimating optimal utilization
from a review of evidence-based clinical guidelines. Cancer 2005;
104: 1129–37.
5 Frödin JE, Jonsson E, Möller T, Werkö L. Radiotherapy in Sweden:
a study of present use in relation to the literature and future trends.
Acta Oncol 1996; 35: 967–79.
6 The Swedish Council on Technology Assessment in Health Care
(SBU). Radiotherapy for cancer. Acta Oncol 1996; 35 (suppl 6): 9–23.
7 Bentzen SM, Heeren G, Cottier B, et al. Towards evidence-based
guidelines for radiotherapy infrastructure and staffing needs in
Europe: the ESTRO QUARTS project. Radiother Oncol 2005;
75: 355–65.
8 International Atomic Energy Agency. DIRAC (DIrectory of
RAdiotherapy Centres). http://www-naweb.iaea.org/nahu/dirac/
default.asp (accessed Sept 11, 2012).
9 United Nations Statistics Division. Composition of macro
geographical (continental) regions, geographical sub-regions, and
selected economic and other groupings. http://unstats.un.org/
unsd/methods/m49/m49regin.htm (accessed Sept 11, 2012).
10 European Commission. EUROSTAT database. http://epp.eurostat.
ec.europa.eu/portal/page/portal/eurostat/home (accessed Sept 11,
2012).
e86
11 International Agency for Research on Cancer. Globocan 2008.
http://globocan.iarc.fr/ (accessed Sept 11, 2012).
12 The World Bank. Gross national income per capita 2011, Atlas
method and PPP. http://databank.worldbank.org/databank/
download/GNIPC.pdf (accessed Sept 10, 2012).
13 Particle Therapy Cooperative Group. Particle therapy facilities in a
planning stage or under construction. http://ptcog.web.psi.ch/
newptcentres.html (accessed Sept 11, 2012).
14 Lievens Y, van den Bogaert W, Kesteloot K. Activity-based costing:
a practical model for cost calculation in radiotherapy.
Int J Radiat Oncol Biol Phys 2003; 57: 522–35.
15 Levin V, Tatsuzaki H. Radiotherapy services in countries in
transition: gross national income per capita as a significant factor.
Radiother Oncol 2002; 63: 147–50.
16 Abdel-Wahab M, Rosenblatt E, van der Merwe D, Pynda Y,
Meghzifene A. Changes in the availability of radiation oncology
services in Africa; a report from the IAEA. http://m.rsna.org/
rsna2011/program/event_display.cfm?em_id=11030300 (accessed
Dec 18, 2012).
17 Tatsuzaki H, Levin CV. Quantitative status of resources for radiation
therapy in Asia and Pacific region. Radiother Oncol 2001; 60: 81–89.
18 Zubizarreta EH, Poitevin A, Levin CV. Overview of radiotherapy
resources in Latin America: a survey by the International Atomic
Energy Agency (IAEA). Radiother Oncol 2004; 73: 97–100.
19 Guedea F, Venselaar J, Hoskin P, et al. Patterns of care for
brachytherapy in Europe: updated results. Radiother Oncol 2010;
97: 514–20.
20 Esco R, Palacios A, Pardo J, et al. Infrastructure of radiotherapy in
Spain: a minimal standard of radiotherapy resources.
Int J Radiat Oncol Biol Phys 2003; 56: 319–27.
21 Ruggieri-Pignon S, Pignon T, Marty M, Rodde-Dunet MH,
Destembert B, Fritsch B. Infrastructure of radiation oncology in
France: a large survey of evolution of external beam radiotherapy
practice. Int J Radiat Oncol Biol Phys 2005; 61: 507–16.
22 International Atomic Energy Agency. Setting-up a radiotherapy
programme: clinical, medical physics, radiation protection and
safety aspects. http://cancer.iaea.org/documents/Ref5-
TecDoc_1040_Design_RT_proj.pdf.(accessed Dec 18, 2012).
23 International Atomic Energy Agency. Planning national
radiotherapy services: a practical tool. http://www-pub.iaea.org/
books/iaeabooks/8419/Planning-National-Radiotherapy-Services-A-
Practical-Tool (accessed Dec 18, 2012).
www.thelancet.com/oncology Vol 14 February 2013