RESEARCH AND DEVELOPMENT
36 CANCER CONTROL 2021
complete in their registry.
IIn February 2019, over 300 cancer registries were invited
to complete online questionnaires to identify those which
have data at the required level of completeness, or were
willing to improve their data. Analyses of the questionnaires
were presented in Lisbon, Portugal (May 2019), in Vancouver,
Canada (June 2019) and in Moscow, Russian Federation
(September 2019). By September 2019, 123 cancer
registries in 42 countries (4 lower-middle-income, 14 uppermiddle-income, 24 high-income)
had submitted at least one
questionnaire, but most registries (80%) submitted all three.
Cancer registries have continued to submit questionnaires
to be considered for participation. By mid-September 2021, we
had received 374 questionnaires from 136 cancer registries
(Figure 2). The impact of the COVID-19 pandemic is visible.
For most of the data items that we wish to examine, the
answer as to whether they are collected routinely was
categorised as "Yes", "No" or "Unknown" and, if "Yes", the
expected level of completeness was categorised as less than
25%, 25%-49%, 50%-74%, 75%-100%, or unknown. Several
questions invited free-text replies.
The questionnaire on breast cancer included 71 questions,
which produced more than 300 separate data items (variables).
The questionnaire on cervical cancer included 62 questions
(300 variables), while the questionnaire on ovarian cancer
included 54 questions (over 250 variables).
We will include cancer registries with the highest availability
and completeness of the required data. Following analysis
of the questionnaires for each cancer, the data collection protocol was finalised in November 2019. The protocol enables
cancer registries to start their own database for the study, but
following rigorous agreed rules.
We defined "basic information" as demographic data (e.g.,
age) and tumour data (e.g., the anatomic site and microscopic
appearance of the tumour). We evaluated three criteria for the
inclusion of cancer registries in VENUSCANCER:
J Basic information, plus data on stage at diagnosis and
treatment: for all three cancers, slightly more than 50%
of cancer registries collected these data with adequate
completeness.
J Basic information, plus data on stage at diagnosis,
treatment and molecular biomarkers: for all cancers,
around 30% of cancer registries collected these data with
adequate completeness.
J Basic information plus data on stage at diagnosis, molecular
biomarkers, treatment and socioeconomic status: for all
cancers, only 5% of cancer registries collected these data
with adequate completeness.
However, after considering the willingness of colleagues
to collect additional data for individual women from clinical
records, these percentages rose to 76%-79%, 44%-51%, and
9%-14%, respectively.
If we include only registries that can provide all the variables
we want to analyse at the required level of completeness
(option 3), the worldwide scope of the project would be
severely constrained. Therefore, we selected the criteria that
will enable us to conduct a study on patterns of care with the
Figure 2: Number of registries that have submitted questionnaires or data
0
20
40
60
80
100
120
140
160
04/02/19 04/07/19 01/12/19 29/04/20 26/09/20 23/02/21 23/07/21 20/12/21
Questionnaires
Data sets
4
14
80
25-49% 50-74% 75-100%
Data completeness
Surgery (81/114)
22
13
84
Less than
25%
25-49% 50-74% 75-100%
Data completeness
Surgical procedure (63/81 )
8
15
71
Radiotherapy (73/114)
5 8
15
71
Chemotherapy (73/114)
0
20
40
60
80
100
Percentage of registries
0
20
40
60
80
100
Percentage of registries
25-49% 50-74% 75-100% Less than 25-49% 50-74% 75-100%