Transcript Population mixing and childhood leukaemia

Cancer trends in England and Wales
Dr Heather O Dickinson
http://www.staff.ncl.ac.uk/heather.dickinson/
Department of Child Health
University of Newcastle
Deaths by cause (all ages),
England & Wales, 1998
other 18%
cancer 25%
diseases of
respiratory
system 16%
diseases of circulatory system 41%
Deaths by cause and age group, 1998
cancer
Age 1-4 yrs
Age 5-24 yrs
circulatory system
respiratory system
accidents
nervous system
congenital anomalies
other
Age 25-74 yrs
Over 75 yrs
Adult cancer, diagnosed 1986-90
stomach
bladder
prostate
rectum
colon
lung
Affluent
Deprived
breast
0%
20%
40%
60%
5-year survival
80%
Adult cancer, diagnosed 1986-90
stomach
bladder
prostate
rectum
colon
lung
Affluent
Deprived
breast
ALL CANCERS
0%
20%
40%
60%
5-year survival
80%
Target
In July 1999, the UK government
set a ‘tough but attainable’ target:
to reduce the death rate from
cancer in people under 75 by at
least a fifth by 2010 (compared
with 1997)
- saving up to 100,000 lives
Cancer mortality, under age 75 yrs
Annual mortality
per 100,000
200
100
0
1950 1960 1970 1980 1990 1998
Year
Male deaths from cancer, 1998
lung
prostate
colorectal
stomach
oesophagus
bladder
0
5,000
10,000
15,000
20,000
Female deaths from cancer, 1998
breast
lung
colorectal
ovary
pancreas
stomach
oesophagus
0
5,000
10,000
15,000
20,000
Cancer mortality, males, age 45-74 yrs
Annual mortality
per 100,000

+
lung
stomach


colorectal
prostate
400
200
0
1950
1960
1970
1980
Year
1990 1998
Cancer mortality, females, age 45-74 yrs
Annual mortality
per 100,000

+
lung
stomach

o
colorectal
breast
150
100
50
0
1950
1960
1970
1980
Year
1990 1998
Death rates from cancer
depend on:
 incidence
–several years ago
 survival
–over the past few years

We can decrease the
incidence through
prevention.

We can improve
survival through better
treatment.
Can starting prevention
strategies now affect
the incidence enough to
reduce the death rate
by 2010?
Smoking accounts for
 over one third of cancer deaths
– lung, mouth, larynx, oesophagus
and other cancers

about one fifth of other deaths
– mainly from circulatory and
respiratory disease
Survival has improved.
 If it continues to improve,
roughly 24,000 deaths will be
avoided by 2010
 If survival for everyone were as
good as survival of the most
affluent, about 41,500 deaths
would be avoided.

Eliminating social class
differences - in both
incidence and survival would almost certainly
save more lives in the
next decade than
innovative treatments.
Annual mortality
per 100,000
Cancer mortality, children 0-14 years
8
6
4
2
0
1950 1960 1970 1980 1990 2000
Year
Children’s cancer, diagnosed 1986-90
Affluent
Deprived
Brain and spinal tumours
Acute lymphoblastic leukaemia
0%
20%
40%
60%
5-year survival
80%
Children’s cancer - a success story
Why?
 many childhood malignancies are
chemosensitive - and among the
first for which curative
chemotherapy was developed
 rare disease - so a manageable
problem
Children’s cancer - a success story
Why?
 treatment at regional centres
 cross-speciality communication
 evidence-based treatment
 national collaboration in
treatment protocols
 most patients entered into
clinical trials

Can adult cancer be treated as
successfully as children’s
cancer?

Can we give everyone the best
care, irrespective of their social
status?
Good statistics are the
crucial underpinning
of government policy.
Statistics are needed for:
 valid target setting
 planning service
delivery
 audit of performance
High quality statistics:
 accurate
 complete
 timely
What sort of statistics?
 incidence }
{ age
 mortality
} by { sex
 survival
}
{ tumour type
How do we use the statistics?
 to analyse trends
 to analyse factors affecting trends
 to predict the effects of these
factors as the age structure of the
population changes
Health care has taken the
lead in calling for evidence
based decisions;
government policy likewise
needs to be determined by
a firm knowledge base.
I work on the epidemiology of children’s
cancer.
I previously taught English to people from
other countries - mainly Bangladesh and
Pakistan - who had settled in England. I
integrated health education into my English
teaching. This work made me more aware of
the inequalities in society, both within
England and between different countries.
Heather Dickinson
 Learning
objectives - to understand:
 factors influencing cancer incidence
and survival
 stratifying by age, sex, social class
 national statistics on disease
 Performance
objectives - to assess:
 national trends in disease rates
 targets for reduction in mortality