Transcript Assessing the Effect of Clinical Guidelines on Cancer

Assessing the Effect of Clinical
Guidelines on Cancer Services
for Colorectal Cancer
UKACR Annual Conference, London
29th September 2004
Lou Gonsalves, Colin Brooks, Carly Mellors, Gill Lawrence
West Midlands Cancer Intelligence Unit
[email protected]
tel: 0121 414 7711
fax: 0121 414 7712
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Acknowledgements
WMCIU Registration and Data Quality teams
Henry Gowen
Lillian Somervaille - WMPHO
Richard Wilson - South Birmingham PCT
Gavin Rudge - University of Birmingham
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Colorectal cancer in the UK

3rd most commonly-diagnosed cancer
• over 35,000 cases per year
• increasing

2nd most common cause of cancer death
• over 16,000 deaths per year (>10% of all cancer deaths)

5 year survival just under 50%
• prognostic factors • stage at diagnosis
• admission method into hospital
• quality of surgery
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Improving Outcomes in
Colorectal Cancer

cancer service guidance
published by DH/NICE
• initially published in 1997
• updated June 2004
West Midlands Cancer Intelligence Unit

recommendations on
appropriate treatment
and care of people with
colorectal cancer

evidence-based
www.wmpho.org.uk/wmciu
Improving Outcomes in
Colorectal Cancer











Patient-centred care
Access to appropriate services
Multi-disciplinary teams
Diagnosis
Surgery and histopathology
Radiotherapy in primary disease
Adjuvant chemotherapy
Anal cancer
Follow-up
Recurrent and advanced disease
Palliative care
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Key Questions
Has clinical practice changed in the West
Midlands since the introduction of the clinical
guidelines?
Have outcomes improved between 1998 and
2002?
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Hospital Episode Statistics
(HES) data
What are HES data?

information on admitted patient care delivered by
NHS hospitals in England

sub-set of the data submitted by NHS Trusts to the
NHS-Wide Clearing Service (NWCS)

‘episode’ defined as a period of admitted patient care
under a particular consultant within a single hospital
provider (several episodes  spell)

do not include out-patient treatments
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Hospital Episode Statistics
(HES) data
What is recorded?

patient demographics
• postcode, date of birth, sex, NHS number

episode details
• diagnosis (7)(14 from 2002/3), OPCS4 codes (4)(12
from 2002/3), admission date, admission method, procedure
date, discharge date, discharge destination, episode start
and end dates, hospital code
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Linking data sets




Cancer registration database at WMCIU
HES database at West Midlands Public Health
Observatory (WMPHO)
Match cohort of cancer registry patients against
episodes on HES database
Issues of confidentiality
• identifiable patient information
• Data Protection Act 1998, Health and Social Care Act 2001,
Caldicott Principles
• NWCS Security and Confidentiality Advisory Group protocols
• UKACR and WMCIU Confidentiality Guidelines
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Linkage Methodology

Step 1 - match WMCIU patients to HES episodes
• on NHS number, postcode, date of birth, sex
• partial matches - corroborate with treatment data

Step 2 - data quality checks
• group HES episodes into spells and compare to WMCIU
treatment dates

Step 3 - compare treatments between HES and
WMCIU
• OPCS4 codes
• some treatments not expected to match
– diagnostic procedures, non-cancer related procedures (HES)
– treatments in private/voluntary organisations, out-patient
treatments (WMCIU)
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Linkage Methodology

Step 4 - produce composite database of combined
information
• demographic details
• tumour characteristics
• treatment details
– validated OPCS4 codes
– admission method
– length of stay in hospital
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Data
Tumours registered at WMCIU :

with colorectal cancer diagnosis
(ICD10 codes C18-C20)

diagnosed in the periods:
• 1st January - 31st December 1998
• 1st January - 31st December 2002

Resections: OPCS4 codes H04 - H11 and H33
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Diagnosis and surgery
1998
2002
Tumours diagnosed
3382
3192
Surgical interventions
2619
2294
81.52%
91.54%
Surgical interventions
‘matched’ to HES episodes

recording of NHS numbers in HES data has improved

not all interventions are expected to match
• day surgery on out-patient basis, private hospitals
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Admission method
(matched cases)
1998
2002
4%
26%
27%
69%
74%
elective
West Midlands Cancer Intelligence Unit
emergency
not known
www.wmpho.org.uk/wmciu
Histopathology reporting
IOG:

“The histopathologist should search for as many
lymph nodes as possible in the excised specimen
...and the number found should be audited.”

Measure: “The proportion of histopathology reports
which give the degree of involvement of surgical
margins...the number of lymph nodes examined and
the number involved.”
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Histopathology reporting
(matched cases)
% with margins
reported
Elective
% with margins
reported
Emergency
West Midlands Cancer Intelligence Unit
1998
2002
82.34%
87.95%
84.52%
89.69%
76.28%
83.92%
www.wmpho.org.uk/wmciu
Excision margins
(matched cases, resections only)
IOG:
 Measure: “Proportion of “curative” resections...with
involved surgical margins.”
% with
involved
margins
1998
2002
Elective
10.52%
7.81%
Emergency
11.95%
9.49%
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Surgical caseload
IOG:

“Surgery should be undertaken by specialist
colorectal cancer surgeons”
• 1997: “An average figure of one or two radical colorectal
resections per month has been suggested as a minimum
number.”
• 2004: “Each surgeon in the MDT should carry out a minimum
of 20 colorectal resections with curative intent per annum.”
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Surgical caseload
100%
100%
90%
90%
80%
80%
<5
70%
5-9
60%
50%
10-19
40%
>20
30%
% treated by surgeon
% treated by surgeon
(matched cases, resections only)
1998
2002
70%
60%
50%
40%
30%
20%
20%
10%
10%
0%
0%
elective
emergency
Admission method
West Midlands Cancer Intelligence Unit
elective
emergency
Admission method
www.wmpho.org.uk/wmciu
Conclusions

Linking cancer registration and hospital episode
statistics data facilitates the monitoring of cancer
services in the West Midlands

The introduction of clinical guidelines for the
management of colorectal cancer patients has
resulted in some changes in clinical practice
• better reporting of excision margins
• lower rates of margin involvement at first surgery
but
• little change in surgical specialisation
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu
Any questions?
West Midlands Cancer Intelligence Unit
www.wmpho.org.uk/wmciu