1. cunliffeanalytics Cancer indicator trend analysis Stevenage locality Summary of practice level cancer indicators 2010 to 2012 Version 2.1 August 2013

2. 2 Introduction letter TBC

3. 3 Contents Page Introduction – purpose of the report4 Screening indicators5 Percentage of females aged 50–70 screened for breast cancer in last 36 months Percentage of females aged 25–64 attending cervical screening within target period Percentage of persons aged 60–69 screened for bowel cancer in last 30 months Screening indicator performance vs demographics Two week wait indicators16 Two Week Wait referral ratio Percentage of Two Week Wait referrals with cancer Percentage of new cancer cases treated which are Two Week Wait referrals Two week wait indicator performance vs demographics Emergency admission indicator27 Rate of emergency admissions with cancer per 100,000 population Proportion of persons diagnosed with cancer via an emergency admission Appendices33 Definitions for indicators and demographics.

4. 4 Introduction Purpose of the report The purpose of this report is to provide a three year summary of the key diagnosis and referral indicators for practices across Stevenage locality. Eight key indicators are reviewed at CCG, locality and practice level, highlighting how the activity rates have changed over the last three years, in relation to the current national targets and recommended ranges. The key indicators are: Percentage of females aged 50–70 screened for breast cancer in last 36 months Percentage of females aged 25–64 attending cervical screening within target period Percentage of persons aged 60–69 screened for bowel cancer in last 30 months Two Week Wait referral ratio Percentage of Two Week Wait referrals with cancer Percentage of new cancer cases treated which are Two Week Wait referrals Rate of emergency admissions with cancer per 100,000 population Proportion of persons diagnosed with cancer via an emergency admission Please note that this report is based on a small number of practices and therefore the locality level percentages shown are sensitive to volatile changes. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit 2010 to 2012 Maps contain: Ordnance Survey data © Crown copyright and database right 2012 Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012.

5. 5 Percentage of females aged 50–70 screened for breast cancer in last 36 months Aim to be above the national target (70%). Consider actively encouraging patients to participate in screening programmes with letters or opportunistic prompts. GPs can be influential here. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit Definition: The number of females aged 50-70 registered to the practice screened adequately in previous 36 months divided by the number of eligible females (aged 50-70) on last day of the review period. (See appendix for full definition) Indicator source(s): Data was extracted from the NHAIS via the Open Exeter system. Data was collected by the NHS Cancer Screening Programme.

6. 6 CCG average remains just below the national average but above the national target of 70% for the last two years. The number of practices within the Stevenage locality achieving the national target has remained at 44% for the last two years. Targets achieved for 3 years Summary statistics 1 National target > 70% Maps contain: Ordnance Survey data © Crown copyright and database right 2012, Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012 201020112012 England mean average71.8%72.5% CCG mean average69.3%72.5%71.4% Locality practice min63.2%60.8%53.2% Locality practice max79.3%74.3%79.7% Practices above national target 1 8(9)4(9) Practices above national target (%)88.9%44.4% Locality range and CCG mean average Rate distribution – has the profile changed? National target Key  CCG mean average Locality range — National target >70% Key 2010 2011 2012 Note: Published year shown, 2010 refers to 3 year coverage for 2007/08 to 2009/10, 2011: 2008/09 to 2010/11, 2012: 2009/10 to 2011/12 Key Targets achieved  3  2  1  0

7. 7 Four out of nine practices within the Stevenage locality achieved the 70% target in 2012. One practice failed to meet the target for the last three years. Note: Published year shown, 2010 refers to 3 year coverage for 2007/08 to 2009/10, 2011: 2008/09 to 2010/11, 2012: 2009/10 to 2011/12 Key  Year on year increase  Year on year decrease Above national target Below national target National target > 70% Practice scores 2010 to 2012 Year on year rateAt or above target Difference over 2 years (pp 1 ) 201020112012201020112012 E82111Symonds Green HC74.1%  70.2%  79.7%  9.4% E82086King George78.3%  73.5%  77.2%  3.7% E82056Shephall Way73.7%  67.4%  72.49%  5.1% E82005Stanmore Road HC75.4%  70.4%  70.9%  0.5% E82089Chells Way79.3%  74.3%  69.7%  -4.6% E82039St Nicholas HC73.5%  69.8%  65.8%  -4.0% E82093Bedwell MC70.3%  66.4%  61.4%  -5.0% E82641Manor House70.6%  63.4%  56.1%  -7.3% E82624Canterbury Way63.2%  60.8%  53.2%  -7.6%

8. Percentage of females aged 25–64 attending cervical screening within target period Aim to be above the national target (80%). Consider actively encouraging patients to participate in screening programmes with letters or opportunistic prompts. GPs can be influential here. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit Definition: The overall cervical screening coverage: the number of women registered at the practice screened adequately in the previous 42 months (if aged 24-49) or 66 months (if aged 50-64) divided by the number of eligible women on last day of review period. (See appendix for full definition) Indicator source(s): Data was extracted from the NHAIS via the Open Exeter system. Data was collected by the NHS Cancer Screening Programme.

9. 9 CCG average remains steady and has been above the national average for the last three years but below the national target of 80%. No practices within the Stevenage locality achieved the 80% target in 2012. Summary statistics 1 National target > 80% 201020112012 England mean average75.4%75.6%75.3% CCG mean average78.2%78.4%78.2% Locality practice min65.3%71.3%72.1% Locality practice max78.2%80.3%79.4% Practices above national target 1 0(9)1(9)0(9) Practices above national target (%)0.0%11.1%0.0% Locality range and CCG mean average Rate distribution – has the profile changed? National target Key  CCG mean average Locality range — National target >80% Key 2010 2011 2012 Note: Published year shown, 2010 refers to 3.5 or 5.5 year coverage for 2004/05Q3 to 2009/10, 2011: 2005/06Q3 to 2010/11, 2012: 2006/07Q3 to 2011/12 Targets achieved for 3 years Key Targets achieved  3  2  1  0 Maps contain: Ordnance Survey data © Crown copyright and database right 2012, Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012

10. 10 Eight out of nine practices within the Stevenage locality failed to meet the 80% target for the last three years. Note: Published year shown, 2010 refers to 3.5 or 5.5 year coverage for 2004/05Q3 to 2009/10, 2011: 2005/06Q3 to 2010/11, 2012: 2006/07Q3 to 2011/12 Practice scores 2010 to 2012 Year on year rateAt or above targetDifference over 2 years (pp 1 ) 201020112012201020112012 E82111Symonds Green HC78.2%  80.3%  79.4%  -0.9% E82089Chells Way77.5%  78.4%  78.2%  -0.1% E82056Shephall Way77.5%  78.2%  77.3%  -0.9% E82039St Nicholas HC78.1%  77.9%  75.9%  -2.0% E82005Stanmore Road HC72.7%  76.1%  75.6%  -0.4% E82641Manor House76.1%  77.8%  75.2%  -2.6% E82086King George77.0%  76.7%  74.5%  -2.2% E82624Canterbury Way65.3%  71.3%  72.3%  1.0% E82093Bedwell MC75.4%  73.9%  72.1%  -1.8% Key  Year on year increase  Year on year decrease Above national target Below national target National target > 80%

11. 11 Percentage of persons, 60–69, screened for bowel cancer in last 30 months Aim to be above the national target (60%). Consider actively encouraging patients to participate in screening programmes with letters or opportunistic prompts. GPs can be influential here. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit Definition: The number of persons aged 60-69 registered to the practice screened adequately in the previous 30 months divided by the number of eligible persons on last day of the review period. (See appendix for full definition) Indicator source(s): Bowel Cancer Screening System (BCCS) via the Open Exeter system. Data was collected by the NHS Cancer Screening Programme.

12. 12 201020112012 England mean average40.2%51.5%57.4% CCG mean average54.2%55.1%58.3% Locality practice min40.8%44.7%45.8% Locality practice max55.8%57.0%58.5% Practices above national target 1 0(9) Practices above national target (%)0.0% CCG average has increased over the last 3 years, but remains below the national target of 60%. 2012 saw an increase in the proportion 60-69 year olds being screened across the Stevenage locality. Summary statistics 1 National target > 60% Locality range and CCG mean average Rate distribution – has the profile changed? National target Key  CCG mean average Locality range — National target >60% Key 2010 2011 2012 Note: Published year shown, 2010 refers to 2.5 year coverage for 2007/08Q3 to 2009/10, 2011: 2008/09Q3 to 2010/11, 2012: 2009/10Q3 to 2011/12 Targets achieved for 3 years Key Targets achieved  3  2  1  0 Maps contain: Ordnance Survey data © Crown copyright and database right 2012, Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012

13. 13 No practices within the Stevenage locality achieved the 60% target during the last three years. Seven out of nine practices have increased year on year for the last three years. Note: Published year shown, 2010 refers to 2.5 year coverage for 2007/08Q3 to 2009/10, 2011: 2008/09Q3 to 2010/11, 2012: 2009/10Q3 to 2011/12 Practice scores 2010 to 2012 Year on year rateAt or above targetDifference over 2 years (pp 1 ) 201020112012201020112012 E82089Chells Way55.1%  54.9%  58.5%  3.6% E82086King George55.6%  57.0%  58.5%  1.5% E82005Stanmore Road HC55.8%  56.7%  56.9%  0.2% E82641Manor House52.4%  53.5%  56.5%  3.0% E82056Shephall Way46.3%  47.3%  52.1%  4.7% E82039St Nicholas HC44.8%  46.8%  51.1%  4.2% E82624Canterbury Way40.8%  45.2%  50.8%  5.7% E82093Bedwell MC46.0%  47.2%  49.3%  2.0% E82111Symonds Green HC47.5%  44.7%  45.8%  1.1% Key  Year on year increase  Year on year decrease Above national target Below national target National target > 60%

14. 14 Screening indicator performance vs demographics Percentage of females aged 50–70 screened for breast cancer in last 36 months Percentage of females aged 25–64 attending cervical screening within target period Percentage of persons aged 60–69 screened for bowel cancer in last 30 months

15. 15 Practices within the Stevenage locality failing to meet the screening targets tend to have a lower level of new cancer cases Number of practicesPopulation aged 65+Deprivation New cancer casesCancer deathsCancer prevalence

16. Two Week Wait referral ratio (Indirectly age standardised ) Aim to be referring within 20% of the England average two week wait referral rate. Rates outside this range may indicate over/under use of the two week wait referral route. You may wish to audit your referrals against NICE cancer referral guidance. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit Definition: The number of Two Week Wait referrals where cancer is suspected multiplied by 100,000 divided by the list size of the practice in question. Indicator source(s): Trent Cancer Registry based on Cancer Waiting Times data for England, 2011/12, held on the DH Cancer Waiting Times Database.

17. 17 The proportion of practices within the Stevenage locality achieving the best practice range of 80% to 120% decreased in 2012 to 11% of practices (1 in 9). Summary statistics 1 Best practice range = 80% to 120%, practices with less than 5 referrals excluded. 201020112012 England mean average100.0% CCG mean averagen/a Locality practice min42.8%49.7%34.5% Locality practice max135.4%86.7%82.7% Practices within best practice range 1 4(9)2(9)1(9) Practices within best practice range (%) 44.4%22.2%11.1% Locality range and CCG mean average Referral ratio distribution – has the profile changed? Best practice range Key Locality range — Best practice range = 80% to 120% Key 2010 2011 2012 Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12 Stevenage practices achieving the best practice range for 3 years Key Targets achieved  3  2  1  0 Maps contain: Ordnance Survey data © Crown copyright and database right 2012, Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012

18. 18 One out of nine practices within the Stevenage locality were within the best practice range of 80% to 20% in 2012. Five practices failed to achieve the best practice range for the last three years. Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12 Practice scores 2010 to 2012 Year on year rateWithin best practice rangeDifference over 2 years (pp 1 ) 201020112012201020112012 E82093Bedwell MC96.9%  69.9%  82.7%  12.8% E82005Stanmore Road HC88.7%  69.8%  77.8%  8.1% E82039St Nicholas HC84.5%  86.7%  77.8%  -8.9% E82111Symonds Green HC86.2%  80.2%  73.7%  -6.5% E82624Canterbury Way135.4%  49.7%  59.2%  9.5% E82086King George53.0%  67.2%  52.8%  -14.4% E82056Shephall Way59.3%  57.9%  52.6%  -5.3% E82089Chells Way42.8%  51.7%  47.8%  -3.8% E82641Manor House54.1%  53.0%  34.5%  -18.5% Key  Year on year increase  Year on year decrease Within best practice range Outside best practice range Best practice 80% to 120%

19. Percentage of Two Week Wait referrals with cancer Aim to have conversion rate between 8-14%. Rates outside this range may indicate over/under use of the two week wait referral route. You may wish to audit your referrals against NICE cancer referral guidance. There is no target number for referral as this depends on practice size and demographics. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit Definition: The ‘conversion rate’, i.e., the proportion of Two Week Wait referrals that are subsequently diagnosed with cancer: the number of new cancer cases treated in 2011/12 who were referred through the two week wait route divided by the total number of Two Week Wait referrals in 2011/12. Indicator source(s): Trent Cancer Registry based on Cancer Waiting Times data for England, 2011/12, held on the DH Cancer Waiting Times Database.

20. 20 The number of practices within the Stevenage locality achieving the best practice range of 8% to 14% has remained steady for the last three years. Summary statistics 1 Best practice = 8% to 14%, practices with less than 5 referrals excluded. 201020112012 England mean average11.2%10.9%10.6% CCG mean average10.5%10.6%11.6% Locality practice min7.9%5.7%5.0% Locality practice max22.6%20.7%19.2% Practices within best practice range 1 2(8)3(7)2(7) Practices within best practice range (%) 25.0%42.9%28.6% Locality range and CCG mean average Indicator distribution – has the profile changed? Key  CCG mean average Locality range — Best practice = 8% to 14% Key 2010 2011 2012 Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12 Best practice range Stevenage practices achieving the best practice range over 3 years Key Targets achieved  3  2  1  0 Maps contain: Ordnance Survey data © Crown copyright and database right 2012, Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012

21. 21 Two of the seven practices within the Stevenage locality achieved the best practice range of 8% to 14% within 2012. Two practices failed to achieve the best practice range for the last three years. Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12, ratios based on less than 5 referrals are not shown. Practice scores 2010 to 2012 Year on year rateWithin best practice rangeDifference over 2 years (pp 1 ) 201020112012201020112012 E82089Chells Way21.1%  19.2%   0.0% E82056Shephall Way16.3%  20.7%  18.8%  -1.9% E82093Bedwell MC15.7%  11.0%  16.6%  5.5% E82086King George13.9%  17.8%  16.2%  -1.5% E82005Stanmore Road HC7.9%  13.2%  12.8%  -0.5% E82111Symonds Green HC22.6%  16.4%  10.9%  -5.5% E82039St Nicholas HC8.9%  9.4%  8.0%  -1.4% E82641Manor House---- E82624Canterbury Way15.2%--- Key  Year on year increase  Year on year decrease Within best practice range Outside best practice range Best practice 8% to 14%

22. Percentage of new cancer cases treated which are Two Week Wait referrals Aim to be above the line and have more of your cancer cases diagnosed through the two week wait referral route. Consider doing the RCGP cancer diagnosis audit. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit Definition: The proportion of new cancer cases treated who were referred through the Two Week Wait route. Indicator source(s): Trent Cancer Registry based on Cancer Waiting Times data for England, 2011/12, held on the DH Cancer Waiting Times Database.

23. 23 CCG average has increased year on year and has achieved the recommended minimum of 40% for the last two years, but remains below the national average. Summary statistics 1 Recommended minimum = 40%, practices with less than 5 referrals excluded. 201020112012 England mean average42.9%45.3%46.5% CCG mean average39.5%41.9%43.5% Locality practice min26.3%20.0%33.3% Locality practice max55.6%52.9%80.0% Practices above recommended min. 1 3(9)5(9)6(8) Practices above recommended min. (%) 33.3%55.6%75.0% Locality range and CCG mean average Indicator distribution – has the profile changed? Key  CCG mean average Locality range — Recommended minimum 40% Key 2010 2011 2012 Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12 Recommended minimum = 40% Stevenage practices achieving the recommended minimum of 40% over 3 years Key Targets achieved  3  2  1  0 Maps contain: Ordnance Survey data © Crown copyright and database right 2012, Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012

24. 24 Six out of eight practices within the Stevenage locality achieved the recommended minimum of 40% in 2012. One practice failed to achieve 40% for the last three years. Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12, ratios based on less than 5 referrals are not shown. Practice scores 2010 to 2012 Year on year rateAt or above targetDifference over 2 years (pp 1 ) 201020112012201020112012 E82641Manor House33.3%  22.2%  80.0%  57.8% E82093Bedwell MC42.3%  37.8%  58.9%  21.2% E82086King George32.7%  44.8%  44.4%  -0.3% E82089Chells Way32.2%  41.4%  42.9%  1.5% E82005Stanmore Road HC26.3%  47.7%  41.1%  -6.6% E82056Shephall Way33.3%  52.9%  40.0%  -12.9% E82111Symonds Green HC54.5%  40.9%  33.3%  -7.6% E82039St Nicholas HC28.6%  37.1%  33.3%  -3.8% E82624Canterbury Way55.6%  20.0%-- Key  Year on year increase  Year on year decrease Above recommended minimum Below recommended minimum Recommended minimum = 40%

25. 25 Two week wait indicator performance vs demographics Two Week Wait referral ratio (Indirectly age standardised ) Percentage of Two Week Wait referrals with cancer Percentage of new cancer cases treated which are Two Week Wait referrals

26. 26 Practices within the Stevenage locality failing to meet two week wait best practice ranges/ recommended minimum tend to have a higher level of deprivation and a lower level of new cancer cases Number of practicesPopulation aged 65+Deprivation New cancer casesCancer deathsCancer prevalence Excludes practices with low volumes

27. Rate of emergency admissions with cancer, per 100,000 population Aim to minimize the number of cancer patients requiring emergency admissions. Try to proactively manage cases. Consider using the RCGP Significant Event Audit to reflect on cases. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit Definition: The number of persons admitted to hospital as an inpatient or day-case via an emergency admission multiplied by 100,000 divided by the number of persons in the practice list, expressed as a rate per 100,000 persons. Indicator source(s): Hospital Episode Statistics (HES) data for 1st March 2011 to 29th February 2012 was taken from the UKACR “Cancer HES” offload originally sourced from the NHS Information Centre for Health and Social Care HES dataset.

28. 28 CCG average varies year on year but remains below the recommended maximum of 578 and below the national average. The maximum rate for the Stevenage locality has increased year on year Summary statistics 1 Recommended maximum = national average (587 in 2012), practices with less than 5 admissions excluded. 201020112012 England mean average691583587 CCG mean average519459502 Locality practice min131152233 Locality practice max616656857 Practices below recommended max. 1 8(9) 5(9) Practices below recommended max. (%) 88.9% 55.6% Locality range and CCG mean average Rate distribution – has the profile changed? Key  CCG mean average Locality range — Recommended maximum = 587 Key 2010 2011 2012 Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12 Recommended maximum = 587 Stevenage practices achieving recommended maximum rate of 587 over 3 years Key Targets achieved  3  2  1  0 Maps contain: Ordnance Survey data © Crown copyright and database right 2012, Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012

29. 29 Five out of nine practices in the Stevenage locality were below the recommended maximum of 587 in 2012. Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12, rates based on less than 5 admissions are not shown. Practice scores 2010 to 2012 Year on year rateAt or below targetDifference over 2 years (pp 1 ) 201020112012201020112012 E82089Chells Way616  486  857  371 E82111Symonds Green HC318  534  682  148 E82056Shephall Way442  418  680  262 E82093Bedwell MC380  560  667  107 E82624Canterbury Way328  656  552  -103 E82005Stanmore Road HC540  474  513  39 E82086King George344  568  508  -60 E82039St Nicholas HC362  409  449  40 E82641Manor House131  152  233  82 Key  Year on year increase  Year on year decrease Below recommended maximum Above recommended maximum Recommended range <National average (587 in 2012)

30. Proportion of persons diagnosed with cancer via an emergency admission Aim to have as few emergency presentations of cancer and more of the cases detected through managed referral routes. Consider using the RCGP significant Event Audit to reflect on cases and using Risk Assessment Tools to help guide investigation and referral. Data source: GP Practice Profiles for cancer, Cancer Commissioning Toolkit Definition: Proportion of persons diagnosed via an emergency, managed referral or other route. Indicator source(s): Routes to Diagnosis project database

31. 31 CCG average remains around the maximum of the recommended range (0% to 20%) Summary statistics 1 Recommended range = 0% to 20%, practices with less than 5 admissions excluded. 201020112012 England mean average23.7%23.8% CCG mean average20.4%20.9% Locality practice min0.0%5.9% Locality practice max47.2%66.7% Practices within recommended range 1 1(5)2(4) Practices within recommended range (%) 20.0%50.0% Locality range and CCG mean average Indicator distribution – has the profile changed? Key  CCG mean average Locality range — Recommended range = 0% to 20% Key 2010 2011 Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12 Recommended range - 0% to 20% Stevenage practices achieving recommended range of 0% to 20 % over 2 years Note: 2012 data not available across all CCGs Key Targets achieved  2  1  0  Not shown (low volumes) Maps contain: Ordnance Survey data © Crown copyright and database right 2012, Royal Mail data © Royal Mail copyright and database right 2012, National Statistics data © Crown copyright and database right 2012

32. 32 Two out of four practices in the locality were within the recommended range of 0% to 20% in 2011. Note: Published year shown, 2010 refers to 2009/10, 2011: 2010/11, 2012: 2011/12, rates based on less than 5 admissions are not shown. Practice scores 2010 to 2012 Year on year rateAt or below targetDifference over 2 years (pp 1 ) 201020112012201020112012 E82093Bedwell MC47.2%  30.9%-- E82005Stanmore Road HC28.2%  27.4%-- E82086King George36.0%  16.4%-- E82089Chells Way-15.3%-- E82641Manor House---- E82624Canterbury Way---- E82111Symonds Green HC---- E82056Shephall Way17.9%--- E82039St Nicholas HC34.5%--- Key  Year on year increase  Year on year decrease Within recommended range Outside recommended range Recommended range = 0% to 20% Note: rates based on less than 5 admissions are not shown.

33. APPENDIX Indicator definitions

34. 34 Percentage of females aged 50–70 screened for breast cancer in last 36 months Aim to be above the national target (70%). Consider actively encouraging patients to participate in screening programmes with letters or opportunistic prompts. GPs can be influential here. Indicator definition Number: The number of females aged 50 to 70 registered to the practice who were screened adequately in the previous 36 months. Rate or proportion: 3-year screening coverage %: The number of females registered to the practice screened adequately in previous 36 months divided by the number of eligible females on last day of the review period. Method: Data was taken from the Open Exeter system without further processing. The data extracted represents the situation at April 2011, and covers the period 2009/10-2011/12. Binomial confidence intervals are calculated using the Wilson score method 1. Source(s): Data was extracted from the NHAIS via the Open Exeter system. Data was collected by the NHS Cancer Screening Programme. Interpretation: Women are invited for screening for the first time between their 50th and 53rd birthdays and every three years thereafter up to but not including their 71st birthdays. Over this 21 year window a woman who responds to each invitation should be screened 7 times. This indicator measures the fraction of this pool of eligible women who have been screened adequately, at least once, in the three years before April 2011. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

35. 35 Percentage of females aged 25–64 attending cervical screening within target period Aim to be above the national target (80%). Consider actively encouraging patients to participate in screening programmes with letters or opportunistic prompts. GPs can be influential here. Indicator definition Number: The number of women registered at the practice screened adequately in the previous 42 months (if aged 24-49) or 66 months (if aged 50-64) Rate or proportion: The overall cervical screening coverage: the number of women registered at the practice screened adequately in the previous 42 months (if aged 24-49) or 66 months (if aged 50-64) divided by the number of eligible women on last day of review period. Method: Data was taken from the Open Exeter system without further processing. The data extracted represents the situation at April 2011, and covers the period 2006/07Q3-2011/12. Binomial confidence intervals are calculated using the Wilson score method 1. Source(s): Data was extracted from the NHAIS via the Open Exeter system. Data was collected by the NHS Cancer Screening Programme. Interpretation: Women aged 25-49 are invited for routine screening every 3 years and women aged 50-64 are invited for routine screening every 5 years. This indicator gives a combined coverage for the full age range so that it counts women aged 25-49 screened within a period of 3.5 years and women aged 50-64 within a period of 5.5 years prior to the report date and combines the counts to give the final measure. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

36. 36 Percentage of persons, 60–69, screened for bowel cancer in last 30 months Aim to be above the national target (60%). Consider actively encouraging patients to participate in screening programmes with letters or opportunistic prompts. GPs can be influential here. Indicator definition Number: The number of persons aged 60 to 69 registered to the practice who were screened adequately in the previous 30 months. Rate or proportion: 2.5-year screening coverage %: The number of persons registered to the practice screened adequately in the previous 30 months divided by the number of eligible persons on last day of the review period. Method: Data was taken from the Open Exeter system without further processing. The data extracted represents the situation at April 2011, and covers the period 2009/10Q3-2011/12. Binomial confidence intervals are calculated using the Wilson score method 1. Source(s): Data was extracted from the Bowel Cancer Screening System (BCCS) via the Open Exeter system. Data was collected by the NHS Cancer Screening Programme. Interpretation: This indicator measures the fraction of this pool of eligible people who have been screened adequately in the previous 2.5 years. Caution should be used in interpreting the data as not all CCGs had full implementation of the programme in the recorded period. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

37. 37 Two Week Wait referral ratio (indirectly age standardised) Aim to be referring within 20% of the England average two week wait referral rate. Rates outside this range may indicate over/under use of the two week wait referral route. You may wish to audit your referrals against NICE cancer referral guidance. Indicator definition Number: The number of Two Week Wait (GP urgent) referrals where cancer is suspected for patients registered at the practice in question in 2011/12. Rate or proportion: The crude rate of referral: the number of Two Week Wait referrals where cancer is suspected multiplied by 100,000 divided by the list size of the practice in question. Method: Patient level Cancer Waiting Times (CWT) data (including patient identifiers) was downloaded from the DH Cancer Waiting Times Database by the Trent Cancer Registry. Each patient was traced to a GP Practice using the Open Exeter Batch Tracing Service. Two Week Wait Referrals were identified for patients with a date first seen on the CWT database in 2011/12. All records with a ‘Referral Priority Type’ of 3 (Two Week Wait) were counted, excluding patients referred for non- cancer breast symptoms. Poisson confidence intervals are calculated using Byar’s approximation 1. Source(s): Trent Cancer Registry based on Cancer Waiting Times data for England, 2011/12, held on the DH Cancer Waiting Times Database. Interpretation: The number of Two Week Wait referrals with a suspicion of cancer, whether or not cancer was subsequently diagnosed. This indicator may be expected to be higher in practices with an unusually high proportion of persons of 65+ years of age, due to the higher incidence of cancer at these ages. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

38. 38 Percentage of Two Week Wait referrals with cancer Aim to have conversion rate between 8-14%. Rates outside this range may indicate over/under use of the two week wait referral route. You may wish to audit your referrals against NICE cancer referral guidance. There is no target number for referral as this depends on practice size and demographics. Indicator definition Number: The number of Two Week Wait referrals treated for cancer for patients registered at the practice in question. Rate or proportion: The ‘conversion rate’, i.e., the proportion of Two Week Wait referrals that are subsequently diagnosed with cancer: the number of new cancer cases treated in 2011/12 who were referred through the two week wait route divided by the total number of Two Week Wait referrals in 2011/12. Method: Patient level Cancer Waiting Times data (including patient identifiers) was downloaded from the DH Cancer Waiting Times Database by the Trent Cancer Registry. Each patient was traced to a GP Practice using the Open Exeter Batch Tracing Service. Patients on the CWT database who had received a cancer diagnosis were identified as those patients receiving a first treatment in 2011/12, i.e. with ‘Cancer Treatment Event Type’ of 01 (First definitive treatment for a new primary cancer) or 07 (First treatment for metastatic disease following an unknown primary). It was not possible to directly identify which referrals were subsequently diagnosed with cancer. Therefore, the proportion of referrals diagnosed with cancer was calculated by dividing the number of patients receiving a first treatment in 2011/12 who were referred through the two week wait route by the number of two week wait referrals. Most of the Two Week Wait referrals first seen in 2011/12 who were diagnosed with cancer will have started treatment in 2011/12 but a small number will have started treatment in 2011/12 and a small number of patients who started treatment in 2011/12 will have been first seen in 2010/11. For a very small number of practices, this may result in a ‘conversion rate’ of more than 100% being calculated. Binomial confidence intervals are calculated using the Wilson score method 1. Source(s): Trent Cancer Registry based on Cancer Waiting Times data for England, 2011/12, held on the DH Cancer Waiting Times Database. Interpretation: The number of Two Week Wait referrals with a suspicion of cancer, in which cancer was subsequently diagnosed. The proportion is the ‘conversion rate’ for the practice. This varies by cancer type and so will depend on the case-mix of cancers diagnosed in persons registered at the practice. Either an unusually high or an unusually low conversion rate may merit further investigation. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

39. 39 Percentage of new cancer cases treated which are Two Week Wait referrals Aim to be above the line and have more of your cancer cases diagnosed through the two week wait referral route. Consider doing the RCGP cancer diagnosis audit. Indicator definition Number: The number of patients registered at the practice who have a date of first treatment in 2011/12 on the cancer waiting times system. Rate or proportion: The proportion of new cancer cases treated who were referred through the Two Week Wait route. This is calculated as the number of persons referred as a Two Week Wait referral who were subsequently diagnosed with cancer divided by the total number of patients registered at the practice who have a date of first treatment in 2011/12 on the cancer waiting times system. Method: Patient level Cancer Waiting Times data (including patient identifiers) was downloaded from the DH Cancer Waiting Times Database by the Trent Cancer Registry. Each patient was traced to a GP Practice using the Open Exeter Batch Tracing Service. Binomial confidence intervals are calculated using the Wilson score method 1. Source(s): Trent Cancer Registry based on Cancer Waiting Times data for England, 2011/12, held on the DH Cancer Waiting Times Database. Interpretation: This indicator shows the proportion of cancers that were first diagnosed following a two week wait referral. This varies by cancer type and so will depend on the case-mix of cancers diagnosed in persons registered at the practice. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

40. 40 Rate of emergency admissions with cancer, per 100,000 population Aim to minimize the number of cancer patients requiring emergency admissions. Try to proactively manage cases. Consider using the RCGP Significant Event Audit to reflect on cases. Indicator definition Number: The number of persons admitted to hospital as an inpatient or day-case via an emergency admission, with a diagnostic code that includes cancer. Rate or proportion: The number of persons admitted to hospital as an inpatient or day-case via an emergency admission multiplied by 100,000 divided by the number of persons in the practice list, expressed as a rate per 100,000 persons. Method: All emergency admissions with an invasive, in-situ, uncertain or unknown behaviour, or benign brain cancer (ICD-10 C00-C97, D00-D09, D33, and D37-48) present in any of the first three diagnostic fields were extracted from the inpatient HES database. Source(s): Hospital Episode Statistics (HES) data for 1st March 2011 to 29th February 2012 was taken from the UKACR “Cancer HES” offload originally sourced from the NHS Information Centre for Health and Social Care HES dataset. Interpretation: The number and crude rate per 100,000 persons of emergency in-patient or day-case admissions, sourced from HES data, with a diagnosis that includes cancer. These may occur at any stage of the cancer pathway and will include persons diagnosed with cancer in prior years. This indicator may be expected to be higher in practices with an unusually high fraction of persons of 65+ years of age, due to the higher incidence of cancer at these ages. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012)

41. 41 Aim to have as few emergency presentations of cancer and more of the cases detected through managed referral routes. Consider using the RCGP significant Event Audit to reflect on cases and using Risk Assessment Tools to help guide investigation and referral. Indicator definition Number: Number of persons diagnosed via an emergency route, as defined by the Routes to Diagnosis project methodology 1 Rate or proportion: Number of persons diagnosed via an emergency route divided by the number of persons with any categorised route to diagnosis. Method: The data for the pool of patients diagnosed with cancer (ICD-10 C00-C97 excluding C44) in 2008 cancer registry records was examined. These were linked at a patient level to the Routes to Diagnosis In brief, the Routes to Diagnosis project method was that data sources of Screening, Inpatient HES, Outpatient HES, and Cancer Waiting Times were used to trace the history of each patient diagnosed with cancer in the year 2008. Patient histories in the datasets above prior to diagnosis were used to categorise the route that the patient took to arrive at the point of diagnosis. Eight main routes were defined in the Routes to Diagnosis project, these are aggregated into three broad routes in these Practice Profiles – Emergency Presentation, Managed Presentation, and Other Presentation. Emergency presentations are those initiated by an emergency event of some type, Managed Presentations consist of those following a routine or Two week Wait referral from a GP, Other Presentations are those via screening, death certificate only, Inpatient Elective, Other outpatients, and Unknown. See the Routes to Diagnosis Project for further information 1. Binomial confidence intervals are calculated using the Wilson score method 2. Source(s): Routes to Diagnosis project database. Interpretation: The number of persons who present as an emergency. The rate is the estimated fraction of all presentations that are emergencies, though patients who were diagnosed with multiple independent cancers in the same year were excluded. Aggregated data may give slightly different totals for England than previously published as it applies only to those patients who can be traced to a practice database. Proportion of persons diagnosed with cancer via an emergency admission Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 Routes to Diagnosis methodology, available online at: http://www.ncin.org.uk/publications/routes_to_diagnosis.aspxhttp://www.ncin.org.uk/publications/routes_to_diagnosis.aspx 2 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

42. APPENDIX Demographic definitions

43. 43 Practice Population aged 65+ Indicator definition Number: The number of persons registered at the practice aged 65+. Rate or proportion: The percentage of persons registered at the practice aged 65+, defined by the number of persons registered at the practice divided by the list size of the practice. Method: Data is taken from the Attribution Dataset, extracted April 2011. The number of persons aged 65+ is the sum across the population in the 65-69, 70-74, 75-79, 80-84, and 85+ age-bands. The fraction of the practice population aged 65+ is calculated by dividing the number aged 65+ by the list size of the practice sourced from the 2011/12 QOF data. Binomial confidence intervals are calculated using the Wilson score method 1. Interpretation: The percentage of the population over the age of 65 may be expected to have a significant effect on the burden of cancer in the practice population. The percentage of the population is taken as at April 2011 and will not reflect changes since then. Source(s): Data sourced from the Attribution Dataset provided by the South East Public Health Observatory. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

44. 44 Socio-economic deprivation Indicator definition Number: The estimated quintile of deprivation of the practice. Rate or proportion: The estimated income domain score for the practice, which is the percentage of the practice list that is income deprived1. Method: Index of Multiple Deprivation (IMD) scores for each deprivation domain have been estimated for each practice by the English Public Health Observatories using the Index of Multiple Deprivation (IMD) 2010 by Lower Super Output Area (LSOA) 2. Briefly, the overall socio-economic deprivation of the practice is estimated by averaging the socio-economic deprivation of each person on the practice list based on their LSOA of residence. Practices were ranked nationally by Income Domain score and allocated into equal population quintiles (1 being coded as the most affluent quintile, and 5 as the most deprived quintile). Binomial confidence intervals are calculated using the Wilson score method 3. Interpretation: Several common cancers have a known dependence on the socio-economic status of the population. A more deprived population may be expected to have a higher incidence rate of lung cancer but lower incidence rates of prostate and breast cancer. Source(s): Data provide by the English Public Health Observatories. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 The English Indices of Deprivation 2010. Communities and Local Government. Available online at: http://www.communities.gov.uk/publications/corporate/statistics/indices2010 http://www.communities.gov.uk/publications/corporate/statistics/indices2010 2 GP practice IMD 2007 – Calculation Notes, South East Public Health Observatory, 2010. 3 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

45. 45 New cancer cases Indicator definition Number: The number of persons diagnosed with any invasive cancer excluding non-melanoma skin cancer (ICD-10 C00- C97, excluding C44) in 2010 Rate or proportion: The crude incidence rate per 100,000 persons: the number of new cases diagnosed multiplied by 100,000 divided by the practice list size. Method: All invasive cancers diagnosed in 2010 registered by cancer registries and present in the 2010 Office of National Statistics analysis dataset were included. These patients were matched to a GP surgery by tracing them by NHS number to find their current and previous practice. Persons were allocated to their practice at their time of diagnosis. If this was not possible (for example, due to the patient having moved practice more than once in the time between diagnosis and trace) they were not included. The resultant total number of cancer diagnoses across England is 93% of the Office of National Statistics total number of cases for the country. Source(s): Office of National Statistics 2010. Each patient was traced to a GP Practice using the NHS Personal Demographics Service. Interpretation: This indicator gives the number of new cases and incidence rate of invasive cancer (excluding non-melanoma skin cancer) in the practice population, as estimated from cancer registry data for calendar year 2010. Cancer registry data includes persons diagnosed solely through their death certificate or who died shortly after an emergency presentation in secondary care, so may be larger than number of persons known to the practice. However, as 7% of cases could not be traced to a specific practice and are not included numbers at an individual practice may be undercounted by approximately this much. Numbers of cases may also fluctuate year to year meaning that caution should be used in comparing this indicator to other indicators such as the number of new cancer cases treated in 2011/12 taken from the Cancer Waiting Times database. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012)

46. 46 Cancer deaths Indicator definition Number: The number of deaths with an underlying cause of death which is any invasive cancer (ICD-10 C00-C97) in 2011/12. Rate or proportion: The crude mortality rate per 100,000 persons: the number of deaths due to invasive cancer multiplied by 100,000 divided by the practice list size. Method: Records of all deaths in England occurring in 2011/12 were downloaded from the Primary Care Mortality Database. These were filtered on the Underlying Cause of Death by ICD-10 code to exclude all deaths not due to invasive cancer (ICD- 10 C00-C97)) and aggregated to GP Practices using the built-in practice codes. Binomial confidence intervals are calculated using the Wilson score method 1. Source(s): The Primary Care Mortality Database, which is a collaborative project between the Office of National Statistics and the Information Centre. Interpretation: This indicator gives the number of cancer deaths and crude mortality rate in the practice. Numbers of cases may fluctuate year to year meaning that caution should be used in comparing this indicator to other indicators such as the number of new cancer cases in 2010. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457

47. 47 Prevalent cancer cases Indicator definition Number: The number of persons registered on the practice cancer register. Rate or proportion: The proportion of persons on the practice cancer register: the number of persons on the practice cancer register divided by the practice list size. Method: Data is taken from the QOF dataset without further processing. Binomial confidence intervals are calculated using the Wilson score method 1. Source(s): Data sourced from the cancer prevalence field of the QOF 2011/12 data 2. Interpretation: The prevalence data is taken from QOF data for 11/12, and originally sourced from each practice’s cancer register. Recording methodology varies by practice and may underestimate the true cancer prevalence. Source: NCAT General Practice Profiles for cancer: meta-data for profile indicators (Version 3.0, December 2012) 1 APHO Technical Briefing 3: Commonly used public health statistics and their confidence intervals. Available online at: www.apho.org.uk/resource/view.aspx?RID=48457www.apho.org.uk/resource/view.aspx?RID=48457 2 2011/12 QOF data. Available online at: http://www.ic.nhs.uk/webfiles/publications/002_Audits/QOF_2011-12/Practice_Tables/QOF1112_Pracs_Prevalence.xlshttp://www.ic.nhs.uk/webfiles/publications/002_Audits/QOF_2011-12/Practice_Tables/QOF1112_Pracs_Prevalence.xls