Presentation originally given to the NE GP Training Scheme in 2014 on the quality of Irish health care statistics and biasing factors

1. Biases in Irish Health Service
Statistics
William Behan
June 2014

2. How are Irish GPs Presented?
 High annual payment per public patient
 High private fees
 Low workload due to significant financial barrier to
GP attendance
 High payment per public consultation
 Under supply of GPs => Best paid doctors in Europe
 Only moderate national over 65 years old influenza
immunisation rate

4. High annual payment per public
patient
Includes GMS Superannuation
2008/09 Missing GPVC patients

5. • Consumer Price Index increased by 24% 2002-2014
• Average pay increase in the public health service 2002-2013 = 50%.
• Junior nurses and administrators starting in 2002 who have not
been promoted now earn over 60% more than their starting salary.
• 2013 4th
FEMPI reduction of 7.5% “fair and reasonable”
High annual payment per public
patient

6. How are Irish GPs Presented?
 High annual payment per public patient
High private fees
Low workload due to significant financial barrier to GP
attendance
High payment per public consultation
Under supply of GPs => Best paid doctors in Europe
Only moderate national over 65 years old influenza
immunisation rate
Based on under reflection of GMS
patient numbers and overstatement
of GMS income by the PCRS
(adding in non-GMS payments such
as Mother and Child Scheme,
Primary Immunisations, Heartwatch
and Methadone payments)

7. How are Irish GPs Presented?
 High annual payment per public patient
 High private fees / significant financial barriers
 Low workload
 High payment per public consultation
 Shortage of GPs => Best paid doctors in Europe
 Only moderate national over 65 years old influenza
immunisation rate

8. High private fees
The financial cost to patients of visiting a
GP in Ireland: a cross-sectional survey, S
O’Connell, W Cullen. Ir J Med Sci. 2001 Jan-
Mar;170(1):45-8.
•4 survey centres Summer 2000: Dublin suburban,
Dublin south city, Dublin north city and Rural town
centre
•384 surveyed, 319 had seen their GP in the previous
12 months and 79% usually paid for their visits to the
doctor.
•They paid between IEP12 and IEP40 on their last
visit to the GP. Median and mode IEP20, mean
IEP21.79.

9. Private patients pay a fee-for-service, and the average fee
in 2002 was €33 (Indecon Economic Consultants, 2003).

10. “with the full consultation
fee perhaps being between
€60 and €80.”
High private fees
Health Systems in Transition
Ireland:
Health System Review 2009
European Observatory on Health Systems
and Policies.

11. High private fees
National Consumer Agency Doctors and
Dentists Survey May 2010
•123 practices nationally.
•The average price, across all survey locations, for a
routine examination is €51.
•At national level the range is significant, from €35
to €70.

12. High private fees
thejournal.ie 2013
“GP fees in Dublin are the most
expensive in the country
Aug 22 5:57 PM 10,070 Views 81 Comments
THE AVERAGE NATIONAL cost of going to see the doctor is
€46.26, according to a new survey by Laya Healthcare.
The online research poll of 501 parents with children under
18 who hold private health insurance found that Dublin
has the most expensive GP visit costs, with an average
consultation fee of €53.29. People in the west and
northwest are paying the lowest GP costs at an average

13. High private fees
The strength of primary care in Europe:
an international comparative study
BJGP November 1, 2013 vol. 63 no. 616 e742-e750
“The highest (formal) payments in the public system
exist in Ireland, where patients without a medical card
(about 70% of the population) pay €60 to €80 for each
general practice visit, with no reimbursement. ”

14. High private fees/ Quality of GP
Country
The structure of primary care
Primary
care
governance
Economic
conditions
of primary
care
Primary
care
workforce
development
Ireland Weak Weak Strong
Country
The service-delivery process of primary care
Overall
primary
care system
strength
Access to
primary
care
Continuity
of primary
care
Coordinatio
n of primary
care
Comprehen
siveness of
primary
care
Ireland Weak Strong Weak Medium Weak
The strength of primary care in Europe: an international comparative
study
BJGP November 1, 2013 vol. 63 no. 616 e742-e750

16. High private fees/Accessibility
Unmet care needs for medical examination by
income level, European countries, 2011

18. Increase in life expectancy 1997-2007 and potential
gains in 2007 through greater efficiency

19. GDPHE: GDP Health Expenditure compared to clinical outputs

20. The Competition Authority

22. W Behan, 2014

23. 2001 CSO attendance rate GMS 8.9; Private 3.46 and All patients 5
2013 IMJ attendance rate GMS 7.72; Private 3.35 and All patients 5.17

24. How are Irish GPs Presented?
High annual payment per public patient
High private fees / significant financial barriers
Low workload
High payment per public consultation
Shortage of GPs => Best paid doctors in Europe
Only moderate national over 65 years old influenza
immunisation rate
International perception of high private fees is driven by the
2009 Health Systems in Transition Ireland: Health System Review
fee of €60-80 per consultation and references to “standard fees”
rather than the OECD working paper recommendation of average of
high, low & no fees which does not occur in Ireland.
OECD Health at a Glance 2013 clearly indicates that Irish GPs are
among of the most accessible in Europe
French/Belgian GPs have opportunities for extra income/charges on
top of their standard fees
In the case of Belgian GPS – they receive an annual payment to
support their practice so private fees are really just a top up on their
annual income

25. How are Irish GPs Presented?
 High annual payment per public patient
 High private fees / significant financial barriers
 Low workload
 High payment per public consultation
 Shortage of GPs => Best paid doctors in Europe
 Only moderate national over 65 years old influenza
immunisation rate

26. Low consultation rate: TCA

27. Low consultation rate: Budget
Perspectives 2012, ESRI

28. Low consultation rate: OECD
HEALTH CARE ACTIVITIES
Doctors consultations, Number per capita
2006 2007 2008 2009 2010 2011 2012 2011 (or
nearest year)
Australia 6.1 6.3 6.4 6.6 6.6 6.7 6.9 6.7
Austria 6.7 6.8 6.9 6.9 6.9 6.9 .. 6.9
Belgium 7.1 7.2 7.5 7.6 7.4 7.4 .. 7.4
Canada 7.6 7.7 7.8 7.7 7.4 .. .. 7.4
Germany 7.9 8.1 8.6 9.2 9.9 9.7 .. 9.7
Ireland .. 3.3 .. .. 3.8 .. .. 3.8
Netherlands 5.6 5.7 5.9 5.7 6.6 6.6 .. 6.6
United
Kingdom 5.1 5.0 5.9 5.0 .. .. .. 5.0
United States
3.8 4.0 3.9 4.1 .. .. .. 4.1
OECD
AVERAGE 6.6
Source: OECD Health Statistics 2013
http://stats.oecd.org/Index.aspx?DataSetCode=HEALTH_PROC

29. Central Statistics Office Surveys
 CSO 2001 Survey = 19.5 m consults pa GP
 CSO 2007 Survey = 12.2 m consults pa GP
 CSO 2010 Survey = 14.6 m consults pa GP
 ? Could this be right in a situation of rising
population, more elderly, rising doctor
numbers, more demanding patient
population.

30. Are Irish adult general practice consultation rates
as low as official records suggest?
A cross sectional study of 6 general practices. IMJ Nov/Dec 2013
Behan W, Molony D, Beame C and Cullen W.
 Study period Oct 2012-13
 Reviewed 27,080 files, 20,706 >18
 Pop comparable to national
 Practices 4 corners of Ireland
 All using CompleteGP and appointments
 Direct contact pt/dr—surgery, hse, phone
 OOH & Phone – one practice
 Rural, Mixed, Town, City

31. Work not included
 Repeat prescriptions
 Forms filling
 Giving laboratory results
 Secretarial support work
 Organising appointments with ancillary
 Reports, PMAs, the “GOOD letter”
Could be as much work again for which the GP is responsible

32. Counterintuitive 2007 and 2010
CSO GMS attendance rates
W Behan, 2014

33. Findings of Study
Table 2 Key study findings and

34. Counterintuitive 2007 and 2010
CSO GMS attendance rates
W Behan, 2014

35. ALL PATIENTS 1
ALL UNDER 70 2
ALL OVER 70 3
GMS 4
GMS Discretionary 5
GMS DVC 6
PRIVATE only 7
GMS UNDER 70s 8
GMS OVER 70s 9
Counterintuitive 2007 and 2010
CSO GMS attendance rates
W Behan, 2014

36. ALL PATIENTS 1
PRIVATE only 2
GMS 3
ALL UNDER 70 4
ALL OVER 70 5
GMS UNDER 70s 6
GMS OVER 70s 7
GMS Discretionary 8
GMS DVC 9
GP Attendance Rates
W Behan, 2014

37. ALL PATIENTS 1
PRIVATE only 2
GMS 3
ALL UNDER 70 4
ALL OVER 70 5
GMS UNDER 70s 6
GMS OVER 70s 7
GMS Discretionary 8
GMS DVC 9
GP Attendance Rates
W Behan, 2014

38. ALL PATIENTS 1
PRIVATE only 2
GMS 3
ALL UNDER 70 4
ALL OVER 70 5
GMS UNDER 70s 6
GMS OVER 70s 7
GMS Discretionary 8
GMS DVC 9
GP Attendance Rates
W Behan, 2014

39. Counterintuitive 2007 and 2010
CSO GMS attendance rates
W Behan, 2014

40. Survey of 249 Patients Recollection of GP
Attendance WONCA 2013
Patients < 65 yrs estimated attendance rates accurately
Patients 65-74 yrs under estimated their attendance rates by 20%
Patients 75+ yrs under estimated their attendance rates by 30%
This results in a total under estimation of consultation rates (Public &
Private) by 18%
This study shows a clear degradation of recollection of
consultations which predominantly affects those
patients 70 years and older.
Younger patients come less often and
remember more and older patients come more
often and remember less!

41. Difference in GP attendance rates can
be explained by study methodology
 1 year recollection surveys are the 2001 Living in Ireland
Survey (sample size 6,521), 2007 (sample 21,253) & 2010
(sample 15,673) CSO surveys.
 Audit of patient records are produced by the PCRS fee-per-
item consultation rates (7 GPs), 2008/9 QRESEARCH (4.3
million patients) and IMJ Nov 2013 audit of 6 practices (20,706
adults)
Survey Audit Records/
1 Year Recall 2 Week recall
ALL PATIENTS 3.1 5.2
GMS 5.3 8.4
PRIVATE only 2.2 3.4

42. Difference in GP attendance rates can
be explained by study methodology
W Behan, 2014

43. Does eliminating fees at point of access
affect Irish General Practice attendance
rates in the under 6 years old population?
A cross sectional study at six general practices
IMJ April 2014. Behan W, Molony D, Beame C, Cullen W.
Population-27,080 ; 1931 <6; 5,814 consultations,
750,000 predicted extra visits PA or 230 consults/ WTE GP
before extra unnecessary work demanded in contract.

44. How are Irish GPs Presented?
High annual payment per public patient
High private fees / significant financial barriers
Low workload
High payment per public consultation
Best paid doctors in Europe
Shortage of GPs promotes higher individual income
Only moderate national over 65 years old influenza
immunisation rate
• Utilising 1 year recollection, asking the wrong
question and sometimes utilising proxy respondents
which are well established methods of under-
reflecting GP attendance rates by about 40%.
• Clear consistency between 2001 LIIS, 2007 & 2010
CSO surveys which all demand 1 year recollection vs
2001 CSO 2 week survey, Qresearch 2009 audit, NUI
Galway 2010 audit and IMJ 2013 Adult & 2014 under 6
audits which all closely correlate with each other and
are re very dissimilar to 1 year recollection surveys.

45. The OECD HEALTH COMMITTEE: Health Accounts
Experts 2009 and Health Working Papers 2010 directly
quotes Charles F. Cannell's work stating (health survey
methodology) :
“non-sampling biases can be large”
1. The number of events forgotten increases proportionately
with the length of the recall period (upwards of 30-50%).
2. Events with less salience or impact on the individual are
more likely to be forgotten.
3. Proxy respondents tend to report 20% fewer events.
NB: If the question is unclear or does not reflect the subject
being studied, it will deliver an unreliable result
Why was the 2001 CSO methodology requiring 2 week
recollection changes to 1 year recollection for the 2007
and 2010 surveys when 2006 Eurostat EHIS protocol was
to require recollection over 4 weeks?

46. How are Irish GPs Presented?
 High annual payment per public patient
 High private fees / significant financial barriers
 Low workload
 High payment per public consultation
 Shortage of GPs => Best paid doctors in Europe
 Only moderate national over 65 years old influenza
immunisation rate

50. TCA and ESRI 2009/10 Reports /
Conference
 GP visit €50-55 for private patients
 2005 CSO data: doctors fees averaged €193 per
household and GP fees account for 70% overall
doctors fees.
 2000-2009 CPI rose 30% but GP fees rose 87%
 In 2008, GPs received €65 for every visit made by a
public patient (private patients €51 in 2010) => Irish
GPs more highly remunerated for seeing public
patients than private patients
 2011 conference: GPs received €120 for every new
over 70s visit (€640 capitation 2007/5.3 visits p.a.CSO 2007)

51. Average GP fee calculation
Total GP income for all GP consultations
Total number of consultations in GP
Average standard GP consultation fee
Total GP income for all consultations
Number of clinical consultations with
GP only, not including nurse and
telephone clinical consultations
W Behan, 2014

52. Average GP fee calculation
Total GP income for all private consultations
Total number of private consultations in GP
Total GP income for all public consultations
Total number of public consultations in GP
Average fee per consultation
Average Private fee per consultation
Average Public fee per consultation
Total GP income for all clinical consultations
Total number of GP, practice nurse and
clinical consultations W Behan, 2014

53. How are Irish GPs Presented?
High annual payment per public patient
High private fees / significant financial barriers
Low workload
High payment per public consultation
Shortage of GPs => Best paid doctors in Europe
Only moderate national over 65 years old influenza
immunisation rate
 Exaggerated total GMS payments and under-
reflection GMS attendance rates and numbers
 IMJ 2013 Audit accurately predicted
subsequently published 2012 fee-per-item and
N.I. consultation rates (assuming all R.of I.
private patients attend at GPVC rates)

54. How are Irish GPs Presented?
 High annual payment per public patient
 High private fees / significant financial barriers
 Low workload
 High payment per public consultation
 Shortage of GPs => Best paid doctors in Europe
 Only moderate national over 65 years old influenza
immunisation rate

55. Low number of GPs per 1,000 pop:
Budget Perspectives 2012, ESRI

56. Low consultation rate: Budget
Perspectives 2012, ESRI

57. Low consultation rate: TCA

58. Shortage of doctors promotes
higher individual income
W Behan, 2014

59. “Best Paid Doctors in Europe”

60. “Best Paid Doctors in Europe”
1. The European Observatory Report (2012)
2. GPs and the Irish primary care system: towards
Universal Primary Care? (2014) Oireachtas Library
and Research Service
“there are approximately 14 million visits to GPs annually”
“cited Irish costs as about €60 to €80
per visit with no reimbursement…overstatement”
“OECD figures suggest there is considerable scope
for a reduction in GP income in Ireland”

61. “Best Paid Doctors in Europe” – D. Molony version
 Payments 2,400 GMS GPs 2011 (including subsidies) was €447,750,479
or €186,563 per GP (best paid doctors in Europe)
 Excluding Nurse and Secretary subsidies to 2400 full Time GMS GPs
 in 2011 – 361,682,940 (not all nurse and secretary hours are subsidised)
 Approx 800 GPs employed by the GMS GPs --average cost of €100,000 per
year per GP. – €80,000,000
 Leaves €281,682,940 for 2,400 GMS GPs or €117,367.89 per GP in 2011
 FEMPI 2012 – Minus 10% or €105,631.10 per GMS GP
 FEMPI 2013 – Minus 7.5% or €97,708.77 per GMS GP
 This 2.27 times the Average Industrial Wage (The average industrial wage is
€43,101.76 Feb 2013).
 While this does not include private income it is before costs of Rent, Rates, IT,
Phone, Heating, Electricity, Broadband, Insurance, stationary, required training,
cost of Coops, none-subsidised staff.

62. “Best Paid Doctors in Europe”
 Health at a Glance 2011 OECD Indicators
(2009 Multiple of the Average Industrial Wage)
– “In Ireland, the data for self-employed GPs
include practice expenses, resulting in an over-
estimation” (3.5xAIW)
 Our estimation Ireland:
– 2011 is 2.8 x AIW
– 2013 is 2.27 x AIW
 This puts the Irish GP well below the Germans (3.7), Mexico (3.5),
United Kingdom (3.4), Netherlands (3.0), Canada (3.0), Iceland (3.0),
Denmark (2.7), Austria (2.7), Slovenia (2.3), France (2.1 but Net
instead of Gross Income) [2011 or 2009 data]

63. How are Irish GPs Presented?
 High annual payment per public patient
 High private fees
 High payment per public consultation
 Low workload / significant financial barriers
 Shortage of GPs => Best paid doctors in Europe
 Only moderate national over 65 years old
influenza immunisation rate

64. Only moderate national over 65 years
old influenza immunisation rate
2011/ 2010/ 2009/ 2008/ 2007/
2012 2011 2010 2009 2008
Epi-Insight/
OECD 63.8 53.8 70.1 61.7
HSE Media
Office 56.5 60.1 51.6 58.2 57.2
W Behan, 2014

65. How are Irish GPs Presented?
 High annual payment per public patient
 High private fees
 High payment per public consultation
 Low workload / significant financial barriers
 Shortage of GPs => Best paid doctors in Europe
 Only moderate national over 65 years old influenza
immunisation rate
 More recent reporting on Irish General Practice

66. More recent reporting on Irish
General Practice
“‘Ghost’ medical card bill costs us €210m”
“Between 2010 and 2013 GP numbers have increased by 570”
“Extending Universal General Practice to the state will cost
an extra €320 million.”
Currently (2014) €440 m. is spent on GP p.a. including non-
GMS, add €320 m. => total spend c. €750 m.
HSE spend €13 billion = 67% total health spend of c.€19b.
750/19,000 = 4% total health budget
UK GPs are given £10 b. not including pensions and ooh out
of NHS budget of £108.5 b + £27b. Private (7.3% total health
budget) to provide 330 million consultations and are seeking
11% = £33 per consultation
Universal GP = 28 m. consultations = €26.80 per con.

67. More recent reporting on Irish
General Practice
“‘Ghost’ medical card bill costs us €210m”
2012 PWC Medical Card Report figure was based on information
being submitted by the PCRS to PWC that appears to have been
heavily biased.
This €210 figure took attention away from the PWC Report claims
that the PCRS did not have the IT systems in place prior to
centralisation of medical card applications and local offices were not
processing medical card applications for many months prior to
centralisation in an apparent unofficial work to rule.
“Between 2010 and 2013 GP numbers have increased by 570”
Many older GPs were entitled to be on the register, but did not apply, are
now retiring and being replaced by younger GPs.
Younger GPs automatically joint the specialist register on qualification.
US data indicates that younger GPS work less than their older colleagues
independent of the increase feminisation of GP

68. How do state representatives
present Irish GPs?
 High annual payment per public patient
 High private fees vs France/Belgium
 High payment per public consultation
 Low workload / significant financial barriers
 Shortage of GPs => Best paid doctors in Europe
 Only moderate national over 65 years old influenza
immunisation rate
WHO BENEFITS FROM THE CULTIVATION OF
ERRONEOUS HEALTH STATISTICS?
FIGURES INFORMING CHANGE MUST BE
RELIABLE AND CONVINCING

69. Future Problems
EHIS Ireland 2014-2015

70. Research bias
Research bias, also called
experimenter bias, is a process where
the scientists performing the research
influence the results, in order to
portray a certain outcome.

71. Will eliminating fees at point of access affect
Irish General Practice attendance rates in
the under 6 years old population?
Irish Government figures disagree with April 2014 IMJ
audit and UK Under 6s GP attendance figures
.
W Behan, 2014

72. Will eliminating fees at point of access affect
Irish General Practice attendance rates in
the under 6 years old population?
Irish Government figures disagree with April 2014 IMJ audit,
Lifeways 2006 and UK 2009 Under 6s GP attendance figures
.
W Behan, 2014

73. Irish public health expenditure is falling as a
proportion of total expenditure on health

74. Irish expenditure on health fell dramatically between 1980
and 2000 while the rest of the world increased its
expenditure as a proportion of GDP

75. W Behan, 2014