Australian Government, Australian Government Actuary

3. Data

3.1 Data collection

3.1.1 For the purpose of preparing this report, certain data were collected from the MIIs and MDOs by Medicare Australia during late 2007 and early 2008 including:

  • details of practitioners who were identified as having become eligible for membership of the Scheme before 30 June 2007 (provided in February 2008);
  • details of claims/incidents notified to MIIs and MDOs by 30 June 2007 which might be claims eligible for reimbursement under the Scheme;
  • details of ROC support payments;3
  • actuarial estimates of that part of the cost of claims relating to incidents which occurred before 30 June 2007 (whether notified or not) and which is expected to be reimbursed under the Scheme;
  • actuarial estimates of that part of the future claims cost of medical incidents projected to be notified during the 2007-08 to 2010-11 financial years which is expected to be reimbursed under the Scheme; and
  • actuarial estimates of that part of the future claims cost of medical incidents occurring during 2007-08 which is expected to be reimbursed under the Scheme.

3.1.2 This report also utilises other data and information including that previously provided to Medicare Australia for the purpose of section 34ZW of the Medical Indemnity Act.

3.2 Data verification

3.2.1 The results in this report rely on information provided by MIIs and MDOs. This information is regarded as the most suitable information available for the current purpose.

3.2.2 Steps were taken to ensure, as far as practicable, that the information provided was prepared on a basis suitable for the purpose. Despite this, it is not possible to guarantee that the information provided is free from material error. The information was not independently audited. As was the case in previous years, there were some notable disparities in the data provided, some of which could not be readily explained. Moreover, there were some inconsistencies between data provided for this review and that provided for the previous review. All of this means that figures and estimates provided in this report need to be treated with some caution.

3.2.3 Historically, MDOs have not maintained data in a form which is directly amenable to ROC analysis. For example, it has not been possible to establish a comprehensive list of doctors who were eligible for the Scheme on 1 July 2004. This is not a criticism of the MDOs. It simply reflects that their business and information systems were not developed with a scheme like the Run-Off Cover Scheme in mind. However, in order to monitor the operation of the Scheme effectively, accurate and timely data is clearly important.

3.2.4 Certain information was sought from industry actuaries. Guidance was provided as to the nature of the data, calculations and information required. Discussions with industry actuaries were held to supplement the data provided.

3.2.5 As was the case last year, a range of assumptions was used by industry actuaries. Although some significant assumptions differ by only a few percentage points between actuaries, substantially different estimates of Scheme costs are produced. This reflects the highly uncertain nature of estimates of the costs of the Scheme.

3.2.6 It is to be expected that many of the data issues encountered will diminish in time. This is likely to take a few years. Until data issues subside, Scheme projections will be subject not only to the considerable inherent uncertainty which surrounds medical indemnity insurance business, but also to additional uncertainty associated with the amount and quality of the available data.

3.2.7 In general, the results in this report blend estimates provided by industry actuaries with other actuarial estimates based on data provided by the MIIs and assumptions and models developed within this office.

3.3 Eligible practitioners

3.3.1 Practitioners can become eligible for the Run-Off Cover Scheme by means of permanent retirement at age 65 years or older, cessation of practice for three years, death, permanent disability or maternity leave. In addition, practitioners who have worked under a subclass 422 (Medical Practitioner) or 457 (Business (Long Stay)) visa under the Migration Regulations 1994 become eligible for the Scheme when they have permanently ceased medical practice in Australia and ceased to reside in Australia.

3.3.2 Appendix 2 describes the test of eligibility for the Scheme and the process of issuing and notifying compulsory run-off cover to eligible practitioners. Eligible practitioners are entitled to receive notification of the terms and conditions of compulsory run-off cover from their MII. MIIs are also required to notify Medicare Australia of the details of the compulsory run-off cover provided, including the name of the practitioner and the date from which the cover took effect.

3.3.3 There are inherent lags involved in notification of the details of eligible practitioners to Medicare Australia. As a result, it will be possible only to estimate the number of practitioners who are eligible for the Scheme at any time. For example, there will often be a delay between the time that a practitioner becomes eligible for the Scheme and the time when the insurer becomes aware of this. More generally, it is also very possible that there will be circumstances where an insurer is unsure of the eligibility status of a practitioner indefinitely; for example, where a practitioner has not renewed their insurance for, say, three years. For all of these reasons, the numbers of eligible practitioners reported by insurers need to be treated with caution.

3.3.4 The number of practitioners eligible for the Scheme in this report is based on

  • data produced manually by the medical indemnity industry relating to practitioners identified as having become eligible between 1 July 2004 and 30 June 2007; and
  • industry estimates of practitioners eligible for the Scheme as at 1 July 2004, provided for the purpose of the 2004-05 report.

3.3.5 Table 1 summarises the data provided by the industry.

Table 1: Run-Off Cover Scheme eligible practitioners

Practitioners eligible for the Scheme as at 1 July 2004 2,112
Practitioners who became eligible for the Scheme during the 2004-05 financial year 442
Practitioners who became eligible for the Scheme during the 2005-06 financial year 574
Practitioners who became eligible for the Scheme during the 2006-07 financial year 603
Total number of practitioners who had become eligible for the Scheme by 30 June 2007 3,731

3.3.6 According to the data provided by the industry, 442 practitioners became eligible for cover under the Scheme during 2004-05, 574 during 2005-06 and 603 during 2006-07. These numbers are somewhat lower than corresponding estimates made by the industry last year and substantially lower than the number of practitioners that would be expected to be eligible based on our projection models.

3.3.7 There was some apparent inconsistency in the industry data provided for this report. For example, the number of doctors identified by the industry as having been eligible for the Scheme as at 1 July 2004 was less than the number identified as becoming eligible during 2006-07. This seems improbable to us since doctors who were eligible at 1 July 2004 would have become eligible over a much longer time frame than one year. As a result, we have based our estimate of the number of doctors eligible at 1 July 2004 in Table 1 on previous industry estimates. The estimate of 2,112 doctors eligible for the Scheme at 1 July 2004 is subject to considerable uncertainty.

3.3.8 Table 2 below illustrates the break-up of the 2004-05 to 2006-07 new entrants by reason of eligibility, based on the data provided by the underwriters. The relativities are compared to those implied by our model.

Table 2: Run-Off Cover Scheme new entrants by reason of eligibility

  New entrants (per cent)  
  Industry data Model
  2004-05 2005-06 2007-08 2007-08
Retired 41 49 36 39
Maternity 17 25 31 23
Permanent disability 5 3 3 3
Died 19 10 12 9
Other(a) 3 7 5  
Resigned 15 6 13 27
Total 100 100 100 100

(a) Overseas trained doctors who had permanently ceased practice under a 422 or 457 visa.

3.3.9 The relativities in Table 2 show some consistency by year of eligibility. However, there were significant differences in the break-up by reason of eligibility for each of the underwriters.

3.3.10 Costs of the Scheme are very sensitive to the assumed retirement pattern. For this investigation, we have adjusted the retirement decrements assumed in our models given the sensitivity of cost estimates to retirement rates. We have not adjusted other decrements. Our projections imply a similar proportion of retirees to industry data for 2004-05 to 2006-07 (about 39 per cent of all eligible doctors). However, our assumed retirement rates remain significantly higher than the observed rates in absolute terms. We do not believe that the reported information is sufficiently reliable to justify a greater reduction in our assumed rates.

3.3.11 It will be important to continue to monitor the aggregate reporting rates as more complete information becomes available.

3.4 Claims eligible for ROC indemnity payments

3.4.1 MIIs and MDOs are entitled to reimbursement from the Australian Government for the costs of claims which:

  • are first notified to the MII or MDO on or after 1 July 2004;
  • relate to a practitioner who is eligible under the Scheme at the date of notification (see Section 3.3); and
  • meet the other requirements for ‘payable claims’ (see Appendix 3).

3.4.2 MIIs provided details of individual medical incidents which they have identified as potentially being eligible for the Scheme. The data provided was not wholly consistent with that provided for last year’s report. Moreover, there were some apparent internal inconsistencies within the data. It is quite possible that other medical incidents have been notified to MIIs since 1 July 2004 which were not included in the data but which will be eligible for the Scheme. It is also possible that some of the incidents notified will not be eligible for the Scheme. Accordingly, these numbers should be treated with caution.

3.4.3 As at 30 June 2007, MIIs and MDOs reported 71 medical incidents relating to eligible medical practitioners since the commencement of the Scheme. 25 of these were identified as formal medical indemnity claims with an incurred cost of $3.6 million. Another seven of these were identified as likely to convert into formal medical indemnity claims which may then become payable under the Scheme. 13 incidents relate to the 2004-05 new entrants to the Scheme, 18 relate to the 2005-06 new entrants, and eight relate to the 2006-07 new entrants. The other 32 incidents relate to those practitioners who were eligible at the commencement of the Scheme on 1 July 2004.4

3.4.4 The data implies that 29 new incidents (including 25 new claims) were identified during 2006-07 which relate to prior periods. The implied case estimate development of claims present in the 2005-06 data was approximately 16 per cent.5

3.4.5 The number of medical incidents notified to MIIs and MDOs which could potentially give rise to a future ROC indemnity payment is lower than perhaps might have been expected. At this point we are only relying minimally on the data, given the inconsistencies in it and its small volume.

3.5 ROC indemnity payments

3.5.1 ROC indemnity payments are the payments made by the Australian Government to MDOs and MIIs as reimbursement of the costs of eligible claims.

3.5.2 No ROC indemnity payments had been made by 30 June 2007. However, indemnity payments were made in 2007-08 (after the effective date of this investigation).

3.5.3 The Scheme also provides for payments in respect of compliance costs and internal claims handling costs under the ROC Claims and Administration Protocol (section 34ZN of the Medical Indemnity Act).

3.5.4 During 2007-08, Medicare Australia will make payments to MIIs in relation to 2006-07 compliance costs of around $1 million.

3.5.5 The Commonwealth’s own administration costs are Budget-funded and so are not considered in this report.

3.6 ROC support payments

3.6.1 ROC support payments are paid to Medicare Australia in the form of an annual lump sum imposed as a tax on each MII from 1 July 2004 under the MI ROCSPA.

3.6.2 The amount of ROC support payments is calculated using a method set out in the MI ROCSPA. Appendix 1 describes the calculation in detail. Very briefly, it is based on:

Applicable rate (premium income less taxes and charges) (1 + applicable rate).

3.6.3 For most MIIs the applicable rate is currently 8.5 per cent. In order to provide equivalence on a present value basis, a slightly higher percentage applies to one insurer, AMIL, whose policy year is a calendar year and which remits its ROC support payment six months after the other MIIs.

3.6.4 Table 3 below summarises the ROC support payments received during the 2006-07 financial year.

Table 3: ROC support payments

    ROC support payments ($’m)
Paid 31 December 2006 AMIL 10.136
Paid 30 June 2007 MIPSi 3.423
  QBE 0.222
  MDANI 4.686
  MIGA 2.906
  PIICA 5.216
  Total 26.588(a)

(a) Numbers do not add exactly due to rounding.

3.6.5 In order to provide full transparency for practitioners, MIIs are required to attribute ROC support payments to individual policyholders. Each premium notice specifies the amount that has been included in the policyholder’s invoice to meet the MII’s ROC support payment obligations. All amounts are reported to Medicare Australia, which maintains a record of each practitioner’s total run-off cover credit. Interest is applied to this balance annually at the short-term bond rate in accordance with section 34ZS of the Medical Indemnity Act.

3.6.6 Part 2, Division 2B, Subdivision E of the Medical Indemnity Act provides for certain payments, should the Scheme ever be wound up without alternative arrangements being put in place. Thus, doctors who were still practising at the time of the wind-up of the Scheme would be entitled to have an amount not exceeding their total run-off cover credit paid to their nominated medical indemnity provider. Practitioners who were eligible for the Scheme at the time of its wind-up would not be entitled to any refund but would continue to be covered for any future claims that might emerge.

3.6.7 Figure 2 below summarises the contribution to ROC support payments by age of practitioner. Note that age and gender were not available for a minority of doctors. However, the shape of the graph is similar to that produced in last year’s report. The proportion of ROC support payments is greater than the proportion of practitioners for doctors in their 40s up to their mid to late 60s. This reflects the low level of premiums for interns, trainees and hospital indemnified doctors aged in their 20s and 30s. The proportion of ROC support payments tends to diminish at higher ages. This provides some support for the suggestion that doctors tend to wind down their practice hours and possibly perform fewer risky medical procedures (for example, surgery) as they approach retirement, resulting in lower premiums.

Figure 2: Contribution to ROC support payments by age

Figure 2: Contribution to ROC support payments by age

3.6.8 Figure 3 below summarises the contribution to ROC support payments by area of specialisation. ISA specialty codes were not available in relation to a small minority of doctors. However, the profile of contributions is similar to that produced in last year’s report.

3.6.9 Medical indemnity insurance premiums tend to be risk-based. Thus, practitioners operating in risky areas of specialisation are likely to incur the highest premiums and, accordingly, the highest ROC support payment liabilities. The largest ROC support payments are for obstetricians, gynaecologists, neurosurgeons, cosmetic/plastic/reconstructive surgeons, orthopaedic surgeons, and general surgeons. Medical practitioners not otherwise classified (including interns, trainees and hospital indemnified doctors) have the smallest ROC support payments.

Figure 3: Contribution to ROC support payments by specialisation

Figure 3: Contribution to ROC support payments by specialisation


3 A database of ROC support payments is maintained by Medicare Australia.

4 Differs from the corresponding number reported last year of 51.

5 72 of the 114 incidents reported in our 2005-06 report did not appear in the current data set. The incurred cost of these incidents of $0.5 million was mostly attributable to 7 civil claims. We have assumed that these incidents were incorrectly flagged as Scheme claims as at 30 June 2006.

Next: Financial Management of the ROC Scheme
Previous: Background
Return to: Table of Contents