Chapter 2

Audit Procedures

Medicare Australia

2.1        Since the inception of the PSR it has received 767 referrals from Medicare Australia ("Medicare").  This represents a long term average of 45 per annum.[1]  As of June 2011 there were 95 000 health practitioners providing services that attract a Medicare benefit.[2]  The practitioners referred to PSR represent less than 0.1 of one per cent of all medical practitioners.[3] Outcomes from the PSR's work during 2004-05 to 2010-11 include:

• 70 matters in which decisions to take no further action were taken (21 per cent of matters);
• 166 Agreements in which a practitioner acknowledged inappropriate practice were negotiated and entered into (49 per cent of matters);
• 103 final determinations were made in which a practitioner was found to have engaged in inappropriate practice by a Committee of their peers (30 per cent of matters).[4]

2.2        The committee received a significant amount of evidence focussed on the audit procedures utilised by Medicare that identify practitioners with profiles significantly different from that of their peers.  There was a particular focus on the role of statistics and the methodology employed to highlight anomalies.

2.3        The committee undertook an inquiry in March 2009 into Compliance Audits on Medicare Benefits and reflected in that report on some of the issues raised.

2.4        In response to the specific concerns raised throughout the current inquiry, on 26 September 2011 the committee sent written questions to the Department of Human Services (DHS), which is responsible for compliance activity relating to MBS and PBS including audits of services billed to Medicare.[5] The committee received a response on 5 October 2011.   

2.5        The Medicare Practitioner Review Program is a five step process which uses a practitioner's practice profile to identify whether they are rendering services significantly differently to their peers. The Program literature explains how the practitioner's profile is established:

The Profile is a suite of Medicare and PBS data which reflect the services rendered (provided by or on behalf of the practitioner) or initiated (eg requests for pathology or diagnostic imaging) under Medicare and PBS prescribing for that practitioner over a specified time frame.[6]

The department uses sophisticated data mining techniques to analyse large amounts of information and derive statistically valid methods of identifying health practitioners with profiles or practices significantly different from their peers. These methods take into account potential differences in claiming patterns between health practitioners based on such factors as patient demographics and location. Data mining tools are customised for each health practitioners group; and these tools are currently used to review general practitioners and specialists.

All profiling techniques rely on input from Medicare Medical Advisers (qualified medical practitioners with current practice experience). At all stages of analysis, Medicare Medical Advisers provide advice on the nature of the risk and conclusions drawn from the analysis.[7]

2.6        Much of the discussion in the public hearing and in written evidence discussed the representation of the profile statistics in terms of a bell or 'Gaussian' curve.  The suggestion was that the further to the right of that curve a practitioner's profile lies, the more likely that inappropriate practice is occurring.  Dr Ruse explained the assumptions used in the methodology:

Medicare deals with a huge volume of statistics regarding doctor claims. Because of varying individual doctor work patterns the frequency distribution for each item is not the classical bell shaped curve. There is a long right tail. Some of those doctors might just have chosen to work very long hours (for standard consultation item outliers). Some might have a particular clientele of patients generating a lot more than the average number of specific investigations or treatments.

However, as a doctor's position on the curve moves further right, it was assumed that the possibility of inappropriate categorisation, of consultation items, or inappropriate use of investigations or treatments, or inappropriate professional input, because of time constraints, (all potentially present anywhere in the curve) might be rising. I think that assumption, as a basis for further enquiry, but not a final judgement, was appropriate. The amount of sub standard professional practice that I have seen in my Committee work is, I am sure, much higher than that seen in more statistically "normal" practitioners.[8]

2.7        There were a number of suggestions that practitioners were identified purely on the use of statistics, and that this process included no context.  In the public hearing Mr Watt from the Australian Doctors Union (ADU) specifically cited an example that implied that there was no profiling involved other than the use of basic statistics:

...a simple, logical profiling of practices would explain, from the outset, why a female doctor at a female clinic specialising in treating females is ordering so many pap smears, as opposed to simply saying, 'Wow—look how many she's ordering compared to the bloke who is up the road!'[9] 

2.8        Dr Reece, Chairman of the ADU, also stated in his opening statement his analysis of the Medicare audit process: 

It is just a repeated line of, 'We notice that you are a statistical outlier, so you are different from your colleagues; therefore, we have got concerns.' When you ask them, 'What exactly is your concern?' they say, 'We're just letting you know that you're in this five or 10 per cent of people who do,' for instance, 'more long consultations than other doctors.' Then you might say back to them, 'How many would you like me to do? They say, 'No. That's not a problem. Just what your peers would find acceptable.'[10]

2.9        Dr Masters from the ADU also commented:

I think Medicare are suggesting to us that they would be very happy if we had this bell curve and there was a straight line right down the middle. They do not seem to like anybody on the end of the bell curve and they seem to want to interrogate them.[11]    

2.10      The response received through DHS described how Medicare assessed the information produced by its profiling techniques.  Their process includes:

• identifying unusual patterns of item usage and item combinations;
• reviewing top claimants by a range of fields including demographic and claim or service types; and
• identifying and applying patterns learned from previous cases of non-compliance.[12]

2.11      Dr Cootes, Acting Director of the PSR rebutted suggestions that Medicare do not take a number of factors into account.  He cited a submission from Medicare to the Community Affairs Committee's inquiry in 2009 into Compliance Audits on Medicare Benefits:

Medicare Australia uses sophisticated technology to compare factors including total benefits, services, patient demographics and prescribing of pharmaceuticals. The profiling system is adaptive and takes into account factors such as number of days worked and area of practice.[13]

2.12      That response from Medicare to the committee's inquiry goes on to say:

There are four broad situations in which a provider’s claims may be identified for audit. These are when:

• a provider has used an item with a medium to high risk of non-compliance;
• a provider’s individual claiming statistics appear to be unusual or irregular;
• a provider’s claiming statistics are significantly different to their peers; or
• a provider has been identified through ‘tip-offs’ and information received.

In each of these situations, Medicare Australia recognises there are often many acceptable reasons for claiming behaviour. Medicare Australia’s approach is not to assume an incorrect claim but to raise the concern with the provider and allow the provider the opportunity to explain their situation.[14]

2.13      The committee was concerned with the example used by the ADU that a women's health clinic would be subject to Medicare or PSR review for over-servicing in relation to pap smears. It requested that PSR review their records to discover if this was a genuine case. PSR responded:

There are 18 pap-smear specific item numbers in the Medicare Benefits Schedule.  A search of the PSR case management system returned no finding of a referral to PSR from Medicare Australia in relation to concerns around these Pap-smear items.

PSR has no record of any of these items ever being reviewed by PSR for any practitioner.[15]

2.14      The committee also heard that the auditing methodology unjustly identified practitioners who were particularly innovative and busy. The Australasian Integrative Medicine Association discussed their concerns at the public hearing:

From AIMA's observation, the problem starts with Medicare auditing for the longer consultations, which subsequently leads to the PSR referral. GPs who see more patients with chronic diseases or multiple health problems, such as AIMA members, are more likely to use longer consultation item numbers. Using the longer consultation item numbers, such as items 36 and 44 that I mentioned, more than the general body of peers is an indicator for Medicare auditing, and it has been noted by the PSR director in annual reports that it can lead to referral to the PSR.[16]

2.15      With respect to AIMA's evidence on the extent of the auditing of their members, the committee is not persuaded that they are being unfairly targeted due to their practice profile. AIMA's evidence was that a small number of its members were audited. The number appeared to be in line with the average across all professions,[17] showing that AIMA members were not particularly likely to attract the attention of Medicare and the PSR.

2.16      The question of whether the auditing methodology is appropriate is more complex.  The committee accepts the assumptions that premise the methodology, in that by focussing attention on the statistical outliers it is more likely to identify inappropriate practitioners.  It notes the admission by Dr Webber and others[18] that 'a low-volume, inappropriately practicing doctor would probably not be identified using the Medicare auditing methodology',[19] however given the scale of the data administered by Medicare it concedes that there are going to be practitioners who slip through the net.[20] The committee accepts Medicare's contention that it is using advanced data mining and analysis. It encourages Medicare to continue to try and develop techniques that have the capacity to uncover inappropriate MBS item use and PBS prescribing practices other than those identified as statistical outliers using current methods.  

Recommendation 1

2.17      The committee emphasises the importance of communicating the methodology utilised by Medicare Australia to the wider medical community.  The committee recommends that Medicare Australia publish its current auditing methodology and any subsequent improvements to the methodology as they come on stream.

2.18      The committee is not convinced by the ADU's suggestion that sophisticated audit profiling does not take place. It notes that the use of unfounded allegations in relation to over-servicing of pap smears was unhelpful to the conduct of the inquiry.   

Medicare Education and Advisory processes

2.19      The issue of whether practitioners have access to information, education and advice over the interpretation and application of MBS items arose during the inquiry. 

2.20      There were suggestions made in evidence that some MBS items are difficult to interpret and therefore adhere to.  Dr Masters from the ADU recounted a situation following the establishment of his multidisciplinary clinic: 

I was concerned to get the Medicare numbers right for this clinic. They are not straightforward. So I sent quite a lot of information to Medicare asking for help. I said: 'Are these odd numbers right? Is what I am going to charge right?' It took months to get a reply. I got a reply saying: 'We cannot give you an answer, Dr Masters. We suggest you contact the AMA and the college of GPs.' I contacted the AMA and the college of GPs—and I think I put this in my submission—and they said: 'We are not here to interpret the Medicare schedule. That should be done by Medicare.' Medicare will not do it. The PSR will not do it. The AMA will not do it. The college of GPs will not do it. And we get fined.[21]

2.21      MDA National also provided evidence regarding the interpretation of some MBS items:

We believe that there should be greater consultation with the profession, including the relevant colleges, in developing MBS item descriptors and the associated explanatory notes. In developing MBS item descriptors and the associated notes, we feel that feedback should be actively sought from these groups and the PSR where problems are identified. We believe that improved processes should also be put in place to enable individual practitioners to obtain clarity about the use of specific MBS items.[22]

2.22      Mr Dahm from the ADU also discussed the MBS Item interpretation, and made an interesting point on how the Australian Tax Office issues rulings and interpretations for complex situations.

...many practice managers report an inconsistency in verbal advice provided to them by employees of Medicare with very little reference to any written rulings. Yet they stand accused for misunderstanding the said rules or interpretations that have not been published or circulated widely to practices and even their own Medicare advisers...

The Australian Tax Office issue public rulings on a variety of tax matters, especially matters considered ambiguous and at a high risk of misinterpretation or perceived fraud. We find it unusual that given the apparent high incidence of fraud in the medical profession that similar rulings and interpretations are not reported by Medicare and are not included on the Governments website or in various education programs.[23]

2.23      The committee was concerned that there may not be sufficient educative and advisory processes in place to provide practitioners with confidence in their interpretations of MBS Item Descriptors.  It put the issue to AMA, DoHA and also to Medicare.

2.24      The AMA concurred that there was an issue around interpretation and outlined the steps they took on behalf on their members:

We have made strong representations to Medicare and any doctor, any person who bills Medicare now, can ask for a written interpretation, which they can be expected to receive and hold. Once they have got that written interpretation, I would expect that they would be able to submit that as a piece of evidence if they were called into question. I am not sure of the timing of that, but that is the case now.[24]

2.25      DoHA were concerned by suggestions that people were not given appropriate information from either the Department of Health and Ageing, or Medicare.  They advised that:

There are a number of avenues that are available to practitioners to receive advice about the interpretation of Medicare Benefits Schedule (MBS) items. The Department of Human Services (DHS) maintains a dedicated Provider Enquiry line that provides advice to practitioners and practitioners can also call the Department.[25]

2.26      The response from DoHA is consistent with the information the committee received from Medicare in response to specific questions on this issue.  Medicare said it provides information in the following manner:

The Department of Human Services - Medicare program has available formalised education resources, which include a range of self paced eLearning programs available online 24 hours a day. These programs include:

• Medicare and You – for new health professionals;
• Medicare and You – MBS primary care items;
• Medicare and You – for Dentists;
• Medicare and You – treatment for skin lesions; and
• Medicare and You – Chronic Disease Management for GPs.[26]

2.27      It also outlined how it communicates information on particularly complex MBS items:

Where the Department of Human Services – Medicare program is aware of ambiguity or difficulty in using an MBS item, it works with the Department of Health and Ageing to clarify the issue.

Once the issue is clarified it is communicated to health professionals. This can be directly to an individual, through professional associations, newsletters, quick reference guides and fact sheets.

2.28      Importantly however Medicare does not provide advice on the appropriate clinical practice for specific MBS items:

All clinical decisions are for the professional judgement of the medical practitioner.[27]

2.29      Dr Cootes, Acting Director of the PSR, advised the committee:

Around that whole issue of the items and people complaining that they are not sure where to place an item or how to itemise a particular consultation, there is a need for a little bit of realism here. Doctors are professionals; they have certain responsibilities put on them. The AMA Code of Ethics implores them to be a little bit sensible in their use of the community's resources. The new code of conduct produced by the Medical Board of Australia makes the same point even more strongly. I do not know that you can ever codify these things. Just in the GP domain, I think the total number is 120 million. There are 120 million interactions between a GP and an Australian each year, and if anyone can codify those unambiguously into four brief descriptions then good luck to them. So it falls back on the professional responsibility of the GP, and the system in the main does work effectively.[28]

2.30      While the committee agrees that it is the practitioner's responsibility to make clinical judgements and decisions in relation to MBS items, we are of the view that as much advice and information as possible should be as accessible to the practitioners. The production of quick reference guides and factsheets are particularly useful and the committee commends Medicare on this initiative.  The committee suggests that the department, in consultation with practitioner representative bodies keep a watching brief on the accessibility and currency of information sources.

Navigation: Previous Page | Contents | Next Page