R v Sharma & Agrawal[2017] QCA 209

SUPREME COURT OF QUEENSLAND

1. MORRISON JA:  I have read the reasons of McMeekin J and agree with those reasons and the orders his Honour proposes.
2. PHILIPPIDES JA:  I agree with McMeekin J.
3. McMEEKIN J:  The appellants, Anil Sharma and Priya Agrawal, were jointly charged with one count of fraud to the value of $30,000 or more (count 1).  Agrawal was also charged with one count of attempted fraud (count 2).  Following a seven day trial before a judge of the District Court and a jury, the appellants were each convicted of the charges that they faced.
4. The charges were brought pursuant to ss 408C(1)(c) and 408C(2)(d) of the Criminal Code.  In relation to count one the prosecution relied on the party provisions – ss 7 and 8 of the Criminal Code.
5. The appellants were not represented at trial but are represented on appeal.

THE CONVICTION APPEAL

The grounds of appeal

6. The original ground alleged in each of the respective notices of appeal that the verdicts were unreasonable were abandoned by each appellant.
7. Leave was sought by Sharma, and granted, to argue the following two grounds:
   1. The learned trial judge’s failure to direct the jury as to the drawing of inferences from circumstantial evidence resulted in a miscarriage of justice;
   2. The learned trial judge erred in directing the jury as to the weight of the expert evidence with respect to the handwriting on the HICAPS receipts, thereby occasioning a miscarriage of justice.
8. Agrawal had filed a timely appeal against conviction but subsequently abandoned that appeal.  She sought and obtained leave to reinstate that appeal and to argue the first of the grounds urged by Sharma.  Agrawal also applies for leave to appeal her sentence on the ground that it is manifestly excessive.  She was sentenced to six years imprisonment on count one.

Background Facts

9. The appellants were each directors of Lifeline Healthcare Pty Ltd trading as Lifeline Dental, obviously a dental practice.  Agrawal was a qualified dentist.  Sharma was the manager of the practice.  He was involved in administration and reception duties.  Sharma was a qualified dentist in India before coming to Australia but had not successfully completed the requirements to become qualified locally as a dentist.  No-one else was involved in the practice.  The parties were in a de facto relationship for some unidentified time, at least according to their respective cross-examinations of a witness.  Sharma suggested the relationship had ended by June 2011.^[1]
10. The prosecution case on count one concerned 1,915 dental treatments allegedly carried out on one Kamla Sharma, the mother of the appellant Anil Sharma, over the period from 9 June to 14 November 2011.^[2]  The alleged treatments included but were not limited to 1,104 five surface fillings (an uncommon form of filling in which all five surfaces of the one tooth need filling, the evidence being that one can have a one, two, three, four, or five surface filling), 7 four surface fillings, 80 three surface fillings, 34 two surface fillings, 108 one surface fillings, 14 teeth extractions, 17 bleaches, seven fluoride treatments, 72 recontour fillings, 181 pulp tests, and 260 radiographs.^[3]
11. Claims were submitted to Medibank Private Limited (the well-known health insurer) totalling $376,725.  Of that sum a total of $177,900.90 was paid out to Lifeline Healthcare.^[4]  The prosecution case was that many of the alleged treatments were either excessive, unreasonable, or in some cases impossible, an example being that treatments were claimed to have been carried out on teeth previously extracted.
12. The evidence led from a forensic odontologist at the trial was that the benefit to the appellants from treatments that fell into that category of excessive, unreasonable or impossible was $169,480.50.^[5]  A forensic accountant analysed the deposits made to the practice account.  The monthly totals of the deposits^[6] that on the prosecution case related to claims that were excessive, unreasonable, or impossible were:

13. The amounts claimed that were said to fall into the category of reasonable treatments was, by contrast, $8,420.40.^[7]  It should be noted that the categorisation does not mean that the treatments in fact occurred.  It meant that the odontologist was not prepared to say, simply from his examination of the records, that the various treatments were “excessive, unreasonable, or impossible”.
14. In respect of count one there was no contest at the trial that because of claims lodged from the practice, monies had been deposited into Lifeline Healthcare’s bank accounts by Medibank Private.  The facts established that, apart from junior assistants, this was a two person practice.  The appellants were the only two persons in a position to have made the claims.  Each of the appellants argued at trial that it was the other who had perpetrated the fraud.
15. The prosecution case on count 2 was that Agrawal had attempted to obtain payment in December 2011 for dental work allegedly performed, but not actually performed, on her own parents.  Because of the suspicions of the insurance company, monies were not paid out.

Ground 1

16. The first complaint is that the trial judge did not, in terms, give the jury the standard circumstantial evidence direction.  It is submitted that this was a circumstantial case and the direction was essential to avoid injustice.  Reliance is placed on Shepherd v The Queen (1990) 170 CLR 573 at 578; R v Perera [1986] 1 Qd R 211 at 217.  The direction contended for would be along these lines:

Circumstantial evidence is evidence of circumstances which can be relied upon not as proving a fact directly but instead as pointing to its existence.  It differs from direct evidence, which tends to prove a fact directly:  typically, when the witness testifies about something which that witness personally saw, or heard.  Both direct and circumstantial evidence are to be considered.

To bring in a verdict of guilty based entirely or substantially upon circumstantial evidence, it is necessary that guilt should not only be a rational inference but also that it should be the only rational inference that could be drawn from the circumstances.

If there is any reasonable possibility consistent with innocence, it is your duty to find the defendant not guilty.  This follows from the requirement that guilt must be established beyond reasonable doubt.^[8]

17. This was a circumstantial case and the direction would usually be given.  But here the cogency of the complaint depends on what was in issue at the trial.  The short answer to the complaint is that if it is not possible to frame a reasonable hypothesis consistent with innocence, there was no point to the direction.  Rather, it would have needlessly complicated a simple case.
18. The elements of the offence created by s 408C(1)(c) and 408C(2)(d) of the Criminal Code are:
    1. That the relevant defendant had induced another (here Medibank Private) to deliver property (here the deposits of money to a bank account) to any person (here Lifeline Healthcare);
    2. The action was done dishonestly;
    3. The amount induced to be paid exceeded $30,000.
19. After describing the distinction between evidence that directly proved a fact and inferences drawn from facts proved, his Honour gave the usual direction that before convicting, the jury must be satisfied beyond reasonable doubt of each element of the offence.  His Honour explained the elements of the offence accurately.  His Honour also explained the various possible ways the prosecution said that each was involved under the party provisions of the Code and there is no complaint about that summary.  Was anything more required?
20. The only issue that either defendant agitated before the jury was whether the defendant in question was involved in what was an obvious fraud, whether as principal or, under the party provisions, by counselling or procuring or aiding the other to commit the fraud.  There was no challenge mounted at trial to the prosecution evidence regarding the elements that I have identified as (b) and (c).  That is, it was accepted implicitly that whoever had done this was involved in an obvious fraud.  Each in effect said that the compelling evidence pointed to the other.
21. To put the arguments in context I will detail some of that evidence but the relevant point is that the issues were very limited and that is of considerable significance.

How was the fraud carried out?

22. The following applies to count one.  Sharma’s mother had a policy with Medibank Private.  It was called “Visitors Health Insurance” and was a “Top 85” cover.  That cover entitled her to 85% of the scheduled fee in respect of any treatment without a cap.^[9]  The policy commenced on 21 May 2011.^[10]  To effect the fraud several things were necessary – two of which were knowledge of the existence of such a cover and access to the mother’s benefit card.  To gain access to the insurance fund through the HICAPS system, it was necessary to swipe the benefit card held by the member (ie Sharma’s mother’s card) on a machine at the practice, as well as sign the HICAPS form.^[11]  It was necessary to use the dentist’s provider number, in this case a number personal to Agrawal.^[12]  This swiping must necessarily have been done by a person familiar with item numbers and the dentist’s provider number.  The monies were paid by Medibank Private directly into the practice account.

The scale of the fraud

23. The extent and nature of the fraud is a relevant fact.  To give some idea of the nature of the evidence I will mention a few matters.  Again this applies to count one.  The prosecution evidence commences on the 9^th June 2011 and there are suspicious treatments then.  For example the first two treatment days for which claims are made involve the removal of the mother’s four front teeth but dentures later prepared do not include dentures for these supposedly extracted teeth.^[13]  Nor does there appear any record of these attendances in the appointments book, the first appointment being nearly two weeks later.  There are consent forms for the extraction of teeth in the practice records.  These however postdate these alleged extractions by five months.  The clinical notes start on the 22^nd June and record extensive problems but a claim was made for a comprehensive check-up at the supposed appointment two weeks before but without a clinical note being made.  There is little doubt that the fraud commenced in June.
24. There are suspicious treatments in June, July and August but take the period starting on 12 September.  It is relevant to a submission made by Sharma to the jury.  The expert odontologist Mr Forrest, an Associate Professor at Griffith University, explained that the claims showed that on that date various surfaces of the patients’ teeth were filled despite the fact they had been previously filled.  Further radiographs were also taken despite the fact that there had been 14 previous radiographs.  The evidence, not unexpectedly, was that the principle dentists apply when deciding whether to use radiation was summarised in an acronym ALRARA – as little as possible, as little as reasonably achievable.^[14]  Then on 13 September the appointment book showed a half hour appointment^[15] and the following treatment was claimed as performed^[16]:

• A crown was removed from 1-6 and 2-7 but without the usual work to cover up the damage thereby exposed;^[17]
• Various fillings were performed on 2-4, 2-8, 3-8, 3-7, 3-6, 3-5, 3-4 and 2-5;
• Surface fillings are performed on 4-4 and 4-5 replacing previous fillings (making ten fillings in all that day);
• A scale and clean was undertaken;
• A pulp test performed;
• A mouth photo taken;
• A face photo taken;
• Study models done;
• Letter of referral;
• A desensitisation undertaken;
• Topical fluoride applied;
• Four further radiographs performed.

25. Professor Forrest when asked whether that amount of treatment was realistically feasible in half an hour replied “absolutely not”.^[18]  He said that a scale and clean and filling alone would take a half hour.
26. The following day (the 14^th) there were a further 19 fillings (so making 29 fillings in two days) as well as a pulp test and topical fluoride applied as well as eight radiographs taken.  The repeat procedures were said to be unnecessary and the amount of treatment in the time not possible.^[19]
27. On the following day (the 15^th) there was an appointment for one hour.  Four radiographs were claimed and 11 fillings (now 40 fillings in three days), seven surfaces having been filled previously.  Again the expert gave unchallenged evidence that the treatment was in the doubtful category.^[20]
28. This brings us to the 16^th September.  On that day four radiographs are claimed as performed and 22 one surface fillings (62 fillings in four days) all within a one hour appointment.^[21]  Apart from the apparently needless radiographs, and the impossibility of performing such extensive work in an hour, and apart from the staggering number of fillings, some were performed on teeth supposedly previously extracted.  I have mentioned the supposed extraction of the four front teeth above.  According to the claims made on 9 June a tooth identified as 1-1 was extracted.^[22]  On 14 June teeth identified as 2-1, 3-1 and 4-1 were shown as extracted.  These numbers identify the two front upper and two front lower teeth.  They were all extracted by mid-June.  Yet on 16 September these teeth are each shown as being subjected to a filling.^[23]
29. The events of 14 November again provide compelling evidence of fraud.  On that day Sharma and his mother left Australia.  Immigration cards show that they passed through the immigration checkpoint at the airport at around midday.  Claims are made for dental treatment on that day involving six radiographs, 10 pulp tests, and 32 five-surface fillings.  Mr Forrest said that the work, if performed, would take several hours, he making clear that a responsible dentist would not attempt so many fillings in the one day due to the need for care in not disrupting biting patterns and joints.  No one would contemplate such treatment on the day they were to undertake a long international flight.
30. There are no claims for treatment for Sharma’s mother after this day.

The case against Sharma

31. The most cogent evidence against Sharma is the presence of his signature on the 1,915 claim forms.  As his Honour pointed out it is not simply that treatment occurred and that claims were lodged.  Initially the lodgements might be innocent enough and might be explicable, although given the nature of the claims that much is doubtful.  It is the cumulative nature of the treatments that must be brought into account.
32. Once it be accepted that it was his signature on the documents then it is not possible to conceive of a reasonable hypothesis consistent with innocence.  None was put to the jury by Sharma and none was advanced before this Court.  True it is that a finding of guilt involves inferential reasoning to arrive at a conclusion of dishonesty but those inferences inevitably followed from proof of the fact of his signature appearing on the claim forms for treatment that could not possibly have been needed or, in many cases, performed.
33. The evidence of the handwriting expert establishing that the signatures on the 1,915 claim forms were Sharma’s and not a forgery was expressed in terms of probability and not certainty.  I will detail the evidence in discussing ground 2.  But it was not the only fact proved implicating Sharma.  The jury were required of course to look at all relevant facts.  Again, even leaving out the signatures on 1,915 forms, it is not possible to conceive of a reasonable hypothesis consistent with innocence that accounts for all the facts.
34. The prosecution case that Sharma was involved as principal, or as aider, counsellor or procurer, leaving to one side the allegation that it was his signature on the 1,915 HICAP claims/receipts lodged, relied on proof of the following facts:
    1. He was himself a qualified dentist in India, albeit not yet qualified in Australia, but presumably familiar with dental treatments and practise;
    2. It was his signature on documents held in the practice records completed as a guardian for his mother, being a medical questionnaire and tooth extraction consent of forms;
    3. He performed the role of the business manager and was able to and did access the bank account of the practice;
    4. The practice bank account showed the deposits obtained from the fraudulent HICAPS claims;
    5. The scale of the fraud was such that the monies thereby obtained and which were banked to the practice account were bound to be noticed and well in excess of that experienced before the frauds commenced;
    6. Despite the business being a marginally loss making one, he withdrew monies from the account including a withdrawal of $20,000 which was indicative of two things – his knowledge of the state of the accounts and confidence that the practice could cope without the monies;
    7. There were a number of withdrawals of significant sums to his credit over the months;
    8. As practice manager he would have access to the records of the practice, and to the machine by which claims were made;
    9. The clinical records of the practice recorded treatments that were plainly excessive, unreasonable or impossible;
    10. To perpetrate the fraud the perpetrator had to have access to his mother’s Medibank benefit card;
    11. He was the one authorised to make claims on his mother’s behalf – the evidence was that she was unable to speak English;
    12. The fraudulent claims ended on the day he and his mother left the country.
35. Even if one assumes in Sharma’s favour that he was unaware of the lodging of the claims – that this was Agrawal’s doing – he remains liable as a party to the offence if he was aware that monies were being obtained fraudulently yet used the money so obtained within the practice, or took the benefit of the monies personally, or with that knowledge continued to permit the business equipment to be used to process the false claims, or continued to permit his mother’s benefit card to be used to make the claims, or completed forms to pretend that the supposed treatment was requested by his mother.  I will set out why, in my view, the jury were quite entitled to find that Sharma was involved in these various ways.
36. The evidence showed that Sharma acted as the business manager of the practice.  He assisted an independent accountant in the preparation of Business Activity Statements in each of the quarters prior to the October BAS.  The practice was running at a loss until the monies were received from the over servicing.^[24]  One would expect that the business manager, and one of the only two directors of the company and one of two signatories to the account, would be acutely aware of the state of finances of any business, let alone a marginal business.  As the fraudulent claims were paid Lifeline Health’s account commenced to show a healthy credit balance.
37. That Sharma was well aware of the state of the finances is quite clear.  Substantial monies from Lifeline Health - $20,000 - were transferred to Sharma’s personal account on 4 July 2011.  That same day an amount of $31,990 was transferred into the account, apparently from the Australian Taxation Office.  The significance of this is not only that a very significant sum went to Sharma personally but more importantly that Sharma was quite evidently well aware of the state of the finances.  The money simply was not there to be drawn the day before.  And the taking out of such a large sum suggests some confidence that the business could afford to get by without those monies.  That apparent confidence was not misplaced.  It was not a confidence that could have been engendered by the consistent but modest earnings (leading to an overall small loss) shown in the two quarterly BAS statements that preceded the payment and that Sharma had assisted in preparing.
38. There were further transfers to Sharma’s personal Citibank account on 12 October and 14 November of $6,000 and $5,000 respectively^[25] and the bulk of these monies were transferred overseas.  Substantial sums were used to keep the practice going eg $40,600 to one item of expense,^[26] $25,000 of which occurred before Sharma left to go overseas.
39. Other pieces of evidence also pointed to Sharma’s involvement.  They are less compelling taken individually but relevant.
40. The evidence was silent as to who had possession of the mother’s benefit card used to effect the transactions.  In the normal course one would expect that if the card was not in the mother’s possession then, in the circumstances pertaining, her son and purported guardian would be in possession of the mother’s private health insurance card, not her treating dentist.  There is no evidence that Agrawal was in a relationship with Sharma’s mother that would suggest that she would have possession of such a card.  There is the further issue that it is difficult to see how Agrawal would have the card over an extended period (for five months) or regularly from time to time without Sharma’s knowledge.  There is no evidence that she did access the card or had any opportunity to do so.
41. The claims concerning Sharma’s mother ceased on 14 November, the day Sharma and his mother left Australia at about midday.  The cessation of the use of the card that day is consistent with Sharma taking it with him.  It might also be consistent with Agrawal appreciating that making a claim for treatment on someone who was out of the country at the time was bound to be discovered if there was any investigation.  But the same comment could be made about the whole fraud as the evidence of the forensic odontologist demonstrated.
42. Further evidence that Sharma was involved is in his signing of various documents related to the treatment which formed part of the records of the business.  He signed a medical questionnaire form on behalf of his mother.  He signed four “Consent for Extraction of Teeth” forms for his mother bearing dates 5, 7, 8 and 9 November 2011 respectively involving collectively the extraction of 17 teeth.^[27]  The timing is curious.  Sharma and his mother left the country on 14 November.  Normally one does not sign a consent for treatment unless one is contemplating undergoing it.  It is hardly credible that one would contemplate having 17 teeth extracted in the week before one was to fly to India.  I put to one side the fact recorded in the clinical records that Sharma’s mother was suffering from pancreatic cancer and in receipt of chemotherapy.  If the imminent international flight was not in contemplation on the dates the forms bear then it suggests a very sudden decision to depart the country – supporting the inference that Sharma decided to flee which the prosecution and Agrawal urge (and which I ignore for present purposes).  Or that the forms bear false dates.  None of this helps Sharma.
43. The only evidence that went against the proposition that Sharma was involved was the evidence that pointed to Agrawal’s involvement.  But that did not serve to explain in any way why the compelling evidence against Sharma became less incriminating.
44. Taken together, as the facts must be, they are compelling.  Against these compelling features the matters relied on by Sharma are almost inconsequential.  His counsel summarised them as follows, with my responses following:
    1. The expensive treatments were from October on – that is not accurate. See the summary of deposits and the analysis of the five days in September above.  And there was much more.  The more egregious claims did occur later but the claims made in September (and indeed earlier) were indefensible and obviously fraudulent;
    2. Agrawal was responsible for forging a letter purporting to be a confession by him – true and a possible indicator of her guilt.  But that does not exonerate Sharma.  This evidence incidentally serves to strengthen the evidence that it was Sharma’s signature on the claim forms.  If Agrawal was capable of forging a fluent signature on 1,915 claim forms sufficient to fool the expert and over a period of five months why was the signature on the letter so evidently different?;
    3. Agrawal acted suspiciously in relation to a search of the premises – perhaps true to an extent but of doubtful relevance even in Agrawal’s case;
    4. Agrawal used some of the funds – again true and again an indicator of Agrawal’s guilt.  But the reasoning applies equally to Sharma;
    5. The accountant gave evidence that Sharma was more involved with the preparation of the BAS statements before the quarter in which the huge spike in income occurred – again true.  But the evidence is all but irrelevant when one considers Sharma’s inevitable prior knowledge of the state of the accounts.
45. There is no reasonable hypothesis consistent with innocence that I can conceive of that takes account of all the facts.

The case against Agrawal

46. A useful starting point in the case against Agrawal is count 2, the attempted fraud.  Sharma was out of the country at the time the impugned claims were made.  There is no evidence or even suggestion that he has anything to do with the facts underlying this count.
47. The count concerns treatment allegedly given to Agrawal’s parents in December 2011.  The claims made are for attendances on six days.  The claims made for one parent are effectively mirrored by the claims made for the other.  For her mother there were 288 treatments involving 51 radiographs with 187 “five surface fillings”.  Five surface fillings are not common.  On one day, the 6^th, 29 such fillings were claimed as being performed – ie 29 of the 32 teeth were supposedly subjected to five surface fillings.^[28]  Fifty-one radiographs in six days (between the 6^th and the 12^th) is inexplicable.  Apart from the extraordinary improbability of a need for such treatment there are practical problems.
48. The expert called said that with such multiple treatments on the one day, if ever necessary and that would be very rare, they would need to be done under general anaesthetic.^[29]  There was no suggestion of that here.  The claims made suggest treatment in the dental surgery as usual.
49. The expert said, and he was not challenged, that a five surface filling would take at least 30 minutes and more probably 45 minutes.^[30]  The appointment book shows Agrawal’s mother as attending on the 6^th for one 45 minute appointment.  Other patients are shown as having been treated that day.  On the same day Agrawal’s father had 28 such fillings.^[31]  So to complete the 47 fillings allegedly performed on the mother and the father would take in the order of at least 23 and a half hours.
50. Any filling impacts on the dynamics of the structure of the mouth.  A person’s mouth is adjusted to closing to a certain plane – too high a filling will mean pain or potentially so on closing because of sudden impact with the tooth above or below.^[32]  Joints can be effected.  So in the usual course care is taken to see how a filling goes before embarking on others where they might have an effect.  For Agrawal to ignore the dictates of prudent care on her own parents emphasises the improbability of the treatments.
51. The anaesthetic required to perform such multiple treatments would involve multiple injections with the numbing of the entire mouth and tongue.  The amount of fluid and the discomfort would be quite out of the ordinary.
52. The next day (the 7^th) Agrawal’s mother again had a five surface filling on a tooth subjected to the same treatment the day before.  As well she had two five surface fillings on the same tooth on the same day.  The expert was unable to conceive of a circumstance where that could be necessary.^[33]
53. And so the evidence continued for a further four days of improbable treatments.  The forensic odontologist said that he could not imagine the scenario in which the treatment allegedly performed would be expected treatment in six days.  His opinion was not challenged.  The point is that the attempted fraud is undeniable.  The question is who perpetrated it?
54. Following the claimed, treatments a statement of account on the practice letterhead was prepared for each patient.  Who prepared the statements of account?  The dates of treatment matched the dates on which appointments had been written into the appointments book.  Who did that?  It is certain that it cannot be Sharma as he is out of the country.  So the perpetrator must be someone with dental knowledge as the claims made are detailed, and someone with access to the practice documents.  There is no evidence (or suggestion from Agrawal) of any other person being employed in the practice who was in a position to bring these records into being.
55. The scheme used on this occasion by whoever was the perpetrator was to prepare a statement of account showing the treatments as having been performed, and showing that the account had been paid in full with a receipt acknowledging payment.  A claim was then made on the insurer for a refund.  Who is to profit from all this?  Who, other than Agrawal could have prepared the documents?  The insurer in the meantime had grown suspicious and stopped any payment being made.
56. Various statements of account were prepared for each day of treatment including one on the 12^th December 2011 for services rendered on that date.^[34]  Agrawal made four phone calls to the offices of Medibank Private Limited through this period.  In those calls she pretended to be her mother.  She enquired about payment of the claims.^[35]  She also attended at the insurer’s office on 12 December enquiring as to why monies the subject of the statements of account (including that of the 12^th) had not been paid to her parents.^[36]  She told the officer of the insurance company that she was enquiring in her capacity as the daughter of the patients and expressly disclaimed, when asked, that she was the service provider.^[37]  This assertion of the officer was not challenged in cross examination let alone by evidence.  When later questioned by police she initially denied that she had made the calls but eventually admitted the pretext.
57. If the treatments had never taken place and if the statements of account and alleged receipt of monies were therefore false and forged by someone other than Agrawal how is it that Agrawal knew of them to make the calls?  And why deny she was the service provider when making a legitimate enquiry on behalf of one’s parents?
58. There is no reasonable hypothesis consistent with innocence that I can conceive of and that takes account of all these facts.  There was a concession by counsel that he could not.  The starting premise then is that Agrawal committed the attempted fraud the subject of count 2.
59. Turning to count one - if one starts with that premise of guilt in respect of count 2 then there is quite an extraordinary coincidence.  Without involvement in Sharma’s frauds, and presumably on this premise without knowledge of them, Agrawal has independently decided to defraud the insurer herself, and to do so in a similar way by pretending to treat family members.  Two fraudsters in a two person practice but acting independently may not be impossible but the notion stretches credulity.
60. The most cogent evidence implicating Agrawal in count one is again the benefit obtained.  A loss making practice suddenly is not.  The profits are extraordinary by comparison to what has gone before.  The BAS statements show that in each of the three preceding quarters the practice is grossing a little under $40,000.^[38]  In the period from October to December the gross is shown as exceeding $156,000.^[39]  The HICAP system automatically deposits monies to the practice account usually within a day or two of the benefit being claimed.  So there was a steady crediting of monies to the practice account.  Agrawal, one of the two directors of the company and one of the two signatories to the account, must on this premise of innocence not notice that suddenly and inexplicably her gross income has expanded significantly.  That defies belief.
61. Adopting the hypothesis that Agrawal was initially ignorant of the claims being lodged by Sharma how did she remain ignorant?  The claims go on for five months.  Significant sums are taken out of the practice account and transferred to her account.  The hypothesis of innocence requires that she remains unaware of her expanding and inexplicable wealth.  Any enquiry would have quickly revealed the true situation.
62. There is the further feature that to make the claims Sharma had to use Agrawal’s provider number.  Agrawal told the investigating officers that there was an “access user limitation” and Sharma did not know her access.^[40]  The conclusion that she must have at least permitted this access to occur in respect of treatments not performed from some point is inescapable.
63. That Agrawal was well aware of her new found wealth is suggested by what she did.  In November 2011 Agrawal determined to enter into a contract to purchase land and dwelling for $500,000.^[41]  A deposit of $1,000 was paid on 25 November 2011.^[42]  On 6 December Agrawal paid a deposit of $19,000.^[43]  Each sum was drawn from the practice account.  It will be recalled that Sharma left the country on 14 November.  The inescapable inference, as one would expect, is that Agrawal was familiar with the state of the accounts.  Six months before her practice was losing money.  On 8 June, and so the day before the prosecution alleges that the suspicious treatments commenced, there was a credit balance of $5,802 in the practice account.^[44]  On 25 November the balance was in excess of $120,000.^[45]
64. Further evidence implicating Agrawal is found in the clinical records.  The clinical records contain a detailed statement of the various treatments performed.  In the normal course one would expect the dentist to make the clinical notes, not the practice manager.  On this premise of innocence Sharma has gained access to the records – perhaps a possibility – but falsified them without Agrawal noticing.  There are at least three problems with the hypothesis.  The first is that Agrawal made plain in her interview with police and in cross examinations that she claimed to have done some of the treatments.  She told the police officer that she had treated Sharma’s mother on between five and seven occasions “maybe less”.^[46]  If true, she presumably had to access the records of the patient then.  How did she not notice the extensive and detailed entries?  If not true, why lie?  A second problem is that the entries are generally in excellent English.  Sharma’s spoken English was nowhere near that level.  Agrawal’s spoken English was excellent.  The third problem is that the records make plain that it is Sharma who is responsible for authorising the treatment.  So on 22 June the record includes: “[the mother] does not speak english, communication through the son Anil Sharma.  All consents to be signed by Anil Sharma on behalf of [the mother].  Anil Sharma has accepted all responsibilities in relation to patient [the mother’s] treatment and payments.”^[47]  One suspects that if Sharma was doing the falsifying he would minimise, not maximise, his involvement in blatantly fraudulent treatments.
65. I note too that Agrawal’s signature appears on the consent for extraction forms that I have mentioned above, immediately adjacent to Sharma’s signatures.  These forms were found in the clinical records.  It is difficult to conceive that these consents are anything but false documents.  They are consistent with Agrawal’s involvement in the fraudulent scheme in November.
66. I leave out of account the fact of the forged letter mentioned earlier implicating Sharma.  While Sharma was overseas and investigations commenced a letter was prepared in which Sharma purportedly admitted his responsibility for the claims that had been made in respect of his mother.  Agrawal claimed to have received the letter in the post and brought it to the attention of the police.  The handwriting expert thought the signature on the document was not Sharma’s but a forgery.  His opinion was not contested.  Who other than Agrawal had an interest in throwing suspicion on Sharma and forging a signature, presumably a signature that she knew well?  But this does not add greatly, if at all, to the prosecution case as the forgery could be explained by panic at a false accusation being made.
67. The innocent hypothesis advanced was that Sharma carried this out without any involvement from Agrawal.  The hypothesis does not explain any of the matters I have mentioned.

Principles relevant to the appeal against conviction

68. The overall question of course is whether the summing up taken as a whole has resulted in an injustice.  But subject to that there are only two relevant principles to bear in mind.
69. The first is that there is no rule of law or practice that a circumstantial direction must be given in every case where inferences are asked to be drawn from proved facts: see Shepherd v The Queen (1990) 170 CLR 573 at 578 where Dawson J (Mason CJ, Toohey and Gaudron JJ agreeing) said:

“The learned trial judge gave the customary direction that, where the jury relied upon circumstantial evidence, guilt should not only be a rational inference but should be the only rational inference that could be drawn from the circumstances: see Hodge’s Case (1838) 2 Lewin 227; 168 ER 1136; Peacock v The King (1911) 13 CLR 619; Plomp v The Queen (1963) 110 CLR 234.  Whilst a direction of that kind is customarily given in cases turning upon circumstantial evidence, it is no more than an amplification of the rule that the prosecution must prove its case beyond reasonable doubt.  In many, if not most, cases involving substantial circumstantial evidence, it will be a helpful direction.  In other cases, particularly where the amount of circumstantial evidence involved is slight, a direction in those terms may be confusing rather than helpful.  Sometimes such a direction may be necessary to enable the jury to go about their task properly.  But there is no invariable rule of practice, let alone rule of law, that the direction should be given in every case involving circumstantial evidence.  It will be for the trial judge in the first instance to determine whether it should be given. As Barwick CJ, speaking for the Court, observed in Grant v The Queen (1975) 11 ALR 503 at 504:

‘Where the circumstances of the case seem to require that some such direction be given, the summing up regarded as a whole may prove to be, and generally may be likely to be, inadequate.  On the other hand, having regard to the circumstances of the case and the nature of the summing up, the failure to give the special direction may not in a particular case result in an inadequacy of the summing up as a whole.  It may none the less be concluded from the terms of the summing up that the jury were fully instructed.’”

70. Counsel have helpfully drawn our attention to several decisions of this Court in which the need for a circumstantial evidence direction was considered.
71. The essential question is whether the summing up taken as a whole was inadequate: R v Sharp [2012] QCA 342.  So in R v Reynolds^[48] Morrison JA approached the issue by asking whether the giving of the direction would have made a difference.  Where there were several innocent hypotheses open he resolved that issue in favour of the appellant.  That is the approach that I perceive is pertinent here.  It is the approach adopted by de Jersey CJ in R v Dolley^[49] where the point relevant here was under consideration.  In response to a submission that the full circumstantial evidence direction should have been given de Jersey CJ said:

“The judge clearly instructed the jury that proof of the relevant awareness was an essential element of the charge to be established beyond reasonable doubt.  The direction concerning the need to exclude other rational hypotheses consistent with innocence is but a logical elaboration upon the Crown's obligation to establish guilt beyond reasonable doubt.

Certainly, that is an essential direction, where what is sought to be inferred, involves a matrix of facts and circumstances, as, for example, how or whether a murder has been committed in a case where no body has been found.  That is but one example, but in a case like this, where the fact to be inferred is itself but one element of the offence, the direction that, in order to convict, that fact must be inferred beyond reasonable doubt, adequately directs the jury to the test to be applied, because obviously, if the inference is drawn beyond reasonable doubt, then, ipso facto, all other reasonable possibilities must have been excluded.  In my view, the appeal against conviction should, for those reasons, be dismissed.” (my emphasis)

72. The second principle of significance here is that neither this Court nor the jury need to be concerned with mere conjecture.  The hypotheses that are of significance are those that are reasonable.  An hypothesis that ignores the facts is not a reasonable one.  The principles were usefully summarised by Brennan and Gaudron JJ in Knight v The King (1992) 175 CLR 495 at 509:

“In Barca v The Queen^[50] Gibbs, Stephen and Mason JJ said: “When the case against an accused person rests substantially upon circumstantial evidence the jury cannot return a verdict of guilty unless the circumstances are ‘such as to be inconsistent with any reasonable hypothesis other than the guilt of the accused’: Peacock v The King.^[51]  To enable a jury to be satisfied beyond reasonable doubt of the guilt of the accused it is necessary not only that his guilt should be a rational inference but that it should be ‘the only rational inference that the circumstances would enable them to draw’: Plomp v The Queen^[52]; see also Thomas v The Queen^[53].  However, ‘an inference to be reasonable must rest upon something more than mere conjecture.  The bare possibility of innocence should not prevent a jury from finding the prisoner guilty, if the inference of guilt is the only inference open to reasonable men upon a consideration of all the facts in evidence.’ (Peacock v The King^[54].)  These principles are well settled in Australia.” (my emphasis)

Conclusion re Ground 1

73. There was only one element in contention: was the jury satisfied beyond reasonable doubt that the relevant defendant was involved in this fraud either as principal, procurer, counsellor, aider or abettor?  No attempt was made by either appellant to formulate a reasonable hypothesis consistent with innocence and consistent with all the features of the evidence.  If no such hypothesis can be formulated there was little point to a direction that the jury exclude such a hypothesis before finding the relevant defendant guilty.
74. The usual direction that before convicting the jury must be satisfied beyond reasonable doubt of each element of the offence was quite sufficient in the circumstances.  The only question was: are you satisfied beyond reasonable doubt that this defendant was involved in this dishonesty in one of the ways characterised by the prosecution?  To say more would only confuse a simple case.
75. Before leaving the case it is worth noting that the only two people who knew the true state of affairs chose not to reveal what they knew.  I have not brought that into account in my analysis.  But I have no doubt that it is a relevant matter: Weissensteiner v The Queen (1993) 178 CLR 217; R v Neilan [1992] 1 VR 57.  In the former case Mason CJ, Deane and Dawson JJ said at 227-228:

“In particular, in a criminal trial, hypotheses consistent with innocence may cease to be rational or reasonable in the absence of evidence to support them when that evidence, if it exists at all, must be within the knowledge of the accused.”

76. If there was any evidence capable of supporting an innocent hypothesis then it was within the knowledge of the appellants and they chose not to reveal it.  What that evidence might be I cannot conceive.  There is no merit in this ground.

Ground 2 - Evidence of Lau

77. This ground is argued by Sharma alone.  It concerns the evidence that the signatures on the HICAP receipts was that of Sharma.  The evidence that it was Sharma’s signature, and so not that of Agrawal, on the 1,915 forms came from a witness claiming expertise in the study of handwriting, Mr Lau.  There was no challenge to the witness’ expertise.
78. There was no cross examination of Mr Lau and no evidence was given by either appellant.  Mr Lau’s unchallenged opinion, based on a detailed comparison, was that it was “more likely” that the signature matched the specimen that was shown to be Sharma’s signature and that this was “a strong opinion”.  Nor did Sharma submit to the jury that they should disregard the expert’s opinion.  Indeed he expressly relied on it to assert that Agrawal had forged his signature on a letter in which he purportedly confessed to having made the fraudulent claims.
79. English is evidently not Mr Lau’s first language.  Mr Lau had charts and specimens on display for the jury as he gave his evidence.  After much explaining of the methodology Mr Lau expressed his opinions in the following passage:

“Now, having a back slant signature, having the character A done a similar fashion, connection done a similar fashion, and the line density of the L, and the construction, and the skill, and the rhythm, all in combination – therefore, I am able to form opinion – an indication.  Because, I said before, there are limitations: (a) it is a short signature, and (b) I don’t have the similar type of initial to compare like with like.  That was the basis I’m able to form a stronger opinion than what I had before on the previous two signatures.  I can only with limitation consider the combination of the similarity features, and [indistinct] specimen signature to exactly compare like with like.  That was the reason I’m able to form a strong opinion, other than the [indistinct] opinion that you could say it’s more likely has been done by the writer of the specimen signatures.

Okay.  So just to summarise that, when you considered the combination of features you outlined your conclusion was that it was more likely that the person who did the specimen signatures did those signatures as well, and the limitation is because you were unable to compare like-type documents with like.  Is that right?---That’s correct.”^[55]

80. To understand the significance of the statements that I have highlighted it is necessary to quote an earlier passage where Mr Lau explained the hierarchy of the levels of opinion:

“Now, the opinion we have in the – in all the government laboratory we have different level opinion.  That means on either side of the inconclusive.  So when we compare specimen and question signature, we cannot form opinion, we call it inconclusive.  In between the one above, because there limitation, we form a qualified opinion that signature was written or not written by the person of the specimen-writer.  And then the highest level is done by the writer of the specimen.  That means there’s no limitations and there’s sufficient evidence to support the proposition that the question signature was written by the specimen-writer.  So they’re the two level above the inconclusive, two below the inconclusive.”^[56]

81. So the only level above “more likely” is the “highest level … done by the writer of the specimen”.  The limitations that prevented the opinion being at the level of absolute certainty were the shortness of the signature and the inability to compare like documents.
82. There was no cross examination of the expert and no submission made by Sharma that the jury should not accept his opinion, nor any submission as to why they might have reservations about it.  As counsel for the respondent points out Sharma did cross-examine some witnesses and did put propositions where he wished to make his point.  He was not ignorant of his rights.
83. Any person, no matter how dull, would appreciate the significance of their signature appearing on fraudulent claims.  There is no reason to think that Sharma was not of at least average intelligence.  He plainly made a deliberate decision not to put in issue the fact that his signature was said to appear on the forms.  Sharma gave no evidence contradicting that opinion, he of course being in the best position to know if the signature was his or not.  In fact what he did say to the jury, as best I can make out – his spoken English was not good:

“I was authorised person [indistinct] to sign Medibank slips.  Slips was presented to me from what I remember once or twice a week and my job was to ask my mother …”^[57]

The trial judge then intervened as Sharma was giving evidence from the Bar table.  What is evident is that Sharma was very much aware of the evidence concerning the “Medibank slips” and did not seek to advance any proposition that he was ignorant of them at the relevant time, or of the significance of his signature appearing on them.

84. Sharma now complains that the learned trial judge overstated the evidence of the handwriting expert Mr Lau in this passage of his summing up (with my emphasis):

“As I understand it, the Crown case is put on the basis that in an evidentiary sense, that is, established or tendered to be established against Mr Sharma, particularly by his signing of the HICAPS slips that are found in the house – and you’ll remember that the uncontradicted evidence of the handwriting expert is that they were his genuine signatures on those slips – involving him, so says the Crown, in the presentation of those claims to Medibank Private and, therefore, the processing of the claims.”^[58]

85. That passage needs to be read with the directions that the learned trial judge gave to the jury that they were not obliged to accept the opinions of an expert.  His Honour said:

“However, the fact we refer to such witnesses as being expert does not mean their evidence must be automatically accepted.  You, at the end of the day, are the sole judges of the facts and you’re entitled to assess and accept and reject any such opinion evidence as you see fit.  It is up to you to give such weight to the opinions of the expert witnesses as you think they should be given, having regard in each case for the qualifications of the witness and whether you thought them to be impartial or partial to either side and the extent to which their opinion accords with whatever other facts you find to be proved in the case.  This is a trial by jury not a trial by expert, so it is up to you to decide what weight or importance you give to their opinions or, indeed, whether you accept their opinion at all.

It is also important to remember that an expert’s opinion is based on what the expert witness has been told of the facts.  If those facts have not been established to your satisfaction, the expert’s opinion may be of little value at the end of the day.  However, I should say in the context of all of this, the thrust of which largely is to be careful about expert evidence, I suppose – however, in this case, the expert witnesses are uncontradicted and substantially unchallenged, really.  You may struggle to recall any particular challenges made to them in cross-examination, with one exception that I will come to.  Therefore, you ought not to reject the views expressed by the experts unless the matters on which the opinion is based have not been proved to your satisfaction or you consider there is other evidence which casts doubt on the expert’s conclusions.  And indeed, it seems the bulk of the expert evidence might not be under any great, significant challenge.”^[59]

86. Given the state of the evidence that I have described and of the matters put before the jury for their deliberation his Honour’s statement reflected the reality of the situation.  His directions made it clear that this was a question of fact for the jury.  The signatures on the claim forms and the specimen signature were there for them to form their own opinion.  They are remarkably alike and unlike the signature that was said to be a forgery.  There was no demonstrated reason as to why the signatures on the HICAP receipts should not be treated as Sharma’s genuine signatures.

Conclusion re Ground 2

87. It would not have availed Sharma one iota for his Honour to have detailed the evidence that it was the expert’s uncontradicted “strong opinion” that it was “more likely” that each of the 1,915 signatures was Sharma’s genuine signature and as to why he held that view.
88. There was no miscarriage of justice in the summary his Honour gave.

AGRAWAL’S SENTENCE APPLICATION

89. The application for leave to appeal the sentence may be shortly disposed of.
90. Counsel for Agrawal conceded that the sentence imposed was appropriate given the information before the Court below.  The success of the application depended on the introduction of new evidence namely a report by a psychiatrist.  In my view leave should not be given to adduce that evidence for several reasons:
    1. The report is based on an uncritical acceptance of Agrawal’s view of the facts.  For example the psychiatrist records: “the applicant…failed to adequately represent herself, including explaining the treatment plan for both her parents”.  And “[s]he was able to express an appropriate level of remorse and regret for engaging in the extensive dental work for her parents eg ‘I took it too far.  I wasn’t thinking straight.’”  I refer to the analysis of count 2 above.  There is not the slightest prospect that Agrawal in fact performed the work that she claimed for.  There was no evidence that she had done any of the work.  The factual basis for the psychiatrist’s assumptions is wrong.  There is no reason to accept Agrawal’s expressions of remorse and regret as true.  The psychiatrist thought those expressions to be relevant.
    2. The report sets out in considerable detail Agrawal’s personal history.  If true, it is an unfortunate one.  But the premise underlying the opinion is that it is true.  The prosecution have had no opportunity to test the accuracy of any of Agrawal’s claims.  There is no reason to have any confidence in what she claims.  To the extent that Agrawal descends to dealing with the facts that were before the Court her version is largely inaccurate.
    3. The psychiatrist assumes that Agrawal was an unwilling tool in the hands of an oppressive and controlling Sharma.  Agrawal told the psychiatrist that she had “limited knowledge of [Sharma’s] behaviour”.  The events after Sharma’s departure strongly suggest that was not so.  The jury’s verdict on count 2 is not consistent with that premise.  Sharma is entitled to be sentenced on the same factual assumptions as Agrawal where their actions interact.  It is reasonable to assume that Sharma would not accept the premise given that his defence involved a rejection of the claims made by Agrawal.
    4. The psychiatrist purports to reach a diagnosis of Agrawal’s psychological state as it was in 2011 and the effect on her motivations.  He diagnoses a post-traumatic stress disorder and persistent depressive condition.  The opinions were based on an interview conducted in January 2017.  It long post-dated the events in issue.  The personal history relied on concerned events in India that long pre-date the fraud the subject of the charge.  The time gaps alone invite scepticism of an accurate diagnosis.  Uncritical acceptance of the opinions in the report with the prosecution being denied the chance to test it would not be in the interests of justice.
    5. It would be very advantageous for this Court to have the benefit of the trial judge’s views of Agrawal’s functioning in the course of the trial.  His Honour was in a very good position to assess Agrawal’s personality, apparent psychological functioning and her interaction with Sharma after a seven day trial.  Admission of the report at this stage would deny this Court that opportunity.
91. It was not submitted, and it could not have been, that the sentence imposed was outside the exercise of a sound sentencing discretion.  There is no good reason shown to accept psychiatric evidence now.  The application for leave to appeal the sentence should be refused.

Orders

92. I propose the following orders:
    1. The appeals against conviction by each appellant be dismissed;
    2. Agrawal’s application for leave to appeal against sentence be refused.