1 This is a review of a continuing detention order made under the Dangerous Sexual Offenders Act 2006 (WA) (DSO Act) by Jenkins J on 24 September 2009: Director of Public Prosecutions (WA) v McGarry [No 2]

3 The application for leave to apply for a special review was not opposed by the Director of Public Prosecutions. On 25 August 2016, Jenkins J granted leave to apply for a special review and made directions regarding the obtaining and filing of reports for the purposes of the review. Those reports were obtained and the review was heard over two days on 12 and 13 September 2016.

4 The issue to be determined on a special review is the same as that to be determined on an annual review: s 33(1) DSO Act. On any such review the court must determine whether a person who is the subject of a continuing detention order remains a serious danger to the community: s 33(1) DSO Act. If the person is no longer a serious danger to the community the court must rescind the continuing detention order. However, if the court finds that the person remains a serious danger to the community it must either expressly decline to rescind the detention order or rescind the order and make an order that the person be released into the community on conditions that the court considers appropriate: s 33(2) DSO Act.

5 In making a decision as to whether a person who is a serious danger to the community should continue to be detained or be released on a supervision order the paramount consideration is the need to ensure adequate protection of the community: s 33(3) DSO Act.

6 On this review Mr McGarry did not dispute that he remained a serious danger to the community. All of the expert evidence remains unchanged in that regard. The issue for determination is whether Mr McGarry can be released into the community on a supervision order with conditions that will adequately protect the community from the risk that he would commit further serious sexual offences.

7 It was submitted on behalf of Mr McGarry that suitable accommodation is available in the community and that he can now be released on a supervision order that would ensure adequate protection of the community. Whether the protection that could be afforded by a supervision order will be 'adequate' requires a qualitative assessment. It is necessary to have regard to the nature of the risk, the likelihood of the risk being realised and the likely affect that the proposed supervision order could have in reducing or obviating that risk. In this respect, the word 'adequate' has some similarities to the word 'unacceptable' as used in s 7 of the DSO Act: see in regard to s 7 Director of Public Prosecutions (WA) v GTR [2008] WASCA 187 [26] [27] (Steytler P & Buss JA) and Director of Public Prosecutions (WA) v Williams [2007] WASCA 206; (2007) 35 WAR 297 [63] [65] (Wheeler JA).

8 Mr McGarry's personal history and his history of offending were detailed by Jenkins J in Director of Public Prosecutions for Western Australia v McGarry [2009] WASC 226 [14] [41]. His history of sexual offending commenced in 1985. It started with offences of wilful exposure but over time he moved on to more serious offending which involved the commission of indecent acts upon children with whom he was in a familial relationship.

9 On 19 April 1991 Mr McGarry was sentenced to a total of 7 years and 8 months' imprisonment for 21 counts of indecent assault on a person under the age of 16 years, four counts of sexual penetration of a person under the age of 16 years and seven counts of wilful exposure. He was released on parole in February 1994.

10 In April 1994 Mr McGarry commenced offending against his daughter, then aged 12, by touching her indecently and masturbating in front of her. On 9 August 1994 he was convicted of two counts of indecently dealing with his daughter and sentenced to a total of 2 years' imprisonment to be served cumulatively on the balance of the earlier sentence he was yet to serve. On this occasion he was not made eligible for parole. He was released from prison on 26 February 1996.

11 In December 1997 Mr McGarry saw a photograph of an 11yearold girl in a local newspaper. He used the telephone directory to discover her home telephone number and address. He then went to the girl's home and entered the rear yard. He observed the girl and her 14yearold sister through the windows of the house and obtained their attention by tapping on the window. As the 11yearold girl approached the window he exposed himself and masturbated. He then left the premises but made several telephone calls to the girl's home in the following days. On each occasion he impersonated a police officer and discussed the offence that he had committed. As he did so he masturbated. On 30 October 1998 he pleaded guilty to indecently dealing with a child under the age of 13 years and to two summary offences of impersonating a police officer. He was initially sentenced on 16 December 1998 to 5 years' imprisonment but this was reduced to 3 years' imprisonment on appeal.

12 Following his release from prison Mr McGarry again became preoccupied with offending behaviour. On 16 April 2002 he hid in some bushland and waited for a young girl to walk to school along a nearby path. As a 14yearold girl walked past he came out of the bushes and grabbed her. He dragged her into the bushland and forced her onto the ground. He initially threw a jumper over her face so that she could not see him. He indecently assaulted her, then knelt next to her and masturbated until he ejaculated. He then rummaged through the girl's backpack apparently looking for money, and after saying that he would return in a short while, he fled the scene.

13 On 12 March 2003 Mr McGarry approached an 8yearold girl who was playing in a park near her home. He called her over to him and asked her how old she was. After she told him her age he offered her money in return for her showing him her pants. He told her that he was a counsellor and that this activity was part of his job. The girl screamed and ran from him. The matter was reported to the police. A short time later police located Mr McGarry walking through the grounds of a local primary school towards his vehicle. He was then observed buttoning up the front of his pants. His vehicle was searched and a number of pornographic magazines were located. He was then charged with both this offence and also that which had occurred on 16 April 2002. Following an appeal he was sentenced to a total of 5 years and 24 days' imprisonment.

14 Mr McGarry's sentence was due to expire on 2 January 2009. Before that day, on 7 November 2008, the State filed an application under the DSO Act. On the hearing of the application, the DPP sought that a supervision order be made. Jenkins J considered the available evidence and was satisfied that the community could be adequately protected by a supervision order provided that conditions, which she described as extremely onerous, were imposed: DPP v McGarry [2009] WASC 226 [123].

15 Mr McGarry signed the supervision order on 17 August 2009 and was released from custody shortly thereafter. On the same day an officer from the Western Australian Police Services Sex Offenders Management Squad interviewed Mr McGarry and explained the conditions of the supervision order to him. Those conditions included a prohibition on Mr McGarry having face to face contact with his daughter until the contact was prearranged via telephone. A further condition prohibited him from being at a shopping centre without a reasonable excuse at any time on weekends, except for specified purposes.

16 Within a short time after his release, Mr McGarry breached the terms of his supervision order. Those breaches involved having face to face contact with his daughter by attending a hospital at a time when he knew she would be there. He also learned that the daughter of one of his former partners was working at a shop in a shopping centre. He then went to that shop on two occasions that did not breach the order, but went there a third time on the weekend in breach of the order. There was evidence on the occasions he attended that he had acted in a way that his former stepdaughter, who was present on those occasions, found to be intimidating.

17 The DPP brought proceedings for contravention of the supervision order pursuant to s 23 of the DSO Act. Jenkins J found the contraventions proved: Director of Public Prosecutions for Western Australia v McGarry [No 2] [2009] WASC 287. In doing so, her Honour rejected explanations advanced by Mr McGarry for the contravention. In regard to meeting with his daughter at the hospital, he said that he had forgotten that he had been told she would be attending the hospital that day. In this regard, her Honour said:

I do not accept that Mr McGarry forgot that his daughter was going to the hospital that day. There is no evidence to that effect apart from his self serving comments to the police that he did not know that she was going to be there. This assertion is contradicted by her statement that she told him that she was going to be at the Hospital that day. It seems unlikely that, less than a week later, he would have forgotten that information. Further, it seems unusual behaviour to go to the Hospital to see a former partner, during her working day, without contacting her to make an appointment. On the balance of probabilities I find that at the very least, Mr McGarry went to the Hospital knowing that it was a possibility that he would see his daughter, without prior arrangement via the telephone. Given the number of phone calls and attempts made by Mr McGarry to meet with his daughter in the period after his release, I find that Mr McGarry was interested in meeting with his daughter. Probably, Mr McGarry thought that an apparently chance meeting with her at the Hospital would be desirable. When she did make contact with him, he breached his order by maintaining and extending the meeting by giving her a lift [17].

18 In regards to attendance at the shopping centre, Mr McGarry said that he thought the prohibition only related to a shopping mall and because he was able to enter the shop by an external door he did not believe that the condition applied. He also said that he had a reasonable excuse for attending the shop because he needed to buy a doona and quilt cover because he was cold. Her Honour rejected both of these explanations and concluded that:

I reject Mr McGarry's contention that he had a reasonable excuse for being on the premises of the shopping centre on Saturday to buy a doona and cover. First, he had other warm bedding available to him. Secondly, if he truly needed further warm bedding before Monday, 24 August 2009, when he could have legitimately visited a shopping centre, he could have sought the assistance of a charity. Thirdly, he had last been in the store the afternoon of the previous day. No explanation has been proffered as to why he did not buy the bedding on that occasion. Fourthly, on each of the three occasions that Mr McGarry went to the shop he used the opportunity to make contact or try to make contact with the complainant. It was only when Mr McGarry found out that the complainant worked at the shop that he started to shop there. This contact and the lack of a reasonable excuse for it given the age difference between the complainant and Mr McGarry, their lack of a recent familial or social relationship and the indifferent attitude displayed by the complainant to Mr McGarry on these occasions points strongly to the inference that Mr McGarry was using the visits to the store, at least in part, as an excuse to make contact, which he would not otherwise have had, with the complainant [31].

19 There was also evidence that Mr McGarry had purchased and eaten food at a fastfood outlet. At the time he did so he was sitting within 50 m of a children's playground situated within the outlet. He sat in a position overlooking the playground. When asked about this incident he told police that he had not thought about the condition which required him to not be within 50 m of a children's playground when he sat where he did.

20 Her Honour noted that at the time Mr McGarry came before her he had commenced antilibidinal medication and had remained on it for some months. Tests at that time indicated that his testosterone levels had fallen to a very low level as compared with the average levels for men of his age. It was hoped that this would reduce his risk of reoffending. It was, however, noted that without very close supervision it was unlikely that Mr McGarry could be relied upon to take his medication or to report the presence of feelings of sexual arousal. Her Honour then said:

Thus, the fact that Mr McGarry is on medication and it is working to reduce his testosterone levels is but one strategy to reduce Mr McGarry's risk of reoffending. I am satisfied that the psychiatrists were of the view, as were and am I, that along with a requirement for Mr McGarry to take his medication there was a need for very close supervision and control of Mr McGarry in the community. Dr Wynn Owen in his report of 15 February 2009 said that risk management in the community, even under close supervision, will not reduce Mr McGarry's risk of reoffending unless major constraints are applied. The conditions on Mr McGarry's supervision order were designed to provide that constraint.

I do not accept Mr McGarry's contention that the fact that he is taking antilibidinal and antidepressant medication means that he will not commit a serious sexual offence [67] [68].

21 Her Honour then referred to the circumstances in which she made the supervision order and said:

The conditions of the supervision order were not just to try and reduce Mr McGarry's physical or psychological desire to offend. They were also for the purpose of making sure that the authorities could appropriately monitor Mr McGarry so as to reduce the risk of him reoffending. They were also for the purpose of controlling or constraining Mr McGarry's movements so that he would not put himself into high risk situations which may tempt him to reoffend.

I was of the opinion then and remain of the opinion that a multi pronged approach was necessary to reduce Mr McGarry's risk of offending to an acceptable level which would enable him to be released into the community under a supervision order.

The fact that his proven contraventions and likely contraventions of the supervision order do not directly relate to serious sexual offending, does not mean that they should be given little weight in my determination of this application. As I said in my original decision, it is only if all the conditions of the supervision order are complied with that I can be satisfied that the community will be adequately protected from the risk that Mr McGarry will commit a serious sexual offence.

Now that I am satisfied that Mr McGarry has not complied with those constraints and is likely not to comply with them all in the future, I conclude that there is an unacceptable risk that, if I do not make a continuing detention order Mr McGarry will commit a serious sexual offence [74] [77].

22 One of the matters that was of concern to Jenkins J in regard to whether Mr McGarry would comply with conditions in the future was his general conduct after his release on the supervision order. Her Honour found that the contraventions indicated that Mr McGarry was not prepared to, and not able to, live a cautious enough life to ensure compliance with a supervision order. She noted that he had not taken steps to comply with conditions in circumstances where his obligations were obvious.

23 There are a number of conclusions that can be drawn as to the nature of the risk. Firstly, Mr McGarry's victims have largely been prepubescent girls, although the age range has been from 8 to 16 years of age. Secondly, the victims have included not only children who Mr McGarry has been able to access and groom in a domestic context but also children who are otherwise complete strangers. Thirdly, some of his offending has involved a significant degree of preparation and planning. Fourthly, he has demonstrated a willingness to exploit what he perceives as weaknesses or ambiguities in the conditions of a supervision order. Fifthly, he has demonstrated a willingness to engage in behaviour notwithstanding that he knows that it has been associated with an increased risk of reoffending. These factors are relevant both to the nature and degree of risk and to an assessment of the likely effectiveness of any supervision order.

24 Based on Mr McGarry's history of offending, the risk is that he will commit a serious sexual offence against a young female. The possible victim could be a child, but also extend to girls of 14 to 16 years old. There is no fixed pattern to his offending and this makes predicting specific risk scenarios a difficult task. The risk relates not only to children who he may be in a familial or domestic relationship with, it extends to children who are strangers and who he has seen only briefly before becoming obsessed with them. His offending is not likely to be opportunistic, rather it could involve significant planning and preparation. The offending could involve the use of psychological coercion as well as physical violence.

25 In the circumstances described, the nature of the risk is a serious one and the class of potential victims is very wide. The variety of offending conduct engaged in means Mr McGarry could be difficult to predict and also, consequentially, difficult to manage. These difficulties are significantly magnified by his proven resistance to management and willingness to breach the terms of supervision.

26 I have referred earlier to reasons that have been given following annual reviews in 2012, 2013, 2015 and earlier this year. On each of these occasions the issue has been whether release on a supervision order could provide adequate protection to the community. On each occasion the conclusion has been that it could not and the continuing detention order has not been rescinded.

27 In DPP v McGarry [No 7], the 2015 annual review, I concluded that the factors that contribute to Mr McGarry's risk of reoffending included a sustained deviant sexual interest and a continuing sexual desire. Other factors that were relevant included psychopathy, narcissism, and antisocial personality traits. The nature of those factors meant that psychotherapy had limited benefit. Unless the risk could be reduced by some other means it was not of a nature as to be capable of being adequately managed in the community.

28 I noted at the time of the 2015 review that the only reasonable option for reducing the risk to a manageable level was antilibidinal therapy. Mr McGarry had previously undertaken a course of such therapy but had ceased it due to significant adverse health effects. As at the date of the 2015 review he was seeking to resume medication but had been unable to find a doctor who was willing to prescribe it. I concluded at that time that until Mr McGarry had been prescribed the medication, and been on it for some period of time to determine whether it was effective in reducing the risk of reoffending, it was impossible to make a determination as to whether a supervision order was a viable option. It was also unclear at that stage whether a programme could be developed by Mr McGarry in consultation with his doctors to take the medication, monitor its effectiveness and mitigate its side effects.

29 There was another factor which made a supervision order at the time of the 2015 review a nonviable option. That was that there was no suitable accommodation to which Mr McGarry could be released. Two suggested options had been assessed and found to be inappropriate either because of the proximity to a school or because the property was in an area not capable of being monitored by GPS tracking. Both of the psychiatrists who gave evidence at that review stressed the importance of suitable accommodation and employment in reducing the risk of reoffending by minimising exposure to risk factors and boredom.

30 At the last annual review, DPP v McGarry [No 8] [2016] WASC 82, the position in regard to antilibidinal medication had changed. Mr McGarry had been prescribed Androcur, an oral antilibidinal medication, by a prison doctor. He commenced on Androcur on 26 September 2015 at a dose rate of 25 mg per day. He had undertaken blood tests to monitor the effects of Androcur, both in reducing the male hormone testosterone and any possible adverse side effects. Prior to commencing on Androcur his testosterone levels were 21 nanomoles per litre of blood (the normal range is 10 to 35 nanomoles per litre). Two months after commencing on Androcur, testing undertaken on 26 November 2015 showed that there had been a marked reduction in testosterone to 3.8 nanomoles per litre.

31 Two psychiatrists, Dr Wynn Owen and Dr Wojnarowska, provided reports and gave evidence at the last annual review. Dr Wynn Owen concluded that Mr McGarry's risk of future serious sexual offending was high but that the risk could be addressed by use of antilibidinal therapy. He considered that such medication was an essential component in managing any such risk in the community. Mr McGarry had reported a significant reduction in his libido following his resumption of Androcur. Notwithstanding this Dr Wynn Owen suggested that an increase in the dose rate could be trialled in order to see whether the amount of active testosterone could be reduced still further. He acknowledged that this carried with it some risk of increased side effects. He was of the view that continued use of medication was an essential and effective management of risk and that there should be a mechanism for reassessing the risk in the event that Mr McGarry became noncompliant with the medication regime.

32 Dr Wojnarowska also considered that antilibidinal medication was an important factor in reducing the risk of reoffending and that its continued use would be essential if he was to be released into the community. She did not recommend an increase in the dose rate, in part due to the fact that the existing dose rate had been apparently effective, but also because increasing the dose rate could, in her view, be counterproductive. An increased dose rate may make Mr McGarry more prone to lethargy and boredom. Dr Wojnarowska said that boredom was a factor which could increase risk. Both psychiatrists were of the view that with continued use of antilibidinal medication and other appropriate conditions Mr McGarry was capable of being adequately managed in the community.

33 There was another factor that was relevant to the suitability of a supervision order at the time of the last annual review. That factor was, again, the availability of suitable accommodation. A number of accommodation options were considered, but none of them was appropriate. The proximity of children or former victims or the unavailability of GPS tracking in respect of these options made management considerably more difficult and increased the risk of offending. This was so even taking into account the antilibidinal medication.

34 In the circumstances that existed at the time of the last annual review, I concluded that Mr McGarry was then suitable for release on a supervision order but only if that could be to a place where the conditions of release could be effective. At that time no such place had been found and there was no reason to believe that a suitable place was then available. Accordingly, I declined to rescind the continuing detention order.

35 At the hearing of this review the DPP tendered a book of materials. There was no objection to the tender. The book included the following:

1. a Department of Corrective Services 'Incidents and Clearances' Report relating to the period 1 January 2016 to 19 August 2016;
2. an individual management plan prepared by the Department of Corrective Services dated 12 May 2016;
3. health records for the applicant for the period 11 January 2016 to 2 August 2016;
4. a bone mineral density report for testing conducted on 16 May 2014;
5. blood test reports for testing conducted on 26 November 2015, 18 June 2016, 25 August 2016 and 6 September 2016;
6. a desktop analysis prepared by the WA Police Sex Offender Management Squad in respect of proposed accommodation;
7. a community supervision assessment by Ms Jane Henshall, a senior community corrections officer with the Department of Corrective Services, dated 24 August 2016;
8. an updated community supervision assessment report by Ms Henshall dated 8 September 2016;
9. a treatment progress report by Ms Vanessa Rankin, a senior clinical psychologist for the Department of Corrective Services, dated 2 September 2016; and
10. a psychiatric report and updated report by Dr Peter Wynn Owen dated 24 August and 8 September 2016 respectively.

36 A written report by Dr Wojnarowska was unable to be incorporated into the book of materials and was separately tendered at the hearing. That report is dated 9 September 2016. The applicant also tendered an affidavit of his solicitor, Ms Mara Barone, sworn on 10 August 2016 and the book of materials tendered at the last annual review.

37 The respondent called Dr Wynn Owen, Dr Wojnarowska, Detective Senior Constable Gary PayetSmith and Ms Henshall to give oral evidence. Mr McGarry elected not to give evidence or adduce any evidence, other than that referred to.

38 Whilst this special review was sought upon the basis that suitable accommodation had now been found, it could not merely be assumed that all other factors remained the same as at the last annual review. In assessing the risk of reoffending and the appropriateness of a supervision order it was necessary to consider all of the circumstances that now exist.

39 As has been noted earlier, at previous reviews antilibidinal medication had been identified as an essential component of any plan to reduce and manage the risk of reoffending. If medication could reduce testosterone to a low level the expectation was that this would also reduce libido and any consequential desire to commit sexual offences. At the time of the last annual review, Mr McGarry had been on a 25 mg per day dose of Androcur for several months and testing indicated that that dose had been effective in reducing his testosterone. The belief at that time was that the low levels achieved had stabilised and were capable of being maintained on the existing dose rate. This provided a basis for the opinion of the psychiatrists that the medication had been effective and would continue to be so if Mr McGarry was released on a supervision order.

40 Since the last annual review, Mr McGarry has continued on the 25 mg per day dose rate. However, further testing has shown fluctuations in his levels of testosterone. In particular, following the last annual review there was a marked increase in testosterone levels at the next test conducted on 18 June 2016. The normal range for testosterone is between 10 and 35 nanomoles per litre. Prior to commencing the most recent regime of Androcur Mr McGarry's testosterone levels were between 17.9 and 21.7 nanomoles per litre. After commencing Androcur, these levels dropped to 3.8 nanomoles per litre when tested on 26 November 2015. However, the next test conducted on 18 June 2016 showed that testosterone levels had more than doubled to 8 nanomoles per litre. Subsequent tests on 26 August and 6 September 2016 produced results of 4.7 nanomoles per litre and 5.0 nanomoles per litre respectively. It is noteworthy that these last two tests occurred after the application for a special review had been filed and pursuant to orders of the court.

41 Dr Wynn Owen considered that the fluctuation that occurred on 18 June 2016 was 'quite surprising' and 'quite significant'. He said that some variations could occur due to factors such as over drinking of water, dehydration or prolonged or extreme exercise but such variations would not be expected to exceed 0.5 0.7 nanomoles per litre. The degree of variance that occurred here was abnormal and Dr Wynn Owen considered that a possible explanation for it could be noncompliance. At ts 758, he said:

Well, compliance could be not swallowing the whole tablet or some sort of mistake occurring between sort of cup and lip. Or, it could be something which is wholly conscious, as in secreting the medication and spitting it out later, for example. How - all I'm mentioning is that there's a fluctuation there that needs to be investigated and probably would not be wanted in the levels of testosterone approach the normal on a regular basis, and that we had to consider all of the possibilities or possible causes of those fluctuations.

42 Dr Wynn Owen said that when Mr McGarry was asked about the fluctuation, he had said that he had been unwell. Dr Wynn Owen said that there were a range of illnesses that could affect renal or liver function and that this in turn could affect the levels of testosterone. However, Dr Wynn Owen noted that Mr McGarry was reasonably healthy and that even in the case of severe influenza a fluctuation of the degree here would not be expected (ts 758). Dr Wynn Owen had also examined the prison medical records and noted that there were no presentations to doctors or nurses that would explain the fluctuation.

43 Whilst Dr Wynn Owen could not say with any certainty that deliberate noncompliance was an explanation for the fluctuation, it was a possibility that could not be dismissed. He suggested that an alternative medication could be trialled which would avoid the risk of noncompliance. DepoProvera is a contraceptive that has also been found to be effective as an antilibidinal treatment. This medication is delivered by way of intramuscular injection on a monthly basis. He said that the frequency with which injections needed to be administered varied from individual to individual. It would be necessary to administer a test dose and to then determine the effect by undertaking regular blood or urine tests. DepoProvera acts in a very similar way to Androcur and Dr Wynn Owen expected that within one or two doses good control could be achieved. The drug has the same profile of side effects as does Androcur and these would also have to be monitored. Because DepoProvera would be administered by a doctor any risk of noncompliance could be avoided.

44 Dr Wynn Owen said that DepoProvera could be trialled whilst continuing on a course of Androcur. He said that this transition could be managed safely in the community. He envisaged that the changeover would be monitored very tightly, meaning that there would be very regular testing for at least the first four or five weeks. If a 12month period established consistent results with relatively small fluctuations it would then be reasonable to move out to a quarterly testing regime (ts 778).

45 Dr Wojnarowska also considered that the fluctuations were significant and greater than would be expected from merely physiological factors that interfere with the metabolising of medication. She also thought that a move to DepoProvera was appropriate. She said that injectable medication was deliverable in a much more consistent form and reduces risks associated both with noncompliance and problems with absorption due to gastric illness. She also considered that a transition to DepoProvera could occur in the community. Dr Wojnarowska said that if noncompliance with Androcur had occurred that was a matter relevant in terms of assessing risk. That would be particularly so in respect of Androcur which depends upon the person taking the tablets on a daily basis.

46 In my view, the fluctuation in testosterone levels is a matter of considerable concern. The psychiatrists have identified the importance of reducing testosterone levels to a low and stable amount as being critical to the management of Mr McGarry in the community. The fact that the levels have fluctuated to a considerable degree is a matter of concern in itself, whatever the cause.

47 It was suggested in submissions that the rise to 8 nanomoles was still below the normal range. The implication was that this fluctuation did not represent any increase in risk. That, however, is not consistent with the evidence of the psychiatrists who saw the fluctuation as being significant. The fact that it has occurred, whatever the explanation for it is, indicates that the current medication regime has not been effective in the consistent way considered necessary.

48 It was suggested in submissions that the 8 nanomole reading was an 'outlier'. I do not accept this submission. Neither of the psychiatrists was of the view that this reading was unreliable or should be ignored. In any event the testing over the last eight months has not occurred with any degree of frequency or consistency such that any clear pattern in the movement of testosterone could be discerned.

49 If the reason for the fluctuation was deliberate noncompliance, this would raise real concerns regarding the viability of management in the community. At present Mr McGarry's Androcur medication is dispensed by the prison authorities. The opportunities and incentives to avoid taking the medication are reduced in that environment. However, if released into the community he would obtain Androcur on prescription and be expected to take the tablets without supervision. Whilst there may be a risk that noncompliance would be detected by random testing, the risk of such testing does not appear to have acted as sufficient incentive in a custodial environment. Furthermore, Mr McGarry has proven not to be a reliable or willing subject of external controls.

50 It was submitted on behalf of Mr McGarry that the evidence did not establish that the fluctuation was due to noncompliance. It is true that the psychiatrists could not say with any certainty that this was the explanation. However other possible explanations, such as illness, exercise or dehydration, were discounted as being likely causes of a variation of this amount. Deliberate noncompliance was considered to be a real possibility and it was for this reason that both psychiatrists favoured a move to injectable DepoProvera.

51 It was submitted on behalf of Mr McGarry that a transition to DepoProvera could occur safely in the community. Both psychiatrists gave evidence to this effect. There are, however, a number of uncertainties in this process.

52 Firstly, Mr McGarry has not been trialled on DepoProvera and its effectiveness in reducing his testosterone level is not known. Whilst the evidence indicated that DepoProvera works in a similar way to Androcur and there was no reason to believe that it would not be effective, both psychiatrists recommended a transition. This transition would enable the effectiveness of DepoProvera to be assessed whilst maintaining the existing dose rate of Androcur. If there was no uncertainty as to the effectiveness of DepoProvera then this transition would be unnecessary. This raises the distinct possibility that DepoProvera might not be effective in the way anticipated. In this event it would be necessary to fall back on Androcur, in respect of which there has been an unexplained fluctuation, one possibility for which may be noncompliance. Whilst the risk that DepoProvera may not be effective may be small, it is a risk that needs to be weighed in the context of the nature of any offence that may be committed if the risk of reoffending is realised.

53 Secondly, if the fluctuation did not occur due to noncompliance then there is another concern. As noted earlier, DepoProvera works in the body in a similar way to Androcur. In these circumstances, in the absence of any explanation for the fluctuation, the possibility of fluctuations occurring on DepoProvera could not safely be excluded. The fluctuation that occurred in June was significant, both because of the size of it and because it brought Mr McGarry close to the normal range for testosterone. The possibility that a variation of this degree could have a significant affect upon libido and, therefore, on the risk of offending, has not been discounted.

54 Thirdly, there is also an uncertainty regarding side effects that may be caused by DepoProvera. Whilst the range of side effects caused by that medication is similar to Androcur, the degree to which they may be suffered is not yet known. In the past the occurrence of significant side effects of a larger dose of Androcur were such as to cause Mr McGarry to cease his treatment. At present it cannot be known whether an effective dose of DepoProvera would result in side effects to the same or similar degree. If that occurred and the medical advice was to discontinue DepoProvera and revert to Androcur, the difficulties with ensuring compliance would remain unresolved.

55 Whilst DepoProvera is likely to be more effective in avoiding the risk of noncompliance and also of obviating some physiological impediments to absorption, I am not satisfied that the suitability and effectiveness of that medication can be safely assumed before it has been trialled in a custodial environment. The risk that it will be ineffective may be small but that risk needs to be considered in the context of a man who has been shown to be unreliable and manipulative. The fluctuation in testosterone levels leaves me with a significant concern that releasing Mr McGarry on Androcur would adequately protect the community and the evidence in respect of DepoProvera does not yet establish that it is a safe and appropriate alternative. What would be required is a trial of DepoProvera in a custodial environment together with a regime of regular testing both of testosterone levels and of possible side effects. This would have to occur over a period sufficient enough to establish that longterm use of DepoProvera would be a safe and effective management tool.

56 In August a rental property in suburban Perth became available to Mr McGarry. The owner of the property was a Government department and a lease of six or 12 months duration was available. The property consists of a single story brick dwelling on a reasonably large block. There is a laneway to the rear of the property though no access is possible due to a rear fence. It is proposed that Mr McGarry would live at the property on his own, though there is a possibility that he could be joined there by a woman with whom he has been corresponding.

57 A spatial analysis was conducted by police from the Sex Offender Management Squad. This involved officers visiting the property on a small number of occasions, making observations as to the surroundings and taking photographs of those surroundings. There is a café and shopping area, several parks and public spaces, 12 schools and eight childcare centres within easy walking distance of the property . One of the schools is a primary school which is approximately 500 m from the residence. One of the childcare centres is located approximately 600 m away from the residence.

58 Police officers observed what appeared to be children's toys in the back garden of a neighbouring house. No children were observed, although the toys had moved when the police visited on a second occasion. There is a solid fence between the properties, although the evidence was that the height of that fence was not such as to prevent an adult seeing into the neighbouring garden. A platform or cubby house was also observed in the backyard of a house to the rear of the property. A child's bicycle was also observed at that property, though no children were seen at the time of the police visits.

59 In crossexamination, Detective Senior Constable PayetSmith, who gave evidence of the police spatial analysis, was asked whether he had kept a log of pedestrians walking along the street at the front of the property and whether there was any comprehensive analysis of the usage of the laneway to the rear of the property. He said that no log or analysis had been undertaken and that his visits had been relatively brief. He said that it would take a large amount of resources to properly conduct such an analysis. He said it was also impractical to make inquiries with neighbours as to whether they had children because to do so would necessitate revealing the purpose of those inquiries.

60 Police officers also established that there are two convicted child sex offenders residing in close proximity to the property, including one who lives approximately 50 m away. It could not be established that Mr McGarry knew either of these offenders, though he had been incarcerated in the same prison as one of them. A notice requiring him not to consort with these people pursuant to s 557K of the Criminal Code would need to be served on Mr McGarry.

61 In her community supervision assessment, Ms Henshall said that the surrounding area of the property raised concerns. In particular the laneways and accessibility to private properties could encourage behaviours in Mr McGarry related to his sexual offending 'such as watching, fixating, preoccupations and planning especially during periods of boredom'. Whilst global positioning system (GPS) tracking could create exclusion zones, Ms Henshall said that this may not be adequate to ensure the safety of the community. It was considered that 'significant observation of Mr McGarry' by police would be required.

62 In general terms Dr Wynn Owen considered that the property was suitable. He did consider that there was some risk of boredom if Mr McGarry lived alone, although initially he expected that there would be a high degree of structure in any supervision arrangements. Boredom could also be alleviated if the woman with whom Mr McGarry has been corresponding came to live with him. He also has a supportive nephew and sister.

63 Dr Wynn Owen's views in relation to the proposed accommodation was summarised by him in the following terms:

I do understand that it is extraordinarily difficult to source accommodation of this sort and there were two - two things that stood out to me in relation to the reports. One was the proximity of other offenders. Having said that, if that was a person that Mr McGarry knew, that might make a difference because talking to somebody of that sort may well emphasise underlying beliefs that condone sexual violence or sexual offending. I don't know that that's the case and nor does Mr McGarry. All we know is that the - that the person within 50 metres and I think one other have been in the same prison as him but that doesn't tell us that he's had any contact with them but there would be a risk, I believe, if that person was known. The other issue is of course the proximity to schools or playgroups.

Does that concern you? - It does but it's very difficult to find anywhere in a residential area that doesn't have some proximity to a school so 500 metres is - is actually a reasonable distance. The issue I had and tried to address with Mr McGarry was that around regular movements around his house in that I particularly was aware that from seeing an image of a young girl on the front of a community newspaper, he then followed it up and - and a number of offences were committed and he said to me that at that time with his thinking, he could become obsessed.

Yes? - He has said that he wouldn't go in the front yard anyway even if the street was a place that - that students would walk up and down and that he didn't believe that there was a risk there but my - my concern would have to be around somebody who passes regularly that might attract his attention. Having said that, all streets are thoroughfares and it would be very very difficult to find one that didn't have people walking around regularly to and fro.

...

Is that what you had in mind that in an apartment, there wouldn't necessarily be the traffic, the pedestrian traffic passing? - That's correct, your Honour. Rather than a residence facing straight onto the street.

...

What about neighbours' houses that are in sight; the front yard or the backyard of a neighbour and the activity that goes on there. What if there are children who are playing in the backyard on a regular basis? - Look, I think all - all of those are potential issues. However, in any residential area, there's no control over who lives there, and that will always be an issue with any placement of any offender in the community, unfortunately. Again, I wouldn't want to jump to a conclusion, it's really just about how can we reduce the risk of something occurring that may be similar to something that's occurred in the past [ts 764 765].

64 Dr Wojnarowska was also of the view that the accommodation was suitable. She said:

I don't have any specific concerns about the accommodation. It appears to be situated in a sort of medium socioeconomic environment. That means that you will have a mixture of children who are very well supervised and those who are less supervised, but that's probably as good as he can get in terms of accommodation.

Did you have concerns and I think you heard Dr Wynn Owen's evidence when I put to him the question of whether he had any concerns about laneways surrounding the house, do you have any concerns about that? - I understand that he can't really get to the back lane without leaving the house from the front. I don't think he's got access unless he jumps over a fence which appears to be quite high.

And what about its proximity to local schools, the closest primary school being 500 metres and the closest child care centre being 600 metres? - Well, again I do agree with Dr WynnOwen's testimony that the proximity is not as important as frequency with which he would be viewing children and that could be managed by creating curfew times and also the exclusion zones [ts 818].

65 I accept that the nature of the risk in the case of Mr McGarry and the difficulties in managing him mean that finding accommodation is extremely difficult. The issue, however, is not whether this accommodation is the best that can be found but whether, together with all of the conditions that could reasonably be included on a supervision order, it would enable adequate protection of the community from the risk of reoffending. There are a number of issues with the proposed residence which need to be considered. The availability of management controls to reduce risk factors is a relevant consideration. A risk does not become acceptable merely because it cannot be further reduced.

66 The proximity of the house to schools, day care centres and recreation areas where children may be present is a matter of concern. It is true that it would be difficult to find a suburban residence that was not within reasonably close proximity of such facilities. Exclusion zones monitored by GPS may be effective to some extent. However, there would still be a concern regarding the movement of children along the street or the laneway behind the house. Mr McGarry's offending conduct has included becoming fixated on children and using force to take a child from a public thoroughfare.

67 It was submitted on behalf of Mr McGarry that no logs or analysis had been undertaken regarding movements of children in the vicinity of the house. This is true but there is evidence of the existence and proximity of schools and the movement of children along the street and laneway could not reasonably be excluded. I note in this regard that on a previous occasion Dr Wynn Owen had suggested an apartment or unit as being appropriate accommodation, at least in part because it was less likely to afford the opportunity of observing passing children.

68 A similar concern arises in respect of the possible presence of children in nearby houses. It was submitted on behalf of Mr McGarry that the only evidence in this regard was inconclusive. I accept that that is so, however the evidence does not allow me to conclude that children are not present in houses that are close by. The proposed residence affords the possible opportunity to observe children in their own gardens whilst remaining hidden. Given Mr McGarry's prior offending profile this is a matter of concern.

69 The fact that Mr McGarry would be living at the property alone raises issues of boredom and that he could engage in fixated thinking without distractions. Dr Wynn Owen was of the view that single occupancy was less desirable in the longer term than shared accommodation. On the other hand, I do accept that a highly structured supervision regime may mitigate the risk of boredom and that Mr McGarry hopes that he may be joined by the woman with whom he has been corresponding.

70 As I have noted on previous occasions, accommodation is not a factor to be viewed in isolation, rather it is one of a range of considerations that need to be taken into account when determining whether the community can be adequately protected from the risk of reoffending. The proposed accommodation avoids a number of the problems that made previous accommodation options unsuitable. However, it is not without its problems and I have referred to a number of them. Those problems have to be seen in the context of the issues that have been raised earlier regarding testosterone levels.

71 In the course of evidence, counsel for Mr McGarry crossexamined the witnesses regarding a number of the conditions that were proposed in the event that Mr McGarry was to be released in to the community on a supervision order. There were suggestions made regarding the wording of some conditions and the necessity for them. Some of those suggestions were accepted, others were not.

72 One issue was with conditions that required Mr McGarry to disclose the existence of friendships with persons who had custody of children. It was pointed out that what was meant by a friendship was unclear and that perceptions in this regard could differ. The purpose of these conditions was, however, clear. That is to ensure that Mr McGarry does not obtain trusted or privileged access to children without the knowledge of those who would be managing him.

73 I accept that the wording of some proposed conditions was open to interpretation and could be improved. However, any problem with such conditions is exacerbated in the case of Mr McGarry because of his proven tendency to take advantage of what he perceives to be ambiguity and to engage in risk taking behaviour despite knowing of the efforts made to prevent it.

74 It is unnecessary for me to engage in a detailed analysis of the proposed conditions because I am not satisfied that any conditions would presently ensure adequate protection of the community in light of the evidence I have referred to regarding testosterone levels and accommodation.

75 The fluctuations in the level of testosterone whilst on Androcur are such that I have significant concerns as to the effectiveness of that drug. If the reason for the fluctuations is noncompliance then a move to DepoProvera is a possible solution, however that medication has not yet been trialled. The risk that DepoProvera will not be effective may be small, but even a relatively small risk can be unacceptable where the nature of the possible future offending is serious, the person is at high risk of reoffending and the person is difficult to manage. In the absence of any adequate explanation for the fluctuation in testosterone levels, I am not satisfied that the community could be adequately protected by releasing Mr McGarry before there has been a trial of DepoProvera.

76 For these reasons, I expressly decline to rescind the continuing detention order.