Response

The recent guest editorial by Professor Rob Poole, ‘Remembering Lysenko; when ideology and science meet’, ^{Reference Poole1} asserts that the continued use of community treatment orders (CTOs), following the conclusive findings of the OCTET study that these do not work, is based on pseudoscience. The editorial states: ‘Anti-science and ideology come together slowly, and UK pseudoscience has had unforeseen consequences’, which applies to CTOs, and that ‘There are current examples that willfully ignore empirical evidence within British psychiatry’, including CTOs. And also,

These criticisms apply to over 75 jurisdictions and many scientists, psychiatrists, families and patients who support CTOs. The editorial has accused them all, primarily based on three small, flawed studies, of falling prey to a consequence of pseudoscience, which is ignoring evidence.

The editorial dismisses positive findings for CTOs from non-randomised controlled studies by relying primarily on the OCTET study. ^{Reference Burns, Rugkåsa, Molodynski, Dawson, Yeeles and Vazquez-Montes2} The editorial states that ‘Burns and his colleagues managed to complete the OCTET Randomised Controlled Trial which conclusively showed that CTOs have no advantage over existing measures in preventing relapse, a finding that has been shown in similar studies elsewhere in the world’ [emphasis added]. However, only two other randomised control trials (RCTs) exist, both in one country, and these have significant methodological limitations. ^{Reference Segal and Badran3}

Burns is reported as stating ‘…CTOs should be abandoned or that another RCT should be mounted to confirm or refute the results.’

Neither occurred.

However, the ethical and methodological challenges in OCTET illustrate why further RCTs have not been mounted and why non-RCT studies can be valid. ^{Reference O’Reilly and Vingilis4}

The OCTET trial did not fully test the real-world CTO question. In practice, the clinical question is: among patients whose history and current presentation make them strong candidates for a CTO, including those who would not voluntarily agree to be randomised in a research experiment, do CTOs reduce relapse, readmission or disengagement compared with otherwise similar patients not placed on a CTO?

OCTET found no difference in readmission rates within 12 months. For the CTO group, it was 36% (59 out of 166), and 36% (60 out of 167) for the control group (non-CTO, leave). However, limitations were substantial in both sampling and execution. The sampling was not typical of CTO populations because, first, potential CTO patients were excluded from the study if clinicians could not risk their patient not being on a CTO; second, of the 442 qualifying patients, 91 refused to be in the study, 9 lacked capacity and 6 were not eligible, meaning that 24% (336 out of 442) were excluded, although arguably they are typical of CTO patients; third, potential CTO patients had to be less severe so that they could ethically be randomised to the non-CTO leave group; fourth, the CTO group was not compared with a similar voluntary group but with another compulsion group (non-CTO leave). The sample does not appear to be typical of CTO populations in practice.

The execution of the study resulted in several apparent protocol violations. For example, because this was an intention-to-treat study, if a patient assigned to the CTO group was not put on a CTO they were still counted as if they had received one. The substantial violations in the execution of the study included, first, 25% (42 out of 166) did not receive a CTO despite allocation to the CTO group; second, 40 of the 167 patients in the non-CTO leave control group (24%) had to be placed on a CTO but remained in the non-CTO control group for analysis; and third, the findings of this study compared two coercive regimes rather than providing a comparison of CTOs with non-CTOs. For these reasons, and also differences in CTO requirements in other jurisdictions, these findings cannot be generalised to other jurisdictions.

A comparable drug study would be unlikely to declare a drug ineffective with this level of serious protocol violations. These violations include 25% of the treatment group not receiving the treatment and 24% of the placebo group having to be put on the drug but counted as if they were in the placebo group. It would be even more unlikely to reach a negative conclusion when professional and government bodies worldwide were using that drug with some evidence that it was effective.

The editorial asserted that, since OCTET there has been ‘no further research’, presumably meaning RCTs. This is because a true comparison of CTO patients versus comparable non-CTO patients would be unethical; just as randomisation of dangerous psychotic patients to involuntary admission and treatment versus not admitting or treating them would be unethical. However, considerable research using ethical scientific methods has occurred. ^{Reference Segal5} For example, Kisley, lead author of the Cochrane reviews, has, in a recent review of systematic reviews and meta-analyses of both quantitative and qualitative studies, concluded: ‘Whereas uncontrolled evidence suggested benefits, results were more equivocal from controlled studies and randomised controlled trials showed no effect.’ ^{Reference Kisely, Zirnsak, Corderoy, Ryan and Brophy6} The positive findings included a reduction in mortality and positive opinions of families and psychiatrists about CTOs. Kisley also referenced a large controlled before-and-after study with 14 726 patients over a 10-year period. The conclusion was that ‘CTOs in New Zealand are associated with increased community care, and increased dispensing of psychiatric medication. Patients with Psychotic Disorders also experienced reduced frequency and length of admissions while under a CTO.’ ^{Reference Beaglehole, Newton-Howes and Frampton7}

We agree with an assessment by respected researchers of all types of CTO studies, including OCTET, that, in their view, such evidence is sufficient to justify more widespread implementation of out-patient commitment. ^{Reference Swanson and Swartz8}