Davoren Mary, Hennessy Sarah, Conway Catherine, Marrinan Seamus, Gill Pauline, Kennedy Harry G
Department of Psychiatry, Trinity College Dublin, Dublin, Ireland.
National Forensic Mental Health Service, Central Mental Hospital, Dundrum, Dublin 14, Ireland.
BMC Psychiatry. 2015 Mar 28;15:61. doi: 10.1186/s12888-015-0433-x.
Detention in a secure forensic psychiatric hospital may inhibit engagement and recovery. Having validated the clinician rated DUNDRUM-3 (programme completion) and DUNDRUM-4 (recovery) in a forensic hospital, we set out to draft and validate scales measuring the same programme completion and recovery items that patients could use to self-rate. Based on previous work, we hypothesised that self-rating scores might be predictors of objective progress including conditional discharge. We hypothesised also that the difference between patients' and clinicians' ratings of progress in treatment and other factors relevant to readiness for discharge (concordance) would diminish as patients neared discharge. We hypothesised also that this difference in matched scores would predict objective progress including conditional discharge.
In a prospective naturalistic observational cohort study in a forensic hospital, we examined whether scores on the self-rated DUNDRUM-3 programme completion and DUNDRUM-4 recovery scales or differences between clinician and patient ratings on the same scales (concordance) would predict moves between levels of therapeutic security and conditional discharge over the next twelve months.
Both scales stratified along the recovery pathway of the hospital, but clinician ratings matched the level of therapeutic security more accurately than self ratings. The clinician rated scales predicted moves to less secure units and to more secure units and predicted conditional discharge but the self-rated scores did not. The difference between clinician and self-rated scores (concordance) predicted positive and negative moves and conditional discharge, but this was not always an independent predictor as shown by regression analysis. In regression analysis the DUNDRUM-3 predicted moves to less secure places though the HCR-20 C & R score dominated the model. Moves back to more secure places were predicted by lack of concordance on the DUNDRUM-4. Conditional discharge was predicted predominantly by the DUNDRUM-3.
Patients accurately self-rate relative to other patients however their absolute ratings were consistently lower (better) than clinicians' ratings and were less accurate predictors of outcomes including conditional discharge. Quantifying concordance is a useful part of the recovery process and predicts outcomes but self-ratings are not accurate predictors.
被拘留在安全的法医精神病医院可能会抑制参与度和康复进程。在一家法医医院验证了临床医生评定的邓德拉姆 - 3（项目完成情况）和邓德拉姆 - 4（康复情况）量表后，我们着手起草并验证患者可用于自评的、测量相同项目完成情况和康复项目的量表。基于之前的研究工作，我们假设自评分数可能是包括有条件出院在内的客观进展的预测指标。我们还假设，随着患者临近出院，患者与临床医生在治疗进展及其他与出院准备相关因素（一致性）方面的评分差异会减小。我们还假设，匹配分数之间的这种差异会预测包括有条件出院在内的客观进展。
在一家法医医院进行的前瞻性自然观察队列研究中，我们检验了自评的邓德拉姆 - 3项目完成量表和邓德拉姆 - 4康复量表的得分，或者临床医生与患者在相同量表上的评分差异（一致性）是否能预测未来十二个月内治疗安全级别之间的转换以及有条件出院情况。
两个量表都沿着医院的康复路径进行了分层，但临床医生的评分比自评更准确地匹配了治疗安全级别。临床医生评定的量表预测了向安全性较低病房的转移、向安全性较高病房的转移以及有条件出院情况，但自评分数未能做到。临床医生与自评分数之间的差异（一致性）预测了正向和负向转移以及有条件出院情况，但回归分析表明，这并不总是一个独立的预测指标。在回归分析中，邓德拉姆 - 3量表预测了向安全性较低地方的转移，尽管HCR - 20风险评估量表的C&R得分在模型中占主导地位。邓德拉姆 - 4量表上缺乏一致性预测了向安全性较高地方的转回。有条件出院主要由邓德拉姆 - 3量表预测。
相对于其他患者，患者能够准确地进行自评，然而他们的绝对评分始终低于（优于）临床医生的评分，并且在预测包括有条件出院在内的结果方面准确性较低。量化一致性是康复过程中的一个有用部分，并且可以预测结果，但自评不是准确的预测指标。
