Validation of an algorithm-based definition of treatment resistance in patients with schizophrenia

SCHIZOPHRENIA RESEARCH

卷:197

Article

Ajnakina, Olesya¹  Horsdal, Henriette Thisted^2,3,4  Lally, John^1,5,6  MacCabe, James H.^1,7,8  Murray, Robin M.^1,7,8  Gasse, Christiane^2,3,4  Wimberley, Theresa^2,3,4 

[1] Kings Coll London, Inst Psychiat Psychol & Neurosci, Dept Psychosis Studies, London, England

[2] Aarhus Univ, Aarhus BSS, Natl Ctr Register Based Res, Aarhus, Denmark

[3] IPSYCH, Aarhus, Denmark

[4] Lundbeck Fdn Initiat Integrat Psychiat Res, Aarhus, Denmark

[5] Beaumont Hosp, Royal Coll Surg Ireland, Dept Psychiat, Dublin, Ireland

[6] St Vincents Univ Hosp, Univ Coll Dublin, Sch Med & Med Sci, Dept Psychiat, Dublin, Ireland

[7] South London & Maudsley NHS Fdn Trust, Mental Hlth Biomed Res Ctr, NIHR, London, England

[8] Kings Coll London, London, England

关键词: Treatment resistance;    Clozapine;    Schizophrenia;    Validation;    Sensitivity;    Specificity;    Positive predictive value;    Negative predictive value;   

DOI  :  10.1016/j.schres.2018.02.017

来源: Elsevier

【 摘 要 】

Large-scale pharmacoepidemiological research on treatment resistance relies on accurate identification of people with treatment-resistant schizophrenia (TRS) based on data that are retrievable from administrative registers. This is usually approached by operationalising clinical treatment guidelines by using prescription and hospital admission information. We examined the accuracy of an algorithm-based definition of TRS based on clozapine prescription and/ or meeting algorithm-based eligibility criteria for clozapine against a gold standard definition using case notes. We additionally validated a definition entirely based on clozapine prescription. 139 schizophrenia patients aged 18-65 years were followed for a mean of 5 years after first presentation to psychiatric services in South-London, UK. The diagnostic accuracy of the algorithm-based measure against the gold standard was measured with sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). A total of 45 (32.4%) schizophrenia patients met the criteria for the gold standard definition of TRS; applying the algorithm-based definition to the same cohort led to 44 (31.7%) patients fulfilling criteria for TRSwith sensitivity, specificity, PPV and NPV of 62.2%, 83.0%, 63.6% and 82.1%, respectively. The definition based on lifetime clozapine prescription had sensitivity, specificity, PPV and NPV of 40.0%, 94.7%, 78.3% and 76.7%, respectively. Although a perfect definition of TRS cannot be derived from available prescription and hospital registers, these results indicate that researchers can confidently use registries to identify individuals with TRS for research and clinical practices. (C) 2018 Elsevier B.V. All rights reserved.

【 授权许可】

Free