Applications of temporal kernel canonical correlation analysis in adherence studies

Adherence to medication is often measured as a continuous outcome but analyzed as a dichotomous outcome due to lack of appropriate tools. In this paper, we illustrate the use of the temporal kernel canonical correlation analysis (tkCCA) as a method to analyze adherence measurements and symptom levels on a continuous scale. The tkCCA is a novel method developed for studying the relationship between neural signals and hemodynamic response detected by functional MRI during spontaneous activity. Although the tkCCA is a powerful tool, it has not been utilized outside the application that it was originally developed for. In this paper, we simulate time series of symptoms and adherence levels for patients with a hypothetical brain disorder and show how the tkCCA can be used to understand the relationship between them. We also examine, via simulations, the behavior of the tkCCA under various missing value mechanisms and imputation methods. Finally, we apply the tkCCA to a real data example of psychotic symptoms and adherence levels obtained from a study based on subjects with a first episode of schizophrenia, schizophreniform or schizoaffective disorder.

[1]  L. Osterberg,et al.  Adherence to medication. , 2005, The New England journal of medicine.

[2]  Robert A McNutt,et al.  Decompensated heart failure: symptoms, patterns of onset, and contributing factors. , 2003, The American journal of medicine.

[3]  S. Siegel.  Extended release drug delivery strategies in psychiatry: theory to practice. , 2005, Psychiatry (Edgmont (Pa. : Township)).

[4]  R. Horne.  Adherence to Medication: A Review of Existing Research , 1998 .

[5]  J. Lieberman,et al.  Predictors of medication discontinuation by patients with first-episode schizophrenia and schizoaffective disorder , 2002, Schizophrenia Research.

[6]  J. Schafer,et al.  Missing data: our view of the state of the art. , 2002, Psychological methods.

[7]  K. Nuechterlein,et al.  Risperidone nonadherence and return of positive symptoms in the early course of schizophrenia. , 2011, The American journal of psychiatry.

[8]  D W Bates,et al.  Practical approach to determining costs and frequency of adverse drug events in a health care network. , 2001, American journal of health-system pharmacy : AJHP : official journal of the American Society of Health-System Pharmacists.

[9]  Richard A. Johnson,et al.  Applied Multivariate Statistical Analysis , 1983 .

[10]  P. Rodgers,et al.  Medication nonadherence: Part II--A pilot study in patients with congestive heart failure. , 1998, Managed care interface.

[11]  K. Nuechterlein,et al.  Classifying episodes in schizophrenia and bipolar disorder: Criteria for relapse and remission applied to recent-onset samples , 2006, Psychiatry Research.

[12]  Michael I. Jordan,et al.  Kernel independent component analysis , 2003 .

[13]  P. McDonnell,et al.  Hospital Admissions Resulting from Preventable Adverse Drug Reactions , 2002, The Annals of pharmacotherapy.

[14]  Kenji Fukumizu,et al.  Statistical Consistency of Kernel Canonical Correlation Analysis , 2007 .

[15]  C. Camp,et al.  A Fluctuating Pattern of Over- and Under-Adherence for HAART: A Case Study from a Videophone Intervention Project , 2011, Clinical gerontologist.

[16]  J. Lieberman,et al.  Predictors of relapse following response from a first episode of schizophrenia or schizoaffective disorder. , 1999, Archives of general psychiatry.

[17]  Klaus-Robert Müller,et al.  Temporal kernel CCA and its application in multimodal neuronal data analysis , 2010, Machine Learning.

[18]  R. Liberman,et al.  Brief Psychiatric Rating Scale (BPRS) Expanded Version (4.0): Scales, Anchor Points, and Administration Manual , 1993 .

[19]  P. Weiden,et al.  The expert consensus guideline series: adherence problems in patients with serious and persistent mental illness. , 2009, The Journal of clinical psychiatry.

[20]  Klaus-Robert Müller,et al.  Relationship between neural and hemodynamic signals during spontaneous activity studied with temporal kernel CCA. , 2010, Magnetic resonance imaging.

[21]  Shotaro Akaho,et al.  A kernel method for canonical correlation analysis , 2006, ArXiv.

[22]  H. Möller,et al.  Relapse prevention in first-episode schizophrenia--maintenance vs intermittent drug treatment with prodrome-based early intervention: results of a randomized controlled trial within the German Research Network on Schizophrenia. , 2011, The Journal of clinical psychiatry.

[23]  D. Rubin.  INFERENCE AND MISSING DATA , 1975 .

[24]  Kristopher J Preacher,et al.  On the practice of dichotomization of quantitative variables. , 2002, Psychological methods.