BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Using Rasch Analysis in Neurological and Community Rehabilitation 
 -  The SARDINE Project (Secondary Analysis of Routine Data in Neurorehab E
 valuations) - Dr Andrew Bateman (Dept of Psychiatry and Oliver Zangwill Ce
 ntre\, Ely)
DTSTART:20130430T153000Z
DTEND:20130430T163000Z
UID:TALK42975@talks.cam.ac.uk
CONTACT:Luning Sun
DESCRIPTION:On the Matilda Bay Club website Ben Wright commented  "as long
  as primitive counts and raw scores are routinely mistaken for measures by
  our colleagues in Social\, Educational and Health research\, there is no 
 hope of their professional activities ever developing into a reliable or u
 seful science. We owe it to them\, and to ourselves\, to teach them how to
  construct measures which work as well as the ubiquitous physical measures
  by which they manage their everyday living\, so that they can do a better
  job in making sense out of the profusions of data which they collect so e
 nthusiastically". [Wright 1999 http://www2.wu-wien.ac.at/marketing/mbc/mbc
 .html]\n\nAs an NHS Manager I am  indeed faced with "profusions of data" a
 nd having benefited from "how to" analysis training\, I have been working 
 away to understand the Patient Reported Outcomes we are collecting routine
 ly in the various clinical settings in Cambridgeshire Community Services N
 HS Trust.\n\nAt the Oliver Zangwill Centre for Neuropsychological Rehabili
 tation we have examined the properties a brain injury symptom checklist (t
 he EBIQ)\, the Dysexecutive Questionnaire (DEX)\, a self-criticism tool\, 
 and most recently an assessment of emotion perception.   These have been w
 ith relatively small sample sizes.  In the wider community services the ge
 neric Health Related Quality of Life tool the EQ5D-5L has been adopted\, w
 ith a substantial flow of data (>500 patients per month).\n\nIn each case 
 the benefits of a Rasch approach have been clear: 1) examination of catego
 ry probability curves and demonstration of disordered thresholds points to
  the difficulties many patients have distinguishing between the Likert cat
 egories conceived by questionnaire designers.  The outcomes enable a reaso
 ned  approach to item re-scoring 2)  item properties revealed by tests of 
 Item Bias have uncovered DIF that has provided interesting insights into c
 hanges between cultures\, pathologies and age-groups. 3) evaluation of the
  targetting of a tool contributes to the confidence one can have in the as
 sessments we are using.  The analyses enable improvements in the tools\, i
 n the scoring of the tools and prioritisation of services.  I will illustr
 ate each of these points.\n\nIn each case the findings have been both clin
 ically informative at the same time as challenging measurement assumptions
  underlying the tools.  The overall objective is to achieve a mapping of t
 he outcome space for community rehabilitation and to be able to describe c
 hanges that are achieved by our patients.  \n\nThe challenges ahead includ
 e the need to automate analyses for clinical activity monitoring\, to expl
 ain the findings clearly and thereby enable NHS service efficiency.  Wider
  collaboration between the clinical and psychometric communities is encour
 aged.
LOCATION:Seminar Room\, Department of Psychology\, Downing Site\, Cambridg
 e
END:VEVENT
END:VCALENDAR