BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//jEvents 2.0 for Joomla//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:STANDARD
DTSTART:20171203T150000
RDATE:20180325T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:Europe/London GMT
END:STANDARD
BEGIN:STANDARD
DTSTART:20181028T010000
RDATE:20190331T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:Europe/London GMT
END:STANDARD
BEGIN:STANDARD
DTSTART:20191027T010000
RDATE:20200329T020000
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:Europe/London GMT
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20180325T020000
RDATE:20181028T010000
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:Europe/London BST
END:DAYLIGHT
BEGIN:DAYLIGHT
DTSTART:20190331T020000
RDATE:20191027T010000
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:Europe/London BST
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:2b7ca1b859510c0f6f624f3f478c0b01
CATEGORIES:RSS sections/special interest groups, Medical Section
CREATED:20181106T175001
SUMMARY:Multiple imputation 40 years on, where are we now?
LOCATION:15 Hatfields, Chadwick Court, London, SE1 8DJ
DESCRIPTION;ENCODING=QUOTED-PRINTABLE: About the Event\n\nIt has been exactly 40 years since the seminal pap
er introducing multiple imputation to handle the problem of missing data wa
s published. Since then, this approach has led the way in developing princi
pled statistical methodology to address this problem. The approach has also
proved to be widely appealing, with uptake popular in many disciplines. In
this session we aim to bring together state of the art developments in mul
tiple imputation research in the area of medical statistics. In particular,
we will look at the use of Multiple Imputation beyond in RCTs. The se
ssion speakers comprise some of the world leaders and experts in the field.
\n\n\n15:00 – 15:40\n\n\nSpeaker: Ian White, University College London\n\n\
nTitle: Multiple imputation: the universal panacea, and its limitation
s\n\n\nAbstract: Ian will review some of the developments in theory and sof
tware that led to multiple imputation being seen by some people as a univer
sal solution to missing data problems. He will then explain why it isn’t, d
iscussing alternatives to multiple imputation, difficulties of imputing mul
tilevel data, and avoiding the untestable missing at random assumption.\n\n
\n \n\n\n15:40 – 16:20\n\n\nSpeaker: Tra Pham, University College Lond
on\n\n\nTitle: Population-calibrated multiple imputation for a binary/categ
orical covariate in categorical regression models\n\n\nAbstract: Multiple i
mputation (MI) has become popular for analyses with missing data in medical
research. The standard implementation of MI is based on the assumption of
data being missing at random (MAR). However, for missing data generated by
missing not at random (MNAR) mechanisms, MI performed assuming MAR might no
t be satisfactory. For an incomplete variable in a given data set, its corr
esponding population marginal distribution might also be available in an ex
ternal data source. We show how this information can be utilised in the imp
utation model to calibrate inference to the population by incorporating an
appropriately calculated offset termed the "calibrated-? adjustment". We de
scribe the derivation of this offset from the population distribution of th
e incomplete variable and show how, in applications, it can be used to clos
ely (and often exactly) match the post-imputation distribution to the popul
ation level. Through analytic and simulation studies of a binary/categorica
l covariate in categorical regression models, we show that our proposed cal
ibrated-? adjustment MI method can give the same inference as standard MI w
hen data are MAR, and can produce more accurate inference under two general
MNAR mechanisms. The method is used to impute missing ethnicity data in a
type 2 diabetes prevalence case study using UK primary care electronic heal
th records. Calibrated-? adjustment MI represents a pragmatic approach for
utilising available population-level information in a sensitivity analysis
to explore potential departures from the MAR assumption.\n\n\n \n\n\n1
6:20 – 16:50: Break for refreshments\n\n\n \n\n\n16:50 – 17:30\n\n\nSp
eaker: James Carpenter, London School of Hygiene and Tropical Medicine\n\n\
n16:50 – 17:30\n\n\nSpeaker: James Carpenter, London School of Hygiene and
Tropical Medicine\nTitle: Sensitivity analysis for missing trial outcomes:
what can it do for you?\n\n\nAbstract: James Carpenter (1,2) and Suzie
Cro (3)James Carpenter (1,2) and Suzie Cro (3)\n\n\n1. Department of Medic
al Statistics, London School of Hygiene & Tropical Medicine\n2. MRC Cli
nical Trials Unit at UCL\n3. Imperial Clinical Trials Unit, Imperial Colleg
e London\n\n\nMissing outcome data are almost inevitable in clinical trials
, for example due to inter-current events such as treatment withdrawal, tre
atment switching or loss to follow-up. In such settings, the analysis can o
nly proceed on the basis of an untestable assumption about the missing outc
ome data. In applications, it is therefore important to understand the robu
stness of conclusions to a range of plausible assumptions about the distrib
ution of the missing outcomes.\nIn this talk, we outline two approaches for
this: (a) reference based imputation, where missing outcomes are imputed b
y reference to other patient groups [1], and (b) eliciting expert opinion o
n the distribution of missing values and incorporating this into the analys
is [2]. For each approach, we discuss the assumptions made about the missin
g data, implementation using multiple imputation, and give an illustrative
application. We conclude with a discussion of the pros and cons of each app
roach, and how they may be used to address some of the challenges raised by
the ICH-E9 addendum on estimands.\n\n\nReferences:\n[1] Cro, S., Carpenter
, J. R. and Kenward, M. G. (2018) Information-anchored sensitivity analysis
: theory and application. Journal of the Royal Statistical Society, Series
A. https://rss.onlinelibrary.wiley.com/doi/epdf/10.1111/rssa.12423\n[2] Mas
on, A. J., Gomes M., Grieve, M. Ulug, P., Powell, J. T. and Carpenter J. R.
(2017). Development of a practical approach to expert elicitation for tria
ls with missing health outcomes: application to the IMPROVE trial. Clinical
Trials, 14, 357-367. https://doi.org/10.1177/1740774517711442\n\n\n17:30 –
18:00, Panel discussion\n\n\nTea and coffee will be available from 2.30pm\
n\n\nRegistration is required - the event is free to RSS Fellows with a £25
registration fee for non-Fellows\n\n\n\n\n\n \n
X-EXTRAINFO:The meeting will be preceded by the Annual General Meeting of the Medical S
ection\n
DTSTAMP:20190918T151355Z
DTSTART;TZID=Europe/London:20181204T150000
DTEND;TZID=Europe/London:20181204T180000
SEQUENCE:0
TRANSP:OPAQUE
END:VEVENT
END:VCALENDAR