Meadow should never have been brought before the GMC

Sir, The Sally Clark case is in the news again.[1] Throughout, the
debate has
focused – incorrectly – on Meadow’s evidence about the rarity of two SIDS
cases occurring in the same family. We write to emphasise another, more
important issue that the case raised about the interpretation of evidence
in
Court. Equally importantly, we wish to highlight the broader implications;
the
same issue arises in other contexts, e.g. care proceedings for infants
failing
to thrive where parental neglect is suspected.

As an illustration, we reviewed an 11 month infant with failure to
thrive
growing below the 1st weight velocity centile (slowest-growing 1% of the
population).[2] Failure to thrive (FTT) was the key reason that Social
Services
sought a Care Order to remove the baby from the parents, on the grounds of

neglect. The growth pattern was so rare that the parents were deemed
unlikely to be innocent.

Legal proceedings in the two cases, ours and Sally Clark’s, gave
weight to the
rarity of the event (<_1 and="and" _1="_1" per="per" _73="_73" million="million" _3="_3" respectively="respectively" though="though" the="the" second="second" figure="figure" was="was" incorrect="incorrect" for="for" at="at" least="least" three="three" reasons="reasons" _4.="_4." however="however" it="it" is="is" flawed="flawed" reasoning="reasoning" to="to" base="base" parental="parental" innocence="innocence" or="or" guilt="guilt" on="on" proportion="proportion" of="of" population="population" in="in" which="which" rare="rare" event="event" occurs.="occurs." quite="quite" a="a" different="different" statistic="statistic" that="that" relevant="relevant" court="court" proceedings.="proceedings." critical="critical" issue="issue" chance="chance" i="i"/>in such
rare cases that the parents are innocent – in what proportion of
FTT
cases, or in what proportion of double unexplained infant deaths, are the
parents blameless?

This probability of parental innocence can range from 100% (no
parents
guilty) to 0% (all guilty). In Sally Clark’s case the relevant figure,
i.e. the
proportion of families with two unexplained deaths in which the parents
were
considered innocent, has been estimated as 82-90%.[4] In our own FTT
example the probability of innocence is similar; Skuse et al [5] found
that
over 90% of 2609 cases of faltering growth were not related to parental
neglect or abuse. So, in stark contrast to the initial assumption of
guilt, based
on the rarity of the event, the parents in both situations were nine times
more
likely to be innocent than guilty.

Thus in Court proceedings, rarity of presentation is not relevant.
The key
issue is the statistical chance, in the presence of this rare
presentation, that
the accused party is innocent. This principle that we highlight applies
equally
in other medico-legal areas.

1. Dyer C. Meadow should never have been brought before the GMC. BMJ
2006;332:439.

2. Cole TJ. Conditional reference charts to assess weight gain in
British
infants. Arch Dis Child 1995;73:8-16.

3. Fleming P, Blair P, Bacon C, Berry PJ, editors. Sudden unexpected
deaths in
infancy: the CESDI SUDI studies 1993-1996. London: The Stationery Office;
2000, p 92.

4. Hill R. Multiple sudden infant deaths – coincidence or beyond
coincidence?
Paediatr Perinat Epidemiol 2004;18:320–326.

5. Skuse DH, Gill D, Reilly S, et al. Failure to thrive and risk of
child abuse: a
prospective study. J Med Screen 1995;2:145-149.

Competing interests:
None declared