Clinical effectiveness of health visitor training in psychologically informed approaches for depression in postnatal women: pragmatic cluster randomised trial in primary care
Sir, I think there may be some problems with the reporting of this
trial. The protocol, accessible here http://www.controlled-
trials.com/isrctn/pf/92195776 describes a three-arm trial but the report
is of a two-arm one. Figure 1, the trial flow diagram, is particularly
misleading. The two arms receiving cognitive behaviour and person centred
therapy appear to have been combined for the primary and most other
analyses, although this was not mentioned in the protocol.
The primary outcomes [sic] in the protocol, “changes in symptoms,
health outcomes, hospital admissions, NHS service use, family well-being
and infant progress to eighteen months” are quite different from the
primary outcome in the paper “proportion of women with a six week EPDS
>= 12, who had a six month EPDS score >= 12.
These alterations in the comparison groups, primary outcome and in
the time point of assessment might make some readers suspect data
dredging.
The authors are also rather quick to dismiss the higher loss to
follow-up in the treatment group 133/404 (33%) compared with controls
44/191 (23%). This is unlikely to have occurred by chance. If more
severely affected participants were more likely to be lost to follow-up,
as has been reported many times in other studies, this would bias the
results in favour of the treatment.
Finally, it is disappointing that, with one exception, the only
outcomes reported were questionnaire scores. The exception, self-reported
anti-depressant prescription, is only mentioned in the discussion section
and has a different denominator to the rest of the results. Presumably
suicides, suicide attempts, and hospital admissions etc. will be reported
elsewhere. This is not good reporting practice.
I remain to be convinced that the two interventions are effective.
Rapid Response:
Trial reporting problems
Sir, I think there may be some problems with the reporting of this
trial. The protocol, accessible here http://www.controlled-
trials.com/isrctn/pf/92195776 describes a three-arm trial but the report
is of a two-arm one. Figure 1, the trial flow diagram, is particularly
misleading. The two arms receiving cognitive behaviour and person centred
therapy appear to have been combined for the primary and most other
analyses, although this was not mentioned in the protocol.
The primary outcomes [sic] in the protocol, “changes in symptoms,
health outcomes, hospital admissions, NHS service use, family well-being
and infant progress to eighteen months” are quite different from the
primary outcome in the paper “proportion of women with a six week EPDS
>= 12, who had a six month EPDS score >= 12.
These alterations in the comparison groups, primary outcome and in
the time point of assessment might make some readers suspect data
dredging.
The authors are also rather quick to dismiss the higher loss to
follow-up in the treatment group 133/404 (33%) compared with controls
44/191 (23%). This is unlikely to have occurred by chance. If more
severely affected participants were more likely to be lost to follow-up,
as has been reported many times in other studies, this would bias the
results in favour of the treatment.
Finally, it is disappointing that, with one exception, the only
outcomes reported were questionnaire scores. The exception, self-reported
anti-depressant prescription, is only mentioned in the discussion section
and has a different denominator to the rest of the results. Presumably
suicides, suicide attempts, and hospital admissions etc. will be reported
elsewhere. This is not good reporting practice.
I remain to be convinced that the two interventions are effective.
Competing interests:
None declared
Competing interests: No competing interests