Effectiveness of antipsychotic treatments in a nationwide cohort of patients in community care after first hospitalisation due to schizophrenia and schizoaffective disorder: observational follow-up study

Dear Sirs,

In his rapid response ”Stop irresponsible conclusions” Peter Moleman
claims that “authors should prove that there is no difference between the
groups of patients receiving different drugs” and “no relevant data are
available about these patient characteristics”. Of course groups of
patients receiving different drugs differ from each other and, of course,
we have taken this into account in the statistical analysis. If Moleman
had looked at the Figure and the Table he would have noticed that in
addition to crude relative risk, also adjusted relative risks are
presented. E.g., in the Figure text we explained that relative risks of
rehospitalization are adjusted for sex, age at onset of follow-up, number
of previous relapses, duration of first hospitalization, calendar year,
and length of follow-up by a multivariate regression and propensity score
method (fully adjusted column). This propensity score method is generally
considered the best method to adjust the effect of confounding factors in
observational studies. A more comprehensive description of demographic and
clinical variables is shown in Table 1 of the longer online version of the
manuscript (DOI 10.1136/bmj.38881.382755.2F). Clozapine use was associated
with the lowest risk of all cause discontinuation of initial medication,
and Moleman asks “what are the authors comparing here?” We compared the
risk of discontinuation among those patients who bought clozapine within
30 days of discharge after their first hospital treatment period vs. those
patients who bought haloperidol. (The first antipsychotic medications used
in the community.)

Mercy Ochuko-Emore states in the response “Compliance: a Factor” that
“perphenazine depot outperforming both old and new generation oral
antipsychotic may be as a result of poorer compliance with oral medication
and nothing to do with effectiveness”. We have addressed this issue in our
article: Our results show that depot perphenazine is associated with
markedly better outcome than oral perphenazine. We stated in our original
version of the manuscript that this is probably explained by the route of
administration (depot being superior to oral), but this part of text was
omitted in the edition process (due to word limit). We would like to state
that there is a difference between efficacy and effectiveness: Compliance
may have nothing to do with efficacy, but it is a crucial factor
contributing to effectiveness.

In their response “Antipsychotics are not equally effective”, Peter
Haddad and Omair Niaz suggest that, concerning route of administration,
depot medications are superior to oral medications. We totally agree with
this comment.

Jari Tiihonen
Department of Forensic Psychiatry
University of Kuopio
Niuvanniemi Hospital
FI-70240 Kuopio, Finland
E-mail: jari.tiihonen@niuva.fi

Competing interests:
None declared