Meta-synthesis of qualitative studies of psychosis 1
This is an Author’s manuscript of the paper published under the following Citation:
MCCARTHY-JONES, S., MARRIOTT, M., KNOWLES, R., ROWSE, G. and THOMPSON,
A.R., 2013. What is psychosis? A meta-synthesis of inductive qualitative studies
exploring the experience of psychosis. Psychosis, 5 (1), pp. 1-16. ISSN 1752-2439
The original file in published format is available online as follows:
http://dx.doi.org/10.1080/17522439.2011.647051
Meta-synthesis of qualitative studies of psychosis 2
Abstract
Qualitative studies have played an important role in elucidating the lived experience of
psychosis and there has recently been an increase in the number of such studies. There is
now an urgent need to draw together the findings of these studies. This paper performed a
meta-synthesis of inductive qualitative peer-reviewed research into psychosis. Ninety-eight
articles were identified for systematic appraisal. Four themes, ‘Losing’, ‘Identifying a need
for, and seeking, help’, ‘Rebuilding and reforging’, and ‘Better than new: gifts from
psychosis’, were identified. The important implications these themes for clinicians and future
research are examined upon. These findings also highlight that the experience of psychosis is
much more than simply just hallucinations and/or delusions.
Keywords: Human needs, recovery, schizophrenia.
Meta-synthesis of qualitative studies of psychosis 3
The past decade has seen the emergence of a significant body of research that seeks to
explore individual’s experience of psychosis. Methodologically, qualitative research is most
suited to this aim, and in particular inductive qualitative research whose findings are derived
primarily from in-depth ‘data’ derived from engagement with individuals (Willig & Stainton-
Rogers, 2008). The findings from these qualitative studies of psychosis offer a rich and deep
source of information that can help mental health professionals understand their client’s
experiences.
A limitation of individual qualitative studies of psychosis is the extent to which they
actually influence policy. Studies may lack impact due to their very nature of being rich
individualized accounts or concerns over their generalizability (Morse et al., 1998). Walsh
and Downe (2005) have highlighted the danger of such research being marginalised by
policymakers and clinicians if there is no attempt to connect between studies. One way to
address this is through meta-synthesis, a technique which aims to make qualitative findings
more accessible for practical application by amalgamating individual studies (Zimmer, 2006),
and which may also help create new knowledge through identifying new overarching themes.
This paper aimed to perform a meta-synthesis of peer-reviewed inductive qualitative research
into lived experiences of psychosis. In doing so it attempted to formalise and re-conceptualize
themes from these studies to create new, more broadly applicable knowledge of the meaning
of psychosis, likely to be of direct use to mental health professionals, policymakers, and
service-users themselves. The selection of inductive qualitative studies involving systematic
analysis of first-hand accounts was not intended to suggest a superiority of these studies
either over other research designs, but simply to provide an understanding of a specific
section of literature that has a unique, but not comprehensive, contribution to make to the
understanding of psychosis and how individuals relate their experiences.
Meta-synthesis of qualitative studies of psychosis 4
Method
Inclusion criteria were that the study should be an inductive qualitative study of some aspect
of psychosis, be based on analyses of first-hand accounts by the person with psychosis, and
be published in an English language peer-reviewed journal between January 2000 and May
2010. Exclusion criteria were that the study was a single person case-study, an ethnographic
study (e.g., solely based on participation/observation), or if it lacked evidence that its
conclusions reflected inductive analysis grounded in the data. We limited our review to post-
January 2000 to generate a quantity of information that was practically synthesizable and also
to acknowledge that changes in society and practice may lead experiences of psychosis to
differ in different eras.
Relevant papers were identified by a search of the MEDLINE and Scopus databases.
The search string employed was: (schizophren* OR psychosis OR psychotic OR hallucin*
OR delusion*) AND (“in-depth interview” OR qualitative* OR interpretive* OR transcribe*
OR “focus group” OR “grounded theory” OR hermeneuti* OR open-ended OR narrative*).
Reference sections and citation reports from the papers selected for inclusion in this paper
were also examined. This search identified 2,635 potentially relevant papers. The abstracts of
these papers were then examined. We initially erred on the side of inclusiveness when
identifying relevant qualitative literature (Dixon-Woods et al., 2007), identifying 175 papers
as potentially relevant. The full-texts of these papers were re-examined in detail and the
formal inclusion/exclusion criteria applied. This led to the exclusion of a further 77 papers,
leaving 98 papers for inclusion. A descriptive table of these studies is available from the
authors on request. Although the quality of papers varied widely we did not feel it necessary
to exclude any of these papers on grounds of quality.
We followed the method of analysis outlined by Walsh and Downe (2005), beginning
by identifying what we perceived as key metaphors, phrases, ideas, and concepts in each
Meta-synthesis of qualitative studies of psychosis 5
study, adding our own words or paraphrases. These were then sorted into broader themes.
The next stage was reciprocal translation, in which we tried to transpose the findings of an
individual study into those of another, using metaphors and concepts that were applicable to
both, whilst being careful to make sure that differences were not glossed over or subsumed.
Finally, translated concepts were synthesized to elucidate more refined meanings, exploratory
theories and new concepts.
Validity of themes was addressed by having initial themes developed independently
by two of the researchers. These were then critically explored, and modified through
discussion with the other authors. We addressed internal validity by including original
quotations from respondents, and external validity by triangulating our findings with the
quantitative literature. A reflexive approach was taken throughout the study by considering
how our own training, theoretical positions and personal beliefs were likely to impact our
analyses.
Results
The 98 studies synthesised had a total of 1,945 participants (M = 19.92, SD = 22.77). Of the
studies that reported specific data on age and gender, there was a mean participant age of
34.48 years (SD = 9.00) with 55% of participants being male. The ratio of studies that used
first episode patients or patients from early intervention services to studies that employed
older, more chronic patients was approximately 1:3. Four themes relating to the experience of
psychosis were inductively generated.
Theme 1: Losing
This theme centred on the devastating losses psychosis can result in, ranging from the loss of
the very existential ground under one’s feet, to the loss of the people in one’s heart.
Sub-Theme I: The loss of consensual reality