Schizophrenia Bulletin vol. 33 no. 1 pp.

192–199, 2007
doi:10.1093/schbul/sbl016
Advance Access publication on August 7, 2006

Toward Understanding the Insight Paradox: Internalized Stigma Moderates the

Association Between Insight and Social Functioning, Hope, and Self-esteem
Among People with Schizophrenia Spectrum Disorders

Downloaded from https://academic.oup.com/schizophreniabulletin/article/33/1/192/1922956 by guest on 08 October 2023

Paul H. Lysaker1–3, David Roe4, and Philip T. Yanos5,6 Relative to persons with other psychiatric disorders, per-
2
Day Hospital 116H, 1481 West 10th Street, Roudebush VA sons with schizophrenia spectrum disorders are often un-
Medical Center, Indianapolis, IN 46202; 3Department of Psychi- aware or willfully contest that they have what others
atry, Indiana University School of Medicine, Indianapolis, IN think to be a mental illness.1,2 Taken as a whole, this phe-
46202; 4Department of Psychiatric Rehabilitation and Behavioral nomenon, often referred to as ‘‘lack of awareness’’ or
Health Care, University of Medicine and Dentistry of New Jersey, ‘‘poor insight,’’ has shown a pattern of apparently con-
Scotch Plains, NJ; 5Department of Psychiatry, University of
Medicine and Dentistry of New Jersey, New Jersey Medical School, tradictory associations with outcome. For instance, poor
Newark, NJ; 6Institute for Health, Health Care Policy, and Aging insight has been linked to poorer treatment adherence,3,4
Research, Rutgers University, New Brunswick, NJ poorer clinical outcome,5 poorer social function,6–8 voca-
tional dysfunction,9 and difficulties developing working
relationships with mental health professionals.10 On the
Research has paradoxically linked awareness of illness to other hand, greater insight has been associated with
both better function outcomes and lesser hope and self- higher levels of dysphoria,11–14 lowered self-esteem,15
esteem. One possible explanation for these findings is that and decreased well-being and quality of life.16–21
acceptance of having schizophrenia may impact outcomes One possible explanation for the seemingly contradic-
differently depending on the meanings the person attaches tory findings is that the impact of the acceptance of
to this acceptance, particularly whether he or she accepts schizophrenia depends on the meanings persons attach
stigmatizing beliefs about mental illness. To explore this to schizophrenia.22 For example, if one believes the illness
possibility we performed a cluster analysis of 75 persons means that he or she is not capable of achieving valued
with schizophrenia spectrum disorders based on single social roles, then awareness could lead to hopelessness
measures of insight using the Positive and Negative Syn- and less motivation to persevere. However, if one does
drome Scale, internalized stigma using the Internalized not believe that this illness precludes chances for a satis-
Stigma of Mental Illness Scale, and compared groups on fying life, then awareness of the illness may be a key part
concurrent assessments of hope and self-esteem. Three of negotiating the challenges posed by the symptoms
groups were produced by the cluster analyses: low in sight/ and diagnostic label. Supporting this view is research
mild stigma (n = 23), high insight/minimal stigma (n = 25), suggesting ‘‘role engulfment,’’ or the extent to which
and high insight/moderate stigma (n = 27). As predicted, a person adopts the identity of a ‘‘mental patient,’’23 is
analysis of variance–comparing groups revealed that the associated with both better insight24 and poorer social
high insight/moderate stigma group had significantly the functioning.25 Internalized stigma,26 a related construct
lowest levels of hope on the Beck Hopelessness Scale and reflecting the degree to which a person has internalized
self-esteem using the Multidimensional Self-esteem Inven- societally endorsed stigmatizing beliefs about mental ill-
tory. As predicted, the high insight/minimal stigma group ness (eg, people with mental illness are violent and unable
also had significantly less impaired social function than the to function independently), has been found to take on
other groups. Implications for assisting persons to come to personal relevance when a person is diagnosed with men-
cope with awareness of illness and stigma are discussed. tal illness27 and has been linked to depressed mood.26
Also consistent with this is a longitudinal qualitative
Key words: schizophrenia/stigma/quality of life/insight/ study that reported that the transformation of ‘‘sense
hope/social function/narrative of self’’ from one of ‘‘patienthood’’ to ‘‘personhood’’ was
an important element in the process of improving func-
tioning in a group with severe mental illness.28
1
To whom correspondence should be addressed; e-mail: While these findings support the view that the mean-
[email protected]. ings a person attaches to having a mental illness can
Ó The Author 2006. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved.
For permissions, please email: [email protected].
192
 Toward Understanding the Insight Paradox

Stigma rejected + Illness accepted higher hope/self-esteem, better adaptation to illness better social functioning
Stigma accepted + Illness accepted lower hope/self-esteem, better adaptation to illness poorer social functioning
Stigma accepted + Illness rejected higher hope/self-esteem, poorer adaptation to illness poorer social functioning

Fig. 1. The interactions of stigma and insight and their relationship to hope, self-esteem, and social function.

have important implications for how awareness impacts Disorders, Fourth Edition (DSM-IV) (SCID34)-confirmed
outcomes, stigma’s impact on the effects of insight on DSM-IV diagnoses of schizophrenia (n = 40) or schizo-
outcome in schizophrenia has yet to be explicitly studied. affective disorder (n = 35) were recruited from a compre-
To investigate this we have sought to determine whether hensive day hospital at a VA Medical Center (n = 55) and

Downloaded from https://academic.oup.com/schizophreniabulletin/article/33/1/192/1922956 by guest on 08 October 2023

we could distinguish groups of persons with schizophre- local Community Mental Health Center (CMHC; n = 20)
nia who varied according to level of insight and internal- for a larger survey of the prevalence of anxiety symptoms
ized stigma and then to see whether these different groups in schizophrenia. All participants were receiving ongoing
differed in hopefulness, self-esteem, and social function. outpatient treatment and were in a postacute or stable
Social functioning includes a person’s capacity for social phase of their disorder, defined as no hospitalizations
relations and the breadth of social relations and is a major or changes in medication or housing in the last month.
outcome of interest for people with schizophrenia.29 Participants with a history of mental retardation docu-
Hopefulness has been found to predict problem-oriented mented in a chart review were excluded. Participants
coping when accompanied by insight30 and is by matter had a mean age of 48.25 years (SD = 7.01), a mean ed-
of definition an appraisal of the future and thus a context ucational level of 12.57 (SD = 2.05), and a mean of
in which persons make decisions about how to pursue life 8.73 (SD = 9.04) lifetime hospitalizations with the first
goals. Self-esteem has been found to predict life satisfac- occurring on average at the age of 25.26 years (SD =
tion and is related to positive outcome among people with 8.37). Thirty-six participants were Caucasian, 37 African
schizophrenia.31–33 American, and 2 Latino.
Accordingly, we made 3 hypotheses. We first predicted Participants were recruited through contact with pri-
that clustering participants in a treatment sample by in- mary providers who informed patients whom they be-
sight and internalized stigma would result in 3 groups: (a) lieved met study criteria. Of those 86 participants who
those who held stigmatizing beliefs about mental illness consented to participate, 11 completed only portions
and believed they had a mental illness; (b) those who did of the assessments and failed to attend follow-up ap-
not endorse having internalized stigmatizing beliefs pointments to complete the assessments. It is estimated
about mental illness and believed they had a mental ill- that the total possible number of persons who could
ness; and (c) those who endorsed stigmatizing beliefs have been recruited was approximately 450 (300 patients
about mental illness but did not have insight or were in the VA and 150 in the CMHC).
not aware of their mental illness. As illustrated in figure
1, we secondly predicted that the group that rejected
stigma but acknowledged their illness and the group Instruments
that accepted stigma but rejected their illness would The Positive and Negative Syndrome Scale (PANSS35) is
have more hope and self-esteem than the group that ac- a 30-item rating scale completed by clinically trained re-
cepted stigma and acknowledged their illness. Third, as search staff at the conclusion of chart review and a semi-
also illustrated in figure 1, we predicted that the group structured interview. For the purposes of this study, 2 of
that rejected stigma and acknowledged their illness would the 5 PANSS factor analytically derived components36
have the best social function of all 3 groups. Here we rea- were used: positive symptoms, which includes symptoms
soned that the lesser hope and self-esteem of the group such as hallucinations and delusions, and negative symp-
that held stigmatizing beliefs about mental illness and be- toms, which includes symptoms such as lack of affect. We
lieved they had a mental illness could erode social func- also included the PANSS insight and judgment item that
tion just as the unawareness of illness in the third group provides a rating of 1 to 7 which reflects global awareness
could erode social function by virtue of its limiting that of symptoms, treatment need, and consequences of ill-
group’s adaptation to illness. We finally planned to com- ness. Assessment of interrater reliability for this study
pare groups on symptom levels for exploratory purposes. was found to be high to excellent with intraclass correla-
tions for blind raters observing the same interview rang-
ing from .84 to .93.
Methods The Quality of Life Scale (QOLS37) is a 21-item scale
completed by clinically trained research staff following
Participants a semistructured interview and chart review. For the pur-
Sixty-four men and 11 women with Structured Clinical poses of this study, we were interested in 2 of the 4 factor
Interview for Diagnostic and Statistical Manual of Mental scores of the QOLS that are most intimately tied to social
193
P. H. Lysaker et al.

function. The first, ‘‘interpersonal relations,’’ measures The Multidimensional Self-esteem Inventory (MSEI39)
the frequency of recent social contacts and includes sep- is a 116-item self-report measure that assesses individuals’
arate assessments, for example, of frequency of contacts self-perception of their overall social value. Respondents
with friends and acquaintances. The second, ‘‘intrapsy- rate items on a 5-point scale according to the degree or
chic foundations,’’ measures qualitative aspects of inter- frequency with which each item applies to them. The
personal relationships and includes assessments, for MSEI offers t scores based on a community sample.
example, of empathy for others. High to excellent inter- The t scores are normalized scores with a mean of 50
rater reliability was found for the 2 QOLS factor scores and a standard deviation of 10. The mean t score for
for this study, with intraclass correlations for blind raters this sample was 44.56 with a standard deviation of
observing the same interview ranging from .85 to .93. Al- 10.33. This suggests that participants’ reported level of

Downloaded from https://academic.oup.com/schizophreniabulletin/article/33/1/192/1922956 by guest on 08 October 2023

though originally created to assess negative symptoms in self-esteem was approximately 0.5 standard deviations
schizophrenia, the QOLS has been widely used to study lower than those of persons in a broad community sam-
social function in this population.7 ple. Because the MSEI has been largely used in samples of
The Internalized Stigma of Mental Illness Scale persons without psychosis, internal consistency was ex-
(ISMIS26,38) is a 29-item paper and pencil questionnaire amined in this sample. Examination of items comprising
designed to assess subjective experience of stigma. It the total score revealed a highly significant degree of in-
presents participants with first-person statements and ternal consistency (coefficient a = .90).
asks them to rate on a 4-point Likert scale regarding The Beck Hopelessness Scale40 is a 20-item question-
whether they ‘‘Strongly disagree,’’ ‘‘Disagree,’’ ‘‘Agree,’’ naire that asks participants to endorse statements as
or ‘‘Strongly agree’’ with statements related to having true or false as applied to them. Individual items are
a mental illness. Items are summed to provide 4 major then summed to provide an overall index of hope or
scale scores: alienation, which reflects feeling devalued its absence. Examples of items include, respectively,
as a member of society; stereotype endorsement, which ‘‘Things just won’t work out the way I want them to’’
reflects agreement with negative stereotypes of mental ill- and ‘‘I might as well give up because I can’t make things
ness; discrimination experience, which reflects current better for myself.’’ This scale has been used successfully
mistreatment attributed to the biases of others; and social with a wide range of psychiatric, medical, and community
withdrawal, which reflects avoidance of others because of populations.41
mental illness. The fifth additional score, stigma resis-
tance, asks about participant’s perceived ability to deflect
Procedures
stigma. All scales scores are calculated as averages with
higher scores suggesting graver experiences of stigma. Ev- All procedures were approved by the research review
idence of acceptable internal consistency, test-retest reli- committees of Indiana University and the Roudebush
ability, factorial, and convergent validity have been VA Medical Center. Following informed consent, diag-
reported, including links with morale and well-being.26,38 noses were determined using the SCID conducted by
The instrument was presented to persons in its written a clinical psychologist. Following the SCID, participants
form with research assistants available to assist if partic- were administered the PANSS and QOLS interviews,
ipants were confused about the meaning of any item. MSEI, and ISMIS. A research assistant was available
Of note, items of ISMIS differ from the QOLS and to assist participants if there were difficulties reading
PANSS assessments of social function detailed above be- or understanding the questionnaires. PANSS and
cause they involve explicit attributions of abstract states QOLS ratings were performed blind to responses to
of impoverished function due to discrimination on the the MSEI and ISMI. QOLS and PANSS interviews
basis of mental illness or acceptance that mental illness were conducted by trained research assistants with a min-
renders persons to a lesser status. The QOLS, for in- imum of a BA degree in a field related to psychology.
stance, asks how often a person socialized with a close
friend and the PANSS may involve asking a person about
Results
the absence or presence of interest in others. The ISMIS,
on the other hand, asks if the person socializes less ‘‘be- Means and standard deviations for all measures are pre-
cause my mental illness might make me look or behave sented in table 1. Analyses were conceived as the follow-
weird’’ and whether they believe in general that ‘‘Men- ing 4 steps: First, to determine whether the first 4
tally ill persons should not get married.’’ Items of the subscales of the ISMIS (alienation, stereotype endorse-
ISMIS also differ from the PANSS assessments of diffi- ment, discrimination experience, and social withdrawal)
culties trusting people or accepting as true conspiracies could be combined into a total score for the purposes of
again them; the ISMIS asks whether the person feels dis- using it in a cluster analyses, internal consistency was de-
criminated against ‘‘because I have a mental illness’’ or termined. This revealed a high degree of internal consis-
whether they ‘‘feel out of place in the world because I tency, coefficient a = .86, which was reduced to .80 if the
have a mental illness.’’ fifth score was included. Correlations among all 5 ISMIS
194
 Toward Understanding the Insight Paradox

Table 1. Mean and Standard Deviations K-Means cluster analysis is a nonhierarchical form of
cluster analysis appropriate when hypotheses exist re-
Standard garding the number of clusters contained in a sample.
Instrument Score Mean Deviation It produces the number of clusters as initially called
ISMIS Alienation 2.31 0.65
for, minimizing variability within clusters and maximiz-
ing variability between clusters. We chose this procedure
ISMIS Stereotype endorsement 1.99 0.54
rather than rationally defining groups in order to deter-
ISMIS Discrimination experience 2.42 0.69 mine, in an exploratory and statistical manner, whether
ISMIS Social withdrawal 2.30 0.66 we could detect participants who demonstrated patterns
ISMIS Stigma resistance 2.17 0.52 of these scores as hypothesized rather than as we had arti-

Downloaded from https://academic.oup.com/schizophreniabulletin/article/33/1/192/1922956 by guest on 08 October 2023

PANSS Positive component 15.93 5.65 ficially defined them ahead of time. To give the groups
PANSS Negative component 17.36 5.24
contextual meaning, however, we assigned labels with
reference to the meaning of the insight and stigma scores.
PANSS Insight and judgment item 3.29 1.39
First, we chose to categorize groups as ‘‘low insight’’ if
QOL Interpersonal relations 19.92 7.39 the PANSS insight item was equal or greater than 4 of
QOL Intrapsychic foundations 22.57 5.20 7 and scores of less then 4 as good insight. This catego-
BHS Total 14.16 5.45 rization has been used elsewhere30 and reflects the differ-
MSEI Total 30.28 7.05 ence between general vs minimal awareness that
something is wrong. To describe a group’s stigma level,
Note: ISMI, Internalized Stigma of Mental Illness Scale; we decided that groups with scores of 2 or less would be
PANSS, Positive and Negative Syndrome Scale; QOL, Quality labeled as ‘‘minimal stigma’’ because such scores indicate
of Life; BHS, Beck Hopelessness Scale; MSEI, general disagreement with items. Scores of greater than 2
Multidimensional Self-esteem Inventory. but less than 2.5 were chosen to be labeled as ‘‘mild
stigma,’’ as these scores reflect agreement of roughly
scores revealed that the first 4 scores were all significantly less than half of the ISMIS items. Scores of greater than
related to one another (P < .05), while the fifth, stigma 2.5 but less than 3 were chosen to be labeled as ‘‘moderate
resistance, was significantly correlated only with stereo- stigma’’ because these scores reflect either agreement of
type endorsement. These findings were taken as support more than half of ISMIS items or strong agreement on sev-
to not include the fifth score in the summary score. eral. Scores of greater than 3 were chosen to be labeled as
In the second phase of analyses, insight and stigma ‘‘severe stigma’’ as these scores reflect either agreement
scores were separately correlated with symptoms, social with all ISMIS items or strong agreement on many items.
function, hope, and self-esteem. These revealed that inter- The cluster analysis produced 3 groups which, based
nalized stigma was significantly related (P < .05) to pos- on insight and stigma levels, we have labeled low in-
itive (r = .36) and negative (r = .26) symptoms, as well as sight/mild stigma (n = 23), high insight/minimal stigma
hope (r = ÿ.45), self-esteem (r = .54), and QOLS interper- (n = 25), and high insight/moderate stigma (n = 27). As
sonal relations (r = ÿ.34). Insight was related to positive revealed in table 2, these groups did not differ signifi-
(r = .24) and negative (r = .29) symptoms, as well as self- cantly in age, education, or hospitalization history.
esteem (r = .26). Chi-square analyses additionally found that groups did
In the third phase, ISMIS and PANSS insight and not differ in proportion of participants with schizoaffec-
judgment items were standardized into z scores and tive disorder or schizophrenia or in proportion of partic-
a K-Means cluster analysis was performed to identify ipants from the VA medical center or the CMHC.
3 homogenous participant groups based on these scores. Analysis of variance also found no significant differences
Cluster analysis is a method of classifying people into ty- in raw stigma or insight score for participants from the
pologies by determining clusters of participants that dis- VA medical center compared with the participants
play small within-cluster variation relative to the from the CMHC. Examination of symptoms, however,
between-cluster variation.42–44 In cluster analysis, each found that the high insight/minimal stigma group had
participant is assigned to a cluster and participants are significantly lower levels of positive and negative symp-
moved from one cluster to another until terminating con- toms on the PANSS.
ditions are met. In essence, a cluster analysis is similar in In the third phase of analyses we compared hope, self-
some respects to both factor analysis and discriminant esteem, and social function between groups. As revealed
function analysis. It differs primarily from factor analysis in table 3, the high insight/minimal stigma group had sig-
in that its end is the determination of orthogonal groups nificantly better interpersonal function on the QOLS
of participants rather than orthogonal groups of varia- than either of the other 2 groups. The high insight/mod-
bles, and it differs from a discriminant function analysis erate stigma group, by contrast, reported significantly
in that determining group assignment is the goal and not poorer self-esteem and lesser hope than either of the other
known ahead of time. 2 groups. Given that the groups differed significantly on
195
P. H. Lysaker et al.

Table 2. Background and Symptoms Among Groups

Group 1, Low Group 2, High Group 3, High

Insight/Mild Stigma Insight/Minimal Insight/F = Moderate Group Comparisons,
(n = 23) Stigma (n = 25) Stigma (n = 27) ANOVA P < .05

Age 48.00 (7.65) 49.60 (6.40) 47.22 (7.04) 0.76 NS

Education 12.00 (1.47) 13.04 (2.25) 12.63 (2.23) 1.56 NS
Lifetime 10.35 (9.63) 7.84 (7.22) 9.33 (10.27) 0.19 NS
hospitalizations

Downloaded from https://academic.oup.com/schizophreniabulletin/article/33/1/192/1922956 by guest on 08 October 2023

Age of first 23.30 (5.44) 25.20 (7.03) 26.78 (10.89) 0.99 NS
hospitalization
Stigma score 2.19 (0.30) 1.80 (0.40) 2.75 (0.54) 5.93** 3>1>2
PANSS insight and 4.83 (0.87) 2.40 (0.76) 2.81 (1.11) 40.08** 1 > 2, 3
judgment
PANSS positive 17.70 (5.01) 12.92 (4.31) 17.22 (6.25) 6.11* 2 < 1, 3
PANSS negative 19.22 (5.35) 14.80 (3.89) 18.15 (5.47) 5.28* 2 < 1, 3

Note: ANOVA, analysis of variance; PANSS, Positive and Negative Syndrome Scale; NS, not significant.
*P < .01; **P < .001.

positive and negative symptoms, analyses comparing so- degree to which persons internalize stigmatizing views
cial function, hope, and self-esteem were repeated with about mental illness. As predicted, a cluster analysis of
positive and negative symptom scores included as cova- persons in a stable phase of illness revealed 2 groups
riates. In these analyses the groups continued to differ on of persons relatively aware of having a mental illness:
hope and self-esteem (F2,68 = 4.72, P < .05; F2,68 = 9.35, one group that did and another that did not endorse hav-
P < .001), with the high insight/moderate stigma group ing self-stigmatizing beliefs about their condition. Also,
again in post hoc comparisons having significantly poor- as predicted, persons with high insight who endorsed
er hope and self-esteem. When symptoms were statisti- self-stigmatizing beliefs had lower levels of self-esteem
cally controlled for, however, no significant differences and hope and fewer interpersonal relationships than
were found between groups on the QOLS interpersonal those with high insight who rejected stigmatizing beliefs.
relations scores (F2,62 = 0.53, P = NS). Finally, as predicted, the cluster analyses produced a third
group that demonstrated low awareness and also en-
dorsed stigmatizing beliefs, though to a lesser degree
Discussion
than did the high insight/moderate stigma group. This
In the current study we examined the hypothesis that the group also had more self-esteem and hope than the group
effects of awareness of illness in schizophrenia on self- with high insight and moderate stigma but did not differ
esteem, hope, and functioning would be affected by the from them in social functioning. This last finding may

Table 3. Hope, Self-esteem, Symptoms, and Social Function Among Groups

Group 2, High Group 3, High

Group 1, Low Insight/Minimal Insight/F = Moderate Group Comparisons,
Insight/Mild Stigma Stigma Stigma ANOVA P < .05

Beck Hopelessness 14.80 (4.46) (n = 20) 16.33 (4.10) (n = 24) 11.65 (6.38) (n = 26) 5.39** 3 < 1, 2
Scale
MSEI self-esteem 32.95 (7.92) (n = 21) 32.92 (4.41) (n = 24) 25.69 (6.07) (n = 26) 11.20*** 3 < 1, 2
total
QOLS interpersonal 18.06 (6.18) (n = 17) 23.29 (8.08) (n = 21) 18.36 (6.79) (n = 25) 3.54* 2 > 1, 3
relations
QOLS intrapsychic 21.65 (4.21) (n = 17) 24.33 (5.00) (n = 21) 21.72 (5.20) 1.86 NS
foundations (n = 25)

Note: ANOVA, analysis of variance; MSEI, Multidimensional Self-esteem Inventory; QOLS, Quality of Life Scale.
*P < .05; **P < .01; ***P < .001; NS, not significant.

196
 Toward Understanding the Insight Paradox

suggest that both the acceptance of stigma or unaware- With replication, our findings may have several clinical
ness of illness may lead to social isolation. However, it is implications. First, it may be useful to consider interven-
also possible that an underlying risk factor for both poor tions that decrease internalized stigma. Warner,52 for in-
insight and social isolation, such as neurocognitive im- stance, has suggested that it is just as important for
pairment,9 may explain these relationships. interventions to assist in developing a sense of mastery
Exploratory comparisons of symptom levels between as it is to help enhance insight. This is consistent with a re-
the groups revealed that groups with higher insight cent intensive case study that suggested that as a person
and minimal stigma had significantly lower levels of pos- with schizophrenia recovered, he first evolved a greater
itive and negative symptoms than either of the 2 other sense of personal agency before developing a more com-
groups. This suggests several possible interpretations. plex grasp of his illness .53 It is also consistent with a study

Downloaded from https://academic.oup.com/schizophreniabulletin/article/33/1/192/1922956 by guest on 08 October 2023

First, it is possible that having more symptoms makes using a different sample that suggested that hope was
one the target of more stigma and therefore makes it more closely tied to sense of agency than illness aware-
more difficult to reject these beliefs. Second, it is possible ness.54 Perhaps, if dysfunctional beliefs stemming from
that with more symptoms it becomes harder to formulate social stigma impact life in such an enduring manner, tai-
positive beliefs about oneself and, therefore, to reject so- lored interventions could be devised to help persons com-
cietal beliefs about schizophrenia. It is also possible that bat these self-stigmatizing beliefs. Future interventions
having less internalized stigma allows one to manage and research could be directed to help persons with
symptoms more effectively, thereby reducing their sever- schizophrenia overcome their negative beliefs and find
ity. Regardless of which interpretation is valid, when newer and more adaptive ways to think of themselves
symptoms were controlled for, groups continued to and their futures, thus allowing for the acceptance of
differ on self-esteem and hope, although notably group mental illness to have fewer devastating effects. One pos-
differences in interpersonal relatedness were no longer sibility is for research to study ways to facilitate the
significant once initial differences in symptoms were efforts of persons with schizophrenia to replace self-
accounted for. stigmatizing beliefs and transform their narratives and
While the cross-sectional nature of this study precludes experience themselves as active protagonists in their
reaching any conclusions regarding causality, results do own lives with realistic appraisals of their strengths and
suggest hypotheses and directions for future research. deficits.55,56
First, consistent with the initial hypotheses of this study Of note there may be additional parallels here with
it appears that persons who both accept that they have earlier studies of differing styles of framing past experi-
a mental illness and who accept stigmatizing beliefs about ence of mental illness as persons recover. Specifically,
mental illness are at risk for having low self-esteem and McGlashan and colleagues57–59 have suggested the exis-
expecting less of their futures. These persons, regardless tence of a continuum of recovery styles. On one end of
of whether they adhere to treatment, may show little their continuum lies ‘‘integration,’’ which is characterized
‘‘motivation’’ to pursue personal or rehabilitation by patients showing interest in their psychotic experien-
goals,45 having little hope that they can find a satisfying ces, appearing eager to discuss and learn more about
life. This group, substantial in size (roughly a third of our them and to gain a perspective of them. On the other
sample), may pose a great challenge to mental health pro- end of the continuum is ‘‘sealing over,’’ characterized
viders, who struggle to encourage these clients to pursue by patients’ denial of the existence and/or severity of their
vocational and other goals, when they, the clients them- illnesses, expectations to return rapidly to normal func-
selves, lack hope for the future. tioning, and poor ability to recall or describe the phase of
Our findings, consistent with others,16–18 also poten- acute psychosis. Perhaps these findings coupled with the
tially raise a challenge to the view that insight, as tradi- results of this study point to the reality that, for some,
tionally defined, is always desirable for people with forgetting or recasting the past in such a way as to
schizophrenia. In fact, some with stigmatizing beliefs deny previous difficulties may be adaptive. As a corollary,
about mental illness may find accepting their illness bur- it may also be that forcing or urging persons who have
densome in some ways. This is consistent with the find- sealed over to do otherwise may not be helpful, and re-
ings suggesting that benefit from psychiatric treatment is search that more fully explores what meets their needs is
related to the meanings persons assign to both their ill- needed.
ness and the treatment itself.46–48 It is also consistent Importantly, there are several limitations to the study.
with observations that well-being for persons who suffer Given its cross-sectional nature, no conclusions can be
from a wide range of chronic medical illnesses is a func- drawn regarding causality and alternative explanations
tion of the degree to which they experience a balance of the findings cannot be ruled out. For instance, it is
among bodily concerns or symptoms, the sense of co- possible that having less hope and self-esteem makes per-
herence of their personal narrative or biography, and sons more vulnerable to stigma or that acceptance of
conceptions they have formed about the nature of stigma and low self-esteem is the product of other bio-
their identity.49–51 logical or sociocultural variables not measured here.
197
P. H. Lysaker et al.

Generalization of findings is also limited by sample com- 17. Kravetz S, Faust M, David M. Accepting the mental illness
position. Participants were mostly men in their 40s, all of label, perceived control over the illness, and quality of life.
Psychiatr Rehabil J. 2000;23:323–332.
whom were involved in treatment. It may well be that
18. Moore O, Cassidy E, Carr A, O’Callaghan E. Unawareness
a different relationship exists among insight, internalized of illness and its relationship with depression and self-decep-
stigma, hope, and self-esteem among younger persons tion in schizophrenia. Eur Psychiatry. 1999;14:264–269.
with schizophrenia, in a predominantly female sample, 19. O’Mahony PD. Psychiatric patient denial of mental illness as
or among persons who decline treatment. Thus, more re- a normal process. Br J Med Psychol. 1982;55:109–118.
search that involves collecting data at multiple time 20. Pyne JM, Bean D, Sullivan G. Characteristics of patients with
points with broader samples is necessary. We also did schizophrenia who do not believe they are mentally ill. J Nerv
not assess the nature of stigmatizing experiences. More Ment Dis. 2001;189:146–153.

Downloaded from https://academic.oup.com/schizophreniabulletin/article/33/1/192/1922956 by guest on 08 October 2023

‘‘fine-grained’’ assessments of stigma including apprais- 21. Schwartz RC. Self-awareness in schizophrenia: its relation-
als of the nature of stigmatizing experiences are necessary ship to depressive symptomatology and broad psychiatric
impairments. J Nerv Ment Dis. 2001;189:401–403.
to replicate and confirm the importance of the findings
22. Roe D, Kravetz S. Different ways of being aware of and ac-
noted here. knowledging a psychiatric disability. A multifunctional narra-
tive approach to insight into mental disorder. J Nerv Ment
References Dis. 2003;191:417–424.
23. Lally SJ. Does being in here mean there is something wrong
1. Amador XF, Strauss DH, Yale SA, Gorman JM. Awareness with me? Schizophr Bull. 1989;15:253–265.
of illness in schizophrenia. Schizophr Bull. 1991;17:113–132. 24. Williams CC, Collins A. Factors associated with insight
2. David AS. Insight and psychosis. Br J Psychiatry. 1990; among outpatients with serious mental illness. Psychiatr
156:798–805. Serv. 2002;53:96–98.
3. Bartko G, Herczeg I, Zador G. Clinical symptomatology and 25. McCay EA, Seeman MV. A scale to measure the impact of
drug compliance in schizophrenia. Acta Psychiatr Scand. a schizophrenic illness on an individual’s self-concept. Arch
1988;77:74–76. Psychiatr Nurs. 1998;12:41–49.
4. Cuffel BJ, Alford J, Fischer EP, Owen RR. Awareness of ill- 26. Ritsher JB, Phelan JC. Internalized stigma predicts erosion of
ness in schizophrenia and outpatient treatment compliance. morale among psychiatric outpatients. Psychiatry Res. 2004;
J Nerv Ment Dis. 1996;184:653–659. 129:257–265.
5. Schwartz RC. The relationship between insight, illness and 27. Link BG, Frances TC, Struening E, Shrout PE, Dohrenwend
treatment outcome in schizophrenia. Psychiatric Q. 1998; BP. A modified labeling theory approach to mental disorders:
69:1–22. an empirical assessment. Am Sociol Rev. 1989;54:400–423.
6. Francis JL, Penn DL. The relationship between insight and 28. Roe D. Progressing from patienthood to personhood across
social skill in persons with severe mental illness. J Nerv the multidimensional outcomes in schizophrenia and related
Ment Dis. 2001;189:822–829. disorders. J Nerv Ment Dis. 2001;189:691–699.
7. Lysaker PH, Bell MD, Bryson GJ, Kaplan EZ. Psychosocial 29. Yanos PT, Moos RH. Determinants of functioning and well-
function and insight in schizophrenia. J Nerv Ment Dis. being in schizophrenia: an integrated model. Clin Psychol
1998;186:432–436. Rev. In press.
8. Olfson M, Marcus SC, Wilk J, West JC. Awareness of illness 30. Lysaker PH, Campbell K, Johannesen JK. Hope, awareness
and nonadherence to antipsychotic medications among per- of illness, and coping in schizophrenia spectrum disorders.
sons with schizophrenia. Psychiatr Serv. 2006;57:205–211. J Nerv Ment Dis. 2005;193:287–292.
9. Lysaker PH, Bryson GJ, Bell MD. Insight and work perfor- 31. Bradshaw W, Brekke JS. Subjective experience in schizo-
mance in schizophrenia. J Nerv Ment Dis. 2002;190:142–146. phrenia: factors influencing self-esteem, satisfaction with
10. Frank A, Gunderson JG. The role of the therapeutic alliance life, and subjective distress. Am J Orthopsychiatry. 1999;
in the treatment of schizophrenia. Arch Gen Psychiatry. 69:254–260.
1990;47:228–236. 32. Markowitz FE. The effects of stigma on the psychological
11. Amador XF, Strauss DH, Yale SA, et al. Awareness of illness well-being and life satisfaction of persons with mental illness.
in schizophrenia. Arch Gen Psychiatry. 1994;51:826–836. J Health Soc Behav. 1998;39:335–347.
12. Dixon M, King S, Steiger H. The contribution of depression 33. Roe D. A prospective study on the relationship between
and denial towards understanding the unawareness of symp- self-esteem and functioning during the first year after being
toms in schizophrenic out-patients. Br J Medical Psychol. hospitalized for psychosis. J Nerv Ment Dis. 2003;191:
1998;71:85–97. 45–49.
13. Mintz AR, Dobson KS, Romney DM. Insight in schizophre- 34. Spitzer R, Williams J, Gibbon M, First M. Structured Clinical
nia: a meta-analysis. Schizophr Res. 2003;61:75–88. Interview for DSM IV. New York, NY: Biometrics Research;
14. Thompson EH. Variations in the self-concept of young adult 1994.
chronic patients: Chronicity reconsidered. Hosp Community 35. Kay SR, Fiszbein A, Opler LA. The Positive and Negative
Psychiatry. 1988;39:771–775. Syndrome Scale (PANSS) for schizophrenia. Schizophr Bull.
15. Warner R, Taylor D, Powers M, Hyman R. Acceptance of 1987;13:261–276.
the mental illness label by psychotic patients: effects on func- 36. Bell MD, Lysaker PH, Beam-Goulet JL, Milstein RM,
tioning. Am J Orthopsychiatry. 1989;59:389–409. Lindenmayer JP. Five-component model of schizo-
16. Hasson-Ohayon I, Kravetz S, Roe D, Weiser M. Insight into phrenia: assessing the factorial invariance of the Positive
severe mental illness, perceived control over the illness, and and Negative Syndrome Scale. Psychiatry Res. 1994;
quality of life. Compr Psychiatry. 2006;194(7):538–42. 52:295–303.

198
 Toward Understanding the Insight Paradox

37. Heinrichs DW, Hanlon TE, Carpenter WT. The Quality of 49. Albrecht GL, Devleiger PJ. The disability paradox: high qual-
Life Scale: an instrument for assessing the schizophrenic def- ity of life against all odds. Soc Sci Med. 1999;48:977–988.
icit syndrome. Schizophr Bull. 1984;10:388–396. 50. Lambert BL, Street RL, Cegala DJ, Smith DN, Kurtz S,
38. Ritsher JB, Otilingam PG, Grajales M. Internalized stigma of Schofield T. Provider patient communication patient centered
mental illness: psychometric properties of a new measure. care and the mangle of practice. Health Commun. 1997;9:
Psychiatry Res. 2003;121:31–49. 27–43.
39. O’Brien EJ, Epstein S. MSEI: The Multidimensional Self- 51. Lambert BL, Levy NA, Winer J. Keeping the balance and
esteem Inventory Professional Manual. Lutz, Fla: Psychologi- monitoring the self-system: towards a more comprehensive
cal Assessment Resources, Inc; 1998. model of medication management in psychiatry. In: Brashers
40. Beck AT, Weissman A, Lester D, Trexler L. The measure- D, Goldsmith D, eds. Communication and the Management of
ment of pessimism: the hopelessness scale. J Consult Clin Psy- Health and Illness. Mahwah, NJ: Erlbaum: In press.

Downloaded from https://academic.oup.com/schizophreniabulletin/article/33/1/192/1922956 by guest on 08 October 2023

chol. 1974;42:861–865. 52. Warner R. Recovery From Schizophrenia: Psychiatry and Po-
41. Haatainen KM, Tanskanen A, Kylma J, et al. Life events are litical Economy. New York, NY: Routledge; 1994.
important in the course of hopelessness-a 2-year follow-up 53. Lysaker PH, Davis LD, Eckert GJ, Strasburger A, Hunter N,
study in a general population. Soc Psychiatry Psychiatr Epi- Buck KD. Changes in narrative structure and content in
demiol. 2003;38:436–441. schizophrenia in long term individual psychotherapy: a single
42. Carpenter WT, Bartko JJ, Carpenter CL, Strauss JS. Another case study. Clin Psychol Psychother. 2005;12:406–416.
view of schizophrenia subtypes: a report from the interna- 54. Lysaker PH, Buck KD, Hammoud K, Taylor AC, Roe D.
tional pilot study of schizophrenia. Arch Gen Psychiatry. Associations of symptom remission, psychosocial function
1976;33:508–516. and hope with qualities of self experience in schizophrenia:
43. Dillon WR, Goldstein M. Multivariate Analysis: Methods comparisons of objective and subjective indicators of recov-
and Applications. New York, NY: John Wiley & Sons; 1984. ery. Schizophr Res. 2006;82:241–249.
44. SPSS. SPSS Base 10.0 User’s Guide. Chicago, Ill: SPSS, Inc; 55. Roe D, Davidson L. Self and narrative in schizophrenia: time
1999. to author a new story. J Med Humanit. 2005;31:89–94.
45. Rusch N, Corrigan PW. Motivational interviewing to im- 56. Lysaker PH, Lysaker JT. A typology of narrative impoverish-
prove insight and treatment adherence in schizophrenia. Psy- ment in schizophrenia: implications for understanding the
chiatr Rehab J. 2002;26:23–33. processes of establishing and sustaining dialogue in individual
46. Deegan PE. The importance of personal medicine: a qualita- psychotherapy. Couns Psychol Q. 2006;19:57–68.
tive study of resilience in people with psychiatric disabilities. 57. Levy ST, McGlashan TH, Carpenter WT. Integration and
Scand J Public Health. 2005;66:29–35. sealing-over as revocery styles from acute psychosis: metapsy-
47. Lacro J, Dunn L, Dolder C. Prevalence of and risk factors for chological and dynamic concepts. J Nerv Ment Dis.
medication nonadherence in patients with schizophrenia: 1975;161:307–312.
a comprehensive review of recent literature. J Clin Psychiatry. 58. McGlashan TH, Levy ST, Carpenter WT. Integration and
2002;63:892–908. sealing over. Clinically distinct recovery styles from schizo-
48. Tait L, Birchwood M, Trower P. Predicting engagement with phrenia. Arch Gen Psychaitry. 1975;32:1269–1272.
services for psychosis: insight, symptoms and recovery style. 59. McGlashan TH, Carpenter WT. Does attitude towards psy-
Br J Psychiatry. 2003;182:123–128. chosis relate to outcome? Am J Psychiatry. 1981;138:797–801.

199