Blame, Shame, and Contamination: The Impact of Mental Illness and
Drug Dependence Stigma on Family Members
Patrick W. Corrigan
Illinois Institute of Technology
Amy C. Watson
University of Illinois, Chicago
Frederick E. Miller
Evanston Northwestern Healthcare and Northwestern
Family members of relatives with mental illness or drug dependence or both report that they
are frequently harmed by public stigma. No population-based survey, however, has assessed
how members of the general public actually view family members. Hence, the authors
examined ways that family role and psychiatric disorder influence family stigma. A national
sample (N ? 968) was recruited for this study. A vignette design describing a person with a
health condition and a family member was used. Family stigma related to mental illnesses,
such as schizophrenia, is not highly endorsed. Family stigma related to drug dependence,
however, is worse than for other health conditions, with family members being blamed for
both the onset and offset of a relative’s disorder and likely to be socially shunned.
Keywords: stigma, family, mental illness, drug dependence, other health conditions
The New Freedom Commission (Hogan, 2003) high-
lighted stigma as a major barrier to the mental health goals
of Americans and recommended concerted efforts to change
public opinion and to diminish prejudice. Although the New
Freedom Commission report was not explicit about family
stigma, it clearly recognized that barriers to family partici-
pation significantly impede mental health care. This paper
addresses the complexity of what is called family stigma.
Stigma not only harms many people with mental illness or
drug abuse or both but also injures family members who are
associated with these individuals. Goffman (1963) called
this courtesy stigma, the prejudice and discrimination that is
extended to people not because of some mark (e.g., mental
illness, disorder) that they manifest but rather because they
are somehow linked to a person with the stigmatized mark.
Surveys have shown that family members with relatives
who have mental illness or drug dependence disorders re-
port significant experience with family stigma. However, to
our knowledge, no survey based on a national sample has
been conducted to determine whether the public, in fact,
endorse stigmas about family members. The goal of this
study is to examine family stigma in a sample drawn from
the general adult public. We first review the stigma expe-
rienced by families: blame, shame, and contamination.
Then we examine how family members in various roles—
parents, children, spouse, and siblings—interact with
Public Stigma Applied to People With Mental
Illness and Drug Dependence Disorders
Common stereotypes about people with mental illness
seem to parallel those with drug dependence and include
dangerousness and blame (Angermeyer, Matschinger, &
Corrigan, 2004; Link, Phelan, Bresnahan, Stueve, & Pesco-
solido, 1999). Generally, research shows that psychiatric
disorders are viewed as more blameworthy than physical
health conditions such as cancer and heart disease (Corrigan
et al., 1999; Weiner, Perry, & Magnusson, 1988). Research
has focused on stereotypes related to attributions about
personal responsibility and blame (Corrigan, 2000). Inves-
tigators have found that research participants who blame
relatives for the onset of the relatives’ mental illness or drug
dependence are more likely to react angrily to those rela-
tives, to withhold help, to avoid them socially, and to
support coercive mental health services (Corrigan et al.,
1999, 2000; Corrigan, Markowitz, Watson, Rowan, & Ku-
biak, 2003; Corrigan, & Miller, 2004). Research that com-
pares the public stigma of mental illness to drug dependence
consistently shows that persons with drug dependence are
judged to be more responsible for their disorder (Corrigan et
al., 1999; Link et al., 1999; Weiner, Perry, & Magnusson,
1988). The difference between perceptions of drug depen-
1In this article, we distinguish the stigma experienced by people
with psychiatric disorders from family stigma by labeling the
former “primary stigma.”
Patrick W. Corrigan, Institute of Psychology, Illinois Institute of
Technology; Amy C. Watson, School of Social Work, University
of Illinois, Chicago; Frederick E. Miller, Department of Psychiatry
and Behavioral Sciences, Northwestern Healthcare and Northwest-
Correspondence concerning this article should be addressed to
Patrick W. Corrigan, Institute of Psychology, Illinois Institute of
Technology, 3424 South Slate Street, Chicago, IL 60616. E-mail:
Journal of Family Psychology
2006, Vol. 20, No. 2, 239–246
Copyright 2006 by the American Psychological Association
dence and mental illness is expected to emerge when family
stigma is the dependent measure.
Stigma and Family Members
The themes of blame and shame are seen in surveys of
families of individuals with psychiatric disorders in which
family members discuss their experience with family
stigma. Large-scale studies have shown that between a
quarter and a half of family members believe that their
relationship with a person with mental illness should be kept
hidden or is otherwise a source of shame to the family
(Angermeyer, Schulze, & Dietrich, 2003; Phelan, Bromet,
& Link, 1998; Ohaeri & Fido, 2001; Phillips, Pearson, Li,
Xu, & Yang, 2002; Thompson & Doll, 1982; Shibre et al.,
2001; Wahl & Harman, 1989). Shame seemed to be linked
to blaming the family for the member’s psychiatric disorder.
Findings from a group of 178 family members showed that
about 25% worried that other people might blame them for
their relatives’ mental illness (Shibre et al., 2001).
Blame and shame seem to lead to discrimination in the
form of social avoidance. Three large studies reported that
about a fifth to a third of family members reported strained
and distant relationships with extended family or friends or
both because of a relative with mental illness (Oestman &
Kjellin, 2002; Shibre et al., 2001; Struening et al., 2001;
Wahl & Harman, 1989). However, another study found a
much smaller rate, with only 10% of a sample reporting
occasional avoidance by a few people (Phelan et al., 1998).
Note that all the studies we reviewed examined family
stigma from the perspective of the family, that is, whether
family members perceive members of the general public
stigmatizing them because of their relatives with mental
illness. Hence, the first goal of this study was to conduct a
survey using a national sample to determine how a subset of
the American public actually views family stigma. In addi-
tion to descriptive statistics representing the endorsement of
family stigma, our survey also included a comparison be-
tween the family stigma of mental illness and a physical
health condition for which patients are frequently blamed:
emphysema (Chapple, Ziebland, & McPherson, 2004). Con-
sistent with research on primary stigma (Weiner et al.,
1988), we expected the family stigma of mental illness to be
more severe than that related to emphysema. Also note that
these previous studies limited their research to the family
stigma that stems from a family member with mental ill-
ness; we were unable to find any studies examining how
family members of people with drug dependence disorders
experience family stigma. An additional goal of this study
was to examine family stigma for drug dependence disor-
ders. Consistent with the research on primary stigma, family
stigma due to drug dependence is hypothesized to be worse
than for mental illness.
How Stigma Varies by Family Role
Family stigma may vary by family role: parent, spouse,
sibling, or child (Corrigan & Miller, 2004). Struening et al.
(2001) examined this question in terms of parents in two
different samples. Almost half of one sample (N ? 281),
comprised mostly of mothers, reported some concern about
being blamed for their children’s mental illness. Typically,
blame is attributed to bad parenting skills; for example, the
mother’s incompetence led to the child developing a mental
illness. Results from a second sample (N ? 180) reported by
Struening et al. (2001) reported the same concerns though at
a lower rate: about 10% of mothers experienced being
blamed. Siblings and spouses are often blamed for family
members who mismanage their illness. In describing causal
attributions about human behavior, Weiner (1995) distin-
guished between onset and offset attributions. As applied to
health conditions, onset attributions answer questions re-
garding how a set of symptoms started. Offset attributions
reflect the regular recurrence of symptoms (e.g., the treat-
ments a person must participate in to experience a cure).
Siblings and spouses are often blamed for a relative’s dis-
ease offset; namely, they fail to help the person with mental
illness adhere to treatment such that the person unnecessar-
ily relapses. A study of 164 siblings hinted at this stigma;
survey participants were concerned about relatives with
mental illness remaining adherent to treatment regimens and
perceptions that relapse was somehow the participants’ fault
(Greenberg, Kim, & Greenley, 1997). Unlike the kind of
responsibility experienced by parents, sibling blame seems
to mirror public expectations that family members who are
somehow currently associated with adult children with men-
tal illness (e.g., siblings) or who have opted to live with the
adult (e.g., spouses) have greater responsibility for current
status. This is evident in the reduced shame experienced by
family members who do not live with the relative with
mental illness as compared to those who do (Phelan et al.,
The child of a person with mental illness is often viewed
as contaminated by the parent’s mental illness. One inves-
tigation attempted to test this finding using a more carefully
controlled vignette experiment (Mehta & Farina, 1988).
Results showed that students portrayed in the vignettes as
having a father who is depressed, alcoholic, or an ex-convict
were viewed as having more difficulty than the other
groups. Another study illustrated the complexity of contam-
ination on children, in this case, of parents with alcoholism
or mental illness (Burk & Sher, 1990). A sample of 570
adolescents was more likely to rate teenagers with stigma-
tized parents as more socially negative than teens with
parents who do not abuse alcohol and do not have a mental
A Comparison of Family Stigma and Primary
Finally, the data in this paper provide an answer to a
fundamental question about family stigma: how bad is it?
One way to address this question is by comparing the family
stigma applied to mental illness and drug dependence versus
that experienced by emphysema. A second question,
though, is how bad is family stigma for a specific health
condition compared to corresponding primary stigma? We
answer this question by comparing responses made by the
240CORRIGAN, WATSON, AND MILLER
sample toward people with mental illness or drug depen-
dence disorder against the same research participants’ atti-
tudes about the family member.
The data for this study come from the Family Stigma Survey
collected by Time-Experiments for the Social Sciences (TESS;
NSF Grant 0094964, Diana Mutz and Arthur Lupia, Investigators).
TESS uses a national online research panel recruited by Knowl-
edge Networks (KN). KN recruits for its sample via list-assisted
random digit-dialing techniques on a sample frame consisting of
the entire United States’ telephone population. Recruits are pro-
vided free WEB-TV access in return for completing surveys that
are sent to them via e-mail weekly.
For this study, KN randomly identified and solicited 1,307
individuals from its overall panel for the Family Stigma Survey
from March 26, 2004 to April 8, 2004; 74% completed the survey
(N ? 968). The sample was 51.9% female, with a mean age of 47.0
years (SD ? 16.5, range ? 18–95). The sample was 72.5% White,
11.7% Black, 11.0% Mexican Americans, and 4.8% other. Of the
sample, 15.8% had less than a high school education, 32.1% were
high school graduates, 27.8% had completed some college, and
24.4% had a bachelor’s degree or higher. Postsurvey stratification
weights were used to adjust sample demographics to values con-
sistent with the 2000 U.S. Census. Variables used to determine
stratification weights include gender, age, race/ethnicity, geo-
graphic region in the United States, and level of education. Data
reported in this paper represent weight-corrected cases. Despite
efforts to attain a true probability sample, there are limits to the KN
approach. Most prominent of these are limiting the sample to
phone-bearing households that do not have the Internet but wish to
do so via WEB-TV.
Each respondent was randomly assigned to read a vignette that
varied across four conditions: disease of the person with the
disorder, role of the corresponding family member, gender of the
person with the disorder, and gender of the family member. One
such vignette follows:
[John Smith/Joan Smith] is the [father/mother/son/daughter/
brother/sister/husband/wife] of [Frank/Fran] Smith, a 30-year-old
[man/woman] with [schizophrenia/drug dependence/emphysema].
[Frank/Fran] lives with [his or her] family and works as a clerk at
a nearby store. [Frank/Fran] has been hospitalized several times
because of [his or her] illness. The illness has disrupted [his or her]
The quality of specific terms used to describe health conditions
can influence the reaction of respondents. For example, problems
related to “psychiatric disorder” are broader than the idea of
mental illness alone and include areas such as drug dependence
(Martin, Pescosolido, & Tuch, 2000). We addressed this problem
by providing respondents with types of mental health problems as
listed in the DSM. Moreover, we adopted labels from the
MacArthur Mental Health Module of the 1996 General Social
Survey (GSS) for the two psychiatric conditions in order to facil-
itate comparison with previous research (Pescosolido, Monahan,
Link, Stueve, & Kikuzawa, 1999). Mental illness was “schizophre-
nia” and drug dependence was “drug dependency.” Based on
earlier research by Weiner et al. (1988) on attributions across
health conditions, we decided on emphysema as the comparison
physical health disorder. Consistent with the labels of the GSS
MacArthur Module, we decided on a label that represented a
specific disorder rather than a generic category. We chose emphy-
sema on the possibility that its connection with smoking might
increase the level of blame associated with it (Chapple, Ziebland,
& McPherson, 2004).
Family roles were limited to four dominant ones found in
previous research on family stigma (Corrigan & Miller, 2004):
parents, children, siblings, and spouses. Some evidence suggests
that family stigma may vary by the gender of the family role; for
example, mothers may be stigmatized more harshly than fathers
(Corrigan & Miller, 2004; Lefley, 1992). Hence, vignettes ran-
domly varied the gender of the family member. In like manner,
gender of the person with the health disorder was also randomly
varied by vignette. We decided not to vary other sociodemograph-
ics of vignette participants because earlier research, for the most
part, failed to show them to be relevant in stigmatizing people with
psychiatric disorders (Pescosolido et al., 1999).
To capture the effects of these many variations on several
outcomes by means of direct questioning would be cumbersome
and time-consuming. This dilemma can easily be managed, how-
ever, by using a factorial survey design (Rossi & Nock, 1982)
similar to those employed by previous national probability surveys
of attitudes toward persons belonging to stigmatized groups (Link
et al., 1999; Phelan et al., 2000). In vignette experiments that
incorporate factorial survey designs, respondents are presented
with descriptions of a fictional person who varies across theoret-
ically relevant dimensions (i.e., type of health condition, hire
versus promotion decision, gender, age, and ethnicity). The differ-
ent versions of the vignette that are created in this way are
randomly assigned to respondents. Respondents then answer ques-
tions about the described person in terms of relevant outcome
variables. In addition to manipulating and then measuring the
effects of several independent variables at one time, this factorial
survey design also allowed the strengths of the experimental
method to be brought to bear in our study. By experimentally
manipulating one of the individual’s characteristics (e.g., whether
NAME’s health condition is psychotic disorder, drug dependence,
or emphysema) while holding all others constant, we were able to
attribute any differences in attitudes and behavioral intentions
specifically to variation in that characteristic. Because respondents
were assigned randomly to vignettes, differences in responses
could be attributed to variations in the stimulus rather than to
variations in respondents’ characteristics. The profiles required to
accomplish a factorial survey design can be produced using
distribution-generating algorithms available in standard statistical
After reading the vignette, respondents were instructed to re-
spond to 14 items using seven-point Likert scales (e.g., 7 ?
strongly agree). Seven of the items were about the person with the
health disorder, while seven were about the family member. The
first seven items were from the short form of the Attribution
Questionnaire, which has been shown to be a reliable and valid
measure of primary stigma (Corrigan et al., in press; Corrigan et
al., 2003, 2002). For example, “it is [Frank’s/Fran’s] own fault that
[he or she] is in the present condition.” These studies include
confirmatory factor analyses which support the reliability and
content validity of measures. The items selected from the short
form of the Attribution Questionnaire represented the single item
that loads most into the seven factor solution of the confirmatory
The selection of seven items reflecting family stigma was based
on our review of relevant content areas from three sources. First,
we reviewed the common themes that describe the primary stigma
of mental illness and drug dependence used in prior research
(Corrigan, 2005). Although this information largely influenced the
first seven items relevant to how the public views the person with
the health disorder, we also considered it in developing items
reflecting family stigma. Second, we reviewed the common themes
that family members have used to describe their experience with
family stigma (Corrigan & Miller, 2004). Third, we conducted a
focus group of family members to augment our list of items
reflecting family stigma. During a 60-minute session, seven mem-
bers of families with a person with psychiatric disorder (57.1%
female, including and across all four family roles) answered ques-
tions about their general understanding of stigma and prejudice.
They also responded to examples of stigma applied to their family
member with psychiatric disorder and to examples of stigma
applied to them as family members. Analyses of the responses of
focus group participants endorsed the themes of blame, shame, and
contamination found in our literature review. In addition, a content
analysis of transcripts of the focus group yielded additional family
stigma items, such as onset responsibility (family member to
blame for person getting disorder), offset responsibility (family
member to blame for person relapsing), pity, contamination (ill-
ness could rub off), shame, incompetence (the family member was
not very good as a parent, sibling, spouse, or child), and avoidance
(the respondent would not want to socialize with the family mem-
ber). For example, “[John/Joan] bears some responsibility for [his
or her] [insert relationship] originally getting ill.”
Items assessing primary stigma were always presented to re-
search participants before family stigma items to prime stereotypes
related to the health condition. Items within each domain (i.e.,
primary stigma and family stigma) were presented in random
The research questions guiding this paper suggest a three-
step approach to analysis. First, we conducted a series of
descriptive analyses to examine how the public endorses
family stigmata for families that have members with one of
these three health conditions. This included an inferential
analysis to examine how family stigma varied by disorder.
Second, we examined how stigma varied by family role.
Third, we determined how family stigma compared to pri-
mary stigma by comparing the mean score on selected
primary stigma items to the scores on family stigma items.
How Does Family Stigma Vary by Health
Mean and standard deviations of responses to the seven
overall family stigma survey items is summarized in the
Overall row of Table 1. Results of a oneway MANOVA
with the seven items as dependent variables were signifi-
cant, F(14, 1886) ? 13.31, p ? .001. Subsequent oneway
ANOVAs showed that all seven items differed significantly
across the three health conditions (F ranges from 3.95, p ?
.05 to 54.18, p ? .0001). Post hoc Tukey’s test examined
the differences between pairs of health conditions. Results
suggest that families of people who are drug dependent are
viewed in the most stigmatizing manner, that is, these
families are viewed as more responsible for onset (drug
dependence-schizophrenia, p ? .0001; drug dependence-
emphysema, p ? .001) and offset of the disorder (drug
dependence-schizophrenia, p ? .0001; drug dependence-
emphysema, p ? .001), more likely to be contaminated (drug
dependence-schizophrenia, p ? .0001; drug dependence-
emphysema, p ? .001), more ashamed of afflicted family
member (drug dependence-schizophrenia, p ? .0001; drug
dependence-emphysema, p ? .001), and less competent in
their family role (drug dependence-schizophrenia, p ?
.0001; (drug dependence-emphysema, p ? .001). Fami-
lies of people with drug dependence and schizophrenia
were viewed as more pitiable (p ? .05) than those with
Additional analyses examined whether survey partici-
pants endorsed any specific family attitudes higher than
others. Results of a within-Group ANOVA for the subgroup
of survey participants randomized to the schizophrenia vi-
gnettes, with the seven family stigma items as dependent
variables, was significant, F(6, 1842) ? 108.19, p ? .001.
Subsequent contrasts showed the group most agreed with
withholding pity (p ? .05), viewing the person as incom-
petent in his or her family role (p ? .05) and as socially
avoiding the family member (p ? .05) compared to the
remaining four items. The within-Group ANOVA for the
group assigned to drug dependence was also significant,
F(6, 1950) ? 52.42, p ? .001. Subsequent contrasts showed
withholding pity (p ? .05) and contamination (p ? .05) as
the two items most highly endorsed by survey participants.
How Does Family Stigma Vary With Family Role?
Table 1 also includes the mean and standard deviations of
participant responses to family stigma items organized by
family role and health condition. Results of a 4 ? 3 (family
role by health condition) MANOVA with the seven family
stigma items as dependent variables yielded significant
main effects for family role, F(21, 2805) ? 10.18, p ? .001,
as well as a significant interaction, F(42, 5628) ? 2.53, p ?
.001. Subsequent 4 ? 3 ANOVAs were then conducted for
the seven family stigma items individually. Significant main
effects for family role were found for onset blame (F ?
2.45, p ? .05), contamination (F ? 8.09, p ? .001), offset
blame (F ? 3.95, p ? .05), and withholding pity (F ? 1.35,
ns). Subsequent post hoc Tukey’s tests examined pairwise
differences. Results showed that parents and spouses are
viewed to be more responsible for the onset (p ? .05) of the
person’s schizophrenia, drug dependence, and emphysema
than children and siblings. Schizophrenia, emphysema, and
drug dependence were likely to contaminate children (p ?
.05) more than other family roles. Parents are viewed as
more responsible for the person’s schizophrenia or drug
dependence relapse than children (p ? .05). Generally,
siblings were the least pitied of the four groups.
How Does Family Stigma Compare to Primary
The final question examined in this paper was how the
public endorses family stigma compared to primary stigma.
Means and standard deviations of participant responses to
242CORRIGAN, WATSON, AND MILLER
the seven primary stigma items are summarized in Table 2
by health condition. Results of a oneway MANOVA were
significant, F(14, 1890) ? 99.56, p ? .001. Subsequent
ANOVAs for each of the seven items were all significant; F
ranged from 7.85, p ? .05 to 319.02, p ? .001. Post hoc
Tukey’s test showed that primary stigma for drug depen-
dence was the worst. The sample rated the person with drug
dependence as more dangerous (p ? .05), fearful (p ? .05),
blameworthy (p ? .05), anger arousing (p ? .05), and likely
to be avoided (p ? .05). They were also rated as less pitiable
Differences in Courtesy Stigma Rated by Health Conditions
1. Family member bears some responsibility for person originally getting ill.
2. Person’s illness could rub off on family member.
M SD SD SD
3. When person relapses, it may be family member’s fault.
4. Family member should feel ashamed about person’s illness.
5.was not a very good family member to person with illness.
6. I would not want to socialize with family member.
I would NOT be likely to pity family member.a
siblings, and spouse.
aThis item is reverse scored from the original survey item. The original stated: “I would be likely
to pity family member.”
Findings are listed as a whole (overall) and subdivided by family role: parents, children,
(p ? .05) and worthy of help (p ? .05). The person with
schizophrenia was rated as more dangerous (p ? .05),
fearful (p ? .05), and likely to be avoided (p ? .05) than the
person with emphysema. Results showed that the person
with emphysema was viewed as more responsible (p ? .05)
for his or her disorder than the person with schizophrenia.
Table 3 represents the mean and standard deviation of the
total scores for primary stigma and for family stigma. Re-
sults of a 3 x 2 ANOVA suggested that primary stigma was
endorsed more highly than family stigma, F(1, 939) ? 409,
p ? .001. Post hoc Tukey’s tests examined the differences
between primary and family stigma across the health con-
ditions. In all cases where paired tests yielded significant
results, survey participants endorsed primary stigma more
than family stigma. In terms of schizophrenia, participants
rated primary stigma greater. Total score was also ranked
more highly for drug dependence.
Surveys of family members of people with mental illness
or drug dependence conclude that family members experi-
ence significant family stigma. They report being blamed
for the onset of their relative’s disorder, held responsible for
relapse, and being viewed as an incompetent family member
(Corrigan & Miller, 2004). This has led to feelings of shame
and contamination. Our study was guided by the following
question: How does a sample of the American public actu-
ally view family members of people with mental illness or
drug dependence? Several interesting trends emerged. First,
the public does not seem to highly endorse family stigma of
mental illness or drug dependence. A second way to deter-
mine the depth of family stigma is to compare it to primary
stigma using the same vignette. Results suggest the sample
in this study was less likely to endorse stigmatizing attitudes
about family members compared to people who directly
experience the health conditions. Survey participants, on
average, produced higher overall stigma scores for people
with drug dependence disorders and with schizophrenia
than other family members. Survey participants were also
more likely to avoid people with schizophrenia and with
drug dependence disorders compared to their family.
A second question asked the following: Despite the low
level of family stigma, does the public discriminate among
health conditions? Results suggest that families with a rel-
ative with drug dependence disorder are viewed most
harshly. Compared to the vignettes of people with schizo-
phrenia and emphysema, results showed that family mem-
bers with relatives who were drug dependent were blamed
more for the onset of their relatives’ conditions and for
relapses, although this latter difference was not significant
between the vignettes on drug dependence and those on
schizophrenia. Family members in the vignette on drug
dependence were viewed as more likely to be contaminated
Differences in Primary Stigma Rated by Health Conditions
1. I feel NO pity for the person with
2. The person with mental illness is
likely to be dangerous.
3. I feel scared of the person with
4. It is the person’s own fault that
he/she is in the present condition.
5. I feel angry toward the person
with mental illness.
6. If you knew him/her, how likely
is it that you would NOT help the
person with mental illness.a
7. If you know him/her, how likely
is it that you would stay away
from the person with mental
aThese items are reverse scored from the actual survey item. The original stated: “I feel pity for
the person with mental illness.” and “If you knew him/her, how likely is it that you would help the
person with mental illness?”
Selected Within-Group Differences in Courtesy and
Primary Stigma Across Health Conditions
244CORRIGAN, WATSON, AND MILLER
by the disorder, more shameful, and more likely to be
Family stigma of people with mental illnesses like
schizophrenia was less deleterious. We chose emphysema
as a physical health condition because of the perception that
people suffering with this disorder are more blameworthy
because of a past smoking history (Chapple et al., 2004).
And, in fact, the primary stigma related to blame was more
highly endorsed for emphysema than schizophrenia. How-
ever, no difference was found in blaming family members
for emphysema compared to schizophrenia. In fact, the only
family stigma item that differed significantly across these
two groups was for pity, with the schizophrenia family
members viewed as more pitiable than the emphysema
Feeling shame because of mental illness or public stigma
might have significant impact on illness career (Pescosolido
& Boyer, 1999). Research has shown that people with
greater stigma are later to admit their illness, less likely to
begin treatment, and more likely to drop out of treatment
prematurely (Corrigan, 2000). Shame and self-blame are
frequently the stigma that lead to diminished illness career.
Clinicians may opt to adjust treatments so that stigma does
not become a barrier to participation. For example, clini-
cians could avoid words that are likely to elicit stigma.
These include references to onset and offset responsibility
and to feelings of shame.
Combined with our earlier findings, these results suggest
that families with a relative who is dependent on drugs are
viewed in a stigmatizing manner by the public while those
with a relative with mental illness are not. What might
account for the difference between this public survey and
the perceptions of families with mental illness? The differ-
ence may represent a history effect, that is, family stigma
has diminished over the five plus years since these family
surveys were completed (Phelan et al., 1998; Wahl & Har-
man, 1989). This seems unlikely given other recent data that
suggest that primary stigma has actually worsened over the
past four decades (Phelan, Link, Stueve, & Pescosolido,
2000). Alternatively, the disparity between family percep-
tions and public report may represent the effects of social
desirability. Members of the general public are unwilling to
endorse the family stigma that, in fact, exists and to risk
Social scientists have tested a variety of implicit mea-
sures that measure stigma without the influence of the
desirability effect (Fazio, Jackson, Dunton, & Williams,
1995; Greenwald & Banaji, 1995). Future research should
include implicit measures to determine if the low rate of
family stigma in terms of mental illness is still evident. The
difference between public perceptions and family reports
may represent self-stigma. Namely, families with a relative
with mental illness may internalize prejudice (Corrigan &
Miller, 2004; Corrigan & Watson, 2002) which, in turn,
diminishes the self-esteem and self-efficacy of family mem-
bers. If self-stigma accounts for family reports of family
stigma, then future research should find a significant asso-
ciation between internalized stigma and a family member’s
perception of how others stigmatize him or her.
A third goal of this study was to determine whether
family stigma varies by role. Does the public view parents,
siblings, children, and spouses differently? Results suggest
a difference in perception, but the specific difference de-
pends on the type of stigma. Results showed that adults in
a family with an immediate relationship with the person
with a health condition—parents and spouses—are more
likely to be viewed as responsible for the health condition.
This effect was found for all three conditions, although
parent and spouse blame was significantly worse for the
relative with drug dependence. Parents were also viewed as
significantly more responsible for the relapse for children in
the two psychiatric vignettes on psychiatric conditions.
Children were more likely to be viewed as contaminated by
all three disorders than the other groups, with drug depen-
dence once again showing statistically more contamination
than schizophrenia or emphysema. In all, these findings
suggest that the public distinguishes between the role of
parents and children in terms of stigma and judges family
role most harshly in families that have a person with drug
Despite these positive findings, this study is limited. First,
the technology used to recruit and test research participants
resulted in some bias in the sample. Research participants
were required to be phone users with sufficient interest in
web-mail to be willing to sign up for KN efforts. This group
was obviously not an unbiased subset of the population.
Second, the measures used in the study were mostly single
items with no clear information about their psychometrics.
This problem could be partially overcome when future
studies are able to replicate these findings using the same
Advocacy groups have sought to erase family stigma and
associated discrimination, that is, they are to blame for their
loved one’s illness and should be kept away from the person
with mental illness, especially during treatment. Efforts by
parental advocacy groups like the National Alliance for the
Mentally Ill (NAMI), the largest group of this kind in the
United States, have diminished the stigma and promoted
appropriate affirmative action: increase parental involve-
ment in treatment. Antistigma efforts have included protest,
asking participants to suppress their negative attitudes about
a group; education, contrasting the myths of mental illness
with the facts; and contact, decreasing stigma by fostering
interactions between a person with mental illness and a
group where such stereotypes might exist. Most of the
research on stigma change has focused on decreasing pri-
mary stigma. However, primary stigma seem like adequate
candidates for also diminishing family stigma.
Angermeyer, M. C., Matschinger, H., & Corrigan, P. W. (2004).
Familiarity with mental illness and social distance from people
with schizophrenia and major depression: Testing a model using
data from a representative population survey. Schizophrenia Re-
search, 69, 175–182.
Angermeyer, M. C., Schutze, B., & Dietrich, S. (2003). Courtesy
stigma: A focus group study of relatives of schizophrenia pa-
tients. Social Psychiatry and Psychiatric Epidemiology, 38, 593–
Burk, J. P., & Sher, K. J. (1990). Labeling the child of an alco-
holic: Negative stereotyping by mental health professionals and
peers. Journal of Studies on Alcohol, 51, 156–163.
Chapple, A., Ziebland, S., & McPherson, A. (2004). Stigma,
shame, and blame experienced by patients with lung cancer:
Qualitative study. British Medical Journal, 328, 1470–1473.
Corrigan, P. W. (2000). Mental health stigma as social attribution:
Implications for research methods and attitude change. Clinical
Psychology-Science and Practice, 7, 48–67.
Corrigan, P. W. (2005). On the stigma of mental illness: Implica-
tions for research and social change. Washington, DC: Ameri-
can Psychological Association.
Corrigan, P. W., Lurie, B., Goldman, H., Slopen, N., Medasani,
K. M., & Phelan S. (in press). How adolescents perceive the
stigma of mental illness. Psychiatric Services.
Corrigan, P. W., Markowitz, F. E., Watson, A. C., Rowan, D., &
Kubiak, M. A. (2003). An attribution model of public discrimi-
nation towards persons with mental illness. Journal of Health
and Social Behavior, 44, 162–179.
Corrigan, P. W., & Miller, F. E. (2004). Shame, blame, and
contamination: A review of the impact of mental illness stigma
on family members. Manuscript submitted for publication.
Corrigan, P. W., River, L. P., Lundin, R. K., Uphoff-Wasowski,
K., Campion, J., Mathisen, J., et al. (1999). Predictors of partic-
ipation in campaigns against mental illness stigma. Journal of
Nervous and Mental Disease, 187, 378–380.
Corrigan, P. W., River, L. P., Lundin, R. K., Uphoff-Wasowski,
K., Campion, J., Mathisen, J., et al. (2000). Stigmatizing attri-
butions about mental illness. Journal of Community Psychology,
Corrigan, P. W., Rowan, D., Green, A., Lundin, R., River, L. P.,
Uphoff-Wasowski, K., et al. (2002). Challenging two mental
illness stigmas: Personal responsibility and dangerousness.
Schizophrenia Bulletin, 28, 293–310.
Corrigan, P. W., & Watson, A. C. (2002). The paradox of self-
stigma and mental illness. Clinical Psychology-Science and
Practice, 9, 35–53.
Fazio, R. H., Jackson, J. R., Dunton, B. C., & Williams, C. J.
(1995). Variability in automatic activation as an unobstrusive
measure of racial attitudes: A bona fide pipeline? Journal of
Personality and Social Psychology, 69, 1013–1027.
Goffman, E. (1963). Stigma: Notes on the management of spoiled
identity. Englewood Cliffs, NJ: Prentice Hall.
Greenberg, J. S., Kim, H. W., & Greenley, J. R. (1997). Factors
associated with subjective burden in siblings of adults with
severe mental illness. American Journal of Orthopsychiatry, 67,
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cogni-
tion: Attitudes, self-esteem, and stereotypes. Psychological Re-
view, 102, 4–27.
Hogan, M. F. (2003). New freedom commission report: The pres-
ident’s new freedom commission: Recommendations to trans-
form mental health care in America. Psychiatric Services, 54,
Link, B. G., Phelan, J. C., Bresnahan, M., Stueve, A., & Pescoso-
lido, B. A. (1999). Public conceptions of mental illness: Labels,
causes, dangerousness, and social distance. American Journal of
Public Health, 89, 1328–1333.
Martin, J. K., Pescosolido, B. A., & Tuch, S. A. (2000). Of fear
and loathing: The role of “disturbing behavior,” labels, and
causal attributions in shaping public attitudes toward people with
mental illness. Journal of Health and Social Behavior, 41, 208–
Mehta, S. I., & Farina, A. (1988). Associative stigma: Perceptions
of the difficulties of college-aged children of stigmatized fathers.
Journal of Social and Clinical Psychology, 7, 192–202.
Oestman, M., & Kjellin, L. (2002). Stigma by association: Psy-
chological factors in relatives of people with mental illness.
British Journal of Psychiatry, 181, 494–498.
Ohaeri, J. U., & Fido, A. A. (2001). The opinion of caregivers on
aspects of schizophrenia and major affective disorders in a ni-
gerian setting. Social Psychiatry and Psychiatric Epidemiology,
Pescosolido, B. A., & Boyer, C. A. (1999). How do people come
to use mental health services? Current knowledge and changing
perspectives. In A. V. Horwitz & T. L. Scheid (Eds.), The
sociology of mental health and illness (pp. 392–411). New York:
Cambridge University Press.
Pescosolido, B. A., Monahan, J., Link, B. G., Stueve, A., &
Kikuzawa, S. (1999). The public’s view of the competence,
dangerousness, and need for legal coercion of persons with
mental health problems. American Journal of Public Health, 89,
Phelan, J. C., Bromet, E. J., & Link, B. G. (1998). Psychiatric
illness and family stigma. Schizophrenia Bulletin, 24, 115–126.
Phelan, J. C., Link, B. G., Stueve, A., & Pescosolido, B. A. (2000).
Public conceptions of mental illness in 1950 and 1996: What is
mental illness and is it to be feared? Journal of Health and Social
Behavior, 41, 188–207.
Phillips, M. R., Pearson, V., Li, F., Xu, M., & Yang, L. (2002).
Stigma and expressed emotion: A study of people with schizo-
phrenia and their family members in China. British Journal of
Psychiatry, 181, 488–493.
Rossi, P. H., & Nock, S. L. (Eds.) (1982). Measuring social
judgments: The factorial survey approach. Beverly Hills, CA:
Shibre, T., Negash, A., Kullgren, G., Kebede, D., Alem, A.,
Fekadu, A., et al. (2001). Perception of stigma among family
members of individuals with schizophrenia and major affective
disorders in rural Ethiopia. Social Psychiatry and Psychiatric
Epidemiology, 36, 299–303.
Struening, E. L., Perlick, D. A., Link, B. G., Hellman, F., Herman,
D., & Sirey, J. A. (2001). Stigma as a barrier to recovery: The
extent to which caregivers believe most people devalue consum-
ers and their families. Psychiatric Services, 52, 1633–1638.
Thompson, E. H., & Doll, W. (1982). The burden of families
coping with the mentally ill: An invisible crisis. Family Rela-
tions, 31, 379–388.
Wahl, O. F., & Harman, C. R. (1989). Family views of stigma.
Schizophrenia Bulletin, 15, 131–139.
Weiner, B. (1995). Judgments of responsibility: A foundation for a
theory of social conduct. New York: The Guilford Press.
Weiner, B., Perry, R. P., & Magnusson, J. (1988). An attributional
analysis of reactions to stigmas. Journal of Personality and
Social Psychology, 55, 738–748.
Received October 5, 2004
Revision received February 24, 2005
Accepted March 29, 2005 ?
246 CORRIGAN, WATSON, AND MILLER