Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jul 1.
Published in final edited form as: J Nerv Ment Dis. 2021 Jul 1;209(7):510–517. doi: 10.1097/NMD.0000000000001327

Social Identities of Persons with Schizophrenia and Social Functioning: Individual and Family Caregiver Perspectives

Vanesa Pérez a, Luisa Elena Hernández Melo b, María del Carmen Lara-Muñoz c, Alex Kopelowicz d, Jodie Ullman e, Steven Regeser López f
PMCID: PMC8239249  NIHMSID: NIHMS1670129  PMID: 34170860

Abstract

We apply social identity theory and self-categorization theory to examine the role of social identities in relation to the recovery of persons with schizophrenia. We assess whether illness-based and non-illness-based identities held by both those with schizophrenia and their caregivers explain additional variance in social functioning in persons with schizophrenia beyond the previously established predictors of negative symptoms and theory of mind. Sixty Mexican-origin adults diagnosed with schizophrenia and their family caregivers were obtained through an outpatient mental health clinic located in either Los Angeles, California or in Puebla, Mexico. A three-step hierarchical regression indicated that identity endorsements, from both the perspective of the person with schizophrenia and their caregiver, and negative symptomatology are significant independent predictors of social functioning. Specifically, greater endorsement of non-illness identities both for the person with schizophrenia and also their caregiver are associated with higher social functioning. Illness identity plays an important role in the path to recovery.

Keywords: Schizophrenia, Mexican-origin, social identity, family caregivers, social functioning

Introduction

The consumer-led recovery movement (Bellack, 2006) has helped redefine recovery from schizophrenia as a process that includes participation in the social world (e.g., vocational fulfillment). In consultation with many stakeholders, the U.S. Substance Abuse and Mental Health Services Administration (2020) further refined the definition as “A process of change through which individuals improve their health and wellness, live a self-directed life, and strive to reach their full potential.” Social functioning is an integral component of recovery. In identifying the predictors of social functioning, quantitative researchers have attended primarily to negative symptoms and social cognition (e.g., Javed and Charles, 2018). Negative symptoms are of particular interest as they account for difficulty in many social domains, such as practicing independent living (Harvey et al., 2019). Social cognition, a multifaceted construct referring to the mental operations underlying social behavior, has also emerged as a robust predictor of functional outcomes (Bora, 2017; Green, 2016). Specifically, many symptoms typical of schizophrenia are hypothesized to be accounted for by difficulties in theory of mind, which is the cognitive ability to accurately attribute mental states – such as thoughts, beliefs, and intentions – to others. Theory of mind plays a key role in individuals being able to explain, manipulate, and predict behaviors (Frith, 1992; Sprong et al., 2007).

First-person accounts of serious mental illness point out that the recovery process is incremental, and part of a complex subjective process embedded in a rich social context (Jenkins et al., 2005; Ridgway, 2001). In contrast, quantitative researchers’ growing attention to recovery and social functioning has largely overlooked the role of the social context in shaping the everyday lives of persons with schizophrenia (Santos et al., 2018). Social psychologists have identified multiple paths in which social connectedness is associated with well-being, health, and mental health in the general public (e.g. Cohen, 2004). In the present study, we draw from social identity theory (Tajfel, 1978; Tajfel et al., 1979) and self-categorization theory (Turner et al., 1987) to consider the role that social identities or group memberships have on the social functioning of persons with schizophrenia. Social identity refers to the individual’s knowledge that they belong to certain social groups as well as the emotional valence and personal significance in being a member of that group (Turner, 1975).

Narratives of individuals with schizophrenia and related disorders reveal a distinction between the ill self and well self (Davidson & Strauss, 1992; Jenkins & Carpenter-Song, 2005). Other studies indicate a reduction of their past identities to a single focus on their illness (e.g., Morgades-Bamba et al., 2019). In fact, Estroff (1989, p. 189) refers to schizophrenia as an “…I am illness, one that is joined with social identity and perhaps with inner self, in language and terms of reference.” What is missing from the study of self-construal and self-identity within schizophrenia is the consideration of family caregivers’ construals of their loved ones with serious mental illness. The social identity that others hold of individuals with schizophrenia may also play a role. For instance, caregivers who prioritize medication management and symptom reduction may serve to strengthen their relative’s illness identity. In contrast, those caregivers who attend to their loved ones as wage earners, friends, romantic partners, or family members, may serve to bolster their other identities, and in turn, their social functioning. Identity construals may serve as an important prosocial mechanism to consider in recovery.

The social identities of those with schizophrenia may differ from the identities that their caregivers hold of their loved ones. When thinking about the future, multiple social identities appear to be more salient for those with schizophrenia whereas illness/wellness related identities appear to be more salient for caregivers (Santos et al., 2018). A more direct test would help clarify whether persons with schizophrenia and their caregivers do indeed differ in the social identities that they hold.

Drawing on participants for whom families play a significant role in their daily lives provides an ideal context to first test whether caregivers’ construals matter. Accordingly, we chose to study families of Mexican origin. Prior research in California indicates that persons with schizophrenia of Latinx origin (mainly of Mexican heritage) are more likely to live with family members than persons with schizophrenia from other ethnic/racial groups (Snowden, 2007). In addition, prior research on the illness course of schizophrenia among Mexican Americans point out the important role of family factors (Jenkins & Karno, 1992). Another reason for studying families of Mexican origin is that there has been little attention to the study of recovery among ethnic/racial minority group members. It is important to extend the reach of the recovery-based research literature to a broad range of community groups.

In the present cross-sectional study, we aim to extend the quantitative study of social functioning by examining the social identities that individuals with schizophrenia hold of themselves, as well as the social identities that their caregivers hold of them. We first test the hypothesis that identity construals will differ for caregivers than for those with schizophrenia. We expect caregivers to endorse relatively more illness identities than non-illness identities when compared to those with schizophrenia (Santos et al., 2018). We then examine the interrelations of social identities with negative symptoms, theory of mind, and social functioning. Given past research, we first test the hypothesis that more severe negative symptoms and poorer theory of mind will be associated with worse social functioning. In addition, given social identity theory, we expect that the endorsement of relatively more non-illness identities for both persons with schizophrenia and their caregivers will be associated with better social functioning. We also explore whether social identities explain additional variance of social functioning above that explained by negative symptoms and theory of mind. Finally, we include gender and age in our analyses, as they are relevant social identities and are associated with social functioning (Gould et al., 2012; Usall et al., 2007).

Methods

Participants

Our sample consisted of 60 individuals (61.7% male; M = 38.90 years, SD =11.33) with schizophrenia and their corresponding 60 familial caregivers (78.3% female; M = 54.80 years, SD =15.67). Participants were recruited from two public outpatient clinics, one located in Southern California, United States (n = 21) and the other in Puebla, México (n =39). Please see table 1 for a complete summary of the total sample demographics.

Table 1:

Demographics

Individuals with Schizophrenia
(N=60)		 Caregivers
(N=60)
Variable n % M SD n % M SD
1. Gender
Male 37 62% 13 22%
Female 23 38% 47 78%
2. Age 38.9 11.33 54.8 15.67
3. Years of Education 10.44 3.36 7.35 4.63
4. Employment
Yes 16 27% 34 57%
No 44 73% 26 43%
5. Place of Birth
U.S. 12 20% 5 8%
Mexico 48 80% 55 92%
6. Home Ownership
Yes 6 10% 16 27%
No 54 90% 44 73%
Open in a new tab

Eligible participants: a) had a clinic-based diagnosis of schizophrenia, b) were between 18 and 65 years of age, c) were of Mexican origin, d) were fluent in English, Spanish or both languages, and e) had a close relative willing to participate in the study. Caregiving relatives were those identified as the primary caregiver and could be either a mother, father, sibling, son, or daughter. Both the person with schizophrenia and their caregiver received monetary compensation for their participation. The research protocol was approved by the University of Southern California’s Office for the Protection of Human Subjects.

Measures

Social functioning.

We administered The Role Functioning Scale (RFS) that assesses 4 domains of functioning: working productivity, independent living/self-care, immediate social network relationships, and extended social network relationships (Goodman et al., 1993). Raters applied a scale of 1 (severely limited level) to 7 (optimal level) for each domain. The points of the rating scale were clearly delineated. For example, independent living/self-care referred to “Management of household, eating, sleeping, hygiene care.” A “7” referred to “Optimal care of health/hygiene; independently manages to meet personal needs and household tasks.” The internal consistency of the RFS was .88 for our sample. We also applied the Strauss-Carpenter Levels of Functioning Scale (Strauss and Carpenter, 1977); the internal consistency of the four items regarding social relations and useful work was .67. The RFS and the 4 items of the Strauss-Carpenter are significantly related (r = .75, p < .001). We carried out our analyses using the RFS given its relatively high reliability. We report the mean item ratings to facilitate interpretation of the data.

Symptomatology.

The Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987), which applies a 1 (absent) to 7 (extreme) severity scale for each symptom, was administered. The negative symptom scale demonstrated good reliability in our sample (7 items, α = .80). We report the mean item ratings to help interpret the degree of severity.

Theory of mind.

We used the Hinting task (Corcoran et al., 1995) for which interviewers read 10 brief stories and asked participants to infer the intentions behind the indirect speech of the main character in each story. Two bilingual raters, who had no role in collecting the data, were trained and rated all the Hinting Task responses. Their inter-rater reliability was .77. Discordant ratings were discussed with the Principal Investigator and a consensus was reached. Raters did not know whether the Spanish-speaking participants resided in Los Angeles or Puebla. We also made a few minor adjustments to the original British and Spanish language versions provided by the original authors so that our participants would fully understand the stories. We report the mean of the participants’ total score across the 10 stories.

Identity construals.

A semi-structured interview was used to derive the identity construals held by the individual with schizophrenia and the familial caregiver. The interview consisted of open-ended questions designed to gather qualitative data about the daily lives and interactions of the individual with schizophrenia and their family. The interviews were adapted to the informant’s perspective. For example, when asking about the daily routine, the interviewer asked the person with schizophrenia, “How do you pass your day?”, versus “How does (name of relative) pass their day?” when speaking with the caregiver. Other questions addressed: shared activities between the person with schizophrenia and their family, the behavior of the individual with schizophrenia when they did not feel well, and what each person most appreciated of one another. The average length of the interviews was 24 minutes for individuals with schizophrenia and 30 minutes for family caregivers.

The interviews were audio-recorded, transcribed, and coded. A coding manual was created for this study in which identity which was rated on a 5-point scale ranging from −2 (high sickness identity and low social role) to +2 (low sickness identity and high social role). Please see table 2 for examples that reflected each of the 5 points on the continuum from the individual with schizophrenia and caregiver’s perspectives.

Table 2:

Coded examples of identity construals

Code Individual with Schizophrenia Caregiver
−2 = high illness-based role and a low social-based role. Interviewer (I): “How much time do you spend with your family? Do your cousins visit, or what does your family do?”
Individual with schizophrenia (IS): “No, they don’t visit because of my illness.”
I: “They don’t come?”
IS: “Only when my sister comes, I mean. My nephews make me nervous. I tell my mom, ‘Mom, my nephews make me nervous and I don’t want them in the house.’ That’s why we don’t have friends. That’s why family doesn’t visit. I take a lot of medication. I get sad, you know. I get sad. My father tells me, ‘go the park.’ ‘No.’ ‘Why not?’ ‘There is a lot of people. I don’t like it. I don’t like being around people. I like, like, I’m a loner, you know?” 		“Yes, I would like to give him more affection, but I get scared because of his illness. I do hug him sometimes, but with fear.”
−1 = moderate illness-based role and low social-based role. I: “And what kind of work do you do?”
IS: “Um I tried to serve food and all that, but…but it scared me to see so many people, so many persons, and I got paranoid or something and felt like I said nasty things. Ultimately, I just quit.” 		“Sometimes my wife takes him to do laundry, I’ll take him to the store, or I take them to the park, To the park that is right here. Like that. We take him to the doctor, with my wife, so they can give him his medication. Um he always wants to eat out at a restaurant, or my wife makes him something here [at home]. He’s like a child.”
0 = moderate in both illness-based and social-based roles. “On Monday, for example, on Monday I try to clean since I don’t do chores on Saturday or Sunday. On Monday I start… while they are not here and neither is my son, nor my mom, I start to clean the kitchen, Then I clean up over here in the living room and then the bathroom and then I make my son’s bed. Then I start to review what is English. And suddenly I start to read the bible. On Sundays never miss church of the Catholic church. I’m Catholic. I go to the clinic Monday through Friday from… I’m over there between 8 and 8:30. Every day.” From sibling: “Because of her illness, my mom gets riled up and, at times, she gets in a bad mood when she sees a mess and starts to argue. Besides the illness, she is very understanding, we support each other, and she is very cheerful. When she’s not sick, I leave my kids with her for her to watch and she loves them very much.”
1 = low illness-based role and moderate social-based role. IS: “Um, when I’m not feeling well? I feel a little anxious, um stressed out maybe, uh, sometimes I feel like I’m not doing enough with my life, not doing what I’m capable of. Sometimes I feel I’m not living to my full potential, and that kind of stresses me out. I feel like I can do more with my day.”
I: But what’s stopping you from not doing that? Aside, you know, from…”
IS: “Uh sometimes I feel it’s the medication that slows me down, but sometimes I feel a little lazy, or just sometimes I don’t want to get up.”
I: “And you think that’s due to the medicine?”
IS: “I think so. I mean before I used to be pretty active. But lately, after going, as I’m going to the gym and everything, I’ve been feeling more active, so I think that helps.”
I: “How often do you go to the gym? Do you go every day or?”
IS: “We, we try to go five days a week, so Monday through Friday we go to the gym, like from 9pm to 11pm.” 		“Almost, almost every day. Because sometimes, I don’t know, there are days – not always- that he gets dressed, but he doesn’t shave. But most of the times he wears a suit, puts cologne, and bathes, and even if he is just going to see about his medication or to visit his brothers, he gets dressed up like that. Very well-dressed.”
2 = low illness-based role and high social-based role. “In the morning I wake up like around 7:30, maybe 8 and then I, you know do my activities of daily living. And then I go to… um… sometimes I go to the library, you know, go use the computer to job hunt. Then I come back and I eat, my food. Throughout my life I’ve been normal so it was something that surprised me but I learned to deal with it.” “We see him as a normal person even though we know he is in treatment. I think and say that he will be a normal person, because he has his family, he has his two daughters…”
Open in a new tab
*

Coded examples were translated from the original transcription in Spanish.

Two pairs of raters with no prior contact with the data were trained on the coding scheme. One pair of raters coded responses from the person with schizophrenia and the other pair of raters independently coded responses from the familial caregiver. Inter-rater reliabilities were computed for the pair of coders’ overall rating for each participant (ICC = .77 for persons with schizophrenia and ICC = .78 for caregivers). Discordant ratings were discussed with the Principal Investigator to reach consensus.

Procedure

During the initial phase, a team of four research assistants (RAs) applied the semi-structured interview and measures to six dyads in Puebla, Mexico. Three of the RAs were bilingual Latinx-Americans and had been trained in the United States. They participated in a summer research training program in Mexico, but their permanent residence was in the US. The fourth RA was trained and resided in Mexico. One of the US RAs and the Mexican RA then served as a key member of their respective research teams in Los Angeles and Puebla.

In Los Angeles, the RA was joined by another BA-level RA and an experienced psychiatric nurse, both native Spanish-speakers of Latinx background. In Mexico, the RA and a third-year psychiatry resident collected the data. The psychiatric nurse, who served as a clinical rater on previous studies of schizophrenia (e.g., Kopelowicz et al., 2012), and the psychiatry resident administered the PANSS. The resident was trained by the senior Mexican investigator, an expert in measurement (Lara-Muñoz & Ortega-Soto, 1995).

The two research teams visited with the family dyads at their homes, and on some occasions, at the clinic. The study team carried out separate semi-structured interviews with each member of the dyad. Assessments lasted from 1-2 hours and all participants received compensation. In Los Angeles, the interviews were conducted in the participant’s preferred language—Spanish or English.

Results

Assumptions and missing data

Means, standard deviations, skewness and kurtosis were obtained for all measured variables prior to hypothesis testing to ensure the quality of the data. None of the variables exceeded conventional cutoffs of ±2 for skewness (range = −.22 to .57) and ±7 for kurtosis (range = −.98 to −.22), suggesting data met assumptions of normality (West, Finch, & Curran, 1995). Data were also screened for outliers using a criterion of z +/− 3.3, p < .001 and no univariate outliers were detected.

Overall, there were very few cases missing data. The total sample size was 60 dyads (60 persons with schizophrenia and 60 caregivers). One case was missing the age of the person with schizophrenia (1.7%). Negative symptoms, social functioning, and the social identity of both parties were missing in four cases (6.7%). Missing data for the variables used in the study were evaluated and no patterns of missingness were detected, therefore we assumed that the data were MCAR (missing completely at random) and used the EM algorithm to impute the missing data (Schafer and Graham, 2002). The following analyses were conducted with imputed data.

Mexican American vs. Mexican Subsamples

Differences among the Mexican American (n=21) and Mexican subsamples (n=39) were assessed with a Bonferroni correction to adjust for Type 1 error given the multiple tests (.05/7 = .007). As noted in Table 3, there were no significant group differences with regard to age of the person with schizophrenia, age of caregiver, social functioning, theory of mind, and the identity construals of those individuals with schizophrenia and their caregivers (p-values = .02-.45). In addition, there were no group differences regarding the gender of persons with schizophrenia (χ [1] =.34, p = .56) or their caregivers (χ [1] =.09, p = .77). The only significant difference was found with regard to negative symptoms. Mexican Americans (M = 2.53; SD = 0.75) were rated as having less severe negative symptoms than Mexicans (M =3.33; SD = 0.98; t [58] = - 3.25, p = .002; d = .92). The negative symptoms of both groups were on the low end, however, with Mexican Americans reporting on average between minimal and mild symptoms and with Mexicans reporting mild symptoms. Given that the difference in negative symptoms has marginal clinical significance and given the small subsample sizes, we opted to collapse the two groups.

Table 3:

Descriptive Statistics for Key Study Variables by Country

U.S. Sample
n = 21		 Mexico Sample
n = 39		 Total Sample
N = 60		 t-test
Variable M SD M SD M SD df t
1. Age of IS 36.23 12.87 40.05 10.32 38.72 11.32 58 −1.25
2. Age of CG 52.38 14.25 56.10 16.41 54.80 15.67 58 −.88
3. Negative Symptoms 2.53 .75 3.33 .98 3.05 .98 58 −3.25*
4. Theory of Mind 7.90 4.70 8.90 4.95 8.55 4.85 58 −.75
5. Identity-IS .06 1.51 .36 1.09 .26 1.25 31.57a −.82
6. Identity-CG .40 1.34 −.36 1.08 −.09 1.22 58 2.40
7. Social Functioning 4.42 1.69 3.89 1.03 4.08 1.31 28.22a 1.29
Open in a new tab

Note: (IS) = Individual with schizophrenia; (CG) = Caregiver

*

p < .007 (based on Bonferroni type adjustment)

a

Degrees of freedom based on Welch’s t-test given unequal variances in the 2 groups

Social Identities by Persons with Schizophrenia and Caregivers

A paired t-test indicated that persons with schizophrenia (M = .26, SD = 1.25, n = 60) held relatively more non-illness social identities than their caregivers (M = - .09, SD= 1.22, n = 60. Although this difference was not significant, t (59) = 1.79, p = .08, there was a small effect size for the difference in the two groups identity construals (d = .20).

Interrelations of Predictors and Social Functioning

As hypothesized, more severe negative symptoms were related to poorer social functioning (r = −.52; p < .001) whereas stronger theory of mind was associated with higher social functioning (r = .33; p < .01). In addition, more negative symptoms were associated with worse theory of mind (r = −.30; p <.05).

Also consistent with our hypotheses, both the individual with schizophrenia and caregiver’s identity construals were significantly related to social functioning (individuals with schizophrenia r = .52, p < .001; caregivers r = .58, p < .001). Higher non-illness social identities were associated with greater theory of mind for individuals with schizophrenia r = .35, p < .01. This association was not significant for caregivers (r = .23, p = .08). Negative symptoms were significantly and inversely related to a higher non-illness social identity as endorsed by the individual with schizophrenia (r = −.28, p < .05) and their caregiver (r = −.45, p < .001). Please refer to table 4 for a complete summary of correlations.

Table 4:

Correlations among Key Study Variables

Variable 1 2 3 4 5 6 7 8 9
1. Gender (IS) --
2. Age (IS) .08 --
3. Gender (CG) −.17 −.10 --
4. Age (CG) −.22 .30* −.18 --
5. Negative symptoms −.09 .09 −.10 .19 --
6. Theory of mind .23 .11 −.05 .00 −.30* --
7. Identity (IS) .03 .12 −.11 −.08 −.28* .35* --
8. Identity (CG) .12 −.15 .11 −.31* −.45* .23 .25 --
9. Social functioning .14 .01 −.09 −.13 −.52* .33* .52* .58* --
Open in a new tab

Note: (IS) = Individual with schizophrenia; (CG) = Caregiver; N = 60

*

p < .05

Testing a Model of Social Functioning: Symptoms, Theory of Mind, and Identity Construals

A three-step hierarchical regression was utilized to further test the hypothesis that negative symptoms and theory of mind are associated with social functioning, and that identity construals of both persons with schizophrenia and their caregivers are related to social functioning over and above the variance explained by negative symptoms and theory of mind. Gender and age of individuals with schizophrenia and their caregivers were first entered in the regressions to control for variability accounted by these demographic characteristics. Together they accounted for 4% of the variability in social functioning, but this was not a significant contribution to the regression model, F (4, 55) = .57, p = .68. At step two, negative symptoms and theory of mind accounted for an additional 29.2% of the variability in social functioning and this change in R2 was significant, F (2, 53) = 11.56, p < .001. In particular, negative symptoms independently predicted lower social functioning, β = −.47; t = −3.93, p <.001. Theory of mind, however, was not a significant predictor when negative symptoms were included in the model, β = .17; t = 1.42, p = .16. Adding identity construals from the perspective of both the individual with schizophrenia and their caregivers at step three explained an additional 22.6% of the variance in social functioning and this change in R2 was also significant, F (2, 51) = 13.00, p < .001. When all variables were included in this step, theory of mind was not a significant independent predictor of social functioning. However, negative symptoms, β = .27; t = - 2.43, p < .05, identity construals from the perspective of the individual with schizophrenia, β = .33; t = 3.11, p < .01, and from the perspective of the caregiver, β = .40; t = 3.58, p < .01, were all significant independent predictors of social functioning. That is, fewer negative symptoms and a greater non-illness social identity predicted higher social functioning. Altogether, the variables accounted for a total of 55.7% of the variance in social functioning. See table 5 for a complete summary of estimates.

Table 5:

Summary of Hierarchical Regression Analysis for Predicting Social Functioning

Model 1:
Demographics		 Model 2:
Demographics, symptoms and theory of mind		 Model 3:
Demographics, symptoms, theory of mind, and identity construals
Variable B SE B β B SE B β B SE B β
Gender (IS) .26 .37 .10 .06 .33 .02 .14 .27 .05
Age (IS) .00 .02 .03 .00 .01 .03 .00 .01 .01
Gender (CG) −.31 .43 −.10 −.43 .37 −.14 −.34 .31 −.11
Age (CG) −.01 .01 −.13 −.00 .01 −.07 .00 .01 .06
Negative Symptoms −.63 .16 −.47* −.35 .14 −.26*
Theory of Mind .05 .03 .17 .01 .03 .03
Identity (IS) .34 .11 .33*
Identity (CG) .42 .12 .40*
R2 .04 33.2 55.7
F for change in R2 .57 11.56* 13.00*
Open in a new tab

(IS) = Individual with schizophrenia; (CG) = Caregiver

*

p < .05

Discussion

We found support for the application of social identity theory (Tajfel et al., 1979) to help understand how the social world can influence the social functioning of persons with schizophrenia. Our findings point out that the illness and non-illness social identities that persons with schizophrenia hold for themselves are related to their social functioning. Specifically, reporting relatively more non-illness social identities than illness social identities is associated with greater engagement in their everyday social life. A similar social identity pathway was observed in family caregivers’ identity construals. As caregivers emphasize the non-illness social identities over the illness social identity, those with schizophrenia exhibited higher social functioning. Our findings are consistent with other applications of social identity theory to health and well-being. For example, stroke sufferers who had a greater number of group memberships prior to their stroke were better adjusted after the stroke than those who had fewer group memberships (Haslam et al., 2008). Compared to those with few social identities, having multiple social identities may contribute to greater social functioning when faced with health challenges because although there will be a loss of some group memberships (e.g., wage earner), others will remain intact (e.g., husband).

One possible reason why we were able to identify the role of identity construals in social functioning is that we used a Mexican-origin sample. Prior research has documented the importance of families in the lives of persons of Mexican origin living in the United States (Aguilera et al., 2010; Jenkins, 1991; Lopez et al., 2004). It may be that in families with strong family ties, the social identities that family caregivers hold of their loved ones with schizophrenia may promote social functioning. This highlights the potential value of studying recovery processes in diverse communities.

Implications

The findings have both theoretical and clinical implications. One prominent theoretical approach to family caregiving focuses on attribution theory (Weiner, 1993). The theory argues that caregivers’ attributions to illness should result in more sympathy and less anger, and, in turn, more support and improved outcomes. It is clear that attributions are related to affective reactions (either criticism or warmth, e.g., Weisman et al., 1998), and affective reactions are related to clinical outcomes (more symptoms or relapse, Butzlaff and Hooley, 1998). Attributions themselves, however, are not related to outcomes (e.g., Lopez et al., 2004). In particular, judging the person as not responsible for their illness related behavior is not related to improved outcomes as the theory suggests. The findings from the current study offer a plausible explanation for why this is the case. Illness attributions may reduce the number of social identities caregivers hold for their loved one and at the same time augment a social identity (a person with serious mental illness) that is stigmatized. Based on social identity theory, attributing behavior to an illness may not be conducive to specific positive outcomes, such as social functioning. Although there are clear benefits to illness attributions (more warmth and less criticism), promoting an illness identity may undermine the potential benefits by limiting opportunities for both the person with schizophrenia and their familial caregiver to work towards behavioral changes that will ultimately lead to improved social functioning. See Lebowitz and Applebaum (2019) for a review of the potential negative implications of biomedical explanations for mental disorders.

With regard to clinical implications, the promotion of recovery includes addressing the social skills necessary to carry out everyday life activities (e.g., Liberman et al., 1986). Social identity theory argues that learning skills is more than simply expanding one’s behavioral repertoire. It includes promoting identity resources as substantial and concrete assets. Learning job-seeking skills, for example, is likely to strengthen one’s identity as a wage earner, and, in turn, promote further engagement in job seeking. Social identity theory would argue for monitoring the number of non-illness social identities persons with schizophrenia have and taking steps to strengthen those identities. This does not necessarily imply that efforts should be made to eliminate the illness identity. An illness identity may be needed to help individuals, family members, and mental health providers promote recovery. Estroff (1989, p. 193) asked two relevant questions: “Is good prognosis associated with a separation from self, a preserving of person who has but is not an illness? Or does the individual fare better by embracing and incorporating schizophrenia as within and of the self?” Our data support the importance of preserving the person who has but is not an illness. By proactively encouraging ways to strengthen multiple non-illness social identities, clinicians (and caregivers) can help to reduce the salience of the illness identity and its possible deleterious effects.

Strengths and limitations

There are a number of strengths to our findings. First, the bivariate correlations and the first two steps of the regression analysis replicated, at least in part, past results of the association between negative symptoms, theory of mind, and social functioning. The fact that social identities proved to be significant independent predictors of social functioning, even when accounting for negative symptoms and theory of mind in the analyses, highlights the additional and unique contribution social identities make in explaining variability in the social functioning aspect of recovery. Second, our measure of social identity considers that individuals can endorse varying levels of both illness- and non-illness identities. Another strength is that we include social identity construals from two key informants: the person with schizophrenia and their familial caregiver. Reports from the self and from significant others, such as familial caregivers, may offer distinct information regarding social identity as informants often vary given the context in which they observe behavior. Our findings revealed that both informants’ perception of the person with schizophrenia’s social identity independently predicted social functioning. It is important to note that the perceptions from both informants were not significantly correlated (r = .25, p > .05). Thus, the social identities endorsed by both informants were not merely reflections of each other and most likely captured meaningful differences, yet still independently predicted social functioning. Future studies should assess informant agreement and discrepancies in measuring social identity as it relates to clinical outcomes (De Los Reyes et al., 2013).

This study is not without limitations. First, there is likely some overlap in the measure of identity construals and social functioning. There are times when references to identity include some mention of the specific behaviors associated with social functioning. For example, a caregiver might point out that their daughter is “a hard worker” referring to the fact that she manages a small hamburger stand outside the house on the weekends. Here the social construal of a “hard worker” is closely linked to the specific behavioral account that would contribute to her rating of social functioning. It is important to point out, however, that the identity construals of relatives with schizophrenia and their caregivers differed at least to some degree as they were not significantly correlated, and that each informant’s construal independently contributed to social functioning. Furthermore, a key strategy of our coding procedure focused on utilizing identity labels (i.e., “loner”, “normal, and “like a child”) to help guide our codes knowing that identity construals are related to, yet distinct from, clinical measures of social functioning. For example, a person with schizophrenia may go to school or hold a job but may still be predominantly described as a person with illness by their caregiver. These factors argue in favor of the interpretation that the construction of the ill relative’s identity is associated with social functioning independent of the overlap between the measures of identity construals and social functioning. If identity construals were largely informed by the persons’ level of functioning, then we would not expect any differences in identity construals between the two groups nor would we expect the two groups’ identity construals to independently predict social functioning. Further contributing to the independence of the measures is that the raters of social construals and social functioning were completely independent.

Another limitation involves the cross-sectional nature of the study. It is not clear if a non-illness identity leads to greater social functioning or vice-versa. It may very well be that the relationship is bidirectional. Greater non-illness identities may contribute to improved social functioning and greater social functioning may lead to greater non-illness identities. A prospective study, particularly one early in the illness course, would help shed light on whether there is a bidirectional relationship between identity construals and social functioning. In depth qualitative studies also have the potential to shed further light on the relationship between self and caregiver construals and recovery.

Conclusion

Although there is increased attention to recovery and social functioning of persons with schizophrenia, this line of quantitative research has given little attention to how the social world can help shape social functioning. The present study applies social identity theory and identifies concrete ways of bringing the social world to the study of social functioning and the promotion of recovery. In particular, the illness and non-illness identities that persons with schizophrenia hold for themselves and that caregivers also hold for them may play important roles in the path to recovery from schizophrenia.

Acknowledgements:

The authors thank Daisy Aceves, José Calderon, Katie Cobian, Edny Josefina Vásquez Bautista, Richard Franco, Natalia Jaramillo, Kitsya Macias, Jose Miguel Ruiz, Athenai Ximena Sánchez Millán, and Aiden Vaewsorn for their contributions.

Disclosures: The authors declare no conflict of interest. This research was funded by National Institute of Minority Health and Health Disparities (T37 MD003405) and the Foundation for Psychocultural Research.

References

  1. Aguilera A, López SR, Breitborde NJ, Kopelowicz A, & Zarate R (2010). Expressed emotion and sociocultural moderation in the course of schizophrenia. Journal of Abnormal Psychology, 119(4), 875–85. 10.1037/a0020908 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bellack AS (2006). Scientific and consumer models of recovery in schizophrenia: concordance, contrasts, and implications. Schizophrenia Bulletin, 32(3), 432–442. 10.1093/schbul/sbj044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bora E (2017). Relationship between insight and theory of mind in schizophrenia: a meta-analysis. Schizophrenia research, 190, 11–17. [DOI] [PubMed] [Google Scholar]
  4. Butzlaff RL, & Hooley JM (1998). Expressed emotion and psychiatric relapse: a meta-analysis. Archives of General Psychiatry, 55(6), 547–52. [DOI] [PubMed] [Google Scholar]
  5. Cohen S (2004). Social relationships and health. American Psychologist, 59(8), 676–684. [DOI] [PubMed] [Google Scholar]
  6. Corcoran R, Mercer G, & Frith CD (1995). Schizophrenia, symptomatology and social inference: investigating “theory of mind” in people with schizophrenia. Schizophrenia Research, 17(1), 5–13. 10.1016/0920-9964(95)00024-G [DOI] [PubMed] [Google Scholar]
  7. Davidson L, & Strauss JS (1992). Sense of self in recovery from severe mental illness. The British Journal of Medical Psychology, 65, 131–45. [DOI] [PubMed] [Google Scholar]
  8. De Los Reyes A, Thomas SA, Goodman KL, & Kundey SMA (2013). Principles underlying the use of multiple informants’ reports. Annual Review of Clinical Psychology, 9(1), 123–149. 10.1146/annurev-clinpsy-050212-185617 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Estroff SE (1989). Self, identity, and subjective experiences of schizophrenia: In search of the subject. Schizophrenia Bulletin, 15(2), 189–196. [DOI] [PubMed] [Google Scholar]
  10. Frith CD (2014). The cognitive neuropsychology of schizophrenia. Psychology press. [Google Scholar]
  11. Goodman SH, Sewell DR, Cooley EL, & Leavitt N (1993). Assessing levels of adaptive functioning: the role functioning scale. Community Mental Health Journal, 29(2), 119–131. 10.1007/BF00756338 [DOI] [PubMed] [Google Scholar]
  12. Gould F, Bowie CR, & Harvey PD (2012). The influence of demographic factors on functional capacity and everyday functional outcomes in schizophrenia. Journal of Clinical and Experimental Neuropsychology, 34(5), 467–75. 10.1080/13803395.2011.651102 [DOI] [PubMed] [Google Scholar]
  13. Green MF (2016). Impact of cognitive and social cognitive impairment on functional outcomes in patients with schizophrenia. The Journal of Clinical Psychiatry, 77(Suppl 2), 8–11. 10.4088/JCP.14074su1c.02 [DOI] [PubMed] [Google Scholar]
  14. Harvey PD, Deckler E, Jarskog LF, Penn DL, & Pinkham A (2019). Predictors of social functioning in patients with higher and lower levels of reduced emotional experience: social cognition, social competence, and symptom severity. Schizophrenia Research, 206, 271–276. 10.1016/j.schres.2018.11.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Haslam C, Holme A, Haslam SA, Iyer A, Jetten J, & Williams WH (2008). Maintaining group memberships: social identity continuity predicts well-being after stroke. Neuropsychological Rehabilitation, 18(5–6), 671–691. [DOI] [PubMed] [Google Scholar]
  16. Javed A, & Charles A (2018). The importance of social cognition in improving functional outcomes in schizophrenia. Frontiers in Psychiatry, 9. 10.3389/fpsyt.2018.00157 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jenkins JH (1991). Anthropology, expressed emotion, and schizophrenia. Ethos, 19(4), 387–431. [Google Scholar]
  18. Jenkins JH, & Carpenter-Song E (2005). The new paradigm of recovery from schizophrenia: cultural conundrums of improvement without cure. Culture, Medicine and Psychiatry: An International Journal of Comparative Cross-Cultural Research, 29(4), 379–413. [DOI] [PubMed] [Google Scholar]
  19. Jenkins JH, Strauss ME, Carpenter EA, Miller D, Floersch J, & Sajatovic M (2005). Subjective experience of recovery from schizophrenia-related disorders and atypical antipsychotics. The International Journal of Social Psychiatry, 51(3), 211–27. [DOI] [PubMed] [Google Scholar]
  20. Jenkins JH, & Karno M (1992). The meaning of “expressed emotion”: Theoretical issues raised by cross-cultural research. American Journal of Psychiatry, 149, 9–21. [DOI] [PubMed] [Google Scholar]
  21. Kay SR, Fiszbein A, & Opler LA (1987). The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophrenia bulletin, 13(2), 261. [DOI] [PubMed] [Google Scholar]
  22. Kopelowicz A, Zarate R, Wallace CJ, Liberman RP, Lopez SR, & Mintz J (2012). The ability of multifamily groups to improve treatment adherence in Mexican Americans with schizophrenia. Archives of General Psychiatry, 69(3), 265–73. 10.1001/archgenpsychiatry.2011.135 [DOI] [PubMed] [Google Scholar]
  23. Lara-Muñoz MC, Ortega Soto H, 1995. ¿La clinimetría o la psicometría? Medición en la práctica psiquiátrica. (Clinimetrics or psychometrics. Measurement in clinical psychiatry). Salud Mental 18(4), 33–40. [Google Scholar]
  24. Lebowitz MS, & Appelbaum PS (2019). Biomedical explanations of psychopathology and their implications for attitudes and beliefs about mental disorders. Annual Review of Clinical Psychology, 15, 555–577. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Liberman RP, Mueser KT, Wallace CJ, Jacobs HE, Eckman T, & Massel HK (1986). Training skills in the psychiatrically disabled: learning coping and competence. Schizophrenia Bulletin, 12(4), 631–47. [DOI] [PubMed] [Google Scholar]
  26. López SR, Nelson HK, Polo AJ, Jenkins JH, Karno M, Vaughn C, & Snyder KS (2004). Ethnicity, expressed emotion, attributions, and course of schizophrenia: family warmth matters. Journal of Abnormal Psychology, 113(3), 428–39. [DOI] [PubMed] [Google Scholar]
  27. Morgades-Bamba CI, Fuster-Ruizdeapodaca Ma José, & Molero F (2019). The impact of internalized stigma on the well-being of people with schizophrenia. Psychiatry Research, 271, 621–627. 10.1016/j.psychres.2018.12.060 [DOI] [PubMed] [Google Scholar]
  28. Ridgway P (2001). Restorying psychiatric disability: learning from first person recovery narratives. Psychiatric Rehabilitation Journal, 24(4), 335–343. 10.1037/h0095071 [DOI] [PubMed] [Google Scholar]
  29. Santos MM, Kopelowicz A, & López SR. (2018). Recovery from schizophrenia: the case of mexican-origin consumers and family caregivers. The Journal of Nervous and Mental Disease, 206(6), 439–445. 10.1097/NMD.0000000000000826 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Schafer JL, & Graham JW (2002). Missing data: our view of the state of the art. Psychological Methods, 7(2), 147–77. [PubMed] [Google Scholar]
  31. Snowden LR (2007). Explaining mental health treatment disparities: ethnic and cultural differences in family involvement. Culture, Medicine and Psychiatry: An International Journal of Comparative Cross-Cultural Research, 31(3), 389–402. 10.1007/s11013-007-9057-z [DOI] [PubMed] [Google Scholar]
  32. Sprong M, Schothorst P, Vos E, Hox J, & Van Engeland H (2007). Theory of mind in schizophrenia: meta-analysis. The British Journal of Psychiatry, 191(1), 5–13. [DOI] [PubMed] [Google Scholar]
  33. Strauss JS, & Carpenter WTJ (1977). Prediction of outcome in schizophrenia. III. Five year outcome and its predictors. Archives of General Psychiatry, 34(2), 159–63. [DOI] [PubMed] [Google Scholar]
  34. Tajfel HE (1978). Differentiation between social groups: Studies in the social psychology of intergroup relations. Academic Press. [Google Scholar]
  35. Tajfel H, & Turner JC (1979). An Integrative Theory of Intergroup Conflict. In Austin WG & Worchel S (Eds.), The Social Psychology of Intergroup Relations (33-48). Monterey, CA: Wadsworth. [Google Scholar]
  36. Turner JC (1975). Social comparison and social identity: some prospects for intergroup behaviour. European Journal of Social Psychology, 5(1), 1–34. 10.1002/ejsp.2420050102 [DOI] [Google Scholar]
  37. Turner JC, Hogg MA, Oakes PJ, Reicher SD, & Wetherell MS (1987). Rediscovering the social group: A self-categorization theory. Basil Blackwell, Cambridge, MA. [Google Scholar]
  38. Usall J, Haro JM, Araya S, Moreno B, Muñoz PE, Martínez A, Salvador L, 2007. Social functioning in schizophrenia: what is the influence of gender? The European Journal of Psychiatry, 21(3), 199–205. [Google Scholar]
  39. U.S. Substance Abuse and Mental Health Services Administration (2020). Recovery support tools and resoures. Downloaded from internet November 13, 2020. https://www.samhsa.gov/brss-tacs/recovery-support-tools-resources [Google Scholar]
  40. Weiner B (1993). On sin versus sickness. a theory of perceived responsibility and social motivation. The American Psychologist, 48(9), 957–65. [DOI] [PubMed] [Google Scholar]
  41. Weisman AG, Nuechterlein KH, Goldstein MJ, & Snyder KS (1998). Expressed emotion, attributions, and schizophrenia symptom dimensions. Journal of Abnormal Psychology, 107(2), 355–359. [DOI] [PubMed] [Google Scholar]
  42. West SG, Finch JF, & Curran PJ (1995). Structural equation models with nonnormal variables: Problems and remedies. In Hoyle RH (Ed.), Structural equation modeling: Concepts, issues and applications (pp. 56–75). Newbury Park, CA: Sage. [Google Scholar]

RESOURCES