Testing the Influence of “Social Support Centered Model” Vs. “Symptom Centered Model” On Functioning in Individuals with Schizophrenia

Background: Social support and psychiatric symptoms are widely identified in the literature as having a significant impact on treatment outcomes and quality of life for people with schizophrenia (Rudnuck & Kravez, 2001). Social support is associated with increased functioning and recovery, while the absence of social support is related to increased psychiatric symptoms and reduced potential for full community integration (McCorkle, 2007). Symptoms related to schizophrenia impact an individual's functioning in family relationships, social relations, work, and independence. However, how symptoms and social support act together to influence functional outcomes has not been studied.

Purpose: The purpose of the study was to test competing models to explore the impact of symptomatology and social support on functional outcomes for individuals with schizophrenia. Our first model “Social Support Centered model” hypothesizes that symptoms will have an indirect effect on functioning through social support, thereby making social support a central target for intervention. Our second model, “Symptom Centered model”, hypothesizes that social support will have an indirect effect on functioning, through symptoms, thereby making symptom control and management a central target for intervention.

Methods: The sample consisted of 166 ethnically diverse individuals diagnosed with schizophrenia who were recruited upon admission to four highly similar community-based psychosocial rehabilitation interventions as part of an NIMH funded protocol. The following measures were used. The Brief Psychiatric Rating Scale-Extended Version (BPRS-E) is a measure of psychiatric symptoms. The Role Functioning Scale (RFS) measures functional outcome in four domains: family, independent living, social, and work. The Medical Outcomes Study (MOS) Social Support Survey is a measure of social support. A series of path analyses were conducted to test the two competing hypotheses across the four outcomes domains.

Results: The path analyses found that the Social Support Centered model showed better fit to the data than the Symptom Centered model for the two social functional outcome variables: family functioning (χ²=.26, df=1 p =.610; RMSEA=.000; CFI =1.000, AIC = 16.260 and social functioning (χ²=0.08, df=1, p=.781; RMSEA=.000; CFI=1.000, AIC =16.08). However, the Symptom Centered model showed better fit to the data for the two instrumental functional outcomes: work functioning (χ²=.118, df=1, p=.731; RMSEA=.000; CFI=1.000, AIC =16.11) and independent living (χ²=.05, df=1, p =.824; RMSEA=.000; CFI=1.000, AIC =16.05). The incremental chi-square test was also used to assess differences in fit across models.

Conclusions/Implications: Our first finding implies that the impact of symptoms on social functional outcomes is largely indirect and through their impact on social support. This indicates a need for interventions and research that focus most centrally on social support, and secondarily on symptoms, in order to improve family and social functioning. Our second finding suggests that the Symptom Centered model is more applicable to the instrumental functioning outcomes. This implies that interventions and research should focus most centrally on symptom improvement, and secondarily on social support, in order to show impact on work and independent living outcomes. Overall, the findings suggest that interventions need to be differentially focused on symptoms or social support depending on the targeted outcome.