Is Personal Recovery Distinct for Young Adults?: (Re)Conceptualizing Dimensions of the Recovery Assessment Scale with Confirmatory Factor Analysis

Background and Purpose: Young adulthood is a period where serious mental illnesses (SMI) commonly emerge yet engagement in treatment remains low. The adult mental health system has expanded its focus from clinical recovery (symptom reduction) to personal recovery. The Recovery Assessment Scale (RAS) was developed using adult narratives of personal recovery and posits five underlying dimensions: 1) personal confidence and hope; 2) willingness to seek help; 3) goal and success orientation; 4) reliance on others; and 5) no domination by symptoms. This study evaluates the relevance and generalizability of these original five-factors as applied to young adults with SMI, a developmentally distinct subpopulation. It then considers the implications for (re)conceptualizing and measuring unique dimensions of young adult personal recovery.

Methods: The sample includes 121 young adults with SMI involved in an NIMH-funded study. We used the 20-item RAS with response options on a 5-point Likert scale. A confirmatory factor analysis (CFA) was conducted using structural equation modeling (SEM) in Mplus using robust maximum likelihood estimation. Missing data were treated using full information maximum likelihood. The five-factor structure was evaluated using sample-size adjusted (Swain correction) indices of global fit (e.g., RMSEA, CFI) and localized ill-fit (modification indices). Composite reliabilities of each subscale were calculated using the hierarchical omega method.

Results: The sample is 66% male; mean age is 26.7 (SD=3.9), and 94% identify as Black, Latinx, or Multiracial. The five-factor model demonstrated poor model fit (χ²=258.3 [DF=160], p<0.001; RMSEA=0.07; CFI=0.87). Modification indices revealed three theoretically meaningful correlated errors that were sequentially included in a modified model. While the modified five-factor model yielded improved fit (χ²=207.5 [DF=157], p<0.01; RMSEA=0.05; CFI=0.93), the correlation between factors one (personal confidence and hope) and three (goal and success orientation) was problematically high (0.94) and one item had a low factor loading (<0.30). From a developmental lens, there was a theoretical justification for evaluating a modified four-factor model. The four-factor model yielded good fit to the data (χ²[DF=143]=182.02, p<0.05; RMSEA=0.047; CFI=0.94). All items loaded onto their respective factor with most being above 0.70 (range: 0.45-0.84, p < 0.001) and the correlations between the factors ranged from 0.28-0.59 (p<0.05). The four subscales demonstrated acceptable composite reliabilities (omegas: 0.64-0.92).

Conclusions and Implications: Findings support a four-factor structure of the RAS with a combined factor (personal confidence and hope/goal and success orientation) among our sample of young adults. Although the relationship between these factors may be distinct among middle-aged adults (population for whom the RAS was initially developed), they may be more intertwined for young adults with SMI. For example, a young adult’s personal confidence and hope may be more closely associated with their future goals, compared to middle-aged adults whose work and relationships may be more stable and thus more distinct from their sense of confidence and hope. Findings may relate to developmental processes in young adulthood such as solidifying one’s identity, independence, employment/education goals, and autonomy. More research is needed to explore factors that may influence personal confidence and hope and how they relate to future goals.