Functioning assessment short test (FAST): validity and reliability in adults with Autism Spectrum Disorder

A sample of participants with ASD was recruited from the monitored population within the Comprehensive Care Program for Autism Spectrum Disorder (PAITEA), from the Vall d’Hebron Hospital’s Psychiatry Department. All participants were diagnosed using the Autism Diagnostic Observational Schedule-2 [28] by trained clinicians (L.GG and J.L.M). Inclusion criteria included: 1) being aged between 18 and 65 years; 2) diagnosis of ASD according to DSM-5 diagnostic criteria [2]. Respondents’ language and cognitive barriers and ASD levels 2 and 3 were considered as exclusion criteria [2]. The nonclinical control group (NC) was recruited at the Barcelona Clinic Hospital’s Bipolar and Depressive Disorders Unit [29]. NC inclusion criteria were (1) age between 18 and 65 years, (2) no current or past psychiatric history as collected from medical records, and (3) non-reported history of family psychiatric disorder.

After informing the participants and obtaining their written consent, the investigator recorded demographic (age and gender) and clinical variables (psychiatric diagnosis history, if any) and administered the FAST scale [24]. The FAST is an interviewer-administered instrument, designed to be used by a trained clinician. The evaluation time-window refers to the last two weeks. The areas evaluated with the FAST are six, including: 1) Autonomy, related to the ability of living on their own, taking care of themselves (physical appearance, hygiene …); 2) Occupational functioning, capacity to get and keep a paid job and being efficient at it, working in the field in which the patient was educated, and earning according to the level of his/her position; 3) Cognitive functioning, being able to concentrate while reading or watching a movie, solve problems, remember simple data and learn new information; 4) Financial issues, being capable to manage their own finances; 5) Interpersonal relationships, refers to the capacity of keeping friendships, getting well with family, involvement in social activities, sexual relations, and being able to express and defend own interests; 6) Leisure Time, capacity to exercise and enjoy a hobby. Items (24 in total) are rated on a 4-point scale, ranging from 0 (no difficulty) to 3 (severe difficulty). All scores are added to obtain a global punctuation for which high scores indicate poorer functioning [24].

Additionally, the Sheehan’s Disability Scale (SDS) [23] was also administered. It is a three-item instrument widely used in clinical settings, which measures the severity of disability in three inter-related domains: work, family life/home responsibilities, and social/leisure activities. The patient rates the extent to which responsibilities are impaired by his or her symptoms on a 10-point visual analog scale. Clinicians are recommended to pay special attention to patients who score 5 or higher on any of the three scales because such high scores are associated with significant functional impairment. The SDS has shown good validity and reliability in Spanish clinical population [30].

To study the normality of the FAST variables, Shapiro–Wilk coefficient was applied. The results showed significant deviations from normality, precluding the use of parametric tests. The reliability coefficient (Cronbach’s alpha) was used to examine the internal consistency of the FAST items in each domain and the full scale. Cronbach’s alpha measures were considered minimally acceptable if α = .65, acceptable if α = .7, and optimal if α = .8 [31]. Spearman’s correlation coefficient was calculated to examine the correlations between the FAST and the SDS scores. To explore intergroup differences between ASD versus controls and between younger ASD adults (18–25 years) versus older ASD adults (= > 26 years), nonparametric tests (U Mann-Whitney test) for each FAST domain were conducted. A confirmatory factor analysis (CFA) with the original six-factor structure of the FAST was studied. When evaluating the fit of the CFA and stability models to the data, the root mean square error of approximation (RMSEA) and comparative fit index (CFI) were used. RMSEA values below .06 indicate a good fit to the data and CFI values close to .95 are acceptable [32]. The optimal cut-off point of the measure was analyzed by the area under the ROC curve (AUC). The AUC values were considered minimally acceptable when lower than .7, acceptable when between .7 and .8, optimal when between .8 and .9, and excellent when above .9 [33]. Data analyses were carried out with the statistical package IBM SPSS 19.0 and AMOS 26.0 for Windows. The alpha level was set at p < .05.

Cronbach’s alpha showed excellent value for the FAST total score (α = .97) in the total sample of participants, and also for the ASD group (α = .91), whereas acceptable values were found for the NC group (α = .76). Table 1 shows internal consistency results for each FAST subscale.

Spearman’s correlations analyses were conducted to study convergence between the FAST total score with the SDS scores. Results showed significant small to moderate correlations with each of three SDS domains: Work (r = .54, p < .001), Social Life (r = .27, p = .02), and Family Life (r = .31, p = .01). FAST Occupational Functioning also showed a moderate correlation with SDS Work domain (r = .61, p < .001); FAST Interpersonal Relationships correlated with the SDS Social domain (r = .46, p < .001); and the FAST Family Life item (Item 20) correlated with the SDS Family Life domain (r = .68; p < .001).

Table 2 shows the results of the intergroup mean differences analysis. The ASD group showed significantly higher scores in the FAST total score (Z = − 10.23, p < 0.001), and in all the subscales when compared to NC participants. When compared by age, younger ASD adults showed significantly lower scores in the FAST Autonomy subscale (Z = − 2.31, p = .02), whereas older ASD adults showed significantly lower scores in the FAST Interpersonal Relationships subscale (Z = − 2.27, p = .02). No significant differences in the FAST total score were found between age groups (Table 3).

The study of the internal structure of the FAST determined a poor model fit to the original six-factor structure (Fig. 1). The RMSEA value was .104, (90% CI [.94, .11]; p = .00), which is below the acceptable threshold, whereas the CFI value was .90, indicating poor model fit.

The AUC was .98, 95% CI [0.96, 1] which, being close to 1, indicates excellent discriminant capacity. A total FAST score above 12 reached best sensitivity (96%) and specificity (94%) scores. Positive and negative predictive values were .96 and .94, respectively. Figure 2 shows the ROC curve for the FAST total score comparing the ASD and NC groups.

The findings of this study should be interpreted with caution, considering some limitations that have been identified. First, given that patients and controls were not matched by gender, the number of males in the ASD group is much higher than in the NC group. This is due to a higher prevalence of ASD in male population [41]. Further studies should address gender differences in the adaptive functioning of adults with ASD levels. Also, in relation to age, the older ASD group comprised a wide range of ages (26–65 years). Considering the great variability that can occur throughout the different stages of adult life (middle adult vs. older adult), future studies should explore these differences with greater accuracy. Finally, the FAST might not be appropriate for ASD patients of Levels 2 and 3 due to the content of the items examined. Moreover, differences in cognitive domains (e.g., executive function), as well as adaptive functioning in our sample could not be reported as many of the participants lacked this information. Future research could be conducted in order to adapt some items regarding these variables and thus making the FAST more suitable for the entire ASD spectrum. Among others, items like support needs, communication skills, and cognitive and behavioral difficulties could be taken into account.