European Psychiatry If you feel you can’t, you won’t: the role of www.cambridge.org/epa subjective and objective cognitive competence on psychosocial functioning in depression Muriel Vicent-Gil1 , Joan Trujols1 , Maria Serra-Blasco2 , Research Article 3,4,5 1 Guillem Navarra-Ventura , Dolors Puigdemont , Carlo Alemany1 , Cite this article: Vicent-Gil M, Trujols J, Serra- 1 1 6 Blasco M, Navarra-Ventura G, Puigdemont D, Sara Crivillés , Javier de Diego-Adeliño , Narcís Cardoner and Alemany C, Crivillés S, de Diego-Adeliño J, 1 Cardoner N, Portella MJ (2023). If you feel you Maria J. Portella can’t, you won’t: the role of subjective and 1 objective cognitive competence on Sant Pau Mental Health Research Group, Institut d’Investigació Biomèdica Sant Pau (IIB-Sant Pau), Hospital de la Santa psychosocial functioning in depression. Creu i Sant Pau, Centro de Investigación Biomédica en Red de Salud Mental, Instituto de Salud Carlos III (CIBERSAM- European Psychiatry, 66(1), e83, 1–10 ISCIII), Universitat Autònoma de Barcelona (UAB), Barcelona, Spain; 2eHealth ICOnnecta’t Program and Psycho- https://doi.org/10.1192/j.eurpsy.2023.2454 Oncology Service, Institut Català d’Oncologia, L’Hospitalet de Llobregat, Spain; 3Institut d’Investigació Sanitària Illes Balears (IdISBa), Hospital Universitari Son Espases, Palma, Spain; 4Institut Universitari d’Investigació en Ciències de la Received: 04 May 2023 Salut (IUNICS), Universitat de les Illes Balears, Palma, Spain; 5Centro de Investigación Biomédica en Red de Enfermedades Revised: 27 July 2023 Respiratorias, Instituto de Salud Carlos III (CIBERES-ISCIII), Madrid, Spain and 6Department of Psychiatry and Legal Accepted: 21 August 2023 Medicine, Institute of Neuroscience, Universitat Autònoma de Barcelona (UAB) Keywords: cognition; depression; executive function; Abstract psychosocial functioning; self-appraisal Background. The purpose of this exploratory study is to examine the role of sociodemographic, Corresponding author: clinical, and cognitive – both objective and subjective – factors in overall and in specific domains Narcís Cardoner; of psychosocial functioning, in patients with depression at different clinical states of the disease Email: ncardoner@santpau.cat (remitted and non-remitted). Methods. A sample of 325 patients with major depressive disorder, 117 in remission and 208 in non-remission, were assessed with a semi-structured interview collecting sociodemographic, clinical, cognitive (with neuropsychological tests and the Perceived Deficit Questionnaire), and functional (Functioning Assessment Short Test) characteristics. Backward regression models were conducted to determine associations of global and specific areas of functioning with independent factors, for both clinical states. Results. Residual depressive symptomatology and self-appraisal of executive competence were significantly associated with psychosocial functioning in remitted patients, in overall and some subdomains of functioning, particularly cognitive and interpersonal areas. While depressive symptoms, executive deficits and self-appraisal of executive function were significantly related to functional outcomes in non-remitted patients, both in overall functioning and in most of subdomains. Discussion. This study evidences the strong association of one’s appraisal of executive compe- tence with psychosocial functioning, together with depressive symptoms, both in remitted and non-remitted patients with depression. Therefore, to achieve full recovery, clinical management of patients should tackle not only the relief of core depressive symptoms, but also the cognitive ones, both those that are objectified with neuropsychological tests and those that are reported by the patients themselves. Introduction Depression has always been associated with difficulties in daily functioning [1, 2]. According to the classification manual for mental disorders DSM-5, one of the diagnostic criteria is that depressive symptomatology causes clinically significant distress and/or impairment in social, © The Author(s), 2023. Published by Cambridge work-related, or other important areas of one’s life. In this regard, there is scientific evidence University Press on behalf of the European upon the presence of psychosocial dysfunction in the acute phase of the illness [3, 4]. Different Psychiatric Association. This is an Open Access studies have shown how work performance, and family and social relationships are altered in article, distributed under the terms of the depressed patients [5]. Contrary to expectations however, these functional difficulties also remain Creative Commons Attribution licence (http:// creativecommons.org/licenses/by/4.0), which in stages of clinical improvement. Some studies have shown that patients with residual depressive permits unrestricted re-use, distribution and symptoms may have the same functional impairment as patients in the severe phase of the reproduction, provided the original article is disease. Therefore, psychosocial functioning remains impaired in periods of clinical remission properly cited. [3, 6, 7]. To this point, the scientific research carried out in this field has focused on understanding which factors might be related to the functional impairment of patients suffering from depres- sion. Clinical characteristics such as the stage of the illness (i.e., being in non-remission or in remission), showing residual depressive symptomatology, having a higher number of depressive https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press 2 Vicent-Gil et al. episodes throughout the course of the illness, as well as an earlier written informed consent after a comprehensive explanation age of illness onset, have been associated with a poorer psychosocial of the study. functioning [8–11]. Similarly, cognitive dysfunction seems to play a key role in functional impairment. On one side, some studies link Sociodemographic and clinical assessment objective difficulties in the cognitive domains of attention, verbal memory, and executive function with poorer functional outcomes A semi-structured interview designed ad hoc was used to collect [6, 12, 13]. On the other side, very few studies have taken into sociodemographic and clinical information from all participants, account patients’ appraisal of their own cognitive performance, and including age, sex, years of schooling, estimated IQ, age at illness this little evidence only relates subjective measures of inattention to onset, number of depressive episodes, and depressive symptom- poorer work performance [9, 14]. However, it has not been studied atology. Clinical variables were double-checked with clinical rec- whether these factors related to psychosocial functioning vary ords and corrected if discrepant. Estimated IQ was assessed with the according to clinical stages, and this is indeed a matter of interest Vocabulary subtest of the Wechsler Adult Intelligence Scale for clinical research as it allows the development of much more version-IV (WAIS-IV; Wechsler, 2008). Hamilton Depression Rat- targeted interventions with the aim of achieving patients’ full ing Scale (HDRS-17) [19, 20] was used to evaluate depressive recovery [4, 15]. symptoms. Patients with a HDRS-17 score of 7 or less were con- Another issue to take into consideration is that in clinical sidered to be in full clinical remission (remitted patients) and practice not all depressed patients show the same pattern of func- patients with scores of 8 or higher were classified as non-remitted tional impairment, meaning that patients could present a global (non-remitted patients). impairment in daily activities, or impairment in one or some specific areas of psychosocial functioning. This functional hetero- Cognitive assessment geneity has been little addressed in previous research [3, 5, 6], as most studies use total scores on questionnaires and scales as proxies Subjective cognitive function of functional outcome [10]. In order to take current treatments a The patients’ appraisal of their own cognitive functioning, referred step further, it may be necessary to know which functional areas are as subjective cognition, was assessed by means of the Perceived particularly affected in each patient. Realizing whether there are Deficit Questionnaire (PDQ-20) [21–23]. The PDQ-20 is a 20-item difficulties at work, whether social activities are diminished, or self-report measure of cognitive dysfunction that assesses the most whether social life cannot be enjoyed is crucial in the development affected cognitive domains: attention, retrospective memory, pro- of interventions tailored at each of the clinical stages of the disorder. spective memory, and planning and organization. Each item is Not only it is relevant to know what is affected, but also which rated from 0 (never) to 5 (almost always); so higher scores indicate factors may determine these functional differences in patients. For a higher perception of suffering from cognitive deficits. instance, cognitive symptoms have been widely associated with poorer psychosocial functioning in general [16, 17], but little evi- Objective cognitive function dence has been gathered on specific cognitive problems within Objective cognition was measured with a battery of neuropsycho- specific psychosocial subdomains [18]. In this regard, both subject- logical tests covering the domains of attention, verbal memory, and ive cognitive complaints and objective neuropsychological per- executive function, to assemble the same cognitive domains formance may account for possible differences in specific included in the PDQ. Attention was measured using the Forward functional difficulties. Digit Span subtest (WAIS-IV) [24] and the Trail Making Test Part To address the issue of functional heterogeneity in depression, A (TMT-A) [25]. The Rey Auditory Verbal Learning Test [26] was the aim of the present exploratory cross-sectional study is to used to assess Verbal Memory, specifically the first trial, immediate examine whether sociodemographic, clinical, and cognitive (object- recall (sum of trials 1–5), and delayed recall subscales. Finally, ive and subjective) characteristics play a role in overall functioning executive function was tested with the Trail Making Test Part B and in the different areas of psychosocial functioning, in different (TMT-B) [25], the phonemic verbal fluency PMR test, adapted for clinical states of the disease (remitted and non-remitted patients). Spanish speaking population [27, 28], the number of categories from the Wisconsin Card Sorting Test [29], and the Digit Symbol Substitution subtest (WAIS-IV) [24]. Materials and methods Participants Functional assessment A total of 325 patients with a diagnosis of major depressive The psychosocial functioning was evaluated through the Function- disorder (MDD; DSM-5 criteria) aged between 18 and 65 were ing Assessment Short Test (FAST) [30, 31]. This scale was designed recruited from the psychiatry departments of the Hospital de to detect functional difficulties in patients with mental disorders, la Santa Creu i Sant Pau (Barcelona, Catalonia, Northeast specifically for bipolar disorder patients. FAST consists of 24 items Spain) and of the Hospital Universitari Parc Taulí (Sabadell, that assess six specific areas of psychosocial functioning: Catalonia, Northeast Spain). Patients were excluded if they (i) Autonomy refers to the patient’s ability to decide and carry met any of the following exclusion criteria: (i) an estimated out daily activities, (ii) Occupational functioning is related to the intelligence quotient (IQ) lower than 85, (ii) any neurological capacity to be employed as well as to maintain an adequate work disease or medical condition presenting cognitive deficits, and performance, (iii) Cognitive functioning refers to the ability to (iii) a history of substance abuse disorder except nicotine. The concentrate, to learn and memorize information, and to solve study followed the ethical principles of the Declaration of problems, (iv) Financial issues are related to the ability to manage Helsinki and the principles of Good Clinical Practice, com- their own money properly, (v) Interpersonal relationships refers to plied with current data protection law, and approved by the the rapport with family, partners, and friends, involvement in social ethics committee of each hospital. All participants gave activities, having satisfactory sexual relationships, and the ability to https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press European Psychiatry 3 defend one’s own interests, and (vi) Leisure time refers to doing functioning, non-remitted patients indicated greater functional sport or exercise and maintaining hobbies. All items are scored difficulties, not only in the overall functioning, but also in each of from 0 to 4, with higher scores suggesting greater functional dis- the different functional subdomains. In terms of objective cogni- ability. A previous study had determined that FAST total score from tive performance, significant differences were observed between 12 to 20 indicated a mild impairment in some areas of psychosocial the two study groups, with higher scores shown in the remitted functioning and scores from 21 to higher represented moderate to patients group. In subjective cognition, difference was observed in severe impairment in most areas of functioning [31]. the self-perceived executive functioning, showing a greater per- ception of deficit in those non-remitted patients. Statistical analyses Statistical analyses were carried out by means of the Statistical Bivariate relationships of FAST total and subdomains scores Package for Social Sciences, version 26. This study used the raw with sociodemographic, clinical, and cognitive characteristics scores of the FAST total score as well as the different scores of the Pearson’s correlation coefficients and p-values are presented in different functional subdomains that comprise this scale: auton- Table 2 for both, remitted and non-remitted patients. Those sig- omy, occupational, cognitive functioning, financial issues, inter- nificant associations were to be included in the regression models. personal relationships, and leisure time. Three measures of the PDQ were also used, including in the study PDQ attention, PDQ memory (retrospective plus prospective), and PDQ executive func- Backward regression models predicting psychosocial tion. To reduce the number of neuropsychological variables and functioning following the recommendations to address cognitive impairment The PDQ Executive was a significant independent factor in all the [32], three cognitive domains (Attention, Verbal Memory, and regression models run in the sample of remitted patients, except in Executive Function) were defined by averaging T-scores of different the FAST Occupational model. HDRS-17 was found to be signifi- neuropsychological tests mentioned above (by summing scores and cant in the regression models of FAST total score, FAST Occupa- dividing by the number of included tests). tional, and FAST Interpersonal, while the Executive Function was A descriptive analysis of sociodemographic, clinical, functional, only significant for the FAST Occupational model (data are shown and cognitive measures was performed using absolute and relative in Table 3). frequency measures for categorical variables and calculating mean In non-remitted patients, the significant factors were PDQ and standard deviation statistics for continuous variables. Differ- Executive and HDRS-17 in all of the regression models (except ences between remitted and non-remitted patients were assessed by HDRS-17 in the FAST Leisure model). Executive function was means of t-tests for continuous variables and chi-square tests for observed in most regression models; however, it reached statistical categorical variables. A series of multiple linear regression analyses significance only in specific cases, such as the FAST total score, were carried out to identify the independent factors associated with FAST Cognitive, and FAST Financial models. The FAST Occupa- the FAST total and subdomains scores in remitted patients and in tional model also included PDQ Memory as a significant factor, as non-remitted patients, separately. To perform these linear regres- well as verbal memory in the FAST Leisure model and being female sion analyses, Pearson correlations were used to examine bivariate in the FAST Cognitive model (data are shown in Table 4). A relationships between FAST scores and the following variables: age, graphical representation of variable contributions to regression sex, years of schooling, estimated IQ, age at illness onset, depressive effects can be found in Figure 1. episodes, HDRS-17, Attention, Verbal Memory, Executive Func- tion, PDQ Attention, PDQ Memory, and PDQ Executive. Variables associated with each FAST assessment at an α level of 0.05 in the Discussion bivariate analyses were considered candidate predictors of the corresponding FAST assessment and incorporated into the subse- This study identifies the factors that are independently associated quent linear regression analysis, for each subsample separately. with the distinct facets of psychosocial functioning at different Multiple linear regression analyses were performed with a back- stages of major depression. Depressive symptoms, executive diffi- ward procedure. The estimated minimum sample size to detect a culties, and subjective executive complaints appear to be the sig- significant difference greater than or equal to 7 units [33], and a nificant factors of functional outcomes in non-remitted patients, common standard deviation of 12, was 38 remitted subjects and while only residual depressive symptoms and self-appraisal of 77 non-remitted subjects, accepting an alpha risk of 0.05 and a beta executive competence are significantly associated with psychosocial risk of 0.2 in a two-sided test. functioning in remitted patients. Interestingly, these factors are common predictors of the different subdomains of FAST in both non-remitted and remitted patients with MDD. Considering that in Results the last decade, the treatment goal of depression has evolved from exclusively trying to reduce depressive symptomatology to improve Participant characteristics and differences between remitted patients’ functioning [4, 15], identifying which clinical and cogni- and non-remitted patients tive variables are related to this functional recovery may be an Sociodemographic, clinical, functional, and cognitive character- essential aim in the intervention design. istics of the patients at clinical remission (n = 117) and in non- One clear fact that emerges from the present data is the clinical remission (n = 208) are shown in Table 1. In detail, non-remitted value of the patient’s appraisal of their own cognitive functioning. patients showed fewer years of schooling as well as a lower The use of subjective measures of cognition in clinical practice has estimated IQ. Likewise, non-remitted patients had a higher degree been proposed as part of the assessment of patients’ cognitive of depressive symptomatology than remitted patients as clinically complaints. Yet studies attempting to correlate these subjective expected due to grouping variable. At the level of psychosocial measures with neuropsychological tests results have not been https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press 4 Vicent-Gil et al. Table 1. Sociodemographic, clinical, functional, and cognitive characteristics between remitted and non-remitted patients Statistics Remitted patients (n = 117) Non-remitted patients (n = 208) t or χ 2 p Age, years 51.56 (9.73) 52.55 (7.69) 1.03 .312 Sex, female (n, %) 76 (65) 151 (72.6) 2.08 .150 Years of schooling, n 13.36 (3.88) 10.5 (3.65) 43.96 <.001 Estimated IQ, T score 55.16 (8.82) 49.75 (7.93) 31.95 <.001 Age at illness onset, years 37.39 (12.85) 40.03 (11.47) 3.62 .058 Depressive episodes, n 2.57 (1.52) 2.54 (1.53) 0.028 .868 HDRS-17, total score 4.04 (2.07) 17.96 (6.54) 500.37 <.001 FAST, total score 15.74 (9.67) 39.27 (14.25) 253.34 <.001 FAST autonomy 1.73 (1.99) 5.27 (3.33) 110.42 <.001 FAST occupational 4.16 (4.62) 11.1 (4.97) 153.08 <.001 FAST cognitive 4.36 (2.9) 9 (3.78) 132.97 <.001 FAST financial 0.43 (0.85) 1.53 (1.83) 37.89 <.001 FAST interpersonal 3.65 (2.91) 8.11 (4.03) 110.82 <.001 FAST leisure 1.79 (1.55) 3.85 (2.07) 88.11 <.001 Attention, T score 48.39 (7.48) 41.65 (7.79) 57.27 <.001 Verbal memory, T score 45.34 (8.63) 40.23 (9.63) 22.54 <.001 Executive function, T score 48.96 (6.15) 42.87 (7.37) 53.19 <.001 PDQ, total score 23.22 (13.63) 42.18 (13.94) 140.7 <.001 PDQ attention 10.07 (4.37) 10.8 (4.72) 1.89 .170 PDQ memory 14.22 (7.79) 15.59 (8.59) 2.02 .156 PDQ executive 5.95 (4.11) 11.82 (4.04) 156.14 <.001 Abbreviations: FAST, Functioning Assessment Short Test; HDRS-17, Hamilton Depression Rating Scale; IQ, intelligence quotient; p, significance level; PDQ, Perceived Deficit Questionnaire. Note: Values represent mean scores (standard deviation) unless otherwise specified. entirely successful as there is a significant discrepancy between to psychosocial functioning itself, whatever the depressive symp- those measures [34–36]. These discordances have given rise to tomatology is. In contrast, items related to attention or memory do doubts about their use in clinical routine. One possible explanation not appear to exert a direct impact on patients’ functioning. This for this lack of agreement may be the fact that subjective and could be explained by the fact that these items describe cognitive objective cognitive measures do not assess the same capabilities. aspects more closely aligned with the constructs assessed by neuro- In other words, the neuropsychological test is done in a controlled psychological tests. Another possibility is that the PDQ scale might context and without distracting stimuli, whereas the subjective not have been developed entirely from the patient’s perspective and assessments reflect everyday situations in which there are a multi- it may not be identifying the most common attentional and mem- tude of stimuli, as well as interaction with other people, that can ory difficulties in patients with depression [40, 41]. Nevertheless, influence their performance. Therefore, it is reasonable to think more research is needed to understand what role self-perceived about a possible direct relationship between subjective measures of cognitive performance plays in everyday life. cognition with the daily functioning as observed in the present It is important to bear in mind that not only subjective cognitive study. However, there are not as many studies that relate subjective appraisal but also the objective cognitive impairment has a relevant measures of cognition to psychosocial functioning in depression, role in determining daily functioning of patients with MDD. There and even fewer which are specific as to which subjective cognitive is some empirical evidence that cognitive impairment (such as domain determines this relationship. Regardless of the few that deficits in attention and processing speed, executive function, and exist, overall perceived cognitive impairment is known to be asso- verbal learning) mediates functional outcomes in depression [6, ciated with worse functional outcomes [34, 37–39] as it is also seen 42–44]. In the current study, in non-remitted patients, objective in the current findings. These results highlight the relevance of cognition measures remain clearly associated with worse psycho- counting with what patients believe or think about their cognitive social functioning, along with subjective cognition and clinical competence in clinical practice, as these beliefs or thoughts may severity. Specifically, objective executive function is the only cog- determine what they do to regain an effective level of functioning. nitive domain that is associated with an overall psychosocial func- According to the current results, the perception of cognitive deficits tioning, and with the different areas of functioning (with the is limited to self-appraised difficulties in executive function, with no exception of FAST leisure), in agreement with previous research apparent association with attention or memory. This finding sug- [6, 42, 43]. gests that PDQ questions directed at one’s perception of organiza- According to foregoing scientific literature, executive function- tion, planning, and problem-solving skills are more directly related ing also plays a role in the degree of functional impairment of https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press European Psychiatry Table 2. Pearson’s correlation coefficients (p-values) between FAST total and FAST subdomains and possible explanatory factors Age at Years of Estimated illness Depressive Verbal Executive PDQ PDQ PDQ a Age Sex schooling IQ onset episodes HDRS-17 Attention memory function attention memory executive REMITTED PATIENTS .166 .117 .035 .084 .132 .277 .411 .107 .162 .191 .029 .118 .565 FAST TOTAL (.074) (.207) (.711) (.371) (.155) (.003) (<.001) (.254) (.083) (.043) (.753) (.206) (<.001) .077 .007 .081 .042 .012 .015 .206 .024 .003 .002 .018 .028 .411 FAST autonomy (.407) (.939) (.385) (.655) (.898) (.871) (.026) (.797) (.977) (.984) (.847) (.761) (<.001) .205 .036 .139 .049 .072 .182 .360 .066 .196 .212 .032 .132 .200 FAST occupational (.026) (.697) (.135) (.605) (.441) (.049) (<.001) (.481) (.035) (.024) (.734) (.155) (.030) .086 .203 .061 .086 .160 .210 .277 .034 .094 .086 .117 .090 .500 FAST cognitive (.356) (.028) (.511) (.359) (.085) (.023) (.002) (.719) (.314) (.366) (.208) (.335) (<.001) .114 .116 .073 .039 .114 .042 .111 .109 .084 .112 .126 .083 .343 FAST financial (.221) (.212) (.434) (.679) (.219) (.650) (.232) (.245) (.371) (.236) (.175) (.375) (<.001) .092 .029 .149 .139 .038 .151 .332 .053 .123 .032 .037 .096 .329 FAST interpersonal (.325) (.759) (.110) (.136) (.684) (.104) (<.001) (.573) (.189) (.738) (.689) (.303) (<.001) .078 .020 .009 .133 .033 .144 .238 .069 .058 .101 .020 .049 .459 FAST leisure (.406) (.827) (.926) (.155) (.725) (.121) (.010) (.463) (.538) (.289) (.831) (.601) (<.001) NON-REMITTED PATIENTS .040 .016 .202 .170 .027 .093 .574 .279 .333 . 337 .055 .016 .544 FAST TOTAL (.564) (.823) (.003) (.014) (.698) (.183) (<.001) (<.001) (<.001) (<.001) (.427) (.822) (<.001) .106 .063 .168 .068 .082 .040 .451 .184 .249 .245 .085 .019 .473 FAST autonomy (.129) (.367) (.015) (.331) (.236) (.568) (<.001) (.008) (<.001) (.001) (.221) (.782) (<.001) .046 .142 .097 .083 .004 .016 .332 .146 .174 .201 .067 .144 .270 FAST occupational (.507) (.040) (.165) (.234) (.958) (.819) (<.001) (.035) (.012) (.008) (.339) (.038) (<.001) .088 .147 .189 .190 .031 .075 .506 .213 .312 .256 .047 .055 .526 FAST cognitive (.208) (.034) (.006) (.006) (.659) (.282) (<.001) (.002) (<.001) (.001) (.499) (.429) (<.001) .080 .017 .155 .159 .036 .057 .375 .233 .173 .227 .136 .066 .389 FAST financial (.252) (.809) (.026) (.022) (.602) (.416) (<.001) (.001) (.013) (.003) (.051) (.343) (<.001) .093 .006 .151 .059 .020 .103 .524 .217 .198 .257 .084 .081 .431 FAST interpersonal (.184) (.929) (.030) (.397) (.772) (.138) (<.001) (.002) (.004) (.001) (.229) (.244) (<.001) .034 .1 .144 .136 .035 .028 .381 .264 .220 .230 .071 .117 .437 FAST leisure (.627) (.149) (.039) (.051) (.615) (.683) (<.001) (<.001) (.001) (.002) (.311) (.091) (<.001) Abbreviations: FAST, Functioning Assessment Short Test; HDRS-17, Hamilton Depression Rating Scale; IQ, intelligence quotient; PDQ, Perceived Deficit Questionnaire. a Strickly speaking, the correlation coefficients appearing in this column correspond to point–biserial correlation values (the point–biserial correlation is, however, mathematically equivalent to the Pearson correlation). Variables in bold were considered candidate predictors and were incorporated in the subsequent linear regression analysis 5 6 Vicent-Gil et al. Table 3. Backward regression models to assess the relationship between possible explanatory factors and functioning measures remitted patients Outcome measures Regressors Statistics F df p R2 FAST total Depressive episodes β = .80, p = .092, RW = .041 19.25 4, 108 <.001 .427 HDRS-17 β = 1.01, p = .004, RW = .105 Executive function β= .21, p = .073, RW = .026 PDQ executive β = 1.09, p < .001, RW = .254 FAST autonomy HDRS-17 23.41 1, 115 <.001 .169 PDQ executive β = .2, p < .001, RW = .169 FAST occupational Age β = .08, p = .051, RW = .034 8.27 3, 109 <.001 .195 Depressive episodes HDRS-17 β = .64, p = .002, RW = .113 Verbal memory Executive function β= .17, p = .010, RW = .048 PDQ executive FAST cognitive Sex 21.07 2, 114 <.001 .269 Depressive episodes HDRS-17 β = .21, p = .079, RW = .048 PDQ executive β = .32, p < .001, RW = .221 FAST financial PDQ executive β = .07, p < .001, RW = .117 15.29 1, 115 <.001 .117 FAST interpersonal HDRS-17 β = .37, p = .004, RW = .086 11.73 2, 114 <.001 .171 PDQ executive β = .18, p = .005, RW = .084 FAST leisure HDRS-17 30.64 1, 115 <.001 .210 PDQ executive β = .17, p < .001, RW = .210 Abbreviations: β, beta; df, degrees of freedom; F, F test; FAST, Functional Assessment Short Test; HDRS-17, Hamilton Depression Rating Scale; p, significance level; PDQ, Perceived Deficit Questionnaire; R2, variance explained; RW, raw relative weight. Note: The regressors of the final model are marked in bold with their corresponding parameter estimates. patients in clinical remission, and it determines functioning in outcomes, but also which factors are related to each of the different specific functional areas such as in occupational functioning, in areas that constitute a person’s overall functioning. In fact, it is subjective cognition, and in leisure time [6]. However, the results of observed that depressive symptomatology, executive function, and the present study contradict such previous findings since the per- the patient’s perception of memory and executive functioning ceived cognitive functioning seems to be the only cognitive deter- influence occupational functioning in non-remitted patients, minant in full clinical remission. A possible explanation could be whereas when the patient is in clinical remission, only residual that remitted patients showed better executive functioning when depressive symptomatology and the objective measure of executive explored with objective tests, and thus no relation could be found functioning are determinants. These results underline the need of with different psychosocial subdomains. In this same line, two executive skills to function properly at work at all stages of the previous studies have observed a stronger correlation between disease. In other words, achieving good executive performance subjective cognition to psychosocial functioning rather than with seems to be of utmost importance to reach optimal occupational objective cognitive variables [34, 37], and they have also confirmed functioning. the importance of subjective cognitive performance in the restor- The present findings suggest that in order to achieve a full ation of daily activities once the depressive episode is resolved. It is functional recovery, psychotherapeutic interventions should not relevant to note that different characteristics such as demographics only aim at improving psychosocial functioning through cognitive related to sex and socioeconomic status together with the symptom enhancement, but it should also take into account the patient’s severity across study samples, or the cognitive domains used in perception of their cognitive functioning, especially in those stages previous research, as well as the grouping based on patients’ cog- of greater clinical stability in which the greatest predictor of psy- nitive profiles (ranging from global impairment to mild or no chosocial functioning will be the patient’s own perception of their impairment), might have interfered with the understanding of executive abilities. In clinical practice, it is observed that there is a the relationship between cognition (either objective or subjective) profile of patients who, despite being in clinical remission, have a and psychosocial functioning [16, 45–48]. Being female seemed to significant number of complaints about not having recovered their play a discrete role in FAST cognitive results, indicating that gender previous level of functioning, as if certain executive skills have not perspective should be taken into consideration in future studies that been recovered and make it difficult for patients to carry out day-to- address cognitive, psychosocial, and clinical outcomes altogether. day activities. Consequently, it may be necessary to develop strat- It must be pointed out that the present study has not only egies to better manage their self-perception, and not only interven- assessed which variables are the most related to functional tions directed at cognitive training. https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press European Psychiatry 7 Table 4. Backward regression models to assess the relationship between possible explanatory factors and functioning measures in non-remitted patients Outcome measures Regressors Statistics F df p R2 FAST total Years of schooling 53.13 3, 169 <.001 .521 Estimated IQ HDRS-17 β = .88, p < .001, RW = .229 Attention Verbal memory Executive function β= .36, p = .001, RW = .071 PDQ executive β = 1.34, p < .001, RW = .221 FAST autonomy Years of schooling 28.48 3, 169 <.001 .336 HDRS-17 β = .16, p < .001, RW = .139 Attention Verbal memory Executive function β= .06, p = .053, RW = .035 PDQ executive β = .30, p < .001, RW = .173 FAST occupational Sex 8.95 4, 168 <.001 .171 HDRS-17 β = .2, p = .001, RW = .077 Attention Verbal memory Executive function β= .81, p = .094, RW = .026 PDQ memory β = .09, p = .034, RW = .051 PDQ executive β = .19, p = .037, RW = .017 FAST cognitive Sex β = 1.47, p = .003, RW = .021 25.03 5, 167 <.001 .463 Years of schooling Estimated IQ HDRS-17 β = .21, p < .001, RW = .179 Attention β = .08, p = .069, RW = .013 Verbal memory Executive function β= .13, p = .006, RW = .04 PDQ executive β = .38, p < .001, RW = .210 FAST financial Years of schooling 14.18 3, 169 <.001 .201 Estimated IQ HDRS-17 β = .05, p = .006, RW = .095 Attention Verbal memory Executive function β= .03, p = .04, RW = .032 PDQ executive β = .13, p < .001, RW = .116 FAST interpersonal Years of schooling 28.73 3, 169 <.001 .338 HDRS-17 β = .26, p < .001, RW = .204 Attention Verbal memory Executive function β= .07, p = .053, RW = .038 PDQ executive β = .28, p < .001, RW = .132 FAST leisure Years of schooling 28.77 2, 170 <.001 .209 HDRS-17 Attention Continued https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press 8 Vicent-Gil et al. Table 4. Continued Outcome measures Regressors Statistics F df p R2 Verbal memory β= .04, p = .004, RW = .033 Executive function PDQ executive β = .21, p < .001, RW = .176 Abbreviations: β, beta; df, degrees of freedom; F, F test; FAST, Functional Assessment Short Test; HDRS-17, Hamilton Depression Rating Scale; p, significance level; PDQ, Perceived Deficit Questionnaire; R2, variance explained; RW, raw relative weight. Note: The regressors of the final model are marked in bold with their corresponding parameter estimates. Figure 1. Relative importance of sociodemographic, clinical, and cognitive characteristics for predicting psychosocial functioning (FAST scores). Nonetheless, there are some limitations worth mentioning that stages of the disease (remission vs. non-remission) are being might affect the understanding of the observed results. First, some considered, and unfortunately not all clinical scenarios are being clinical variables as anxiety, level of self-efficacy, personality traits, represented in terms of the psychosocial functioning of patients. or the depressive symptomatology itself might influence the Due to the design of the study is not possible to analyse every patient’s perceived cognition [21], and in turn, these variables depression stage as well as the potential variables that could be might impact on patients’ psychosocial functioning. It is possible influencing the outcomes. that the fact that the self-appraisal in organization and planning In conclusion, this study provides evidence on the relevance of skills that remain significant in clinical remission may be due to patients’ self-awareness of their cognitive functioning in carrying the presence of those psychological characteristics. Second, the out daily activities, from work performance to interpersonal rela- subjective cognition measurement scale itself (PDQ) was not tionships. These findings have significant implications for the initially developed for depression and, although it was subse- clinical management of patients MDD. It highlights the importance quently adapted and validated for this population [23], it may of incorporating cognitive tests (both subjective and objective) and not fully represent the most common cognitive difficulties in psychosocial functioning assessments into routine clinical practice. depressed patients, especially in the domains of attention and These aspects are often overlooked but are crucial for a compre- memory. Third, the psychosocial functioning scale (FAST) was hensive understanding of each individual’s condition. By integrat- created for bipolar disorder and has not been validated for the ing these assessments, clinicians can identify key factors that may depressive population, so it may not reflect the functional diffi- hinder treatment adherence, impede finding the most suitable culties encountered by patients with depression. Fourth, the FAST treatment for each patient, and ultimately prevent some individuals scale is only validated as a total outcome measure, and not from achieving full recovery. Consequently, it is important to designed to consider the different functional areas that comprise continue researching on this issue to develop interventions aimed it. However, previous studies [6] have used the different func- at achieving full recovery [49], which are based not only on redu- tional subdomains as part of the statistical analysis because of the cing depressive symptomatology, but also on improving objective great clinical interest in knowing exactly what is affected in cognition, specially including patients’ appraisal of their own cog- depressed patients, in order to target interventions to these par- nitive functioning. ticular difficulties. Fifth, given the lack of an observable measure of psychosocial functioning, the current findings may have been Data availability statement. The data that support the findings of this study influenced by the subjective appraisal itself. Finally, only two are available from the corresponding author upon reasonable request. https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press European Psychiatry 9 Acknowledgments. We thank the patients who participated in this study. We study. Neuropsychiatr Dis Treat. 2020;16:589–95. https://doi.org/10.2147/ also thank the staff of the Department of Psychiatry of the Hospital de la Santa NDT.S224754. Creu i Sant Pau and of the Hospital Parc Taulí for their kind collaboration in [14] Buist-Bouwman MA, Ormel J, de Graaf R, de Jonge P, van Sonderen E, patient recruitment. Alonso J, et al. Mediators of the association between depression and role functioning. Acta Psychiatr Scand. 2008;118:451–8. https://doi.org/ Financial support. This research was financed by the Spanish Ministry of 10.1111/j.1600-0447.2008.01285.x. Health Instituto de Salud Carlos III (Grant No. PI17/00056), by the Health [15] Lam RW, Filteau MJ, Milev R. Clinical effectiveness: the importance of Research and Innovation Strategic Plan (PERIS) (Grant No. SLT006/17/00177) psychosocial functioning outcomes. J Affect Disord. 2011;132:S9. https:// of the Generalitat de Catalunya, and supported by the CIBERSAM annual doi.org/10.1016/j.jad.2011.03.046. budget of G21 and by the Department of University of Generalitat de Catalunya [16] Lam RW, Kennedy SH, McIntyre RS, Khullar A. Cognitive dysfunction in (Grant No. 2017 SGR 1343) and by Lundbeck España. major depressive disorder: effects on psychosocial functioning and impli- cations for treatment. Can J Psychiatry. 2014;59:649–54. Competing interest. The authors declare no competing interests. [17] Chakrabarty T, Hadjipavlou G, Lam RW. Cognitive Dysfunction in Major Depressive Disorder: assessment, impact, and management. J Life- long Learn Psychiatry. 2016;14:194–206. https://doi.org/10.1176/appi. References focus.20150043. [18] McIntyre RS, Cha DS, Soczynska JK, Woldeyohannes HO, Gallaugher LA, [1] Cabello M, Mellor-Marsá B, Sabariego C, Cieza A, Bickenbach J, Ayuso- Kudlow P, et al. Cognitive deficits and functional outcomes in major Mateos JL. Psychosocial features of depression: a systematic literature depressive disorder: determinants, substrates, and treatment interven- review. J Affect Disord. 2012;141:22–33. https://doi.org/10.1016/ tions. Depress Anxiety. 2013;30:515–27. https://doi.org/10.1002/da.22063. j.jad.2011.12.009. [19] Bobes J, Bulbena A, Luque A, Dal-Ré R, Ballesteros J, Ibarra N. Evaluación [2] Malhi GS, Mann JJ. Depression. Lancet. 2018;392:2299–312. https://doi. psicométrica comparativa de las versiones en español de 6, 17 y 21 ítems de org/10.1016/S0140-6736(18)31948-2. la escala de valoración de Hamilton para la evaluación de la depresión. [3] Godard J, Baruch P, Grondin S, Lafleur MF. Psychosocial and neurocog- Med Clin. 2003;120:693–700. https://doi.org/10.1157/13047695. nitive functioning in unipolar and bipolar depression: a 12-month pro- [20] Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry. spective study. Psychiatry Res. 2012;196:145–53. https://doi.org/10.1016/ 1960;23:56–62. https://doi.org/10.1136/jnnp.23.1.56. j.psychres.2011.09.013. [21] Strober LB, Binder A, Nikelshpur OM, Chiaravalloti N, DeLuca J. The [4] Fiorillo A, Carpiniello B, De Giorgi S, La Pia S, Maina G, Sampogna G, et al. perceived deficits questionnaire: perception, deficit, or distress? Int J MS Assessment and management of cognitive and psychosocial dysfunctions Care. 2016;18:183–90. https://doi.org/10.7224/1537-2073.2015-028. in patients with major depressive disorder: a clinical review. Front [22] Sullivan M, Edgley K, DeHousx E. A survey of multiple sclerosis, part 1: Psychiatry. 2018;9:1–8. https://doi.org/10.3389/fpsyt.2018.00493. perceived cognitive problems and compensatory strategy used. Can J [5] Mehta S, Mittal PK, Swami MK. Psychosocial functioning in depressive Rehabil. 1990;4:99–105. patients: a comparative study between major depressive disorder and [23] Lam RW, Lamy FX, Danchenko N, Yarlas A, White MK, Rive B, et al. bipolar affective disorder. Depress Res Treat. 2014;2014:302741. https:// Psychometric validation of the perceived deficits questionnaire- doi.org/10.1155/2014/302741. depression (PDQ-D) instrument in US and UK respondents with major [6] Knight MJ, Air T, Baune BT. The role of cognitive impairment in psycho- depressive disorder. Neuropsychiatr Dis Treat. 2018;14:2861–77. https:// social functioning in remitted depression. J Affect Disord. 2018;235:129– doi.org/10.2147/NDT.S175188. 34. https://doi.org/10.1016/j.jad.2018.04.051. [24] Wechsler D. Weschler Adult Intelligence Scale (WAIS-IV). 4th ed. San [7] Romera I, Pérez V, Menchón JM, Polavieja P, Gilaberte I. Optimal cutoff Antonio, TX: Pearson; 2008. point of the Hamilton Rating Scale for Depression according to normal [25] Tombaugh TN. Trail Making Test A and B: normative data stratified by levels of social and occupational functioning. Psychiatry Res. 2011;186: age and education. Arch Clin Neuropsychol. 2004;19:203–14. https://doi. 133–7. https://doi.org/10.1016/j.psychres.2010.06.023. org/10.1016/S0887-6177(03)00039-8. [8] Romera I, Perez V, Menchón JM, Delgado-Cohen H, Polavieja P, Gilaberte [26] Rey A. L’examen clinique en psychologie. Paris: Presses Universitaires de I. Social and occupational functioning impairment in patients in partial France; 1941. versus complete remission of a major depressive disorder episode. a six- [27] Casals-Coll M, Sánchez-Benavides G, Quintana M, Manero RM, Rognoni T, month prospective epidemiological study. Eur Psychiatry. 2010;25:58–65. Calvo L, et al. Estudios normativos españoles en población adulta joven https://doi.org/10.1016/j.eurpsy.2009.02.007. (proyecto NEURONORMA jóvenes): Normas para los test de fluencia verbal. [9] McIntyre RS, Soczynska JZ, Woldeyohannes HO, Alsuwaidan MT, Cha Neurologia. 2013;28:33–40. https://doi.org/10.1016/j.nrl.2012.02.010. DS, Carvalho AF, et al. The impact of cognitive impairment on perceived [28] Pena-Casanova J, Quinones-Ubeda S, Gramunt-Fombuena N, Quintana- workforce performance: results from the International Mood Disorders Aparicio M, Aguilar M, Badenes D, et al. Spanish Multicenter Normative Collaborative Project. Compr Psychiatry. 2015;56:279–82. https://doi.org/ Studies (NEURONORMA Project): norms for verbal fluency tests. Arch Clin 10.1016/j.comppsych.2014.08.051. Neuropsychol 2009;24:395–411. https://doi.org/10.1093/arclin/acp042. [10] Sheehan D V., Nakagome K, Asami Y, Pappadopulos EA, Boucher M. [29] Heaton RK. Wisconsin Card Sorting Test manual. Odessa: Psychological Restoring function in major depressive disorder: a systematic review. J Assessment Resources; 1981. Affect Disord. 2017;215:299–313. https://doi.org/10.1016/ [30] Rosa AR, Sánchez-Moreno J, Martínez-Aran A, Salamero M, Torrent C, j.jad.2017.02.029. Reinares M, et al. Validity and reliability of the Functioning Assessment [11] Wilson S, Hicks BM, Foster KT, McGue M, Iacono WG. Age of onset and Short Test (FAST) in bipolar disorder. Clin Pract Epidemiol Ment Heal. course of major depressive disorder: associations with psychosocial func- 2007;3:1–10. https://doi.org/10.1186/1745-0179-3-Received. tioning outcomes in adulthood. Psychol Med. 2015;45:505–14. https:// [31] Bonnín CM, Martínez-Arán A, Reinares M, Valentí M, Solé B, Jiménez E, doi.org/10.1017/S0033291714001640. et al. Thresholds for severity, remission and recovery using the Function- ing Assessment Short Test (FAST) in bipolar disorder. J Affect Disord. [12] Cambridge OR, Knight MJ, Mills N, Baune BT. The clinical relationship 2018;240:57–62. https://doi.org/10.1016/j.jad.2018.07.045. between cognitive impairment and psychosocial functioning in major [32] Miskowiak K, Burdick K, Martinez-Aran A, Bonnin C, Bowie C, Carvalho depressive disorder: a systematic review. Psychiatry Res. 2018;269:157– A, et al. Methodological recommendations for cognition trials in bipolar 71. https://doi.org/10.1016/j.psychres.2018.08.033. disorder by the International Society for Bipolar Disorders Targeting [13] Castellano S, Torrent C, Petralia MC, Godos J, Cantarella RA, Ventimiglia Cognition Task Force. Bipolar Disord. 2017;19:614–26. https://doi.org/ A, et al. Clinical and neurocognitive predictors of functional outcome in 10.1111/bdi.12534. depressed patients with partial response to treatment: one year follow-up https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press 10 Vicent-Gil et al. [33] Torrent C, Del Mar Bonnin C, Martínez-Arán A, Valle J, Amann BL, [41] Trujols J, Duran‐Sindreu S, Portella MJ. Patient‐reported outcome meas- González-Pinto A, et al. Efficacy of functional remediation in bipolar ures (PROMs) to guide clinical care: recommendations and challenges. disorder: a multicenter randomized controlled study. Am J Psychiatry. Med J Aust. 2022;217:110. https://doi.org/10.5694/mja2.51614. 2013;170:852–9. https://doi.org/10.1176/appi.ajp.2012.12070971. [42] Kiosses DN, Alexopoulos GS. IADL functions, cognitive deficits, and [34] Potvin S, Charbonneau G, Juster RP, Purdon S, Tourjman SV. Self-evaluation severity of depression: a preliminary study. Am J Geriatr Psychiatry. and objective assessment of cognition in major depression and attention 2005;13:244–9. https://doi.org/10.1097/00019442-200503000-00010. deficit disorder: implications for clinical practice. Compr Psychiatry. 2016; [43] Baune BT, Air T. Clinical, functional, and biological correlates of cognitive 70:53–64. https://doi.org/10.1016/j.comppsych.2016.06.004. dimensions in major depressive disorder – rationale, design, and charac- [35] Srisurapanont M, Suttajit S, Eurviriyanukul K, Varnado P. Discrepancy teristics of the cognitive function and mood study (CoFaM-Study). Front between objective and subjective cognition in adults with major depressive Psychiatry. 2016;7:1–9. https://doi.org/10.3389/fpsyt.2016.00150. disorder. Sci Rep. 2017;7:1–7. https://doi.org/10.1038/s41598-017-04353-w. [44] Woo YS, Rosenblat JD, Kakar R, Bahk WM, McIntyre RS. Cognitive [36] Serra-Blasco M, Torres IJ, Vicent-Gil M, Goldberg X, Navarra-Ventura G, deficits as a mediator of poor occupational function in remitted major Aguilar E, et al. Discrepancy between objective and subjective cognition in depressive disorder patients. Clin Psychopharmacol Neurosci. 2016;14: major depressive disorder. Eur Neuropsychopharmacol. 2019;29:46–56. 1–16. https://doi.org/10.9758/cpn.2016.14.1.1. https://doi.org/10.1016/j.euroneuro.2018.11.1104. [45] Evans VC, Iverson GL, Yatham LN, Lam RW. The relationship between [37] Sumiyoshi T, Watanabe K, Noto S, Sakamoto S, Moriguchi Y, Tan KHX, neurocognitive and psychosocial functioning in major depressive disorder. J et al. Relationship of cognitive impairment with depressive symptoms and Clin Psychiatry. 2014;75:1359–70. https://doi.org/10.4088/JCP.13r08939. psychosocial function in patients with major depressive disorder: cross- [46] Pu S, Noda T, Setoyama S, Nakagome K. Empirical evidence for discrete sectional analysis of baseline data from PERFORM-J. J Affect Disord. neurocognitive subgroups in patients with non-psychotic major depres- 2019;258:172–8. https://doi.org/10.1016/j.jad.2019.07.064. sive disorder: clinical implications. Psychol Med. 2018;48:2717–29. [38] Wang G, Si TM, Li L, Fang Y, Wang CX, Wang LN, et al. Cognitive https://doi.org/10.1017/S003329171800034X. symptoms in major depressive disorder: associations with clinical and [47] Vicent-Gil M, Keymer-Gausset A, Serra-Blasco M, Carceller-Sindreu M, functional outcomes in a 6-month, non-interventional, prospective study de Diego-Adeliño J, Trujols J, et al. Cognitive predictors of illness course at in China. Neuropsychiatr Dis Treat. 2019;15:1723–36. https://doi.org/ 12 months after first-episode of depression. Eur Neuropsychopharmacol. 10.2147/NDT.S195505. 2018;28:529–37. https://doi.org/10.1016/j.euroneuro.2018.02.001. [39] Zhang BH, Feng L, Feng Y, Xin LM, Zhu XY, Tan YL, et al. The effect [48] Vicent-Gil M, Portella MJ, Serra-Blasco M, Navarra-Ventura G, Crivillés of cognitive impairment on the prognosis of major depressive disorder. S, Aguilar E, et al. Dealing with heterogeneity of cognitive dysfunction in J Nerv Ment Dis. 2020;208:683–8. https://doi.org/10.1097/NMD. acute depression: a clustering approach. Psychol Med. 2021;51:2886–94. 0000000000001180. https://doi.org/10.1017/S0033291720001567. [40] Trujols J, Portella MJ, Iraurgi I, Campins MJ, Siñol N, de Los Cobos JP. [49] IsHak WW, Bonifay W, Collison K, Reid M, Youssef H, Parisi T, et al. The Patient-reported outcome measures: are they patient-generated, patient- recovery index: a novel approach to measuring recovery and predicting centred or patient-valued? J Ment Heal 2013;22:555–62. https://doi.org/ remission in major depressive disorder. J Affect Disord. 2017;208:369–74. 10.3109/09638237.2012.734653. https://doi.org/10.1016/j.jad.2016.08.081. https://doi.org/10.1192/j.eurpsy.2023.2454 Published online by Cambridge University Press