Table 2.
Studies on depression/anxiety as independent variables and QoL outcomes.
| First Author (Year) | Disorder or Symptoms Analyzed; QoL Domains Analyzed | Research Question Regarding QoL | Methods | Results |
|---|---|---|---|---|
| Årdal (2013) [45] | Controls and patients in the acute phase of recurrent MD and FU (DSM-IV, HDRS); SF-36 (physical functioning, role physical, vitality, bodily pain, mental health, role emotional, social functioning, general health, as well as summary scores PCS, MCS and total score) | (a) Whether QoL scores differ between MD patients and healthy comparisons across domains over time. (b) Whether QoL in patients with recurrent MDD differed between acute phase and recovery. |
(a) ANOVA (b) Paired-sample t-tests |
(a) There was a significant interaction effect between time, QoL domain and group, indicating that QoL scores differed between MD patients and controls over time. Compared to the healthy control group, the MDD group had reduced QoL in all domains at BL and reduced QoL in several domains at FU (significant for general health, social, emotional role, mental health, PCS, MCS and total score). (b) In the MD group, QoL scores significantly improved during recovery from recurrent MDD in most domains (significant for physical functioning, physical role, vitality, social functioning, role emotional, mental health, PCS, MCS and total score). |
| Buist-Bouwman (2004) [46] | Onset, acute phase and subsequent remission from MDE (CIDI); comorbid anxiety disorder (CIDI); SF-36 (physical functioning, physical role, vitality, pain, psychological health, psychological role, social functioning and general health) | (a) Whether incident MDE and recovery from MDE are associated with changes in QoL and whether pre- and post-morbid QoL scores in the MD group differ from the comparison group without MD. (b) In the subgroup with worse QoL after MDE: whether the severity of depression and number of depressive episodes were associated with worse QoL. (c) Whether comorbid anxiety during MDE is associated with reduced QoL (i.e., lower QoL after MDE compared to before MDE). |
(a)–(c) Multivariate logistic regression | (a) Incident MDE was associated with a drop in QoL (significant for vitality, psychological, psychological role and social functioning). Subsequent recovery from MDE was associated with an improvement in QoL (significant for physical role, vitality, psychological health, psychological role, social functioning and general health). Comparing pre- and post-morbid levels, QoL did not differ or was higher after MDE in some domains (significantly higher for psychological health and psychological role). Moreover, before MD onset, QoL was significantly lower compared to healthy controls in all domains. After remission from MDE, QoL scores in nearly all domains (not significant for psychological role) were significantly lower compared to healthy controls. (b) About 40% of the MDE group had worse QoL after recovery from MDE compared to pre-morbid levels. The severity of depression was associated with worse QoL only for the psychological health domain, but no other domains. The number of depressive episodes was not significantly associated with worsening QoL in any domain. (c) In the MDE cohort, comorbid anxiety was associated with a significant reduction in QoL (significant for physical role and psychological health). |
| Cabello (2014) [47] | Chronic MD (AUDADIS interview; summary score of the number of symptoms to identify severity); SF-12, “disability” (i.e., domain-specific reduced QoL, defined as score ≤ 25th percentile in the subscale; physical functioning, physical role, bodily pain, general health, vitality, social functioning, emotional role and mental health) | (a) Whether chronic MD is associated with the incidence/persistence of “disability” (i.e., reduced QoL) in a general population sample. (b) Whether the severity of depressive symptoms is associated with the incidence/persistence of “disability” (i.e., reduced QoL) in the MD subgroup of the sample. |
Both (a) and (b) Generalized Estimating Equations and logistic regressions | (a) In the general population, chronic MD was a significant risk factor for the persistence of disability (i.e., reduced QoL) in all domains and of the incidence of disability (i.e., reduced QoL) in all domains except for the physical role. (b) In the chronic MD subgroup, the severity of depressive symptoms was associated with the persistence of disability (i.e., reduced QoL) (significant for general health, social functioning, emotional role and mental health) and not significantly associated with the incidence of reduced QoL in any domain. |
| Cerne (2013) [48] | Number of depressive episodes over time according to CIDI; number of episodes of panic and other anxiety syndromes over time (PHQ); SF-12 (PCS, MCS) | Whether the pooled number of (a) depressive episodes over time, (b) panic and anxiety episodes over time are are associated with the pooled QoL over time. |
(a) and (b) Multivariate linear regression | (a) A higher number of depressive episodes over time was associated with lower pooled PCS and MCS. (b) a higher number of pooled panic episodes over time was associated with a lower mean MCS but not PCS. A higher number of pooled other anxiety syndrome episodes over time was not associated with the mean MCS or PCS. |
| Chin (2015) [49] | Depression according to PHQ-9 (>9), clinician’s diagnosis; SF-12v2 (PCS, MCS) | (a) Whether depressive symptoms and a clinician’s detection of depression at BL are associated with QoL at FU. (b) Whether a clinician’s detection of depression at BL is associated with a change in QoL. |
(a) Multivariable non-linear mixed-effects regression (b) Independent t-tests |
(a) Depressive symptoms and a clinician’s detection of depression at BL were not predictive of QoL at FU. (b) A clinician’s detection of depression at BL was related to change (improvement) in MCS, but not PCS over time in a primary care sample screened as positive for depression. |
| Chung (2012) [50] | Depression diagnosis and symptoms (DSM-IV, HRSD17 depression scale, HADS depression scale); anxiety symptoms (HRSD17 anxiety scale, HADS anxiety scale; SF-36 (physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional, mental health, PCS and MCS) | (a) Whether BL depressive symptoms are associated with QoL at FU. (b) Whether BL depressive symptoms or changes in depressive symptoms are associated with changes in QoL over time. (c) Whether BL anxiety symptoms are associated with QoL at FU. (d) Whether BL anxiety symptoms or changes in anxiety symptoms are associated with changes in QoL over time. |
(a)–(d) Hierarchical regression | (a) BL depressive symptoms were not associated with any QoL domain at FU. (b) BL depressive symptoms were not associated with changes in any QoL domain over time. Changes in depressive symptoms were significantly associated with changes in some QoL domains over time (significant for: general health, vitality, mental health and MCS). (c) BL anxiety symptoms were not associated with any QoL domain at FU. (d) BL anxiety symptoms were not associated with changes in any QoL domain over time. Changes in anxiety symptoms were significantly associated with changes in some QoL domains over time (significant for: bodily pain, general health and mental health). |
| Diehr (2006) [51] | Depression according to CIDI, CES-D (>16); QLDS, WHOQOL-Bref (environmental, physical, psychological and social), SF-12 (PCS, MCS) | (a) Whether the quartile of change in depressive symptoms is associated with changes in QoL. (b) Whether the remission of depression at FU is associated with changes in QoL. |
Regression | (a) No/little change in CES-D associated with changes in QoL over time (significant for SF-12 MCS). Every other quartile of change in depressive symptoms was significantly associated with changes in QoL in most scales/domains (significant for: QLDS, all domains of WHOQOL-Bref and SF-12 MCS), meaning a higher reduction in depressive symptoms was associated with a higher increase in QoL, and more severe depressive symptoms were associated with a reduction in QoL. (b) Remission of depression at FU was associated with improvement in all QoL measures and domains (SF-12, QLDS and WHOQOL-Bref). There was no significant change in QoL in those with persistent clinical depression at FU. |
| Hajek (2015) [21] | Depressive symptoms (GDS); EQ-VAS | Whether an initial change in depressive symptoms is associated with a subsequent change in QoL in the whole sample and by sex. | Vector autoregressive models | No significant association between an initial change in depression score and a subsequent change in QoL was found for the whole sample or stratified by sex. |
| Hasche (2010) [52] | Depression status at BL (according to DIS diagnosis and CES-D ≥ 9); SF-8 (PCS, MCS) | (a) Whether depression status groups at BL differed according to QoL at FU. (b) Whether depression status groups at BL differed according to QoL changes in score over time. |
(a) t-tests (b) Linear mixed effects regression models |
(a) At 6- and 12-month FU, those with and without depression at BL differed significantly in QoL scores, with the depression group reporting lower QoL at FUs (significant for MCS and PCS). (b) While depression at BL was significantly related to improvements in MCS (but not PCS) scores over time, those with depression still reported lower QoL compared to those without. |
| Heo (2008) [53] | Depression (BDI ≥ 10); SF-36 (decrease in total score over time) | Whether FU depression is associated with a reduction in QoL over time. | Binary logistic regression | Depression at FU was associated with a significant reduction in QoL total score over time. |
| Ho (2014) [54] | Depression (according to GDS ≥ 5); SF-12 (PCS, MCS) | Whether depression at BL is associated with QoL at FU. | Linear regression | BL depression was associated with lower QoL at FU (significant for MCS and PCS). |
| Hussain (2016) [55] | Depressive disorders (SCID, MINI); current PTSD, specific phobias, other anxiety disorders (SCID, MINI); WHOQOL-Bref (general QoL and hrqol) | (a) Whether current depressive disorders at BL predict QoL at FU. (b) Whether current PTSD, specific phobias and other anxiety disorders at BL predict QoL at FU. |
(a) and (b) Multiple linear regression | (a) Depressive disorders at BL predicted reduced QoL at FU (significant for general QoL and hrqol). (b) PTSD, but not specific phobias or other anxiety disorders, predicted reduced general QoL at FU. None of the anxiety disorders predicted hrqol at FU. |
| Joffe (2012) [56] | Lifetime history of depression (according to SCID); anxiety disorder (according to SCID); SF-36 (impaired QoL according to 25th percentile of SF-36; social functioning, role emotional, role physical, pain and vitality) | (a) Whether a lifetime history of depression is associated with impaired QoL during FU. (b) Whether a prior lifetime history of anxiety disorder (compared to no depression or anxiety) is associated with reduced QoL during FU. (c) Whether a lifetime history of comorbid depression and anxiety is associated with impaired QoL during FU. |
(a)–(c) Repeated measure multilevel regression | (a) A history of depression only was associated with reduced QoL during FU (significant for social functioning and pain). (b) Prior lifetime history of anxiety disorder was associated with reduced QoL (significant for physical role). (c) A history of comorbid anxiety and depression was associated with reduced QoL during FU (significant for social functioning, emotional role, physical role and pain). |
| Johansen (2007) [57] |
Level of PTSD symptoms according to IES-15; WHOQOL-Bref (physical health, psychological health, social relationships and environment) | Whether PTSD symptoms predict QoL at FU. | Structural equation model | More severe PTSD symptoms predicted QoL at FU (significant positive association between FU1 and FU2). |
| Kramer (2003) [58] | Current or lifetime depression/PTSD (according to Q-DIS); SF-36 (energy/fatigue, emotional role, general health, mental health, pain, physical functioning, physical role and social) | Whether QoL outcomes over time differed among the disorder groups. | Random/fixed effects model | There was no significant interaction between time and diagnostic group (no depression/PTSD, PTSD, depression and comorbid depression/PTSD) on QoL. Comparing the adjusted means for all three times among the disorder groups showed significant differences between the groups in most domains. In comparison, those with depression at BL reported reduced QoL over time in several domains compared to the PTSD group and the group without PTSD/depression. In comparison, those with PTSD only showed higher QoL compared to those with depression or comorbid depression/PTSD in several domains. |
| Kuehner (2009) [20] | Depressive symptoms (MADRS); WHOQOL (overall, physical, psychological, social and environmental) | Whether the lag in levels of depressive symptoms predicts future levels of QoL and whether the association differs by group (formerly depressed inpatients vs. community controls). | Time-lagged linear models | Higher depressive symptoms predict future lower QoL (significant for social). The association was not moderated by group status. |
| Kuehner (2012) [59] | Depression score (according to MADRS, FDD-DSM-IV); WHOQOL-Bref (physical, psychological, social and environment) | Whether the lag in depressive symptoms predicted QoL at FU. | Hierarchical, time-lagged linear models | Higher depressive symptoms significantly predicted lower QoL at FU (significant for physical and psychological). |
| Lenert (2000) [60] | Remission or persistent depression (according to DSM-III criteria, DIS); SF-12 (PCS, MCS) | Whether the remission of depression (compared to no remission) is associated with changes in QoL over time. | OLS regression | Remission of depression was associated with improved QoL (significant for MCS) at FU1 and FU2. |
| Mars (2015) [61] | Asymptomatic, mild and high symptoms of depression (according to SCAN); EQ-5D (without anxiety/depression item) | Whether depression symptom trajectories over time (asymptomatic, mild symptoms and chronic–high symptoms) are associated with QoL at FU. | Latent class growth analysis with distal outcome models | QoL at FU differed significantly among different depression symptom trajectories, with persons from the the chronic–high depressive symptom class showing lower QoL scores relative to the asymptomatic class. |
| Moutinho (2019) [62] | Depression at BL (according to DASS cut-off: 9); anxiety at BL (according to DASS anxiety scale cutoff: 7); WHOQOL-Bref at FU (physical, psychological, social and environment) | (a) Whether BL depression predicted QoL at FU. (b) Whether BL anxiety predicted QoL at FU. |
(a) and (b) Stepwise linear regression | (a) Depression at BL was significantly associated with reduced QoL at FU (significant for psychological functioning, social functioning and environmental). (b) Anxiety at BL was associated with reduced QoL at FU (significant for physical). |
| Ormel (1999) [63] | Depression at BL (according to CIDI); “disability” (i.e., reduced QoL according MOS SF 6-item physical functioning scale ≥ 2) | Whether depression at BL is associated with the onset of disability (i.e., reduced QoL) during FU. | Logistic regression models | Compared to the non-depressed group, people with depression at BL showed higher odds for the onset of disability (i.e., reduced QoL) during FU (significant for 12-month FU, but not 3-month FU). |
| Pan (2012) [64] | Depressive symptoms (CES-D); WHOQOL-Bref-TW (overall score, physical, psychological, social and environmental) | Whether depressive symptoms were associated with QoL over time. | Linear mixed-effects models | Higher depressive symptoms were associated with lower QoL in MDD patients (significant for overall score, physical, psychological, social and environmental). |
| Panagioti (2018) [65] | Depressive symptoms (MHI-5); WHOQOL-Bref (physical, psychological, environmental and social) | Whether depressive symptoms at BL are associated with changes in QoL over time. | Multivariate regression models | Higher depressive symptoms at BL were associated with a decline in QoL over time (significant for physical and psychological). |
| Pakpour (2018) [66] | Dental anxiety at BL (MDAS); PedsQL 4.0 general hrqol and oral hrqol scale at FU | Whether dental anxiety at BL predicted oral- and general-health-related QoL at FU. | Structural equation modeling | Dental anxiety at BL was no significant direct predictor of generic QoL at FU and was significantly associated with worse oral-health-related QoL at FU. |
| Pyne (1997) [67] | MD-diagnosis (SCID/SADS) and depressive symptoms (HAM-D); QWB | Whether group status over time (community controls, continuously non-depressed patients, incident depression patients and continuously depressed patients) is associated with changes in QoL. | Repeated measure analysis (ANOVA) | There was no significant interaction term between group status and time, indicating that changes in QoL did not differ between the groups. At both points in time, QoL differed significantly among all groups, except between the incident depression and continuous depression group. |
| Remmerswaal (2020) [68] | OCD course (SCID), Y-BOCS, BDI, BAI over time; EQ-5D over time | (a) Whether OCD symptom severity and QoL over time were associated. (b) Whether QoL over time differs between OCD course groups (chronic, intermittent and remitting) and general population norms. (c) Whether OCD symptom severity, anxiety and depressive symptoms over time are associated with changes in QoL over time in patients with OCD. |
(a) Pearson’s correlation (b)–(c) Linear mixed models |
(a) QoL over time and OCD symptom severity were significantly correlated. (b) The QoL of OCD patients was significantly lower compared to general population norms, except the QoL of the intermittent OCD group at FU1, where there was no significant difference compared to the general population. When comparing the OCD course groups, the chronic OCD group had a significantly lower QoL over time compared to the other groups. The remitting group had moderately improved until FU1 and a small QoL improvement between FU1 and FU2 relative to the chronic group. (c) In those with a remitting OCD, only more severe symptoms of comorbid anxiety and depressive symptoms, but not OCD symptom severity over time, were significantly associated with a lower QoL over time. |
| Rhebergen (2010) [69] | MD-/dysthymia-/DD diagnosis at BL and subsequent recovery at FU (according to CIDI); comorbid anxiety at BL (CIDI); SF-36 (physical health summary score) | Whether QoL trajectories over time differ between: (a) different depression status groups who achieved remission (MDD, dysthymia and double depression) and a comparison group without mental health disorders. (b) The different depression status groups. (c) Whether comorbid anxiety at BL in a sample recovering from depression is associated with changes in QoL. |
(a)–(c) Linear mixed models | (a) There was a significant interaction between group status and time. More specifically, compared to changes in QoL over time in people without a mental health diagnosis, QoL improved over time in those with MDD and DD, but not dysthymia. All depression diagnosis groups showed a significantly lower QoL compared to the no diagnosis group at all waves. (b) Considering the depression groups, only the interaction term between dysthymia and time until FU1 was significant. Those with dysthymia had a significantly lower QoL compared to those with MDD at FU1. This difference was not significant at FU2. (c) Comorbid anxiety disorder at BL in people who recovered from depression over time was not associated with a significant change in QoL over time. |
| Rubio (2014) [14] | First episode of incident MDD (AUDADIS-IV) at FU; incident GAD, social anxiety disorder, PD, specific phobia (AUDADIS-IV); SF-12 (MCS) | Whether incident MDD is associated with changes in QoL over time compared to: (a) people without history of MDD, (b) without history of any mental health disorder, (c) and whether the association differed by gender. Whether incident anxiety disorders are associated with changes in QoL over time: (d) compared to no history of the specific anxiety disorder, (e) compared to no history of any psychiatric disorder, (f) and whether the association differed by gender. |
Linear regression model | (a) Incidence of MDD (compared to no MDD) was associated with a significant decrease in QoL until FU. (b) Incidence of MDD (compared to no mental health disorder) was associated with a significant decrease in QoL until FU. (c) The association did not vary by gender. (d) Incidence of all anxiety disorders (with comorbid disorders; ref: no history of anxiety disorder) was associated with a significant decrease in QoL over time. (e) Incident anxiety disorders were not significantly associated with QoL when only considering “pure” anxiety without any comorbidities (ref: no history of any psychiatric disorder). (f) The association did not vary by gender. |
| Rubio (2013) [15] | Remission from MDD, dysthymia (AUDADIS-IV); Remission from GAD, PD, SAD, specific phobia (AUDADIS-IV); SF-12 (MCS) | Whether remission from depression (MDD, dysthymia) is associated with: (a) changes in QoL over time (compared to non-remitted cases), (b) QoL at FU (compared to people with no history of a specific depressive disorder), (c) QoL at FU, when only considering depressive disorders without any psychiatric comorbidity (compared to people without any lifetime psychiatric diagnosis). Whether remission from anxiety disorders are associated with: (d) changes in QoL over time (compared to non-remitted cases), (e) QoL at FU (compared to people with no history of a specific anxiety disorder), (f) QoL at FU, when only considering anxiety disorders without any psychiatric comorbidity (compared to people without any lifetime psychiatric diagnosis). |
(a)–(f) Linear regression models | (a) Remission from MD and dysthymia was associated with a significant positive change in QoL compared to non-remitted cases. (b) Remission of MD and dysthymia was associated with significantly lower QoL at FU compared to people without history of a specific diagnosis. (c) Remission of MD and dysthymia was associated with significantly lower QoL at FU compared to people without any lifetime psychiatric diagnosis. (d) Remission from SAD and GAD was associated with significant positive changes in QoL compared to non-remitted cases. (e) Remission of PD, SAD, specific phobia and GAD was associated with significantly lower QoL at FU compared to people without history of a specific diagnosis. (f) Remission of “pure” PD, SAD, specific phobias and GAD was associated with significantly lower QoL at FU compared to people without any lifetime psychiatric diagnosis. |
| Rozario (2006) [70] | Depressive symptoms (GDS); SF-12 (MCS and PCS) | Whether depressive symptom severity was associated with QoL change profiles over time (no change, declined and improved groups). | Multinomial logistic regression | There was no significant association between depressive symptom severity and QoL change score profiles at FU. |
| Sareen (2013) [71] | Depression trajectory groups over time (according to AUDADIS-IV); anxiety disorder trajectory groups over time (according to AUDADIS-IV); SF-12 (MCS and PCS) | (a) Whether depression trajectory groups (no past year disorder/no suicide attempt at FU, remission without treatment, persistent disorder/comorbidity/suicide attempt/treatment) differed according to QoL at FU. (b) Whether anxiety disorder trajectory groups (no past year disorder/no suicide attempt at FU, remission without treatment, persistent disorder/comorbidity/suicide attempt/treatment) differed according to QoL at FU. |
(a) and (b) Multiple linear regression models | (a) QoL at FU differed among the different depression trajectory groups (MCS was significant for all groups: no disorder > remitted disorder > persistent disorder; PCS: no disorder > remitted disorder; remitted disorder < persistent disorder). (b) QoL at FU differed among the different anxiety trajectory groups (MCS was significant for all groups: no disorder > remitted disorder > persistent disorder; PCS: no disorder > persistent disorder, remitted disorder > persistent disorder). |
| Shigemoto (2020) [72] | PTSD symptoms (PCL-C); Q-LES-Q (psychosocial and physical) | Whether previous PTSD symptoms are associated with QoL at FU. | Longitudinal structural equation model | Previous PTSD symptoms were associated with physical QoL at FU1, but not FU2 or psychosocial QoL at both FUs. |
| Siqveland (2015) [73] | Depressive symptoms (according to the depression scale from the GHQ-28); PTSD symptoms (PCL-S); WHOQOL-Bref (global and hrqol) | (a) Whether depressive symptoms at BL are associated with QoL at FU. (b) Whether PTSD symptoms at BL are associated with QoL at FU. |
(a) and (b) Multiple mixed effects regression analyses | (a) Higher depressive symptoms at BL were associated with reduced QoL at FU. (b) PTSD levels at BL were not significantly associated with reduced QoL at FU. |
| Spijker (2004) [74] | Depression status (CIDI); Comorbid anxiety (CIDI); SF-36 (social, role emotional) | (a) Whether depression status over time (non-depressed, recovered or depressed (including persistent, relapsing course)) is associated with QoL at FU. Whether comorbid anxiety is associated with QoL at FU (b) in a group with persistent depression and (c) in a group recovered from depression. |
ANOVA | (a) QoL at FU was significantly reduced in depressed samples compared to the non-depressed group, and lower in the persistently depressed compared to the recovered group (significant for: role emotional and social). Among the depressed subgroups, there was no significant difference between a persistent or a relapsing course regarding QoL at FU. (b) In the persistently depressed group, comorbid anxiety was significantly associated with reduced QoL at FU (significant for role emotional and social). (c) In those who recovered from depression, comorbid anxiety was significantly associated with reduced QoL (significant for role emotional). |
| Stegenga (2012) [75] | MDD status according to CIDI (remitted, intermittent and chronic); SF-12 (PCS and MCS) | Whether MDD course (remitted, intermittent and chronic) is associated with QoL over time. | Random coefficient analysis | While change in QoL over time did not differ between course groups, QoL at BL (MCS) was lower in those with a chronic course compared to those who remitted from BL. |
| Stegenga (2012) [76] | MDD (CIDI); anxiety syndromes (panic disorder and others, PHQ); SF-12 (PCS) | (a) Whether MDD at BL predicts change in QoL over time. (b) Whether anxiety syndrome at BL (compared to no psychiatric diagnosis) predict changes in QoL over time. (c) Whether comorbid anxiety and MDD at BL (compared to no psychiatric diagnosis) predict changes in QoL over time. |
(a)–(c) Random coefficient model | (a) While changes in QoL over time did not differ significantly between those with MDD at BL and those without any psychiatric diagnosis, QoL at BL was lower in those with depression. (b) While changes in QoL over time did not differ significantly between those with anxiety syndrome at BL and those without any psychiatric diagnosis, QoL at BL was lower in those with anxiety compared to those without any psychiatric diagnosis. (c) While changes in QoL over time did not differ significantly between those with comorbid anxiety and MDD at BL and those without any psychiatric diagnosis, QoL at BL was lower in those with comorbid anxiety and MDD compared to those without any psychiatric diagnosis. |
| Stevens (2020) [77] | Posttraumatic stress symptoms (VETR-PTSD); SF-36 (MCS, PCS, physical functioning, bodily pain, general health, role physical, role emotional, mental health, vitality and social functioning) | Whether PTSS at BL is associated with QoL at FU. | Generalized estimating equations | Higher BL PTSS was significantly associated with lower QoL (PCS and MCS) at FU. Using a Bonferroni-corrected alpha value, only the domains of mental health, vitality and social functioning at FU were significantly associated with BL PTSS symptoms. The interaction between time and PTSS at BL was not significant, indicating that PTSS had the same effect on QoL outcomes at both FUs. |
| Tsai (2007) [78] | Increased post-traumatic stress symptoms (DRPST); MOS SF-36 (physical functioning, role physical, pain, general health, vitality, social functioning, role emotional, mental health, PCS and MCS) | (a) Whether different PTSS trajectory groups over time (persistent PTSS, recovered, delayed and persistently healthy) differed in QoL at FU. (b) Whether increased post-traumatic stress symptoms at BL predicted QoL at FU. |
(a) ANOVA (b) Multiple regression models |
(a) At FU, those who were persistently healthy had the highest QoL scores (significantly higher compared to the persistent group in all domains; significantly higher than the recovered group for: pain, general health, vitality, mental health and MCS; significantly higher compared to delayed PTSS in all domains). In addition, those with delayed PTSS (significantly lower than the recovered group in all domains except physical functioning) and those with persistent PTSS (significantly lower than recovered group in all domains) had the lowest QoL overall. (b) Increased PTSS at BL was not significantly associated with QoL at FU. |
| Vulser (2018) [79] | Depressive symptom levels (CES-D score), depression status (CES-D ≥ 19); SF-12v2 (role emotional and social) | Whether depressive symptoms or depression status at BL are associated with QoL at FU. | Generalized linear models | Both the level of depressive symptoms at BL as well as depression status at BL were associated with QoL at FU (significant for: role emotional and social). |
| Wang (2000) [80] | Depressive symptoms (SCL-90 subscale); anxiety symptoms (SCL-90 subscale); WHOQOL-Bref (total) | (a) Whether depressive symptoms at BL were associated with QoL at FU. (b) Whether anxiety symptoms at BL were associated with QoL at FU. |
(a) and (b) Stepwise regression | (a) Higher depressive symptoms at BL were associated with reduced QoL at FU. (b) Anxiety symptoms BL were not included in the final stepwise regression model. |
| Wang (2017) [81] | Depressive disorder course groups (CIDI); anxiety disorder course (CIDI); SF-36 (MCS, PCS) | (a) Whether QoL at FU differs between three different course groups of depressive disorders (1. no disorder at BL and no suicide attempt until FU; 2. remitted without treatment; 3. persistent disorder/treatment/developed psychiatric co-morbidity/suicide attempt until FU). (b) Whether QoL at FU differs between three different course groups of anxiety disorders (1. no disorder at BL and no suicide attempt until FU; 2. remitted without treatment; 3. persistent disorder/treatment/developed psychiatric co-morbidity/suicide attempt until FU). |
(a) and (b) Multiple linear regression | (a) Those with depression at BL that remitted without treatment had lower QoL at FU (significant for MCS and PCS) than those without the disorder and higher QoL at FU (significant for MCS) than those with a persistent disorder. (b) Those with anxiety at BL that remitted without treatment over time had lower QoL at FU than those without the disorder and higher QoL (MCS, but not PCS) than those with a persistent disorder. |
| Wu (2015) [82] | Depressive symptoms according to CDI; social anxiety symptoms (SASC); QOLS | (a) Whether depressive symptoms at BL are associated with QoL at FU. (b) Whether social anxiety symptoms at BL are associated with QoL at FU. |
(a) and (b) Multivariate stepwise forward regression | (a) Higher depressive symptoms at BL were significantly associated with reduced QoL at FU. (b) Higher social anxiety symptoms at BL were not significantly associated with QoL at FU. |
Abbreviations: QoL = quality of life; MD = major depression; FU = follow-up; DSM = Diagnostic and Statistical Manual of Mental Disorders; HDRS = Hamilton Depression Rating Scale; PCS = Physical Component Score; MDS = Mental Component Score; MDD = major depressive disorder; ANOVA = analysis of variance; BL = baseline; MDE = major depressive episode; CIDI = Composite International Diagnostic Interview; SF-36 = Short Form 36; AUDADIS = Alcohol Use Disorders and Associated Disabilities Interview Schedule; SF-12 = Short Form 12; PHQ = Patient Health Questionnaire; SF-12v2: Short Form 12, Version 2; HRSD = Hamilton Rating Scale for Depression; HADS = Hospital Anxiety and Depression Scale; QLDS = Quality of Life in Depression Scale; EQ-VAS = EQ Visual Analogue Scale; DIS = Diagnostic Interview Schedule; BDI = Beck Depression Inventory; SCID = Short Children’s Depression Inventory; MINI = Mini-International Neuropsychiatric Interview; PTSD = post-traumatic stress disorder; hrqol = health-related quality of life, IES-15 = Impact of Event Scale 15; Q-DIS = Quick Version of the Mental Health’s Diagnostic Interview Schedule; MADRS = Montgomery–Åsberg Depression Rating Scale; FDD-DSM-IV = Fragebogen zur Depressionsdiagnostik nach Diagnostic and Statistical Manual of Mental Disorders IV; SCAN = Schedule for Clinical Assessment in Neuropsychiatry; DASS = Depression Anxiety Stress Scales; MOS SF = Medical Outcomes Study Short Form; CES-D = Center for Epidemiological Studies Depression Scale; WHOQOL-Bref-TW = WHOQOL-Bref Taiwan Version; MHI-5 = Mental Health Inventory 5; OCD = obsessive compulsive disorder; Y-BOCS = Yale–Brown Obsessive Compulsive Scale; BAI = Beck Angst Inventar; DD = depressive disorder; PD = psychiatric disorder; SAD = social anxiety disorder; Q-LES-Q = Quality of Life Enjoyment and Satisfaction Questionnaire; GHQ-28 = General Health Questionnaire 28; PCL-S = Post-traumatic Stress Disorder Checklist Scale; VETR-PTSD = Vietnam Era Twin Registry Posttraumatic Stress Disorder; DRPST = Disaster-Related Psychological Screening Test; SCL-90 = Symptomcheckliste bei psychischen Störungen 90; SASC = SpLD Assessment Standards Committee; QOLS = Quality of Life Scale; CDI = Children’s Depression Inventory.