This cross-sectional study of 282 patients within the POKAL consortium provides further evidence of significant social environment deficits among depressed inpatients and outpatients. Over half of the patients with depression in our sample were at risk of social isolation, with a prevalence four times higher than that of the general population. Similarly, social disability levels were elevated in patients with depression compared with the general population. Notably, no differences in social disability were observed between socially isolated and non-isolated patients. Socioeconomic and social environment characteristics were similar between inpatients and outpatients. The quality appraisal suggests that the WHODAS and the LSNS are suitable questionnaires for patients with depression. The WHODAS showed only minor weaknesses, while the SCTS demonstrated major shortcomings.

Every second patient—whether inpatient or outpatient—reported being at risk of social isolation according to the LSNS, compared with every eighth person in the general population.23 This finding might be attributed to the already impaired interpersonal functioning of the patients with depression caused by depression itself on one hand, or it may reflect a pre-existing trait that predisposes individuals to develop depression. Furthermore, the high levels of social isolation could be associated with the high prevalence of childhood trauma in our sample, which might be associated with a poorer ability to integrate socially in later life and therefore explain our results (Eder et al, A network analysis approach to loneliness, social support, and adverse childhood experiences in the context of adult depression and treatment response, under revision). Interestingly, depression severity and social disability did not differ in our sample between socially isolated and non-isolated patients. Moreover, another study found that patients with depression with larger social networks achieved better outcomes after 4 weeks (Eder et al, under revision), and this underscores the need for routinely monitoring social networks as part of depression management. Such assessments could help identify individuals at higher risk of depression and facilitate the integration of social interventions.25

Our cohort showed higher levels of social disability (in the WHODAS) compared with both the general population and individuals with pre-existing mental health conditions25 26 (online supplemental figure 3). This may be explained because our patients were assessed during a current depressive episode, which tends to intensify impairments. Despite the assumption that inpatients are more severely affected, no differences were found in our sample between inpatients and outpatients. This could indicate a potential underestimation of disability because of recall or selection bias, as only less severely ill individuals may have participated in the study. The elevated social disability in our study aligns with previous research showing that social impairments are both common and debilitating in depression.1 6 As these impairments can persist for years, even after recovery,27 detecting and addressing them could improve the patients’ quality of life.

Despite the association between neighbourhood social cohesion and adolescent depression,9 its assessment is not routinely performed in clinical settings. Our findings revealed no differences in social cohesion between inpatients and outpatients, but the lack of comparable data for the general population highlights the need for further research.

The quality appraisal of the LSNS, WHODAS and SCTS indicated that the LSNS and WHODAS are suitable for clinical use in patients with depression, while the SCTS is not suitable in the clinical context.

The importance of social aspects in mental health is well-established, as demonstrated by this study and the recent EU-wide Eurobarometer survey, which identified social environment as a key contributor to mental well-being.28 Also, national institutes such as the German Robert Koch Institute recognise social aspects as important,29 however they lack practical recommendations or structured guidance on selecting and using standardised questionnaires to assess these aspects. Introducing such guidance could potentially be provided by national institutes, medical or public health associations or consortiums to improve nationwide monitoring, providing a more comprehensive understanding of the role and impact of social factors on mental health. In the face of a lack of qualitative data, we tried to build the foundation for guiding structure for a context-specific appraisal of the questionnaires.

Generally, measuring social cohesion poses challenges due to the variety of definitions and approaches to operationalise this concept, as reflected in the vast number of proxy measures for social cohesion.30 Selecting context-appropriate questionnaires helps ensure that relevant information is not overlooked in the treatment of patients with depression. Certainly, other potentially suitable questionnaires measuring similar social environment aspects, such as the Global Assessment of Functioning, the Social Functioning Scale or the Social Network Index, could have been considered. Furthermore, other relevant social environment constructs, such as subjective social isolation, connectedness and discrimination, could have been of interest. There are contexts in which these questionnaires may be of better use, such as in urban planning and social services. The questionnaires selected within this study were prioritised due to the possibility of directly translating them into treatment or management decisions and recommendations.

The strengths of this study are (1) the comparison of inpatients and outpatients with depression, demonstrating similar levels of social deficits regardless of the treatment setting and (2) a context-specific quality appraisal of questionnaires, providing practical insights into their suitability.

Our findings are not generalisable due to the following limitations: (1) the small sample size from a limited geographical area, where social factors may vary profoundly by region and may be assessed differently in other countries, (2) patients were assessed during the COVID-19 pandemic, when social isolation levels were higher compared with prepandemic conditions. These data were further compared with prepandemic data, which may have inflated the prevalence and the differences in social isolation levels. Also, inpatients were recruited in the highly social clinical environment which may have influenced patients’ responses, introducing recall or perception bias, (3) the cross-sectional design of this study does not allow causal inferences, nor insights into longitudinal timing effects. Social deficit, especially social isolation, ‘is developmentally intertwined with the experience of poor mental health’, consequently social isolation should not be viewed merely as a risk factor or outcome of mental health problems but rather as an integral part of the phenotypic profile of depression.31

Further research should involve larger cohort studies to further investigate the relationship between social environment and depression. Assessing additional constructs such as loneliness, connectedness and discrimination could provide a more comprehensive understanding. Furthermore, the development or refinement of tools to measure social cohesion could enhance its assessment and application in both clinical and public health contexts.