Table 3.
Longitudinal studies on social relationships and mental health and/or wellbeing in disability
| First author, year, reference | Country | Follow up time, number of waves | Social relationship measures | Mental health / wellbeing measures | Quality rating | Participant characteristics | Key findings a | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| N | Mean age | % males | Disability | Disease duration | |||||||
| Social support | |||||||||||
| Benka 2012 [35] | Slovakia (EURIDISS) | 4 years, 4 | Social support SSQS | Psychological distress GHQ-28 | Medium | 116 | 47.6 | 15.5 | Rheumatoid arthritis | 0–4 years after diagnosis | Emotional and instrumental social support increased over time. Emotional support T1-T3 significantly negatively associated with psychological distress at T4. Instrumental support T1-T3 not associated with distress at T4. |
| Costa 2013 [81] | Portugal | 2 years, 2 | Social support AIMS 2 | Depression DASS | Medium | 55 | 55.2 | 20.0 | Rheumatoid arthritis | First 2 years of disease progression | Low social support at T1 positively associated with depression at T2. |
| Curtis 2004 [84] | Ireland | 1 year, 2 | Social Support MOS SSS | Anxiety, depression AIMS
Positive and negative affect PANAS |
Low | 52 | 60.0 | 0.0 | Rheumatoid arthritis | 13 years | Cross-sectional analyses at T1 and T2 showed no significant association of perceived social support with depression, anxiety, positive affect or negative affect when controlling for disease status and perceived stress. |
| Demange 2004 [78] | France, the Netherlands, and Norway (EURIDISS) | 3 years, 3 | Social support, social companionship SSQT | Psychological distress GHQ | High | 542 | 52.5 | 31.0 | Rheumatoid arthritis | 0–4 years after diagnosis | Social support did not change over time. Cross-sectional relationships between social support and psychological distress were significant but no longitudinal within subject variation was associated with baseline social support or changes in social support over time. Social companionship decreased over time. Cross-sectional relationships between social companionship and psychological distress were significant but no longitudinal within-subject variation in distress was associated with baseline social companionship. |
| Doeglas 2004 [82] | Netherlands (EURIDISS) | 3 years, 4 | Social support, social companionship SSQT | Depression GHQ | High | 264 | 53.0 | 35.0 | Rheumatoid arthritis | 0–4 years after diagnosis | Level of social support did not change significantly over time. Social support at T1 was significantly inversely associated with depression at T4. Social companionship at T1 did not show a significant association with depression at T4. |
| Evers 1997 [83] | Netherlands | 1 year, 2 | Social support IRGL | Anxiety, depression IRGL | Low | 91 | 57.0 | 30.0 | Rheumatoid arthritis | Shortly after diagnosis | Perceived social support at T0 was significantly inversely associated with anxiety and depression at T0 but not T1. |
| Evers 2002 [85] | Netherlands | 5 years, 3 | Social support IRGL | Anxiety, depression IRGL | Low | 78 | 57.0 | 30.0 | Rheumatoid arthritis | Shortly after diagnosis | Perceived social support at baseline was not significantly associated to depression at 3 or 5 years. |
| Hilari 2010 [79] | UK | 6 months, 3 | Social support MOS SSS | Psychological distress GHQ | High | 87 | 69.3 | 56.0 | Stroke | At stroke onset | Social support was significantly inversely associated with psychological distress at T1. |
| Sit 2007 [36] | Hong Kong | 6 months, 2 | Social support, social companionship SSQT | Depression CES-D | Medium | 95 | 67.0 | 51.6 | Stroke | At stroke onset | Information support and social companionship at T1 but not T0 were significantly negatively associated with depression at T1. |
| Strating 2006 [77] | Netherlands (EURISIDD) | 8 years, 5 | Social support, social companionship SSQT | Psychological distress GHQ | High | 129 | 51.0 | 29.0 | Rheumatoid arthritis | 0–4 years after diagnosis | Social companionship did not change significantly over time whereas emotional support decreased significantly between T4 and T5. Both social support and social network (T1-T3) were insignificantly associated with psychological distress at T4 and T5, when distress at T1-T3 was entered into the model. |
| Townend 2007 [76] | Australia | 3 months, 3 | Social support MSPSS | Depression HADS | Medium | 125 | 75.6 | 49.0 | Stroke | At stroke onset | Social support increased over time. Social support at T1 and T3 was negatively associated with depression at T1 and T3. |
| Van Leeuwen 2010 [31] | Netherlands | 1 year, 2 | Social support SSL-12 | Life satisfaction NV | Medium | 190 | 40.6 | 74.7 | Spinal cord injury | At the start of active rehabilitation | Everyday social support and support in problem situations decreased over time, whereas esteem support remained stable. Everyday social support was positively associated and support in problem situations was negatively associated to life satisfaction over time, in particular in persons with high levels of distress. |
| Van Leeuwen 2012 [32] | Netherlands | 5 years, 3 | Social support SSL-12 | Life satisfaction NV | Medium | 162 | 39.0 | 72.6 | Spinal cord injury | At the start of active rehabilitation | Only everyday social support had a significant positive association with life satisfaction. Esteem social support and support in problem situations showed no significant association with life satisfaction. |
| Social network | |||||||||||
| Evers 1997 [83] | Netherlands | 1 year, 2 | Social network IRGL | Anxiety, depression IRGL | Low | 91 | 57.0 | 30.0 | Rheumatoid arthritis | Shortly after diagnosis | Social network was not associated with anxiety or depression at T0 and only with depression at T1. |
| Evers 2002 [85] | Netherlands | 5 years, 3 | Social network IRGL | Anxiety, depression IRGL | Low | 78 | 57.0 | 30.0 | Rheumatoid arthritis | Shortly after diagnosis | Social networks at baseline was not significantly associated to depression at 3 or 5 years. |
| Relationship quality | |||||||||||
| Robinson 1999 [90] | USA | 2 years, 3 | Social functioning SFE | Depression HAMD | Medium | 50 | 60.0 | 66.0 | Stroke | 3–6 months after stroke onset | Relationship with significant other was significantly inversely associated with depression at T0. No measures of social functioning with significant other, family or children was associated with depression at T1 and T2. |
EURIDISS: European Research on Incapacitating Diseases and Social Support
a Key findings are from multivariate results. + indicates significant positive association (p ≤ 0.05); − significant negative association (p ≤ 0.05), 0 no association; (+) positive but non-significant trend; (−) negative but non-significant trend
Abbreviations for social relationship measures: AIMS 2 Arthritis Impact Measurement Scale 2, IRGL Impact of Rheumatic Diseases on General Health and Lifestyle, MOS SSS Medical Outcomes Study Social Support Scale, MSPSS Multidimensional Scale of Perceived Social Support, SFE Social Functioning Examination, SSL Social Support List, SSQS Social Support Questionnaire for Satisfaction, SSQT Social Support Questionnaire for Transactions
Abbreviations for mental health and wellbeing measures : AIMS Arthritis Impact Measurement Scale, CES-D Centre for Epidemiological Studies Depression, DASS Depression, Anxiety and Stress Scale, GHQ-28 GeneralHealth Questionnaire-28, HADS Hospital Anxiety and Depression Scale, HAMD Hamilton Depression Scale, IRGL Impact of Rheumatic Diseases on General Health and Lifestyle, PANAS Positive and Negative Affect Scale