Research Appraisalobservational

Association of social isolation and loneliness with cancer prognosis: an analysis of the UK Biobank

Journal of affective disordersLiu, Qiaoxue, Zhao, Yingsuo, Xue, Huiwen et al.15 Apr 2026DOI

Clinical Snapshot

65CEBM

Evidence: Moderateobservational

PICO Framework

P — Population	Cancer patients from UK Biobank cohort (n=64,772), median follow-up 6.7 years
I — Intervention	Social isolation (9.6% of cohort) and loneliness (4.8% of cohort) as exposures
C — Comparator	Cancer patients without social isolation or loneliness
O — Outcomes	All-cause mortality, cancer-specific survival across multiple cancer types

Bottom Line

This large UK Biobank analysis demonstrates that social isolation and loneliness are independent prognostic factors in cancer patients, associated with 18-21% increased mortality risk. The study's strength lies in its substantial sample size and robust outcome measurement through death registry linkage. However, the healthy volunteer bias inherent in UK Biobank and lack of detailed confounding adjustment limit generalisability. While the findings suggest psychosocial interventions might improve cancer outcomes, the observational design cannot establish causality. Clinicians should consider screening for social isolation in cancer patients, particularly those with breast, prostate, colon, and bladder cancers where associations were strongest. The modest effect sizes suggest social support interventions would complement, not replace, standard oncological care. Further research with external validation and intervention studies is needed before implementing systematic screening programs.

Evidence: Moderate

Key Findings

P Value: Both associations P < 0.001
Effect Size: Social isolation HR 1.21, loneliness HR 1.18
Primary Outcome: All-cause mortality among cancer patients
Nnt Or Sensitivity: 21% increased mortality risk for social isolation, 18% for loneliness
Confidence Interval: Social isolation 95% CI 1.16-1.26, loneliness 95% CI 1.11-1.25

Clinical Application

Psychosocial screening tools readily available in clinical practice, though intervention strategies require development Relevant to Australian cancer care given similar healthcare system and population demographics. Could inform Cancer Australia guidelines and psycho-oncology services under Medicare Cancer patients in developed healthcare systems, particularly those with colon, melanoma, breast, prostate, kidney, and bladder cancers

Abstract

BACKGROUND: Evidence is scarce for the association between social isolation and loneliness and prognosis of any cancer or other common cancer types. METHODS: To address this gap, we performed a cohort study from 2006 to 2022 using data from the UK Biobank (UKB). Social isolation and loneliness were identified based on the questionnaire of the UKB. Cases of death were obtained from the death registry, where both primary and contributory causes were recorded using ICD-10 codes. The Cox proportional hazard models were used to estimate hazard ratios (HRs) and associated 95% confidence intervals (CIs) of death in relation to social isolation and loneliness among patients with cancer. RESULTS: Among cancer patients studied, 6206 (9.6%) were socially isolated and 3130 (4.8%) reported loneliness. Over a median follow-up of 6.7 years, there were 2378 deaths (38.3%) among exposed patients and 16,566 deaths (29.1%) among unexposed patients. Social isolation was associated with poorer cancer prognosis overall (HR 1.21, 95% CI 1.16-1.26, P < 0.001), while loneliness correlated with a similar risk (HR 1.18, 95% CI 1.11-1.25, P < 0.001). Social isolation specifically impacted survival in colon, melanoma, breast, prostate, kidney, and bladder cancers, as well as myeloproliferative neoplasms. Loneliness was linked to poorer outcomes in breast, prostate, and bladder cancers. CONCLUSIONS: Our findings suggested that social support for these vulnerable patients might have the potential to improve prognosis.

References

1.Liu, Q., Zhao, Y., Xue, H., Wang, Y., Wei, H., & Liu, Q. (2026). Association of social isolation and loneliness with cancer prognosis: an analysis of the UK Biobank. Journal of Affective Disorders, 121049. https://doi.org/10.1016/j.jad.2025.121049

Share:X LinkedIn

Related Research

Systematic Review

15CEBM

Cancer medicine

Applications of Artificial Intelligence in Cancer Diagnosis and Treatment

1 Aug 2026

other

65CEBM

Acta oncologica (Stockholm, Sweden)

User perceptions of machine learning models as decision support for colorectal cancer multidisciplinary team conferences (AID-SIM-2): a qualitative simulation study.