The long-term psychosocial outcomes of UK armed forces personnel who sustained serious combat injuries during deployment to Afghanistan are largely unknown. We aimed to assess rates of probable post-traumatic stress disorder (PTSD), depression, anxiety, and mental health-associated multimorbidity in a representative sample of serving and ex-serving UK military personnel with combat injuries, compared with rates in a matched sample of uninjured personnel. This analysis used baseline data from the ADVANCE cohort study, in which injured individuals were recruited from a sample of UK armed forces personnel who were deployed to Afghanistan and had physical combat injuries, according to records provided by the UK Ministry of Defence. Participants from the uninjured group were frequency-matched by age, rank, regiment, deployment, and role on deployment. Participants were recruited through postal, email, and telephone invitations. Participants completed a comprehensive health assessment, including physical health assessment and self-reported mental health measures (PTSD Checklist, Patient Health Questionnaire-9, and Generalised Anxiety Disorder-7). The mental health outcomes were rates of PTSD, depression, anxiety, and mental health-associated multimorbidity in the injured and uninjured groups. The ADVANCE study is ongoing and is registered with the ISRCTN registry, ISRCTN57285353. 579 combat-injured participants (161 with amputation injuries and 418 with non-amputation injuries) and 565 uninjured participants were included in the analysis. Participants had a median age of 33 years (IQR 30–37 years) at the time of assessment. 90·3% identified as White and 9·7% were from all other ethnic groups. The rates of PTSD (16·9% [n=89] vs 10·5% [n=53]; adjusted odds ratio [AOR] 1·67 [95% CI 1·16–2·41], depression (23·6% [n=129] vs 16·8% [n=87]; AOR 1·46 [1·08–2·03]), anxiety (20·8% [n=111] vs 13·5% [n=71]; AOR 1·56 [1·13–2·24]) and mental health-associated multimorbidity (15·3% [n=81] vs 9·8% [n=49]; AOR 1·62 [1·12–2·49]) were greater in the injured group than the uninjured group. Minimal differences in odds of reporting any poor mental health outcome were noted between the amputation injury subgroup and the uninjured group (AOR range 0·77–0·97), whereas up to double the odds were noted for the non-amputation injury subgroup compared with the uninjured group (AOR range 1·74–2·02). Serious physical combat injuries were associated with poor mental health outcomes. However, the type of injury sustained influenced this relationship. Regardless of injury, this cohort represents a group who present with greater rates of PTSD than the general population, as well as increased psychological burden from multimorbidity.