Studies describing posttraumatic stress disorder (PTSD) as a result of physical illness and its treatment were reviewed. PTSD was described in studies investigating myocardial infarction (MI), cardiac surgery, haemorrhage and stroke, childbirth, miscarriage, abortion and gynaecological procedures, intensive care treatment, human immunodeficiency virus (HIV) infection, awareness under anaesthesia, and in a group of miscellaneous conditions. Cancer medicine was not included as it had been the subject of a recent review in this journal. Studies were reviewed in terms of the prevalence rates for PTSD, intrusive and avoidance symptoms, predictive and associated factors and the consequences of PTSD on healthcare utilization and outcome. There was considerable variability both in the study methodology and design and in the results. The highest prevalence rates were identified in patients treated in intensive care units (ICUs) and those with HIV infection. Irrespective of the physical illness, posttraumatic symptomatology is more common than PTSD caseness. Existing characteristics of the patient may well predispose individuals to the development of PTSD as do other factors such as poor social support and negative interactions with healthcare staff. Generally, the severity of the illness itself is not predictive of PTSD. Issues relating to sampling, attrition, diagnosis, the course of symptoms, aetiological pathways, and the consequences of the disorder are discussed. The presence of PTSD most probably influences the patient's use of healthcare resources and may affect their clinical outcome.