The Scottish veterans health study: a retrospective cohort study of 57,000 military veterans and 173,000 matched non-veterans

Bergman, Beverly P. (2015) The Scottish veterans health study: a retrospective cohort study of 57,000 military veterans and 173,000 matched non-veterans. PhD thesis, University of Glasgow.

Full text available as:
PDF (edited version, 3rd party copyright removed)
Download (16MB) | Preview
Printed Thesis Information:


Although the health of military personnel who have taken part in specific conflicts has been studied throughout the 20th century, there is a paucity of evidence on the long-term overall impact of military service on health. This thesis describes the establishment of and findings from the Scottish Veterans Health Study, a retrospective cohort study comparing the health outcomes of veterans with those of people with no record of service, in order to determine whether the long-term health of military veterans living in Scotland differed from that of people who had never served in the armed forces.

The study population comprised all 57,000 military veterans born between 1945 and 1985 who were resident in Scotland both before and after military service, together with a 3:1 comparison group of 173,000 people with no record of service, matched for age, sex and postcode sector of residence. The demographic data were extracted from the National Health Service Central Registry database and were linked electronically to the National Health Service Scottish Morbidity Record and national vital records data for acute and psychiatric hospital admissions, psychiatric day-case admissions, cancer registrations and death certificate data. Survival analysis was used to determine hazard ratios for those health conditions and outcomes considered to be of a priori interest, overall, by sex, by birth cohort and by length and period of service, both univariately and after adjusting for deprivation.

Veterans were at significantly increased risk of cardiovascular disease compared to non-veterans overall, and of acute myocardial infarction, stroke and peripheral arterial disease specifically. Subgroup analysis showed the increased risk to be confined to veterans born between 1945 and 1959, reducing in more recent birth cohorts. The risk was highest in veterans who left after only a short period of service (Early Service Leavers), whilst those who served for longest exhibited a similar risk of cardiovascular disease to all non-veterans. Veterans were at no higher overall risk of cancer than non-veterans, although there were major differences in the risk of specific cancers, which changed over time. The oldest veterans had an increased risk of cancer of the lung, oropharynx and larynx, oesophagus and stomach; the risks of these cancers reduced in more recent birth cohorts. The 1960-1964 birth cohort showed an increased risk of both bladder cancer and pancreatic cancer in comparison with non-veterans. There were increased risks of ovarian cancer in veteran women compared with non-veterans, and of breast cancer in longer-serving women. The risk of cervical cancer decreased in more recent birth cohorts. There were no differences in the risk of colorectal cancer or prostate cancer in veterans, overall or in any subgroup. There was no clear evidence of increased risk of lymphohaematopoietic cancer in veterans. Veterans were at increased risk of motor neuron disease, but not of multiple sclerosis. Veterans were at increased risk of peptic ulcer disease for all birth cohorts up to the mid-1960s but not thereafter; the risk was highest in those with the shortest service. Hepatitis C was less common in veterans than in non-veterans, in all subgroups. Analysis of mental health outcomes showed that the greatest burden of ill-health was among Early Service Leavers, whilst veterans who completed at least a minimum length of engagement were not at increased risk compared with non-veterans, except for post-traumatic stress disorder. The results for post-traumatic stress disorder, in both veterans and non-veterans, demonstrated a complexity which could not be reconciled with any operational exposure or conventional clinical pattern, but which may have reflected a ‘hidden iceberg’ of unmet need in the late 1990s which was uncovered by increasing awareness of the condition. Longer service was generally associated with better mental health. Veterans were at no greater risk of suicide than non-veterans; the risk was independent of length of service. Veteran women exhibited a risk profile for mental health outcomes which more closely resembled that of veteran men; this was especially marked for suicide. Veterans were not at increased risk of alcoholic liver disease overall; the only subgroup to show an increase in risk was Early Service Leavers who had completed training, and there was also evidence of increased risk of some alcohol-related cancers in trained Early Service Leavers.

Older veterans demonstrated an increased risk of smoking-related ill-health, including cardiovascular and respiratory disease and the smoking-related cancers, which is consistent with reported high rates of military smoking in the 1960s and early 1970s. Overall, there has been an improvement in health of veterans compared with the non-serving population in more recent generations, suggesting that the increased emphasis on health promotion and physical fitness in the armed forces since the late 1970s has been effective. Major alcohol problems were no more common in veterans than in the wider community, and were most likely to affect those who left earliest, although not those who left whilst still in training. Longer service was generally associated with better long-term health. Early Service Leavers had poorer health outcomes than longer-serving veterans, but the ability to stratify by length of service demonstrated that the poorest outcomes were in those who did not complete initial training. It is likely that their long-term health outcomes have been predominantly influenced by pre-service and post-service health and behavioural factors which, at a pre-service level, may have also contributed to their failure to complete the minimum military engagement, rather than by their short period of military service. The early period of service appears to act as an extension to the screening process for entry to service, filtering out those who prove least suited to service. The Early Service Leavers therefore form a ‘less healthy leaver’ group which is the counterpart to the longer-serving ‘healthy worker effect’; their status as veterans means that they can be identified within the community, unlike most other occupational leaver groups, but their poorer long-term health is unlikely to be due to military occupational factors. Improved understanding of the determinants of veterans’ health will inform the provision of appropriate health and community services to meet their needs.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Military veterans, armed forces, early service leavers, retrospective cohort studies, survival analysis, long-term health outcomes, smoking, alcohol, cardiovascular, cancer, mental health, healthy worker effect
Subjects: R Medicine > R Medicine (General)
R Medicine > RA Public aspects of medicine > RA0421 Public health. Hygiene. Preventive Medicine
Colleges/Schools: College of Medical Veterinary and Life Sciences > Institute of Health and Wellbeing > Public Health
Supervisor's Name: Pell, Prof. Jill P.
Date of Award: 2015
Depositing User: Dr Beverly P Bergman
Unique ID: glathesis:2015-7144
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 09 Mar 2016 11:19
Last Modified: 30 Jan 2020 11:38

Actions (login required)

View Item View Item


Downloads per month over past year