Longitudinal investigations of associations between offending and health are rare. Studies which have investigated this relationship principally focus on mental health within samples of incarcerated offenders. Therefore, the physical health of offenders outside secure settings, henceforth ‘community offenders’, form an under-researched and potentially vulnerable group.

To address the limited knowledge on the relationship between community offenders and health, this thesis utilised access to a unique prospective longitudinal study, The Cambridge Study in Delinquent Development (CSDD), and conducted two global systematic reviews and meta-analyses. This aim of this thesis is to use these data sources to longitudinally, systematically, and meta-analytically investigate to what extent community offenders have poorer physical health across the life-course and die prematurely compared to non-offenders.

Chapter 1 provides an introduction together with a comprehensive review of the current evidence and theorizations behind the health-crime relationship. This highlights the limitations and gaps in the literature to date and sets out how this thesis seeks to contribute to this debate. Within Chapter 2, the first central focus, in contrast to the conviction and risk factor analyses previously performed, is on temporary versus persistently antisocial persons, based on three offending trajectories (see: Moffitt, 1993; Jolliffe, Farrington, Piquero, McLeod, & Van de Weijer, 2017b): Life-Course-Persistent, Adolescence-Limited, and Late-Onset offenders. These three typologies constitute qualitatively distinct types of person and their different offending pathways may bear differential risks for adult health. By using data from the CSDD, the new approach in this thesis investigates the longitudinal impact of criminal behaviour on physical health problems in self-reports and General Practitioner (GP) data by testing the following hypothesis: individuals who commit offences earlier in their lives and have prolonged criminal careers will suffer from greater physical health problems than individuals who have short criminal careers. These early onset individuals, it is further hypothesised, will also have greater odds of injury and hospitalization than Late-Onset offenders, who in turn will have greater odds than Non-offenders.

These CSDD longitudinal analyses found that, when considering organic illnesses (respiratory tract, cardiovascular, musculoskeletal, skin, allergic, gastrointestinal and infectious illnesses) and hospitalizations (the number of hospital visits), the impact of offending on health becomes more serious if offending persists beyond adolescence.

The second focus of Chapter 2 was to conduct further analyses investigating the relationship between psychosocial risk factors at age 8-10, antisocial personality (ASP) at ages 18, 32 and 48, and poor physical health (based on self-reports and GP records). These analyses found that high ASP scores at ages 18, 32 and 48 were related to a high prevalence of hospitalization. They also found that, according to GP records, high ASP scores at age 32 were related to poor physical health, and high ASP scores at age 48 were related to more mental illness and disabling medical conditions. These three CSDD analyses also highlight age-specific health implications related to the ages at which offenders begin and end their delinquent behaviour. Unfortunately, these findings could not be supplemented through a systematic review and meta-analysis, due to the paucity of existing studies. The worst consequence of poor physical health is premature mortality, so this element of the health-crime relationship was subsequently investigated in Chapter 3.

In Chapter 3, a systematic review and meta-analysis sought to establish whether community offenders die prematurely compared to non-offender community and population comparison samples. Thirty-six studies met the inclusion criteria (N= 1,116,614). Premature mortality is a significant issue for non-incarcerated offenders in general (OR= 3.42), and for ex-prisoners in particular (OR= 4.51). Offenders were more likely to die from unnatural violent causes (OR= 3.97) and natural causes (OR= 2.06) than non-offenders, with a meta-regression revealing that time at risk was not a significant factor (z= -0.01, p= 0.12). These results suggest that the rates of premature mortality previously found for offenders do not just reflect the impact of mental illness on these individuals, but rather that offending and its correlates may have a significant physiological impact on the body. Suicide was of particular interest when considering the causes of premature mortality in community offenders, one of the most prominent causes of death in offenders (Fazel, Benning, & Danesh, 2005) and males globally (WHO, 2018). A further systematic and review and meta-analysis was therefore conducted to investigate this significant element of the health-crime relationship.

In Chapter 4, a second systematic review and meta-analysis sought to establish whether community offenders were more likely to commit suicide compared with community and general population comparison groups. Fifteen studies met the inclusion criteria (N= 602,347) and highlight that non-incarcerated offenders are significantly more likely to commit suicide compared with non-offenders (OR= 4.54), with time at risk being a non-significant factor. Ex-prisoners had a high likelihood of suicide (OR= 4.18), but not as high as offenders who had not been incarcerated (OR= 7.62).

Chapter 5 presents limitations concerning the studies conducted in Chapters 2, 3 and 4, in addition to providing recommendations for future research. Although this thesis could not provide directional or causal conclusions, Chapter 6 argues that the evidence provided suggests that the antisocial lifestyles that offenders lead when out of secure environments pose a significant risk to physical health. It is likely that the antisocial lifestyle of offenders causes processes which damage their health over time, and several criminological, epidemiological and medical theories are discussed to explain these links.

Overall, community offenders form a vulnerable group who require targeted interventions to reduce the incidence of poor physical health, mortality and suicide across the lifespan. Yet, the prior research and literature reviewed within this thesis demonstrates that community offenders form an under-researched group, with methodologically limited research conducted to date. Until there is further understanding of the health-crime relationship, the nature of these interventions remains impossible to comment upon. Future directions for new research are discussed which aim to produce further robust evidence on the relationship between community offenders and poor health, together with the differences between individual prospective longitudinal work and population level meta-analytic findings. These approaches should seek to establish causality and directionality of relationships and inform the design of future interventions.

The findings of this thesis, with respect to the health-crime relationship, should be viewed as a future public health challenge and continue to inform the evidence from which targeted interventions can be developed, with the aspiration of improving the health and life chances of community-based offenders.