Epidemiological results

Some text in this chapter has been reproduced with permission from Lüchtenborg et al.42 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) licence, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The text below includes minor additions and formatting changes to the original text.

Cohort study of cancer incidence

The prison population increased in size during the 20-year period, with the number of men in prison increasing from 60,252 in 1998 to 77,825 in 2017. The number of women in prison is much lower overall but also increased from 3139 to 4007 (a 28% increase) during the same period. The proportion of male and female prisoners over 50 years of age doubled from 17% and 17% in 1998 to 34% and 35% in 2017, respectively. The number of malignant and pre-malignant tumours diagnosed in prison increased sixfold from 19 in 1998 to 171 in 2017. We identified 2015 tumours in 1964 adult persons in prison during the 20-year period, of which 1556 were diagnosed in men and 459 were diagnosed in women. More than half of cancers in men are made up of prostate (19%), lung (15%), colon and rectal (9%), testis (7%) and bladder (4%) cancer. Most diagnoses in women are cervical cancer in situ (76%), whereas cervix and breast cancer account for 5% of all diagnoses each and 20% and 22% of invasive tumours, respectively.

The age-standardised incidence rate of cancer for men in prison was lower than that for men in the general population during the early study period but increased towards the end of the study period to a level similar to the general population (Figure 1). Poisson regression showed that during the study period the incidence of invasive cancer for men in prison increased [incidence rate ratio per year 1.05, 95% CI 1.04 to 1.06, during 1999–2017 compared with 1998].

FIGURE 1 Age-standardised incidence rates (per 100,000 person-years) for cancer in men in the general and prison populations in England by diagnosis year. (Figure not reproduced here; see the original article.)

The incidence rate ratios in Figure 2 show that over the 20-year period the cancer incidence was lower in both women and men in prison, although not statistically significantly so for women. Incidence rate ratios for the most common male cancers showed statistically significantly lower rates of colon, rectal, prostate and testis cancer in the prison population, but no differences in bladder and lung cancer incidence. Invasive cancers are relatively rare in women in prison, but it is striking that cervical cancer in situ is diagnosed around twice as often in prison as in the general population.

FIGURE 2 Age-standardised incidence rate ratios for common cancer and for all cancers combined in English prisons compared with the general population by gender between 1 January 1998 and 31 December 2017. (Figure not reproduced here; see the original article.)

Matched cohort study of treatment and survival

Table 1 gives an overview of the prison cohort and matched general population cohort. It shows that the matching on age, gender, tumour stage and diagnosis year worked well. There were differences in ethnicity, with fewer individuals with an Asian ethnicity and more with a black, mixed or other ethnic background in the prison cohort than in the general population controls. No significant difference in levels of comorbidity was found. Individuals diagnosed with cancer in prison were far less frequently diagnosed via screening than were matched general population cases. In an analysis where we randomly matched prison cases to cases from the general population on age, gender, tumour site and diagnosis year, but not tumour stage, we found no significant difference between the frequency of stage IV tumours among people diagnosed in prison and that among the general population (28% vs. 26%; p = 0.067).

Hospital admissions

Excluding day cases and regular night and day attenders, we found that fewer patients among those diagnosed with cancer in prison had HES recorded hospital admissions than among those diagnosed in the community (40% vs. 46%; p < 0.001). When admitted, prison patients had slightly longer admissions using the same definitions than community controls, where the median bed-days for prison cases was 8 (interquartile range 3–17 bed-days) compared with 6 (interquartile range 3–13 bed-days) among community controls (p = 0.024).

Curative treatment

Table 2 shows the OR of the receipt of any treatment with curative intent, curative surgery, radiotherapy and systemic anti-cancer therapy of people diagnosed with cancer in prison compared with those in the general population. The prison population is less likely to undergo treatment with curative intent than the general population (OR 0.63, 95% CI 0.53 to 0.75), but the difference is most pronounced for major resections (OR 0.64, 95% CI 0.53 to 0.78). Adjusting the analyses for route to diagnosis in the model attenuated the association (OR 0.73, 95% CI 0.60 to 0.88).

Survival

The survival of patients diagnosed with cancer in prison is lower than that of patients diagnosed with cancer in the community. One-year survival was 71% (95% CI 68% to 74%) among prison diagnoses compared with 74% (95% CI 73% to 76%) among community diagnoses. Five-year survival estimates were 54% (95% CI 50% to 58%) and 56% (95% CI 55% to 58%), respectively.

Table 3 shows the results from the Cox proportional hazards analysis. Being diagnosed with cancer in prison carries a small but significantly increased risk of death [hazard ratio (HR) 1.16, 95% CI 1.03 to 1.30, model A]. Stratification by treatment with curative intent showed that differences in treatment explained almost half the increased risk (adjusted HR 1.09, 95% CI 0.97 to 1.23, model B). Further adjustment for diagnosis route attenuates the association further (HR 1.05, 95% CI 0.94 to 1.18, model D).