In post-industrial countries, ethnic minorities suffer poorer health and premature deaths. The present study examined ethnic differences in life expectancy and related features among elite heavyweight boxers.
Compared with white boxers, non-white boxers tend to die younger with excess neurological and accidental deaths, and they have lower social positions in later life. Sporting authorities should reappraise the wisdom of permitting head injuries in sport and monitor and support the health and wellbeing of sports men and women after retirement.
Studies have shown that those who are attracted to boxing often come from inner cities and single-parent households, and have limited formal education [15, 16]. Blacks have been disproportionately drawn to boxing for a number of reasons: historically, boxing has been a relatively egalitarian sport where members of different ethnic groups can participate on a fairly equal level; it requires relatively limited facilities and equipment and consequently there tends to be a higher density of boxing clubs in minority neighbourhoods ; often, the lack other social opportunities makes boxing attractive to young men from minority groups [18, 19]. Floud et al.  argued that professional boxers have tended to come from socially deprived backgrounds where those who grow tallest commonly have better life-opportunities and thus experience upward social mobility. Choosing to become a boxer seems to offer a route to the rapid acquisition of substantial wealth and status [15, 21]. Although there is a body of evidence on racial disparity in mental and physical health, life expectancy [22,23,24], and mortality  among the general population, no previous comprehensive study has assessed how ethnicity could influence the health and life expectancy of these elite boxers once their sporting career is over.
We sought to determine whether there exist ethnic differences in life-course after retirement from boxing, including (1) life expectancy, (2) cause of death, (3) occupation, and (4) substance abuse and convictions in white and non-white male heavyweight boxers who have contested for World Championship titles from 1889 to 2019.
Demographic information of male boxers recorded at the time of the championship contests was collected, including date of birth and date of death, cause of death, occupation after boxing career, substance abuse and convictions, champion/challenger status (champions of any of the recognised World boxing titles, principally World Boxing Council, World Boxing Association, World Boxing Organization and International Boxing Federation , and unsuccessful challengers who contested but never prevailed in any of the World title bouts), nationality (categorised as North America, Europe, and others), and ethnicity (American white; African-American; European white; and others including mixed race, Hispanic, Samoan, and African) . Weight and height were also collected and BMI calculated as weight/height squared.
Ethnicity was grouped into whites and non-whites; nationality into North America, Europe, and others; and occupation into non-manual, manual, and unemployed. Causes of death include neurological disorders (dementia, stroke), accidents (involving aeroplane, automobile and road traffic, and work accidents), cancers of any aetiology, homicides (murder by any method), metabolic disorders (diabetes, liver and kidney failure), infections (pneumonia, sepsis), cardiac conditions (coronary heart disease, heart failure), and natural causes (due to old age). Reported occupations (day jobs) were classified into non-manual jobs (successful business, managers, public officials, shop ownership, established acting career), manual jobs (doorman, truck driving, training boxers), and unemployed (no day job). Five boxers who died during their boxing career were not included in the analysis in relation to occupational status. Substance abuse includes those who have been reported to have a problem with alcoholism and use of illicit drugs, and convictions include illegal activities that resulted in imprisonment.
Chi-squared tests were performed to assess differences in the proportion of deaths in different categories of age of death or cause of death between ethnic groups. Cox regression and Kaplan-Meier survival curves were performed to examine the risk of early death and logistic regression causes of death in white (referent group) and non-white boxers. Data were adjusted for age at contest, nationality, champion/challenger status, BMI, and number of boxing matches (or rounds). Analyses were conducted using IBM SPSS Statistics for Windows, version 23.0 (IBM Corp., Armonk, NY, USA). The null hypothesis was rejected when p
All 239 boxers who contested heavyweight championships of the World between 1889 and 2019 were identified. Two who were stripped of the titles after being found positive for the use of banned substances were excluded, leaving 237 for analysis. 35.0% were champions and 65.0% were unsuccessful challengers. 65.0% of all boxers came from North America, 18.3% from European countries, and 10.5% from elsewhere. Of these boxers, 52.7% were white and 47.3% non-white. 10.5% of boxers were known to have a history of substance abuse or convictions, 59.6% had non-manual jobs, and 24.0% manual jobs, while 10.6% were unemployed. Neurological disorders (15.5%) and cardiac conditions were among the highest causes of death (16.4%).
By 2019, 110 boxers (75 whites, 34 non-whites) were known to have died (1 white boxer with unknown date of death). Non-white boxers died at an earlier age than whites boxers (mean SD = 59.8 14.2 years versus 67.3 16.4 years, p = 0.018). In comparison with published data for the general US population, the average life expectancy of white males was 75.7 years and black males was 69.5 years in 2006, and rose to 76.1 years for whites and 71.5 years for blacks a decade later in 2017 . Compared with their respective racial groups in the general US population, non-white boxers had 9.7 years shorter life expectancy and white boxers had 8.4 years shorter life expectancy in 2006, while non-white boxers had 11.7 years shorter life expectancy and white boxers had 8.8 years shorter life expectancy in 2017.
Occupations after the end of boxing career were available for analysis from news archives for 104 deceased boxers (71 white, 33 non-white). Overall, 60.2% of them had had non-manual jobs; 24.3% manual jobs, with 10.7% reported as unemployed; and 4.9% as not known. Among white boxers, 71.8% had non-manual jobs, 19.7% had manual jobs, and 2.8% were unemployed or involved in illegal trades. In comparison, among non-white boxers only, 34.4% had non-manual jobs and 34.4% had manual jobs, while 28.1% were unemployed or involved in illegal trades (Fig. 3a; Table 2).
Based on news archives, a total of 9 boxers (8.3%) were reported to have problems with substance abuse and 7 (6.4%) had a criminal conviction. The proportions of those with substance abuse were similar in white and non-white boxers (8.0% versus 8.8%) but non-white boxers had higher proportions of conviction, 17.6% versus 1.3% (Fig. 4a). Substance abuse was highest among the non-manual group (55.6%) and unemployed boxers (33.3%). Convictions were not reported in the non-manual group, but two convictions were reported in the manual group (8.0%) and five convictions were reported among the unemployed (45.5%) (Fig. 4b).
Proportions of boxers reported in the media to have problems with substance abuse or convicted of crimes by ethnic groups (white bars = white boxers and grey bars =n on-white boxers); group differences: χ2 = 10.5, p = 0.005 (a), and by occupation (white bars = no reported substance abuse or conviction, grey bars = substance abuse and black bars = conviction of crimes; group difference: χ2 = 40.1, p b)
The present study found that among elite heavyweight boxers, non-white boxers died at earlier ages and were more likely to have died from neurological disorders and accidents. Non-white boxers also had higher rates of manual jobs or unemployment and a greater frequency of reported criminal convictions. These findings indicate that despite championship wealth and status, ethnic differences still negatively affected social position and health for non-white boxers once their boxing careers are over. Our findings of ethnic differences in athletes mirror well known disparities in the general populations of the UK  and the USA .
The most striking observations specific to boxers were the increased risk of death from neurological disorders and accidents. These have been shown to exist among ethnic groups in general , but the risk appears to be exaggerated for boxers. Studies of patients admitted to hospital with traumatic brain injury has shown that compared with white patients, blacks and Hispanics have a higher mortality  and are more likely to be cared for by a junior doctor (resident) on acute admissions . Other studies have found that non-whites were less likely to be insured and therefore less likely to be able to afford rehabilitation [46, 47]. On the other hand, one study has revealed that even having the same insurance coverage (Medicare), blacks and Hispanics were less likely than whites to be transferred to facilities to receive a higher level of rehabilitation . These health care inequalities could therefore explain the higher risk of death from neurological causes among non-white boxers observed in our study.
There may be differences in upbringing between whites and non-whites, which would have influenced on their post-boxing environment. Generally, non-white boxers face greater risk of exposure to adverse environmental factors during their life time, away from boxing.