Oskarsson S, Kuja-Halkola R, Latvala A, Andersson A, Garcia-Argibay M, Bertoldi BM, et al. (2021) Low autonomic arousal as a risk factor for reoffending: A population-based study. PLoS ONE 16(8): e0256250. https://doi.org/10.1371/journal.pone.0256250
Abstract
Background Low resting heart rate (RHR) and low systolic blood pressure (SBP) are associated with criminal behavior. However, knowledge is lacking about their predictive value for reoffending.
Aim We aimed to examine associations of RHR and SBP with reoffending in a large population-based sample.
Methods We conducted a cohort study of all convicted male conscripts born in Sweden 1958–1990 (N = 407,533). We obtained data by linking Swedish population-based registers. Predictor variables were RHR and SBP, measured at conscription which was mandatory until 2010 for men at age 18. The outcome variable was reoffending, defined as criminal convictions (any crime, violent crime and non-violent crime), obtained from the National Crime Register. We used survival analyses to test for associations of RHR and SBP with reoffending, adjusting for pertinent covariates such as socioeconomic status, height, weight and physical energy capacity.
Results In fully adjusted Cox regression models, men with lower RHR (≤60 bpm) had higher risk of reoffending (any crime: HR = 1.17, 95% CI: 1.14, 1.19, violent crime: HR = 1.23, 95% CI: 1.17, 1.29, non-violent crime: HR = 1.16, 95% CI: 1.14, 1.19), compared to men with higher RHR (≥ 82 bpm). Men with lower SBP (≤80 mmHg) had higher risk of reoffending (any crime: HR = 1.19, 95% CI: 1.17, 1.21, violent crime: HR = 1.16, 95% CI: 1.12, 1.20, non-violent crime: HR = 1.20, 95% CI: 1.18, 1.22), compared to men with higher SBP (≥138 mmHg).
Conclusions Low autonomic arousal is associated with increased risk of reoffending. RHR and SBP should be investigated further as potential predictors for reoffending as they each may have predictive value in risk assessment protocols.
Discussion
In a sample of 407,533 convicted male conscripts in Sweden born between 1958 and 1990, we found that lower RHR and lower SBP at age 18 were associated with an increased risk of reoffending. These results replicate findings from previous work conducted on risk factors for criminal behavior [5, 9], and extend these findings in important ways. Specifically, our findings replicate prior work demonstrating that individual differences in autonomic arousal, measured as RHR and SBP, are associated with criminal behavior [5, 9]. However, earlier work has not considered the prior criminal histories of study participants. Our study focused on men with a history of criminal convictions and showed that these indicators of autonomic arousal predicted subsequent reoffending. This finding is important because it provides evidence that variations in autonomic arousal are associated not only with the likelihood of offending [5, 9], but also with the persistence and repetitiveness of offending.
Our findings suggest that low RHR and low SBP should be further investigated and considered as potential predictors to be included in risk assessment protocols. All estimates in the present study remained statistically significant, with some in fact strengthened, after adjusting for pertinent covariates, indicating that RHR and SBP contribute uniquely to prediction of reoffending. Individuals with low RHR who engage in criminal behavior may not benefit from traditional treatment programs to the extent they are, according to theoretical accounts based on low fear [4] and stimulation-seeking propensities [4, 16], less able to learn from their experiences. In particular, low fear of punishment may reduce the effectiveness of conditioning. As evidence for this, youth with disruptive behavior disorders have been found to profit less from behavioral treatment if they also exhibit low RHR [30]. From this perspective, individuals who engage in criminal behavior who also have low RHR may require tailored interventions targeting their under-aroused autonomic nervous system to prevent them from reoffending. Low RHR and low SBP may have limited importance in themselves as predictors of reoffending, but together with more established predictors (e.g., history of criminal behavior, employment, psychiatric disorder) [2], they could add importantly to the identification of individuals at high risk for reoffending who could be prioritized for intervention programs. We encourage future research aimed at incorporating autonomic arousal measures into models for predicting reoffending.
It is well-known that criminal behavior runs in families [23], and that measures of autonomic arousal are heritable [31]. In addition, findings from recent research provide evidence that the association between low RHR and criminal behavior is substantially attributable to genetic influences [9]. This evidence suggests that autonomic arousal as indexed by low RHR and SBP levels may be one of the biological factors underlying transmission of criminal behavior from one generation to the next [8]. Taken together, our findings provide further impetus for considering autonomic arousal variations in etiologic models of reoffending.
The results from the present study are novel, as only one previous small-scale study has tested for an association between low RHR and reoffending [10]. In contrast to our results, this study did not find an association between low RHR and reoffending, potentially due to lack of statistical power. However, this study did find that weak heart rate reactivity and an elevated heart rate variability within a stressor task predicted a higher likelihood of reoffending during follow up. The current study is therefore the first to demonstrate an increased risk of reoffending in a large-scale population-based sample of men with prior conviction histories. Our finding of an increased risk for reoffending among previously convicted men with low SBP is also novel, as no study has examined this association before. Future research should replicate these findings in different settings and populations.
The findings of the present study should be considered in the light of some limitations. RHR and SBP were measured concurrently using an arm-cuff monitor, after subjects had rested for 5 to 10 minutes [21]. This procedure differs from the conventional laboratory method of measuring cardiac activity from skin-surface electrodes attached to the limbs or torso, which may yield cleaner data [32]. However, use of an arm-cuff is the most prevalent method for obtaining these measures clinically, and is standard practice in clinical settings in Sweden [21]. Further, no other information pertaining to the collection of RHR and SBP data was available in the Swedish Military Conscription Register. Therefore, factors such as time of recording and room temperature that may influence cardiovascular measurements could not be controlled for. It also warrants mention that valid RHR data were missing for 43% of the overall registry sample. However, sensitivity analysis including only men with valid RHR and SBP data yielded comparable estimates. This is in line with a prior study examining low RHR in relation to criminal behavior, which showed that excluding the portion of the sample who lacked RHR data did not affect the observed associations [5].
A further limitation is that our study included only men, so it remains to be seen whether our findings generalize to women. Although previous research has shown that low autonomic arousal is associated with criminal behavior for both men and women [7], we cannot draw the same conclusion with respect to reoffending. It is also important to bear in mind that our study focused on men with official convictions only, who may not be representative of all men who have engaged in criminal acts within Sweden.
In conclusion, our findings demonstrate evidence for associations for two distinct indices of autonomic arousal, RHR and SBP, with risk of reoffending, even after adjusting for possible extraneous confounds. These findings indicate that low RHR and low SBP should be further examined as predictors of reoffending, as they may help to improve identification of individuals at risk for repeated criminal justice involvement. Further evidence for the predictive value of autonomic arousal measures would support their inclusion in risk assessment protocols, as a basis for targeting case management and intervention efforts.