Allahyari Elahe, Moshtagh Mozhgan
Department of Epidemiology and Biostatistics, School of Health, Social Determinants of Health Research Center, Birjand University of Medical Sciences, Birjand, Iran.
Social Determinants of Health Research Center, Faculty of Health, Birjand University of Medical Sciences, Birjand, Iran.
Biomedicine (Taipei). 2021 Mar 1;11(1):26-33. doi: 10.37796/2211-8039.1031. eCollection 2021.
Maintaining and improving prisoners' health and their rehabilitation can be effective steps towards eliminating health inequalities and approaching the UN's Sustainable Development Goals. Accordingly, identifying protective factors and health barriers of this vulnerable group and changing the prison into an environment that can deliver health interventions tailored to the needs of inmates can provide the basis for attaining justice in health.
This study builds on an artificial neural network model to determine the effect of demographic, psychological, criminological, and physical activity factors on prisoners' general health.
The study detected the patterns between variables using a neural network with nine inputs and one output. To determine the neural network with the minimum sum of squared errors, we evaluated the performance of all neural networks using varying algorithms and numbers of neurons in the hidden layer. For this purpose, the analysis of the data of 149 prisoners aged between 16 and 61 years was performed using SPSS-22 software.
The optimal neural network model was useful in predicting prisoners' general health. In this model, the variables of occupation, life expectancy, age, and hope of acceptance were identified as the first most significant factors with 19.25, 17.45, 15.98, and 15.16 percentages, respectively, whereas the cause of incarceration, education level, type of drug misuse, and physical activity were the second most important factors with 8.82, 8.38, 7.91, and 7.04 percentages, respectively.
Experiencing psychosocial pressures related to incarceration is more severe for the marginalized and disadvantaged individuals, persons in very young or old age ranges, and those with no history of incarceration, which can increase the likelihood of impaired health for these inmates. Consideration of the prisoners' needs in proportion to their characteristics and provision of emotional and spiritual support of the inmates, especially in the early stages of incarceration, can help shape an effective adjustment process and select appropriate and efficient strategies.
维护和改善囚犯的健康及其康复状况,可能是消除健康不平等现象并迈向联合国可持续发展目标的有效举措。因此,识别这一弱势群体的保护因素和健康障碍,并将监狱转变为能够提供符合囚犯需求的健康干预措施的环境,可为实现健康公平奠定基础。
本研究基于人工神经网络模型,以确定人口统计学、心理学、犯罪学和身体活动因素对囚犯总体健康的影响。
该研究使用一个具有九个输入和一个输出的神经网络来检测变量之间的模式。为了确定具有最小平方误差总和的神经网络,我们使用不同的算法和隐藏层中的神经元数量评估了所有神经网络的性能。为此,使用SPSS - 22软件对149名年龄在16至61岁之间的囚犯的数据进行了分析。
最优神经网络模型有助于预测囚犯的总体健康状况。在该模型中,职业、预期寿命、年龄和被接纳的希望等变量被确定为最重要的因素,其贡献率分别为19.25%、17.45%、15.98%和15.16%,而监禁原因、教育水平、药物滥用类型和身体活动则是第二重要的因素,贡献率分别为8.82%、8.38%、7.91%和7.04%。
对于边缘化和弱势群体、非常年轻或年老的人群以及没有监禁史的人来说,与监禁相关的社会心理压力更为严重,这可能会增加这些囚犯健康受损的可能性。根据囚犯的特点考虑他们的需求,并为囚犯提供情感和精神支持,特别是在监禁的早期阶段,有助于形成有效的调整过程并选择合适且高效的策略。