The COVID-19 pandemic, hereafter referred to as the pandemic, has had a significant impact on the physical and mental health of the prison population, as a particularly vulnerable group [1]. This was due to pre-existing prison conditions (e.g., overcrowding) [2], highlighting poor ventilation [3], inadequate resources and healthcare [4], and to many inmates having pre-existing physical and mental health needs [5].

The main difficulties encountered during imprisonment, besides the pandemic, include: rejection by society, placing the prisoner in an inferior social category, being under the influence of other people, and being deprived of autonomy, facilities, services, sexual contact, or a sense of security. These difficulties were exacerbated by the disruption of their way of life, particularly in terms of family contact, occupational, leisure and social activities, decision-making opportunities, and by the lack of privacy in their shared living space. This constant deprivation of basic needs can lead to frustration, conflicts, negative experiences, and poor mental health [6]. The prison system differs across countries. The majority of prisoners globally are men (93%), but over the last 20 years the number of women in prison has shown a greater increase (33% increase) than that of men (25%). There are large differences in incarceration rates across regions, ranging from almost 600 inmates per 100,000 inhabitants in North America, to around 50 in South Asia. In 2019, the global mean was estimated at 152 incarcerated persons per 100,000 population. Over the last 20 years, North America, Sub-Saharan Africa, and Eastern Europe have reduced long-term imprisonment rates -up to a 27% decrease since 2000- but Latin America and Australia/New Zealand have increased them -up to 68%. Prisoners in half of the countries are in overcrowded facilities, at more than 100% of capacity [1].

As social beings, people need interaction, which has proven positive effects on the mental health [7, 8]. Conversely, social isolation can have long-lasting negative consequences [8–10], especially in vulnerable populations [11]. In the long term, solitary confinement, referring to the physical and social isolation of an individual, is considered torture [12] and is associated with psychological consequences such as increased levels of anxiety and stress, sleep disruption, loss of emotional control, paranoia, cognitive disturbances, and increased self-harm, suicide [8, 9], depression, and aggression [8, 13].

Following recent infectious outbreaks such as SARS, Influenza A/N1H1, and Ebola, evidence has emerged on the impact of such threats on psychological wellbeing [14]. In this line, greater psychopathological morbidity, primarily anxiety and depression, has been identified during early stages of these outbreaks [14–17]. In this sense, there is evidence of the influence that information has had on reactions of fear, anxiety, and depression, as well as on the application of adaptive protective behaviours [16, 18, 19].

The development of protective behaviours, such as avoiding social contact or adopting hygiene patterns, has been described as related to the amount of information received, to individual attitudinal factors, and to levels of anxiety. Among the attitudinal factors, perceived susceptibility, perceived severity of the disease, and assumptions about the efficacy of the recommended behaviours were highlighted [15, 16, 18, 19]. When describing the relationship between protective behaviours and anxiety and depression, it is generally described that higher levels of anxiety are associated with more appropriate and consistent protective behaviours [15, 20], although other studies have found opposing data [18, 21]. And regarding the information received, it must be clear and consistent in order to encourage the implementation of protective behaviours [19–22].

From here, in the early phases and during the quarantine period of the COVID-19 pandemic in 2020, research was developed on the psychological effects on emotional wellbeing and the sense of coherence [20, 23, 24]. By integrating data on the impact of other outbreaks, evidence suggested that a large-scale lockdown would lead to a range of psychological consequences. Early data from the epidemic originating in China indicated the importance of protecting mental health, noting high levels of anxiety. In addition, protective factors such as adequate health information and protective measures were identified [21]. On the other hand, sadness, grief, and fear have been reported as emotional reactions. The relationship between fear and the high rates of infection, morbidity, and mortality associated with the disease has been described. Thus, high levels of fear can lead to maladaptive behaviours and emotional responses [25]. The Fear of COVID-19 Scale (FCV-19S) was validated in English to measure the level of fear [26], as well as in Italian [27] and Spanish [25].

Limited published research has identified that isolation and anxiety related to the risk of infection were common issues in all studies examining the impact of COVID-19 on the mental health of imprisoned populations [28]. The adverse effect of fear of COVID-19, the effects of social distancing and isolation, and the disruption of visits, as well as the reduction or discontinuation of mental health services, are described as having a detrimental effect on the mental wellbeing of inmates [29]. Finally, social isolation in prisons has been identified as aggravating prisoners’ mental health problems and even suicide rates, and poor communication about the pandemic by prison staff is reported to contribute to increased stress and anxiety among inmates [30]. It should be noted that, in the pandemic, isolated prisoners often spent their time in a single cell for 22.5–24 h a day and that solitary confinement could last for months or years, for an indeterminate length of time [31].

In conclusion, and in accordance with the above, this study is considered useful in order to describe how COVID-19 has affected the levels of different mental health variables in prison inmates. In addition, as a secondary and/or derived objective, the aim is to identify predisposing and precipitating factors of the psychological consequences of the pandemic, always with the goal of improving the public health of the inmate population, especially their mental health.

The initial search strategies identified a total of 430 references, which were then screened according to the topic of this review. A total of 13 studies were finally selected (Figure 1), 8 of which were quantitative and 5 qualitative.

Three articles were found in the United States [39, 45, 46], two in Italy [40, 47], one in Ethiopia [37], one in Peru [41], one in Portugal [43], one in China [44], one in Poland [38], and three in the United Kingdom [42, 48, 49].

Anxiety was studied in 10 of the 13 articles included [37–39, 41, 43–46, 49]; depression and mood in 9 [37–41, 43–45, 49] predisposing and precipitating factors in 7 [37, 42, 43, 45, 46, 48, 49]; cognitions and complex emotions in 8 [37, 38, 40, 42, 43, 46, 47, 49]; basic emotions in 6 [38, 39, 41, 43, 47, 49]; stress in 5 [38, 39, 41, 43, 48]; differences in mental health between sexes or by sexual orientation in 4 [41, 43, 45, 48]; and suicide in 2 [41, 48].

In 10 of the 13 articles selected, the sample consisted only of prison inmates, of which 2 were women and 1 article included transgender women and non-binary persons. Two articles included recently released persons, both men and women, who were surveyed about their experience in prison.

Finally, one of the included articles also surveyed health professionals in addition to the inmates, although only the inmates’ responses were included in the differentiated results.

After applying the JBI critical appraisal tool, medium-high scores were obtained in both cross-sectional observational studies and qualitative studies.

By country, it was found that the factors studied differed across studies. Among the countries from which only one study was included, in Ethiopia [37], the factors assessed were depression, anxiety, and associated factors. In Poland [38], changes and stressors related to COVID-19 were studied, which moods and emotions were most common after the pandemic, and which of the selected factors determined the positive and negative mood of inmates. In Peru [41], levels of psychological distress among female inmates were studied. In Portugal [43], perceptions of fear of COVID-19 and the psychological impact of the pandemic were assessed in a sample of young adult male inmates. In China [44], anxiety levels were compared across inmates pre- and post-COVID-19, and the causes of changes in anxiety levels were analysed.

Among the countries where several studies were included, in Italy [40, 47], knowledge, attitudes, and preventive practices towards COVID-19 were assessed from the perspective of incarcerated persons. Also, the study explored the psychological effects of confinement during the early stages of the pandemic.

In the United States [39, 45, 46], anxiety symptoms, depressive symptoms, and self-rated health were compared before and during the pandemic. Another study described the COVID-19 risk mitigation strategies implemented and their impact on the mental health of incarcerated women. In addition, another study explored the changing conditions of confinement in the early months of the pandemic, as well as how incarcerated persons experienced the pandemic in terms of mental and physical health, safety, and trust in correctional institutions. Studies in the UK [42, 48, 49] assessed whether there had been changes in mental health as a result of the pandemic and what this had contributed to, the understanding the impact of reduced routines and other changes. Also, whether those who required mental health support felt supported by the minimum mental health services in the prison. Another study explored how a socially excluded gender, the transgender and non-binary (TGNB) minority group, experienced and coped with the stressors of the pandemic. Another study assessed how healthcare in prisons had changed in response to COVID-19 and how these changes were experienced.

Finally, only in Peru, Portugal, the US, and the UK were sex/gender differences specifically analysed.

Table 5 shows the characteristics of each of the 13 studies included in this review.

The aim of this study was to describe how COVID-19 affected the levels of different mental health variables in prison inmates. To this end, it attempted to provide a comprehensive response, not only based on key variables, but also by sequencing their structure and relationships in a factorial explanatory nexus with a global approach, within an aetiological framework, which also included differences in sex and gender.

In general, the mental health of prisoners has deteriorated over the pandemic [50]. As shown in the systematic review by Williams et al. [51], prisoners were generally found to have higher rates of SARS-CoV-2 infection, poorer clinical outcomes, and greater mental health deterioration during the pandemic than the general population.

On the one hand, based on the basic emotions found, such as fear of COVID-19, 49.7% of the variance was explained in one study [43], which can be justified as a natural response to a perceived threat. On the other hand, fear was also associated with complex emotions and cognitions that had been identified in studies, such as worry, psychological pain [47], abandonment, frustration [49], disturbing memories, and feeling isolated, left out, and insecure [42]. All of these factors can precipitate the response to fear, as they are perceived as threatening. Similarly, they may also be related to loneliness in women [41] due to the lack of attention received and a solitary confinement for the whole day, lasting for a long time [49]. In a complementary way, another study identified phenomena of alienation and worry, together with other emotions and problems when self-rating mood and emotions, indicating alexithymia [42]. These types of negative emotions have already been confirmed in other studies, namely, worry, psychological pain, fear [29], and fear of dying [49].

On the other hand, confinement may have worsened contact with relatives, levels of anxiety about contagion, and trust in professionals, both in incarcerated and recently released people, as is the case [46], as the convergence of these factors could involve cyclical relationships between psychological antecedents and consequences, given that most of the possible interactions took place with the professionals. In this sense, the inmates’ trust in the prison as an institution and in themselves in this respect was perceived as low [40]. Previous studies have already noted that low self-esteem, lack of confidence, uncertainty, a sense of survival, and hopelessness may precede the development of signs and symptoms of anxiety and depression. In addition, they may precede the cognitive triad that predicts possible depression, including negative thoughts about oneself, the world, and the future [52]. This would explain the results of the study, where worsening depression increased anxiety more than other related factors such as social support [39]. Affected self-esteem and confidence, linked to uncertainty and survival, may underlie anxiety and depression [52].

Depressive symptoms also increased during the pandemic [39, 40, 43], less likely in the absence of severe insomnia and in the presence of strong social support [37]. Also, substance use doubled the likelihood of developing depressive symptoms [37], given previous high dependence [42] perhaps exacerbated by difficulties in access, cancelled visits and releases, and fewer appointments for psychotropic medication, as reported in previous studies [53]. In other studies, inmates were found to have a mood that was categorised as average [38, 42], though rated as worse than in other groups. This may be a consequence of being immersed in uncontrollable events, generating a state of acquired helplessness. According to the Learned Helplessness theory, in the face of uncontrollable events, individuals reduce their responses because they expect them not to influence the consequences; they find it difficult to learn that the consequences depend on their responses, and with aversive consequences, behavioural and physiological disorders of anxiety and fear occur, followed by depression [54].

Regarding the anxiety found, it was often associated with distress, as observed in a study with women where 34.6% could be distressed due to the risk of contagion and to restrictions [41]. In men, they were moderately to severely distressed by fear of infection and, on the other hand, by uncertainty [43]. At a general level, distress was also detected along with anxiety, alienation, and pain, as it triggers negative emotions, anxiety, and depression [38], increases generalised anxiety [39] due to restrictions [41, 48], risk of infection [29, 38, 41, 43], as well as fear, uncertainty [43], and other negative emotions [38]. All of these factors may favour the persistence of the concern beyond the initial threat and may condition its severity, which could be related to the acquired helplessness mentioned above.

In women, and among common findings with men, it was found that the reduction in appointments made it more difficult to access mental healthcare, which was already limited [45], and anxiety was observed to increase when they were unable to have contact with deceased loved ones or infected family members and friends due to the suspension of visits [41]. Among transgender women and non-binary persons (TGNB), additional stressors such as solitary confinement for reasons of sex and gender and reduction of gender-affirming care, of social support from the TGNB community, and of specific mental healthcare were experienced [48].

Accordingly, in two studies, moderate anxiety was found in nearly the 40% of individuals [42, 43]. Generalised anxiety was identified at 51.7% in one study [40] and at 66.9% in another one, being more severe and persistent in case of disproportionate worry about infection and severe insomnia [37], all descriptive symptoms of generalised anxiety. Probably, the restrictive and overcrowded conditions in Ethiopia, where the study with the highest score was conducted, may explain these results compared to those found in Italy, also because of the different assessment tools used, GAD 2 and 7, respectively, the calculated size, and the characteristics of the sample. It is striking that in the study by Zhang et al. [44] in prisoners with a history of anxiety, this was reduced during the pandemic, and appeared in those with lower levels of education, with COVID-19, depression, or insomnia. This contrasts with the study in which symptoms were found to increase [39], and with another one where inmates noted that the decrease in routines affected their autonomous programming [42]. This could be improved, as recommended in previous studies, by facilitating prison phone calls [30, 31] and enabling appropriate distancing, clear communication, risk assessment, and mental healthcare [30].

Regarding suicide in the selected studies, and following the TGNB study by Suhomlinova et al. [48], restrictions led to higher suicidal and anxiety patterns, coinciding with results found in a study in men, which predicted insomnia, addictions, self-harm, and self-harm attempts [38]. However, in another study involving women [41], suicidal ideation was low, perhaps because the pandemic would have helped to value their life and their support role in the family. It is known that isolation can aggravate previous mental conditions [31] and increase self-harming and suicidal practices [9, 10], correlating with hopelessness [52], the main precipitant of suicide, and with the ultimate expression of acquired helplessness [55].

Among the strengths of this review are the wide coverage of types of studies, which allows for a broader view of the phenomenon under study, as well as the fact that it is a little-studied topic with a population group difficult to access. In addition, it should be added that a scientific methodology has been followed so that it can be replicated, and the quality of the studies finally selected has been assessed. Previous related reviews are based more on describing the restrictive conditions derived from the pandemic in prisons as predisposing and precipitating factors for the mental health of inmates. However, there are few studies that describe and explain such effects and relate them to those factors in a way that allows for prevention and promotion of people’s mental health in similar situations, especially among particularly vulnerable population groups.

At the study level, limitations exist due to the different methodologies used, which have not allowed a review with meta-analysis to be carried out.