Mental health outcomes in times of economic recession: a systematic literature review

Background Countries in recession experience high unemployment rates and a decline in living conditions, which, it has been suggested, negatively influences their populations’ health. The present review examines the recent evidence of the possible association between economic recessions and mental health outcomes. Methods Literature review of records identified through Medline, PsycINFO, SciELO, and EBSCO Host. Only original research papers, published between 2004 and 2014, peer-reviewed, non-qualitative research, and reporting on associations between economic factors and proxies of mental health were considered. Results One-hundred-one papers met the inclusion criteria. The evidence was consistent that economic recessions and mediators such as unemployment, income decline, and unmanageable debts are significantly associated with poor mental wellbeing, increased rates of common mental disorders, substance-related disorders, and suicidal behaviours. Conclusion On the basis of a thorough analysis of the selected investigations, we conclude that periods of economic recession are possibly associated with a higher prevalence of mental health problems, including common mental disorders, substance disorders, and ultimately suicidal behaviour. Most of the research is based on cross-sectional studies, which seriously limits causality inferences. Conclusions are summarised, taking into account international policy recommendations concerning the cost-effective measures that can possibly reduce the occurrence of negative mental health outcomes in populations during periods of economic recession.


Background
Economic recessions have been estimated to significantly affect the population's health and wellbeing, which applies, in particular, to vulnerable groups of people [1][2][3][4][5]. In countries that have been hardest hit by the latest recession, which started in 2007, the living and working conditions have substantially worsened [6]. Work became more precarious and unemployment rates increased as a result of the slowdown in global growth and consequent deterioration of the labour markets [7]. For instance, almost half of the citizens of Europe reported knowing someone who had lost his/her job as a direct result of the crisis [8]. Rates of involuntary part-time employment have also been rising since the beginning of the recession [9]. Overall, people are more fearful about losing their employment [8] since competition for jobs is rising and finding work quickly is perceived as unlikely. It is estimated that labour markets will take time to improve even though there are prospects for economic recovery [6]. Levels of poverty and social exclusion have worsened, mainly in groups that were already at risk [10]. During this recession, more people have been reporting being at risk of being unable to cope with unexpected expenses and even facing difficulties with paying ordinary bills or buying food over the coming year [8].
It is known that the health of populations is shaped by the socioeconomic context, welfare systems, labour markets, public policies, and demographic characteristics of countries [4]. There are strong reasons to believe that changes in these key determinants may be reflected in the mental wellbeing of populations [11]. Therefore, mental health should be a health area regarded as possibly vulnerable during a recession [12], especially if mental disorders were already highly prevalent even before the crisis began [13]. Nonethless, some authors have argued that associations between contracting economies and levels of well-being may show mixed patterns of both positive and negative impacts [14]. However, this current recession is likely to aggravate and boost mental health problems through growing socioeconomic risk factors such as unemployment, financial strain, debts, and job-related problems [3]. People facing these major life changes are more prone to mental ill-health [15][16][17][18]. It has also been theorised that economic pressure and unemployment have a devastating impact on families, in particular children, since the family is the most important context for their healthy development [19,20].
This paper intends to cover the main sources and types of recent evidence on populations' mental health outcomes in times of economic recession. Specifically to summarize the mental health outcomes and the socioeconomic determinants most frequently addressed by the literature on economic recessions, which groups of people seem to be the most vulnerable, and to determine possible research needs.

Search strategy and definition of terms
A systematic search was performed in Medline, PsycINFO, SciELO, and EBSCO Host. The keywords used for reference tracing were derived from Medical Subject Headings (MeSH) in combination with key terms used in other reviews [2,3,5,21,22].
Two sets of keywords were then used and combined: 1) Recession and socioeconomic terms -"Economic recession" OR "Financial crisis" OR "Recession" OR "Unemployment" OR "Socioeconomic deprivation"; combined with 2) Mental health outcomes -"Mental health" OR "Mental disorders" OR "Suicide" OR "Substance-Related Disorders".
Regarding the recession and socioeconomic terms, besides the logical use of the words "economic recession" and its synonyms, the word "unemployment" was used as it is a widely recognised countercyclical variable, i.e. a phenomenon that increases in recessions [23]. The term "socioeconomic deprivation" was used, on one hand because it is a broad term that includes the characteristics of both social and economic vulnerability that are expected to increase in periods of recession [24], and on the other hand, because of its indisputable negative effect on health [5,11]. Concerning the mental health outcomes, in addition to "mental health", the term "mental disorders" was used because it is a MeSH term that encompasses "all psychiatric illness or diseases manifested by breakdowns in the adaptational process expressed primarily as abnormalities of thought, feeling, and behaviour producing either distress or impairment of function". Although "mental disorders" is a broad term, it does not include suicide, which is known to be associated with major mental health problems [25]. Therefore a specific keyword for that was entered. The term "substance-related disorders" was also included, because using the broad term "mental disorders" did not retrieve papers with clear specific results and this was a MeSH term used in other reference works [26].

Eligibility criteria and data extraction
Two reviewers independently screened all the titles and abstracts. The final articles in this review are a consensual reflection of both reviewers. They only considered studies for inclusion that were original research papers, peer-reviewed, published between 2004 and 2014, written in English or Portuguese, and showing associated results between recession or socioeconomic terms and mental health outcomes. Moreover, the authors excluded all duplicates, small sample investigations (<1000 except for casecontrol studies) for precision reasons and strength of effect sizes [27], research that did not employ validated instruments or used an inappropriate methodology regarding the associations under consideration (e.g. ambiguous variables under study, poor construct validity, and drawing of conclusions without statistical support), and qualitative research. The data extraction from each study was based on the following variables: the setting and country, the sample (N and age), the years examined, the mental health outcome(s) and the socioeconomic determinant(s), and the key associations or effects found. In general, we found significant disparities in the methods, data collection procedures, analyses, and contexts of existing studies that complicated direct comparison of results among studies. Because of this diversity of metrics and outcome variables, it was impossible to apply statistical criteria to the studies and for that reason it was not appropriate to perform meta-analysis of the results.

Mental health outcomes associated with economic recessions
We organised the main results by mental health outcomes and the socioeconomic determinants most frequently addressed by the literature, based on the quality of study design (cohort, case-control, cross-sectional and ecological). The mental health outcomes were clustered into four main groups: 1) psychological wellbeing (measured by continuous variables of mental health distress, selfrated health, and wellbeing or quality of life variables); 2) common mental disorders (assessed by caseness for depression, anxiety, and somatoform disorders); 3) problems related to substance-related disorders (reports on smoking, patterns of alcohol consumption, drug use, and substancerelated harms), and 4) reports on suicidal behaviours (suicide mortality, parasuicidal behaviour, suicidal ideation, and attempts). The socioeconomic determinants retrieved were clustered into three groups by: 1) inter-time variables (preand post-economic recession changes); 2) macroeconomic indicators (rates of unemployment, GDP, home foreclosure rates), and 3) individual-level indicators (employment status, psychosocial job quality and security, household income, perceived financial strain or security, perceived economy/recession stress, deprivation, indebtedness, housing payment problems, socioeconomic status).

Study selection results
At the beginning, 20,502 studies were identified and were first filtered on the basis of being original peer-reviewed research papers and published between 2004 and 2014. The remaining 7351 papers were then screened by two independent reviewers through their titles and abstracts, and the subsequent filtering was performed on the basis of the following inclusion criteria: not being duplicates, written in English or Portuguese, were non-qualitative research, and reported associations between recession or socioeconomic terms and mental health outcomes. The full texts of 183 studies were then analysed. The number of papers excluded was a consequence of combinations of search keywords such as "crisis" and "mental health" or "suicide" that resulted in papers not relevant to the study objective. From the analysis of the 183 full texts, the investigators further excluded studies that used an inappropriate methodology regarding the associations between economic recession and mental health outcomes, including non-validated instruments, or used small samples (<1000, with the exception of case-control studies). After the previously described multistep selection method, 101 papers were used for the present review. Figure 1 shows the progress of selection for the study and the number of articles at each selection stage. Table 1 summarizes the main features of the retrieved studies. Two studies used case-control design, 30 were cohort studies, 40 were cross-sectional studies or repeated cross-sectional studies, one used a mixed cross-sectional and case-control design, and 28 were ecological studies.

Samples and geographical allocation
More than half (66) of the total of 101 studies used national population samples. Out of these 66, 58 were general population samples and the rest were focusing on specific populations. Two used working populations, two used unemployed populations, two used samples of adolescents, one used a sample of patients attending primary care centres, and one used a sample of older adults. Furthermore, 16 studies used crossnational population samples, of which 11 were general population samples and 5 focused only specific populations. Two studies were samples of children and adolescents; one was a sample of working population; one used a sample of patients attending primary care centres, and one used samples of older adults. An additional 19 studies used community or regional samples. Out of these 19, 10 studies used general population, 3 used samples of workers, 1 used a sample of unemployed adults, 2 used samples of adolescents, 1 used a sample of children and parents, 1 used a sample of older adults, and 1 used a sample of hospital patients.
In terms of geographical allocation, 61 studies were conducted in Europe (7 studies were cross-European, 2 studies used samples from both Greece and Poland, and both Denmark and Sweden, 15 used samples from the UK, 8 from Greece, 7 from Sweden, 7 from Spain, 4 from Italy, 3 from Finland, 2 from Iceland, and 1 population sample each from France, Germany, Hungary, Portugal, Slovakia, and Slovenia). Eighteen studies have North American population samples (15 from USA and 3 from Canada); 7 studies were from Australasia countries (five Australian and two from New Zealand). Two studies were South American, one from Argentina and Brazil, and six studies were from Asian countries (three from South Korea, two from Japan, and one study from Hong Kong). In addition, there were seven multicentre studies that used cross continent population samples from various countries.

Pre and post-economic recession changes in psychological wellbeing
Studies comparing the data to pre-recession periods show a consistent aggravation of the mental health status of the populations involved (Table 2).
A longitudinal study from Greece showed that mental health and self-rated health were negatively affected by unemployment during the economic recession (2008-2013), especially among unemployed individuals [28]. A similar result was found in Italy, where the inequalities regarding self-reported health between workers and unemployed individuals were amplified after the onset of the recession [29].
Repeated cross-sectional studies from Greece also showed that the recession period was associated with a significant deterioration of the population's self-reported health and increased odds of poor health when compared to control populations [30,31]. English and Spanish repeated cross-sectional studies indicated that the prevalence of psychological distress significantly increased during the recession period, with a greater impact on men compared to women [32,33]. However, women also reported increased mental distress during the recession, according to a repeated cross-sectional survey from Sweden [34]. In Japan, comparable surveys before and after the economic recession period showed reports of an increase in poor health across people of all socioeconomic ranks [35].
Pre and post-economic recession changes in rates of common mental disorders Regarding morbidity rates for common mental disorders, longitudinal data from Iceland presented aggravated stress levels among the population, though only significant for women and especially if unemployed [36].
Greek comparable data from before and after the recession exhibited a statistically significant rise in the prevalence of depression [37,38]. In Spain, evidence displayed a risk of suffering from depression during a recession that was almost three times higher than before [39]. Similar evidence was also found in Canada and Hong Kong [40,41]. The same Spanish study also showed an increase in the         Examined trends in the total suicide rate before and after the onset of the recession and in relation to unemployment rates.
Because of its ecological nature, the quality of the data is not assessable and no implications can be drawn regarding causality. The results should be interpreted with caution as other variables, independent of economic conditions, may be involved in the precipitation of suicide. prevalence of anxiety disorders [39]. Nevertheless, no changes in the prevalence of anxiety were found in the Canadian working population sample [40].
Pre and post-economic recession changes in substancerelated disorders A cohort study from the USA stated that the overall prevalence of any alcohol use significantly declined during the recession but, conversely, binge-drinking became more frequent [42]. Spanish repeated cross-sectional evidence shows that this recession may have triggered alcoholrelated disorders, since a noteworthy rise of 4.6 % in the abuse of alcohol and dependence on it was observed [39]. Furthermore, available data from Argentina also revealed that people may tend to increase their intake of lowerquality alcohol, which is known to pose additional threats to health [43].

Pre and post-economic recession changes in suicidal behaviours
Longitudinal evidence during the Swedish recession reported a post-recessionary increase in suicide rates among unemployed men [44] suggesting possible delayed effects of the recession on suicidal behaviours. Similarly, Greek cross-sectional data, from before and after the onset of the recession, indicated that the rate of suicidal ideation increased significantly in men [45]. Moreover, in South Korea data from comparable surveys also showed that income gradient-related suicide behaviour was found to have increased in the years after the recession period [46]. Several ecological studies from Spain have reported a substantial growth of suicidal ideation and suicide attempts [47][48][49]. In Italy, an ecological study showed an evident increase in suicide rates among Italian men involved in the labour force after 2007 [50]. Similar evidence comes from the UK, where time-trends analysis displayed an increase in the suicide rate, especially among working-age men [51,52]. In an ecological analysis, Reeves et al. found that most of the European countries experienced a significant rise (6.5 %) in suicide rates after the onset of the recession in 2009 [53]. The same was found in Canada (a rise of 4.5 %) and in the USA (a rise of 4.8 %) [54].

Macroeconomic indicators associated with mental health outcomes
Data from cohort studies focusing on unemployment rates ( Table 3) have shown that high unemployment rates are linked to individuals' worsened mental wellbeing and higher mental distress levels [55][56][57]. Similar evidence was found in a cross-sectional study from the USA [58,59].
A large cross-national ecological study has shown that rises in unemployment among the population are also associated with lower life satisfaction levels, especially among unemployed individuals [58]. Despite this evidence, however, there is a recent ecological study from Spain that suggests that rises in unemployment rates were associated with a decrease in the demand for mental healthcare [60].
Recent ecological studies provide evidence of a strong positive association between unemployment rates and suicidal behaviour. A study that covered 30 countries (European, North American, and Australia) demonstrates that increases in the unemployment rate related to the recession period have a negative impact on suicide, especially in those Eastern and Southern European countries with the least developed social protection systems [61]. Similar evidence has been found in studies focusing solely on European countries [62][63][64][65][66] and studies performed in the USA [67][68][69], as well as in South Korea [70]. In Greece two studies also found strong correlations between unemployment rates and suicide [71,72], though there is one study reporting no correlation and no increase in suicide behaviours [73]. In England, these correlations were only statistically significant at the regional level [74], and in Hungary the correlations were only strong 3 to 5 years after the onset of the recession [75]. Using other macroeconomic indicators, Houle et al. found that the state-level foreclosure rate also correlated to suicide rates in the USA [76] and an Italian study found that the decrease in GDP per person was associated with male suicide [77].

Individual-level indicators associated with mental health outcomes Unemployment
Studies demonstrate that people who lose their job during a recession are more vulnerable to the economic recession. For instance, during the Japanese economic crisis unemployed people were twice as likely to report poor health compared to controls [35]. In Hong Kong the prevalence of a major depression episode increased among the unemployed [41]. In addition, research dealing with the European recession shows a significantly higher risk of depression and mental distress among this group of people compared to the general population [28,29,33,39], although Icelandic and Swedish data showed increased stress levels only for unemployed women [34,36]. When variations in macroeconomic indicators are considered, the unemployed were also more vulnerable to mental health problems and suicidal behaviour [56,57,70].
Several individual-level cohort studies (Table 4) found an association between job loss and poor mental health     The associations between unemployment rate and suicide rates were only statistically significant associations at regional level between 2008 and 2010. National and regional level trends analysis covering the period of recession.
Because of its ecological nature, the quality of data is not assessable and no implications on causality can be drawn. Results should be interpreted with caution as other variables, independent of economic conditions, may be involved in the precipitation of suicide.
[75] National population sample, Hungary         Being unemployed or having work disability were also associated with suicidal ideation in people aged 18-49. The prevalence of suicidal ideation and attempts found in this study is similar to the one found ten years ago, before the economic crisis Representative sample of the national population in Spain during the economic recession. The data was collected in the same way as in the ESEMED study, making it possible to compare current figures with the prevalence found before the crisis.
No causal inference can be made due to the cross-sectional nature of the study.Reported suicidal ideation and suicide attempts can be susceptible to recall bias or to reluctance on the part of respondents to disclose such a sensitive information. outcomes and low life satisfaction, and suggest that this can be both a risk factor for being unemployed and its consequence [78][79][80][81][82][83][84][85][86]. In addition, in cross-sectional studies unemployment has also been associated with psychosomatic symptoms and psychological distress [87][88][89][90]. Two large cohort studies showed that unemployment was associated with depressive symptoms [91,92]. The risk of common mental disorders such as depression and anxiety was also found to be significantly greater in unemployed individuals in several cross-sectional studies [93][94][95][96].
Furthermore, a case-control study from Finland found that being unemployed was a heavy predictor of risky behaviours such as driving under the influence of drugs [97]. Cross-sectional data from the USA and Germany also discovered that unemployment was significantly related to alcohol and drug use [98][99][100]. Additionally, alcohol-attributable deaths rate were determined to be higher among the unemployed population during recession, says an ecological study from South Korea [101].
Suicidal behaviours were also linked to unemployment in several studies. A Canadian case control study found that unemployed individuals have a significantly increased risk of parasuicidal behaviour compared to their matched controls [102]. Likewise, in a large cross-national study, being unemployed was discovered to be a strong risk factor for suicidal ideation and attempts [103]. An Australian study also revealed that, in times of recession, unemployed males commit suicide at 4.62 times the rate of employed men and women 8.44 times more compared with employed females [104]. Also in times of recession a Spanish study states that being unemployed was found to be associated with suicidal ideation [105].

Precarious and insecure work
Working conditions affect mental health (Table 5). Finnish longitudinal data pinpointed mental distress as being stronger among precarious workers with high job insecurity [106]. Nevertheless, there is a Swedish cohort study that found no significant differences in the effects of job insecurity on health between temporary and permanent workers [107]. Cross-sectional data from during the recession in Italy, determined that job stress was significantly related to workers' mental health and fear of the crisis [108]. This was supported by British evidence of an increased risk of depression and anxiety among such employees [109,110].

Debt, deprivation, and financial hardship
Several studies found socioeconomic status and indebtedness to be related to mental health. In the USA, a cohort study, indicated increased incidence of anxiety and mood disorders, and substance use disorders were strongly associated with drops in household incomes [111]. Strong causal conclusions about this matter can also be drawn on the basis of a cohort study from New Zealand that shows a high level of association between inequalities in wealth and psychological distress, stating that people reporting low levels of wealth have three times greater distress than those reporting higher levels of wealth [112]. Longitudinal data also illustrates that housing payment problems and indebtedness have a detrimental effect on mental health [113] and on the onset of depression and anxiety [114]. Income inequality at a regional level was also significantly associated with poorer mental health in a crosssectional study completed in a community sample from Wales, UK [115]. Additionally, low socioeconomic status was related to higher rates of tobacco smoking and the use of cannabis and other illegal drugs compared to people of higher socioeconomic status in a French community-based cohort [116].
Furthermore, a cohort and a cross-sectional study from England found that people facing debt are also at higher risk of depression [110], and are twice as likely to think about suicide [117].
The previously cited studies show that during recession Greek people with serious economic difficulties had 1.33 times higher odds of developing a major depressive episode during the recession [38], and in South Korea well-off people do better in recessions in terms of the prevalence of depression, suicidal ideation, and suicide attempts [46].

Families, children, and older people
The literature also stated that families and children affected by socioeconomic factors might face a decline in their mental health (Table 6). Finnish longitudinal research shows that economic stress can lead to deterioration in children's mental health, mainly through changes in family relationships and parenting quality [118]. A large cross-national study with representative data on adolescents from 31 countries found that the countries most hit by the recession (Ireland and Portugal) faced a rise in psychological health complaints (9-17 %), and this was related to the increase in unemployment rates [119].
In fact, adolescents who perceived themselves as being socioeconomically worse off have a four-times higher likelihood of rating low life satisfaction and quality of life, claims a study from Slovenia [120]. In addition, children with unemployed parents have a higher prevalence of depression, higher rates of psychosomatic symptoms, and lower perceptions of psychological well-being [121][122][123].
Trends in a cohort of Canadian adolescents' total suiciderelated behaviour during periods of recession illustrate that the downward trends in suicidal behaviour stopped after the onset of the recession, though no increase has been reported [124]. Moreover, in the USA, repeated cross-  High neighbourhood income inequality was associated with better mental health in low-deprivation neighbourhoods (P = 0.036). Income inequality at regional level was significantly associated with poorer mental health (P = 0.012).
Uses a continuous measure of mental health symptoms.Large sampling fraction.
No data were available on individual income.Its cross-sectional design removes the possibility of causal inference.    sectional analysis before and after the onset of the crisis revealed that state-level unemployment during the year preceding the survey increased girls' rates of suicidal ideation and suicide plans, but did not affect the suicidal behaviour of boys [125]. Studies focusing on older adults report that those facing distress related to economic slowdown and rates of home foreclosure also had greater depression and anxiety symptoms in Australian cohort studies [126,127].

People with mental health problems
A cross-sectional study comparing data from 27 EU countries before and after the crisis found that individuals with mental health problems were more vulnerable to losing their employment than those without these problems. This evidence is particularly important for people already facing mental health problems because it may indicate that during a recession discriminatory attitudes towards people with chronic mental health conditions may harden, both in the job market and in society, further increasing their suffering and isolation [128].

Discussion
In general, evidence on the impact of economic crises and recessions on mental health is accruing, but comprehensive studies are lacking. Epidemiological data comparing changes in health status before and after a recession are consistent and report negative associations with mental health and increased mental health problems. However, to measure the extension and duration of these impacts and to isolate the exact causal factors appeared to be challenging. There is a preponderance of cross-sectional and ecological studies compared to cohort or case-control studies. This causes great limitations in terms of determining causality between the recession and mental health problems. Nevertheless, the repeated cross-sectional studies helped to better estimate the changes in the population's outcomes before and after the recession period.
In terms of geographical allocation, most of the research is being done in Europe and North America during the period of this review (2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014). Some of the countries hardest hit by the economic recession (Greece, Spain, and Italy) are monitoring changes in the mental health outcomes of their populations, although they are doing so mainly by using repeated cross-sectional surveys or ecological analysis. We found no specific studies from Ireland or Portugal focusing on the effects of the recession on mental health. We strongly believe that research results from these countries could contribute to a better understanding of the consequences of the recession since its impact on mental health varies greatly, depending on how austerity measures and policy responses were implemented. Additionally, there were a very limited number of studies from low and middle-income countries despite the fact that there are strong reasons to believe that these countries are likely to be heavily affected by the recession, especially because any further reductions in these countries' already weak health budgets (mental health services in particular) is likely to be very damaging. We argue that research from these countries finds substantial barriers to publication in widely accessible journals due to possibly material and financial constraints, problems of research design and statistics and thinkable difficulty in writing in English. Thus, we argue that this under-representation of research might result in limited conclusions.
Nonetheless, the studies included in this review confirm that recession periods are feasibly associated with the increased prevalence of psychological distress and common mental disorders, substance disorders, and ultimately suicidal behaviour. Despite being limited to the validity of self-reporting, the data on alcohol misuse behaviour indicates that any increase in its prevalence may be countercyclical and related to unemployment rates. We further add that recessions might result in an increased prevalence of smoking and illicit substance use since the literature indicated this may be a coping mechanism used to help deal with unemployment and economic distress [100]. However, the impact may vary according to the profile of substance users. Recreational users may be more susceptible to cuts in income, therefore reducing abuse, while others who are more dependent may actually adopt riskier patterns of substance misuse, such as injecting or binge drinking, in order to maximise the effects of the substances they have managed to purchase [129]. Further analysis of these fields is still required.
Although reports of growing suicidal ideation and attempts in countries in recession are limited to the complexity of the phenomenon, to the cultural background, and to the quality of the data sets and self-reports, which are susceptible to recall bias, it is consistent with the previous idea that suicide is more common in areas of high socioeconomic deprivation, social fragmentation, and unemployment [4]. Futhermore, a great proportion of the evidence from this review shows that unemployment, precarious work, debt, and financial deprivation are significantly associated with mental health problems. Determinants as such are well-known driving forces for widening health inequities, and put some groups of people at higher risk of suffering the impact of the economic recession. The influence of these factors on mental health has been widely recognised in the past [15,130,131]. Therefore, special attention should be given to people facing economic pressure and unemployment.
Indirect data supports the view that families and children may be disproportionally affected by recession, which is consistent with the Family Stress Model [19,20]. Many mental disorders often start in adolescence or young adulthood. Growing up in a challenging environment can put young people in a very vulnerable position [132][133][134]. Upto-date evidence shows that 27 % of young Europeans aged below 18 are at risk of poverty or social exclusion and, considering the growing number of people who are unemployed and in deprivation, these are worrying indicators [135]. Failing to protect the mental health of young people and to capitalise on their energy may indicate that we will possibly face a long-lasting loss of future adult productivity [134,136]. Unexpectedly, there is a substantial research gap on the effects of recession on families and children. A better understanding of these effects could be gained from research focusing on how job losses and economic strain affect family members.

Research and policy implications
Summarising the data from this review gives us a global perspective and allows some hypotheses to emerge that serve as a framework for future research on economic recessions and mental health outcomes: it is plausible that the actual recession increased the population's psychological distress; according to the evidence reviewed, periods of recession correlate with higher prevalence of common mental disorders, substance disorders, and ultimately suicidal behaviour; it may be possible that in order to cope with psychosocial stress people might turn to substance misuse; some key factor seem to make people more vulnerable to the effects of the recession: being unemployed, having a precarious work situation, facing debts and economic strain, and having a pre-existing mental illness; economic recession may also have a severe and long-term impact on mental health in children and young people, especially if they face stress within the family as a result of economic hardship or parental unemployment; some specific differences between countries and regions were found in this review. The authors hypothesise that this may be explained by the socioeconomic response policy to recession (the presence of unemployment benefits or social programmes) which could influence changes in the mental health outcomes of the populations; more research is needed concerning mediating factors between the determinants of a recession and mental health outcomes; more research from countries badly hit by the economic recession and from low and middle income countries is needed; the links between recession and direct effects on health seem to be very complex, and the lagged effects have not been systematically studied because of a lack of longitudinal studies and therefore a scarcity of long data series persists.
Even though the economy can shape populations' mental wellbeing, better mental health can in turn be a major contributor to economic growth [136]. Policies and costeffective measures may affect the extent of the risk factors faced by populations and the occurrence of mental health disorders during and after an economic recession. The World Health Organisation [4] has argued that the mental health effects of economic crises depend on action in five key areas: active labour market programmes family support programmes regulation of the marketing of alcoholic beverages, restrictions on their availability, and taxation provision of quality and equitable access to primary care for those people at high risk of mental health problems debt relief programmes.

Strengths and weaknesses of this study
A language bias might be present since the review was exclusively based on English and Portuguese language research reports. Nonetheless, the potential impact of studies published in other languages in this literature review may be minimal since most of the publications in widely accessible journals are in English.
Another limitation may be the literature search time framed to last 10 years. Although it is an usual procedure [137], it could have limited the inclusion of other important works. Also, given the heterogeneity of the metrics used by the studies we were unable to use quantitative meta-analytic methods and therefore were not capable of identifying statistical patterns.
As included studies have mainly cross-sectional or ecological design, there is a limited space for establishing causal inferences. This is especially important because this gives only evidence of the rough short-term mental health outcomes related to economic recession and specific socioeconomic indicators, but there is still a lack of evidence on the longer-term consequences, particularly if the number of long-term unemployed people continues to grow and social safety nets experience further cuts.
Moreover, despite the fact that most studies are showing negative associations between the recession and levels of mental health, there may be mixed patterns (positive and negative effects of the recession) that are dependent on countries' policies and responses adopted to deal with the recession [14]. Thus, the generalisability of the findings is considerably limited by the uniqueness of the welfare and health systems of each country and its response measures to the economic recession itself. The only way to ascertain whether the economic recession has increased the incidence of poor mental health is to intensify the gathering of empirical evidence from long-term cohort studies [138].
Notwithstanding limitations, the literature review gives a rough approximation of the consequences of the recession, showing an increasing number of people experiencing poor mental health and reporting common mental disorders such as depression and anxiety, substancerelated disorders, and suicidal behaviour, which corroborates with what was found in other reference works [1][2][3].

Conclusions
Quality evidence showing that economic recessions are possibly associated with negative mental health outcomes of populations is growing. This seems especially true for psychological wellbeing, common mental disorders, substance disorders, and suicidal behaviour, despite the fact that the mediation pathways are still undisclosed. There are groups of people that may be especially vulnerable to the effects of recessions: the unemployed, those in debt or facing financial difficulties, people with pre-existing mental health problems, and families with children. It is well known that mental disorders and substance use disorders make major contributions to the global burden of disease in high-income countries and constitute important public health problems. Since economic downturns may possibly exacerbate mental ill-health and suicide risk factors, it is a collective responsibility to take action and reduce these unbearable costs as far as possible. In times of economic constraints countries may want to consider balancing appropriate resources. Structural reforms and the implementation of available cost-effective measures to achieve health and high levels of wellbeing may contribute to a more productive economy and desirable societal assets.