Table 3.

Characteristics of studies included in review relating macroeconomic indicators and mental health outcomes, 2004–2014

Study	Setting	Study design	N Year Age	Socioeconomic determinants	Mental health outcomes	Associations/Effects	Strengths	Limitations
[55]	National population sample, USA	Cohort	26,313 18–59 y	Macroeconomic indicators State level unemployment rate	Psychological Well-being Health Related Quality of Life, Mental Health Component Summary Scale - (SF-12 Health Survey)	Increases in average state unemployment rate worsen individual’s HRQL. During hard economic times mental health decreases more than physical health	Temporal order of exposures under consideration affected all participants at the same time producing stronger causal conclusions.	The effect sizes are relatively small in magnitude. The results and recommendations should not be generalized to other cohorts.
[56]	National population sample, Britain, UK	Cohort	10,264 (1991–2008) 16–65 y	Macroeconomic indicators Local area unemployment rate	Psychological Well-being Mental health distress (GHQ-12)	Mental distress levels among unemployed people are significantly higher than among their securely employed counterparts (2.20; 95 % CI:1.98–2.42). Residence in a high-unemployment area protects against distress if unemployed.	Annual data collected over a 17 year period. Temporal order of exposures, confounders, and the outcome under consideration affected all participants at the same time producing stronger causal conclusions.	Possible bias due to selection effects threat causal inference since those with poor mental health are more likely to subsequently become unemployed.
[57]	National unemployed population sample, Sweden	Cohort	1806 (1996) 1415 (1997) 19–64 y	Macroeconomic indicators Regional unemployment and vacancy rate	Psychological Well-being Mental health distress (GHQ-12)	Significant negative effects of both unemployment rate (−0.22) and vacancy rate (−5.29) on the level of mental health among the unemployed.	Dataset surveyed in two times and shows that higher municipal vacancy rates improved mental health among the unemployed	The cross-sectional result of vacancy rates by longitudinal analysis of change gives some information on this being an effect of ecological modification and not differential health-based selection.
[59]	National population sample of working-age men, USA	Repeated cross-sectional	30,000 (1997) 35,000 (2001)	Macroeconomic indicators Local area unemployment rate	Psychological Well-being Mental health distress caseness (K6)	1 percentage point increase in the local unemployment rate leads to 3.4, 3.3, 2.5, 3.5, 3.5 and 3.8 percentage point increases in responding affirmatively to sadness, hopelessness, worthlessness, restlessness, nervousness, and feelings of effort, respectively.	Gives systematic evidence of the procyclical nature of mental health, in several clusters.	Does not take into account the lagged effect of macroeconomic conditions on mental health. Its cross-sectional design removes the possibility of causal inference.
[58]	Cross-national, 40 European and Anglo-Saxon societies	Ecological study	42,275 (2000–2004) (2005–2007)	Macroeconomic indicators Employment status, GDP, income inequality	Psychological Well-being Life satisfaction	Unemployment lowers substantially the level of life-satisfaction (−0.761 to −0.785 points lower than those employed). GDP per capita and income inequality negatively influence this association.	Large cross-national sample with attention to the macroeconomic variables of countries.	Not all contexts that affect the relationship between unemployment and life-satisfaction may be placed at the national level.
[60]	Regional population sample, Asturias, Spain	Ecological study	2000–2010	Macroeconomic indicators National unemployment rate, GDP	Psychological Well-being hospital records on incidence and prevalence of mental illness	Found a negative correlation of unemployment rate with mental health care demand. Unemployment rate was associated with a decrease in both new and prevalent mental health demand.	Regionally analysis of the association between mental health care demand and the variation of the unemployment rate and GDP.	The series studied ended in 2010, just before the intense years of the crisis. 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.
[61]	Cross-national samples of 30 countries EU, North American and Australia	Ecological study	1960–2012	Macroeconomic indicators National unemployment rate	Suicidal Behaviours National suicide rates	Unemployment rate increase has a detrimental impact on suicide, especially in country groups with the least developed unemployment protection (eastern and southern Europe).	Large cross-national sample covering a period of 52 years.	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.
[62]	Cross-national samples, 29 EU countries	Ecological study	1999–2010	Macroeconomic indicators National unemployment rate	Suicidal Behaviours National suicide rates	A 1 % increase in unemployment rates, suicide rates increase by 0.09. Male suicides increase by 0.21 (per 100,000 male inhabitants). The relationship is positive for women but not statistically significant.	Cross-national 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.
[63]	Cross-national samples of 26 EU countries	Ecological study	1970–2007	Macroeconomic indicators National unemployment rate	Suicidal Behaviours Substance-Disorders National suicide rates National deaths by alcohol abuse	1 % increase in unemployment increases suicide at 0.79 % in ages younger than 65 years (95 % CI 0 · 16–1 · 42; 60–550 potential excess deaths [mean 310]). A more than 3 % increase in unemployment increases suicide in 4.45 % at ages younger than 65 years (95 % CI 0 · 65–8 · 24; 250–3220 potential excess deaths [mean 1740]) and 28 % deaths from alcohol (12 · 30–43 · 70; 1550–5490 potential excess deaths [mean 3500]	Large cross-national sample covering a period of 37 years.	It is limited to 2007. 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.
[64]	Cross-national samples of 23 EU countries	Ecological study	2000–2010	Macroeconomic indicators National unemployment rate	Suicidal Behaviours National suicide rates	A 1 % increase in unemployment rates, suicide rates increase by 34.1 %.	Cross-national 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. The time series is limited to a decade.
[65]	Cross-national samples of 20 EU countries	Ecological study	1981–2011	Macroeconomic indicators National unemployment rate	Suicidal Behaviours National male suicide rates	Male suicide increases significantly 0.94 % with each rise in male unemployment (95 % CI: 0.51–1.36 %)	Large cross-national sample covering a period of 30 years.	Focus only in male suicide. 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.
[66]	Cross-national samples of 8 EU countries	Ecological study	2000–2010	Macroeconomic indicators Unemployment rate and GDP	Suicidal Behaviours National suicide rates	Rise on unemployment rates and decline GDP incresed suicide mortality (Germany +5.3 %, Portugal +5.2 %, Czech Republic +7.6 %, Slovakia +22.7 % and Poland +19.3 %). In low social spending countries, unemployment rate has a stronger effect on suicide.	Cross-national level trends analysis covering the period of recession.	The time series is limited to a decade. The ecological design does not allow for control of potential confounders or effect modifiers. Results should be interpreted with caution as other variables, independent of economic conditions, may be involved in the precipitation of suicide.
[67]	National Population sample, USA	Ecological study	1979–2004	Macroeconomic indicators Unemployment rate and GDP	Suicidal Behaviours National suicide rates	Higher unemployment rates for prime working-age (35–64) men and women are positively correlated with their higher suicide rates	Cross-state level trends analysis	The time series is limited to 2004, higher effects are expected afterwards. The ecological design does not allow for control of potential confounders or effect modifiers. Results should be interpreted with caution as other variables, independent of economic conditions, may be involved in the precipitation of suicide.
[68]	National Population sample, USA	Ecological study	1997–2010	Macroeconomic indicators Employment Rate	Suicidal Behaviours National Suicide Rates	Strong positive association between unemployment rates and total suicide rates over time. Strong explanation among the middle-aged suicides but cannot explain temporal variation in suicide rates among the young and elderly.	Examined trends in the total suicide rate and in the rate disaggregated by sex, age group and time period and include a number of important confounding factors in a multivariate analysis.	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.
[69]	National population sample, USA	Ecological study	1968–2008	Macroeconomic indicators State level unemployment rate	Suicidal Behaviours State level suicide rates	A 1-percentage-point increase in the state unemployment rate was associated with 0.16 (95 % CI: 0.08, 0.24) more suicide deaths per 100,000 population. The presence of generous state unemployment benefit programs buffer the impact of unemployment rates on suicide.	State fixed-effect analysis covering 1968–2008 on suicide rates	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.
[70]	National population sample, South Korea	Ecological study	2003–2011	Macroeconomic indicators National unemployment rate	Suicidal Behaviours National suicide rates	National unemployment rate was positively and significantly associated with the unemployed and employed suicide rate.	National 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. The time series is limited to 8 years.
[71]	National population sample, Greece	Ecological study	1968–2011	Macroeconomic indicators National unemployment rate, government expenditure	Suicidal Behaviours National suicide rates	Unemployment rates and suicide rates were highly correlated (0.45). 1 % increase in unemployment of males (25–44y), increases suicide rates in 4.5 %. Austerity measures and negative economic growth also significantly increase male suicide rates.	Evaluates specific effects of fiscal austerity, among other socio-economic variables, on suicide rates over recession period.	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.
[72]	National population sample, Greece	Ecological study	1991–2011	Macroeconomic indicators Unemployment rate and GDP	Suicidal Behaviours National suicide rates	Suicide rates are positively and significantly correlated with percentage of public debt in GDP and unemployment.	National level trends analysis covering the period of recession and macroeconomic fluctuations	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. The time series is limited to 10 years
[73]	National population sample, Greece	Ecological study	2000–2010	Macroeconomic indicators National unemployment rate, growth rate	Suicidal Behaviours National suicide rates	The correlations between suicidal rates and unemployment and growth rate were about zero. Found no increase in suicidality in Greece during the recession and no relationship of suicidal rates with unemployment rates or growth rate.	Evaluates specific effects of unemployment and growth rates, on suicide rates over recession period.	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. The time series is limited to a decade.
[74]	National population sample, England, UK	Ecological study	1993–2010	Macroeconomic indicators National unemployment rate	Suicidal Behaviours National suicide rates	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	Ecological study	2000–2011	Macroeconomic indicators National unemployment rate	Suicidal Behaviours National suicide rates	Unemployment rates might be associated with suicidality in the general population after 3–5 years after the onset of recession (strong positive correlation at 5 years for the general population (0.78))	National level trends analysis covering the period of recession and suggesting that there is a time lag in the increase of suicide rates.	The time series is limited to a decade 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.
[76]	National population sample, USA	Ecological study	2005–2010	Macroeconomic indicators State level foreclosure rate	Suicidal Behaviours State level suicide rates	The foreclosure crisis has likely contributed to increased suicides (b = 0.04; P < .1). the effects were strongest among the middle-aged people (46–64 years: total foreclosure rate, b = 0.21; P < .001)	State-level analysis covering 2005–2010 on suicide state rates.	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. The time series is limited to 5 years
[77]	National population sample, Italy	Ecological study	2000–2010	Macroeconomic indicators GDP per person	Suicidal Behaviours National suicide rates	The real GDP was associated with the percentage of male completed suicides due to financial problems b = 0.16, p = 0.05).	National level trends analysis covering the period of recession	Only male suicides were considered. The time series is limited to a decade. The ecological design does not allow for control of potential confounders or effect modifiers. Results should be interpreted with caution as other variables, independent of economic conditions, may be involved in the precipitation of suicide.