Open access Research Did psychotropic drug consumption increase during the 2008 financial crisis? A cross-sectional population-based study in Spain Elena Arroyo,1,2 Andrés Cabrera-León,2,3 Gemma Renart,1,2 Carme Saurina,2 Laura Serra Saurina,1,2,4,5 Antonio Daponte,2,3,6 Marc Saez1,2 To cite: Arroyo E, Cabrera- Abstract León A, Renart G, et al. Did Strengths and limitations of the study Objectives and setting Although psychotropic drugs psychotropic drug consumption are used to treat mental health disorders, little evidence ►► Microdata from the Spanish National Health Survey increase during the 2008 financial crisis? A cross- analyses the effects the 2008 economic downturn had 2006–2007 and 2011–2012. sectional population-based on psychotropic drug consumption in the case of Spain. ►► Random effects logistic regressions so the intercept study in Spain. BMJ Open We analyse these effects, considering both gender and is different for the various levels considered. 2018;9:e021440. doi:10.1136/ employment situation. ►► The Bayesian estimation provides greater flexibility bmjopen-2017-021440 Participants We used the microdata from the face-to- (a consequence of its hierarchical strategy). ►► Prepublication history and face cross-sectional population-based Spanish National ►► We use penalising complexity priors. These priors additional material for this Health Survey for two periods: 2006–2007 (n=28 954) and are invariant to re-parameterisations and have ro- paper are available online. To 2011–2012 (n=20 509). Our samples included adults (>15 bustness properties. view these files, please visit years old). ►► The cross-sectional study design has difficulty in the journal online (http://​dx.​doi.​ Methods The response variables are consumption establishing a relationship between explanatory org/​10.​1136/​bmjopen-​2018-​ variables and outcomes. (or not) of antidepressants or sedatives and the 021440). explanatory variables are the year of the survey, Received 2 January 2018 gender and employment status. Covariates are mental Revised 29 October 2018 health problems, mental health index General Health Introduction Accepted 20 November 2018 Questionnaire (GHQ-12) and self-reported health outcome Since the Great Depression in 1930, concern variables such as self-rated health, chronic diseases, about the effects of economic downturns on smoking behaviour, sleeping hours, body mass index, the health of a population has increased1–3 physical activity in the workplace, medical visits during and while there are some controversial the past year, age, region of residence (autonomous conclusions in the literature, most studies communities), educational level, marital status and social class of the reference person. Finally, we include agree that mental health problems worsen interactions between time period, gender and employment during economic crises.4–6 As the current status. We specify random effects logistic regressions and financial crisis represents the worst macro- use Bayesian methods for the inference. economic shock since the Great Depres- Results The economic crisis did not significantly change sion,7 and government-introduced austerity the probability of taking antidepressant drugs (OR=0.56, measures are only aggravating an already 95% CI 0.18 to 2.56) nor sedatives (OR=1.21, 95% CI grave situation (eg, in the third quarter of 0.26 to 5.49). In general, the probability of consuming 2007 the unemployment rate was roughly antidepressants among men and women decreases, but 8.1%, meanwhile by the same quarter of 2011 there are differences depending on employment status. it had increased to approximately 21.3%), The probability of consuming sedatives also depends on and investigating the association between © Author(s) (or their the employment status. economic downturn and mental health is employer(s)) 2018. Re-use Conclusions While the year of the financial crisis is more important than ever.8–14 permitted under CC BY-NC. No not associated with the consumption of antidepressants commercial re-use. See rights The Organization for Economic Co-oper- and permissions. Published by nor sedatives, it has widened the gap in consumption ation and Development countries all expe- BMJ. differences between men and women. Although rienced the effects of the 2008 financial For numbered affiliations see antidepressant use dropped, the difference in crisis, but the Spanish economy was one of end of article. consumption levels between men and women grew the hardest hit and its citizens had to rapidly significantly among the retired, and in the case of Correspondence to come to terms with its consequences. In this sedatives, risk of women taking sedatives increased in all Dr Marc Saez; context,15–19 Spain provides a unique oppor- groups except students. ​marc.​saez@​udg.​edu tunity to analyse the impact on financial Arroyo E, et al. BMJ Open 2018;9:e021440. doi:10.1136/bmjopen-2017-021440 1 Open access crisis has on the health of a population. Spanish data (SNHS) for two periods: 2006–2007 and 2011–2012.30 31 agrees with the association made between a financial The SNHS is a stratified tri-stage sample design represen- crisis and mental health problems.20–25 Although psycho- tative of the Spanish population. Our samples included tropic drugs are used to treat mental health disorders, individuals (>15 years old) responding to both adult and little (and contradictory) evidence analyses the effects household questionnaires (n=28 954 for 2006–2007 and the 2008 economic downturn had on psychotropic drug n=20 509 for 2011–2012). The difference between the consumption. sample sizes is because the SNHS 2006–2007 increased its Barceló et al26 concluded that the increase in psycho- initial sample size in Aragón, Baleares, Cantabria, Galicia, tropic drug consumption post 2009, is significantly higher Murcia and Navarra (six out of the 17 Spanish regions) among those most likely to be unemployed and those who to reach the specific goals of obtaining more accurate were consumers before 2009. Moreover, the use is usually estimations for the smaller geographical areas in those higher for those whose probabilities of consuming other regions (eg, NUTS4 or health areas). However, the SNHS drugs or alcohol are lower (eg, women).27 Interestingly, 2011–2012 did not include those specific goals. Colell et al28 do not find any changes in sporadic use of The response variables are consumption (or not) of hypnotics/sedatives between 2005–2007 and 2009–2011 antidepressants or sedatives, because in the SNHS ques- and while they find unemployed men have a higher prob- tions related to mental health problems these are the ability of reporting sporadic use than employed men do, only ones asked for. this is lower for unemployed women. The explanatory variables are: the year of the survey This paper analyses differences in psychotropic drug (ie, 2006–2007—prior to the economic downturn, and consumption (antidepressants and sedatives), among the 2011–2012—during the financial crisis), gender and Spanish population between 2006–2007 and 2011–2012 employment status. (ie, before and during the financial crisis) by consid- ering two of the most important determinants of health Covariates are age, region of residence (autonomous inequalities: gender and employment status.29 communities), educational level, marital status, social class of the reference person,32 mental health problems (represented by whether the individual has suffered from Methods depression, anxiety or other mental health disorders Data setting in the past 12 months), a mental health index (GHQ- We used the microdata from the face-to-face cross-sec- 12)33 and self-reported health outcome variables such as tional population-based Spanish National Health Survey self-rated health, chronic diseases, smoking behaviour, Table 1 Descriptive statistics* Period 2006–2007 (reference) Period 2011–2012 Use of Use of antidepressants Use of antidepressants Use of Total (%)† (%)‡ sedatives (%)‡ Total (%)b (%)‡ sedatives (%)‡ Variables (n=28 954) (n=2089) (n=3904) (n=20 509) (n=1099) (n=2787) Total 100 5.8§ 10.7¶ 100 4.3§ 11.1¶ Sociodemographic variables Sex  Male (reference) 49.1 3.3 6.7 48.8 1.6 6.7  Female 50.9 8.2 14.5 51.2 6.7 15.3 Current activity  Working (reference) 48.2 3.1 6 44.2 1.9 5.2  Unemployed short-term 3.7 6.1 6.5 6.1 2.6 6.3  Unemployed long-term 2 11.2 15.6 6.7 5 11.4 Retired 24 11.7 23.9 23.1 8 23.1 Studying 4.6 0.9 2.3 6.3 0.3 1.5 Home maker 17.4 10.2 18.9 13.5 9.3 21.1 *Weighted distribution of the sample. †For the categorical variables, data are % in the category. The 2006–2007 survey increased its sample size in some regions to obtain more accurate estimations for smaller geographical areas (eg, NUTS4 or health areas). ‡For the categorical variables, data are % of individuals using drugs for each case. §P<0.01. ¶P=0.735. 2 Arroyo E, et al. BMJ Open 2018;9:e021440. doi:10.1136/bmjopen-2017-021440 Open access Table 2 Mixed logistic estimations of the use of antidepressants and sedatives* Antidepressants Sedatives Exp (β) (95% CI) P value Exp (β) (95% CI) P value Year (2006–2007)  2011–2012 0.56 (0.18 to 2.53) P=0.205 1.21 (0.26 to 5.49) P=0.393 Sex (male)  Female 1.38 (1.19 to 1.61) P<0.01 1.48 (1.33 to 1.65) P<0.01 Current activity (working)  Unemployed short-term 1.14 (0.81 to 1.60) P=0.222 0.71 (0.53 to 0.96) P=0.134  Unemployed long-term 1.58 (1.09 to 2.28) P<0.01 1.25 (0.92 to 1.70) P=0.074  Retired 1.44 (1.21 to 1.71) P<0.01 1.43 (1.25 to 1.63) P<0.01  Studying 0.45 (0.23 to 0.89) P=0.010 0.59 (0.38 to 0.90) P<0.01  Home maker 1.16 (0.98 to 1.37) P=0.041 1.35 (1.19 to 1.53) P<0.01 Interactions†  Sex*year 1.76 (1.25 to 2.47) P<0.01 1.03 (0.84 to 1.27) P=0.385  Activity*year (1) 0.43 (0.15 to 1.19) P=0.052 0.91 (0.51 to 1.62) P=0.376  Activity*year (2) 0.48 (0.24 to 0.95) P=0.017 0.72 (0.45 to 1.15) P=0.084  Activity*year (3) 0.63 (0.43 to 0.95) P=0.013 0.78 (0.63 to 0.97) P=0.011  Activity*year (4) 0.82 (0.10 to 6.83) P=0.427 0.76 (0.29 to 1.95) P=0.283  Activity*year (5) 0.47 (0.10 to 2.09) P=0.159 0.57 (0.28 to 1.17) P=0.064  Activity*year*sex (1) 1.64 (0.55 to 4.90) P=0.186 1.34 (0.72 to 2.50) P=0.179  Activity*year*sex (2) 1.30 (0.65 to 2.60) P=0.229 1.46 (0.93 to 2.29) P=0.052  Activity*year*sex (3) 1.56 (1.01 to 2.42) P=0.023 1.27 (0.99 to 1.63) P=0.029  Activity*year*sex (4) 0.67 (0.06 to 7.94) P=0.377 1.02 (0.36 to 2.93) P=0.483  Activity*year*sex (5) 2.43 (0.54 to 10.96) P=0.124 1.83 (0.88 to 3.78) P=0.052 *Global statistical results of variables of interest from the model adjusted for individual characteristics: sociodemographic variables (age, region of residence, level of education, marital status and social class) as well as subjective health variables (depression, mental health index, self-assessed health, chronic diseases, smoking, hours of sleep, body mass index, physical activity and medical visits). †In table 1, the variables of reference are shown. sleeping hours, body mass index,34 physical activity in the year is equivalent to including a non-linear trend in the workplace and medical visits during the past year. For model. further information, see the appendices material (supple- For random effects at the autonomous community mentary appendix A). level, we assume that they are identical and independent Finally, we include interactions between time period of Gaussian random variables with constant variance.35 36 and gender, time period and employment status and Random effects for year and interaction are modelled by between gender, time period and employment status. a random walk of order 1 (ie, independent increments) for the Gaussian random effects vector.37 Patient and public involvement As the Bayesian estimation provides greater flexibility Patients and or public were not involved in this research. (a consequence of its hierarchical strategy), we chose to Statistical methods do the analyses using a Bayesian framework. In summary, We specify random effects logistic regressions. In mixed the initial uncertainty about the effect measures (ie, model terminology, we allow the intercept to be random ORs) and on extent of their variation, was first expressed effect so it will be different for the various levels consid- through prior distributions. Next, we combined prior ered, that is, year, autonomous community and the distributions with the so-called likelihood (ie, the current interaction between year and autonomous community. data), to obtain posterior distribution for the quantities We allowed heterogeneity at the level of autonomous of interest (again, ORs) and then we summarised the community, thus capturing (unobserved) variables not posterior distributions by point estimates and credible already included in the model which could explain the intervals (analogous to the classical CIs. A small modifi- probability of consuming psychotropic drugs. With the cation in the standard priors (eg, increasing their preci- interaction with year, we allowed for this heterogeneity sion) can imply very different inferences. That is why we to vary along time because allowing random effects for need robust priors such as penalising complexity priors, Arroyo E, et al. BMJ Open 2018;9:e021440. doi:10.1136/bmjopen-2017-021440 3 Open access Figure 1 The risk of consuming antidepressant drugs in 2011–2012 compared with 2006–2007. which are invariant to re-parameterisations and always Antidepressant consumption provide the same inferences.38 One of the advantages Table 2 shows that the year of the survey (which indicates of Bayesian analyses (with respect to the classical (or the period of the economic downturn) is not significant, frequentist) analyses) is that it is the most suitable for indicating that the financial crisis did not significantly accounting for model uncertainty, both in the parameters change the probability of taking antidepressant drugs and in the specification of the models. Furthermore, only (OR=0.56, 95% CI 0.18 to 2.53). Using the results in with Bayesian approach is it possible to model variability table 2, we compute the OR by considering the interac- with relatively sparse data and it is also easier to specify tions of the variables of interest (see figure 1). Then, we more complex scenarios. Here we followed the integrated observe that, in general, the probability of consuming nested laplace approximation (INLA) approach,39 within antidepressants among men and women decreases and is a (pure) Bayesian framework. more marked in men, but differs depending on employ- The sampling weights (corrected for non-response ment status (above all for short-term unemployed). For and improved through ratio and calibration techniques) retired individuals, we find significant differences because, were used in the population estimations. All analyses are although women’s risk of consuming psychotropic drugs carried out with the free software R,40 available through does not change between 2006 and 2011, the risk for men the INLA library.35 39 drops dramatically. However, for the student group, risk of consumption reduces equally for both genders and so we cannot observe any differences between them in 2011. Results In the home maker's group (although 96.6% are women) Table 1 shows the descriptive characteristics of the two this is where the biggest differences between men and samples (SNHS 2006–2007 and SNHS 2011–2012) women can be found because it is the only group where according to the consumption of antidepressants and the risk of consuming antidepressants increases for sedatives for the variables of interest (see online supple- women but drops sharply for men. mentary appendix B of the appendices material avail- able online for more in-depth information). In the total Sedative consumption sample in SNHS 2006–2007 (49.1% men), 5.8% take Table 2 shows that the year of the survey is not statisti- antidepressants and 10.7% sedatives. In SNHS 2011–2012 cally significant and that the crisis has not significantly (48.8% men), 4.3% consume antidepressants and 11.1% changed the likelihood of taking sedatives (OR=1.21, 95% sedatives. CI 0.26 to 5.49). Using the results in table 2, we compute Table 2 shows the results of the mixed models of the the OR by considering the interactions of the variables variables of interest for antidepressants and sedatives. of interest (see figure 2). Again, we observe larger differ- The results from the mixed logistic estimations without ences between women and men when employment status any year interactions concerning adult’s mental health and 2011–2012 are considered. In the case of individuals are available on request. Also, in this case, the adjusted in short-term unemployment, both men and women OR of the year of the survey was not significant for antide- show that between 2006–2007 and 2011–2012 there was pressant consumption or sedative use. an increase in the risk of using sedatives, however, this 4 Arroyo E, et al. BMJ Open 2018;9:e021440. doi:10.1136/bmjopen-2017-021440 Open access Figure 2 The risk of consuming sedative drugs in 2011–2012 compared with 2006–2007. increase is higher for women than men. Nevertheless, sphere and, furthermore, the crisis has had an impact on although this group is where we find the biggest increase the informal sectors of the economy which are occupied in the risk for women, it also contains the only status mostly by women.43 group where there is an increase in risk for men as well. Cost and financial stress may reduce access to health For the long-term unemployed, however, the differences and pharmaceutical services, as vulnerable populations between both groups widen between 2006–2007 and or those who are financially better off increased the use 2011–2012, that is, women increase their risk of using of psychotropic drugs. Regarding the access to health sedatives, whereas men decrease that risk. The trend services, the waiting time is found an important barrier for for retired individuals and home makers is not unlike those who are limited in their daily activities by a chronic the long-term unemployed, whereas for students, both physical or mental health problem or disability, and in groups reduce the risk of consuming sedatives between Spain the waiting time has sharply increased during the 2006–2007 and 2011–2012 and no changes are observed financial crisis.44 in the differences between the two groups in 2011. Like our study, Colell et al,28 did not find significant The results from the mixed logistic estimations of the changes in the sporadic use of hypnotics/sedatives for covariates are provided in the appendices material (see men or women (regardless of employment status) before online supplementary appendix C). and during a financial crisis either, nor did Nicieza-García et al45 confirm that the rise in psychotropic medication Discussion consumption in the region of Asturias was a result of the While the current financial crisis has not statistically financial crisis. However, our results differ from other changed the risk of using psychotropic drugs, our results similar studies. For example, in the case of Pérez-Romero do indicate that women in situations of long-term unem- et al,46 using SNHS data they establish that for both men ployment, retirement or home-making, seem to be at and women the use of these drugs increased during the greater risk than men of using psychotropic drugs during financial crisis. However, they analyse antidepressant and this financial crisis. Results are adjusted by mental health sedative consumption together and only for employees. status, presence of relevant mental health disorders and To explain the differences found in the literature, it is other health conditions associated with mental health important to consider that our analysis is the first to treat status. antidepressants and sedatives separately. While the unex- Some possible explanations are that during financial pected decrease in antidepressant use could be because crisis women intensify their work to compensate for the of a misclassification problem or changes in diagnosis losses in household income41 and their mental health and treatment practices, the existing literature can also deteriorates. In addition, it has been shown that some of be misleading because, unlike our study, the two drugs the most serious and common situations created by the are not treated separately. Furthermore, while our results crisis in question, such as evictions, affect women's mental also show that the financial crisis is not associated with health more than men's.42 Likewise, women in Spain psychotropic drug use, we must consider that we are occupy positions inferior to men in the socio-labour using data reported by individuals and so our response Arroyo E, et al. BMJ Open 2018;9:e021440. doi:10.1136/bmjopen-2017-021440 5 Open access variables may be influenced by misreporting (or under-re- performed by MS. EA, CS, LSS, GR and AC-L built the tables and figures. All authors porting) as the result of the stigma47 often associated with wrote the results and the discussion. The writing and final editing was done by EA, MS, AC-L, AD and CS. All authors reviewed and approved the manuscript. mental health issues and substance abuse. Parry et al48 Funding This work was partly funded by the CIBER of Epidemiology and Public studied the validity of responses from those using psycho- Health (CIBERESP) through the strategic subprogram ‘Crisis and Health’, by the tropic drugs and observed that about 29% of males and Catalan Government Project, AGAUR, ‘Compositional and Spatial Analysis’ (COSDA), 18% of females give invalid responses. They attribute the 2014SGR551, and by the University of Girona through the Research Grant to under-reported answers among men (compared with the Improve the Scientific Productivity of the Research Groups of the University of Girona 2016-2018 (MPCUdG2016/69). The first author carried out this research higher rates of consumption by females, especially in the with the support of the Short Term Grant Abroad for PhD European, CIBER of case of sedatives) to the negative stigma men associate Epidemiology and Public Health (CIBERESP), Spain, the support of the Mobility with these drugs. Fellowship among intra-groups of CIBERESP, CIBER of Epidemiology and Public Our study has some limitations. Even though we use two Health (CIBERESP), Spain, and was also beneficiary of the Grant for Pre-doctoral Studies from the University of Girona (BR 4/2013), Girona, Spain. large samples of individual-level data, we use cross-sec- Competing interests None declared. tional data which does not allow us to conclude direct rela- tionships between the financial crisis and psychotropic Patient consent for publication Not required. drug use. Besides, even though self-reported measures Provenance and peer review Not commissioned; externally peer reviewed. of health are strongly correlated with mortality,49 there Data sharing statement We used Spanish Health Survey data in all cases. Public could be a reporting bias affecting individuals’ self-re- and freely accessible at: http://www.​msssi.​gob.​es/​estadEstudios/​estadisticas/​ encuestaNacional/​encuesta2006.​htm and http://www.​msssi.​gob.​es/​estadEstudios/​ ported responses. Moreover, the (seemingly) marginal estadisticas/​encuestaNacional/​encuesta2011.​htm (both in Spanish). effect the financial crisis has had on mental health prob- Open access This is an open access article distributed in accordance with the lems and psychotropic drug use could be explained by a Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which reluctance to respond to (for some) sensitive questions in permits others to distribute, remix, adapt, build upon this work non-commercially, a personal face-to-face interview. and license their derivative works on different terms, provided the original work is Sample size (more than 40 000 representative interviews properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://​creativecommons.​org/​licenses/​by-​nc/​4.​0/. in Spain before and during the financial crisis), is one of the fundamental strengths of this study and provides meaningful results to discern the effects economic reces- References sions have on mental health. Our key contribution is 1. Ruhm CJ. Are recessions good for your health? Q J Econ to have identified (using gender and unemployment 2000;115:617–50. 2. Marmot MG, Bell R. How will the financial crisis affect health? BMJ status) the groups most vulnerable to a financial crisis. 2009;338:858–60. Policy-makers must take these groups into consideration 3. Bezruchka S. The effect of economic recession on population health. (through research efforts, along with healthcare and CMAJ 2009;181:281–5. 4. Catalano R, Goldman-Mellor S, Saxton K, et al. The health effects of social interventions) when planning strategies and imple- economic decline. Annu Rev Public Health 2011;32:431–50. menting policy to mitigate health inequities. 5. Brenner MH, Mooney A. Unemployment and health in the context of economic change. Soc Sci Med 1983;17:1125–38. Future lines of research could focus on determining 6. Goldman-Mellor SJ, Saxton KB, Catalano RC. Economic contraction the role new medications are currently playing, as their and mental health: a review of the evidence, 1990–2009. International Journal of Mental Health 2010;39:6–31. development and rapid spread is quickly displacing the 7. Jenkins SP, Brandolini A, Micklewright J, et al. The great recession conventional psychotropic drug market. and the distribution of household income: Oxford University Press, 2012. 8. Uutela A. Economic crisis and mental health. Curr Opin Psychiatry Author affiliations 1 2010;23:127–30. Research Group on Statistics, Econometrics and Health (GRECS), Universitat de 9. Zivin K, Paczkowski M, Galea S. Economic downturns and Girona, Girona, Spain population mental health: research findings, gaps, challenges and 2 CIBER of Epidemiology and Public Health, CIBER of Epidemiology and Public Health priorities. Psychol Med 2011;41:1343–8. (CIBERESP), Madrid, Spain 10. McInerney M, Mellor JM, Nicholas LH. Recession depression: 3 mental health effects of the 2008 stock market crash. J Health Econ Escuela Andaluza de Salud Publica, EASP, Granada, Spain 4 2013;32:1090–104. Center for Research in Occupational Health (CiSAL), Experimental and Health 11. Gili M, García Campayo J, Roca M. [Economic crisis and mental Sciences, University Pompeu Fabra, Barcelona, Spain health. SESPAS report 2014]. Gac Sanit 2014;28:104–8. 5 Center for Research in Occupational Health (CiSAL), Experimental and Health 12. McKenzie SK, Imlach Gunasekara F, Richardson K, et al. Do changes Sciences, IMIM (Hospital del Mar Medical Research Institute) Parc de Salut Mar, in socioeconomic factors lead to changes in mental health? Findings Barcelona, Spain from three waves of a population based panel study. J Epidemiol 6 Community Health 2014;68:253–60. Observatorio de Salud y Medio Ambiente de Andalucía (OSMAN), Escuela Andaluza 13. Katikireddi SV, Niedzwiedz CL, Popham F. Trends in population de Salud Pública (EASP), Granada, Spain mental health before and after the 2008 recession: a repeat cross- sectional analysis of the 1991–2010 health surveys of England. BMJ Acknowledgements We appreciate the comments of the attendees at the Open 2012;2:5:e001790. 14. Barr B, Kinderman P, Whitehead M. Trends in mental health XXXIV Reunión Científica de la Sociedad Española de Epidemiología (SEE) 2016, inequalities in England during a period of recession, austerity and on September from 14 to 16, 2016, in Sevilla, Spain and at the Third EuHEA PhD welfare reform 2004 to 2013. Soc Sci Med 2015;147:324–31. student-supervisor and early career researcher conference, from 7 to 9 September 15. National Statistics Institute (INE). Active population survey. 2016. 2016, in Barcelona, Spain. We appreciate the comments from four anonymous Available at http://www.​ine.​es (Accessed 15 Jan 2016). reviewers, which have undoubtedly helped to improve this work. 16. Eurostat. 2016. Available at: http://​ec.​europa.​eu/​eurostat. (Accessed 30 Jan 2016). Contributors EA had the original idea for the paper. EA, MS and AC-L designed 17. de la Rica S, Rebollo-Sanz YF. Gender differentials in unemployment the study. The bibliographical search and the writing of the introduction were by ins and outs during the great recession in Spain. Fedea, Madrid: EA, AC-L, CS, GR and LSS. The choice of methods and statistical analysis was Estudios sobre la economía Española, 2015. 6 Arroyo E, et al. BMJ Open 2018;9:e021440. doi:10.1136/bmjopen-2017-021440 Open access 18. Bentolila S, Cahuc P, Dolado JJ, et al. Two‐tier labour markets in 32. Chilet-Rosell E, Álvarez-Dardet C, Domingo-Salvany A. [Use of the great recession: France Versus Spain. Econ J 2012;122:F15 Spanish proposals for measuring social class in health sciences]. 5–F187. Gac Sanit 2012;26:566–9. 19. National Statistics Institute (INE). Foreclosure Statistics. 2016. 33. Goldberg D. Mental health priorities in a primary care setting. Ann N Available at http://www.​ine.​e s (Accessed 15 Jan 2016). Y Acad Sci 1978;310:65–8. 20. Gili M, Roca M, Basu S, et al. The mental health risks of economic 34. World Health Organization. Body mass classification. 2016 http://​ crisis in Spain: evidence from primary care centres, 2006 and 2010. apps.​who.​int/​bmi/​index.​jsp?​introPage=​intro_​3.​html (Accessed 15 Ap Eur J Public Health 2013;23:103–8. 2016). 21. Urbanos-Garrido RM, Lopez-Valcarcel BG. The influence of the 35. R-INLA project. Bayesian computing with INLA. 2016 http://www.​ economic crisis on the association between unemployment r-​inla.​org/ (Accessed 25 May 2016). and health: an empirical analysis for Spain. Eur J Health Econ 36. R INLA project. Independent random noise model. 2016 http://www.​ 2015;16:175–84. math.​ntnu.​no/​inla/​r-​inla.​org/​doc/​latent/​rw1.​pdf (Accessed 25 May 22. Artazcoz L, Benach J, Borrell C, et al. Unemployment and mental 2016). health: understanding the interactions among gender, family roles, 37. R INLA project. Random walk model of order 1. 2016 http://www.​ and social class. Am J Public Health 2004;94:82–8. math.​ntnu.​no/​inla/​r-​inla.​org/​doc/​latent/​rw1.​pdf (Accessed 25 May 23. Pascual N, Rodríguez I. Effects of unemployment on self-assessed 2016). health and mental health: empirical evidence for Catalonia (2006– 38. Simpson DP, Rue H, Martins TG, et al. Penalising model component 2012). Masters Project: Masters in Health Economics and Policy, complexity: a principled. practical approach to constructing priors. Barcelona Graduate School of Economics, 2013. arxiv 2015;1403. 24. Bacigalupe A, Esnaola S, Martín U. The impact of the Great 39. Rue H, Martino S, Chopin N. Approximate bayesian inference Recession on mental health and its inequalities: the case of for latent gaussian models by using integrated nested laplace a Southern European region, 1997-2013. Int J Equity Health approximations. Journal of the Royal Statistical Society: Series B 2016;15:17. 2009;71:319–92. 25. Bartoll X, Palència L, Malmusi D, et al. The evolution of mental 40. R Core Team.: R: A language and environment for statistical health in Spain during the economic crisis. Eur J Public Health computing. R Vienna, Austria: Foundation for Statistical Computing, 2014;24:415–8. 2015. Available at. (Accessed 25 May 2016). 26. Barceló MA, Coll-Negre M, Coll-de-Tuero G, et al. Effects of 41. Floro MS. Economic restructuring, gender and the allocation of time. the financial crisis on psychotropic drug consumption in a World Dev 1995;23:1913–29. cohort from a semi-urban region in catalonia, Spain. PLoS One 42. Bolívar Muñoz J, Bernal Solano M, Mateo Rodríguez I, et al. 2016;11:e0148594. [The health of adults undergoing an eviction process]. Gac Sanit 27. Mellinger GD, Balter MB, Manheimer DI. Patterns of 2016;30:4–10. psychotherapeutic drug use among adults in San Francisco. Arch 43. Women and men in Spain in 2010. Instituto Nacional de Estadística. Gen Psychiatry 1971;25:385–94. 2010. 28. Colell E, Sánchez-Niubò A, Delclos GL, et al. Economic crisis and 44. Healthcare Services. Impacts of the crisis on access to healthcare changes in drug use in the Spanish economically active population. services in the EU. Eurofound, Dublin, 2013. Addiction 2015;110:1129–37. 45. Nicieza-García ML, Alonso-Lorenzo JC, Suárez-Gil P, et al. Efecto de 29. Catalan Health System Observatory. Economic inequalities in la crisis económica sobre el consumo de psicofármacos en Asturias. health and the utilization of public health services in the population Gac Sanit 2016;30:464–7. of Catalonia [in Catalan]. Barcelona: Agency for Health Quality and 46. Pérez-Romero S, Gascón-Cánovas JJ, de la Cruz-Sánchez E, et al. Assessment of Catalonia (AQuAS): Department of Health. Catalan Recesión económica (2006-2012) y cambios en el estado de salud Government, 2017. de la población española. Salud Pública de México 2016;58:41–8. 30. Ministry of Health, Social Services and Equity: National Health 47. Corrigan PW, Watson AC. Understanding the impact of stigma on Survey in Spain 2006 and 2011. Householder questionnaire and people with mental illness. World Psychiatry 2002;1:15–20. adult questionnaire. 2013 http://www.​msssi.​gob.​es/​estadEstudios/​ 48. Parry HJ, Balter MB, Cisin IH. Primary levels of underreporting estadisticas/​encuestaNacional/​encuesta2006.​htm psychotropic drug use. Public Opin Q 1970;34:582–92. 31. (EHIS). European Health Interview Survey (EHIS). 2016 http://​ec.​ 49. Kaplan GA, Camacho T. Perceived health and mortality: a nine-year europa.​eu/​eurostat/​c/​portal/​layout?​p_​l_​id=​203688&​p_​v_​l_​s_​g_​id=0 follow-up of the human population laboratory cohort. Am J Epidemiol (Accessed 15 Apr 2016). 1983;117:292–304. Arroyo E, et al. BMJ Open 2018;9:e021440. doi:10.1136/bmjopen-2017-021440 7