Article Text
Abstract
Background Women are more likely than men to use mental healthcare (MHC) due to differences in the types of problems and help-seeking behaviours. The consistency of this relationship across European countries, whose MHC organisation differs substantially, is unknown.
Methods Lifetime MHC-use and the type of MHC provider were assessed in 37 289 participants from the EU-World Mental Health (EU-WMH) survey, including 10 European countries (Northern Ireland, The Netherlands, Belgium, Germany, France, Spain, Italy, Portugal, Bulgaria and Romania). Lifetime mood/anxiety disorders (DSM-IV) and severity were evaluated using the CIDI V.3.0.
Results MHC use was significantly higher for women than men in every country except for Romania (overall OR=1.80, 95% CI1.64 to 1.98), while remaining so after adjusting for socioeconomic characteristics (age, income level, employment status, education, marital status; adjusted OR=1.87, 95% CI 1.69 to 2.06) and country-level indicators (MHC provision, private household out-of-pocket expenditure, and Gender Gap Index; adjusted OR=1.89, 95% CI 1.71 to 2.08). Compared with men, women were also more likely to consult general practitioners (GP) versus specialised MHC (OR=1.32, 95% CI 1.12 to 1.56) with high between-country variability. In participants with mood disorder, the gender relationship in MHC use and type of MHC did not change. Conversely, in participants with anxiety disorder, no significant gender relationship in MHC use was observed (adjusted OR=1.21, 95% CI 0.99 to 1.47). Finally, men with severe mental health problems had a significantly higher odds of MHC use (OR=14.70) when compared with women with similar levels (OR=8.95, p for interaction=0.03) after adjusting for socioeconomic characteristics and country-level indicators.
Conclusions Women use MHC and GPs more frequently than men, yet this depends on the type and severity of mental health problems.
- Gender
- Health Behaviour
- Health Services
- Mental Health
Statistics from Altmetric.com
Introduction
Research consistently finds, whether self-reported or based on records, that women are about twice more likely than men to use health services for mental problems in Westernised countries1 ,2 This gender difference is more pronounced in outpatient, general mental health services, while smaller in more specialist ones.2 ,3 Gender differences in mental healthcare (MHC) use are often associated with gender-specific psychological profiles. Impulsive and addictive disorders, which are more frequent among men, are associated with a weaker demand for MHC than anxiety and depressive disorders, which are more frequent among women.4–6 Meanwhile, high social and self-stigmatisation among men7 could avert them away from attaining appropriate MHC8 ,9; whereas, women's higher acceptance of seeking help, compliance with treatment regimens and willingness to adopt the sick role10 could also contribute to gender differences. Additionally, the socialisation and caregiving responsibilities of women11 allow them to become more frequently aware of mental distress symptoms.12
Research on individual risk factors suggests that the common socioeconomic positions of men and women might additionally explain gender differences in MHC use. For example, persons without employment, with lower education levels and poverty are more likely to seek help from a primary healthcare provider, while highly educated and wealthier individuals tend to use more specialised care.13–15 Gender differences in MHC use could, therefore, be expected due to the fact that women generally have a lower socioeconomic position.
While the vast majority of existing research focuses on individual-level indicators, the large variation in MHC use across countries implies that certain national-level indicators might play a role even further in explaining gender differences. A number of studies have demonstrated that the degree of gender inequality in a country is directly proportional to gender differences in mental health,4 ,16 ,17 yet how this relates to MHC use specifically is largely unknown. Other research points towards differences in European healthcare systems, as they diverge substantially in terms of input (expenditure, number of personnel, facilities, etc) and output (number and quality of services delivered, etc), thereby affecting the organisation and accessibility of MHC givers, more specifically medical and non-medical mental health professionals.18–21 The relationship between how these diverse systems finance, provide and govern MHC across European countries and gender differences in MHC use has also not yet been studied.
Therefore, the aim of the study herein is first to determine cross-national variation in MHC use overall, and by general or specialised MHC, between men and women in 10 European countries. We also intend to examine whether gender differences apply to those suffering from the most common and more severe mental disorders. Second, we examine the effect of socioeconomic determinants as well as national differences in gender equality and type of healthcare system (ie, private expenditure on healthcare and level of services provided) in explaining gender differences in MHC use. These objectives are addressed using data from the European Union-World Mental Health (EU-WMH) surveys which collected information on a wide range of mental disorders, sociodemographic characteristics and MHC use.
Methods
Study sample
The WMH Surveys Initiative (http://www.hcp.med.harvard.edu/wmh/) aimed to assess the prevalence of common mental disorders, including their correlates and relationship to service use. The European part comprises six countries participating in a cross-sectional, population-based, household survey (The Netherlands, Belgium, Germany, France, Spain, Italy22) and four other countries conducting individual surveys with similar methodology (Bulgaria, Romania, Northern Ireland and Portugal).
A stratified, multistage, random sample without replacement was drawn in each country. The sampling frame and the number of sampling stages used to obtain the final sample differed across countries. The target population was represented by non-institutionalised adults (aged ≥18 years) identified from a national household list or a list of residents in each country. This list was obtained from the census, local postal registries, or, in the case of France, telephone lists.
Respondents underwent a face-to-face, computer-assisted personal interview, conducted by a trained interviewer, speaking in lay terms23; except for Bulgaria where a paper version was administered. The interview was conducted in two phases for all countries except Romania. Part 1 (given to all respondents) included sociodemographic variables, core diagnostic assessment of mental disorders and suicidal behaviours. All Part 1 respondents who met criteria for any lifetime mental disorder and a 25% probability sample of other respondents, were administered Part 2 of the survey. Details for each participating country are given in online supplementary table S1.
From 2001 to 2005 (except Portugal, 2008), data were obtained on a total of 37 289 respondents, ranging from 2357 in Romania to 5473 in Spain. Response rates varied from 45.9% in France to 78.6% in Spain. The sample was weighted to take into account different selection probabilities within countries (including the Part 2 sample), and poststratification weights to restore specific age and gender distributions of the general population in each country. In this study, data from the Part 1 sample were used for most analyses, except: (1) for Bulgaria, where MHC questions were only asked to the Part 2 sample and (2) when anxiety or alcohol disorders were used, as they were only assessed in Part 2 sample for the majority of countries. For Romania, Part 1 samples were used for the entire analysis.
Assessment of mental disorders
Lifetime presence of mood, anxiety and alcohol use disorders was determined by diagnostic criteria defined in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV). Organic exclusion rules were imposed in making all diagnoses. The WHO-CIDI Field Trials, and later clinical calibration studies, have shown that all these disorders could be assessed with acceptable reliability and validity, and more recent reappraisals carried out in four WMH countries (US, Italy, Spain and France, with total n=468) have demonstrated good agreement between CIDI-3.0 diagnoses and diagnoses based on blinded reinterviews.24 ,25
Mental disorders were grouped together into mood disorders—major depressive episode, dysthymia with hierarchy; and anxiety disorders—panic disorder, agoraphobia without panic disorder, specific phobia, social phobia, agoraphobia without panic disorder, generalised anxiety disorder, post-traumatic stress disorder. Additionally, three levels of severity were defined using a combination of presence of some diagnoses, such as substance disorders or suicidal attempt, number of areas of reduced functioning and its intensity measured by the disorder-specific Sheehan Disability Scales, or having overall functional impairment from any disorder consistent with the Global Assessment of Functioning.26
Assessment of MHC use
Lifetime MHC use was measured using the following question: ‘Did you ever in your life talk to a medical doctor or other professional about any problem with your emotions or mental health?’ Then, respondents were asked questions on the type of provider used. The types of services were divided into six categories: psychiatrist, other mental health specialist, general practitioner (GP), other health professional, complementary alternative medicine (CAM, such as healers, self-help or internet resources), and religious or spiritual advisor.
Demographic and social risk factors
Respondents were asked questions on several demographic variables, including gender and age at interview. Total years of schooling were separated by 0–12 vs ≥13 years. Since France was the only country not to ask participants on the precise number of years, sensitivity analysis was performed excluding France and including 0–11, 12, 13–15 and ≥16 years of schooling. Family household income was determined in relation to country medians (low, low-average, high-average, high). Respondents were asked if they were currently working and if not, for what reason(s). Unemployment was then defined as any person not working, excluding persons on sick leave or with any other health condition preventing them from working. Marital status was assessed by distinguishing married from divorced, widowed and never married respondents. Socioeconomic characteristics of the study population are summarised in online supplementary table S2.
Country-level indicators
The role of each individual country's healthcare system was addressed using two distinct indicators. First, private expenditure was expressed as a percentage of the total expenditure on health. This relative measurement reflects governmental priority towards healthcare given the prevailing level of economic development and wealth. In line with Wendt and Kohl,20 we included the density of healthcare providers for mental health problems, that is, the number of GPs, psychiatrists, clinical psychologists and psychiatric beds per 100 000 habitants. Second, a composite indicator for the level of service provision for mental health problems was included. Both these statistics were obtained from WHO (http://data.euro.who.int/hfadb/). The national level of gender inequality was represented by the 2006 Gender Gap Index (GGI) from the World Economic Forum,27 capturing the gap between men and women in four fundamental categories (economic participation and opportunity, educational attainment, health and survival and political empowerment). Details on how this index is calculated can be found elsewhere (http://www3.weforum.org/docs/WEF_GenderGap_Report_2006.pdf). Information on country-level statistics related to MHC resources is provided in table 1.
Statistical analysis
Prevalence of mental health service was stratified by gender and country. Bivariate logistic regression models were used to construct unadjusted ORs comparing women with men on the following end-points: (1) any MHC use and (2) any lifetime GP versus psychiatrist/other mental health specialist use. Adjusted ORs were then constructed by including socioeconomic variables (age, income level, employment status, education and marital status) in the model above. After pooling countries together, the same models, as shown above, were employed, while another adjusted model was constructed by adding country-level indicators (index of MHC provision, private household out-of-pocket expenditure and GGI 2006).
Gender-specific analyses were then performed on the two end-points above, while specifically examining the association with any lifetime mood disorder, any lifetime anxiety disorder, any alcohol use disorder and severity of mental health problems. Differences in associations between genders were performed by including an interaction term of the mental disorder or severity level and gender in separate models (adjusted by the socioeconomic, country-level indicators and, in all models with a mental disorder, the level of severity), from which gender-specific ORs and 95% CI were directly calculated. Significance of the interaction effect was tested using a t test based on the Taylor series linearised SE. Pooled analysis was then performed as described above while including only participants with any mood, anxiety, or alcohol use disorder.
All analyses accounted for complex sampling design and were performed using STATA software (V.12.1, College Station, Texas, USA). Significance was defined as a p value <0.05.
Results
Gender differences in lifetime MHC use
As shown in table 2 and figure 1A, women were significantly more likely to use any MHC service than men for all countries except Romania. When examining gender ORs (figure 1A), adjusting for sociodemographic features does not substantially change the association between gender and MHC use for individual countries. This observation was the same after pooling countries together (figure 1B). Interestingly, the gender OR in MHC use was slightly higher after adjusting for index of MHC provision (per 1000: adjusted OR=1.66, 95% CI 1.53 to 1.80), private household out-of-pocket expenditure (per 100: adjusted OR=0.99, 95% CI 0.94 to 1.05), and GGI (per %: adjusted OR=1.03, 95% CI 1.02 to 1.05).
Gender differences in the type of MHC use
From table 2, the largest gender differences stemmed from GP use, whereas smaller differences were observed in more specialised services. While the prevalence of MHC provided by spiritual and religious sources were generally low for men and women, gender differences were observed in certain rather religious countries (Germany, Italy and Portugal). When comparing GP with mental health specialists, there were no significant gender differences within European countries except for Northern Ireland and Italy (figure 1C). However, women were significantly more likely to use GPs than mental health professionals after countries were pooled together (p<0.001, figure 1D). This association did not change after adjusting for important socioeconomic factors (p<0.001) as well as after adjusting for index of MHC provision (per 1000: adjusted OR=1.02, 95% CI 0.85 to 1.22), private household out-of-pocket expenditure (per 100: adjusted OR=1.12, 95% CI 1.01 to 1.24), and GGI (per %: adjusted OR=1.02, 95% CI 0.99 to 1.04) (p<0.001).
Gender differences in lifetime MHC use and common mental disorders
In a pooled gender-stratified analysis, we observed that women and men with mood disorders had 6.06-fold and 6.89-fold higher odds, respectively, of seeking MHC than without this disorder (p for interaction=0.2, table 3). In a further analysis where only those with any mood disorder were included (figure 2A), women were still significantly more likely to use any MHC (p=0.001), and more specifically GPs than specialised MHCs (p=0.001), as compared with men. These associations remained consistent after adjusting for socioeconomic factors and severity levels (p<0.001), as well as country-level indicators (p<0.001, figure 2A). Despite the significance in gender differences, large variations were observed between countries, with only Bulgaria, Portugal and France consistently reporting significantly higher levels of MHC use in women (see online supplementary figure S1A).
In a separate pooled gender-stratified analysis, we also observed that women and men with anxiety disorder had 3.15-fold and 4.44-fold higher odds, respectively, of seeking MHC than without this disorder (p for interaction=0.008, table 3). This appeared to influence the gender OR in MHC use when only considering those with any anxiety disorder (figure 2B, p=0.06), even as women were significantly more likely to consult GPs than specialised MHC (p<0.001). The lack of association held when additionally adjusting by socioeconomic factors and severity level (p=0.06); yet gender differences became significant after accounting for country-level indicators (figure 2B, p=0.02). Most notably, France was the only country with consistently significant higher MHC use in women versus men among those with any anxiety disorder (see online supplementary figure S1B).
Finally, women and men with any lifetime alcohol use disorder had 2.33-fold and 3.14-fold higher odds, respectively, of seeking MHC than without this disorder (p for interaction=0.15, table 3). Specifically among those with any alcohol use disorder (figure 2C), women were significantly more likely than men to have any MHC use (p=0.001), without any specific preference for type of MHC received (p=0.4). However, this association disappeared when adjusting for sociodemographic factors (p=0.4) and country-level characteristics (p=0.6). Due to the low numbers of participants with any alcohol use disorders in certain countries, no analyses at the national-level could be performed.
Gender differences in lifetime MHC use with increasing mental disorder severity
In the pooled gender-stratified analysis, we observed that women and men, respectively, had higher odds of any MHC-use at mild (OR=3.84 and 5.12, p for interaction=0.19) and moderate (OR=4.99 and 6.21, p for interaction=0.2) mental health problems. Men with severe mental health problems had a significantly higher odds of MHC-use (OR=14.70) when compared with women with the same level of severity (OR=8.95, p for interaction=0.03, table 3).
Discussion
Despite the large study sample and well-validated instruments, our study does have certain limitations that need to be addressed. First, results were based on self-reported use of healthcare, which could have been over-reported by one gender compared with the other.2 Second, differential response rates between genders could have occurred, where women were more likely to respond than men. Male respondents may have been inadvertently more prone to discuss issues involving mental health, thus more likely to consult. Considering that some diagnostic measures may increase memory of previous healthcare contacts,28 and that most of the WMH questionnaires pertained to internalising disorders, which is less prevalent in men than women, this bias may have been limited. Finally, a large portion of the data was collected between 2001 and 2005, with other countries joining the study at the latest in 2008. We cannot exclude the effect of certain temporal factors.
Notwithstanding these limitations, we observed a significant gender difference in any lifetime healthcare use, with women more likely to consult than men. This finding falls in line with previous reports from Europe and in a large cross-country study.26 ,29 Remarkably, gender differences were consistent across a number of countries, regardless of the diversity of economic development, gender gap and health service infrastructure.
One potential explanation for this finding could be that, since women consult GPs more often, they have more opportunities to discuss their mental health problems with a health professional. Women have also been shown to be more communicative of their problems under these circumstances. Additionally, GPs may carry a certain misguided expectation that women have more mental health problems, thereby more extensively inquiring whether they need help for such problems.30 An alternative explanation could be attributed to gender differences in attitude towards seeking care; where women have a more thorough understanding, are more at ease in talking about, and are more willing to accept help for their mental health problems.5
Interestingly, significant gender differences in MHC use held when controlling for individual sociodemographic variables and country-level indicators of service availability (MHC provision index, % of out-of-pocket) and gender equality. However, when looking at individual countries, one noticeable exception was Romania, where gender differences were negligible. Romania has many similar features to countries like Italy (except for the number of GPs) and Portugal, Spain and Bulgaria; all of which have substantial gender gaps. Yet, Romania stands out in terms of total cost of expenditure, being by far the lowest of any other nation included in the EU-WMH survey. It could be hypothesised that resources would be allocated to patients with more severe cases of mental health problems, in whom gender differences are less apparent, as shown in our analysis.
Once focusing on the type of care, we observed that gender differences mainly apply to conventional providers (medical or non-medical specialist, GP) and much less to non-conventional ones. As previously observed,29–33 GPs were consistently and significantly used more often over other mental health professionals among women than men in almost every country and, while pooling countries together and controlling for, among others, MHC expenditure, , this was still the case. Nevertheless, adjustment of country-level factors may not fully capture the reasons for certain differences. For example, Northern Ireland uses the GP as a gatekeeper allowing access to mental health providers (MHP) under severe cases, whereas Italy has costly access to MHP resulting in the use of other providers.28 Both these countries have similar gender differences in GP versus MHP use. Inversely, in countries like Germany and The Netherlands, where non-medical mental health professionals are readily accessible, there is no gender relationship in GP:MHP ratio. Facilitating GP use may, therefore, contribute to the general preponderance of women in using MHC.
One important finding from this study is that, when considering those suffering from specific mental health problems, gender differences remain significant among only those with any lifetime mood disorder, but not for only those with any lifetime anxiety disorder. Nonetheless, the most heterogeneity between countries was displayed in these particular analyses. Of note, France was the only country with a very strong and significant gender difference among those with anxiety disorder, whereas Germany shows almost the exact opposite. Indeed, France relies more heavily on GPs than other sources of MHC for these types of disorders, whereas Germany uses more talkative and corporal therapies provided by MHP.21 Perhaps the structural organisation in France may have favoured a gender difference in seeking MHC. The extensive differences in MHC strategies among those with anxiety disorder, as shown in online supplementary figures S1C,D, might explain why gender ORs became significant once adjusting for the healthcare system and GGI.
In conclusion, women do use MHC more frequently than men, yet this depends on the type and severity of mental health problems. These findings were fairly consistent across European countries, nevertheless, certain exceptions were noted. Our results are rather concerning and leave to question the adequacy of care for specific genders. Certain deficiencies in detecting mental disorders have been observed among GPs, who would, in turn, offer less adapted care than mental health specialists.34 Additionally, GPs are potentially more likely to prescribe psychotropic drugs instead of offering therapy,35 which remains at a high cost in most European countries, and could result in some women turning to more alternative methods of care. At the same time, since men do not frequent GPs for mental health issues, men with more mild symptoms may not have the resources in order to seek adequate care, putting them at risk for self-medicating via excessive alcohol use.36 Better MHC training among GPs, and different paths to seeking more specialised MHC in men and women, must be stressed.
What is already known on this subject ?
-
Women are more likely than men to use health services for their mental health problems. This gender difference can be attributed to socioeconomic position, how healthcare infrastructure is organised and, most importantly, the types of mental disorder with which men and women present.
What this study adds?
-
Women sought care typically from general practitioners, more often than men, for mental health problems in almost all European countries, even after adjusting for socioeconomic and demographic variables as well as country-level indicators of mental healthcare provision and gender equity. However, these differences were almost non-existent in the presence of any anxiety disorder or when accounting for overall severity of mental health problems. Our results stress the need to establish gender-specific paths in seeking more specialised mental healthcare.
Acknowledgments
We thank the World Mental Health (WMH) Survey Initiative staff for assistance with instrumentation fieldwork and data analysis. A complete list of WMH publications can be found at http://www.hcp.med.harvard.edu/wmh/.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online figures
- Data supplement 2 - Online table 1
- Data supplement 3 - Online table 2
Footnotes
-
Contributors VK-M has direct responsibility for the work, contributing to study conception and design, acquisition of data, analysis and interpretation of data and drafting the article; AB contributed to the analysis, interpretation of data and drafted the article; SvdV contributed to interpretation of data and drafted the article; RdG, JMH, SF, SO'N, JA contributed to the acquisition of data and critically revised the manuscript for important intellectual content; GV contributed to analysis and interpretation of data and critically revised the manuscript; and LW critically revised the manuscript for important intellectual content. All authors approved the final version of the manuscript.
-
Funding This project was funded by the European Commission (Contract QLG5-1999-01042 and SANCO 2008/1308); the Piemont Region (Italy), Fondo de Investigacion Sanitaria, Instituto de Salud Carlos III, Spain (FIS 00/0028), Ministerio de Ciencia Y Tecnologia, Spain (SAF 2000-158-CE), Department de Sanitat, Generalitat de Catalunya, Spain, other local agencies and by an unrestricted educational grant from GlaxoSmithKline. ESEMeD is carried out in conjunction with WHO WMH Survey initiative. These WMH activities were supported by the John D and Catherine T MacArthur Foundation, the Pfizer Foundation, the US Public Health Service (R13-MH066849, R01-MH069864, and R01-DA016558), the National Institute of Mental Health (NIMH R01-MH070884), the Fogarty International Center (FIRCA R01-TW006481), the Pan American Health Organisation, Eli Lilly and Company, Ortho-McNeil Pharmaceutical, Inc., GlaxoSmithKline, and Bristol-Myers Squibb.
-
Competing interests None.
-
Provenance and peer review Not commissioned; externally peer reviewed.