Principal findings

This second Headway-based article revealed a high variation in the responsiveness to MH needs across EU-27 + UK countries, which was shown to negatively correlate with the population MH status – the poorer the MH status, the better the responsiveness. Thus, Sweden, Finland, Denmark, France, and Germany achieved the best performance on responsiveness to their populations’ MH needs, although their performance in MH status was poor. On the other hand, Greece, Slovakia, and Cyprus, which performed well in MH status KPIs, showed the poorest performance in responsiveness to MH needs. At first glance, it seems that countries’ responsiveness failed to mitigate the impact of the determinants of MH. Conversely, one may argue that better-performing countries in responsiveness might do so because of greater preexisting MH needs (e.g., Scandinavian countries) and vice versa (South-European countries). Nevertheless, tackling inequality and bridging the mental healthcare gap should guide future European MH policies.

Responsiveness in healthcare

The WHO defined health systems responsiveness as “how well the health system meets the legitimate expectations of population for non-health enhancing aspects of the health system” [18]; hence, an inherent goal of any health system which must be measured and monitored over time [Reference Mirzoev and Kane19]. More specifically, health system responsiveness encompasses both the system’s ability and capacity to respond and its actual response to medical [Reference Khan, Kagwanja, Whyle, Gilson, Molyneux and Schaay20] and nonmedical issues [18].

In particular, Scandinavian countries, such as Sweden (10), Denmark (8.8), and Finland (8.3), showed the best performance in responsiveness in healthcare, while Romania (1.0), Slovakia (1.1), and Latvia and Bulgaria (1.2) had the poorest results. Specifically, four major domains of healthcare systems across Europe were evaluated, namely workforce, facilities, quality of care, and economic resources.

Regarding workforce, Finland (10), the Netherlands (9.7), and Denmark (8.3) achieved the best performance, while Bulgaria (1.0), Portugal (1.9), and Romania (2.0) showed the lowest scores. As of 2021, on average, for the EU-27 + UK countries, there were 18.8 psychiatrists per 100,000 inhabitants, ranging from 10.2 (Bulgaria) to 28.6 (Germany), that is, an almost threefold variation. Two countries – Spain and Malta – had no child psychiatrists at all. Such a high variability of healthcare professionals’ rates across European countries may have widened since the coronavirus disease 2019 (COVID-19) pandemic [2–Reference Lora, Kohn, Levav, McBain, Morris and Saxena4]. Of concern, there has been a severe shortage of MH professionals as the number of individuals in need increased, which seems to be exacerbated by high levels of staff burnout [Reference O’Connor, Muller Neff and Pitman21]. Thus, it is estimated that by 2030, Europe will face a shortfall of ~600,000 doctors [22]. Although Digital Mental Health approaches may mitigate this [Reference Moreno, Wykes, Galderisi, Nordentoft, Crossley and Jones23–Reference Wies, Landers and Ienca25], their long-term outcomes remain unknown [Reference Cummings, Zhang, Gandré, Morsella, Shields-Zeeman and Winkelmann26]. Therefore, urgent action is needed to enhance staff recruitment and retention rates across Europe [22].

In terms of facilities, Slovenia (10), Italy (9.9), and Cyprus (9.6) were the most resourced countries, whereas the Netherlands (1.0), Germany (1.5), and Belgium (3.4) had the poorest performance. Given the Headway Initiative methodology [15], results of worse-performing countries may only indicate a better provision of inpatient, rather than community-based MH facilities, and not necessarily limited MH facilities.

In keeping with the above, mental healthcare expenditure significantly varied across Europe, with France (13.9%), Germany (13.1%), and Sweden (10.0%) – all of them with good performance in responsiveness – emerging as the principal investors, significantly exceeding the EU-27 + UK average of 5.4%. Hence, many European countries failed to comply with the 2018 Lancet Commission recommendations, according to which national MH budgets should receive between 5% (low- and middle-income countries) and 10% (high-income countries) of the total healthcare [Reference Patel, Saxena, Lund, Thornicroft, Baingana and Bolton27]. Certainly, greater national MH expenditure was linked to better quality of care in psychiatry [Reference Salisbury, Killaspy and King28], including lower suicide rates [Reference Nordentoft and Erlangsen29]. In addition, MH prevention yields a high long-term return on investment [Reference McDaid, Park and Wahlbeck30].

Regarding quality of care, Denmark (10.0), Ireland (9.4), and Sweden (8.7) showed the highest performance, whereas Romania (1.0), Slovakia (1.4), and Bulgaria and Latvia (1.5) had the poorest quality of care. Despite MH becoming a top priority in the political agenda of European governments and institutions [31], up to 6.2% Europeans reported unmet MH needs due to financial issues in 2022, ranging from 1.1% (Romania) to 29.8% (Finland). [5]. These apparent between-country differences, however, may well reflect the quality of data and reporting issues, including the influence of stigma, which tends to discourage people from talking openly about MH [Reference Thornicroft, Sunkel, Alikhon Aliev, Baker, Brohan and El Chammay11].

Responsiveness in workplaces, schools, and society

Regarding responsiveness in workplaces, Germany (10) and Finland and Belgium (9.4) reached the best performance, while Greece (1.0), Slovakia (1.4), and Czechia and Hungary (3.4) had the lowest scores. Overall, almost two in three Europeans with mental disorders were found to be unemployed and get paid, on average, 30% lower wages than their counterparts. Indeed, MH patients face exclusion from work, largely due to stigma and discrimination, which prevents them from recovery [Reference Thornicroft, Sunkel, Alikhon Aliev, Baker, Brohan and El Chammay11, 32]. Labour market integration policies should be promoted [Reference Curtis, Pearce, Cherrie, Dibben, Cunningham and Bambra33, Reference Hammarström, Westerlund, Janlert, Virtanen, Ziaei and Östergren34], which also reduced suicide risk [Reference Gerdtham and Johannesson35], while community-based mental healthcare effects on return-to-work outcomes are less clear [Reference Lundqvist, Brattmyr, Lindberg, Havnen, Solem and Hjemdal36]. Although the WHO developed specific guidelines on MH promotion at work [37], guidelines compliance and their effects on health and social outcomes remain to be established [Reference Malik, Ayuso-Mateos, Baranyi, Barbui, Thornicroft and van Ommeren38].

In terms of responsiveness in schools, France (10.0) and Latvia and Lithuania (8.6) achieved the best performance, whereas Slovakia (1.0), Cyprus (2.2) and Greece (2.4) had the poorest results. It is well-established that for almost two in three MH patients, the illness onset occurs before age 25 years, with a peak at 14.5 years [Reference Solmi, Radua, Olivola, Croce, Soardo and Salazar de Pablo39]. In Europe, full-time compulsory education/training usually lasts 10–11 years and ends at the age of 15–16 years [40]. Our results revealed that the overall performance on responsiveness in schools negatively correlated with the overall MH status and suicide rates – the better the performance at schools, the poorer the MH status, including higher suicide rates. However, up to 20% of European children were shown to experience MH problems during their school years, and a comparable proportion reported feelings of unhappiness and anxiety about the future, which was linked to bullying, challenging schoolwork and loneliness [5]. In addition, EU Countries widely differed in the ability to provide MH prevention and promotion programs at schools, and between 7.5% (Estonia) and up to 21.5% (The Netherlands) of school dropouts were children with mental disorders [41]. Hence, few would question the appropriateness of schools as “the right place at the right time” for early intervention [Reference Fusar-Poli, Correll, Arango, Berk, Patel and Ioannidis12, Reference Arango and Fusar-Poli42]. Consistent with this, targeted prevention programs reduced depression and anxiety symptoms [Reference Werner-Seidler, Spanos, Calear, Perry, Torok and O’Dea43], whereas universal school-based interventions prevented suicidal behaviour in adolescents [Reference Wasserman, Hoven, Wasserman, Wall, Eisenberg and Hadlaczky44]. In addition, school-based MH promotion programs focused on resilience may aid students in managing their own stress [Reference Fenwick-Smith, Dahlberg and Thompson45], which could be delivered through digital technologies [Reference Chen, Ou, Li and Luo46]. MH clinical liaison teams appear to facilitate access to care [Reference Camacho-Rubio, Ulloa, Parellada, Pina-Camacho, Díaz-Caneja C and Sevilla-Cermeño47, Reference McPhail, Thornicroft and Gronholm48], which should result in better long-term outcomes, although this remains to be demonstrated.

Regarding responsiveness in society, Germany (10), France (9.3), and Sweden (8.6) achieved the highest performance, while Greece (1.0), Estonia (2.0), and Romania (3.1) had the poorest performance. Of note, there was a negative association between responsiveness in society, especially MH promotion programs, and MH status – the poorer the MH status, the better the responsiveness. From a societal perspective, stigma and discrimination were identified as the main areas of concern for improving people’s MH [Reference Thornicroft, Sunkel, Alikhon Aliev, Baker, Brohan and El Chammay11]. Stigma was defined as the co-occurrence of stereotyping, separating, status loss, and discrimination in the context of power inequities [Reference Link, Yang, Phelan and Collins49]. While the youth may particularly benefit from antistigma campaigns [Reference Song, Hugh-Jones, West, Pickavance and Mir50], such as the England-based Time-to-Change [Reference Sampogna, Bakolis, Evans-Lacko, Robinson, Thornicroft and Henderson51], especially in the short-term [Reference Sampogna, Bakolis, Evans-Lacko, Robinson, Thornicroft and Henderson51], their long-term effects remain far from clear [Reference Gronholm, Henderson, Deb and Thornicroft52].

Next steps

In the first Headway-based article (Lopez-Morinigo et al., this issue), we examined the MH status of Europeans and its determinants. Interestingly, overall KPI scores on responsiveness in workplaces (r = −0.46, p = .014), schools (r = −0.59, p < .001), and society (r = −0.53, p = .003) negatively correlated with overall KPI scores on MH status – the better the responsiveness, the worse the MH status. On the other hand, the correlation between overall performance in responsiveness in healthcare and MH status did not reach significance (r = −0.34, p = .075). Taken together, these findings deserve some consideration. First, it seems that responsiveness to people’s MH needs may largely occur in workplaces, schools, and society rather than in healthcare systems. Second, there was a negative correlation between MH status and responsiveness – the better the status, the worse the responsiveness. Whether more responsive countries succeeded in meeting preexisting greater MH needs (i.e., a positive result) or they failed to address people’s MH issues (i.e., a negative result) remains to be clarified. Of relevance, overall KPI scores on determinants, which were not linked with MH status KPIs (Lopez-Morinigo et al., this issue), positively correlated with responsiveness in healthcare (r = 0.41, p = .031) and in workplaces (r = 0.40, p = .034). This noted, further theoretical debate about the appropriateness of KPIs for the evaluation of public MH policies is still warranted.

Indeed, the lack of comprehensive, independent, and comparable data poses a major barrier to the development of a monitoring and evaluation/accountability framework in MH policymaking. To address this challenge, within the United Nations Sustainable Development Goals agenda, the Countdown Global Mental Health 2030 was designed, which collected data on 48 indicators from 15 sources covering 193 countries across the globe. Specifically, indicators were clustered around three themes: determinants of MH and factors shaping MH needs and their response [Reference Saxena and Kline53]. Easily-accessible datasets, such as the European Headway Initiative [15] and the Countdown Global Mental Health [Reference Saxena and Kline53], may inform public MH policymaking, including between-country comparisons, whereas researchers may generate novel real-world hypotheses. However, these data-driven approaches are not exempt from criticism, namely the extent to which somehow arbitrarily chosen KPIs really align with the population MH needs and the time required to evaluate MH policy changes.

In particular, this study’s results may contribute to future evidence-based MH policies aimed to address European-level and country-specific challenges. To this end, scientific societies, such as the European Psychiatric Association (EPA) and the World Psychiatric Association (WPA), are committed to providing a strong leadership. This being said, it is worth noting that more than nine in 10 healthcare interventions subject to Cochrane Reviews were not supported by high-quality evidence, while harms do not tend to get published [Reference Howick, Koletsi, Ioannidis, Madigan, Pandis and Loef54]. In addition, health authorities must bridge the gap between research evidence, clinical guideline recommendations, and approved interventions by regulatory bodies, especially in children and adolescents with serious mental illness [Reference Lopez-Morinigo, Leucht and Arango55, Reference Solerdelcoll, Arango and Sugranyes56].

Of relevance, MH policymaking is not exempt from compliance with international codes of ethics in psychiatry [Reference Galderisi, Appelbaum, Gill, Gooding, Herrman and Melillo57]. To this end, the ongoing dialogue between the EPA and the WPA [Reference Sansone, Tyano, Melillo, Schouler-Ocak and Galderisi58] and closer collaboration between national psychiatric associations [Reference Samochowiec, Frydecka, Skonieczna-Żydecka, Schouler-Ocak, Carpinello and Chkonia59] will be essential. Digital MH raises ethical issues about privacy and the use of personal data, which is particularly relevant to children and adolescents [Reference Cratsley and Radden60]. More broadly speaking, public MH also faces the challenge of appropriately timing and targeting interventions, which is crucial for staging models [Reference Fusar-Poli, McGorry and Kane61], not to mention the sometimes blurred boundaries between health and disease in MH [Reference Cratsley and Radden60].

To address the increased MH needs after the COVID-19 pandemic, some MH action plans have been made by European governments. For instance, since 2021, the UK government has offered training grants for senior MH leads to all state schools and colleges, whereas in Spain, a 24-h suicide prevention hotline became available that same year. In 2022, MH services in Belgium were reformed to facilitate access to care from schools, sports facilities, and work. One year later, the 2023 German Centre for Mental Health (DZPG) research network was launched. In Italy, since 2023/2024 a psychologist can be accessed both in schools and in primary care (i.e., “Psicologo di base”) [15].

In light of our results, further evidence-based public MH interventions can be recommended. To enhance responsiveness in healthcare systems, not only is higher MH expenditure required, but also optimization of available resources, such as the so-called digital MH [Reference Witteveen, Young, Cuijpers, Ayuso-Mateos, Barbui and Bertolini62]. Labour market integration policies may make the onset of work-related mental disorders less likely [Reference Curtis, Pearce, Cherrie, Dibben, Cunningham and Bambra33, Reference Hammarström, Westerlund, Janlert, Virtanen, Ziaei and Östergren34]. In schools, MH promotion programs [Reference Chen, Ou, Li and Luo46] and clinical liaison teams [Reference McPhail, Thornicroft and Gronholm48] may become the key primary and secondary (i.e., early detection/intervention) preventive strategies, respectively. Finally, from a societal approach, antistigma campaigns should facilitate professional help-seeking by MH patients [Reference Thornicroft, Sunkel, Alikhon Aliev, Baker, Brohan and El Chammay11]. See Table 6 for further details.

Table 6.

Interventions on KPIs in responsiveness to mental health needs in healthcare, workplaces, schools, and society

Strengths and limitations

The Headway Initiative collected cross-national population-level data on 54 MH-related KPIs, including MH status and determinants (Lopez-Morinigo et al., this issue) and responsiveness of the systems to the population MH needs. Data came from official, authoritative open-access datasets, such as Eurostat, WHO, or the Organisation for Economic Co-operation and Development (OECD), thus ensuring the replicability of the study. Owing to this methodology, some evidence-based public MH interventions were recommended.

However, three main limitations of this study should be acknowledged. First, the Headway methodology relied on national datasets, which differed in the quality of data and data availability (i.e., under- and overreporting issues), including cross-national cultural differences. Further, data on KPIs were collected in different years depending on data availability, that is, from the most recent year when the variable data were available. In addition, it is worth noting that “association does not mean causation” [Reference Hill63], and caution is needed when inferring causality from the above bivariate correlations. Second, further nontested KPIs, such as psychiatric medication availability and compliance and/or funding source of healthcare systems, may have altered the results. Third, the Headway Initiative adopted both analytical and qualitative approaches, which, although unlikely, may have biased the results. In particular, the Headway Mental Health Index aimed to comparatively measure European countries’ relative performance on a number of MH-related KPIs, thus ranking them rather than cross-comparing raw data on the above variables.