The case for prevention

Social psychiatry, once given an exclusion definition of ‘not biological and not psychodynamic psychiatry’, has evolved into public mental health (PMH). Davies and Mehta (Reference Davies and Mehta2013) conceptualised PMH as a triad of Rehabilitation/Treatment (how we spend 95%+ of our working lives), Promotion (raising awareness in schools, workplaces and across media) and the neglected third component of Prevention. Prevention in mental health has the evidence (Jenkins & Ustin, Reference Jenkins and Ustin1998; Davies & Mehta, Reference Davies and Mehta2013; Byrne & Rosen, Reference Byrne and Rosen2014; Campion, Reference Campion2019; Firth et al., Reference Firth, Solmi, Wootton, Vancampfort, Schuch and Hoare2020) but lacks both financial support and participation among mental health professionals. If clinical psychiatry can be compared to a battlefield, as psychiatrists we are drawn to treat the wounded and the dying – rather than activities to build up the defences and personal armour, sound the retreat, and train people to fight back. We have been cowed by years of underfunding such that we would not dare to presume to negotiate a ceasefire. In that frontline firefighting spirit, there are two comprehensive textbooks written for psychiatrists and allied professionals to address poor physical health in people with SMI (Cormac & Gray, Reference Cormac and Gray2012; Taylor et al. Reference Taylor, Gaughran and Pillinger2020). These are the How To of detection and treatment of the physical diseases that blight lives and end them prematurely; both touch on secondary prevention (early detection and treatment): each of hypertension and diabetes are important early intervention points, made even more urgent in the Age of Covid. They are disease/organ-based and neither book has a primary prevention focus. The Lancet Psychiatry Commission has also moved the dial: Firth and colleagues (Firth et al. Reference Firth, Siddiqi, Koyanagi, Siskind, Rosenbaum and Galletly2019) in advocating multilevel actions across health systems to improve common modifiable risk factors, based on compelling evidence with few underresearched areas. We now know more about the shared pathogenesis of psychosis and cardiometabolic disorders (Perry et al., Reference Perry, Stochl, Upthegrove, Zammit, Wareham, Langenberg, Winpenny, Dunger, Jones and Khandaker2021) and this adds urgency to intervene earlier. Activation of phenotype and genetic links are not a counsel of despair – we do not ‘give up’ on alcoholic patients who have strong family histories of addiction. While integrated care and innovation (e.g. electronic cigarettes, digital technologies, Diabetes Prevention Programme: Firth et al., Reference Firth, Siddiqi, Koyanagi, Siskind, Rosenbaum and Galletly2019) will progress the protection of physical health, the Commission concluded that we can only achieve better outcomes through public health approaches (Firth et al., Reference Firth, Siddiqi, Koyanagi, Siskind, Rosenbaum and Galletly2019).

Evidence across mental disorders

Data from the UK and Canada, based on patients engaged in treatment within accessible universal healthcare, is likely an underestimate of the scale of premature deaths (Table 1). As we start to look at solutions, remember that the same endeavours that reduce SMI premature mortality are also beneficial to our patients beyond the 1–2% with psychosis. Intellectual disability (ID) confers an even higher premature mortality than SMI (McCarron et al., Reference McCarron, Carroll, Kelly and McCallion2015): this is a combination of both the condition(s) causing ID and the increase in associated psychiatric morbidities, themselves driving poorer physical health – most offer potential points of interventions (Fig. 1). For example in ID patients, safe psychiatric prescribing is even more relevant as highlighted in the STOMP publication by Branford et al. (Reference Branford, Gerrard, Saleem, Shaw and Webster2019). People with addictions and alcohol/substance misuse die even younger than SMI patients, due to lack of treatment opportunities and/or intermittent abstinence-misuse cycles (Chang et al., Reference Chang, Hayes, Perera, Broadbent, Fernandes, Lee, Hotopf, Stewart and Scott2011; Hayes et al., Reference Hayes, Chang, Fernandes, Broadbent, Lee, Hotopf and Stewart2011; Shield et al., Reference Shield, Gmel, Gmel, Mäkelä, Probst, Room and Jürgen2017; Macloud et al., Reference Macloud, Steer, Tilling, Cornish, Marsden, Millar, Strang and Hickman2019). Misuse increased among a more deprived subpopulation during the first UK lockdown (Marmot et al., Reference Marmot, Allen, Goldblatt, Herd and Morrison2020). Data on life expectancy among people with personality disorder is also severely effected (Fok et al., Reference Fok, Chang, Broadbent, Stewart and Moran2019; Firth et al., Reference Firth, Siddiqi, Koyanagi, Siskind, Rosenbaum and Galletly2019). Improving physical heath in everyone, especially maintaining mobility and reducing alcohol excess, has major positive impacts on the prevention of most types of dementia (Livingston et al., Reference Livingston, Huntley, Sommerlad, Ames, Ballard and Banerjee2020). Anxiety, the most common mental disorder, also confers biological disadvantages with an increased relative risk, RR of 1.41 (95% CI, 1.23 −1.61) and 1.71 (95% CI, 1.18 −2.50) for coronary heart disease and stroke respectively (Emdin et al., Reference Emdin, Odutayo, Wong, Tran, Hsiao and Hunn2016). All interventions listed below will reduce mortality excess and physical illness burden. Across the lifespan the highest rates of depression requiring medical intervention occur in people with long term conditions (LTCs) such as diabetes, heart failure, chronic obstructive pulmonary disease and hypertension. Because multimorbidity is the rule not the exception in people with SMI over 40, our patients are overrepresented in LTC populations, as well as being undertreated.

Fig. 1. Seven drivers of Premature Mortality in people with severe mental illness (SMI).

Proportionate universalism

All activities that protect physical health also benefit mental health; and even small increments of improvement in mental health in populations yield benefits to mitigate and prevent physical disorders. These actions have intergenerational benefits: simple interventions now in SMI patients will support them towards quitting smoking, cooking from ingredients and taking gentle exercise. Their children will directly benefit from all three parents’ changes right across their lifetimes. As a profession, we have not performed well in preventing childhood mental disorder, and prioritising parents with mental disorders is a good start (Cooklin & Gorrell Barnes, Reference Cooklin and Gorrell Barnes2021). Our dying planet and unsustainable human activities are pressing concerns. Again, three interventions cited promote a sustainability agenda. So why don’t we do prevention? The drivers of premature mortality across mental disorders are multifactorial, interacting/symbiotic (Fig. 1) and it can be frustrating for people working in health/voluntary sectors when the benefits of positive changes they negotiate accrue to individuals, their families, parts of other systems and the State some years after their efforts. Public health achieved advances we have taken for granted or forgotten, and we need their leadership to rethink then reverse the drivers of the stolen years (Supplementary Figure 1).

Post-Covid challenge

The UK’s Covid deaths to January 2022, the worst in Europe, reflect a combination of regressive social policies before the pandemic, of which obesity is one driver, and inequalities in infection rates and accessing health treatments (Marmot et al. Reference Marmot, Allen, Goldblatt, Herd and Morrison2020). Back to our psychiatric clinics, and Hughes and colleagues (Hughes et al. Reference Hughes, Bellis, Hardcastle, Sethi, Butchart, Mikton, Jones and Dunne2017) examined the effects of 4 or more ACEs (adverse childhood experiences) across 10 European countries (Supplementary Table 1). None of the adverse outcomes predicted here (odds ratios are associations) will surprise experienced clinicians, and how clinical presentations cluster together sharpens the tools of Prevention. Beginning in childhood, poorer children experience more ACEs and there are fewer mitigating interventions for these. These inequalities are ingrained by adult life, living in more deprived areas with less green spaces, higher volume traffic, more fast food/alcohol and tobacco retailer outlets (Marteau et al., Reference Marteau, Rutter and Marmot2021). Studies link urban air pollution to increased adolescent mental health presentations (Szyszkowicz et al., Reference Szyszkowicz, Zemek, Colman, Gardner, Kousha and Smith-Doiron2020). The scale of what is needed to overcome inequalities and ACEs is overwhelming. We know the challenges of primary prevention will fall to future generations of health professionals. Poverty is toxic to mental health; there is evidence of ‘bidirectional causality’ and that targeted anti-poverty programmes will prevent mental disorders with net positive economic benefits (Ridley et al., Reference Ridley, Rao, Schilbach and Patel2020). We must challenge the consensus that ‘we are so much more positive about mental health these days’, not least among politicians who are happy to cut the ribbon for a new service or raise ‘mental health awareness’ but do nothing to promote universal income, adequate benefits (Boardman, Reference Boardman2020), and ignore proven approaches like housing first (Padgett, Reference Padgett2020). Across systems and nationally, we need sustainable policies that establish a clinical focus on physical health as the new normal. Principles of integrated care are presented in Supplementary Table 2. Learning from smoking, we cannot wait seven decades to address any potential issues with the ‘Alcohol lobby’, supermarkets, ‘Big Tech’, ‘Big Pharma’ and poor medical prescribing practices (Norris et al., Reference Norris, Smith, Doran and Barry2021).

Why me?

Doing things differently carries costs. For clinicians there are time costs – the opportunity costs of discussing physical health when you might be doing other things. Health services are fragmented (Delgrado et al., Reference Delgrado, Binzer, Shah, Ekberg, Arrieta and Allwood2021), and each silo repeats the common mantra about treating the poor physical health of people with SMI: ‘can’t someone else do it?’ Our failures as clinicians to address premature SMI mortality relate to a professional mind-set of low expectations of our patients (to stop smoking, eat healthily, take exercise) and therapeutic nihilism. When adequate trials compare interventions between SMI and controls, for example outcomes in 105 018 patients with myocardial infarction, proactive treatment of cardiovascular disease produces the same better outcomes in SMI as controls (Kugathasan et al., Reference Kugathasan, Horsdal, Aagaard, Jensen, Laursen and Nielsen2018). This 11 year Danish study also showed that each of invasive (percutaneous) cardiac interventions and multiple cardioprotective medications were less likely to be offered to the subgroup with schizophrenia (Kugathasan et al., Reference Kugathasan, Horsdal, Aagaard, Jensen, Laursen and Nielsen2018). The hope now is that our collective response to Covid-19 has changed clinicians’ attitudes. Economists, policy makers and politicians who hear Prevention arguments might focus on the savings from admissions avoided, reduced care costs of poor physical health or (hopefully looking beyond annual budgets or election cycles), the aspiration of improvements in life expectancy. In these islands, health systems are complex and holistic care becomes merely an aspiration, quickly abandoned when something demanded of our service might be left to another part of the system. While we work to fix these perverse disincentives, we need to act now in clinics to reverse the drivers of premature mortality in people with mental disorders. Our postgraduate educational systems have also become fragmented, but we can resist: when a diabetologist or hepatologist trains you, return the favour and teach mental disorders to their teams. Why not you?

The Malevolent Seven

The drivers of Fig. 1 are not the seven deadly sins – what the current UK Government has dubbed ‘lifestyle choices’. The Malevolent Seven are driven by inequalities, mostly economic but minority ethnic status is also a powerful determinant of inequality and exclusion, and they tend to clump together as syndemics (Singer et al., Reference Singer, Bulled, Ostrach and Mendenhall2017). In the prevention of depression-anxiety, Jacka and colleagues (Jacka et al. Reference Jacka, Mykletun and Berk2012) have identified the three most plastic/remediable risk factors as physical inactivity, smoking and diet quality. For Jacka, the central ‘causes of the causes’ poverty-inequality/ACEs are in the ‘too difficult box’, and the paper conceptualises the related negative influences of poor social networks and alcohol/substances as less fruitful intervention points. Others disagree. In rejecting a disease-specific paradigm, Fig. 1 is a useful reconceptualization of the interacting, common antecedents that drive physical disease processes and make mental health even worse. In the clinic, we cannot separate the behaviours of substance misuse from psychological dysfunction – because they arise and interact in the same person. Thankfully, not every patient will have all these drivers and interventions to reduce/mitigate the drivers work within a personalised medicine framework. Once liver impairment starts, other drivers (smoking, alcohol, overweight, prescribed/illegal drugs) damage organs to increasing degrees. LTCs are detected late in people with SMI so we should not see these as sequential comorbidities: Fig. 1 shows shared antecedents. LTCs share aetiologies, and many, for example, essential hypertension, are attenuated by obesity, inactivity, alcohol, and high fat, sugar, salt (FSS) diets. In attending to these markers, we could also reduce or delay dementia (Livingston et al., Reference Livingston, Huntley, Sommerlad, Ames, Ballard and Banerjee2020).

Where to start?

With interventions to reduce alcohol and smoking, try to integrate the individual conversations within broader public health actions (Fig. 2). To achieve reductions in both, it is about pricing, then limits to availability and advertising curbs; what we undertake as health education (much schools-based activity, finger wagging by doctors) has far less impact on alcohol misuse (Babor et al., Reference Babor, Caetano, Casswell, Edwards, Giesbrecht, Graham, Grube, Hill, Holder, Homel, Livingston, Österberg, Rehm, Room and Rossow2010). In individual smoker discussions, this is not about what might happen 10 years away, but the effects of these activities now: impact on finances plus time lost, effects on others (children/elders), and immediate personal consequences: smelling of cigarettes, loss of taste, gateway to other substances. The open question (what matters to you?) replaces the traditional (what’s the matter with you?). When patients cite insomnia, our challenge as clinicians is to achieve coproduction of better sleep without prescribing sedative medication. Ask and acknowledge the negatives but change the narrative from the amount of sleep achieved (problematic clock-watching) to enjoying the benefits of feeling rested each morning after small agreed changes. The UK Biobank Study (123 794 white British participants) has linked genes to behaviours using polygenic risk scoring, and Choi and colleagues (Choi et al. Reference Choi, Stein, Nishimi, Ge, Coleman, Chen, Ratanatharathorn, Zheutlin, Dunn, Koenen and Smoller2021a) examined 106 known modifiable risk factors for incident (new) depression. Findings support clinicians’ instincts (benefits of social networks, frequent confiding in others, exercise, attending a club or pub) but there is a stronger bidirectional relationship between daytime napping and depression. Choi and colleagues (Choi et al. Reference Choi, Chung, Kang, Kwak, Yang, Park and Yim2021b) did not find strong links between total hours slept and depression. Cutting out daytime naps, or limiting this to one 30-min nap each day is one of many sleep hygiene measures. Broaden these to advise cutting out stimulants (nicotine, substances), reducing caffeine, modifying deleterious sedatives (alcohol, substances, medications) and promoting gentle exercise outside and offline, with mindfulness (Fikree & Byrne, Reference Fikree and Byrne2021). If either loneliness or fatigue are spontaneous symptoms, there begins a similar coproduction on pairing social activities with exercise.

Fig. 2. Evidence-based public health interventions to reduce alcohol misuse and smoking, shown with selected (top) and individual (tip) interventions.

Personalised medicine – implementing Prevention across the lifespan

Firth and colleagues (Firth et al. Reference Firth, Solmi, Wootton, Vancampfort, Schuch and Hoare2020) reviewed randomised trials and prospective meta-analyses to show clear benefits (in descending order) of any physical exercise, never smoking or quitting, good sleep and healthy diet. For the first two, we see good evidence of the primary prevention of psychosis and attention deficit hyperactivity disorder (ADHD), and all four have been shown at a population level to reduce adult onset of anxiety, depression and bipolar disorder (Firth et al. Reference Firth, Solmi, Wootton, Vancampfort, Schuch and Hoare2020). Although this meta-review did not include alcohol use, their conclusions on sleep and exercise have since been replicated by Choi and colleagues (Choi et al. Reference Choi, Stein, Nishimi, Ge, Coleman, Chen, Ratanatharathorn, Zheutlin, Dunn, Koenen and Smoller2021a) in the prevention of depression. For individuals with schizophrenia, physical activity interventions are best integrated into other pro-social activities: walking a dog, walking with a neighbour, joining a walking/cycling group or team sport. We need to make opportunities to ask the difficult question: ‘is it ok if we weigh you today?’ There are multiple pathways to obesity (Fig. 3), and healthy weight management requires the best of personalised medicine to maintain current weight, with motivational interviewing to plan sustainable reductions, if agreed. You will not ‘fat shame’ patients into losing weight. Engage and pair other gains (learning one new recipe a week, quitting snacks, eating earlier in the day, improved quality sleep, peer support) that modify diet with the euphoric effects of regular exercise. Sadly, the pandemic has widened inequalities in identifying an increased gradient of more exercise and healthier eating, with less alcohol and substances, in people with greater resources (Marmot et al. Reference Marmot, Allen, Goldblatt, Herd and Morrison2020). For health professionals, personalised medicine demands more clinical time and effort in the patients with lower financial resources. What began as collaborations among Glasgow GPs serving the most deprived areas of Scotland, Deep End, has been emulated internationally including in Dublin (Kiely et al., Reference Kiely, Clyne, Boland, O’Donnell, Connolly, O’Shea and Smith2021).

Fig. 3. Solutions across systems to improve Healthy Weight Management (HWM) over the life course.

One size does not fit all

Unlike the solutions in Fig. 2 driven by indisputable facts (smoking and alcohol excess shorten lives; public smoking bans and raising alcohol prices reduce population consumption) there are less certainties in healthy eating. The challenges in the UK and Ireland of how we get our food are similar: oligopolies that provide and disproportionately discount processed foods high in FSS. Over decades, portion sizes have increased as two thirds of the whole population became overweight, our bodies drive higher calorie intake (and thereby higher sales of unhealthy food) to maintain that weight (Marmot et al. Reference Marmot, Allen, Goldblatt, Herd and Morrison2020). We walk less (with the exception of wilful exercise in people who have adequate leisure time) in crowded cities and more remote, car-dependent rural settings: our environments are obesogenic. Inequalities compound high FSS diets and drive lower levels of exercise (Marmot et al. Reference Marmot, Allen, Goldblatt, Herd and Morrison2020). Solutions here appear more universal than selected (Fig. 3). The model does not assume poverty or childhood obesity always lead to increased weight, or that better off populations are spared – the universal challenge remains our obesogenic environment. This is not a crisis but an opportunity to engage across society to reverse the shift to the new normal of overweight. Changes to taxation will be part of this, more so if the supermarkets do not engage, but this will be more challenging than cigarette taxes/alcohol pricing as we all need food. The success of the UK soft drinks industry levy in 2018 was not that people bought less of them or profits fell but that companies were nudged into the production of lower concentration sugar products – and total sugar consumption fell (Pell et al., Reference Pell, Mytton, Penney, Briggs, Cummins, Penn-Jones, Rayner, Rutter, Scarborough, Sharp, Smith, White and Adams2021). Sadly, this tax is one of the rare successes in English anti-obesity measures over 30 years (14 Government strategies, 689 policies) that have been mostly evidence-free with scant learning from previous attempts, with a focus on individuals rather than the obesogenic environment: 13/14 strategies recognised inequalities as a driver of obesity but only 19% of policies were likely to reduce their impact (Theis & White, Reference Theis and White2021).

Mental health subpopulations’ weight

There are additional challenges in SMI: increased calorie intake and unhealthier foods (processed foods, sugary drinks) paired with more sedentary behaviours (Teasdale et al., Reference Teasdale, Ward, Samaras, Firth, Stubbs, Tripodi and Burrows2019). Obesogenic prescribing of psychoactive medications (see below) is the icing on the cake. When overweight develops, shame and/or osteoarthritis limit movement and other morbidities may have revealed themselves: our medical services from endocrinologists to bariatric surgeons need psychiatrists’ guidance to tailor effective treatments. These services are underdeveloped and as they evolve, we need to make sure our patients are treated fairly, with our input as standard to reduce psychological barriers to life-changing treatments. The services should align with disordered eating, and even among the grim data of Tables 1 and 2, anorexia stands out. There is an additional challenge that pre-pandemic, eating disorders have grown in prevalence and complexity in developed countries (Santimauro et al., Reference Santimauro, Melen, Mitchison, Vos, Whiteford and Ferrari2021), and lockdowns yielded further rises. Useful public health interventions to protect health in most mental disorders (calorie labelling, universal promotion of exercise) may have negative unintended consequences for this patient group. Anorexia treatment requires integrated care: physicians, psychiatrists and other disciplines working together. Parity demands the same resources for mental disorders as those for physical health: and that our patients get equal consideration for physical healthcare.

Rising opioid misuse

There is no evidence to justify opioids in noncancer pain; the only effective drugs in chronic primary pain are antidepressants (Carville et al., Reference Carville, Constanti, Kosky, Stannard and Wilkinson2021). Over the counter (OTC) codeine, poor self-report data (not least obtaining legal opioids by any means) and appropriate opioid prescribing in cancer patients have obscured the challenges of opioid misuse in many countries. The US has the highest prescription rate in the world, with related abundant illegal synthetic fentanyl sources; it experienced further rises since Covid: estimated opioid-related deaths for 2020 are 90 000 (Baumgartner & Radley, Reference Baumgartner and Radley2021). The UK Biobank Study (over 466 000 participants) reported regular opioid use in 5.5%, noting study volunteers as significantly healthier than the general population: though they cannot imply causation, 6.9% and 9.1% of light and heavy opioid users. respectively had died by follow-up; 3.3% of controls died (Macfarlane et al., Reference Macfarlane, Beasley, Jones and Stannard2020). Jani et al. (Reference Jani, Birlie Yimer, Sheppard, Lunt, Dixon and Song2020) excluded OTCs and prescriptions for cancer patients to report almost 2 million new opioid scripts in UK primary care over 12 years. There were linear relationships with age and inequalities (more opioids in poorer and older people, replicated in Ireland) (Norris et al., Reference Norris, Smith, Doran and Barry2021), higher prescribing rates in people with alcohol and liver problems (Fig. 1), and exponential increases over the study period of scripts for codeine, tramadol and oxycodone of respectively 5, 7 and 30-fold (Jani et al., Reference Jani, Birlie Yimer, Sheppard, Lunt, Dixon and Song2020). Long-term opioid use wrecks lives, more so in the 15% who become dependent, and clouding with deaths by deliberate poisoning and fear of legal proceedings drive up underreporting. Seyler and colleagues (Seyler et al. Reference Seyler, Giraudon, Noor, Mounteney and Griffiths2021) have looked at opioid deaths across Europe and with exceptions (synthetic fentanyl in Estonia, very high UK opiate deaths), concluded that there is no European opioid epidemic. While steady official death rates reassure some, European police drug seizures show steep rises in fentanyl, its derivatives and other novel opioids. Another problem postponed is that first prescribers in primary care and pain clinics lack training in reducing opioid doses and opioid substitution therapy (OST). One US survey reported a lack of addictions training in 38%, with no buprenorphine training in 92% (Kirane et al., Reference Kirane, Drits, Ahn, Kapoor, Morgenstern, Conigliaro and Enden2019). An EU study of 10 countries, then including UK, showed low confidence and wide training variations in OST (Fischer & Strover, Reference Fischer and Strover2012). Although opioid misuse has considerable global impacts, the administration of OST is sporadic and inconsistent with the evidence (Jin et al., Reference Jin, Marshall, Degenhardt, Strang, Hickman, Fiellin, Ali, Bruneau and Larney2020).

Self-harm and risk

In psychosocial assessments of people who self-harm, the focus is on reducing subsequent self-harm and the prevention of suicide. This population also carries higher risks of nonsuicide deaths: even at 2–10 years follow-up, natural deaths were 2–7.5 times more frequent (Bergen et al., Reference Bergen, Hawton, Waters, Ness, Cooper, Steeg and Kapur2012). Key findings were rising premature death rates, mostly from (preventable) circulatory diseases and alcohol-related liver damage, strong associations with inequalities, and the highest standardised mortality rates for accidental drug poisoning (Bergen et al., Reference Bergen, Hawton, Waters, Ness, Cooper, Steeg and Kapur2012). This 33-year Danish cohort study (62 922 discharged psychiatric patients; 1.5 m controls) reported high suicide rates peaking in the first 3 months post discharge, comparable rates of accidental deaths across 10 years (1 in 50 men; 1 in 200 women), and significant violent events (as victim and perpetrator): they link violence with alcohol/substance misuse and hazardous social environments. In public health terms, this subpopulation faces an iron triangle of interrelated poverty-inequality, substance use and impulsivity. Moeller et al. (Reference Moeller, Barratt, Dougherty, Schmitz and Swann2001) provide biopsychosocial perspectives on impulsivity, with a focus on ADHD, bipolar disorder and borderline personality disorders (though few clinicians outside the US will follow their guidance to prescribe to reduce impulsive behaviours), but recent work has moved the focus onto depression (Fields et al., Reference Fields, Schueler, Arthur and Harris2021) and a clearer understanding of disinhibition in people with schizophrenia (Starc et al., Reference Starc, Murray, Santimauro, Savic, Diehl, Cho, Srihari, Morgan, Krystal, Wang, Repovs and Anticevic2017). Impulsivity, what patients call the ‘moments of madness’ in which self-harm and other life-changing behaviours happen, is substantially state-dependent and our services work with patients to reduce anxiety, improve mood and psychotic symptoms, curtail alcohol, encouraging problem solving where possible. In our clinics, we need the time and the therapies to support patients’ self-management of impulsivity-disinhibition given both the risk of self-harm and interactions with drivers of adverse outcomes (Fig. 1).

Prevention of physical diseases and suicide

With high clinical caseloads and faster turnovers, the enemy here is ‘either/or’ thinking: either we focus on suicide/self-harm risks or we attend to physical health. Globally, we acknowledge the tragedy of suicide: in 2010, 884 000 deaths across all ages, 5–6% of everyone aged 15–49 who died that year (Chesney et al., Reference Chesney, Goodwin and Fazel2014). That same year reported 232 000 premature deaths in people with diagnosed mental disorders, that is, 8.6 million years of life lost (Chesney et al., Reference Chesney, Goodwin and Fazel2014). Accepting data gaps and differences in reporting/collection across countries, this meta-analysis pooled suicide and premature mortality data for the ¼ million deaths in 1.7 million people with known mental disorders (Table 2). The inclusion of deaths in people with addictions, ID and personality disorders, with comparisons to smokers, highlights the impact of these disorders. Some prevalence rates (alcohol misuse, depression) look underreported, and have likely increased since this analysis; ICD11 reclassifications will make scrutiny of personality disorder outcomes harder. Negotiators for more resources in mental health services can use Table 2 to argue for resourced perinatal and early intervention teams, as well as the protection of addiction services – using additional data on local fatalities. Three points about actively protecting physical health need repeating:

• these actions prevent mental disorders (primary prevention)

• the same actions reduce mental disorder symptom burden (in degree and duration) as well as alcohol misuse and multimorbidity (secondary prevention) and

• engagement with mental health and substance misuse teams reduces risky behaviours (tertiary prevention).

Table 2. Mortality risk in specific mental disorders compared to heavy smoking: adapted from Table 4 of Chesney and colleagues  (Chesney et al. Reference Chesney, Goodwin and Fazel2014)

All three prevent suicide alongside universal Public Health interventions: modifications at suicide hotspots, restrictions on paracetamol sales, restrained media reporting. In this context, it should never be a choice either to reduce suicide risk or improve total health (physical and mental). The scale of at risk populations is greater than existing mental health services’ capacity and we need to train others, use task sharing, and devise novel care pathways to reduce morbidity and mortality.

Integrated care

The latest reconfiguration of health services in England involves the formation of integrated care systems (ICS) – to coordinate care and reduce deferred decisions/missed opportunities to halt LTCs. Psychiatrists are unique among clinicians in hearing more ‘secrets’ from our patients and being ‘let in’ to try to understand the chaos. We get to interrogate the why and how (antecedents and triggers) of addictive and/or risky behaviours. Psychiatric teams already intervene to reduce and mitigate these (Figs. 1–3 and Supplementary Figure 1): long term engagement (stable health professional relationships especially when there are few other constants) makes a real difference to people’s lives. It is the last piece of the jigsaw of what we now know about the preventable premature mortality in our patients: we are uniquely placed to reduce polypharmacy that increases risks of early death (Masand & Gupta, Reference Masand and Gupta2002; Nasrallah, Reference Nasrallah2003; Nihalani et al. Reference Nihalani, Schwartz, Siddiqui and Megna2011; Macloud et al., Reference Macloud, Steer, Tilling, Cornish, Marsden, Millar, Strang and Hickman2019). Our skill too is finding ways to communicate what works, and agreeing relapse prevention plans in the heterogeneous population of people with mental disorders, addictions and ID. Psychiatric services have specialised based on age and disorder, but a transdiagnostic approach to psychiatric symptoms is needed. In addition to the textbooks and Lancet Commmission (Firth et al. Reference Firth, Siddiqi, Koyanagi, Siskind, Rosenbaum and Galletly2019) cited in the second paragraph, WHO (2008) have published their priority recommendations to prevent and intervene. We work across many teams and watching the seven drivers (Fig. 1) in every patient’s management plan benefits individuals and signposts training gaps. At minimum, everyone deserves a medication review, but early, coordinated action will ensure holistic care (Table 3). And if this approach sounds paternalistic, this is never about limiting personal choice but clinicians supporting patients to achieve or maintain agency. This is what Amartha Sen advocates (as quoted by Marmot, Reference Marmot2015: 76):

Table 3. Top 10 actions to reverse rising premature mortality, clinic-based to local services and public health