Executive summary
Workplace mental health has completed the journey from taboo to line item. Employers now fund manager training, expanded behavioral networks, digital care and peer programs at a scale that would have been unrecognizable a decade ago, and utilization is rising fast enough that most large employers count it among their cost drivers.12 What has not moved is who the system reaches. The further up an organization a person sits, the less usable the available support becomes, and the more expensive their untreated distress is to everyone around them.
An estimated 15 percent of working-age adults worldwide live with a mental disorder, and most of them are at work today.01
Employer programs are designed around disclosure, volume and convenience, while senior professionals face career, licensing and reputational costs for using anything the organization can see.16
A confidential clinical tier that sits outside the employer's own systems, matched to the pressures of the role rather than to a generic wellbeing curriculum.
Distress is treated in the phase where treatment is short and ordinary, rather than in the phase where it becomes a succession problem.
The problemA field that grew up around the wrong half of the org chart§
The scale is settled. The World Health Organization estimates that 15 percent of working-age adults had a mental disorder in 2019, and that depression and anxiety cost the global economy about 12 billion working days and one trillion dollars a year in lost productivity.01 In the United States, 23.1 percent of adults, some 59.3 million people, met criteria for any mental illness in 2022, and only 50.6 percent of them received any mental health treatment in the previous year.03 Ask workers directly and the numbers climb further: 76 percent reported at least one symptom of a mental health condition, and 84 percent said workplace conditions had contributed to at least one mental health challenge.02 None of this is contested any longer, which is precisely why the interesting question has changed. The question is no longer whether workplace mental health is a real problem. It is whether the response now being funded reaches the people whose deterioration costs the most.
The response has been substantial and it has been real. Nearly three quarters of large employers reported higher use of mental health and substance use services in 2025, and 72 percent deployed manager training to help supervisors recognize distress and route people toward care.12 That is a genuine achievement, and the evidence supports it as an organizational intervention.11 But every one of those mechanisms assumes the same three things: that someone above the affected person will notice, that the affected person can be seen using the resource without cost, and that the resource is deep enough for what is actually wrong. For a first-line employee, those assumptions frequently hold. For a chief executive, a partner, a department chair or a general counsel, they routinely fail, and they fail quietly, because the population that most needs help is also the population most practiced at not appearing to need it.
A workplace mental health program that senior leaders cannot safely use is not a program with a utilization problem. It is a program with a design problem. CEREVITY clinical whitepaper series, 2026
The evidenceWhat the research shows§
Four bodies of evidence have to be read together to see the shape of the field. The first is population prevalence, which sets the size of the need. The second is what employers are actually buying, which is now measurable year over year. The third is who uses what is bought, which is where the picture starts to break. The fourth is the seniority gradient running through all of it, which is visible in survey data as soon as anyone thinks to disaggregate by level.
15%
of working-age adults worldwide were estimated to have a mental disorder
WHO, 2019 estimate
76%
of US workers reported at least one symptom of a mental health condition
Mind Share Partners, via US Surgeon General, 2022
81%
of CEOs agree companies still view people with mental health issues as weak or burdensome
Businessolver, 2024
50.6%
of US adults with any mental illness received treatment in the past year
NIMH, 2022 data
Read in sequence, those four figures describe a funnel that narrows at exactly the wrong point. Prevalence is high and self-reported symptom burden is higher still. Employer spending has followed. Yet half of the American adults who meet criteria for a mental illness receive no treatment in a given year, and the group most convinced that help-seeking will be read as weakness is the group at the top of the organization. Gallup measured global engagement at 20 percent in 2025, its lowest level since 2020, with manager engagement falling from 31 percent in 2022 to 22 percent in 2025.05 US engagement stood at 31 percent in May 2026.06 Employers see the aggregate decline and fund broader awareness, which is the layer senior leaders are least likely to route through. What the aggregate hides is that the leaders being asked to deliver that awareness are themselves among the most distressed people in the building, and are the least able to say so. Long before this shows up in a wellbeing dashboard it shows up as work that is present but degraded, which is the part of the bill that never arrives as an absence: across eight countries, the costs associated with presenteeism ran five to ten times those associated with absenteeism.19
12 Business Group on Health. (2025). 2026 Employer Health Care Strategy Survey: Executive Summary.
07 Gallup. (2026). Leaders Have Better Lives, Worse Days. 2025 World Poll, n=9,880 leaders.
| Support layer | What it is built to do | Where the evidence says it helps | Where it stops for senior professionals |
|---|---|---|---|
| Manager training | Teach supervisors to notice distress and route people toward care. | The most widely adopted employer action for 2026, run by 72 percent of large employers, and a recommended organizational intervention.12, 11 | A chief executive reports to a board. There is no supervisor above them trained to notice, and disclosure travels upward into governance. |
| EAP counseling | Short-term, employer-funded sessions with an assigned counselor. | Broad reach at low marginal cost. Available to 40 percent of workers in establishments with fewer than 100 workers.14 | Session-limited, employer-contracted and visible on an employer invoice. The barrier attorneys report first is not wanting others to find out.16 |
| Health plan network | Covered outpatient behavioral care through the medical plan. | Seventy percent of firms offering health benefits say their largest plan gives timely access to mental health services.13 | Care generates a claim and a diagnosis code that can follow credentialing, licensure and disclosure questions for years. |
| Digital and app-based tools | On-demand skills content, screening and self-guided modules. | Scales instantly and reaches people who will not book an appointment. | Sixty-eight percent of workers did not use the full value of the wellbeing resources offered, calling access time-consuming, confusing or cumbersome.10 |
| Flexibility and leave | Time and autonomy to recover, protected by policy. | Addresses work design, one of the conditions workers name as causal.02 | Time away is the most legible signal a senior person can send. Taking it is read as a referendum on their capacity to hold the role. |
| Peer support and champions | Normalize the topic through visible colleagues and networks. | Adopted by 47 percent of large employers for 2026, and effective at shifting the conversation.12 | For an executive, the available peers sit on the compensation committee. Disclosure to peers is disclosure to the people who price the role. |
| Independent private-pay care | Confidential clinical treatment outside every employer system. | No claim, no employer record, and depth set by the clinical need rather than a session cap. | It has to be found rather than defaulted into, and the cost sits with the individual or with a deliberate organizational arrangement. |
The frameworkA model you can name and own§
Most debate about workplace mental health collapses into a single yes-or-no question: does the employer support mental health or not. That framing cannot explain why an organization with a well-funded, well-reviewed program still loses a partner to a breakdown nobody saw coming. A gradient is a better description than a binary. Support does not stop at a line; it thins, layer by layer, as the help being asked for becomes more clinical, and it thins faster the more senior the person asking. The model below names the four layers so that a benefits leader can say which one their organization actually reaches, and so that what structured burnout treatment actually involves can be separated from awareness work that shares its vocabulary.
CEREVITY model
The Coverage Gradient
A four-layer description of how workplace mental health support thins as the request becomes more clinical. Each layer reaches fewer people than the one above it, and the falloff is steeper the more senior the person. Naming the layers lets an organization say which one it actually funds.
Awareness
The organization says the words. Training, campaigns, champions and policy language reach almost everyone, cost little per head, and are the layer employers fund first.
Access
A route to care exists on paper: an assistance program, a plan network, an app. Coverage is wide, but it is coverage as entitlement, not as care delivered.
Utilization
Someone actually books, attends and returns. This is the first layer with a real cost of entry, and the first where being seen matters. The drop here is the largest in the model.
Depth
Treatment continues long enough and runs deep enough to change the underlying pattern rather than manage the week. Very few reach it, and almost no one reaches it through a channel their employer can see.
SCHEMATIC Schematic, not measured data.
Illustrative model. CEREVITY clinical whitepaper series, 2026.
The gradient is useful because it turns a vague complaint into a specific budget question. An organization that has funded awareness and access has bought two layers and can say so honestly. It has not yet bought utilization, and it has almost certainly not bought depth. For most of the workforce, the first two layers plus a functioning plan network do a great deal of good. For the twenty or thirty people whose judgment the organization actually depends on, the first three layers are structurally unavailable, and the fourth is the only one that was ever going to work.
By professionHow it presents across roles§
The gradient is universal but its steepness is not. What changes by profession is the price of being seen. In some fields the cost of disclosure is reputational, in others it is regulatory, and in others it is written into the licensing questionnaire. Three well-studied populations show the range, and each has a published evidence base rather than anecdote behind it.
Senior executives and the C-suite
The executive picture is the clearest illustration of why aggregate wellbeing metrics mislead. Gallup's 2025 World Poll found that leaders who manage other managers report higher life evaluation than individual contributors, 43 percent thriving against 32 percent, and simultaneously report more daily stress, 46 percent against 39 percent, more anger, 33 against 21, more sadness, 34 against 23, and more loneliness, 31 against 21.07 A dashboard that tracks thriving alone will report that leadership is fine. Both readings are true at once, and that is the point: senior roles combine genuine reward with a level of daily emotional load that no other group in the organization carries. The disclosure math is equally stark. In a 2024 study of more than 3,000 employees, HR leaders and CEOs, 55 percent of CEOs said they had experienced a mental health issue, up 24 points year over year, while 81 percent of CEOs agreed that companies view people with mental health issues as weak or burdensome, a higher share than among HR leaders or employees.08 The people best placed to normalize help-seeking are the most convinced it will be held against them. Deloitte found the same asymmetry from the other direction: 91 percent of the C-suite believed employees felt they cared about wellbeing, while only 56 percent of employees agreed, and nearly 70 percent of executives were seriously considering quitting for a job that better supported their wellbeing.09 Across the CEREVITY network, this segment more often presents as depletion carried competently for years than as an acute episode.
Individual therapy built for chief executives
Organizational executive mental health benefit for companies
Physicians and health care workers
Health care is the sector where the workplace mental health response has been most serious and the structural barrier most explicit. The direction of travel is genuinely good: physician burnout measured by the American Medical Association fell from a record 62.8 percent in 2021 to 53 percent in 2022 and 48.2 percent in 2023, the first time below half in four years, drawn from more than 12,400 physician responses across 81 health systems. Job satisfaction rose to 72.1 percent over the same period, and the share who felt valued by their organization rose to 50.4 percent.15 That is what a decade of sustained organizational effort looks like, and it should be said plainly. The floor underneath it is still not solid. CDC data show that health workers reporting burnout often or very often rose from 32 percent in 2018 to 46 percent in 2022, poor mental health days in the past month rose from three to five, and the share who had looked for a new job rose from 33 percent to 44 percent.04 Harassment at work doubled from 6 to 13 percent, and those who experienced it reported anxiety at 85 percent against 53 percent among those who did not.04 The disclosure problem here is not reputational, it is administrative: a claim, a diagnosis code and a medication history can surface in credentialing and licensure review for years afterward. That is why physicians disproportionately seek care that leaves no insurance record, and why a hospital's own program, however well built, is the last place many of them will use.
Individual physician-specific psychotherapy
Organizational organizational physician burnout solutions for health systems
Attorneys and legal professionals
The legal profession has the longest-running quantitative evidence base of any white-collar sector, and it reads as a warning. The 2016 study of 12,825 licensed, employed US attorneys found 28 percent reporting mild or higher levels of depression, 19 percent anxiety and 23 percent stress, with 21 percent scoring at a level consistent with problematic drinking, rising to 32 percent among those aged 30 or younger. Only 7 percent had sought help for substance use and 37 percent for a mental health concern, and the barriers respondents named, in order, began with not wanting others to find out, followed by confidentiality and concerns about the effect on their license.16 A 2021 study of 2,863 lawyers randomly sampled from two bars found that 24.2 percent of women and 17.4 percent of men had contemplated leaving the profession because of mental health concerns, stress or burnout.17 A 2023 study of 1,962 lawyers from the same bars found 8.5 percent reporting thoughts of being better off dead or of self-harm several days or more, with high perceived stress associated with roughly 22 times the odds of suicidal ideation, loneliness with 2.8 times the odds and work overcommitment with 2.2 times.18 Loneliness and overcommitment are not incidental findings in a profession organized around billable hours and adversarial isolation. For equity partners and general counsel the confidentiality concern is not abstract, because the same bar that licenses them administers the assistance program many firms point to.
Individual attorney-specific psychotherapy
Organizational an EAP alternative for law firms
The stakesThe cost of inaction§
The cost of an unreached senior population does not arrive as a mental health line item. It arrives as capacity that is present but degraded, as departures that were decided months before they were announced, and as conditions that were cheap to treat in year one and are not in year three. The detailed modeling of those first two belongs in the companion papers on presenteeism and senior turnover. What follows is the shape of the exposure, in the terms an operating committee already uses.
Capacity that is present but degraded
Depression and anxiety account for an estimated 12 billion lost working days a year worldwide, at about one trillion dollars in lost productivity.01 Most of that loss is not absence. Across eight countries the costs associated with presenteeism ran five to ten times those associated with absenteeism, and the United States showed one of the widest ratios in the sample.19 Senior roles are where this is hardest to see and most expensive, because the output being degraded is judgment rather than throughput.
Departures decided long before they are announced
Intent to leave is measurable well ahead of attrition and it runs highest at the top. Nearly 70 percent of C-suite executives were seriously considering quitting for a job that better supported their wellbeing, against 57 percent of employees.09 The share of health workers who had looked for a new job rose from 33 percent in 2018 to 44 percent in 2022.04 Roughly a quarter of women attorneys and 17 percent of men had contemplated leaving the profession over mental health.17
Conditions that get more expensive with time
Half of American adults with any mental illness received no treatment in the previous year.03 In the legal data the untreated share is larger still, with 37 percent having sought help for a mental health concern and 7 percent for substance use.16 Untreated conditions do not hold steady. They compound into sleep loss, relationship strain, substance use and, in the lawyer suicide data, ideation at 8.5 percent with high perceived stress carrying roughly 22 times the odds.18
09, 04, 17 Deloitte and Workplace Intelligence. (2022). n=2,100.
CDC. (2023). Vital Signs: Health Worker Mental Health.
PLOS ONE. (2021). Stress, drink, leave. n=2,863.
The solutionWhat effective care looks like§
Effective care for the population the gradient leaves behind has four properties, and none of them is exotic. It has to be confidential in a structural sense rather than a policy sense, meaning there is no employer record to be discreet about. It has to be clinically real, delivered by an independent licensed clinician rather than a coach or a content library. It has to be matched to the pressures of the role, because a person whose decisions carry other people's livelihoods is not helped by generic stress content. And it has to be available in more than one shape, because a weekly appointment and a single concentrated block of work solve different problems. That last point is why the treatment philosophy behind it starts with session depth rather than session count, and why the weekly hour is treated as one option among several rather than the default unit of care.
CEREVITY is built as a nationwide network of independent licensed clinicians, matched to the person rather than assigned by rota, delivered by secure video on a fully private-pay basis so that no claim, diagnosis code or employer invoice is generated. Sessions run in three formats. Where a pattern has been carried for years and needs sustained work, the extended format gives room for material that an hour keeps interrupting. Where a leader has a narrow window and a decision that will not wait, the intensive format concentrates the work into a single block. None of this depends on a single clinician being available; the network is the capability, and matching is done against the pressures of the role as well as the presenting concern.
ImplementationHow to put it into practice§
The gradient is diagnosable inside a single quarter, using data most organizations already hold. What follows is a sequence a benefits leader or a board committee can run without commissioning new research, and without asking any individual to disclose anything.
- 01
Map your own coverage against the four layers
List every mental health surface the organization funds and assign each one to awareness, access, utilization or depth. Most inventories turn out to be heavily weighted toward the first two, which mirrors the market: 72 percent of large employers ran manager training for 2026 while 44 percent worked with their health plan to expand the clinical network those managers refer into.12 The imbalance is not a failure, but it should be named before anything is added.
- 02
Measure utilization by level, not in aggregate
Ask the vendor for utilization broken out by band or grade, with cells suppressed below a threshold that protects identity. If senior utilization cannot be reported at all, that is itself the finding. Aggregate utilization rates conceal the exact gap this paper describes, because senior populations are small enough to disappear into a company-wide percentage.
- 03
Test the confidentiality claim as a leader would
Trace what a hypothetical senior user would leave behind: whether a claim is generated, who holds the record, whether the vendor is contracted by the employer, and whether the utilization report could ever be small enough to be re-identified. The barriers named most often in the attorney data were not wanting others to find out and confidentiality, ahead of cost.16 A program that cannot pass this trace will not be used by the population it most needs to reach, regardless of its clinical quality.
- 04
Fund the depth layer deliberately, and separately
The fourth layer is the one that cannot be bolted onto the first three, because its defining property is that it sits outside the employer's systems. Fund it as a distinct arrangement with its own confidentiality terms, size it to the small population it is for, and resist the instinct to report on it in the same dashboard as the rest. Then leave the first three layers alone: they are doing real work for the majority of the workforce.
RecommendationsWhere to start§
Clinical
Treat seniority as a clinical variable
Distress in a senior role presents late, presents as competence, and is described in operational language before it is described in clinical language. Assessment should account for the concealment cost the role imposes, not only for symptom count. High life evaluation and high daily distress coexist in this group.07
Clinical
Screen for isolation, not only for mood
Loneliness carried 2.8 times the odds of suicidal ideation in the lawyer data, and leaders report loneliness at 31 percent against 21 percent for individual contributors.18, 07 Where mood is the presenting complaint, private treatment for anxiety and depression should still be sequenced alongside work on the structural isolation of the role.
Structural
Keep the awareness layer and stop asking it to do everything
Manager training, champions and policy language are cheap, popular and genuinely useful, and the evidence supports them as organizational interventions.11 They are not a substitute for clinical depth, and the failure mode is treating high adoption of layer one as evidence that layer four is covered.
Structural
Add a confidential tier rather than replacing what works
For most of the workforce an assistance program plus a functioning plan network is a reasonable answer, which is why the honest comparison between the two models matters more than a verdict. The senior tier is an addition, sized to twenty or thirty people, not a replacement for a benefit that serves thousands.
FAQCommon questions§
What is mental health in the workplace?
Why do workplace mental health programs miss senior executives?
What should employers do about the gap?
How does private-pay billing work?
How is my privacy protected?
MethodologyHow this paper was built§
Methodology
This paper is a narrative evidence review, not a systematic review, and it should be read as a map of the current landscape rather than as a meta-analysis. Sources were identified between June and August 2026 across four categories: multilateral and government statistical series, including the World Health Organization, the National Institute of Mental Health, the Centers for Disease Control and Prevention, the Office of the US Surgeon General and the Bureau of Labor Statistics; peer-reviewed literature located through PLOS, MDPI and Springer; large recurring employer and workforce surveys from Gallup, Deloitte, KFF, Business Group on Health and Businessolver; and professional-association research from the American Medical Association and the legal profession studies published through the National Conference of Bar Examiners. Every external figure in this document carries a numbered superscript that resolves to the reference list, and every reference was retrieved and read in full during the review window. Where a figure originates in one body of research and is reported by another, both are named in the text. Sample sizes vary widely and matter to interpretation. The attorney prevalence figures come from a 2016 study of 12,825 licensed, employed attorneys; the attrition figures from a 2021 study of 2,863 lawyers; the suicide-risk figures from a 2023 study of 1,962 lawyers, all drawn from two US bar populations rather than from a national frame. The physician burnout series reflects more than 12,400 responses across 81 health systems in 2023 and is not a random national sample. The executive wellbeing comparison covers 2,100 respondents split evenly between C-suite and employees in four countries in February 2022. The leader emotion comparison covers 9,880 leaders in Gallup's 2025 World Poll. Four limitations deserve to be stated plainly. First, most workplace mental health data are self-reported and cross-sectional, so they establish association and prevalence rather than causation. Second, response bias is likely to run in opposite directions by seniority: distressed senior professionals are the group least likely to complete a survey that names them. Third, several of the most useful employer surveys are proprietary, conducted by organizations with a commercial interest in the topic, and are reported here as what employers say they do rather than as audited practice. Fourth, prevalence instruments differ across the studies cited, so figures are not directly comparable across professions and are presented alongside their populations rather than pooled. No CEREVITY internal data are quantified in this paper. Where the text describes patterns observed across the CEREVITY network, those are qualitative clinical observations offered as such and not as measured findings. Figures 1, 2 and 4 chart external published data only, with their sources named on each figure. Figure 3 is an unmeasured schematic and is labeled as one.
References
- 01World Health Organization. (2024). Mental health at work. WHO fact sheet, 2 September 2024. who.int
- 02Office of the US Surgeon General. (2022). The US Surgeon General's Framework for Workplace Mental Health and Well-Being. US Department of Health and Human Services. hhs.gov
- 03National Institute of Mental Health. (2024). Mental Illness. NIMH Health Statistics, 2022 NSDUH data. nimh.nih.gov
- 04Centers for Disease Control and Prevention. (2023). Vital Signs: Health Worker Mental Health. cdc.gov
- 05Gallup. (2026). State of the Global Workplace. gallup.com
- 06Gallup. (2026). Indicator: Employee Engagement. gallup.com
- 07Gallup. (2026). Leaders Have Better Lives, Worse Days. 2025 World Poll, n=9,880 leaders. gallup.com
- 08Businessolver. (2024). 2024 State of Workplace Empathy Study. n=3,000+ employees, HR professionals and CEOs, fielded March 2024. businessolver.com
- 09Deloitte and Workplace Intelligence. (2022). The C-suite's role in well-being. n=2,100, fielded 8 to 21 February 2022. deloitte.com
- 10Deloitte. (2023). Workplace well-being research. n=1,274 US workers, fielded November 2022. deloitte.com
- 11World Health Organization. (2022). WHO Guidelines on Mental Health at Work. Published 28 September 2022. who.int
- 12Business Group on Health. (2025). 2026 Employer Health Care Strategy Survey: Executive Summary. businessgrouphealth.org
- 13KFF. (2025). 2025 Employer Health Benefits Survey. kff.org
- 14US Bureau of Labor Statistics. (2025). Employee Benefits in the United States, March 2025. News release, 25 September 2025. bls.gov
- 15American Medical Association. (2024). Physician burnout rate drops below 50% for first time in 4 years. AMA Organizational Biopsy, n=12,400+ physicians, 81 health systems, 2023. ama-assn.org
- 16National Conference of Bar Examiners. (2016). Wellness and the Legal Profession: Implications of the 2016 Landmark Study on the Prevalence of Substance Use and Mental Health Concerns Among US Attorneys. The Bar Examiner. n=12,825. thebarexaminer.ncbex.org
- 17Stress, drink, leave: An examination of gender-specific risk factors for mental health problems and attrition among licensed attorneys. (2021). PLOS ONE, 16(5), e0250563. n=2,863. journals.plos.org
- 18Stressed, Lonely, and Overcommitted: Predictors of Lawyer Suicide Risk. (2023). Healthcare, 11(4), 536. n=1,962. mdpi.com
- 19Evans-Lacko, S., and Knapp, M. (2016). Global patterns of workplace productivity for people with depression: absenteeism and presenteeism costs across eight diverse countries. Social Psychiatry and Psychiatric Epidemiology, 51. link.springer.com
PsyD, Licensed Psychologist
Maria Gonzalez, PsyD is a clinical psychologist licensed in California, New York and Massachusetts, seeing clients by telehealth through CEREVITY's nationwide network of independent licensed clinicians. She works with executives, entrepreneurs, attorneys and accomplished professionals questioning long-held assumptions, on career and identity transitions, on burnout, anxiety and overwhelm, on the particular pressures carried by first-generation professionals, and on grief, loss and relationship strain. Her clinical work draws on psychodynamic therapy, narrative work and acceptance and commitment therapy.
Talk to someone who gets it.
If this paper describes something you recognize, a confidential conversation is the next step. CEREVITY matches you to an independent licensed clinician who works with people in your position.
Schedule consultationPrivate-pay, telehealth, nationwide. Questions: (562) 295-6650