When work gets stressful, the standard advice is familiar: exercise more, eat better, sleep more and cut back on unhealthy habits. But our new research study suggests not all healthy habits offer the same protection from chronic work stress.Using data over 10 years from a long-running national survey of 2,871 Canadian workers, we examined whether five health-related behaviours outside work helped weaken the relationship between work stress and general health over time: nutrition, exercise, sleep quality, alcohol use and smoking frequency.What we found was more uneven — and more interesting — than the usual wellness advice suggests. Some behaviours appeared to offer real stress-specific protection. Others were linked to health overall, but did not seem to buffer the effects of work stress specifically. Some habits protect; others don’tSleep quality stood out most clearly. Nutrition also mattered. Exercise remained good for health overall, but did not buffer the health effects of work stress in the same way once the other behaviours were considered together.For many workers, work stress is chronic. It builds through heavy workloads, difficult or unpredictable schedules, after-hours emails and text messages, and the feeling that work keeps spilling into evenings, weekends and family time. Over time, that kind of stress can wear people down physically and psychologically. Research has linked work stress to burnout, depression, anxiety, fatigue, cardiovascular disease, Type 2 diabetes and mortality.Our study asked: when stressful work conditions persist, are there things people do outside work that actually help protect their health? Our findings suggest the answer is yes, but selectively.Sleep may matter more than people thinkSleep quality stood out as the strongest buffer against the health costs of work stress. Good sleep supports attention, emotional regulation, recovery and the self-control needed to maintain other healthy behaviours in the first place. In that sense, it functions less like one good choice among many and more like a foundational resource.Nutrition also showed a meaningful buffering effect, suggesting that diet may help sustain the physical and psychological reserves needed to cope with sustained strain.The exercise finding pushed against popular assumptions. While more frequent exercise was associated with better general health overall, it did not significantly weaken the relationship between work stress and health. This could reflect the way exercise was measured in the survey, or it could mean exercise helps health in ways that are real but not specifically stress-buffering. Being healthy and being protected from stress are not always the same thing.The alcohol finding was the most unexpected and warrants particular caution. Lower alcohol use was associated with better overall health, as expected. But the data showed that higher work stress was more strongly associated with poorer general health among people who reported lower alcohol use than among those who reported drinking more frequently.This should not be read as evidence that drinking protects people from the health effects of work stress, however. People who drank more frequently still reported worse overall health. More likely, this pattern reflects something our data could not fully unpack, such as prior health conditions, different coping profiles or non-linear patterns in alcohol use and health. Healthy habits don’t excuse unhealthy work designWhen work is chronically stressful, some forms of self-care may protect health more than others. Most importantly, wellness interventions cannot compensate for a job that is structured to exhaust people.Organizations are still responsible for designing healthy workplaces. Employees should not be expected to sleep or meal-prep their way out of excessive workload, unreasonable expectations or poor work design. What our findings suggest is not that individual behaviour replaces organizational responsibility. Rather, certain behaviours may help protect people when work remains stressful and structural change is absent, incomplete or slow to arrive.Our study is explicit that these behaviours should be understood as complementary to, but not substitutes for, broader organizational change.That has practical implications for both workers and employers. For workers, the message is not to do everything perfectly. It’s that some behaviours may offer more protection than others when work stress is high, and sleep deserves to be taken especially seriously.For employers, the lesson is not to moralize wellness or shift responsibility onto individuals. It’s to make protective behaviours easier to sustain by reducing after-hours communication, allowing real on-the-job breaks, improving scheduling and designing work in ways that do not erode recovery.Nick Turner receives research funding from Cenovus Energy Inc., Haskayne School of Business's Future Fund, and the Social Sciences and Humanities Research Council of Canada (SSHRC).Erica Carleton receives funding from the Social Sciences and Humanities Research Council. A. Wren Montgomery and Serra Al-Katib do not work for, consult, own shares in or receive funding from any company or organisation that would benefit from this article, and have disclosed no relevant affiliations beyond their academic appointment.