Guest blog by Dr Mary Obele
The health of the UK workforce is a major policy concern. But, less attention has been given to a more basic question: do our health data systems actually record what people do for work? Work is one of the most powerful influences on adult health, but it remains largely invisible in routine health data.
In occupational medicine the relationship between work and health is rarely abstract. Patterns emerge repeatedly in clinical practice. Musculoskeletal problems are common in physically demanding roles. Mental health difficulties may reflect organisational pressures, workload, job insecurity or wider financial stress. Respiratory disease can follow years of exposure in specific industries. At the same time, many of these conditions also arise outside work. Sports injuries, accidents at home, caring pressures, housing stress and financial strain all shape health outcomes. The challenge is not to assume that work explains everything. It is that we currently lack the data needed to understand how work interacts with these other determinants of health.
A fragmented picture - Systematic information about occupation and workplace exposure is inconsistent. In many primary care records, the occupation field is either empty or recorded as free text that cannot easily be analysed. As a result, links between employment and health outcomes remain difficult to study at scale. Several data sources provide partial insights:
- The Labour Force Survey collects self-reported information on work-related illness, although declining response rates have affected data quality in recent years.
- The Health and Safety Executive supports surveillance systems such as THOR, where clinicians report occupational disease.
- Employers and insurers hold data on sickness absence and compensation.
- National datasets held by the Office for National Statistics already link census, employment and some health records within secure environments.
Each of these contributes valuable information. None provides a comprehensive, routinely accessible picture of how work influences population health.
The result is a fragmented evidence base. Policymakers debate workplace health, sickness absence and labour market participation while the underlying data about work itself remains incomplete.
Policy without measurement - Many current policy discussions assume that the health system already understands work. Debates around Fit Note reform, workplace health strategies and initiatives such as Keep Britain Working all rely on assumptions about how illness affects employment and how workplaces shape health outcomes. Yet the data needed to answer even basic questions is often missing. For example:
- Which occupations experience the highest rates of long-term sickness absence?
- Where does good work appear to protect health?
- Which industries show patterns of early exit from the workforce due to illness?
Without systematic occupational data, these questions are difficult to answer reliably.
The practical barriers - Collecting occupational information sounds straightforward, but the operational reality is more complex. General practitioners work under significant time pressure. Additional data entry during consultations competes directly with clinical care. When occupational recording is optional, completion rates tend to remain low.
Expecting clinicians to record detailed occupational histories during routine consultations is therefore unrealistic. A more practical approach would focus on capturing simple, structured indicators such as broad occupational categories or employment sectors.
Technology can help reduce the burden on clinicians. Digital systems such as the NHS App or patient portals could allow individuals to update occupation information before appointments, in much the same way that insurance applications already collect employment data. Information could also be confirmed periodically when patients log into GP systems or attend clinics. This shifts the data entry process towards patient-reported information while maintaining clinical oversight.
The challenge of classification - Even when occupations are recorded, standardisation presents difficulties. Job titles alone rarely capture actual working conditions. Two individuals with the same title may perform very different tasks in very different environments. Classification frameworks such as the UK Standard Occupational Classification (SOC) and the International Standard Classification of Occupations (ISCO) provide useful structure, but manually searching coding systems during consultations is impractical. Automated coding tools, natural language processing and linkage with employment datasets could significantly improve consistency and reduce the administrative burden. Even relatively high-level occupational categories would substantially improve the analytical value of national health datasets.
Linking what already exists - The UK already has strong foundations for better work-health data. National health records, administrative employment datasets, benefits data and major surveys all contain relevant information. The challenge lies less in creating new datasets and more in coordinating and linking those that already exist. Important linkage work has been undertaken within secure environments by organisations such as the Office for National Statistics and initiatives such as UK longitudinal linkage projects. However, access to these resources remains restricted and fragmented.
Strengthening secure data linkage frameworks and expanding responsible access for bona fide research and policy institutions would unlock much greater analytical value.
International examples demonstrate the potential of such approaches. In several Scandinavian countries, registry-based data systems allow researchers to examine the relationships between employment, health outcomes and social policy over long periods. These systems have provided powerful evidence to guide workplace regulation, prevention strategies and labour market policy.
Trust, privacy and the social contract - Workers may worry that linking health information with occupation could affect employability or insurance. Employers may be concerned about regulatory consequences if patterns of illness become more visible.
These concerns must be addressed through clear governance arrangements and transparent safeguards. It is important to distinguish between identifiable clinical information used for patient care and de-identified datasets used for research and surveillance. Trusted research environments and independent oversight mechanisms play a critical role in maintaining public confidence.
Public support also depends on demonstrating clear benefits. People already provide occupational information to insurers, banks and government surveys because the purpose is understood. Health data collection must similarly show how better information can improve prevention, workplace safety and health outcomes.
The COVID-19 pandemic provided a clear example. Linking occupational data helped identify key workers and prioritise protective measures. Demonstrating tangible public benefit is central to maintaining the social licence for data use.
Unequal access to occupational health - Large organisations often operate well-developed workforce health systems. Smaller employers frequently lack dedicated occupational health support or systematic workforce data. A national approach to occupational health data therefore needs to reflect this imbalance. Otherwise, the health experiences of large sections of the workforce will remain poorly represented.
From data gaps to policy infrastructure - Better occupational data would not only identify risks. It would also reveal where work supports health, highlighting sectors and practices associated with longer working lives, lower illness rates and improved wellbeing. The strategic question is therefore not simply whether occupational data should be recorded, but how the UK builds the infrastructure to use it effectively. Key elements of such an approach could include:
- Routine recording of occupation within electronic health records using broad UK Standard Occupational Classification (SOC) categories, with information captured or updated by patients through digital platforms such as the NHS App.
- Secure linkage between health, employment and administrative datasets to allow population-level analysis of work and health outcomes.
- Clear governance frameworks that separate identifiable clinical data from de-identified research data and maintain public trust.
- Incentives within health systems that support consistent recording of occupational information.
- Expanded access to secure data environments so bona fide researchers and policy institutions can analyse linked work and health data.
What is required now is coordination across health, labour market and data policy. Making work visible in health data is not a technical detail. It is a prerequisite for designing effective policy for a healthy, productive workforce.
Dr Mary Obele is a Specialist in Occupational and Environmental Medicine.