What can the NHS learn from other countries on workforce planning?

Flawed projections yield flawed insights

Workforce modelling and planning will never be an exact science because they require assumptions about how the supply and demand for services will evolve, and what that means for the number of staff needed over the long term. But a key risk when making these projections is that the data on which they rely is inaccurate—leading to imprecise or irrelevant recommendations.

In Australia, national projections have been limited by incomplete information on how clinicians balance different types of work (e.g. clinical versus non-clinical time), and how that affects the overall number of staff needed to meet patient need.2 The Australian health system has also traditionally relied on migration and locum workers to meet demand—but these workers are not always reliably accounted for within data, leading to underestimations of the number of staff required.

In Belgium, a key challenge has been distinguishing between doctors who are registered but not actively working, versus those who are in active practice (with the latter being much lower). Not being able to distinguish between the two in available data has led to inaccurate assumptions about the number of staff available and an undersupply of staff in some key roles down the line.3 The Netherlands addresses this problem somewhat by complementing routine data with staff surveys (at least for GPs) to have a fuller understanding of workloads and actual activity levels.4

Allow for flexibility

Given that projections must grapple with uncertainty, their effectiveness depends on how well they adapt to changes and make up for any imprecision in estimates.

This has been an important lesson from Belgium. There, quotas that limit number of licences available to new physicians can be revised every year in response to revised projection estimates and stakeholder input. And since quotas do not limit the number of training places, only the number of professional licences available, there is flexibility to ensure the required number of physicians can be filled as long as the number of graduates exceeds the quota previously fixed for that year.5

Projections no replacement for aligned policy-making

Even in countries with more robust approaches to workforce planning, shortages in key specialties and uneven distribution of personnel across country are consistent problems. One common issue is that most health workforce planning models are developed separately to policies affecting staff pay and conditions, education, migration, and retirement.6 In any country, workforce projections will only make a difference if acted upon, and can be a futile exercise if disconnected from broader decision-making affecting the recruitment and retention of staff. Indeed, this is a lesson from England’s own chequered history with using workforce projections, which have not always been followed or connected to broader policy thinking.7

In Belgium, a long term challenge has been meeting quotas for GPs while having an oversupply of some specialist groups, as doctors have tended to prefer specialist hospital work. While offering one of the more advanced models for workforce planning globally, the Planning Commission can still be disconnected from policy decisions that determine the pay and attractiveness of general practice relative to other professions / practice locations. Independent reviews of Belgium’s workforce planning approach have consistently recommended strengthening the Planning Commission’s mandate to more directly analyse and inform healthcare policies affecting overall workforce supply and retention.8

A lack of national coordination has also been an issue in Australia. National workforce plans have been undermined by siloed decision-making and key information gaps on matters like training numbers, workforce allocation and migration. This is because local jurisdictions have access to state/territory specific data and develop their own plans and projections for their specific purposes, which can be at odds with national modelling, and have contributed to fluctuations in the supply of medical staff.9 Efforts are in place to establish a new joint medical workforce planning and advisory body to better unite data and decision-making.

So what now for the NHS?

While it is important that the UK government has committed to establishing a more transparent and comprehensive approach to estimating staffing needs within the NHS, the experience of other countries makes clear that this will not solve all its problems. The value of projections depends not only on how well the model can predict and accommodate for different staffing scenarios in the short term, but also on its ability to influence and meaningfully inform policy direction in the long term. Let’s hope that the government doesn’t just stop at developing a model, but also thinks critically about how the recommendations will be implemented, and are balanced with financial and political trade-offs. The NHS may then finally be able to plan better for the future.