The hidden work in health care: unpaid overtime in the NHS

The NHS relies on a workforce of around 1.5 million people to deliver care in England. The NHS annual staff survey shows that for at least the past 20 years, half or more of these people have worked additional unpaid overtime, above and beyond their contracted hours – often referred to as ‘discretionary effort’. This has consistently allowed the health service to treat more patients and offer better quality care.

The COVID-19 pandemic appears to have been a watershed moment for unpaid overtime in the NHS (as it was for many other areas). The Darzi review noted a ‘marked reduction in discretionary effort across all staff groups’, after the stress and exhaustion that NHS workers had endured throughout the pandemic. An NHS England public board paper (from May 2024) also considered how this reduction in overtime might be affecting overall NHS productivity. And the 10 Year Health Plan also highlights a dip in intrinsic motivation as evidenced by falls in discretionary effort. 

But understanding the changes in NHS staff unpaid overtime is complex. Whether people work extra hours depends not only on their willingness or motivation to do so, but also on the demand for their time – driven by patient needs and gaps in staffing. The publicly available data do not allow us to fully assess the relative contribution of each, the interaction between the two, or the impact that working unpaid hours can have on staff themselves. 

How many unpaid hours are NHS staff working?

Figure 1 shows the trend in unpaid hours worked (per week) in the NHS in England since 2019. In 2021, at the height of the pandemic, 57% of NHS staff reported working at least some unpaid overtime – a slight increase from 56% in 2019. Since then, this percentage has steadily declined, falling below pre-pandemic levels to 53% in 2023 and 50% in 2024.

The percentage of staff working 6 or more unpaid hours per week follows a similar pattern, rising from 13% in 2019 to 14% in 2021, before dropping to 12% in 2023 and 11% in 2024.

Other indicators from the NHS staff survey suggest several factors that might contribute to these patterns. The percentage of staff who report being ‘satisfied’ or ‘very satisfied’ with how valued they feel (by their organisation) dropped from 48% in 2019 to 44% in 2024. This decline is consistent with the Darzi review – which finds increasing disengagement among NHS staff, leading to a reduction in discretionary effort. 

However, there are also signs that some pressures on staff may be easing, potentially reducing the need for unpaid hours. For example, the percentage of respondents stating that their organisation ‘does not have enough staff to do their job properly’ has returned to pre-pandemic levels. It was 45% in 2019, increased to 54% in 2022, and fell back down to 44% in 2024 (though of course this is still worryingly high). 

We can estimate what these survey results imply for the total number of unpaid hours across the NHS by applying the percentages from Figure 1 to the total NHS staff headcount. To illustrate the scale, we express these estimates in terms of full-time equivalents (FTE = 37.5 hours per week).

Figure 2 presents the total number of unpaid hours worked by NHS staff in England from 2019 to 2024. The red columns give a very conservative estimate, assuming respondents worked the minimum number of unpaid hours in their response category (0, 6, 11 hours).

Under this assumption, the unpaid hours worked in 2021 were equivalent to the contracted annual hours of 38,500 full-time employees. By 2024, this had fallen to the equivalent of 33,000 full-time employees. 

However, these numbers are likely to be an underestimate, particularly for those selecting ‘up to 5 hours’ (because this category implies at least some unpaid hours were worked, ie not 0). To test the sensitivity of our assumption, we increased the average unpaid hours in each category by 1 hour for those reporting some unpaid hours (1, 7 and 12). The resulting (less conservative) estimates – the grey columns – increase the equivalent number of full-time employees to 59,000 in 2021 and 53,500 in 2024. The actual numbers may be even higher, depending on the true average number of unpaid hours worked in each category. 

The role of workforce growth

While the estimates in Figure 2 reflect the scale of ‘hidden’ overtime, they do not isolate changes in the rate of unpaid hours over time. This is because the NHS workforce has grown by more than a fifth since 2019.

Increases in the workforce can have opposing effects on unpaid hours. A larger workforce will tend to increase the total number of unpaid hours, if at least some of these extra staff work some overtime. However, a larger workforce may reduce the demand for staff to work additional hours, if fewer shifts are left unfilled. 

Figure 3 illustrates the relative contribution of a growing workforce to the overall level of unpaid hours worked. It compares our most conservative estimates of unpaid hours (based on actual number worked) – shown by the red line – with estimates of unpaid hours, had the reported rates of unpaid hours remained at 2019 levels – shown by the blue line. The blue line reflects the level of unpaid hours that would be expected solely due to workforce growth.

During the pandemic, total unpaid hours exceed what would have been expected based on workforce growth, indicating a rise in discretionary effort. In 2021, this difference amounted to the equivalent of an additional 5,000 FTE (the gap between the red and blue lines).

Since 2022, total unpaid hours have dipped sharply, falling back to what might have been expected given workforce growth in 2023, and then falling below what might have been expected in 2024 (by the equivalent of 4,500 FTE). This gap suggests that, had staff continued contributing unpaid hours at the 2019 rate, the 2024 workforce would have effectively had the capacity of an additional 4,500 full-time staff. 

So, while the total number of actual unpaid hours remains higher in 2024 than in 2019, this is principally because the workforce is larger. 

The difference between expected and actual unpaid hours in 2024 may be explained by a reduced willingness to provide discretionary effort. But it might also be at least partially explained by workforce growth reducing the pressure on staff to work additional hours.

What are the implications of unpaid hours for NHS planning?

Neither the inaugural NHS Long Term Workforce Plan (2023) nor the recent 10-year health plan fully acknowledged the reliance of the NHS on discretionary effort. With the NHS working towards 2% year-on-year productivity growth over the next 3 years, further declines in discretionary effort could add to the challenge of achieving this target.

Projections for future health care demand and staffing needs, along with proposed reforms, may implicitly assume that current levels of unpaid hours will remain constant. The validity of this assumption is uncertain. Motivation for discretionary effort may change, either because of previous experiences of working in the NHS, or because new entrants to the workforce are less willing to work unpaid hours. We also don’t know how changes in staffing levels might affect demand for discretionary effort, with more staff reducing the number of gaps in rosters that need to be filled. 

These factors could all affect the number of (contracted) FTE staff needed to deliver a given level of service. The revised long-term workforce plan – due soon – should therefore consider the resilience of their proposals to different assumptions about future patterns of unpaid hours.