The impact of the COVID-19 pandemic on falls and fractures, 4 years on, are we any further forward?

Key Points

• Editorial to accompany Impact of the COVID-19 pandemic on hospital episodes for falls and fractures associated with new-onset disability and frailty in England: a national cohort study.

• Trends around falls and fractures in relation to the timing of UK national lockdowns are complex.

• More research is needed in this area to inform public health policy for future pandemics.

• UK incidence rates for falls admissions steadily increased by 24% from 2013 to 2019.

Most publications and presentations about falls cite the statistic that one-third of people over the age of 65 years, and one in two over the age of 80 years will fall each year [1, 2]. These figures arise from studies, which are over 30 years old and do not reflect the more recent and more concerning data, which demonstrate a steadily increasing falls prevalence. For example, in 2015, the US Health and Retirement Study demonstrated that the 2-year prevalence of falls increased from 28 to 36% between 1998 and 2010 [3]; more recently, an analysis of the US Centres for Disease Control and Prevention database revealed that age-adjusted mortality from falls in people aged over 65 years had increased from 29/100,000 in 1999 to 69/100,000 in 2020 [4]. Alongside this, there has also been a steady increase in falls-associated hospital admissions in the UK. An analysis of Hospital Episode Statistics demonstrated that for those aged over 75 years, there was a 45% increase in admissions between 2008 and 2017 [5].

The reasons why both falls and fall-related mortality are increasing are not clear, but may be related to increased survival of people with chronic diseases, increasing levels of polypharmacy and declining levels of physical activity [6, 7]. During the COVID-19 pandemic, there was a call to action out of concern for the anticipated increase in falls and fractures due to social restrictions and reduced levels of physical activity [8]. Indeed, falls services saw a decline in the age of patients attending services alongside a reduced muscle strength and function between 2019 and 2021 [9].

In Thomas et al.’s analysis of UK admissions for falls and fractures between 2015 and 2022, there is also a year-on-year increase in hospital admissions, up until the COVID-19 pandemic when trends become more complex [10]. Falls due to an external force were included as a fall, and vertebral, ulnar, humeral and lower limb fractures were not included in the analysis of fractures. This should be considered when interpreting the fracture rates as COVID-19 lockdowns occurred in Winter and Spring and types of fractures differ across the year. Radius, ulna and humerus fractures peak in December, but fractures of the lower limb peak in the Summer [11].

Thomas et al. present data for the whole population, including children, but as the risks and mechanisms for falls and fractures are very different for children and older people, this editorial will focus on the data specific to older age groups, provided in the supplementary material. To make inferences about the impact of lockdown on falls and fractures, Thomas et al. used an auto-regressive integrated moving average to generate estimates of expected falls rates from 2020 onwards, using data from 2013 to 2018. In those aged over 65 years, the number of admissions for falls dropped dramatically below expected rates, shortly before, and at the start of the first national lockdown. The falls rates rapidly returned to back to expected levels during the initial lockdown and remained at expected levels up until 2021–2022 (about three months following the last lockdown) when expected falls remained lower than expected, persistently so in those aged over 80 years. Hospital admissions related to fractures also fell lower than expected in the initial stages of the first national lockdown and although not always at significant levels, the rates of fractures appear persistently lower than expected in the older age groups. An interesting comparison is that in the working-age group there were unexpectedly high levels of falls at the end of, or shortly after, all three national lockdowns.

The study authors propose that the rapid drop in falls rates during lockdown are likely a result of reduced activity and time outdoors. However, research on physical activity during and after lockdown is contradictory. One study looking at smartphone-tracked activity in UK adults found that whilst younger people showed a steep decline in activity (37%), those aged over 65 years appeared to remain more active and increased their activity after lockdown [12]. Conversely, the UK’s Household Longitudinal Study of over 3,600 older adults, showed that the proportion of those meeting physical activity targets (self-report), fell by 10% (from 43% to 33%) between September 2020 and January 2021 [13]. This may reflect the functional levels and confidence to be more active when restrictions ended, as history of being active is strongly associated with whether people returned to pre-covid activity levels [14]. A systematic review of studies examining physical activity in older people over the pandemic, including 14 cross-sectional and 11 cohort studies, concluded that there was a reduction in activity with increased sitting time, reduced moderate to vigorous activity and fewer daily steps, leading to a decline in physical fitness [15].

Other contributory factors for the fewer than expected falls in older age groups may include reduced reporting of falls, anxiety around attending healthcare settings for fear of catching COVID-19, or not wanting to burden services who were under great pressure. There was also increased use of Emergency Health Care Plans in the community, which may have resulted in fewer admissions. The subsequent sharp increase in falls and fracture admissions, following lockdown, may represent, at least in part, late presentations of those who were fearful of seeking healthcare. Whether the temporary closure of community-based falls prevention services would result in such rapid increase in falls admission is unknown, but it seems more likely that suspending falls services would have a more longer term impact.

When considering frailty (which includes all ages), Thomas et al. present a different outcome that combines an admission for a fall and fracture, rather than presenting data for falls and fractures separately. In contrast to previous analyses, there is a large increase in falls with fractures in the first and second lockdown, but a sharp decrease in the third lockdown. Lockdowns occurred during times of highest community rates of COVID-19 and falls and fractures were a common presentation of COVID-19 in people with frailty [15, 16]. So, COVID-19 may be the explanation here, with subsequent decline in falls from the effects from vaccination. Following the final lockdown, rates of falls with fractures in those with frailty initially increased after a sharp decline and then returned to expected levels through 2021. If lockdown and deconditioning had the result of increasing falls, we may have expected a continued increase in falls with fractures in those living with frailty, especially if we expected more people to transition into frailty because of deconditioning. Concern about contracting COVID-19 may have caused those with functional limitations to avoid going outdoors once restrictions were lifted. Those with functional limitations are more likely to fall with increased activity; so, this may have been protective behaviour but will likely have contributed to an increased risk of falls once they resumed physical activity [17].

Analysis of hospital admissions for falls is rather limited. Falls are poorly recollected, poorly reported and poorly coded [18]. Fractures secondary to falls are a more robust method of estimating falls trends from health records, although the limited types of fractures included in Thomas et al.’s study perhaps reduce the reliability here (other studies found that hip fractures did not increase after the first lockdown [19, 20]).

Given that the final lockdown ended in March 2021 and the data trends are displayed until December 2021, there is insufficient time to reflect the true impact of temporarily closing falls prevention services and the impact of deconditioning on falls, which one would expect to continue over a longer period. An update on this paper would be welcome, looking at longer term trends to confirm if the pandemic truly led to an ongoing increased rate of falls and fractures; with this analysis considering seasonality, excluding children and refinement of the definition of a fall and fracture. Such an analysis could help clear the muddy waters, but also there is a real need for higher quality studies to inform future public health strategies for older people during pandemics.

Declaration of Conflicts of Interest

None.

Declaration of Sources of Funding

None.

References