The New Zealand Government has pursued a strategy of elimination in response to the threat posed by the COVID-19 pandemic to the public and the health system.¹ In March 2020, a four-level national alert system was implemented to combat the spread of the virus, enabling the Government to communicate its public health messages with the corresponding community restrictions in place under each alert level.² Alert Level 4 (‘Lockdown’), the most restrictive level, was implemented on 25 March at 11.59pm and remained in place until 27 April (Table 1). Government-led messaging was clear that people must stay at home and remain local if they were to venture outside for exercise, to shop for essentials or to look after vulnerable people. Many national and regional organisations actively communicated via multiple media channels for people to observe the restrictions and for those working in essential services to remain aware of the potential for injury.^3,4 As most injuries occur at home, there were injury awareness and prevention focused messages aimed at those undertaking do-it-yourself (DIY) activities while restricted to home.⁵

Table 1: Timeline of selected COVID-19 related events in New Zealand,²⁸ January–12 July 2020.

Internationally, trauma volumes and the causes of injury are reported to have changed significantly in many countries during the early months of the pandemic. Many hospitals have reported an overall reduction in trauma cases as a result of varying public health measures, such as work closures and stay-at-home orders.^6–9 At the same time, there have been reports of potential increases in domestic and paediatric trauma and specific causes of injury,¹⁰ with anecdotal evidence and concerns in New Zealand including increases in injuries related to pedal cycling, activities at home and assault.^11–14

In April 2020, the New Zealand Medical Journal published our findings on variation in volumes and characteristics of trauma patients admitted to Waikato Hospital (New Zealand’s only Level 1 trauma centre verified by the Royal Australasian College of Surgeons) during the first 14 days of Alert Level 4.¹⁵ In response to feedback following this the Midland Trauma System (MTS) conducted this study to:

• Review the variation in injury admissions over all alert levels covering the period 15 February to 10 July 2020 and compare this with 2017–2019 volumes.
• Extend the geographical coverage to include all hospitals in the Midland region.
• Provide information to advise Midland region hospitals on what injury volumes might be expected in any future community lockdowns during a time of pandemic.

The Midland region comprises the Bay of Plenty, Lakes, Tairāwhiti, Taranaki and Waikato district health board (DHB) catchments, with an estimated population of 985,285.¹⁶ The region is the only one in New Zealand to collect data for both major and non-major injury, allowing a clearer picture of the burden of injury severe enough to require admission to hospital.

Methods

A retrospective, descriptive study was conducted on trauma registry data on injured patients of all age groups and injury severities admitted to hospitals within the Midland region. Six time intervals covered the study period 15 February to 11 July 2020: these were pre-Lockdown (15 February–18 March); Lockdown week (19–25 March); Level 4 (26 March–27 April); Level 3 (28 April–13 May); Level 2 (14 May–8 June) and Level 1 (9 June–11 July). We overviewed all intervals (and compared 2020 to the average of 2017–2019) then narrowed our focus to hospital admissions occurring between 26 March to 8 June 2020, which encompasses alert levels 4, 3 and 2. The week of lockdown between 19 March and 25 March 2020 was excluded to allow analysis of a presumed steady state of community behaviour within pre- and during-lockdown phases when national alert levels were escalating and community behaviour was changing dramatically in anticipation of Alert Level 4 lockdown. Likewise, Alert Level 1 was excluded as it is a ‘prepare’ phase with no real community level restrictions on activities and movement compared to pre-COVID times.

Patients were grouped according to age group, ethnicity, gender, cause of injury, place of injury and injury severity. Data were sourced from the Midland Trauma Registry (MTR) and analysed using Microsoft Excel (2016) and IBM SPSS v27.¹⁷ Injury severity was quantified using the Abbreviated Injury Scale (AIS), an anatomical scoring system that ranks injuries from 1 (minor) to 6 (non-survivable).¹⁸ The Injury Severity Score (ISS) is also an anatomical scoring system using a 0–75 scale. The highest AIS scores in each body region are the basis of the ISS, with injuries then categorised as non-major (ISS <12) and major (ISS >13).¹⁹ The International Classification of Disease categories (ICD-10AM sixth edition) are used. Injuries related to the category of ‘inanimate mechanical forces’ (W20–W49) captures a wide range of causes, including being caught, crushed, jammed or pinched in or between objects, contact with sharp glass and contact with powered and non-powered hand tools. Injuries caused by ‘animate mechanical forces’ (W50–W64) include being bitten or struck by animals, contact with plants and contact with another person (excluding assault).²⁰

Consistent with international trauma registries, people sustaining an injury as a result of a pre-existing medical condition or the late effects of injury, or if the injury occurred >7 days prior or the person died prior to arrival in the emergency department, are excluded from the MTR. This study was registered with the Waikato DHB’s Clinical Audit Support Unit (4085DTR200527R1).

Results

Trauma admission volumes across the Midland region fluctuated on a weekly basis in 2020 compared with the corresponding average of the previous three years (Figure 1). The seven-day moving average in 2020 was higher prior to the start of Alert Level 4 and following the end of Alert Level 3, with the significant reduction in the average number of hospital admissions clear during those most restrictive alert levels.

Figure 1: Hospital admissions: Seven-day moving average 15 February–10 July 2020 compared with 2017–2019.

There were two time periods with a statistically significant reduction in volumes in 2020 compared to the 2017–2019 average (Table 2). These were the Alert Level 4 period, where volumes decreased by 36.7% (p<.00001), and Alert Level 3, with a reduction of 16.0% (p=.043). Comparatively, volumes increased during the pre-lockdown period in 2020 (17.8%, p<.00001), during Alert Level 2 (6.5%) and during the first month of Alert Level 1 (13.8%, p=.003).

Table 2: Injury-related hospital admissions during the different levels of community restrictions in 2020 compared with 2017 – 2019 volumes, n=9,712.

Focusing in on the periods during alert levels 4, 3 and 2 (Table 3), we see there was an overall decrease in admission volumes of 18.3%, with a greater reduction for males than females (21.9% and 12.3% respectively). By age group there were significant reductions for those aged 5–14, 15–24 and 65–74 years, with the mean age of those injured being higher in 2020.

Table 3: Trauma admission volumes: 26 March–8 June 2017–19 compared with 2020, n=4,742.

In 2020 a lesser proportion of hospital admissions were related to pedestrian injuries (63.0%, p=.02), while a greater proportion of injuries were due to cycling (11.9%, p=.02) and contact with heat and hot substances (56.2%, p=.0008). Admissions of injured vehicle occupants decreased by 35.8%, but compared to the previous three years, this was not a significant proportional change. There were also fewer admissions related to assault and intentional self-harm (where numbers were low yet important). Admissions of motorcyclists increased slightly (2.2%), and this includes both on-road and off-road crashes.

Proportionally there was an increase in 2020 of injuries occurring at home (up 28.3%, p<.00001) and on the footpath (37.9%, p=.0007), and there was a decrease in injuries occurring at schools (down 75.0%, p<.00001), in sports areas (79.7%, p<.00001) and on or in water (71.4%, p=.0096).

Discussion

Adopting an ambitious ‘containment on the path to elimination strategy’,¹ including strict border control measures to limit the spread of COVID-19, contributed to the New Zealand health system avoiding the situation of a rapidly rising number of COVID-19 cases requiring hospital-level care.^21,22 In the Midland region the community restrictions in place²³ were effective when looking at the reduction in hospital admissions, but only during Alert Level 4 and Alert Level 3. We found an injury rebound with hospital admissions increasing during Alert Level 2 and in the first month of Alert Level 1 compared to the previous three years.

Road traffic crashes are a significant cause of injury and a key public health concern. In the Midland region crashes contribute the second largest volume of hospital admissions (after falls), and in the 2019 calendar year they made up 31% of all major severity injury hospitalisations.¹⁶ For road traffic crashes, we found a significant reduction in admissions for vehicle occupants alongside a small increase for motorcyclists during alert levels 4, 3 and 2. The majority of these injury events happen on the road where we found a significant reduction in trauma volumes over the same period.

There was anecdotal evidence available that the popularity of cycling increased during alert levels 4 and 3, given the lesser volumes of motor vehicles on the road.^24,25 We found an increase in hospital admissions in 2020 compared with the previous three years for all cycling related injuries. The increase in injured cyclists could be due to a mix of new or returning riders taking advantage of perceived safer road environments, or regular riders cycling more often (and presumably remaining close to home). It is not possible to know whether access to cycling increased in the region, nor by how much or therefore how this may have impacted injury volumes.

Internationally, public health interventions implemented to curb spread of the virus, such as social-distancing, staying at home in a ‘bubble’ and restricted travel, are thought likely to contribute to an increase in injury due to assault.^26,27 Hospital admissions related to assault are recorded in the MTR as a cause of injury, and while any such admission is concerning, we do not consider these volumes to be a true reflection of what might have been happening in the community. Other front-line agencies and organisations are much better placed to comment on interpersonal violence during the higher alert levels and whether there were increases in the community that did not result in a hospital admission.²⁸ Hospital admissions are only a small part of this assault picture.

Given the changing amount of time people were spending in different locations, particularly during alert levels 4 and 3, the increase in hospital admissions for injuries occurring at home in 2020 was expected, as was the reduction in injuries happening at schools and sports areas. In early public health messaging, there was a focus on reminding essential workers to take care and reduce pressure on health services. This included targeted messaging to farmers as a key group of essential workers, through organisations such as Federated Farmers.³ We were interested to see only a small increase in all injuries requiring hospital-level care occurring on farms, despite farmers and farm workers continuing their daily work-related activities during national-level restrictions.

In the Midland region, and particularly for Waikato Hospital as the regional trauma centre for high-acuity patients, it is important to understand injury patterns and volumes to enable planning to maintain access to trauma care during any future pandemic situations. In our region, as in many others, there were significant changes to the way people behaved and where they went, what activities they undertook and, importantly, in how healthcare was provided, particularly during the more restrictive alert levels.²⁹ We noted earlier the rebound in trauma hospital admissions during Alert Level 2 and particularly in the first month of Alert Level 1 as restrictions eased, meaning the earlier reductions were relatively short lived. This rebound was important as it coincided with the push by DHBs to catch up on healthcare, such as elective surgeries, which were delayed during the higher alert levels.^30,31 This potential resource squeeze must be taken into account should restrictions return in the Midland region. Both workforce capacity and other system resources are problematic to increase in a short period of time and in a fiscally constrained environment.

Conclusion

Worldwide the COVID-19 pandemic has resulted in an unparalleled disruption to life³² and in New Zealand has had an enormous societal impact.²⁵ However, unlike many countries, New Zealand’s hospital system has not been subject to the strain of large numbers of seriously unwell COVID-19 patients requiring hospital-level care, despite a poor assessment of the country’s pandemic preparedness in 2019.²⁵ In the Midland region, using the volume of injury-related hospital admissions as a measure, we see that the implementation of the national alert level system with unprecedented community level restrictions and astute communication of public health messaging was successful. This was greatly assisted by the vast majority of the public adhering to public health messages.^1,25 For hospital resource planning, the trauma admission rebound evident as restrictions eased is important, particularly in the context of health care delays and subsequent need for healthcare catch up. In any future similar scenario, this could again place additional pressure on the healthcare system across our region.