Impact of COVID-19

The COVID-19 pandemic has affected Australia’s population and health-care system in many ways, including: economic expenditure, mortality, disability, the health workforce and disease surveillance.

Diabetes is one of many conditions correlated with greater health consequences throughout the COVID-19 pandemic including increased risk of complication and mortality (Peric and Stulnig 2020). This section explores the impact of COVID-19 in Australia on diabetes risk factors, new onset diabetes and people living with diabetes.

Data available to assess the impact of COVID on diabetes are still evolving across the continuing pandemic period and are limited by the availability of data covering the relevant time period.

Risk factors

Maintaining a healthy lifestyle, including healthy dietary patterns and being physically active, is important for reducing type 2 diabetes risk and improving outcomes for people living with diabetes. No diabetes-specific data are available for changes in behavioural risk factors across the COVID-19 pandemic. However, the Australian Bureau of Statistics (ABS) Household Impacts of COVID-19 Survey is a longitudinal survey of around 1,000 people aged 18 and over in private dwellings that provides regular data on changes in health behaviours.

According to the ABS Household Impacts of COVID-19 Survey, in April 2020:

  • around 25% of people reported increased consumption of snack foods (for example, chips, lollies, biscuits) and 36% reported decreased consumption of take-away or delivered meals compared with before the pandemic
  • around 58% of people reported increases in their personal screen time on their phone, computer, TV or other device compared with before the COVID-19 pandemic. For the June period, 44% reported increasing personal screentime
  • 41% reported an increase in household chores, gardening, yard work, projects or renovations. For the June period, 25% reported increasing household chores, gardening, yard work, projects or renovations. Similar proportions of people reported increasing and decreasing exercise and other physical activity.

Of those adults who usually drank alcohol:

  • 20% reported increasing their consumption of alcohol and 13% reported decreasing their consumption in the previous 4 weeks compared with before COVID-19 restrictions. In June 2020, a similar proportion of people reported that they increased or decreased their alcohol consumption compared with before the COVID-19 pandemic.

Of those who usually smoked (cigarettes, cigars, e-cigarettes or other tobacco products):

  • 18% reported smoking more in the last 4 weeks compared with before COVID-19 restrictions (2.2% of all respondents)
  • 9.7% reported smoking less (1.2% of the total) (AIHW 2021a).

Diabetes incidence

Some studies have proposed a link between COVID-19, hyperglycaemia and new onset diabetes (Sathish et al 2021). In the 12-month periods to March 2021 and March 2022, the National Diabetes Services Scheme (NDSS) had 118,000 and 119,000 new registrants, respectively. Registrations were higher in both periods than any previous 12 months recorded. These new registrations were an increase of 17% compared with the 12 months recorded to March 2020. The largest increase in registration was found among people with gestational diabetes (Diabetes Australia 2022).

However, these new registrations may be, at least in part, people who were previously diagnosed with diabetes and only registering with the NDSS during the pandemic. The increase in registrations also may be influenced by changes to the NDSS to simplify the usual processes to register (Andrikopoulos and Johnson 2020). Further monitoring is required to assess increases in diabetes diagnosis during the COVID-19 pandemic.

Diabetes deaths

ABS provisional mortality data (which includes only deaths certified by general practitioners (GPs) show that from January to December 2020, the age-standardised death rates suggest that the COVID-19 pandemic in Australia did not lead to an increase in mortality for diabetes (AIHW 2021a). Age-standardised death rates were similar for diabetes in 2020 and between 2015 and 2019.

According to the ABS COVID-19 Mortality data from January 2020 to March 2022, pre-existing chronic conditions such as diabetes were reported on death certificates for 3,600 (73%) of the 4,900 deaths due to COVID-19 (ABS 2022). Diabetes was a pre-existing condition in 20% of the 3,600 deaths.

COVID-19 hospitalisations

In 2020–21, there were over 4,700 hospitalisations in Australia that involved a COVID-19 diagnosis. Around 42% of hospitalisations with a diagnosis of COVID-19 had one or more diagnosed comorbid conditions, such as type 2 diabetes or cardiovascular disease, an increase from 25% in 2019–20. Of the 4,700 hospitalisations involving a COVID-19 diagnosis, the most common comorbid conditions associated with COVID-19 hospitalisations over this period were type 2 diabetes (20%) and cardiovascular disease (which includes coronary heart disease and a range of other heart, stroke and vascular diseases) (20%) (AIHW 2022).

Of those with a recorded comorbid diagnosis of type 2 diabetes:

  • 12% of hospitalisations involved time spent in an intensive care unit, compared with 7.0% of all COVID-19 hospitalisations.
  • 7.1% involved continuous ventilatory support, compared with 3.8% of all COVID-19 hospitalisations.
  • 19% had a separation mode indicating the patient died in hospital, compared with 10.3% of all COVID-19 hospitalisations.

These results may be impacted by people with type 2 diabetes being more likely to be older and therefore more likely to have severe COVID-19.

Health service use

People living with diabetes require regular contact with GPs, endocrinologists and allied health services including dietitians and podiatrists to optimise glucose control and reduce risk of diabetes complications. To limit the spread of COVID-19, restrictions were put in place to contain its impact in the community. By the end of March 2020, non-essential businesses and activities had shut down, with people urged to stay at home. As part of these restrictions, many health services were suspended or required to operate in new or different ways. While this may have limited people’s access to and use of these services, in some cases, new or additional services were made available through changes to health service delivery models, policies and programs (AIHW 2021b).

The initial temporary changes to telehealth items in the Medicare Benefits Schedule (MBS) and bulk billing criteria have bridged these issues, but these services were less suitable for complex care which often require face-to-face consultations. The scope of telehealth services was refined over time to make them more suitable for complex care, however, the scope of telephone services has been reduced since July 2021 to focus on more straightforward services.

People avoiding and/or delaying medical care for diabetes during the COVID-19 pandemic has been an emerging global issue. Research has already shown a significant increase in the frequency of severe diabetic ketoacidosis at presentation of type 1 diabetes during the initial period of COVID-19 restrictions in Australia (Lawrence et al. 2021). Presentations with severe ketoacidosis increased from 5% in the pre-COVID-19 period to 45% during May–March 2020.

Primary health care

MedicineInsight is a longitudinal general practice dataset managed by NPS MedicineWise which was established in 2011. NPS MedicineWise undertook a study looking at the 6 months of March to August 2020 (COVID period) compared with the 6 months of March to August 2019 (pre-COVID period) to assess changes to general practice attendance.

Over the 6-month COVID period, the mean number of clinical encounters per patient among patients without a record of diabetes increased when compared with the pre-COVID period. However, there was no significant change in the mean number of clinical encounters per patient among patients with a record of type 2 diabetes when comparing the 6-month pandemic period to the corresponding 6-month period in 2019. The monthly rate of encounters where the patient had a record of type 2 diabetes fell to 59.6 per 1000 encounters in April 2020 before increasing to approximately 80 per 1000 encounters in subsequent months (Figure 1) (NPS MedicineWise 2020).

Figure 1: Monthly mean clinical encounters among patients with a record of type 2 diabetes: COVID and pre-COVID periods

The chart shows the monthly mean clinical encounters among patients with a record of type 2 diabetes during the COVID and pre-COVID periods. The monthly rate of encounters where the patient had a record of type 2 diabetes fell from 79 to 60 per 1000 encounters in April 2020 before increasing to approximately 80 per 1000 encounters by June 2020, above the levels recorded in the equivalent pre-COVID period in 2019.

Hospitalisations

According to the National Hospitals Morbidity Database (NHMD), hospital separations fell in March–April 2020 for both principal and additional diagnosis of diabetes, which may be associated with the effects of the pandemic’s first wave (Figure 2). Similar results were found across type 1, type 2 and other diabetes.

Figure 2: Diabetes hospitalisations by diagnosis type and month, 2015–16 to 2019–20

The chart shows the number of diabetes hospitalisations by diagnosis type and month between 2015–16 and 2019–20. Hospital separations fell in March–April 2020 for both principal and additional diagnosis of diabetes, with a more notable drop in hospitalisations with an additional diagnosis of diabetes.

In 2019–20, the average number of monthly diabetes-related ED presentations was 1,500. The rate of monthly diabetes-related ED presentations:

Figure 3: Emergency department (ED) presentations with a principal diagnosis of diabetes, July 2019 to June 2021

The chart shows emergency department (ED) presentations per 100,000 population by month between July 2018 and June 2021. The rate of monthly diabetes-related ED presentations ranged between 4.7 and 6.6. presentations per 100,000 population between July 2019 and June 2020. This rate was lowest in April 2020 (4.7 per 100,000 population) and highest in June 2020 (6.6 per 100,000 population).

 

Allied health use

No data are available on the use of allied health services for people living with diabetes, but total attendance for the Australian population are available. According to the Medicare Benefits Scheme:

  • allied health attendance increased during the pandemic period, resulting in a 19% increase between 2019 and 2020. However, the increase cannot be attributed to diabetes.
  • optometry services decreased in April 2020 but had returned to the usual level of service provision by the end of August 2020 (though had not yet compensated for services not conducted in the earlier months). This led to an overall 8.1% fall in the number of services in 2020 compared with 2019 (AIHW 2021a).

Monitoring

Rate of HbA1c testing

According to NPS MedicineWise analysis of MedicineInsight, the rate of HbA1c tests over the 6-months from 1 March 2020 to 31 August 2020 was not significantly different from the pre-COVID period, when looking at all regularly attending patients. However, the rate of HbA1c testing did fall significantly among regularly attending patients with a record of type 2 diabetes despite the rate of type 2 diabetes encounters remaining similar in both time periods. In the pre-COVID period, the average monthly rate of HbA1c testing among patients with a record of type 2 diabetes was 126.1 per 1000 clinical encounters, which fell to 109.0 tests per 1,000 clinical encounters in the COVID period (Table 1).

In April 2020, there was a significant decline in the rate of HbA1c tests performed. The rate of tests for all patients fell from 32 tests per 1,000 clinical encounters in April 2019 to 21 tests per 1,000 clinical encounters. The rate of testing for patients with a record of type 2 diabetes fell from 120 tests per 1,000 clinical encounters in April 2019 to 77 tests per 1,000 clinical encounters in April 2020 (NPS MedicineWise 2020).

Table 1: Rate of HbA1c testing and diabetes encounters per 1,000 clinical encounters in the covid period compared with the same period in 2019
 

1 March–31 Aug 2019
(pre-COVID period)*

1 March–31 Aug 2020
(COVID-period)*

p-value
Rate of HbA1c testing (all patients) 32.3 (29.8, 34.8) 30.4 (5.0, 35.8) 0.43
Median all patients 31.5 (Q1 30.6, Q3 33.1) 32.1 (Q1 29.6, Q3 33.1) -
Rate of HbA1c testing (type 2 diabetes patients) 126.1 (118.6, 133.7) 109.0 (92.1, 125.9) 0.04
Median (type 2 diabetes patients) 123.7 (Q1 120.8, Q3 129.9) 113.4 (Q1 111.3, Q3 115.4) -
Rate of type 2 diabetes encounters 73.3 (69.4, 77.2) 75.4 (66.9, 84.0) 0.57
Median 72.9 (Q1 70.5, Q3 75.8) 79.0 (Q1 74.0, Q3 79.8) -

*Reported as the mean (per 1000 encounters) (95% CI); or median (quartiles)

Source: NPS MedicineWise (2020).

HbA1c testing frequency

Imai et al. (2020) undertook a study of the impact of COVID-19 on HbA1c monitoring using data from over 800 general practices (456 from Victoria and 347 from New South Wales) from January 2018 to December 2020. The volume of HbA1c monitoring in 2020 during the weeks of the first wave of COVID-19 (approximately March-May) decreased by 19.1% in New South Wales and 25.6% in Victoria, compared to the average volume of 2018–2019. Although it was not as large as the first wave, there was another fall in the total HbA1c testing volume in 2020 during the weeks of the second wave (approximately June–September).

The study also examined the number of patients living with type 2 diabetes who had records of HbA1c testing in both 2018 and 2019 (n=22,804 in Victoria, n = 15,399 in New South Wales) and their HbA1c testing frequencies. Approximately 14%–15% of these patients did not have HbA1c testing in 2020 (15.3% in Victoria, 14.1% in New South Wales). The number of patients who had multiple HbA1c tests also decreased in 2020 in both states (Figure 4).

Figure 4: Number of subgroup patients by HbA1c testing frequency. The subgroup patients were those who had records of HbA1c tests in both 2018 and 2019

The chart shows the number of subgroup patients (those living with type 2 diabetes who had records of HbA1c tests in both 2018 and 2019) by HbA1c testing frequency in Victoria and New South Wales. Around 14–15%25 of these patients did not have HbA1c testing in 2020 while the number of patients who had multiple HbA1c tests also decreased in 2020 across both states.

Source: Imai et al. 2020.

Gestational diabetes screening

In 2020, the Australasian Diabetes in Pregnancy Society, the Australian Diabetes Society, the Australian Diabetes Educators Association and Diabetes Australia, jointly provided temporarily revised guidelines for gestational diabetes screening in response to the COVID-19 pandemic. The guidelines aim to reduce both the number of women attending and the amount of time spent at pathology collection centres during times of elevated contagion risk.

The COVID-19 guidelines replace the oral glucose tolerance test (OGTT) with HbA1c testing in the first trimester and Fasting Blood Glucose (FBG) between 24 and 28 weeks for those not diagnosed with gestational diabetes from the initial HbA1c result. For those with a resulting FBG of 4.7–5.0 mmol/L, an OGTT is recommended while an FBG ≥5.1mmol/L is diagnostic of gestational diabetes (Diabetes Australia 2020).

These guidelines (and other COVID-related factors) likely influenced a sharp decline in the number of women receiving an OGTT or oral glucose challenge test (OGCT) for the screening of gestational diabetes throughout April to June 2020 with an overall 11% drop in the annual numbers between 2019 and 2020 (from 184,000 to 163,000, respectively) (Figure 5). This drop coincided with a 25% increase in HbA1c testing for the management of pre-existing diabetes where the patient is pregnant and a 2% increase in claims for HbA1c testing for the diagnosis of diabetes – both MBS items likely used as a substitute for the use of the OGTT for the detection of gestational diabetes during COVID.

Recent retrospective studies have suggested the temporarily revised guidelines could lead to an under detection of gestational diabetes of between 25%–29% (van Gemert et al. 2020; Zhu et al. 2021). With the National Hospitals Morbidity Database (NHMD) being the primary data source to report gestational diabetes incidence in AIHW monitoring reports, the impact on the incidence of gestational diabetes using NHMD data won’t be evident until late 2022 with the release of the 2021–22 NHMD. While MBS data processed to 1 August 2021 indicates the number of women receiving the OGTT and OGCT had returned to pre-COVID levels, further monitoring will determine the impact of successive waves of the COVID-19 pandemic on gestational diabetes screening and subsequent incidence numbers.

Figure 5: Number of women who received MBS-subsidised services for the diagnosis of gestational diabetes and management of pre-existing diabetes in pregnancy, by month, April 2019 to August 2021

The chart shows the number women who received MBS-subsidised services for the diagnosis of gestational diabetes and management of pre-existing diabetes in pregnancy, by month, April 2019 to August 2021. There was a sharp decline in the number of women receiving the oral glucose tolerance test for the screening of gestational diabetes throughout April to June 2020 with an overall 11% drop in the annual numbers between 2019 and 2020 (from 184,000 to 163,000, respectively). By January 2021, the numbers had returned to pre-COVID levels.

Diabetes medicine use

The COVID-19 pandemic has affected both patients and health practitioners in terms of the number of medical services, type of services and the way in which services are delivered (Sutherland et al. 2020). Medication access and supply has also been affected.

Analysis of the total volume of Pharmaceutical Benefits Scheme (PBS) prescriptions dispensed for ATC group A10, Drugs used in diabetes dispensed during 2019–20 and 2020–21 shows little change from 2018–19 when accounting for the expected increase in prescriptions dispensed over time. During 2019–20 and 2020–21, 15.4 and 16.5 million scripts for group A10 were dispensed, compared with 14.3 million for 2018–19 (Figure 6), a 7.8% increase from 2018–19 to 2019–20, and a 7.6% increase from 2019–20 to 2020–21.

Figure 6: Prescriptions dispensed for diabetes medicines, by quarter, 2017–18 to 2020–21

The chart shows the number of prescriptions dispensed for diabetes medicines by quarter between 2017–18 and 2020–21. During 2019–20 and 2020–21, 16.5 and 15.4 million scripts for group A10 were dispensed, compared with 14.3 million for 2018–19, a 7.8% increase from 2018–19 to 2019–20, and a 7.6% increase from 2019–20 to 2020–21.

There were, however, changes in consumer behaviour. An unusually high volume of diabetes scripts was dispensed in March 2020 (1.6 million), coinciding with the introduction of national restrictions, followed by a decrease in April 2020 (1.2 million) (Figure 7).

In March 2020, the Australian Government implemented temporary changes to medicines regulation to support Australians’ continued access to PBS medicines during the COVID-19 pandemic. Some of these changes were in response to the dramatic increase in demand for medicines during early March, which resulted in pharmacies and wholesalers reporting medicine shortages.

The measures included a restriction on the quantity of medicines purchased to discourage unnecessary medicine stockpiling, continued dispensing emergency measures to allow one month supply of a patient’s usual medicines without a prescription, a home delivery service for eligible patients, digital image-based prescriptions to support telehealth medical services, and arrangements for medicine substitution by pharmacists without prior approval from the prescribing doctor (AIHW 2020).

Figure 7: Prescriptions dispensed for diabetes medicines, by month, 2017–18 to 2020–21

The chart shows the number of prescriptions dispensed for diabetes medicines by month between 2017–18 and 2020–2021. An unusually high volume of diabetes scripts was dispensed in March 2020 (1.6 million), coinciding with the introduction of national restrictions, followed by a decrease in April 2020 (1.2 million).