Trends in upper respiratory tract infections and antibiotic prescriptions during the COVID-19 pandemic – a national observational study

Upper airway infections such as acute otitis media, rhinosinusitis and pharyngotonsillitis are among the most common infections and account for a large proportion of antibiotic prescriptions [1,2]. They are usually caused by commensal bacteria residing in the nasopharynx. Commensal bacteria often go from being apathogenic to becoming pathogenic when the host encounters a viral infection, due to a synergistic effect between viruses and bacteria [3].

When the COVID-19 pandemic struck in early 2020, countries all over the world introduced various measures to slow down the spread of the infection. This has provided us with a unique possibility to study the effect of reduced social contacts on upper airway infections. A sharp decline in viral infections such as influenza and RS virus has been shown after the introduction of COVID-19 restrictions [4–6]. A Dutch study found that the number of otitis media patients in primary care was reduced by 63% [7], and a smaller study from the United States also showed a large decline in otitis media diagnoses [8]. A study from the United Kingdom found that the number of tonsillitis episodes in children with recurrent tonsillitis was dramatically reduced during lockdown [9].

Unlike many other countries, Sweden did not impose a strict lockdown, however, a work-from-home policy was introduced wherever possible, and those who could not, was to stay at home at the slightest signs of infection. Though day-care and primary schools stayed open, secondary schools and universities closed. Public gatherings were prohibited, and social distancing measures were applied in public places. Unlike in most countries, face masks were not mandatory outside medical health care settings, though, from Nov 2020, they were recommended on public transport during rush hour. In Sweden, upper airway infections such as otitis media, rhinosinusitis and pharyngotonsillitis are handled by otolaryngologists and paediatricians as well as in primary care.

The purpose of this retrospective study was to compare the incidences of upper airway infections and their complication, and of antibiotic prescriptions in Sweden during 2019 and 2020.

The number of diagnoses of suppurative otitis media (H66), acute rhinosinusitis (J01), acute pharyngotonsillitis (J03), acute mastoiditis (H70), acute peritonsillitis (J36), acute nasopharyngitis (J00) and pseudocroup (J05) were obtained from The Swedish Board of Health and Welfare Statistics Database, a national, open database covering all diagnoses from specialist care at hospitals and private specialists in Sweden. Antibiotic prescription data for airway antibiotics were retrieved from the Swedish E-Health Authority, where all prescriptions – primary and specialist care - are registered. This register does not contain information about diagnoses. Population data were obtained from Statistics Sweden.

As this study only used open, anonymised data together with sales statistics on antibiotics, no ethical approval was needed.

Statistical analyses were performed using Stata 16.1 (Stata Corp LP). Yearly incidence rates per 100.000 inhabitants were calculated, along with 95% confidence intervals. Antibiotic prescriptions were reported for age groups 0–6, 7–19, 20–59, 60–79 and 80+ years, and to be able to match this with data from the diagnosis database, which is divided into 5-year age bands, the age groups 0–19, 20–59, 60–79 and 80+ years were chosen.

There was a substantial decrease in overall incidence of uncomplicated bacterial infections such as otitis media, acute rhinosinusitis and acute pharyngotonsillitis in 2020 compared to 2019 (Table 1, Figures 1–3). The most common of these infections was otitis media, which decreased by 40% in the population as a whole, but most pronouncedly in patients aged 0–19 years, where the incidence was highest. Acute rhinosinusitis was most common in adults, however, the incidence decreased to a similar extent in all age groups. Acute pharyngotonsillitis was most prevalent in children and teenagers, and significant decreases in incidence were seen in all but the oldest age group, where the infection was very rare.

Figure 1. Incidence of otitis media/100.000 inhabitants in specialised outpatient care in 2019 and 2020 in various age groups with 95% confidence intervals.

Figure 2. Incidence of acute sinusitis/100.000 inhabitants in specialised outpatient care in 2019 and 2020 in various age groups with 95% confidence intervals.

Figure 3. Incidence of acute tonsillitis/100.000 inhabitants in specialised outpatient care in 2019 and 2020 in various age groups with 95% confidence intervals.

Table 1. Incidence rate ratio for the population as a whole and for separate age groups (2020 compared to 2019).

	Incidence rate ratio (95% CI)
	All ages	0–19	20–59	60–79	80+
Otitis media	0.60 (0.59–0.61)	0.57 (0.55–0.58)	0.69 (0.66–0.72)	0.65 (0.61–0.69)	0.70 (0.64–0.78)
	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001
Acute sinusitis	0.56 (0.54–0.58)	0.52 (0.46–0.59)	0.56 (0.53–0.59)	0.58 (0.54–0.62)	0.53 (0.44–0.63)
	<0.0001	<0.0001	<0.0001	<0.0001	<0.0001
Acute tonsillitis	0.62 (0.61–0.64)	0.63 (0.60–0.65)	0.62 (0.60–0.65)	0.58 (0.49–0.68)	0.95 (0.56–1.63)
	<0.0001	<0.0001	<0.0001	<0.0001	0.9
Acute mastoiditis	0.7 (0.6–0.8)	0.6 (0.4–0.8)	0.7 (0.5–1.0)	1.1 (0.7–1.7)	0.8 (0.4–1.4)
	0.0001	<0.0001	0.07	0.7	0.4
Peritonsillar abscess	0.8 (0.7–0.8)	0.9 (0.8–1.0)	0.8 (0.7–0.8)	0.8 (0.6–0.9)	0.8 (0.5–1.2)
	<0.0001	0.01	<0.0001	0.0001	0.2

Regarding viral infections, acute nasopharyngitis – a diagnosis which should not require a doctor´s visit, was rarely diagnosed in older children or adults, but was common among the very youngest (<5 years). In this age group, the incidence went down from 413/100.000 in 2019 to 247/100.000 in 2020 (−40%). Similarly, pseudocroup was almost exclusively diagnosed in the youngest age group, where it decreased from 1097/100.000 in 2019 to 437/100.000 in 2020 (−60%).

As for complications to upper airway infections, these decreased, as well. Acute mastoiditis decreased by as much as 40% in children and adolescents, however, in adults, in whom the diagnosis is very rare, no change could be proved (Table 1). Peritonsillitis decreased by 20% in all but the youngest and the very oldest (Table 1).

Penicillin V was the most commonly prescribed antibiotic in all age groups. The second most common were amoxicillin and doxycycline, depending on the age group (Figure 4).

Figure 4. Prescriptions of various antibiotics used for treating airway infections/1000 inhabitants during 2019 and 2020.

The prescription of airway antibiotics decreased dramatically in 2020 compared to 2019 (Figure 4). Prescription rates were highest in children and adolescents, and it was here that the greatest reduction was seen. In those aged under 20, prescriptions for penicillin V, amoxicillin and erythromycin were halved (Table 2). The changes were less pronounced for more rarely used antibiotics. In adults, the largest reduction was seen for erythromycin and, in older adults, for doxycycline (Table 2).

Table 2. Change in antibiotic prescription rates (2020 compared to 2019).

	Rate ratio (95% CI) p-value
Age group	0–19 years	20–59 years	60–79 years	80+ years
Antibiotic
Penicillin V	0.53 (0.53–0.53) <0.0001	0.73 (0.73–0.74) <0.0001	0.79 (0.78–0.79) <0.0001	0.78 (0.77–0.79) <0.0001
Amoxicillin	0.53 (0.53–0.54) <0.0001	0.91 (0.90–0.92) <0.0001	0.77 (0.76–0.78) <0.0001	0.77 (0.75–0.78) <0.0001
Amoxicillin with clavulanate acid	0.68 (0.66–0.69) <0.0001	0.86 (0.84–0.87) <0.0001	0.86 (0.84–0.88) <0.0001	0.92 (0.89–0.95) <0.0001
Erythromycin	0.53 (0.51–0.54) <0.0001	0.61 (0.59–0.63) <0.0001	0.62 (0.59–0.64) <0.0001	0.63 (0.58–0.69) <0.0001
Trimetoprim-sulfametazole	0.89 (0.87–0.91) <0.0001	0.97 (0.95–0.99) 0.001	0.99 (0.97–1.00) 0.06	0.97 (0.94–0.99) 0.02
Doxycycline	0.86 (0.83–0.89) <0.0001	0.76 (0.75–0.77) <0.0001	0.64 (0.63–0.65) <0.0001	0.63 (0.61–0.64) <0.0001
Cefalosporins	1.13 (1.08–1.18) <0.0001	0.88 (0.86–0.90) <0.0001	0.91 (0.88–0.93) <0.0001	0.93 (0.89–0.96) 0.0001

This study showed that the incidences of otitis media, acute rhinosinusitis and acute pharyngotonsillitis decreased by approximately 40% in 2020 compared to 2019, accompanied by a subsequent reduction in antibiotic prescriptions. A significant decrease of complications such as acute mastoiditis and peritonsillitis was also noted.

There may be several explanations for the large decline in upper airway infections during 2020. One is that the spread of viruses was reduced due to social distancing measures, resulting in a reduction of viral and, subsequently, bacterial infections. Another explanation could be that many were reluctant to seek medical health care for fear of becoming infected, or because they did not want to put extra burden on an already strained health care. This could have led to fewer diagnoses being registered and would thus create a false reduction in incidence. However, such an effect ought to be most pronounced for patients with less severe infections. These patients tend to attend primary care rather than specialists, and incidences only represent specialist care, where contacts ought to have been more stable. This assumption is supported by the fact that visits to primary care declined more (43%) than visits to specialised care (−30%) in Sweden during 2020 [10]. The incidence changes during 2020 were probably a combination of all of the above.

The use of face masks has been widely debated during the COVID-19 pandemic. It is interesting to note that the incidence of upper airway infections was almost halved despite the fact that Sweden chose a very liberal approach and did not make face masks mandatory outside medical health care facilities.

Incidences and incidence changes varied between age groups. The reduction in otitis media incidence was less pronounced in older age groups, which might be partly explained by the fact that it is less common in adults, and partly by the fact that the ICD code H66 also includes chronic suppurative otitis media, a diagnosis more prevalent in adults and less likely to be affected by a decline in viral infections. Compared to the study by Hullegie et al. [7], the Swedish reduction in otitis media was less pronounced. This might reflect that the Dutch study contained more specific diagnoses than this study, that it contained primary care data rather than specialised outpatient care, and that social distancing regulations differed between the Netherlands and Sweden.

Regarding acute rhinosinusitis, incidences in 2020 declined similarly in all age groups, though the absolute incidence was higher among adults. Acute pharyngotonsillitis incidence decreased by 40%, however, the infection was rare among the elderly, and no obvious change in incidence was seen in this age group.

A small British study [9], where children with recurrent pharyngotonsillitis were followed longitudinally during lockdown showed an even larger decrease in the number of pharyngotonsillitis episodes. This might be explained by chance, by the fact that the British lockdown was much stricter and by the fact that it only contained at-risk patients, in whom reduced social contacts are likely to be even more influential.

Acute nasopharyngitis – a viral infection – was rarely diagnosed in patients over 4 years, something which seems reasonable considering its benign nature. Among young children, it decreased to a similar extent as the bacterial infections during the first year of the pandemic. Likewise, pseudocroup was very rarely diagnosed in patients over 4 years of age, however, in this group, it decreased by as much as 60%.

The incidence in acute mastoiditis is very low, but despite this, a significant decrease was seen among children. The magnitude was similar to the decrease in otitis media incidence. Peritonsillitis on the other hand, is most common in adults, except for in the very old. This condition decreased by 20% in the two middle age groups.

Despite antibiotic prescription rates being halved, there was a significant decrease in the number of complications, implying that patients with severe infections were not ignored by health care. The E-Health Authority database, from which antibiotic prescription data were retrieved, covers all prescriptions in Sweden, primary care as well as specialist care. Prescriptions are not linked to diagnoses, so it was not possible to verify that a prescription change was associated with a decrease in incidence of a specific diagnosis. However, it has previously been shown that 90% of outpatient antibiotic prescriptions is due to one of the ten most common diagnoses, with lower urinary tract infection being the most common (20%), followed by pharyngotonsillitis (19%), acute otitis media (15%) and acute rhinosinusitis (8%) [11]. It is therefore reasonable to believe that prescription changes of penicillin V and amoxicillin largely correlates to changes in incidence of otitis media, acute rhinosinusitis and acute pharyngotonsillitis. For all three diagnoses, penicillin V is the first-hand choice.

The largest decrease in antibiotic prescription rates was seen among children, in whom upper airway infections in general tend to be more common. Prescription rates for the first-hand choice – penicillin V – were almost halved, and the same was seen for amoxicillin, which is used as the second-hand choice for acute otitis media, and for erythromycin – much more rarely prescribed, but used to treat otitis media in patients allergic to betalactams. The third-hand choice for acute otitis media – amoxiclavulanate – was also prescribed considerably less often during 2020 compared to 2019.

The trend was less pronounced for adult patients, with a 20-30% reduction in penicillin V prescription rates. This resulted in penicillin V being prescribed more often to elderly people than to children, something which has never before been the case. The largest reductions for adults were instead seen for erythromycin and doxycycline, perhaps reflecting that acute rhinosinusitis is more common in adults than in children.

The strength with this study is that it comprises data from an entire country, from databases well known for their accuracy, so trends should be reliable.

As with all register studies, there are also several limitations. Due to the lack of individual data, cause-and-effect relationships are impossible to establish. The decreases in incidence of the studied infections might be partly confounded by changed search patterns, as discussed above. The lack of primary care data in the diagnostics database means that the incidence of uncomplicated infections presented here are not absolute national incidences, however, the incidence ratios should be correct, as reporting to the data base has been stable for many years. For reasons stated above, the absence of primary care data should, under Swedish circumstances, likely lead to an underestimation of the reduced infection incidence. The effects on upper airway infections of social distancing per se ought to be generalisable between countries, however, different policies were adopted in different countries, meaning that the magnitude of the incidence change could vary.

This national study showed an overall decrease by approximately 40% of otitis media, acute rhinosinusitis and acute pharyngotonsillitis during 2020 compared to 2019. Childhood mastoiditis decreased by 40% and peritonsillitis in adults decreased by 20%. The decrease in upper airway infections was accompanied by a similar decrease in antibiotic prescriptions. It will be interesting to study what has happened during 2021, for which data is not yet available, and what has happened after social distancing measures have been released, and – in the case of young children – immunologically naïve persons start to mix again.