Skip to main content
NCBI home page
BMJ Medicine logoLink to BMJ Medicine
. 2025 Sep 4;4(1):e000992. doi: 10.1136/bmjmed-2024-000992

Prescription rates in different groups of outpatients with covid-19 and other acute respiratory infections: comparative observational study based on German routine data

Lena Marie Paschke 1,✉, Kerstin Klimke 1, Maike Below 1
PMCID: PMC12414209  PMID: 40922802

Abstract

Objectives

To identify and quantify prescriptions after a covid-19 infection compared with other acute respiratory infections in previously healthy patients and those with chronic disease.

Design

Comparative observational study based on German routine data.

Setting

Ambulatory care of all residents in Germany with statutory health insurance (88% of the German population).

Participants

Adults receiving a diagnosis of covid-19 or an acute respiratory infection between the fourth quarter of 2020 and the second quarter of 2021 who had rarely (70 797 and 173 822 with covid-19 and acute respiratory infection, respectively) or frequently (900 593 and 1 755 691, respectively) accessed outpatient medical care in the past.

Main outcome measures

Difference in differences in the proportion of prescriptions of relevant drugs before and one year after infection.

Results

In patients who used the healthcare system less frequently before their covid-19 infection than afterwards, increases in prescription rates for antidiabetics (difference in differences 0.23%, P=0.007), antithrombotics (0.71%, P=0.02), and cardiovascular drugs like beta blockers (0.25%, P=0.03) were observed compared with patients with other acute respiratory infections. One year after infection, the difference in antidiabetic prescription rates was highest. Although a peak in antihypertensive prescription rates was observed six months after infection, antithrombotics were predominantly prescribed during the acute phase. Conversely, patients who had already used the healthcare system on a regular basis before their infection showed no significant long term increases in prescription rates across the drug groups analysed.

Conclusions

This study supports findings that diseases such as diabetes and cardiovascular disease are more prevalent after covid-19 than after other acute respiratory infections. Because the effect is apparent in real world data, future societal implications should be considered, including increased disease burden and growing demand for medical care owing to the increasing need for drugs.

Keywords: COVID-19, Diabetes mellitus, Cardiology, Drug therapy, Primary health care

WHAT IS ALREADY KNOWN ON THIS TOPIC

  • The individual risk of requiring diverse therapeutic agents owing to long term sequelae is increased during the six month period after the acute phase of covid-19

  • People with pre-existing conditions are presumably more likely to experience long term sequelae, but young adults and healthy people are also affected

  • Uncertainty remains about the type of long term sequelae and their population related frequency in these two patient groups after a diagnosis of covid-19 compared with other acute respiratory infections

WHAT THIS STUDY ADDS

  • In patients with covid-19 who rarely accessed medical care before their illness, but regularly made visits shortly after, the prevalence of post-acute drugs for diabetes, cardiovascular diseases, and coagulopathy was higher than in patients diagnosed with other acute respiratory infections

  • Increased disease burden and growing demand for medical care because of the increasing need for drugs should be considered in light of ongoing covid-19 waves

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE, OR POLICY

  • It would be beneficial for medical professionals to educate patients about the symptoms of newly emerging diseases (eg, diabetes mellitus) after recovery from covid-19 infection

Introduction

Covid-19 is associated with diverse symptoms in the acute phase and after infection. During the acute phase, some patients have no symptoms, while others have fever, cough, gastrointestinal complaints, or loss of appetite and smell.1 2 Because there is no uniform definition, all symptoms that persist or emerge four weeks after disease onset are defined as post-acute covid-19 symptoms.3 If symptoms persist for three months after the infection or newly appear and last for at least two months, this is referred to as long covid.4 These long term sequelae can occur in several systems of the body, including the cardiovascular, immune, respiratory, or neuropsychological system.5 6 Correspondingly, the probability of using diverse therapeutic agents increases during the six months after the acute phase of covid-19.7

The frequency of occurrence of long covid varies between 1.2% and 20% depending on the definition used and the patient group under consideration.8 Evidence shows that patients with covid-19 place a major additional burden on the healthcare system, using 20% more primary and emergency care resources in the first six months after infection.9 However, it is not yet clear how many patients with covid-19 are permanently dependent on medical care and the reasons why.

Research suggests that older adults and those with pre-existing conditions are more likely to experience long term sequelae,6 10 but young adults and people without pre-existing conditions are also affected.11,13 Research focusing on the long term impact of covid-19 on people who were relatively healthy before their infection is currently limited, even though this would be of interest to society. In addition to the personal consequences, these patients require care from the healthcare system and might have limited availability for work.

The objective of this study is to describe prescription patterns and compare the proportion of prescribed drugs in two patient groups after covid-19 or other acute respiratory infections: firstly, patients who were healthy before the infection but whose health declined after the infection; secondly, patients who presumably had a chronic condition before their infection. The distinction between patients was based on their individual use of the outpatient healthcare system. Patients who rarely used the outpatient healthcare system before covid-19 infection, but regularly afterwards, were assumed to have greater health problems after the infection than before. Patients with frequent visits to a physician before and after their infection were considered to have a chronic disease. Proportions of prescriptions of both patient groups were compared with those of patients with acute respiratory infections using inferential statistics to identify potential correlations between covid-19 infection and prescription rates. Although previous research examined outpatient prescriptions during the acute phase of covid-1914 or up to six months after infection,15 the current study covers one year after infection. In contrast to Mooses and colleagues,15 who examined the individual risk of patients receiving a drug prescription after covid-19 or other acute respiratory infection, our study focuses on the overall proportion of patients requiring therapeutic agents before and after an infection. Population based prescription data from Germany allow this approach to estimate the extent of long term effects on the population.

Methods

Data sources

The study is based on nationwide ambulatory billing claims (VDA, in German Verordnungsdaten Ambulant) and drug prescription data (AVD, in German Arzneiverordnungsdaten) of all residents with statutory health insurance in Germany from 2020 to 2021 (approximately 73.3 million people, 88.1% of the population in 2021). The dataset includes outpatient prescriptions (AVD), diagnoses (VDA), and patient-physician contacts (VDA) with general practitioners and specialty doctors. Data about prescriptions are limited to drugs that were redeemed at a pharmacy. The VDA and AVD data were linked using a patient ID and the quarter of diagnosis and prescription. Because VDA data are only available quarterly, all data were analysed on a quarterly basis.

Diagnoses were identified by the international classification of diseases, 10th revision, German modification (ICD-10-GM). Drugs were identified by anatomical therapeutic chemical (ATC) classification codes and analysed at the second level. Based on treatment recommendations and previous studies, we focused on 16 pharmaceutical groups (table 1). Because we did not want to miss any unexpected effects, all pharmaceutical groups that were prescribed to at least 30 patients each quarter were subsequently examined exploratively.

Table 1. ATC (anatomical therapeutic chemical) groups and their indications.

ATC Description Indication for covid-19
A10 Antidiabetics Diabetes20
A11 Vitamins Anti-inflammatory (acute phase), antioxidant (post-acute): fatigue39 40
B01 Antithrombotic agents Thrombosis41 42
C01 Cardiac therapy Heart failure,43 44 arrhythmia44
C03 Diuretics Heart failure,27 hypertension7 45
C07 Beta-adrenoceptor antagonists Heart failure,27 hypertension,7 45 arrhythmia
C08 Calcium channel blockers Heart failure,27 hypertension,7 45 arrhythmia
C09 Agents with an effect on the renin-angiotensin system Heart failure,27 hypertension7 45
H03 Thyroid therapy Autoimmune disease like Hashimoto's thyroiditis,46 hypothyroidism47
L04 Immunosuppressants Potential immunodysregulation48
M01 Anti-inflammatory and antirheumatic drugs Fever, headache, and myalgias,49 50 arthralgia51
N02 Analgesics Fever, headache, and myalgia,49 50 migraine52
N03 Antiepileptic drugs Neuropathy, neuropathic pain,53 generalised anxiety disorder54
N06 Psychoanaleptics Depression55
N07 Other agents for the nervous system Neuropathy,53 vertigo56
R03 Agents for obstructive respiratory diseases Asthma, chronic obstructive pulmonary disease inflammation57 58
Open in a new tab

Study population

Disease groups

To compare patients with covid-19 and those with other acute respiratory infections, including influenza-like illnesses, we formed two cohorts. Both cohorts were selected based on several criteria (figure 1). A basic prerequisite was patient age ≥18 years. To increase the likelihood of only including patients with accurate data, they were required to have unambiguous sex information and be aged <109 years.

Figure 1. Inclusion and exclusion criteria of study population. ICD, international classification of diseases.

Figure 1

Open in a new tab

The covid cohort included all patients with a covid-19 diagnosis (ICD code U07.1) from the fourth quarter of 2020 to the second quarter of 2021. During this period, patients with suspected covid-19 infection underwent regular testing, along with diagnosis and coding by physicians. Therefore, affected patients could be identified with a high degree of certainty. The earliest quarter between the fourth quarter of 2020 and the second quarter of 2021 with a covid-19 infection was defined as the index quarter per patient. To analyse only patients with a new covid-19 infection, they were excluded if they had a covid-19 diagnosis within two quarters before their index quarter. Because the data did not allow differentiation between isolated covid-19 infection and covid-19 infection with additional acute respiratory infection, patients with covid-19 were not controlled for additional acute respiratory infections.

The acute respiratory infection cohort consisted of all patients who were assigned a confirmed ICD code commonly used for acute upper or lower respiratory tract infections (figure 1) between the fourth quarter of 2020 and the second quarter of 2021. The quarter with the coded infection was defined as the index quarter. Patients with acute respiratory infection were excluded if they had a covid-19 diagnosis coded from the first quarter of 2020 (first occurrence of covid-19 in Germany) to the fourth quarter after their index quarter. They were not controlled for additional acute respiratory infections before or after their index quarter. Because the data do not contain information on whether patients have statutory health insurance, the presence of insurance was verified by at least one physician-patient contact between 2017 and 2019.

Patient groups

To separately analyse patient groups with different morbidities, they were assigned to groups according to their physician visits and frequency of drug prescriptions in the eight quarters before and four quarters after their index diagnosis. Patients who had patient-physician contacts or were prescribed drugs in no more than three of eight quarters before their covid-19 infection were assigned to the sporadic users group. Patients who had patient-physician contacts or drug prescriptions in at least seven of eight quarters were assigned to the frequent users group. In the sporadic user group, the patients of interest were those who showed an increased need for medical care close to the time of their index quarter. Therefore, only patients with visits in two quarters after the index quarter were included. To observe whether their condition improved afterwards, physician-patient contacts in the third and fourth quarters were not mandatory. Frequent users were not expected to need less medical care after the index quarter than before, and so they were only included if they continued to see a physician regularly in each of the following four quarters.

Statistical analysis

All statistical analyses were conducted with R software.16 For each cohort and patient group, we calculated the number of patients with a prescription for each ATC and quarter. For privacy reasons, results for ATC groups with very low numbers of prescriptions (less than five per quarter) are not reported. To identify effects after infection, we observed prescriptions in the post-phase, until four quarters after the individual index quarter (figure 2). To obtain a comparison period that includes quarters independent of the group classification criteria and the early phase of the pandemic (pre-acute phase), prescriptions were also analysed in 15 to 9 quarters before the index quarter (neutral pre-phase).

Figure 2. Study design. Upper panel: three cohorts (cohort 1-3) with covid-19 or other acute respiratory infection (ARI) from fourth quarter of 2020 to second quarter of 2021 were merged to one timeline with −15 to +4 quarters in relation to index (Idx) quarter. Lower panel: patient groups identified based on patient-physician contacts (dark grey) in pre-acute and post-acute phase. Sporadic user defined as maximum of three of eight quarters before index and two quarters after index. Frequent user defined as minimum of seven of eight quarters before index and four quarters after index.

Figure 2

Open in a new tab

To compare the number of prescriptions for the covid-19 and acute respiratory infection groups, the proportion of patients with prescriptions was calculated as a percentage of all patients within each group. We used a difference in differences analysis to compare the proportions before and after the index quarter within and between groups. Disease group (covid-19, acute respiratory infection), time (pre-phase, post-phase), and the interaction effect between both factors were represented as three independent variables in a linear regression model (lm() function from “stats” package16). To quantify the direction and magnitude of the effects, the difference between the proportions in the pre-phase (quarter −15 to −1) and the post-phase (index to +4) for each patient group was calculated and compared. To ensure more reliable inference by accounting for potential violations of homoscedasticity and mild autocorrelation assumptions in the linear regression model, we used robust standard errors, confidence intervals, and P values. The robust covariance matrix was calculated using the HC1 method17 as implemented in the “sandwich” package in R.18 Robust standard errors were derived from this matrix and used for coefficient tests and to compute 95% robust confidence intervals.

If the disease groups differed in age or sex distribution by a standardised mean difference of >0.1, their impact was considered by repeating the analysis with a cohort matched on age and sex. Using the matchit() function from the “MatchIt” package,19 a propensity score based nearest neighbour matching with calliper=0.2 and ration=2 matched a maximum of two patients from the acute respiratory infection group for each patient from the covid-19 group. This study was reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (online supplemental file 1).

Patient and public involvement

There was no direct patient and public participation in the design of this research, as it was based on anonymised secondary data. The findings will be shared with the Associations of Statutory Health Insurance Physicians to inform physicians and thereby indirectly reach patients.

Results

Patient groups

The sporadic user group was much smaller (n=244 619) than the frequent user group (n=2 656 284). As expected, patients with a low number of physician contacts were younger than patients with frequent physician contacts (median 36 and 54.5 years, respectively; figure 3, upper panel). Between patient groups, the sex ratio was reversed, with a greater proportion of men in the sporadic user group (67%) and more women in the frequent user group (63.8%). Patients with covid-19 were on average slightly older than those with acute respiratory infection in the sporadic user group (40.3 and 37.8 years, respectively) and slightly younger in the frequent user group (52.7 and 54.3 years, respectively; see figure 3, lower panels and online supplemental table S1).

Figure 3. Age distribution of patient groups with frequent and sporadic use of healthcare system (upper panel). Boxplots of patient age in frequent (lower left panel) and sporadic (lower right panel) user groups. ARI=acute respiratory infection.

Figure 3

Open in a new tab

Prescriptions

Sporadic users

All ATC groups examined showed similar general prescription patterns with increasing proportions over time (online supplemental table S2, figure 4). In six of the ATC groups, the difference between the pre-phase and post-phase was greater in the covid-19 group than in the acute respiratory infection group (figure 5, online supplemental figure S1, table 2). In a cohort matched for age and sex, the increase in prescription rate was 0.23% higher for antidiabetics (P=0.007), 0.71% higher for antithrombotic agents (P=0.021), 0.12% higher for diuretics (P=0.049), and 0.25% higher for beta-adrenoceptor antagonists (P=0.032).

Figure 4. Mean prescription rates during pre-phase (quarter −15 to −1) and post-phase (index to +4) for sporadic user group on second ATC (anatomical therapeutic chemical) code level. Lower right panel: sum for ATC codes C03, C07-C09. For visualisation purposes, only a selection of ATC codes analysed are shown; all others are included in online supplemental table S2. ARI=acute respiratory infection.

Figure 4

Open in a new tab
Figure 5. Prescription rates over time for sporadic user group on second ATC (anatomical therapeutic chemical) code level. Lower right panel: sum for ATC codes C03, C07-C09. For visualisation purposes, only a selection of ATC codes analysed are shown; all others are included in online supplemental figure S1. ARI=acute respiratory infection.

Figure 5

Open in a new tab
Table 2. Regression results for interaction effect in sporadic user group.
ATC Total cohort Matched cohort
Estimate SE (95% CI) P value Estimate SE (95% CI) P value
A10 Antidiabetics 0.33 0.08 (0.17 to 0.49) <0.001 0.23 0.08 (0.07 to 0.39) 0.007
A11 Vitamins 0.01 0.03 (−0.05 to 0.08) 0.659 −0.0002 0.03 (−0.06 to 0.06) 0.993
B01 Antithrombotic agents 0.87 0.29 (0.27 to 1.46) 0.006 0.713 0.29 (0.12 to 1.31) 0.021
C01 Cardiac therapy 0.03 0.02 (0.00 to 0.06) 0.057 0.01 0.02 (−0.03 to 0.05) 0.562
C03 Diuretics 0.24 0.05 (0.14 to 0.35) <0.001 0.116 0.06 (0.00 to 0.23) 0.049
C07 Beta-adrenoceptor antagonists 0.46 0.10 (0.25 to 0.68) <0.001 0.246 0.11 (0.02 to 0.47) 0.032
C08 Calcium channel blockers 0.27 0.07 (0.11 to 0.42) 0.001 0.117 0.08 (−0.04 to 0.28) 0.15
C09 Renin-angiotensin system agents 0.99 0.33 (0.32 to 1.65) 0.005 0.49 0.34 (−0.20 to 1.18) 0.158
H03 Thyroid therapy 0.1 0.13 (−0.16 to 0.36) 0.447 0.025 0.13 (−0.24 to 0.29) 0.85
M01 Anti-inflammatories/antirheumatics −0.31 0.70 (−1.72 to 1.10) 0.655 −0.414 0.71 (−1.85 to 1.02) 0.562
N02 Analgesics 0.19 0.48 (−0.77 to 1.15) 0.692 0.075 0.48 (−0.90 to 1.05) 0.878
N03 Antiepileptic drugs 0.06 0.04 (−0.01 to 0.13) 0.098 0.034 0.04 (−0.04 to 0.11) 0.355
N06 Psychoanaleptics −0.07 0.16 (−0.40 to 0.25) 0.662 −0.125 0.16 (−0.45 to 0.20) 0.445
R03 Obstructive respiratory disease 0.51 0.59 (−0.69 to 1.71) 0.396 0.398 0.60 (−0.82 to 1.61) 0.51
Antihypertensives (C03, C07-C09) 1.33 0.39 (0.53 to 2.12) 0.002 0.687 0.41 (−0.14 to 1.52) 0.102
Open in a new tab

Robust standard errors (SEs), robust 95% confidence intervals (CIs), and robust P values for interaction effect for total cohort and cohort matched by age and sex.

*

Regression coefficient (corresponds to difference in difference).

ATC, anatomical therapeutic chemical; SE, Standard error.

Because the ATC groups C03 and C07-C09 share a common indication of lowering blood pressure in essential hypertension, they were evaluated together. For this analysis, patients were counted only once, even if they were prescribed several drugs from different C-ATC groups. With a difference of 1.33% between the proportion of prescriptions for patients with covid-19 and those with acute respiratory infection, this group showed the largest numerical discrepancy in the unmatched cohort, but not in the matched cohort. As with antidiabetics, the difference between the disease groups tended to increase from the index quarter to post-index quarters (figure 5). In contrast, the largest difference for antithrombotics was observed in the index quarter.

The exploratory analysis of all ATC groups revealed three general prescription patterns. Some drugs were prescribed more frequently only in the index quarter, probably for the treatment of acute illnesses; this applies to cough and cold remedies (R05) and corticosteroids (H02) in patients with covid-19, and antibiotics (J01) and rhinologics (R01) in patients with acute respiratory infection (online supplemental figure S2 and table S3). The second group, which includes gout remedies (M04) and antianemics (B03), reaches its maximum prescription rate shortly after the index quarter and then declines again. In contrast, the third group continues to increase or reaches a plateau after the index quarter. This pattern is identified in antihypertensives (C02), antiparkinsonian drugs (N04), urological drugs (G04), drugs affecting lipid metabolism (C10), and antiepileptic drugs (N03).

Frequent users

Among frequent users, there was no significant increase in prescription rates after the index event in patients with covid-19 compared with those with acute respiratory infection for any of the ATC groups that were the focus of this analysis (figure 6 and table 3). Only a group of drugs for the nervous system (N07; difference in differences −0.08, P<0.001) showed a significant interaction, with a higher increased prescription rate after the index event in patients with acute respiratory infection than in those with covid-19. A general linear increase in prescription rates resulted in consistently higher values from the index event onwards compared with the period before the index event (figure 7, online supplemental table S4 and figure S3).

Figure 6. Mean prescription rates during pre-phase (quarter −15 to −1) and post-phase (index to +4) for frequent user group on second ATC (anatomical therapeutic chemical) code level. Lower right panel: sum for ATC codes C03, C07-C09. For visualisation purposes, only a selection of ATC codes analysed are shown; all others are included in online supplemental table S2. ARI=acute respiratory infection.

Figure 6

Open in a new tab
Table 3. Regression results for interaction effect in frequent user group.
ATC Estimate SE (95% CI) P value
A10 Antidiabetics −0.22 0.33 (−0.89 to 0.46) 0.52
A11 Vitamins 0.04 0.07 (−0.11 to 0.19) 0.589
B01 Antithrombotic agents 0.82 0.61 (−0.41 to 2.06) 0.183
C01 Cardiac therapy −0.07 0.05 (−0.17 to 0.03) 0.155
C03 Diuretics −0.13 0.44 (−1.02 to 0.77) 0.779
C07 Beta-adrenoceptor antagonists −0.28 0.43 (−1.16 to 0.60) 0.525
C08 Calcium channel blockers −0.5 0.37 (−1.24 to 0.24) 0.178
C09 Renin-angiotensin system agents −1.31 0.75 (−2.83 to 0.21) 0.09
H03 Thyroid therapy −0.53 0.38 (−1.30 to 0.25) 0.178
L04 Immunosuppressants −0.16 0.08 (−0.31 to 0.00) 0.044
M01 Anti-inflammatories/antirheumatics −0.82 0.48 (−1.80 to 0.16) 0.099
N02 Analgesics 0.29 0.77 (−1.27 to 1.84) 0.709
N03 Antiepileptic drugs −0.07 0.19 (−0.45 to 0.31) 0.707
N06 Psychoanaleptics −0.33 0.38 (−1.11 to 0.46) 0.404
N07 Other drugs for the nervous system −0.08 0.01 (−0.11 to −0.05) <0.001
R03 Obstructive respiratory disease 0.48 0.85 (−1.24 to 2.21) 0.575
Antihypertensives (C03, C07-C09) −1.59 1.00 (−3.61 to 0.44) 0.121
Open in a new tab

Robust standard errors (SEs), robust 95% confidence intervals (CIs), and robust P values for interaction effect for total cohort.

*

Regression coefficient (corresponds to differece in difference).

ATC, anatomical therapeutic chemical .

Figure 7. Prescription rates over time for frequent user group on second ATC (anatomical therapeutic chemical) code level. Lower right panel: sum for ATC codes C03, C07-C09. For visualisation purposes, only a selection of ATC codes analysed are shown; all others are included in online supplemental figure S3. ARI=acute respiratory infection.

Figure 7

Open in a new tab

Compared with the period before the index event, the increase in antithrombotic (B01) prescriptions during the index quarter was more than twice as high in patients with covid 19 than in those with acute respiratory infection (+4.96% v +1.88%). This transient effect was temporary (figure 7), and the interaction was not significant (table 3).

As figure 7 shows, there was a notable divergence in the prescription rates for obstructive respiratory diseases between the two disease groups in the period before the index event. On average, patients with covid-19 were prescribed 4.48% fewer drugs from this group than those with acute respiratory infection. This trend continued throughout the observation period, with only the index (2.5%) and the following quarter (3.2%) showing a smaller gap.

The exploratory analysis of all ATC groups reveals two additional ATC groups (A16, H05) in which the prescription rates for patients with covid-19 were marginally higher in the long term than for those with acute respiratory infection (online supplemental figure S4a,b and table S5). Increases during the index quarter, such as for cough and cold remedies (R05), corticosteroids for systemic use (H02), and drugs for gastrointestinal disorders (A03), were temporary.

Discussion

Principal findings

This study shows that among patients who rarely used outpatient healthcare services before their acute illness, the proportion with prescriptions for antidiabetics, antithrombotics, or cardiovascular drugs like beta blockers after their infection was higher in patients with covid-19 than in those with other acute respiratory infections. In patients who were regular users of the healthcare system before disease onset and who might already have a chronic disease, no long term increase in prescription rates was observed in any of the analysed ATC groups that were the focus of the analysis.

Findings in context

We observed an increase in prescriptions for antidiabetic drugs and two cardiovascular drugs, but only in the sporadic user group and not in the frequent user group, which is similar to the findings of Mooses and colleagues.15 Several studies have already shown that there is an increased risk of developing diabetes after covid-19 infection.20,22 Our study shows that, even one year after covid-19 infection, antidiabetic prescriptions among sporadic users were almost a third greater than for patients with acute respiratory infection.

The lack of a long term increase in prescription rates in frequent users, who were on average more than 10 years older than sporadic users, aligns with the study by Mooses and colleagues,15 who found the strongest effects in the 40-64 age group, not the ≥65 age group. Subramanian and colleagues12 also found that long term effects were more pronounced in younger patients, not older patients, after age adjustment. However, there is increasing evidence that certain pre-existing conditions also increase the risk of certain long term symptoms after covid-19.23 As our heterogeneous group of frequent users included patients with all types of conditions that require regular medical care, these effects might not be visible. Additionally, prescription rates in this group are already high, which limits the possibility of identifying new drugs. The potential need for dose escalation owing to disease worsening after covid-19 cannot be detected in the current analysis because of the study design.

Meaning of the study

Our observation that a higher increase in prescription rates was only found in patients who were presumed to be healthy before their infection, but not in those presumed to have a chronic illness, provides evidence that covid-19 can act as a trigger for an (immunological) disease.24,26 While the disease may already be present in patients who were classified as presumably chronic (frequent users), it may be triggered by covid-19 in patients presumed to be previously healthy (sporadic users). Overall, it is important to note that sporadic users of the healthcare system represent a relatively small, select group in which unmeasured factors might be particularly influential, which could limit the generalisability of the findings.

The observed increase in prescriptions for diuretics and beta-adrenoceptor antagonists is consistent with previous evidence indicating that a potential long term consequence of covid-19 is the development of cardiac symptoms, including heart failure, arrhythmias, and hypertension.27,33 Our study design reveals maximum differences in the first and second quarter after disease onset, indicating that some patients with covid-19 might require drugs for the first time after three to six months. A plateau or a slow decline in rates indicates that the incidence of new cardiac symptoms might decrease over time, potentially leading to a reduction in the need for drug treatment in some patients.

Strengths and weaknesses of this study

Although the differences found in prescription rates show an association with covid-19 infection, our type of analysis cannot prove a causal relation. Additionally, the diagnostic difficulties in distinguishing between covid-19 and acute respiratory infections did not allow us to exclude patients with covid-19 and concomitant acute respiratory infections. The observed differences might not be attributable to covid-19 alone, but could also relate to acute respiratory infections or other unaccounted for emerging diseases.

The study design included a large cohort and a one year observation period after infection, providing insight into the frequency and severity of the distinct long term effects of covid-19. The exploratory analysis of temporal trends in prescription rates across all ATC codes adds further evidence to the potential long term effects of covid-19.

The subdivision of patients according to their use of the healthcare system provides insight into the actual proportional change in those who are dependent on a drug prescription. At the same time, the categorisation, and the use of routine data in general pose a risk of bias owing to confounding in unconsidered or unmeasured differences between comparison groups. The selection of groups based on respiratory tract infections does not initially suggest any systematic differences in the two groups. In the group of sporadic users, this assumption is confirmed by similar prescription rates in the period before infection. However, the results for the cohort matched by age and sex indicate that the differences in prescribing of cardiac drugs could be partly explained by group differences. In the frequent users, generally higher prescription rates for obstructive respiratory medications (R03) in patients with acute respiratory infection suggest that patients with pre-existing respiratory diseases were more likely to be found in the acute respiratory infection group than in the covid-19 group. This could be a consequence of patients with pre-existing respiratory disease taking greater precautions to protect themselves against covid-19 infection than those without pre-existing respiratory disease because of the possibly more severe disease course associated with such conditions.34 This difference could have influenced the results because higher prevalence of pre-existing conditions in the acute respiratory infection group might increase the risk of long term consequences after infection.

The analysed period encompasses the second and third pandemic waves in Germany, which were caused by the covid-19 wildtype and alpha variant, respectively. Given that later virus variants, such as omicron, were associated with lower mortality35 and milder disease progression,36 and levels of population immunity have increased, it is possible that lower prescription rates might be the result. However, reinfections could also lead to an increased risk of severe courses and long term consequences,37 38 which in turn could make more patients dependent on drug treatment in outpatient care.

Unanswered questions and future research

This exploratory analysis indicates that covid-19 has the potential to cause secondary diseases. In addition to antidiabetics, cardiac drugs, and antithrombotics, these drugs and their indications should continue to be monitored and investigated in relation to covid-19. Although the observed increases in prescription rates of 0.12% to 0.71% might seem modest, the cumulative impact of this increased need for treatment could be substantial when extrapolated to the large number of people who experienced a single covid-19 infection or several infections. The heightened risk should be considered in the context of future covid-19 infections and the potential impact on the healthcare system should be taken into account.

Supplementary material

online supplemental file 1
bmjmed-4-1-s001.pdf (104.7KB, pdf)
DOI: 10.1136/bmjmed-2024-000992
online supplemental file 2
bmjmed-4-1-s002.pdf (1.8MB, pdf)
DOI: 10.1136/bmjmed-2024-000992

Footnotes

Funding: As the funder, the Central Research Institute of Ambulatory Health Care in Germany had no role in study design, data collection and analysis, or preparation of the manuscript.

Provenance and peer review: Not commissioned; externally peer reviewed.

Data availability free text: The Central Research Institute of Ambulatory Health Care in Germany is not permitted to make publicly available the sensitive health data analysed in the study.

Ethics approval: This study analysed routinely collected, fully pseudonymised claims data. As no personal identifiers were accessible, neither patient consent nor ethics committee approval was required under applicable law.

Data availability statement

No data are available.

References

  • 1.Rodebaugh TL, Frumkin MR, Reiersen AM, et al. Acute Symptoms of Mild to Moderate COVID-19 Are Highly Heterogeneous Across Individuals and Over Time. Open Forum Infect Dis. 2021;8:ofab090. doi: 10.1093/ofid/ofab090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.WHO Coronavirus disease (COVID-19) 2023. [05-Apr-2024]. https://www.who.int/news-room/fact-sheets/detail/coronavirus-disease-(covid-19) Available. Accessed.
  • 3.NICE Overview | COVID-19 rapid guideline: managing the long-term effects of COVID-19 | guidance. 2024. [05-Apr-2024]. https://www.nice.org.uk/guidance/ng188 Available. Accessed. [PubMed]
  • 4.WHO Post COVID-19 condition (long COVID) 2022. [10-Apr-2024]. https://www.who.int/europe/news-room/fact-sheets/item/post-covid-19-condition Available. Accessed.
  • 5.Nalbandian A, Sehgal K, Gupta A, et al. Post-acute COVID-19 syndrome. Nat Med. 2021;27:601–15. doi: 10.1038/s41591-021-01283-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Peter RS, Nieters A, Kräusslich H-G, et al. Post-acute sequelae of covid-19 six to 12 months after infection: population based study. BMJ. 2022;379:e071050. doi: 10.1136/bmj-2022-071050. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Al-Aly Z, Xie Y, Bowe B. High-dimensional characterization of post-acute sequelae of COVID-19. Nature New Biol. 2021;594:259–64. doi: 10.1038/s41586-021-03553-9. [DOI] [PubMed] [Google Scholar]
  • 8.Thompson EJ, Williams DM, Walker AJ, et al. Long COVID burden and risk factors in 10 UK longitudinal studies and electronic health records. Nat Commun. 2022;13:3528. doi: 10.1038/s41467-022-30836-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Valdivieso-Martínez B, Sauri I, Philibert J, et al. Impact of Long-COVID on Health Care Burden: A Case Control Study. JCM. 2023;12:5768. doi: 10.3390/jcm12185768. [DOI] [Google Scholar]
  • 10.Tsampasian V, Elghazaly H, Chattopadhyay R, et al. Risk Factors Associated With Post-COVID-19 Condition: A Systematic Review and Meta-analysis. JAMA Intern Med. 2023;183:566–80. doi: 10.1001/jamainternmed.2023.0750. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Mogensen I, Ekström S, Hallberg J, et al. Post COVID-19 symptoms are common, also among young adults in the general population. Sci Rep. 2023;13:11300. doi: 10.1038/s41598-023-38315-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Subramanian A, Nirantharakumar K, Hughes S, et al. Symptoms and risk factors for long COVID in non-hospitalized adults. Nat Med. 2022;28:1706–14. doi: 10.1038/s41591-022-01909-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Dennis A, Cuthbertson DJ, Wootton D, et al. Multi-organ impairment and long COVID: a 1-year prospective, longitudinal cohort study. J R Soc Med. 2023;116:97–112. doi: 10.1177/01410768231154703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Belleudi V, Finocchietti M, Fortinguerra F, et al. Drug Prescriptions in the Outpatient Management of COVID-19: Evidence-Based Recommendations Versus Real Practice. Front Pharmacol. 2022;13:825479. doi: 10.3389/fphar.2022.825479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Mooses K, Vesilind K, Oja M, et al. The use of prescription drugs and health care services during the 6-month post-COVID-19 period. Sci Rep. 2023;13:11638. doi: 10.1038/s41598-023-38691-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.R Core Team R: a language and environment for statistical computing. R foundation for statistical computing. 2022
  • 17.White H. A Heteroskedasticity-Consistent Covariance Matrix Estimator and a Direct Test for Heteroskedasticity. Econometrica. 1980;48:817. doi: 10.2307/1912934. [DOI] [Google Scholar]
  • 18.Zeileis A. Econometric Computing with HC and HAC Covariance Matrix Estimators. J Stat Soft. 2004;11:1–17. doi: 10.18637/jss.v011.i10. [DOI] [Google Scholar]
  • 19.Ho DE, Imai K, King G, et al. MatchIt: Nonparametric Preprocessing for Parametric Causal Inference. J Stat Soft. 2011;42:1–28. doi: 10.18637/jss.v042.i08. [DOI] [Google Scholar]
  • 20.Banerjee M, Pal R, Dutta S. Risk of incident diabetes post-COVID-19: A systematic review and meta-analysis. Prim Care Diabetes. 2022;16:591–3. doi: 10.1016/j.pcd.2022.05.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Rizvi AA, Kathuria A, Al Mahmeed W, et al. Post-COVID syndrome, inflammation, and diabetes. J Diabetes Complications. 2022;36:108336. doi: 10.1016/j.jdiacomp.2022.108336. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Sathish T, Kapoor N, Cao Y, et al. Proportion of newly diagnosed diabetes in COVID-19 patients: A systematic review and meta-analysis. Diabetes Obes Metab. 2021;23:870–4. doi: 10.1111/dom.14269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Russell CD, Lone NI, Baillie JK. Comorbidities, multimorbidity and COVID-19. Nat Med. 2023;29:334–43. doi: 10.1038/s41591-022-02156-9. [DOI] [PubMed] [Google Scholar]
  • 24.Sher EK, Ćosović A, Džidić-Krivić A, et al. Covid-19 a triggering factor of autoimmune and multi-inflammatory diseases. Life Sci. 2023;319:121531. doi: 10.1016/j.lfs.2023.121531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Wang Y, Guo H, Wang G, et al. COVID-19 as a Trigger for Type 1 Diabetes. J Clin Endocrinol Metab. 2023;108:2176–83. doi: 10.1210/clinem/dgad165. [DOI] [PubMed] [Google Scholar]
  • 26.Zebardast A, Hasanzadeh A, Ebrahimian Shiadeh SA, et al. COVID-19: A trigger of autoimmune diseases. Cell Biol Int. 2023;47:848–58. doi: 10.1002/cbin.11997. [DOI] [PubMed] [Google Scholar]
  • 27.Zuin M, Rigatelli G, Roncon L, et al. Risk of incident heart failure after COVID-19 recovery: a systematic review and meta-analysis. Heart Fail Rev. 2023;28:859–64. doi: 10.1007/s10741-022-10292-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Elseidy SA, Awad AK, Vorla M, et al. Cardiovascular complications in the Post-Acute COVID-19 syndrome (PACS) IJC Heart & Vasculature. 2022;40:101012. doi: 10.1016/j.ijcha.2022.101012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Huseynov A, Akin I, Duerschmied D, et al. Cardiac Arrhythmias in Post-COVID Syndrome: Prevalence, Pathology, Diagnosis, and Treatment. Viruses . 2023;15:389. doi: 10.3390/v15020389. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Rahmati M, Koyanagi A, Banitalebi E, et al. The effect of SARS‐CoV‐2 infection on cardiac function in post‐COVID‐19 survivors: A systematic review and meta‐analysis. J Med Virol. 2023;95:e28325. doi: 10.1002/jmv.28325. [DOI] [PubMed] [Google Scholar]
  • 31.Raman B, Bluemke DA, Lüscher TF, et al. Long COVID: post-acute sequelae of COVID-19 with a cardiovascular focus. Eur Heart J. 2022;43:1157–72. doi: 10.1093/eurheartj/ehac031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Roca-Fernandez A, Wamil M, Telford A, et al. Cardiac abnormalities in Long COVID 1-year post-SARS-CoV-2 infection. Open Heart. 2023;10:e002241. doi: 10.1136/openhrt-2022-002241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Xie Y, Xu E, Bowe B, et al. Long-term cardiovascular outcomes of COVID-19. Nat Med. 2022;28:583–90. doi: 10.1038/s41591-022-01689-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Camacho-Rivera M, Islam JY, Vidot DC. Associations Between Chronic Health Conditions and COVID-19 Preventive Behaviors Among a Nationally Representative Sample of U.S. Adults: An Analysis of the COVID Impact Survey. Health Equity. 2020;4:336–44. doi: 10.1089/heq.2020.0031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Ward IL, Bermingham C, Ayoubkhani D, et al. Risk of covid-19 related deaths for SARS-CoV-2 omicron (B.1.1.529) compared with delta (B.1.617.2): retrospective cohort study. BMJ. 2022;378:e070695. doi: 10.1136/bmj-2022-070695. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Wolter N, Jassat W, Walaza S, et al. Early assessment of the clinical severity of the SARS-CoV-2 omicron variant in South Africa: a data linkage study. The Lancet. 2022;399:437–46. doi: 10.1016/S0140-6736(22)00017-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Adrielle Dos Santos L, Filho P de G, Silva AMF, et al. Recurrent COVID-19 including evidence of reinfection and enhanced severity in thirty Brazilian healthcare workers. J Infect. 2021;82:399–406. doi: 10.1016/j.jinf.2021.01.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Bowe B, Xie Y, Al-Aly Z. Postacute sequelae of COVID-19 at 2 years. Nat Med . 2023;29:2347–57. doi: 10.1038/s41591-023-02521-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Rawat D, Roy A, Maitra S, et al. Vitamin C and COVID-19 treatment: A systematic review and meta-analysis of randomized controlled trials. Diabetes Metab Syndr. 2021;15:102324. doi: 10.1016/j.dsx.2021.102324. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Vollbracht C, Kraft K. Feasibility of Vitamin C in the Treatment of Post Viral Fatigue with Focus on Long COVID, Based on a Systematic Review of IV Vitamin C on Fatigue. Nutrients. 2021;13:1154. doi: 10.3390/nu13041154. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.AWMF S1-leitlinie long/ post-COVID. 2022. [10-Apr-2024]. https://register.awmf.org/de/leitlinien/detail/020-027 Available. Accessed.
  • 42.Bikdeli B, Madhavan MV, Jimenez D, et al. COVID-19 and Thrombotic or Thromboembolic Disease: Implications for Prevention, Antithrombotic Therapy, and Follow-Up: JACC State-of-the-Art Review. J Am Coll Cardiol. 2020;75:2950–73. doi: 10.1016/j.jacc.2020.04.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Medvegy M, Simonyi G. Supplementary Therapeutic Possibilities to Alleviate Myocardial Damage Due to Microvascular Dysfunction in Coronavirus Disease 2019 (COVID-19) Cardiol Ther. 2021;10:1–7. doi: 10.1007/s40119-021-00216-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Nabeh OA, Helaly MM, Menshawey R, et al. Contemporary approach to understand and manage COVID-19-related arrhythmia. Egypt Heart J . 2021;73:76. doi: 10.1186/s43044-021-00201-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Matsumoto C, Shibata S, Kishi T, et al. Long COVID and hypertension-related disorders: a report from the Japanese Society of Hypertension Project Team on COVID-19. Hypertens Res. 2023;46:601–19. doi: 10.1038/s41440-022-01145-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Brancatella A, Viola N, Santini F, et al. COVID-induced thyroid autoimmunity. Best Pract Res Clin Endocrinol Metab. 2023;37:101742. doi: 10.1016/j.beem.2023.101742. [DOI] [PubMed] [Google Scholar]
  • 47.Burekovic A, Halilovic D, Sahbaz A. Hypothyroidism and Subclinical Hypothyroidism as a Consequence of COVID-19 Infection. Med Arch. 2022;76:12–6. doi: 10.5455/medarh.2022.76.12-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Karlsson AC, Humbert M, Buggert M. The known unknowns of T cell immunity to COVID-19. Sci Immunol. 2020;5:eabe8063. doi: 10.1126/sciimmunol.abe8063. [DOI] [PubMed] [Google Scholar]
  • 49.NICE Overview | COVID-19 rapid guideline: managing COVID-19 | guidance. 2021. [22-Apr-2024]. https://www.nice.org.uk/guidance/ng191 Available. Accessed.
  • 50.NIH Nonhospitalized adults: therapeutic management. COVID-19 treat. Guidel. 2024. [22-Apr-2024]. https://www.covid19treatmentguidelines.nih.gov/management/clinical-management-of-adults/nonhospitalized-adults--therapeutic-management/ Available. Accessed.
  • 51.Ciaffi J, Vanni E, Mancarella L, et al. Post-Acute COVID-19 Joint Pain and New Onset of Rheumatic Musculoskeletal Diseases: A Systematic Review. Diagnostics (Basel) 2023;13:1850. doi: 10.3390/diagnostics13111850. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Straburzyński M, Kuca-Warnawin E, Waliszewska-Prosół M. COVID-19-related headache and innate immune response - a narrative review. Neurol Neurochir Pol. 2023;57:43–52. doi: 10.5603/PJNNS.a2022.0049. [DOI] [PubMed] [Google Scholar]
  • 53.Attal N, Martinez V, Bouhassira D. Potential for increased prevalence of neuropathic pain after the COVID-19 pandemic. Pain Rep. 2021;6:e884. doi: 10.1097/PR9.0000000000000884. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Daly M, Robinson E. Depression and anxiety during COVID-19. The Lancet. 2022;399:518. doi: 10.1016/S0140-6736(22)00187-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Mazza MG, De Lorenzo R, Conte C, et al. Anxiety and depression in COVID-19 survivors: Role of inflammatory and clinical predictors. Brain Behav Immun. 2020;89:594–600. doi: 10.1016/j.bbi.2020.07.037. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Korres G, Kitsos DK, Kaski D, et al. The Prevalence of Dizziness and Vertigo in COVID-19 Patients: A Systematic Review. Brain Sci. 2022;12:948. doi: 10.3390/brainsci12070948. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Ramakrishnan S, Nicolau DV, Jr, Langford B, et al. Inhaled budesonide in the treatment of early COVID-19 (STOIC): a phase 2, open-label, randomised controlled trial. Lancet Respir Med. 2021;9:763–72. doi: 10.1016/S2213-2600(21)00160-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Vasanthakumar N. Beta-Adrenergic Blockers as a Potential Treatment for COVID-19 Patients. Bioessays. 2020;42:e2000094. doi: 10.1002/bies.202000094. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

    This section collects any data citations, data availability statements, or supplementary materials included in this article.

    Supplementary Materials

    online supplemental file 1
    bmjmed-4-1-s001.pdf (104.7KB, pdf)
    DOI: 10.1136/bmjmed-2024-000992
    online supplemental file 2
    bmjmed-4-1-s002.pdf (1.8MB, pdf)
    DOI: 10.1136/bmjmed-2024-000992

    Data Availability Statement

    No data are available.

    Articles from BMJ Medicine are provided here courtesy of BMJ Publishing Group

    RESOURCES