key: cord-0795298-2aqwzzr2 authors: Abu-Jamous, B.; Anisimovich, A.; Baxter, J.; Mackillop, L.; Vizcaychipi, M. P.; McCarthy, A.; Khan, R. T. title: Associations of comorbidities and medications with COVID-19 outcome: A retrospective analysis of real-world evidence data date: 2020-08-23 journal: nan DOI: 10.1101/2020.08.20.20174169 sha: c2fa098a2a0141d4af3f0e322ff41c60c8bd50ee doc_id: 795298 cord_uid: 2aqwzzr2 Background: Hundreds of thousands of deaths have already been recorded for patients with the severe acute respiratory syndrome coronavirus (SARS-CoV-2; aka COVID-19). Understanding whether there is a relationship between comorbidities and COVID-19 positivity will not only impact clinical decisions, it will also allow an understanding of how better to define the long-term complications in the groups at risk. In turn informing national policy on who may benefit from more stringent social distancing and shielding strategies. Furthermore, understanding the associations between medications and certain outcomes may also further our understanding of indicators of vulnerability in people with COVID-19 and co-morbidities. Methods: Electronic healthcare records (EHR) from two London hospitals were analysed between 1st January and 27th May 2020. 5294 patients presented to the hospitals in whom COVID status was formally assessed; 1253 were positive for COVID-19 and 4041 were negative. This dataset was analysed to identify associations between comorbidities and medications, separately and two outcomes: (1) presentation with a COVID-19 positive diagnosis, and (2) inpatient death following COVID-19 positive diagnosis. Medications were analysed in different time windows of prescription to differentiate between short-term and long-term medications. All analyses were done with controls (without co-morbidity) matched for age, sex, and number of admissions, and a robustness approach was conducted to only accept results that consistently appear when the analysis is repeated with different proportions of the data. Results: We observed higher COVID-19 positive presentation for patients with hypertension (1.7 [1.3-2.1]) and diabetes (1.6 [1.2-2.1]). We observed higher inpatient COVID-19 mortality for patients with hypertension (odds ratio 2.7 [95% CI 1.9-3.9]), diabetes (2.2 [1.4-3.5]), congestive heart failure (3.1 [1.5-6.4]), and renal disease (2.6 [1.4-5.1]). We also observed an association with reduced COVID-19 mortality for diabetic patients for whom anticoagulants (0.11 [0.03-0.50]), lipid-regulating drugs (0.15 [0.04-0.58]), penicillins (0.20 [0.06-0.63]), or biguanides (0.19 [0.05-0.70]) were administered within 21 days after their positive COVID-19 test with no evidence that they were on them before, and for hypertensive patients for whom anticoagulants (0.08 [0.02-0.35]), antiplatelet drugs (0.10 [0.02-0.59]), lipid-regulating drugs (0.15 [0.05-0.46]), penicillins (0.14 [0.05-0.45]), or angiotensin-converting enzyme inhibitors (ARBs) (0.06 [0.01-0.53]) were administered within 21 days post-COVID-19-positive testing with no evidence that they were on them before. Moreover, long-term antidiabetic drugs were associated with reduced COVID-19 mortality in diabetic patients (0.26 [0.10-0.67]). Conclusions: We provided real-world evidence for observed associations between COVID-19 outcomes and a number of comorbidities and medications. These results require further investigation and replication in other data sets. Background Hundreds of thousands of deaths have already been recorded for 30 patients with the severe acute respiratory syndrome coronavirus (SARS-CoV-2; aka 31 . Understanding whether there is a relationship between comorbidities 32 and COVID-19 positivity will not only impact clinical decisions, it will also allow an 33 understanding of how better to define the long-term complications in the groups at 34 risk. In turn informing national policy on who may benefit from more stringent social 35 distancing and shielding strategies. Furthermore, understanding the associations 36 between medications and certain outcomes may also further our understanding of 37 indicators of vulnerability in people with COVID-19 and co-morbidities. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20174169 doi: medRxiv preprint no evidence that they were on them before, and for hypertensive patients for whom 59 Conclusions We provided real-world evidence for observed associations between 66 COVID-19 outcomes and a number of comorbidities and medications. These results 67 require further investigation and replication in other data sets. 68 Introduction 69 As of the 16 th of August 2020, the COVID-19 pandemic has resulted in more than 70 770,000 deaths worldwide over the course of a few months. Although most of the 71 confirmed cases of COVID-19 infection show mild or no symptoms, a global 72 concern exists in ensuring that healthcare systems can cope with those that require 73 hospitalisation, especially with the coming winter pressure and while preparing for a 74 potential second wave. Correct identification of individuals with higher risk of 75 presentation or of death due to COVID-19 will not only help hospitals better identify 76 those in need of hospitalisation but will also assist the community in identifying the 77 vulnerable who require more careful shielding. 78 Hypertension and diabetes have been shown to be associated with poorer outcome 79 in COVID-19. (Perico, et al., 2020 preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20174169 doi: medRxiv preprint We examined the association of comorbidities with COVID-19 positive presentation 97 as well as with inpatient mortality of COVID-19 patients. This is realised by using 98 Fisher's exact test to compare the outcomes of patients with a given comorbidity 99 (the active arm) and patients without that comorbidity (the control arm). To account 100 for possible confounders, the two arms were matched on age, sex, and the number 101 of admissions of patients in their available history (Figure 1 (b) ). To reduce false 102 discoveries, multiple hypothesis testing was conducted using the Benjamini 103 Hochberg (BH) method. Furthermore, a robustness analysis was carried out by 104 running the same test experiment once with 100% of the available data and 20 105 more times with randomly selected proportions (60%, 70%, 80%, or 90%) of the 106 data. An association is considered significant if its adjusted p-value was smaller 107 than 0.05 in more than 70% of these individual experiments (Figure 1 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20174169 doi: medRxiv preprint and asthma were individually assessed in the analysis (Table 1) . 121 Table 1 . ICD-10 codes used for comorbidity grouping from (Quan, et al., 2005) * Disease group ICD-10 codes Myocardial infarction I21, I22, I252 Congestive heart failure I43, I50, I099, I110, I130, I132, I255, I420, I425, I426, I427, I428, I429, P290 Peripheral vascular disease I70, I71, I731, I738, I739, I771, I790, I792, K551, K558, K559, Z958, Z959 Cerebrovascular disease G45, G46, I60, I61, I62, I63, I64, I65, I66, I67, I68, I69, (Quan, et al., 2005) . The two changes are splitting the "Chronic pulmonary disease" group into "Chronic pulmonary disease excluding asthma" and "Asthma" and adding the "Hypertension" group. We investigated the association of pharmacological therapy with COVID-19 positive 124 presentation and inpatient mortality in COVID-19 patients. To assess the 125 association of a given medication with these end points, we utilised the same 126 stringent statistical approach that was described above for the comorbidities 127 analysis, including matching on age, sex, and number of admissions, as well as the 128 robustness analysis with different proportions of the full dataset ( Figure 1) . 129 However, two further aspects were considered in this analysis; the first was 130 conditioning on comorbidities; that is, the association of each medication with end 131 All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20174169 doi: medRxiv preprint points was assessed separately for patients with different comorbidities (Figure 2) . 132 This is to reduce the possibility of confounding observed association of medications 133 with outcomes by their association with diseases. The definitions of comorbidity 134 ICD-10 codes were based on Table 1 Medications of significant association with end points when prescribed to patients with the comorbidity "disease X" Medications of significant association with end points when prescribed to patients with the comorbidity "disease Y" … All rights reserved. No reuse allowed without permission. perpetuity. preprint (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in The copyright holder for this this version posted August 23, 2020. . https://doi.org/10.1101/2020.08.20.20174169 doi: medRxiv preprint We observed that the comorbidities of hypertension and diabetes have statistically 177 significant associations with the diagnosis of COVID-19 (Table 2) . Moreover, 178 hypertension, diabetes, congestive heart disease, and renal disease show 179 significant association with higher inpatient mortality in COVID-19 patients ( * # is the geometric mean of the odds ratios obtained from the 21 robustness experiments run with different proportions of data as demonstrated in Figure 1 (a) . # is the 95% confidence interval where its limits were calculated using the geometric means of the CI limits of the aforementioned 21 robustness experiments. # POSE: the percentage of significant experiments; that is, the percentage of experiments run over different proportions of data that yielded a significant p-value of 0.05 or smaller. This is a measure of robustness. $ Very small numbers of patients were available for these comorbidities and therefore statistics are not applicable (NA). * # is the geometric mean of the odds ratios obtained from the 21 robustness experiments run with different proportions of data as demonstrated in Figure 1 (a) . # is the 95% confidence interval where its limits were calculated using the geometric means of the CI limits of the aforementioned 21 robustness experiments. # POSE: the percentage of significant experiments; that is, the percentage of experiments run over different proportions of data that yielded a significant p-value of 0.05 or smaller. This is a measure of robustness. $ Very small numbers of patients were available for these comorbidities and therefore statistics are not applicable (NA). 185 Table 4 shows the numbers of medications qualified for testing in each line of 186 analysis given that they were prescribed to more than 10 patients within the relevant 187 time window (Figure 3 ). This table also shows the number of medications, out of all 188 of those qualified for testing, that showed significant and robust association with None M-ST2 other than those that also appeared in M-ST1 Where a BNF section and one of its BNF paragraphs both appear significant, the 202 one that is less specific is omitted from this table for a more concise display. (Table 5) . 208 Biguanides prescribed to diabetic patients after testing positive on COVID-19 were 209 associated with reduced mortality (Table 5) . Interestingly, the more general BNF 210 section of drugs used in diabetes, which includes insulins as well as antidiabetic 211 drugs, showed association with reduced COVID-19 mortality for diabetic patients 212 for whom there is an evidence of prescription before and after a positive COVID-19 213 test (M-LT) ( Table 5) . Finally, low data coverage and bias may cause absence of statistically significant 275 associations with COVID-19 outcomes for some comorbidities or medications. 276 Therefore, no conclusions may be drawn for such cases without further 277 investigation. 278 This study provides an important piece of real-world evidence on associations 280 between co-morbidities and medication prescription, respectively with COVID-19 281 positivity presenting to hospital and inpatient mortality. Identifying these 282 associations can help in the crucial task of defining the vulnerable groups that may 283 benefit from more stringent social distancing especially as lockdown due to COVID-284 19, is relaxed. Nonetheless, observations in this study have to be interpreted with 285 caution due to potential bias and confounders, and confirmatory studies will be 286 required to draw reliable conclusions. 287 This work uses data provided by patients and collected by the NHS as part of their 289 care and support. We believe using patient data is vital to improve health and care 290 for everyone and would, thus, like to thank all those involved for their contribution. All analyses were conducted on data with no personal identifying information. 312 Therefore, informed consent was waived by the ethics committee of the Chelsea & 313 Westminster NHS Foundation Trust, which provided ethical approval for the study. Clinical characteristics of 113 deceased patients with 318 coronavirus disease 2019: retrospective study Comment on "Should COVID-19 Concern 320 Nephrologists? Why and to What Extent? The Emerging Impasse of Angiotensin-converting enzyme inhibitors and 323 angiotensin receptor blockers may be harmful in patients with diabetes 324 during COVID-19 pandemic Should COVID-19 Concern 327 Nephrologists? Why and to What Extent? The Emerging Impasse of Nephron Clinical Practice Coding Algorithms for Defining Comorbidities in ICD-9-CM 330 and ICD-10 Administrative Data Anticoagulant treatment is associated with decreased 332 mortality in severe coronavirus disease 2019 patients with coagulopathy Increase in COVID-19 inpatient survival following 335 detection of Thromboembolic and Cytokine storm risk from the point of 336 admission to hospital by a near real time Traffic-light System Epidemiology of 2019 novel coronavirus in Jiangsu 339 China after wartime control measures: A population-level 340 retrospective study Hypertension, renin-angiotensin-aldosterone 342 system inhibition, and COVID-19 Prevalence of Comorbidities and Its Effects in Patients 347 Infected With SARS-CoV-2: A Systematic Review and Meta-Analysis