Critical evaluation of cohort study results.

Being able to critically evaluate modern cohort studies is important when being presented with claims based on observational evidence. In this review article, key aspects of the cohort design are presented using an example of a cohort study investigating the association between the use of SGLT2 inhibitors and gout. We describe the active comparator, new user design, modern methods used to address confounding, how to identify the most common sources of bias, and how to interpret study results appropriately.

Critical evaluation of articles on case-control studies.

The case-control design is one of the key designs used in observational research. In this review, we discuss common pitfalls of case-control studies and describe how case-control studies can be critically evaluated. We further assert that a well-conducted case-control study provides the same results, precision, and level of evidence as a corresponding cohort study.

Is the sequence ratio an unbiased estimate of the incidence rate ratio? A simulation study.

PURPOSE: We aimed to evaluate the conditions under which the sequence ratio (SR) obtained from a sequence symmetry analysis is an unbiased estimate of the true incidence rate ratio (IRR). METHODS: We simulated cohorts of 1 million individuals who could initiate an exposure drug and experience a very rare, rare, common, or frequent outcome of interest. The outcome rate among exposed individuals was modified by a true incidence rate ratio of 0.2, 0.5, 1.0, 2.0, and 5.0. We further evaluated scenarios where the outcome was fatal and led to immediate censoring or the outcome reduced the rate of initiation of the exposure drug. RESULTS: We found the SR to be close to unbiased for rare, common, and frequent events, except when the true IRR was 5.0 (mean SR 4.94 and 3.74 for common and frequent events). The SR was slightly biased when the outcome was very rare. When the outcome was potentially fatal, the SR was increasingly biased with an increasing probability of death. Likewise, when the outcome reduced the probability of future exposure, the SR was upwards biased. CONCLUSION: The SR is a biased estimate of the incidence rate ratio, when the true IRR is high, the outcome has a high mortality, or when the outcome reduces the probability of future exposure.

Screening of dermatology drugs for aberrant use-patterns: An application of epidemiologic estimates and measures of inequality in drug use.

AIMS: Dermatology treatments require adherence for safe and effective use. Real-world healthcare databases can reveal drug utilization patterns and uncover inappropriate or unexpected use. This study aimed to analyse dermatology drug utilization patterns using epidemiological and inequality measures, leveraging Danish nationwide registries. It also assessed the feasibility of this method for detecting aberrant drug use. METHODS: We formed a 2019 cohort of all patients treated for skin conditions through Danish healthcare registries. We calculated prevalence, incidence rates and treatment duration for dermatological drugs. Inequality in drug utilization was assessed using Lorenz curves, Gini coefficients and other measures. RESULTS: The study encompassed 1 021 255 patients using 94 dermatology drugs. Most usage aligned with 'expected clinical use', but we detected inequality, with some drugs having high Gini coefficients and disproportionate consumption by the top percentile of users. Notable findings included potential inappropriate antibiotic use, excessive topical corticosteroid use and unexpected drug use duration. CONCLUSIONS: In Denmark, dermatology drugs are used primarily as anticipated, with minimal unexpected patterns. Specific follow-up is required to draw conclusions about inappropriate use. This approach demonstrates broad applicability for screening aberrant drug utilization.

A case for screening real-world data for collateral drug benefits: Glucagon-like peptide 1 receptor agonists and bile acid diarrhea.

PURPOSE: Collateral drug benefits are hitherto unknown beneficial effects that might lead to repurposing of already marketed drugs. A randomized controlled trial has found liraglutide to be non-inferior to colesevelam in reducing bile acid diarrhea. We hypothesized that this collateral drug benefit of liraglutide could have been detected using observational data. METHODS: We performed a sequence symmetry analysis (SSA). In the SSA, we indexed individuals on the date of the first prescription of GLP1-RA and restricted the analysis to all individuals who had a first prescription of bile acid sequestrants between 365 days prior to until 365 days after the index date. Sequence ratios (SR), that is, the ratio between counts of persons initiating GLP1-RA first versus last, were calculated, and 95% confidence intervals were obtained. We adjusted for prescribing trends using null-effect SR adjustment. RESULTS: We included 158 individuals, with a median age of 58 years. The trend-adjusted SR was 0.96 (95% confidence interval 0.70-1.31). When stratifying on the type of GLP1-RA (liraglutide or semaglutide), we found results compatible with the previous trial (SRliraglutide 0.75, 0.51-1.10 and SRsemagltuide 1.23, 0.80-1.89). Since BAS also can be used as a cholesterol lowering drug, we repeated the main analysis while excluded statin users, resulting in a stronger association (SR 0.56, 0.33-0.96). CONCLUSION: Using the SSA methodology, we obtained estimates of a collateral drug benefit that were compatible with trial results. These results support the use of epidemiological analyses of observational data as instrument for detecting collateral drug benefits.

Application of the waiting time distribution in a nationwide screening of real-world dermatology drug utilization for aberrant use patterns.

BACKGROUND: Inappropriate use of medicines may have critical consequences from individual, public health, and economic perspectives. Discovering wrongful medicine use may require intentional surveillance or screening. OBJECTIVES: The objectives of this study were to: (i) apply and evaluate the waiting time distribution (WTD) method as a screening tool for identifying aberrant drug use and (ii) evaluate the nationwide use of Dermatology drugs in Denmark for signals of aberrant drug use. METHOD: Dermatology drug use data from the Danish nationwide healthcare registries from 2018 to 2020 were used to produce WTDs that were analyzed for drug use patterns. The method provides estimates of the prevalence and incidence and enables estimation of mean treatment duration, drug relapse, and unexpected drug prescribing. RESULTS: The study included 2 027 889 individual drug users and analyzed 6 141 449 prescriptions. The analysis included approximately 100 dermatology drugs and drug categories and produced 56 WTD drug curves. The WTD patterns and epidemiological estimates confirmed that most drugs are used as intended and revealed few unexpected patterns for further investigation. Three unexpected findings were identified concerning (i) short-term use that would entail suboptimal clinical efficiency for minoxidil, (ii) sub-optimal use of topical tacrolimus, and (iii) potential undesirable increase in short-course doxycycline treatments. CONCLUSION: In Denmark, dermatology drugs are predominantly used as expected, with few unexpected use patterns identified. Targeted specific follow-up on the identified signals is necessary for conclusions about inappropriate use. The findings suggest that the WTD method is applicable for screening for aberrant drug use.

Effectiveness of the quadrivalent live attenuated influenza vaccine against influenza-related hospitalisations and morbidity among children aged 2 to 6 years in Denmark: a nationwide cohort study emulating a target trial.

BACKGROUND: Scant evidence exists on the real-world effectiveness of quadrivalent live attenuated influenza vaccines (LAIV-4) in younger children. We aimed to assess the real-world effectiveness of LAIV-4 against influenza-related hospital contacts and admission and morbidity. METHODS: Using nationwide Danish health-care registries, we designed a cohort study that emulates a target trial, comparing LAIV-4 to no vaccination in children aged 2-6 years. Eligible children vaccinated from Oct 1, 2021, to Jan 15, 2022, were matched to unvaccinated controls in a 1:1 ratio according to demographic characteristics and risk groups for influenza, and followed-up until May 31, 2022. Primary study outcomes any hospital contact for influenza and influenza-related hospital admissions more than 12 h in duration, while hospital admission for respiratory tract infections, or for wheezing or asthma, and antibiotic prescriptions were evaluated as secondary outcomes. We estimated incidence rate ratios (IRRs) and 95% CIs using Poisson regression for each outcome. Vaccine effectiveness was calculated as 1 - IRR. FINDINGS: Among 308 520 Danish children aged 2-6 years, 95 434 vaccinated children were matched with 95 434 unvaccinated children who acted as controls. Receipt of LAIV-4 compared with no vaccination was associated with a reduced IRR of 0·36 (95% CI 0·27 to 0·46) and estimated vaccine effectiveness of 64·3% (53·6 to 72·6) against influenza-related hospital contacts (76 vs 210 events). The corresponding IRR and vaccine effectiveness against influenza-related hospital admissions were 0·63 (0·38 to 1·05) and 36·9% (-5·2 to 62·1; 24 vs 38 events), respectively. LAIV-4 was not associated with reductions in admission rates for respiratory tract infections (IRR 1·14, 95% CI 0·94 to 1·38), wheezing or asthma (1·04, 0·83 to 1·31), or antibiotic prescriptions for respiratory tract infections (0·97, 0·93 to 1·00). Vaccine effectiveness assessed across risk groups for influenza showed similar effectiveness in children with and without coexisting risk factors for severe influenza. INTERPRETATION: LAIV-4 offered moderate protection in younger children against influenza-related hospital contacts during a season dominated by influenza A(H3N2); however vaccination was not associated with reductions in secondary outcomes. This real-world study thereby supports trial evidence of moderate vaccine effectiveness of LAIV-4 against influenza-related outcomes when implementing broad vaccination schedules in younger children. FUNDING: Beckett-Fonden.

Mortality and clinical outcomes following SARS-CoV-2 infection among individuals with haematological malignancies: A Danish population-based cohort study.

OBJECTIVES: We aimed to quantify the risk of death following a positive test for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) among individuals with haematological malignancies, stratified by virus variants and type of malignancy. METHODS: Using the Danish nationwide registries, we conducted a population-based cohort study among individuals who received a discharge diagnosis of haematological malignancies during the 5 years prior to testing positive for SARS-CoV-2 (February 2020-April 2023). All individuals were followed for 30 days after a positive test, and overall and time-stratified case fatality risks (CFR) were estimated. RESULTS: We identified 7154 individuals with a history of haematological malignancies who tested positive for SARS-CoV-2. Among these, we observed 223 deaths, yielding a CFR of 3.1%. The CFR was highest at the beginning of the pandemic (10%) and gradually declined to 1.9% during the period of Omicron BA1/BA2 predominance. The highest CFR was observed among individuals with acute leukaemia (CFR 6.2%, adjusted relative risk 1.95, 95% confidence interval 1.33-2.88) compared to individuals with lymphoma (CFR 3%). CONCLUSIONS: We observed a reduction in the CFR over time, which may be attributed to new treatments, COVID-19 vaccination and the emergence of less aggressive variants.

Comparative effectiveness of heterologous third dose vaccine schedules against severe covid-19 during omicron predominance in Nordic countries: population based cohort analyses.

OBJECTIVE: To investigate the comparative vaccine effectiveness of heterologous booster schedules (ie, three vaccine doses) compared with primary schedules (two vaccine doses) and with homologous mRNA vaccine booster schedules (three vaccine doses) during a period of omicron predominance. DESIGN: Population based cohort analyses. SETTING: Denmark, Finland, Norway, and Sweden, 27 December 2020 to 31 December 2022. PARTICIPANTS: All adults aged ≥18 years who had received at least a primary vaccination schedule of AZD1222 (Oxford-AstraZeneca) or monovalent SARS-CoV-2 wild type (ancestral) strain based mRNA vaccines BNT162b2 (Pfizer-BioNTech) or mRNA-1273 (Moderna), in any combination. MAIN OUTCOME MEASURES: The main outcome measure was country combined risks of covid-19 related hospital admission and death with covid-19 and additional outcomes of covid-19 related admission to an intensive care unit and SARS-CoV-2 infection. During a period of omicron predominance, these outcomes were compared in those who received a heterologous booster versus primary schedule (matched analyses) and versus those who received a homologous mRNA vaccine booster (weighted analyses). Follow-up was for 75 days from day 14 after the booster dose; comparative vaccine effectiveness was calculated as 1-risk ratio. RESULTS: Across the four Nordic countries, 1 086 418 participants had received a heterologous booster schedule of AZD1222+BNT162b2 or mRNA-1273 and 2 505 093 had received a heterologous booster schedule of BNT162b2+mRNA-1273. Compared with the primary schedule only (two doses), the vaccine effectiveness of heterologous booster schedules comprising AZD1222+BNT162b2 or mRNA-1273 and BNT162b2+mRNA-1273 was 82.7% (95% confidence interval 77.1% to 88.2%) and 81.5% (78.9% to 84.2%) for covid-19 related hospital admission and 95.9% (91.6% to 100.0%) and 87.5% (82.5% to 92.6%) for death with covid-19, respectively. Homologous mRNA booster schedules were similarly associated with increased protection against covid-19 related hospital admission (≥76.5%) and death with covid-19 (≥84.1%) compared with previous primary course vaccination only. When a heterologous booster schedule was compared with the homologous booster schedule, vaccine effectiveness was 27.2% (3.7% to 50.6%) for AZD1222+BNT162b2 or mRNA-1273 and 23.3% (15.8% to 30.8%) for BNT162b2+mRNA-1273 schedules against covid-19 related hospital admission and 21.7% (-8.3% to 51.7%) and 18.4% (-15.7% to 52.5%) against death with covid-19, respectively. CONCLUSION: Heterologous booster schedules are associated with increased protection against severe, omicron related covid-19 outcomes compared with primary course schedules and homologous booster schedules.

Sudden Sensorineural Hearing Loss Following Immunization With BNT162b2 or mRNA-1273: A Danish Population-Based Cohort Study.

OBJECTIVE: To compare the occurrence of sudden sensorineural hearing loss following immunization with BNT162b2 (Comirnaty®; Pfizer BioNTech) or mRNA-1273 (Spikevax®; Moderna) to the occurrence among unvaccinated individuals. STUDY DESIGN: Cohort study. SETTING: Nationwide Danish health care registers comprised all Danish residents living in Denmark on October 1, 2020, who were 18 years or older or turned 18 in 2021. METHODS: We compared the occurrence of sudden sensorineural hearing loss following immunization with BNT162b2 (Comirnaty®; Pfizer BioNTech) or mRNA-1273 (Spikevax®; Moderna) (first, second, or third dose) against unvaccinated person time. Secondary outcomes were a first-ever hospital diagnosis of vestibular neuritis and a hearing examination, by an ear-nose-throat (ENT) specialist, followed by a prescription of moderate to high-dose prednisolone. RESULTS: BNT162b2 or mRNA-1273 vaccine was not associated with an increased risk of receiving a discharge diagnosis of sudden sensorineural hearing loss (adjusted hazard ratio [HR]: 0.99, confidence interval [CI]: 0.59-1.64) or vestibular neuritis (adjusted HR: 0.94, CI: 0.69-1.24). We found a slightly increased risk (adjusted HR: 1.40, CI, 1.08-1.81) of initiating moderate to high-dose oral prednisolone following a visit to an ENT specialist within 21 days from receiving a messenger RNA (mRNA)-based Covid-19 vaccine. CONCLUSION: Our findings do not suggest an increased risk of sudden sensorineural hearing loss or vestibular neuritis following mRNA-based COVID-19 vaccination. mRNA-Covid-19 vaccination may be associated with a small excess risk of a visit to an ENT specialist visit followed by a prescription of moderate to high doses of prednisolone.

Cox regression using a calendar time scale was unbiased in simulations of COVID-19 vaccine effectiveness & safety.

BACKGROUND: Observational studies on corona virus disease 2019 (COVID-19) vaccines compare event rates in vaccinated and unvaccinated person time using Poisson or Cox regression. In Cox regression, the chosen time scale needs to account for the time-varying incidence of severe acute respiratory syndrome corona virus 2 (SARS-CoV-2) infection and COVID-19 vaccination.We aimed to quantify bias in person-time based methods, with and without adjustment for calendar time, using simulations and empirical data analysis. METHODS: We simulated 500,000 individuals who were followed for 365 days, and a point exposure resembling COVID-19 vaccination (cumulative incidence 80%). We generated an effectiveness outcome, emulating the incidence of severe acute respiratory syndrome corona virus 2 infection in Denmark during 2021 (risk 10%), and a safety outcome with seasonal variation (myocarditis, risk 1/5,000). Incidence rate ratios (IRRs) were set to 0.1 for effectiveness and 5.0 for safety outcomes. IRRs and hazard ratios (HRs) were estimated using Poisson and Cox regression with a time under observation scale, and a calendar time scale. Bias was defined as estimated IRR or HR-true IRR. Further, we obtained estimates for both outcomes using data from the Danish health registries. RESULTS: Unadjusted IRRs (biaseffectivenes +0.16; biassafety -2.09) and HRs estimated using a time-under-observation scale (+0.28;-2.15) were biased. Adjustment for calendar time reduced bias in Cox (+0.03; +0.33) and Poisson regression (0.00; -0.28). Cox regression using a calendar time scale was least biased (0.00, +0.12). When analyzing empirical data, adjusted Poisson and Cox regression using a calendar time scale yielded estimates in accordance with existing evidence. CONCLUSION: Lack of adjustment for the time-varying incidence of COVID-19 related outcomes may severely bias estimates.

Identifying diabetogenic drugs using real world health care databases: A Danish and Australian symmetry analysis.

AIMS: Drug-induced diabetes is underreported in conventional drug safety monitoring and may contribute to the increasing incidence of type 2 diabetes. Therefore, we used routinely collected prescription data to screen all commonly used drugs for diabetogenic effects. METHODS: Leveraging the Danish nationwide health registries, we used a case-only symmetry analysis design to evaluate all possible associations between drug initiation and subsequent diabetes. The study was conducted among individuals aged ≥40 years with a first-ever prescription for any antidiabetic drug 1996-2018 (n = 348 996). Sequence ratios (SRs) and 95% confidence intervals (CIs) were obtained for all possible drug class-diabetes combinations. A lower bound of the 95% CI >1.00 was considered a signal. Signals generated in Denmark were replicated using the Services Australia, Pharmaceutical Benefits Scheme 10% data extract. RESULTS: Overall, 386 drug classes were investigated, of which 70 generated a signal. In total, 43 were classified as previously known based on the SIDER database or a literature review, for example, glucocorticoids (SR 1.67, 95% CI 1.62-1.72) and ß-blockers (SR 1.20, 95% CI 1.16-1.23). Of 27 new signals, three drug classes yielded a signal in both the Danish and Australian data source: digitalis glycosides (SR 2.15, 95% CI 2.04-2.27, and SR 1.76, 95% CI 1.50-2.08), macrolides (SR 1.20, 95% CI 1.16-1.24, and SR 1.11, 95% CI 1.06-1.16) and inhaled ß2-agonists combined with glucocorticoids (SR 1.35, 95% CI 1.28-1.42, and SR 1.14, 95% CI 1.06-1.22). CONCLUSION: We identified 70 drug-diabetes associations, of which 27 were classified as hitherto unknown. Further studies evaluating the hypotheses generated by this work are needed, particularly for the signal for digitalis glycosides.

Changing characteristics over time of individuals receiving COVID-19 vaccines in Denmark: A population-based descriptive study of vaccine uptake.

AIMS: The Danish authorities implemented a differential rollout of the COVID-19 vaccines where individuals at high risk of COVID-19 were prioritized. We describe the temporal uptake and characteristics of COVID-19 vaccine recipients in Denmark. METHODS: Using nationwide healthcare registries, we identified all Danish residents ⩾5 years of age who received at least one dose of a COVID-19 vaccine from 27 December 2020-29 January 2022. We charted the daily number of newly vaccinated individuals and the cumulative vaccine coverage over time, stratified by vaccine type, age groups and vaccination priority groups, and described characteristics of vaccine recipients during two-month-intervals and in vaccination priority groups. RESULTS: By 29 January 2022, 88%, 86% and 64% of Danish residents ⩾5 years (n=5,562,008) had received a first, second and third dose, respectively, of a COVID-19 vaccine, most commonly the BNT162b2 vaccine (84%). Uptake ranged from 48% in 5-11-year-olds to 98% in 65-74-year-olds. Individuals vaccinated before June 2021 were older (median age 61-70 years vs 10-35 years in later periods) and had more comorbidities such as hypertension (22-28% vs 0.77-2.8% in later periods), chronic lung disease (9.4-15% vs 3.7-4.6% in later periods) and diabetes (9.3-12% vs 0.91-2.4% in later periods). CONCLUSIONS: We document substantial changes over time in, for example, age, sex and medical history of COVID-19 vaccine recipients. Though these results are related to the differential vaccine rollout in Denmark, similar findings are probable in other countries and should be considered when designing and interpreting studies on the effectiveness and safety of COVID-19 vaccines.

Development and Validation of a Nordic Multimorbidity Index Based on Hospital Diagnoses and Filled Prescriptions.

Purpose: To develop the Nordic Multimorbidity Index (NMI), a multimorbidity measure specifically suited to the Nordic health and administrative registry data based on current diagnosis, treatment, and coding practices. Methods: The NMI was developed to predict 5-year mortality in a population-based cohort of randomly sampled Danish residents aged ≥40 years (n = 425,087) followed from 2013 to 2018. Included predictors were selected from hospital diagnoses and filled drug prescriptions based on a combination of subject matter knowledge and a data-driven approach using backwards elimination. The performance of the NMI was assessed in a temporal validation cohort of Danish residents followed from 2007 to 2012 and in six cohorts of new users of selected drugs. The discriminative performance of the NMI, Charlson Comorbidity Index (CCI) and the Elixhauser Comorbidity Index (ECI) was assessed using the c-statistic from logistic regression models with 5-year mortality as dependent variable and the multimorbidity index score, age, and sex as independent variables. Results: The NMI included 50 predictors. In the temporal validation cohort, the c-statistic of the NMI (0.887, 95% CI 0.883-0.890) exceeded that of the CCI (0.871, 95% CI 0.868-0.874) and ECI (0.866, 95% CI 0.863-0.870). In all new user cohorts, the NMI outperformed the other indices with c-statistics ranging from 0.781 (95% CI 0.779-0.784) to 0.838 (95% CI 0.834-0.842). Conclusion: The NMI predicted 5-year mortality in a general Danish population and six cohorts of new users of selected drugs and was superior to the CCI and ECI. The NMI could be preferred over these indices to quantify the level of multimorbidity for, eg, descriptive purposes or confounding control. The NMI should be validated in other patient populations and other Nordic countries.

Risk of adverse events after covid-19 in Danish children and adolescents and effectiveness of BNT162b2 in adolescents: cohort study.

OBJECTIVES: To assess the risk of acute and post-acute adverse events after SARS-CoV-2 infection in children and adolescents in Denmark and to evaluate the real world effectiveness of the BNT162b2 mRNA vaccine (Pfizer-BioNTech) among adolescents. DESIGN: Cohort study. SETTING: Nationwide Danish healthcare registers. PARTICIPANTS: All Danish people younger than 18 years who were either tested for SARS-CoV-2 using reverse transcriptase polymerase chain reaction (RT-PCR) or vaccinated with BNT162b2 to 1 October 2021. MAIN OUTCOME MEASURES: Risk of hospital admissions (any hospital contact of ≥12 hours); intensive care unit (ICU) admissions; serious complications, including multisystem inflammatory syndrome in children (MIS-C), myocarditis, and neuroimmune disorders; and initiating drug treatment and health service use up to six months after being tested. Vaccine effectiveness in vaccine recipients compared with unvaccinated peers was evaluated as one minus the risk ratio at 20 days after the first dose and 60 days after the second dose. RESULTS: Of 991 682 children and adolescents tested for SARS-CoV-2 using RT-PCR in Denmark, 74 611 (7.5%) were positive. The risk of hospital admission with any variant for ≥12 hours was 0.49% (95% confidence interval 0.44% to 0.54%; 361/74 350), and 0.01% (0.01% to 0.03%; 10/73 187) of participants were admitted to an ICU within 30 days of testing positive. The risk of MIS-C within two months of SARS-CoV-2 infection was 0.05% (0.03% to 0.06%; 32/70 666), whereas no participants had myocarditis outside of MIS-C or encephalitis and fewer than five had Guillain-Barré syndrome. In the post-acute phase (1-6 months after infection), participants who tested positive for SARS-CoV-2 showed a 1.08-fold (95% confidence interval 1.06-fold to 1.10-fold) increase in rate of contacts with general practitioners compared with a reference cohort sampled among all children tested for SARS-CoV-2 during the study period. Overall, 278 649 adolescents received BNT162b2. Compared with unvaccinated adolescents, the estimated vaccine effectiveness among 229 799 adolescents vaccinated with one dose was 62% (95% confidence interval 59% to 65%) after 20 days, and among 175 176 vaccinated with two doses was 93% (92% to 94%) after 60 days during a period when delta was the dominant variant. CONCLUSIONS: The absolute risks of adverse events after SARS-CoV-2 infection were generally low in Danish children and adolescents, although MIS-C occurred in 0.05% (32/70 666) of participants with RT-PCR confirmed SARS-CoV-2 infection. In adjusted analyses, rates of general practitioner visits were slightly increased in SARS-CoV-2 positive children and adolescents, which could indicate persisting symptoms. BNT162b2 appeared to be effective in reducing the risk of SARS-CoV-2 infection with the delta variant in adolescents.