Self-controlled case series studies: Just how rare does a rare non-recurrent outcome need to be?

Overcoming time-varying confounding in self-controlled case series with active comparators: application and recommendations

Confounding by indication is a key challenge for pharmacoepidemiologists. Although self-controlled study designs address time-invariant confounding, indications sometimes vary over time. For example, infection might act as a time-varying confounder in a study of antibiotics and uveitis, because it is time-limited and a direct cause of both receipt of antibiotics and uveitis. Methods for incorporating active comparators in self-controlled studies to address such time-varying confounding by indication have only recently been developed. In this paper, we formalize these methods and provide a detailed description for how the active comparator rate ratio can be derived in a self-controlled case series: either by explicitly comparing the regression coefficients for a drug of interest and an active comparator under certain circumstances using a simple ratio approach or through the use of a nested regression model. The approaches are compared in 2 case studies, one examining the association between thiazolidinedione use and fractures and one examining the association between fluoroquinolone use and uveitis, using the United Kingdom's Clinical Practice Research Datalink. Finally, we provide recommendations for the use of these methods, which we hope will support the design, execution, and interpretation of self-controlled case series using active comparators and thereby increase the robustness of pharmacoepidemiologic studies. This article is part of a Special Collection on Pharmacoepidemiology.

Core Concepts: Self-Controlled Designs in Pharmacoepidemiology

One of the key challenges in pharmacoepidemiological studies is that of uncontrolled confounding, which occurs when confounders are poorly measured, unmeasured or unknown. Self-controlled designs can help address this issue, as their key comparison is not between people, but periods of time within the same person. This controls for all time-stable confounders (genetics) and in the absence of time-varying confounding negates the need for an external control group. However, these benefits come at the cost of strong assumptions, not all of which are verifiable. This review briefly introduces the reader to different types of self-controlled study designs, their terminology and highlights key publications through an annotated reference list. We include a practical description of how these designs can be implemented and visualised using recent examples, and finish by discussing recent developments. We hope this review will serve as a starting point for researchers looking to apply self-controlled designs in their own work.

Two assumptions of the prior event rate ratio approach for controlling confounding can be evaluated by self-controlled case series and dynamic random intercept modeling

OBJECTIVES

The prior event rate ratio (PERR) is a recently developed approach for controlling confounding by measured and unmeasured covariates in real-world evidence research and observational studies. Despite its rising popularity in studies of safety and effectiveness of biopharmaceutical products, there is no guidance on how to empirically evaluate its model assumptions. We propose two methods to evaluate two of the assumptions required by the PERR, specifically, the assumptions that occurrence of outcome events does not alter the likelihood of receiving treatment, and that earlier event rate does not affect later event rate.

STUDY DESIGN AND SETTING

We propose using self-controlled case series (SCCS) and dynamic random intercept modeling (DRIM), respectively, to evaluate the two aforementioned assumptions. A nonmathematical introduction of the methods and their application to evaluate the assumptions are provided. We illustrate the evaluation with secondary analysis of deidentified data on pneumococcal vaccination and clinical pneumonia in The Gambia, West Africa.

RESULTS

SCCS analysis of data on 12,901 vaccinated Gambian infants did not reject the assumption of clinical pneumonia episodes had no influence on the likelihood of pneumococcal vaccination. DRIM analysis of 14,325 infants with a total of 1719 episodes of clinical pneumonia did not reject the assumption of earlier episodes of clinical pneumonia had no influence on later incidence of the disease.

CONCLUSION

The SCCS and DRIM methods can facilitate appropriate use of the PERR approach to control confounding.

Importance of Examining Incidentality in Vaccine Safety Assessment

The author believes that the principles of statistical methods for vaccine safety can be divided into three categories: comparison of adverse event incidence rates between vaccinated and unvaccinated groups, analysis of incidentality in the vaccinated group, and a combination of both. The first category includes the cohort study; the second, the self-controlled risk interval design (SCRI); and the third, the self-controlled case series method. A single p-value alone should not determine a scientific conclusion, and analysis should be performed using multiple statistical methods with different principles. The author believes that using both the cohort study and the SCRI for analysis is the best method to assess vaccine safety. When the cohort study may not detect a significant difference owing to a low incidence rate of an adverse event in the vaccinated group or a high one in the unvaccinated group, the SCRI may detect it. Because vaccines must have a higher level of safety than the pharmaceuticals used for treatment, vaccine safety is advisable to be assessed using methods that can detect a significant difference even for any value of the incidence rate of an adverse event. The author believes that the analyses of COVID-19 vaccine safety have areas for improvement because the proportion of papers that used the cohort study and the SCRI was negligible.

Estimation and reduction of bias in self-controlled case series with non-rare event dependent outcomes and heterogeneous populations

The self-controlled case series (SCCS) is a commonly adopted study design in the assessment of vaccine and drug safety. Recurrent event data collected from SCCS studies are typically analyzed using the conditional Poisson model which assumes event times are independent within-cases. This assumption is violated in the presence of event dependence, where the occurrence of an event influences the probability and timing of subsequent events. When event dependence is suspected in an SCCS study, the standard recommendation is to include only the first event from each case in the analysis. However, first event analysis can still yield biased estimates of the exposure relative incidence if the outcome event is not rare. We first demonstrate that the bias in first event analysis can be even higher than previously assumed when subpopulations with different baseline incidence rates are present and describe an improved method for estimating this bias. Subsequently, we propose a novel partitioned analysis method and demonstrate how it can reduce this bias. We provide a recommendation to guide the number of partitions to use with the partitioned analysis, illustrate this recommendation with an example SCCS study of the association between beta-blockers and acute myocardial infarction, and compare the partitioned analysis against other SCCS analysis methods by simulation.

Cholinesterase inhibitors and non-steroidal anti-inflammatory drugs and the risk of peptic ulcers: A self-controlled study

BACKGROUND

Non-steroidal anti-inflammatory drugs (NSAIDs) should be used with caution in adults aged 65 years and older. Their gastrointestinal adverse event risk might be further reinforced when using concomitant cholinesterase inhibitors (ChEIs). We aimed to investigate the association between NSAIDs and ChEI use and the risk of peptic ulcers in adults aged 65 years and older.

METHODS

Register-based self-controlled case series study including adults ≥65 years with a new prescription of ChEIs and NSAIDs, diagnosed with incident peptic ulcer in Sweden, 2007-2020. We identified persons from the Total Population Register individually linked to several nationwide registers. We estimated the incidence rate ratio (IRR) of peptic ulcer with a conditional Poisson regression model for four mutually exclusive risk periods: use of ChEIs, NSAIDs, and the combination of ChEIs and NSAIDs, compared with the non-treatment in the same individual. Risk periods were identified based on the prescribed daily dose, extracted via a text-parsing algorithm, and a 30-day grace period.

RESULTS

Of 70,060 individuals initiating both ChEIs and NSAIDs, we identified 1500 persons with peptic ulcer (median age at peptic ulcer 80 years), of whom 58% were females. Compared with the non-treatment periods, the risk of peptic ulcer substantially increased for the combination of ChEIs and NSAIDs (IRR: 9.0, [6.8-11.8]), more than for NSAIDs alone (5.2, [4.4-6.0]). No increased risks were found for the use of ChEIs alone (1.0, [0.9-1.2]).

DISCUSSION

We found that the risk of peptic ulcer associated with the concomitant use of NSAIDs and ChEIs was over and beyond the risk associated with NSAIDs alone. Our results underscore the importance of carefully considering the risk of peptic ulcers when co-prescribing NSAIDs and ChEIs to adults aged 65 years and older.

Comparison design and evaluation power in cohort and self-controlled case series designs for post-authorization vaccine safety studies

Background

Post-authorization safety studies (PASSs) of vaccines are important. PASSs enable the evaluation of association between vaccination and adverse events following immunization through common study designs. Clinical trials during vaccine development typically include a few thousand to 10,000 participants while a PASS might aim to detect a few adverse events per 100,000 vaccine recipients. While all available data may be utilized, prior consideration of power analyses are nonetheless crucial for interpretation in cases where statistically significant differences are not found.

Methods

This research primarily examined cohort study design and self-controlled case series (SCCS) design, estimating the power of a PASS under plausible conditions.

Results

Both the cohort study and SCCS designs necessitated large sample sizes or high event counts to guarantee adequate power. The SCCS design is particularly suited to evaluating rare adverse events. However, extremely rare events may not yield sufficient occurrences, thereby resulting in low power. Although the SCCS design can more efficiently control for time-invariant confounding in principle, it solely estimates relative measures. A cohort study design might be preferred if confounding can be adequately managed as it also estimates absolute measures. It may be an easy decision to use all the data at hand for either design. We found it necessary to estimate the sample size and number of events to be used in the study based on a priori information and anticipated results.

Nationwide safety surveillance of COVID-19 mRNA vaccines following primary series and first booster vaccination in Singapore

Background

The real-world safety profile of COVID-19 mRNA vaccines remains incompletely elucidated.

Methods

We performed a nationwide post-market safety surveillance analysis in Singapore, on vacinees aged 5 years and older, through mid-September 2022. Observed-over-expected (O/E) analyses were performed to identify potential safety signals among eight shortlisted adverse events of special interest (AESIs): strokes, cerebral venous thrombosis (CVT), acute myocardial infarction, myocarditis/pericarditis, pulmonary embolism, immune thrombocytopenia, convulsions and appendicitis. Self-controlled case series analyses (SCCS) were performed to validate signals of concern, occurring within 42 days of vaccination.

Findings

Elevated risks were observed on O/E analyses for the following AESIs: myocarditis/pericarditis, [rate ratio (RR): 3.66, 95 % confidence interval (95 % CI): 2.71 to 4.94], appendicitis [RR: 1.14 (1.02 to 1.27)] and CVT [RR: 2.11 (1.18 to 3.77)]. SCCS analyses generated corroborative findings: myocarditis/pericarditis, [relative incidence (RI): 6.96 (3.95 to 12.27) at 1 to 7 days post-dose 2], CVT [RI: 4.30 (1.30 to 14.20) at 22 to 42 days post-dose 1] and appendicitis [RI: 1.31 (1.03 to 1.67) at 1 to 7 days post-dose 1]. Booster dose 1 continued to be associated with higher rates of myocarditis/pericarditis on O/E analysis [RR: 2.30, (1.39 to 3.80) and 1.69, (1.11 to 2.59)] at 21- and 42-days post-booster dose 1, respectively. Males aged 12 to 17 exhibited highest risks of both myocarditis/pericarditis [RI: 6.31 (1.36 to 29.3)] and appendicitis [RI: 2.01 (1.12 to 3.64)] after primary vaccination. Similarly, CVT was also predominantly observed in males aged above 50 (11 out of 16 cases), within 42-days of vaccination.

Interpretation

Our data suggest that myocarditis/pericarditis, appendicitis and CVT are associated with primary vaccination using COVID-19 mRNA vaccines. Males at specific ages exhibit higher risks for all three AEs identified. The risk of myocarditis/pericarditis continues to be elevated after booster dose 1.

Hypoglycemia risk with inappropriate dosing of glucose-lowering drugs in patients with chronic kidney disease: a retrospective cohort study

The incidence rates and consequences of inappropriate dosing of glucose-lowering drugs remain limited in patients with chronic kidney disease (CKD). A retrospective cohort study was conducted to estimate the frequency of inappropriate dosing of glucose-lowering drugs and to evaluate the subsequent risk of hypoglycemia in outpatients with an estimated glomerular filtration rate (eGFR) of < 50 mL/min/1.73 m². Outpatient visits were divided according to whether the prescription of glucose-lowering drugs included dose adjustment according to eGFR or not. A total of 89,628 outpatient visits were included, 29.3% of which received inappropriate dosing. The incidence rates of the composite of all hypoglycemia were 76.71 and 48.51 events per 10,000 person-months in the inappropriate dosing group and in appropriate dosing group, respectively. After multivariate adjustment, inappropriate dosing was found to lead to an increased risk of composite of all hypoglycemia (hazard ratio 1.52, 95% confidence interval 1.34, 1.73). In the subgroup analysis, there were no significant changes in the risk of hypoglycemia regardless of renal function (eGFR < 30 vs. 30-50 mL/min/1.73 m²). In conclusion, inappropriate dosing of glucose-lowering drugs in patients with CKD is common and associated with a higher risk of hypoglycemia.

Population-based screening to detect benzodiazepine drug-drug-drug interaction signals associated with unintentional traumatic injury

Drug interactions involving benzodiazepines and related drugs (BZDs) are increasingly recognized as a contributor to increased risk of unintentional traumatic injury. Yet, it remains unknown to what extent drug interaction triads (3DIs) may amplify BZDs' inherent injury risk. We identified BZD 3DI signals associated with increased injury rates by conducting high-throughput pharmacoepidemiologic screening of 2000-2019 Optum's health insurance data. Using self-controlled case series design, we included patients aged ≥ 16 years with an injury while using a BZD + co-dispensed medication (i.e., base pair). During base pair-exposed observation time, we identified other co-dispensed medications as candidate interacting precipitants. Within each patient, we compared injury rates during time exposed to the drug triad versus to the base pair only using conditional Poisson regression, adjusting for time-varying covariates. We calculated rate ratios (RRs) with 95% confidence intervals (CIs) and accounted for multiple estimation via semi-Bayes shrinkage. Among the 65,123 BZD triads examined, 79 (0.1%) were associated with increased injury rates and considered 3DI signals. Adjusted RRs for signals ranged from 3.01 (95% CI = 1.53-5.94) for clonazepam + atorvastatin with cefuroxime to 1.42 (95% CI = 1.00-2.02, p = 0.049) for alprazolam + hydrocodone with tizanidine. These signals may help researchers prioritize future etiologic studies to investigate higher-order BZD interactions.

A modified self-controlled case series method for event-dependent exposures and high event-related mortality, with application to COVID-19 vaccine safety

We propose a modified self‐controlled case series (SCCS) method to handle both event‐dependent exposures and high event‐related mortality. This development is motivated by an epidemiological study undertaken in France to quantify potential risks of cardiovascular events associated with COVID‐19 vaccines. Event‐dependence of vaccinations, and high event‐related mortality, are likely to arise in other SCCS studies of COVID‐19 vaccine safety. Using this case study and simulations to broaden its scope, we explore these features and the biases they may generate, implement the modified SCCS model, illustrate some of the properties of this model, and develop a new test for presence of a dose effect. The model we propose has wider application, notably when the event of interest is death.

The Case Time Series Design

Modern data linkage and technologies provide a way to reconstruct detailed longitudinal profiles of health outcomes and predictors at the individual or small-area level. Although these rich data resources offer the possibility to address epidemiologic questions that could not be feasibly examined using traditional studies, they require innovative analytical approaches. Here we present a new study design, called case time series, for epidemiologic investigations of transient health risks associated with time-varying exposures. This design combines a longitudinal structure and flexible control of time-varying confounders, typical of aggregated time series, with individual-level analysis and control-by-design of time-invariant between-subject differences, typical of self-matched methods such as case-crossover and self-controlled case series. The modeling framework is highly adaptable to various outcome and exposure definitions, and it is based on efficient estimation and computational methods that make it suitable for the analysis of highly informative longitudinal data resources. We assess the methodology in a simulation study that demonstrates its validity under defined assumptions in a wide range of data settings. We then illustrate the design in real-data examples: a first case study replicates an analysis on influenza infections and the risk of myocardial infarction using linked clinical datasets, while a second case study assesses the association between environmental exposures and respiratory symptoms using real-time measurements from a smartphone study. The case time series design represents a general and flexible tool, applicable in different epidemiologic areas for investigating transient associations with environmental factors, clinical conditions, or medications.

Diurnal Temperature Ranges in Relation to Lower Limb Amputation Rate of Diabetic Foot in South Korea: A Population Based Nationwide Study

The evidence for the association between diurnal temperature range (DTR) and diabetic foot amputations is limited. We aimed to investigate the region-specific association between DTR and the amputation rate of diabetic foot in Korean national-wide data. Daily data on DTR and the rate of diabetic foot amputations from 16 provincial capital cities in Korea were obtained (2011-2018). In this study, the latitude ranged from 33°11' N to 38°61' N, and we classified each region according to latitude. Region 1, which was located at a relatively high latitude, included Seoul, Incheon, Gyeonggi-do, and Gangwon-do. Region 2, which was located at a relatively low latitude, included Busan, Ulsan, Gyeonsannam-do, Gwangju, Jeollanam-do, Jeollabuk-do, and Jeju-do. The region-specific DTR effects on the amputation rate were estimated based on a quasi-Poisson generalized linear model, combined with a distributed lag non-linear model based on the self-controlled case series design. The DTR impacts were generally limited to a period of nine days, while significant effects during lag days 7-14 were only found in the cities of Seoul, Incheon, and Gyeonggi-do (10th lag day: RR [95% CI]; Seoul: 1.015, [1.001-1.029]; Incheon: 1.052 [1.006-1.101]; Gyeonggi-do: 1.018 [1.002-1.034]). In the subgroup analysis (according to the latitude), an increase of 1 °C in DTR was associated with the risk of diabetic foot in relatively high latitude regions. DTR has considerable effects on the risk of diabetic foot amputation in various provinces in Korea, and it was particularly affected by latitude. The results can inform the decisions on developing programs to protect vulnerable subpopulations from adverse impacts.

Association Between Third-Generation Fluoroquinolones and Achilles Tendon Rupture: A Self-Controlled Case Series Analysis

PURPOSE

We investigated whether use of third-generation fluoroquinolones is associated with occurrence of Achilles tendon rupture using a case series analysis in which patients served as their own control.

METHODS

We used administrative claims data to identify residents in a single Japanese prefecture who were enrolled in National Health Insurance and Elderly Health Insurance from April 2012 to March 2017 and experienced Achilles tendon rupture after receiving an antibiotic prescription. Antibiotics were categorized into 3 groups: first- and second-generation fluoroquinolones, third-generation fluoroquinolones, and nonfluoroquinolones. We used a conditional Poisson regression model to estimate the incidence rate ratio (IRR) of Achilles tendon rupture during the antibiotic exposure period relative to the nonexposure period for each patient. An exposure period was defined as 30 days from start of the prescription.

RESULTS

Analyses were based on 504 patients with Achilles tendon rupture who had received antibiotic prescriptions. Risk of rupture was not significantly elevated during exposure to third-generation fluoroquinolones (IRR = 1.05; 95% CI, 0.33-3.37) and nonfluoroquinolones (IRR = 1.08; 95% CI, 0.80-1.47). In contrast, risk was significantly elevated during exposure to first- and second-generation fluoroquinolones (IRR = 2.94; 95% CI, 1.90-4.54). Findings were similar across subgroups stratified by sex and by recent corticosteroid use.

CONCLUSIONS

Our analysis showed that third-generation fluoroquinolone use was not associated with an increased risk of Achilles tendon rupture. These antibiotics may be a safer option for patients in whom this risk is elevated, such as athletes.

The science of vaccine safety: Summary of meeting at Wellcome Trust

Vaccines are everywhere hugely successful but are also under attack. The reason for the latter is the perception by some people that vaccines are unsafe. However that may be, vaccine safety, life any other scientific subject, must be constantly studied. It was from this point of view that a meeting was organized at the Wellcome Trust in London in May 2019 to assess some aspects of vaccine safety as subjects for scientific study. The objective of the meeting was to assess what is known beyond reasonable doubt and conversely what areas need additional studies. Although the meeting could not cover all aspects of vaccine safety science, many of the most important issues were addressed by a group of about 30 experts to determine what is already known and what additional studies are merited to assess the safety of the vaccines currently in use. The meeting began with reviews of the current situation in different parts of the world, followed by reviews of specific controversial areas, including the incidence of certain conditions after vaccination and the safety of certain vaccine components. Lastly, information about the human papillomavirus vaccine was considered because its safety has been particularly challenged by vaccine opponents. The following is a summary of the meeting findings. In addition to this summary, the meeting organizers will explore opportunities to perform studies that would enlarge knowledge of vaccine safety.