Methods for constructing treatment episodes and impact on exposure-outcome associations

Differential use of antidiabetic medication related to income, cohabitation and area of residence: a Swedish nationwide cohort study

BACKGROUND

Poor adherence to antidiabetic medication in individuals with type 2 diabetes (T2D) may lead to increased risk of morbidity and mortality. Socioeconomic and demographic factors associated with non-adherence have been mainly identified via cross-sectional studies. To investigate the association between antidiabetic medication adherence and income, cohabitation and area of residence.

METHODS

Register-based cohort study of individuals with T2D living in Sweden and initiating antidiabetic treatment between 2006 and 2022. Confounding adjustment and competing events were accounted for via inverse probability of treatment and censoring weighting. Exposures were disposable income, social income support, cohabitation status and area of residence. Outcomes were antidiabetic medication adherence during the first year from treatment initiation and treatment interruption at 12 and 60 months.

RESULTS

This study included 594 918 individuals with T2D. Low disposable income (adjusted OR: 1.18, 95% CI: (1.14 to 1.21)), social income support (1.09, (1.05 to 1.14)), living in large cities (1.28, (1.24 to 1.31)) and cohabitation (1.09, (1.06 to 1.11)) were associated with non-adherence (proportion of days covered ≤0.2), as compared with high adherence (proportion of days covered >0.8). Consistently, treatment interruption was associated with social income support (relative risk ratio at 12 months: 1.10, (1.06 to 1.14) and at 60 months 1.02 (1.00 to 1.05)), living in large cities (1.13, (1.12 to 1.14); 1.08, (1.07 to 1.08)) and low income (1.05, (1.03 to 1.07); 1.01, (1.00 to 1.02)).

CONCLUSIONS

Low income levels, cohabitation and living in large cities were associated with non-adherence to antidiabetic medication and risk of treatment interruption. The results highlight the need for targeted interventions aiming at improving adherence to treatments both at patient and healthcare system levels.

Treatment patterns in essential tremor: Real-world evidence from a United Kingdom and France primary care database

BACKGROUND AND PURPOSE

Essential tremor (ET) is one of the most common neurological disorders, but information on treatment pattern is still scant. The aim of this study was to describe the demographic and clinical characteristics, treatment patterns, and determinants of drug use in patients with newly diagnosed ET in France and the United Kingdom.

METHODS

Incident cases of ET diagnosed between January 1, 2015 and December 31, 2018 with 2 years of follow-up were identified by using The Health Improvement Network (THIN®) general practice database. During the follow-up, we assessed the daily prevalence of use and potential switches from first-line to second-line treatment or other lines of treatment. Logistic regression models were conducted to assess the effect of demographic and clinical characteristics on the likelihood of receiving ET treatment.

RESULTS

A total of 2957 and 3249 patients were selected in the United Kingdom and France, respectively. Among ET patients, drug use increased from 12 months to 1 month prior the date of index diagnosis (ID). After ID, nearly 40% of patients received at least one ET treatment, but during follow-up drug use decreased and at the end of the follow-up approximately 20% of patients were still on treatment. Among treated patients, ≤10% maintained the same treatment throughout the entire follow-up, nearly 20% switched, and 40%-75% interrupted any treatment. Results from the multivariate analysis revealed that, both in France and the United Kingdom, patients receiving multiple concomitant therapies and affected by psychiatric conditions were more likely to receive an ET medication.

CONCLUSION

This study shows that ET is an undertreated disease with a lower-than-expected number of patients receiving and maintaining pharmacological treatment. Misclassification of ET diagnosis should be acknowledged; thus, results require cautious interpretation.

Association Between Selective Serotonin Reuptake Inhibitors Prevalent Use and COVID-19-Related Mortality: A Retrospective Cohort Study

PURPOSE/BACKGROUND

Since the emergence of the coronavirus disease 2019 (COVID-19), many efforts have been made to prevent and to treat the disease. In this line, the anti-inflammatory effect of selective serotonin reuptake inhibitors (SSRI) as alternatives for treating chronic inflammatory diseases has been studied. There is previous evidence of the usefulness of these drugs for reducing COVID-19 impact.

METHODS/PROCEDURES

We conducted a retrospective single-center cohort study of adult patients with a positive reverse transcriptase-polymerase chain reaction for COVID-19, evaluating the association between SSRI use and in-hospital mortality.

FINDINGS/RESULTS

Of 1689 included patients, 182 (10.8%) were exposed to SSRI. A total of 291 patients died during the hospitalization, representing an in-hospital mortality of 17.2% (95% confidence interval [CI], 15.4%-19.0%): 44 (24.2%) of the exposed to SSRIs versus 247 (16.4%) of those not exposed to SSRIs (crude odds ratio [OR], 1.62; 95% CI, 1.12-2.34; P = 0.009). No independent effect of SSRIs on in-hospital mortality was found when applying either the inverse probability of treatment weighting (OR, 1.15; 95% CI, 0.71-1.89; P = 0.56) or with conventional multivariable analysis 0.81 (95 % CI: 0.28-2.31, P = 0.69).

IMPLICATIONS/CONCLUSIONS

In the present retrospective study of patients hospitalized for COVID-19, prior use of SSRIs did not reduce mortality.

Examining the Association Between Serotonergic Antidepressants and Blood Transfusion Requirements in Orthopaedic Surgery: A Comprehensive Analysis

AIM

The study aimed to investigate the association between the use of serotonergic antidepressants and blood transfusion requirements in orthopedic surgical patients.

METHODOLOGY

A retrospective follow-up study was conducted at Kota Trauma Hospital, Kota, Rajasthan focusing on patients who underwent orthopedic surgeries between November 2021 through December 2022. Patients were categorized into two groups: users of serotonergic antidepressants and users of other antidepressants (non-selective serotonergic antidepressants). The requirement for blood transfusion for both groups was assessed. Covariate factors, such as medication use and comorbidities (e.g., diabetes mellitus, cardiovascular diseases), were examined to control potential confounding variables.

RESULT

A total of 170 subjects with complete medical records were included in the study. The results presented a significant association between the usage of serotonergic antidepressants and blood transfusion requirements (p=0.001). While no significant differences between the two groups were observed in perioperative hemoglobin levels and fluid infusion, there was a significant difference in blood loss and postoperative drainage.

CONCLUSION

Serotonergic antidepressant usage was correlated with increased blood transfusion requirements in orthopedic surgery patients on antidepressants. The study underscores the importance of considering medication factors in perioperative management and highlights potential implications for patient care strategies.

Prenatal Antidepressant Exposure and the Risk of Attention-Deficit/Hyperactivity Disorder in Childhood: Accounting for Misclassification of Exposure

BACKGROUND

Many studies of prenatal antidepressant exposure and the risk of attention-deficit/hyperactivity disorder (ADHD) have done little to reduce bias from exposure misclassification. We assessed the prenatal antidepressant-ADHD effect by incorporating information on repeatedly redeemed prescriptions and redemptions of drug classes commonly used in pregnancy in the analyses to reduce bias from exposure misclassification.

METHODS

Using population-based registries, we conducted a nationwide cohort study of all children born in Denmark from 1997 to 2017. In a former-user analysis, we compared children prenatally exposed, defined by a redeemed prescription by the mother during pregnancy, to a comparison cohort consisting of prenatally unexposed children whose mothers had redeemed a prescription before pregnancy. We incorporated information on repeatedly redeemed prescriptions and redemptions of drug classes commonly used in pregnancy in the analyses to reduce bias from exposure misclassification. We used incidence rate ratios (IRRs) and incidence rate differences (IRDs) as effect measures.

RESULTS

The cohort included 1,253,362 children, among whom 24,937 were prenatally exposed to antidepressants. The comparison cohort consisted of 25,698 children. During follow-up, 1,183 exposed children and 1,291 children in the comparison cohort developed ADHD yielding an IRR of 1.05 (95% confidence interval [CI] = 0.96, 1.15) and an IRD of 0.28 (95% CI = -0.20, 0.80) pr. 1,000 person-years. IRRs from analyses attempting to reduce exposure misclassification varied from 1.03 to 1.07.

CONCLUSIONS

Our results were not consistent with the hypothesized effect of prenatal antidepressant exposure on the risk of ADHD. Attempts to reduce exposure misclassification did not alter this finding.

Persistence to Medications for Benign Prostatic Hyperplasia/Benign Prostatic Obstruction-Associated Lower Urinary Tract Symptoms in the ASL TO4 Regione Piemonte (Italy)

BACKGROUND

Pharmacological treatment of benign prostatic hyperplasia (BPH)/benign prostatic obstruction (BPO)-associated lower urinary tract symptoms (LUTS) aims at improving patients' quality of life by managing urinary symptoms and preventing complications and disease progression. However, continuous use of drugs to treat BPH/BPO-associated LUTS decreases over time. The aim of this retrospective observational study was to describe use of α1-adrenoceptor antagonists (ABs) and steroid 5α-reductase inhibitors (5ARIs) by adult (age ≥ 40 years) men in the ASL TO4, a Local Health Authority in the northern area of the city of Turin (Italy).

METHODS

Persistence measures were adopted as a robust, informative, and feasible way to understand medication-taking behavior and to assess patient compliance.

RESULTS

A total of 4309 men (median age 71 years) were enrolled. Monotherapy was the treatment option prescribed to the largest part of the study population. However, ≥two drugs were prescribed to a substantial proportion of men (23%). Men prescribed alfuzosin or dutasteride had significantly greater persistence, which decreased over time.

CONCLUSIONS

Unmet needs and areas of intervention for healthcare systems aimed at improving the use of drugs for BHP/BPO-associated LUTS in the ASL TO4 Regione Piemonte were identified.

Association of 5α-Reductase Inhibitors With Prostate Cancer Mortality

Importance

There is evidence that 5α-reductase inhibitors (5-ARIs), a standard treatment of benign prostate hyperplasia, are associated with a decrease in the incidence of prostate cancer (PCa). However, studies to date have had conflicting results regarding the association with prostate cancer mortality (PCM).

Objective

To evaluate the association of treatment with 5-ARIs with PCM in men without a prior diagnosis of PCa.

Design, Setting, and Participants

This population-based cohort study was conducted in Stockholm, Sweden, between January 1, 2007, and December 31, 2018, and included 429 977 men with a prostate-specific antigen (PSA) test within the study period. Study entry was set to 1 year after the first PSA test. Data were analyzed from September 2021 to December 2021.

Exposures

After their initial PSA test, men with 2 or more newly dispensed prescriptions of 5-ARI, finasteride, or dutasteride were considered 5-ARI users (n = 26 190).

Main Outcomes and Measures

Primary outcome was PCM. Cox proportional hazards regression models were used to calculate multivariable-adjusted hazard ratios (HRs) and 95% CIs for all-cause mortality and PCM.

Results

The study cohort included 349 152 men. The median (IQR) age for those with 2 or more filled prescriptions of 5-ARI was 66 (61-73) years and 57 (50-64) years for those without. The median follow-up time was 8.2 (IQR, 4.9-10) years with 2 257 619 person-years for the unexposed group and 124 008 person-years for the exposed group. The median exposure to treatment with 5-ARI was 4.5 (IQR, 2.1-7.4) years. During follow-up, 35 767 men (8.3%) died, with 852 deaths associated with PCa. The adjusted multivariable survival analysis showed a lower risk of PCM in the 5-ARI group with longer exposure times (0.1-2.0 years: adjusted HR, 0.89; 95% CI, 0.64-1.25; >8 years: adjusted HR, 0.44; 95% CI, 0.27-0.74). No statistically significant differences were seen in all-cause mortality between the exposed and unexposed group. Men treated with 5-ARIs underwent more PSA tests and biopsies per year than the unexposed group (median of 0.63 vs 0.33 and 0.22 vs 0.12, respectively).

Conclusions and Relevance

The results of this cohort study suggest that there was no association between treatment with 5-ARI and increased PCM in a large population-based cohort of men without a previous PCa diagnosis. Additionally, a time-dependent association was seen with decreased risk of PCM with longer 5-ARI treatment. Further research is needed to determine whether the differences are because of intrinsic drug effects or PCa testing differences.

Rationale and performances of a data-driven method for computing the duration of pharmacological prescriptions using secondary data sources

The assessment of the duration of pharmacological prescriptions is an important phase in pharmacoepidemiologic studies aiming to investigate persistence, effectiveness or safety of treatments. The Sessa Empirical Estimator (SEE) is a new data-driven method which uses k-means algorithm for computing the duration of pharmacological prescriptions in secondary data sources when this information is missing or incomplete. The SEE was used to compute durations of exposure to pharmacological treatments where simulated and real-world data were used to assess its properties comparing the exposure status extrapolated with the method with the “true” exposure status available in the simulated and real-world data. Finally, the SEE was also compared to a Researcher-Defined Duration (RDD) method. When using simulated data, the SEE showed accuracy of 96% and sensitivity of 96%, while when using real-world data, the method showed sensitivity ranging from 78.0 (nortriptyline) to 95.1% (propafenone). When compared to the RDD, the method had a lower median sensitivity of 2.29% (interquartile range 1.21–4.11%). The SEE showed good properties and may represent a promising tool to assess exposure status when information on treatment duration is not available.

Antidepressants Drug Use during COVID-19 Waves in the Tuscan General Population: An Interrupted Time-Series Analysis

In Italy, during the COVID-19 waves two lockdowns were implemented to prevent virus diffusion in the general population. Data on antidepressant (AD) use in these periods are still scarce. This study aimed at exploring the impact of COVID-19 lockdowns on prevalence and incidence of antidepressant drug use in the general population. A population-based study using the healthcare administrative database of Tuscany was performed. We selected a dynamic cohort of subjects with at least one ADs dispensing from 1 January 2018 to 27 December 2020. The weekly prevalence and incidence of drug use were estimated across different segments: pre-lockdown (1 January 2018–8 March 2020), first lockdown (9 March 2020–15 June 2020), post-first lockdown (16 June 2020–15 November 2020) and second lockdown (16 November 2020–27 December 2020). An interrupted time-series analysis was used to assess the effect of lockdowns on the observed outcomes. Compared to the pre-lockdown we observed an abrupt reduction of ADs incidence (Incidence-Ratio: 0.82; 95% Confidence-Intervals: 0.74–0.91) and a slight weekly decrease of prevalence (Prevalence-Ratio: 0.997; 0.996–0.999). During the post-first lockdown AD use increased, with higher incidence- and similar prevalence values compared with those expected in the absence of the outbreak. This pandemic has impacted AD drug use in the general population with potential rebound effects during the period between waves. This calls for future studies aimed at exploring the mid–long term effects of this phenomenon.

Impact of COVID-19 Lockdown, during the Two Waves, on Drug Use and Emergency Department Access in People with Epilepsy: An Interrupted Time-Series Analysis

BACKGROUND

In 2020, during the COVID-19 pandemic, Italy implemented two national lockdowns aimed at reducing virus transmission. We assessed whether these lockdowns affected anti-seizure medication (ASM) use and epilepsy-related access to emergency departments (ED) in the general population.

METHODS

We performed a population-based study using the healthcare administrative database of Tuscany. We defined the weekly time series of prevalence and incidence of ASM, along with the incidence of epilepsy-related ED access from 1 January 2018 to 27 December 2020 in the general population. An interrupted time-series analysis was used to assess the effect of lockdowns on the observed outcomes.

RESULTS

Compared to pre-lockdown, we observed a relevant reduction of ASM incidence (0.65; 95% Confidence Intervals: 0.59-0.72) and ED access (0.72; 0.64-0.82), and a slight decrease of ASM prevalence (0.95; 0.94-0.96). During the post-lockdown the ASM incidence reported higher values compared to pre-lockdown, whereas ASM prevalence and ED access remained lower. Results also indicate a lower impact of the second lockdown for both ASM prevalence (0.97; 0.96-0.98) and incidence (0.89; 0.80-0.99).

CONCLUSION

The lockdowns implemented during the COVID-19 outbreaks significantly affected ASM use and epilepsy-related ED access. The potential consequences of these phenomenon are still unknown, although an increased incidence of epilepsy-related symptoms after the first lockdown has been observed. These findings emphasize the need of ensuring continuous care of epileptic patients in stressful conditions such as the COVID-19 pandemic.

Pharmacological and epidemiological considerations while constructing treatment episodes using observational data: A simulation study

BACKGROUND

The permissible gap method is an extensively used approach for defining episodes of continuous treatment use in pharmacoepidemiology. This method uses the amount of drug redeemed, when available, and researcher-defined temporal gaps to fill the interval between the calculated end of coverage of a redeemed prescription and the date of redemption of the next prescription in the same treatment episode. The final scope is defining periods of continuous use of medications. There are strong pharmacological and epidemiological arguments for adding the gap at the end of each treatment episode. However, the evidence is scarce on the impact that such a practice has on measures of association. This study aims to compare the impact of adding or not adding the researcher-defined gap time to the end of a treatment episode on the incidence of drug discontinuation and the incidence rate for a simulated outcome that occurred during an observational window. Additionally, the study aims at assessing the magnitude of misclassification of exposure time between the two methods.

METHODS

A simulated dataset of 100 patients available in the R package AdhereR that contains 1080 redeemed prescriptions was used. A gap time of 90 days was used for constructing treatment episodes in an observational window of 365 days following the first redeemed prescription. Two approaches were used for defining treatment episodes that were named "gap+" and "gap-" and that respectively add and did not add the gap time at the end of a treatment episode. An outcome was simulated by using an exponential baseline hazard function with scale parameter λ = 0.5 and censoring at time t = 365 days. The incidence rate ratio for the simulated outcome between the two approaches was computed.

RESULTS

The 100 patients were followed for a median time of 183 days (interquartile range, IQR 50-365 days) and a median time of 273 days (IQR 140-365 days), respectively using "gap-" and "gap+". During the first 100 days of the follow-up period, none of the patients was found to discontinue the treatment with the method "gap+" while 38 patients discontinued using the method "gap-". The approach "gap+" exerted a higher incidence rate for the simulated outcome among the exposed (0.98 events/person-years) when compared to the "gap-" (0.82 events/person-years) during defined periods of continuous treatment use. When comparing the two approaches and using the method "gap-" as the reference group, the incidence rate ratio for the simulated outcome was 1.20 (95% confidence interval: CI 0.72-2.02) among the exposed.

CONCLUSIONS

This study showed that not adding the gap at the end of the treatment episodes leads to an overestimation of drug discontinuation, particularly at the beginning of the observational window, and an underestimation of the incidence rate of a hypothetical outcome during the period of exposure to the medication.

The use of uncertain exposure-A method to define switching and add-on in pharmacoepidemiology

PURPOSE

When defining exposure to pharmacological treatments in pharmacoepidemiology, register data often do not provide information regarding if a pharmacological treatment is a switch or an add-on. This study aims to compare two methods defining switching and add-on therapies and their impact on exposure-outcome associations. Additionally, to guide bias reduction, it aims to describe how the methods relate to immortal time bias and selection bias.

METHODS

Cohort study using Swedish population-based health registers to identify antidepressant (AD) prescriptions as exposures while hospitalizations for psychiatric reasons were used as an empirical outcome example. The first method for exposure definition used conditioning on future exposure (FE), the second used the concept of uncertain exposure (UE). To estimate associations between outcome and exposure categories "Use of one AD," "Use of two or more ADs", and "UE" compared to "Unexposed," hazard ratios (HRs) and 95% confidence intervals were estimated using Cox regression adjusted for age and sex.

RESULTS

Using the UE method, 7.2% of time periods were classified as "UE" with a notable proportion of psychiatric hospitalizations (7.7%) occurring during this time, while when using the FE method these hospitalizations were distributed over unexposed time and AD use time. The FE method resulted in slightly higher associations than the UE method. The highest HR was found during "UE": HR (95% CI) 5.54 (5.06-6.07).

CONCLUSIONS

This study suggests that to reduce the potential immortal time bias, selection bias, and exposure misclassification inherent to the FE method, the UE method could be used for identifying switching and add-on therapies. If not used as a main exposure definition, the UE method may be used to investigate the impact of UE time in a sensitivity analysis.

Pharmacoepidemiological methods for computing the duration of pharmacological prescriptions using secondary data sources

PURPOSE

In pharmacoepidemiology, correctly defining the exposure period of pharmacological treatment is a challenging step when information on the time in treatment is missing or incomplete.

METHODS

In this review, we describe several methods for defining exposure to pharmacological treatments using secondary data sources that lack such information.

RESULTS AND CONCLUSION

Several methods for assessing the duration of redeemed prescriptions and combining them into temporal sequences are available. We present a set of considerations to make researchers aware of the potentials and pitfalls of these methods that may aid in minimizing biases in research using these methods. Additionally, we highlight that, to date, there is no one-size-fits-all solution. Thus, the choice of method should be based on their area of applicability combined with a careful mapping to the research scenario under investigation.