Confounding in Observational Studies Evaluating the Safety and Effectiveness of Medical Treatments

Causal effect between gut microbiota and metabolic syndrome in European population: a bidirectional mendelian randomization study

BACKGROUND

Observational studies have reported that gut microbiota composition is associated with metabolic syndrome. However, the causal effect of gut microbiota on metabolic syndrome has yet to be confirmed.

METHODS

We performed a bidirectional Mendelian randomization study to investigate the causal effect between gut microbiota and metabolic syndrome in European population. Summary statistics of gut microbiota were from the largest available genome-wide association study meta-analysis (n = 13,266) conducted by the MiBioGen consortium. The summary statistics of outcome were obtained from the most comprehensive genome-wide association studies of metabolic syndrome (n = 291,107). The inverse-variance weighted method was applied as the primary method, and the robustness of the results was assessed by a series of sensitivity analyses.

RESULTS

In the primary causal estimates, Actinobacteria (OR = 0.935, 95% CI = 0.878-0.996, P = 0.037), Bifidobacteriales (OR = 0.928, 95% CI = 0.868-0.992, P = 0.028), Bifidobacteriaceae (OR = 0.928, 95% CI = 0.868-0.992, P = 0.028), Desulfovibrio (OR = 0.920, 95% CI = 0.869-0.975, P = 0.005), and RuminococcaceaeUCG010 (OR = 0.882, 95% CI = 0.803-0.969, P = 0.009) may be associated with a lower risk of metabolic syndrome, while Lachnospiraceae (OR = 1.130, 95% CI = 1.016-1.257, P = 0.025), Veillonellaceae (OR = 1.055, 95% CI = 1.004-1.108, P = 0.034) and Olsenella (OR = 1.046, 95% CI = 1.009-1.085, P = 0.015) may be linked to a higher risk for metabolic syndrome. Reverse MR analysis demonstrated that abundance of RuminococcaceaeUCG010 (OR = 0.938, 95% CI = 0.886-0.994, P = 0.030) may be downregulated by metabolic syndrome. Sensitivity analyses indicated no heterogeneity or horizontal pleiotropy.

CONCLUSIONS

Our Mendelian randomization study provided causal relationship between specific gut microbiota and metabolic syndrome, which might provide new insights into the potential pathogenic mechanisms of gut microbiota in metabolic syndrome and the assignment of effective therapeutic strategies.

Methodological biases in observational hospital studies of COVID-19 treatment effectiveness: pitfalls and potential

Introduction

This study aims to discuss and assess the impact of three prevalent methodological biases: competing risks, immortal-time bias, and confounding bias in real-world observational studies evaluating treatment effectiveness. We use a demonstrative observational data example of COVID-19 patients to assess the impact of these biases and propose potential solutions.

Methods

We describe competing risks, immortal-time bias, and time-fixed confounding bias by evaluating treatment effectiveness in hospitalized patients with COVID-19. For our demonstrative analysis, we use observational data from the registry of patients with COVID-19 who were admitted to the Bellvitge University Hospital in Spain from March 2020 to February 2021 and met our predefined inclusion criteria. We compare estimates of a single-dose, time-dependent treatment with the standard of care. We analyze the treatment effectiveness using common statistical approaches, either by ignoring or only partially accounting for the methodological biases. To address these challenges, we emulate a target trial through the clone-censor-weight approach.

Results

Overlooking competing risk bias and employing the naïve Kaplan-Meier estimator led to increased in-hospital death probabilities in patients with COVID-19. Specifically, in the treatment effectiveness analysis, the Kaplan-Meier estimator resulted in an in-hospital mortality of 45.6% for treated patients and 59.0% for untreated patients. In contrast, employing an emulated trial framework with the weighted Aalen-Johansen estimator, we observed that in-hospital death probabilities were reduced to 27.9% in the "X"-treated arm and 40.1% in the non-"X"-treated arm. Immortal-time bias led to an underestimated hazard ratio of treatment.

Conclusion

Overlooking competing risks, immortal-time bias, and confounding bias leads to shifted estimates of treatment effects. Applying the naïve Kaplan-Meier method resulted in the most biased results and overestimated probabilities for the primary outcome in analyses of hospital data from COVID-19 patients. This overestimation could mislead clinical decision-making. Both immortal-time bias and confounding bias must be addressed in assessments of treatment effectiveness. The trial emulation framework offers a potential solution to address all three methodological biases.

Azithromycin use increases the risk of sudden cardiac death in patients with hemodialysis-dependent kidney failure

Azithromycin is an antibiotic with QT-prolonging potential commonly prescribed to individuals receiving hemodialysis. Hemodialysis patients have a high prevalence of clinical conditions, such as structural heart disease, that can enhance the pro-arrhythmic effects azithromycin, but were excluded from prior investigations evaluating the cardiac safety of azithromycin. Using data from the United States Renal Data System (2007-2017), we conducted two cohort studies to examine the cardiac safety of azithromycin relative to amoxicillin-based antibiotics (amoxicillin, amoxicillin/clavulanic acid) and levofloxacin (a fluoroquinolone antibiotic known to prolong the QT-interval) in the hemodialysis population. The primary outcome was five-day sudden cardiac death. Using inverse probability of treatment weighted survival models, we estimated hazard ratios, risk differences, and 95% confidence intervals. The azithromycin vs. amoxicillin-based antibiotic cohort included 282,899 patients and 725,431 treatment episodes (381,306 azithromycin and 344,125 amoxicillin-based episodes). Azithromycin vs. amoxicillin-based antibiotic treatment was associated with higher relative and absolute risks of sudden cardiac death, weighted hazard ratio of 1.70 (95% Confidence Interval, 1.36 to 2.11) and weighted risk difference per 100,000 treatment episodes of 25.0 (15.5 to 36.5). The azithromycin vs. levofloxacin cohort included 245,143 patients and 554,557 treatment episodes (387,382 azithromycin and 167,175 levofloxacin episodes). Azithromycin vs. levofloxacin treatment was associated with lower relative and absolute risks of sudden cardiac death, weighted hazard ratio of 0.79 (0.64 to 0.96) and weighted risk difference per 100,000 treatment episodes of -18.9 (-35.5 to -3.8). Thus, when selecting among azithromycin, levofloxacin, and amoxicillin-based antibiotics, clinicians should weigh the relative antimicrobial benefits of these drugs against their potential cardiac risks.

Medication Adherence, Complementary Medicine Usage and Progression of Diabetic Chronic Kidney Disease in Thais

PURPOSE

Non-adherence to medication is receiving more attention as a significant problem common to management of chronic diseases including diabetes and chronic kidney disease (CKD). This study was designed to assess the medication adherence and self-medication in a cohort of Thai patients with diabetic kidney disease, and its association with clinical outcomes.

PATIENTS AND METHODS

Non-dialysis patients with diabetic CKD visiting outpatient's clinics of Siriraj Hospital, the largest tertiary care in Thailand, were asked for participation. Self-administered questionnaire was given to assess medication adherence (the 6-item-medication-taking-behavior measure in Thai), complementary medicine usage, and personal information. Clinical, pharmaceutical, and relevant laboratory data (at current and the last visit of around 12 months) were abstracted from the medical records.

RESULTS

Of the 220 participants eligible (54.1% male, mean age 71.3), 50.9%, 24.1%, and 25% were classified as high-, medium-, and low-medication adherence, respectively. Overall, 24.1% reported self-usage of at least one type of herbal or complementary medicines. The most commonly identified items were cordyceps, cod liver oil, Nan Fui Chao, and turmeric (6 each), with unidentified Thai herbal mixture in 11. On multivariate analysis, late-stage CKD (stage IV-V) was the only independent predictor for low adherence (odds ratio (OR), 5.54; 95% confidence interval (CI), 2.82-10.88). Low adherence was associated with higher blood pressure, lower estimated glomerular filtrate rate (eGFR), and more eGFR decline with greater risk of being rapid CKD progressor (annual eGFR drop >5 mL/min/1.73 m²) [OR, 1.15; 95% CI, 1.06-1.25].

CONCLUSION

Medication taking behavior was a frequently encountered problem in Thai diabetic CKD patients. Increased medication non-adherence was independently predicted by stages of increasing CKD severity, and it was associated with poorer hypertensive control and kidney outcome. Targeting interventions to improve medication adherence should be an important strategy to slow CKD progression among patients with diabetic CKD.