that should be considered when interpreting the results. First, the retrospective nature of the study may introduce biases related to data collection and patient selection. The reliance on hospital re

The gap

Consensus across the literature

Supporting evidence — 4 representative gaps

• Opioids in pain medicine (2022) · doi

  There are several limitations to this trial. The rate of recruitment was slower than anticipated due to COVID-19-related surgical disruptions. The use of patient self-report data is subject to response and recall biases. Participants’ comorbidities and overall comorbidity burden were not collected in this trial. Given comorbidity burden may affect opioid tapering, future studies should collect participants’ comorbidities as well as a composite measure of comorbidity burden such as the Charlson Comorbidity Index. Some differences in baseline opioid doses were observed between the intervention and control groups, which may indicate inadequate randomization. Inadequate randomization may lead to bias in unadjusted analyses. Furthermore, this pilot trial was not statistically powered to detect differences in secondary outcomes, so between-group comparisons for these outcomes should be interpreted with caution. While the involvement of seven hospitals across metropolitan and regional settings of New South Wales, Australia strengthens the generalizability of our findings, the generalizability of our results to other contexts of care remains unknown. Participant recruitment and surgery timing was delayed due to the COVID-19 pandemic; we cannot exclude the potential impact (such as social isolation, prolonged surgery wait times, transfer of 115 Chapter 5 public to private health system care) of the pandemic on the study findings (such as patterns of opioid use and patient-reported outcomes). Behaviour in the usual care arm may have potentially confounded results. The extent of opioid tapering in both the intervention and control arms was measured to account for this. Due to the nature of the intervention, we were unable to blind participants and the research officer to allocation, thus potentially biasing outcome assessment. However, several outcomes independent of both the patient and research officer corroborate the effectiveness of opioid tapering pre-operatively. These include opioid dose prescribed at discharge from acute-care, acute-care length of stay, and the favourable directional changes in acute-care opioid consumption. Blinded statistical analysis was conducted through the use of masked patient identification. An ongoing qualitative study involving participants and clinicians who engaged in the trial will further elucidate factors affecting the acceptability, implementation, and mitigation strategies regarding the study intervention in preparation for a future full-scale RCT.

  Keywords: opioid care trial patient participants comorbidity intervention outcomes burden tapering acute several recruitment covid comorbidities
• Prevalence and clinical correlates of human papillomavirus infection: A study of 7,601 women attending a tertiary referral center in Greece (2026) · doi

  The interpretation of our findings should be considered in light of several limitations. First, the retrospective and cross- sectional design, combined with data collection from a single tertiary referral center, introduces a selection bias. Our results reflect the profile of a high-risk cohort attending a specific institution, and causal inferences cannot be established. Second, due to the retrospective nature of the analysis, data on confounding variables (such as HPV vaccination status, smoking, and detailed sexual history) were not consistently available and could not be included in the statistical models. Specifically, the lack of vaccination status information prevents us from accurately assessing the impact of the 2008 national program on current HPV prevalence and genotype distribution, particularly among the younger cohorts. This absence likely leads to an overestimation of the HPV prevalence in vaccinated populations and should be taken into account during interpretation. This lack of control may influence the observed associations. Third, while our sample size (n = 7,601) is large, its retrospective derivation means that we did not perform a priori sample size calculation. However, the size is considered highly adequate to detect meaningful differences in prevalence and associations, ensuring high statistical power. Fourth, the exclusion of non- diagnostic records, such as those labeled with “inflammation” or “cervical atrophy” [Figure 1], while necessary to maintain the analytic focus on squamous intraepithelial lesions, may have introduced a small selection bias by excluding data potentially related to specific HPV types or age-related immunological status. Stamoulis et al.: HPV prevalence and correlates in Greek womenSUMMARY This retrospective study successfully met its objective of defining the prevalence, genotype distribution, and clinical correlates of HPV infection burden within this specific Greek tertiary care cohort. Our findings confirmed a high overall HPV prevalence of 31% and established a clear burden– response relationship, as multiple infections were significantly associated with the highest Odds for low-grade lesions, particularly among younger women. While the highest risks for HSIL were concentrated among the established oncogenic types (HPV-16, -18, and -33), our analysis also demonstrated significant associations of the low-risk HPV-42 with LSIL and of the non-vaccine-covered HPV-51 genotype with both ASC-US and LSIL. These comprehensive results reinforce the clinical necessity of genotype and infection burden profiling for accurate risk stratification and underscore the critical need for continued age-stratified screening and localized genotype surveillance to inform future prophylactic strategies in Greece. AVAILABILITY OF DATA AND MATERIALS An anonymized version of the dataset analyzed during the current study is available from the corresponding author upon reasonable request. ABBREVIATIONS

  Keywords: prevalence genotype retrospective high risk specific established status among associations size burden interpretation considered tertiary
• Interim Effectiveness of 2025–2026 mRNA-1283 and BNT162b2 COVID-19 Vaccines Against COVID-19-Related Outcomes Among Adults Aged ≥ 65 Years in the United States (2026) · doi

  Although we have identified and adjusted for many baseline confounders, cohort assignment was not randomized, and residual confound- ing may remain, as is inherent to observational study designs. For instance, vaccinated individ- uals may engage in other behaviors associated with a healthier lifestyle, such as more frequent exercise, contributing to their lower rates of hospitalization with COVID-19 and this may not be addressed with the available covariates. Also, the study population only includes indi- viduals engaged with the healthcare system, possibly excluding others due to barriers such as trust, socioeconomic status, or accessibility. The outcome definition of hospitalization with COVID-19 partially mitigates this limitation, as severe outcomes are less influenced by vari- ation in healthcare-seeking behavior. Although the study design requires that vaccinated indi- viduals remain event-free for 7 days following vaccination, the use of a symmetric study design among the unvaccinated cohorts minimizes the introduction of bias between cohorts. As with all claims and/or EHR data, vaccina- tion status may be incompletely captured, which could have led to bias toward the null hypoth- esis and resulted in underestimation of true aVE. In addition, outcomes may be underestimated if people avoid obtaining medical care. While treatment avoidance is possible in both vacci- nated and unvaccinated individuals, patients who avoid vaccination may be more likely to avoid other medically-necessary care. This differ- ence in healthcare seeking behavior would also bias towards underestimating aVE Misclassifica- tion of both exposure and outcomes is also pos- sible, as diagnostic and procedural codes used to define vaccination and COVID-19-related hospi- talization may be subject to coding errors. How- ever, validation studies suggest that outcome misclassification is likely limited; for example, Kadri et al. reported that the ICD-10-CM code U07.1 for COVID-19 had a positive predictive value of approximately 98% when compared with clinical documentation and laboratory- confirmed infection [20]. A further limitation is that for some patients COVID-19 may be an incidental diagnosis that is captured as part of a hospitalization for an unrelated health event. While this is possible, the weighting process should ensure that the odds of an unrelated hospitalization are similar between cases and controls. Therefore, while this may lead to an overestimate of the frequency of COVID-19-related hospitalizations it should not bias the aVE results. Furthermore, the low frequency of hospitalizations, particularly in the subgroup analysis, resulted in large confidence intervals in adjusted aVE. Additional analyses with data from the full season that allows for the capture of more events over a longer follow-up time may help assess the durability of vaccine effectiveness and improve confidence in these findings. Additionally, these data sources may not fully capture all hospitalizations, particularly among populations with limited or fragmented access to care, such as uninsured individuals, underrep- resented groups, or those receiving care outside the insurance network. The patient selection also excluded patients based on receipt of immu- nomodulators used to treat COVID-19; however, this approach likely excluded individuals who received these medications for non-COVID-19 indications. This may affect generalizability and introduce selection bias, though the exclusions were applied to both cases and controls.

  Keywords: covid bias hospitalization care healthcare outcomes vaccination avoid individuals patients likely hospitalizations adjusted remain vaccinated
• Clinical Parameters Determining Mortality in COVID-19 Patients: Effects of PCR Positivity, Vaccination Status, Comorbidities and Thoracic CT Findings on Diagnosis and Treatment Process (2026) · doi

  that should be considered when interpreting the results. First, the retrospective nature of the study may introduce biases related to data collection and patient selection. The reliance on hospital records and diagnostic tests could inaccurate data, have particularly status and regarding vaccination comorbidities. Second, the study only includes patients who presented the Emergency COVID-19 Outpatient Clinic of our hospital, which may not be representative of the broader population. As a result, the findings may not be generalizable patients, particularly those who did not seek care at this facility or those with milder symptoms who were not hospitalized or tested. all COVID-19 incomplete or led to to to

  Keywords: hospital particularly patients covid considered interpreting first retrospective nature introduce biases related collection patient selection