Selection Bias in Population Registry-Based Comparative Effectiveness Research

Identifying the prevalence of Parkinson's disease in Denmark using healthcare registries and self-reported survey data

INTRODUCTION

Existing estimates of PD prevalence in Denmark are lower than those in the rest of Europe and are based on identification via single registries. Hence, are aim was to use a combined registry/self-report survey approach to identify people with PD and also investigate whether using different registry methods led to differences in the accuracy, completeness and characteristics of the identified cohorts.

METHODS

This study had a cross-sectional design using routinely collected health registry data to identify adults, ≥18 years of age and resident in Denmark, with PD from either the Danish National Patient (DNP) registry or Danish Prescription Medicines (DPM) registry. Those identified were asked to confirm their PD diagnosis using a national self-report survey.

RESULTS

13,433 people were identified potentially as having PD via the DNP or DPM registry and sent a survey. Of these, 9094 responded (68 %) of which 85 % confirmed they had PD (n = 7763; 194/100,000; 95%CI:7650-7876). When adjusting for non-respondents, assuming an equal rate of confirmation in respondents and non-respondents, estimated Danish PD population was 11,467 (198.4/100,000; 95 % CI:197.2-199.6). Identification of people using those found in both registries led to 98 % confirming they had PD versus using one registry: DNP 93 % and DPM 88 %. No clear differences in sociodemographic characteristics were found between different registry identification methods.

CONCLUSIONS

Estimated PD population in Denmark was significantly higher than previous Danish estimates and close to existing estimates in other European countries. The most accurate PD population was identified when including those found in both the DNP and DPM registries.

Neutrophil-to-lymphocyte ratio, white blood cell, and C-reactive protein predicts poor outcome and increased mortality in intracerebral hemorrhage patients: a meta-analysis

Objective

Inflammation participates in the pathology and progression of secondary brain injury after intracerebral hemorrhage (ICH). This meta-analysis intended to explore the prognostic role of inflammatory indexes, including neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), white blood cell (WBC), and C-reactive protein (CRP) in ICH patients.

Methods

Embase, PubMed, Web of Science, and Cochrane Library were searched until June 2023. Two outcomes, including poor outcome and mortality were extracted and measured. Odds ratio (OR) and 95% confidence interval (CI) were presented for outcome assessment.

Results

Forty-six studies with 25,928 patients were included in this meta-analysis. The high level of NLR [OR (95% CI): 1.20 (1.13-1.27), p < 0.001], WBC [OR (95% CI): 1.11 (1.02-1.21), p = 0.013], and CRP [OR (95% CI): 1.29 (1.08-1.54), p = 0.005] were related to poor outcome in ICH patients. Additionally, the high level of NLR [OR (95% CI): 1.06 (1.02-1.10), p = 0.001], WBC [OR (95% CI): 1.39 (1.16-1.66), p < 0.001], and CRP [OR (95% CI): 1.02 (1.01-1.04), p = 0.009] were correlated with increased mortality in ICH patients. Nevertheless, PLR was not associated with poor outcome [OR (95% CI): 1.00 (0.99-1.01), p = 0.749] or mortality [OR (95% CI): 1.00 (0.99-1.01), p = 0.750] in ICH patients. The total score of risk of bias assessed by Newcastle-Ottawa Scale criteria ranged from 7-9, which indicated the low risk of bias in the included studies. Publication bias was low, and stability assessed by sensitivity analysis was good.

Conclusion

This meta-analysis summarizes that the high level of NLR, WBC, and CRP estimates poor outcome and higher mortality in ICH patients.

Arterial events in cancer patients treated with apixaban for venous thrombosis

INTRODUCTION

In a recent interventional study of cancer patients with newly diagnosed venous thrombosis (VT), we found a high risk of arterial thrombotic events (AT) during treatment with therapeutic doses of apixaban.

METHODS

Total 298 cancer patients with VT received apixaban as treatment and secondary prophylaxis for up to 36 months. AT was registered as a serious adverse event, and this is a post hoc analysis of risk factors for AT. Clinical risk factors and concomitant medication were assessed through odds ratios (OR) with 95 % confidence interval using multivariate logistic regression. Biomarkers were assessed by non-parametric testing.

RESULTS

AT occurred in 16/298 patients (5.4 %, 95 % confidence interval (CI) 3.1-8.6 %). Median leucocyte count at baseline was higher in patients with AT compared with patients without AT (11 vs. 6.8·109/L, p < 0.01). Clinical factors associated with AT were pancreatic cancer (OR 13.7, 95 % CI 4.3-43.1), ovarian cancer (OR 19.3, 95 % CI 2.3-164.4), BMI <25 percentile (OR 3.1, 95 % CI 1.1-8.8) and previous VT (OR 4.4, 95 % CI 1.4-13.7). Pancreatic cancer had a cumulative incidence of AT of 36 % compared with 0.8 % for all other cancers at 6 months (p < 0.01). Non-steroidal anti-inflammatory drugs (OR 4.9, 95 % CI 1.0-26) and antiplatelet treatment (OR 3.8, 95 % CI 1.2-12.2) were associated with AT.

CONCLUSION

In cancer patients with apixaban treated VT, pancreatic cancer was strongly associated with AT. In addition, ovarian cancer, BMI < 25 percentile, previous VT, antiplatelet treatment, non-steroidal anti-inflammatory drug use and high leucocyte count at baseline were associated with AT. The CAP study is registered with the unique identifier NCT02581176 in ClinicalTrials.gov.

Retrospective comparative effectiveness research: will changing the analytical methods change the results?

In medicine, retrospective cohort studies are used to compare treatments to one another. We hypothesize that the outcomes of retrospective comparative effectiveness research studies can be heavily influenced by biostatistical analytic choices, thereby leading to inconsistent conclusions. We selected a clinical scenario currently under investigation: survival in metastatic prostate, breast, or lung cancer after systemic vs systemic + definitive local therapy. We ran >300 000 regression models (each representing a publishable study). Each model had various forms of analytic choices (to account for bias): propensity score matching, left truncation adjustment, landmark analysis, and covariate combinations. There were 72 549 lung, 14 904 prostate, and 13 857 breast cancer patients included. In the most basic analysis, which omitted propensity score matching, left truncation adjustment, and landmark analysis, all of the HRs were < 1 (generally, 0.60-0.95, favoring addition of local therapy), with all P-values < 0.001. Left truncation adjustment landmark analysis produced results with non-significant P-values. The combination of propensity score matching, left truncation adjustment, landmark analysis, and covariate combinations generally produced P-values that were > 0.05 and/or HRs that were > 1 (favoring systemic therapy alone). The use of more statistical methods to reduce the selection bias caused reported HR ranges to approach 1.0. By varying analytic choices in comparative effectiveness research, we generated contrary outcomes. Our results suggest that some retrospective observational studies may find a treatment improves outcomes for patients, while another similar study may find it does not, simply based on analytical choices. This article is protected by copyright. All rights reserved.

Comparative effectiveness of surgery versus external beam radiation with/without brachytherapy in high-risk localized prostate cancer

BACKGROUND

It remains controversial if radical prostatectomy or definitive radiation therapy produces equivalent outcomes in high-risk localized prostate cancer.

METHODS

We queried The Surveillance, Epidemiology, and End Results (SEER) database for those who received upfront surgery or who were recommended for surgery but instead received radiation. Inverse probability of treatment weighing was used to adjust for covariate imbalance and the weighted Cox proportional hazards model was used to estimate the effects of treatment groups on survival. A meta-analysis was performed to pool estimates from published studies.

RESULTS

Among eligible 62 533 patients, 59 540 had upfront surgery and 2993 patients had upfront radiotherapy. EBRT + BT was associated with a superior cancer-specific survival (CSS) compared with surgery or EBRT alone (HR, 0.55, 95% CI, 0.3-1.0; HR, 0.49, 95% CI, 0.24-0.98, respectively), whereas EBRT was associated with an inferior overall survival (OS) compared with surgery (HR, 1.46, 95% CI, 1.16-1.8). Radiotherapy (EBRT ± BT) was inferior to surgery by OS (HR, 1.63, 95% CI, 1.13-2.34) in patients ≤ 65 years, and was superior to surgery by CSS in patients > 65 years (HR, 0.69, 95% CI, 0.49-0.97). The meta-analysis showed consistent results.

CONCLUSION

EBRT + BT was associated with a significantly better prostate CSS compared with surgery or EBRT. EBRT alone was inferior to surgery by OS.