A narrative review of using prescription drug databases for comorbidity adjustment: A less effective remedy or a prescription for improved model fit?

Impact of comorbidities on prostate cancer-specific mortality: A population-based cohort study

AIM

To assess the impact of comorbidities on prostate cancer mortality.

METHODS

We studied 15,695 South Australian men diagnosed with prostate cancer between 2003 and 2019 from state-wide administrative linked data sets. Comorbidity was measured 1-year before prostate cancer diagnosis using Rx-Risk, a medication-based comorbidity index. Flexible parametric competing risk regression was used to estimate the independent association between comorbidities and prostate cancer-specific mortality. Specific common comorbidities within Rx-Risk (cardiac disorders, diabetes, chronic airway diseases, depression and anxiety, thrombosis, and pain) were also assessed to determine their association with mortality. All models were adjusted for sociodemographic variables, tumor characteristics, and treatment type.

RESULTS

Prostate cancer-specific mortality was higher for patients with a Rx-Risk score ≥3 versus 0 (adjusted sub-hazard ratio (sHR) 1.34, 95% CI: 1.15-1.56). Lower comorbidity scores (Rx-Risk score 2 vs. 0 and Rx-Risk score 1 vs. 0) were not significantly associated with prostate cancer-specific mortality. Men who were using medications for cardiac disorders (sHR 1.31, 95% CI: 1.13-1.52), chronic airway disease (sHR 1.20, 95% CI: 1.01-1.44), depression and anxiety (sHR 1.17, 95% CI: 1.02-1.35), and thrombosis (sHR 1.21, 95% CI: 1.04-1.42) were at increased risk of dying from prostate cancer compared with men not on those medications. Use of medications for diabetes and chronic pain were not associated with prostate cancer-specific mortality. All Rx-Risk score categories and the specific comorbidities were also associated with increased risk of all-cause mortality.

CONCLUSION

The findings showed that ≥3 comorbid conditions and specific comorbidities including cardiac disease, chronic airway disease, depression and anxiety, and thrombosis were associated with poor prostate cancer-specific survival. Appropriate management of these comorbidities may help to improve survival in prostate cancer patients.

Medication-based Comorbidity Measures and Prostate Cancer Treatment Selection

INTRODUCTION

We aimed to assess the association between comorbidities and prostate cancer management.

PATIENTS AND METHODS

We studied 12,603 South Australian men diagnosed with prostate cancer between 2003 and 2019. Comorbidity was measured one year prior to prostate cancer diagnosis using a medication-based comorbidity index (Rx-Risk). Binomial logistic regression analyses were used to assess the association between comorbidities and primary treatment selection (active surveillance, radical prostatectomy (RP), external beam radiotherapy (EBRT) with or without androgen deprivation therapy (ADT), brachytherapy, ADT alone, and watchful waiting (WW)). Certain common comorbidities within Rx-Risk (cardiac disorders, diabetes, chronic airway diseases, depression and anxiety, thrombosis, and chronic pain) were also assessed. All models were adjusted for sociodemographic and tumor characteristics.

RESULTS

Likelihood of receiving RP was lower among men with Rx-Risk score ≥3 (odds ratio (OR) 0.62, 95%CI:0.56-0.69) and Rx-Risk 2 (OR 0.80, 95%CI:0.70-0.92) compared with no comorbidity (Rx-Risk ≤0). Men with high comorbidity (Rx-Risk ≥3) were more likely to have received ADT alone (OR 1.76, 95%CI:1.40-2.21), EBRT (OR 1.30, 95%CI:1.17-1.45) or WW (OR 1.49, 95%CI:1.19-1.88) compared with Rx-Risk ≤0. Pre-existing cardiac and respiratory disorders, thrombosis, diabetes, depression and anxiety, and chronic pain were associated with lower likelihood of selecting RP and higher likelihood of EBRT (except chronic airway disease) or WW (except diabetes and depression and anxiety). Cardiac disorders and thrombosis were associated with higher likelihood of selecting ADT alone. Furthermore, age had greater effect on treatment choice than the level of comorbidity.

CONCLUSION

High comorbidity burden was associated with primary treatment choice, with significantly less RP and more EBRT, WW and ADT alone among men with higher levels of comorbidity. Each of the individual comorbid conditions also influenced treatment selection.

Utility of prescription-based comorbidity indices for predicting mortality among Australian men with prostate cancer

BACKGROUND

Drug prescription registries has become an alternative data source to hospital admission databases for measuring comorbidities. However, the predictive validity of prescription-based comorbidity measures varies based on the population under investigation and outcome of interest. We aimed to determine which prescription-based index of comorbidity has most utility in Australian men with prostate cancer.

METHODS

We studied 25,414 South Australian men diagnosed with prostate cancer between 2003 and 2019 from state-wide administrative linked datasets. The Rx-Risk index, Chronic Disease Score (CDS), Drug Comorbidity Index (DCI) and Pharmaceutical Prescribing Profile (P3) with one year lookback period from prostate cancer diagnosis were evaluated. The predictive ability of each index to determine all-cause deaths within two and five years of prostate cancer diagnosis was compared using the c-statistic from flexible parametric survival models, adjusting for age, socioeconomic status and year of prostate cancer diagnosis.

RESULTS

The Rx-Risk index performed better in predicting two-year (c-statistic = 0.818) and five-year (c-statistic = 0.784) all-cause mortality than P3, CDS and DCI. Including comorbidity measures as continuous scores resulted in a better performance than including them as categories. Grouping scores into four categories (≤0, >0 - ≤1, >1 - ≤2, and >2) resulted in better performance and calibration than using fewer categories.

CONCLUSION

Rx-Risk was validated in Australia and reflects Australian prescribing patterns. It showed better predictive performance for mortality in our study, with a modest improvement over P3, CDS and DCI. For research with prostate cancer populations, we recommend the use of drug-based comorbidity indices that have been validated in a similar population.

Pharmacoepidemiology: An Overview

The aims of this review are to provide a comprehensive overview of the definition and scope of pharmacoepidemiology, to summarize the study designs and methodologies used in the field, to discuss the future trends in the field and new methodologies to address bias and confounding, and finally to give some recommendations to clinicians interested in pharmacoepidemiologic research. Because drug efficacy and safety from randomized clinical trials do not reflect the real-world situation, pharmacoepidemiological studies on drug safety monitoring and drug effectiveness in large numbers of people are needed by healthcare professionals and regulatory institutions. We aim to highlight the importance of pharmacoepidemiologic research in informing evidence-based medicine and public health policy. The development of new designs and methodologies for the generation of valid evidence, as well as new initiatives to provide guidance and recommendations on how to incorporate real-world evidence into the drug development process, are reported on. In addition, we have touched on the implication of artificial intelligence in the management of real-world data. This overview aims to summarize all important aspects to consider when conducting or interpreting a pharmacoepidemiologic study.

Cross-validation of comorbidity items in two national databases in a sample of patients with end-stage kidney disease

BACKGROUND

The use of national medico-administrative databases for epidemiological studies has increased in the last decades. In France, the Healthcare Expenditures and Conditions Mapping (HECM) algorithm has been developed to analyse and monitor the morbidity and economic burden of 58 diseases. We aimed to assess the performance of the HECM in identifying different conditions in patients with end-stage kidney disease (ESKD) using data from the REIN registry (the French National Registry for patients with ESKD).

METHODS

We included all patients over 18 years of age who started renal replacement therapy in France in 2018. Five conditions with a similar definition in both databases were included (ESKD, diabetes, human immunodeficiency virus [HIV], coronary insufficiency, and cancer). The performance of each SNDS algorithm was assessed using sensitivity, specificity, positive predictive values (PPVs), negative predictive values (NPVs), and Cohen's kappa coefficient.

RESULTS

In total 5,971 patients were included. Among them, 81% were identified as having ESKD in both databases. Diabetes was the condition with the best performance, with a sensitivity, specificity, PPV, NPV, and Kappa coefficient all over 80%. Cancer had the lowest level of agreement with a Kappa coefficient of 51% and a high specificity and high NPV (94% and 95%). The conditions for which the definition in the HECM included disease-specific medications performed better in our study.

CONCLUSION

The HECM showed good to very good concordance with the REIN database information overall, with the exception of cancer. Further validation of the HECM tool in other populations should be performed.