Primary versus secondary source of data in observational studies and heterogeneity in meta-analyses of drug effects: a survey of major medical journals

Artificial intelligence-driven prediction of COVID-19-related hospitalization and death: a systematic review

Aim

To perform a systematic review on the use of Artificial Intelligence (AI) techniques for predicting COVID-19 hospitalization and mortality using primary and secondary data sources.

Study eligibility criteria

Cohort, clinical trials, meta-analyses, and observational studies investigating COVID-19 hospitalization or mortality using artificial intelligence techniques were eligible. Articles without a full text available in the English language were excluded.

Data sources

Articles recorded in Ovid MEDLINE from 01/01/2019 to 22/08/2022 were screened.

Data extraction

We extracted information on data sources, AI models, and epidemiological aspects of retrieved studies.

Bias assessment

A bias assessment of AI models was done using PROBAST.

Participants

Patients tested positive for COVID-19.

Results

We included 39 studies related to AI-based prediction of hospitalization and death related to COVID-19. The articles were published in the period 2019-2022, and mostly used Random Forest as the model with the best performance. AI models were trained using cohorts of individuals sampled from populations of European and non-European countries, mostly with cohort sample size <5,000. Data collection generally included information on demographics, clinical records, laboratory results, and pharmacological treatments (i.e., high-dimensional datasets). In most studies, the models were internally validated with cross-validation, but the majority of studies lacked external validation and calibration. Covariates were not prioritized using ensemble approaches in most of the studies, however, models still showed moderately good performances with Area under the Receiver operating characteristic Curve (AUC) values >0.7. According to the assessment with PROBAST, all models had a high risk of bias and/or concern regarding applicability.

Conclusions

A broad range of AI techniques have been used to predict COVID-19 hospitalization and mortality. The studies reported good prediction performance of AI models, however, high risk of bias and/or concern regarding applicability were detected.

Primary and secondary data in emergency medicine health services research - a comparative analysis in a regional research network on multimorbid patients

BACKGROUND

This analysis addresses the characteristics of two emergency department (ED) patient populations defined by three model diseases (hip fractures, respiratory, and cardiac symptoms) making use of survey (primary) and routine (secondary) data from hospital information systems (HIS). Our aims were to identify potential systematic inconsistencies between both data samples and implications of their use for future ED-based health services research.

METHODS

The research network EMANET prospectively collected primary data (n=1442) from 2017-2019 and routine data from 2016 (n=9329) of eight EDs in a major German city. Patient populations were characterized using socio-structural (age, gender) and health- and care-related variables (triage, transport to ED, case and discharge type, multi-morbidity). Statistical comparisons between descriptive results of primary and secondary data samples for each variable were conducted using binomial test, chi-square goodness-of-fit test, or one-sample t-test according to scale level.

RESULTS

Differences in distributions of patient characteristics were found in nearly all variables in all three disease populations, especially with regard to transport to ED, discharge type and prevalence of multi-morbidity. Recruitment conditions (e.g., patient non-response), project-specific inclusion criteria (e.g., age and case type restrictions) as well as documentation routines and practices of data production (e.g., coding of diagnoses) affected the composition of primary patient samples. Time restrictions of recruitment procedures did not generate meaningful differences regarding the distribution of characteristics in primary and secondary data samples.

CONCLUSIONS

Primary and secondary data types maintain their advantages and shortcomings in the context of emergency medicine health services research. However, differences in the distribution of selected variables are rather small. The identification and classification of these effects for data interpretation as well as the establishment of monitoring systems in the data collection process are pivotal.

TRIAL REGISTRATION

DRKS00011930 (EMACROSS), DRKS00014273 (EMAAGE), NCT03188861 (EMASPOT).

Big data analytics in supply chain management: a systematic literature review

Purpose

The purpose of this paper is to analyse and classify the literature that contributed to three questions, namely, what are the benefits of big data analytics (BDA) in the field of supply chain management (SCM) and logistics, what are the challenges in BDA applications in the field of SCM and logistics and what are the determinants of successful applications of BDA in the field of SCM and logistics.

Design/methodology/approach

This paper conducts a systematic literature review (SLR) to analyse the findings of 44 selected papers published in the period from 2016 to 2020, in the area of BDA and its impact on SCM. The designed protocol is composed of 14 steps in total, following Tranfeld (2003). The selected research papers are categorized into four themes.

Findings

This paper identifies sets of benefits to be gained from the use of BDA in SCM, including benefits in data analytics capabilities, operational efficiency of logistical operations and supply chain/logistics sustainability and agility. It also documents challenges to be addressed in this application, and determinants of successful implementation.

Research limitations/implications

The scope of the paper is limited to the related literature published until the beginning of Corona Virus (COVID) pandemic. Therefore, it does not cover the literature published since the COVID pandemic.

Originality/value

This paper contributes to the academic research by providing a roadmap for future empirical work into this field of study by summarising the findings of the recent work conducted to investigate the uses of BDA in SCM and logistics. Specifically, this paper culminates in a summary of the most relevant benefits, challenges and determinants discussed in recent research. As the field of BDA remains a newly established field with little practical application in SCM and logistics, this paper contributes by highlighting the most important developments in contemporary literature practical applications.

Bias in pharmacoepidemiologic studies using secondary health care databases: a scoping review

BACKGROUND

The availability of clinical and therapeutic data drawn from medical records and administrative databases has entailed new opportunities for clinical and epidemiologic research. However, these databases present inherent limitations which may render them prone to new biases. We aimed to conduct a structured review of biases specific to observational clinical studies based on secondary databases, and to propose strategies for the mitigation of those biases.

METHODS

Scoping review of the scientific literature published during the period 2000-2018 through an automated search of MEDLINE, EMBASE and Web of Science, supplemented with manually cross-checking of reference lists. We included opinion essays, methodological reviews, analyses or simulation studies, as well as letters to the editor or retractions, the principal objective of which was to highlight the existence of some type of bias in pharmacoepidemiologic studies using secondary databases.

RESULTS

A total of 117 articles were included. An increasing trend in the number of publications concerning the potential limitations of secondary databases was observed over time and across medical research disciplines. Confounding was the most reported category of bias (63.2% of articles), followed by selection and measurement biases (47.0% and 46.2% respectively). Confounding by indication (32.5%), unmeasured/residual confounding (28.2%), outcome misclassification (28.2%) and "immortal time" bias (25.6%) were the subcategories most frequently mentioned.

CONCLUSIONS

Suboptimal use of secondary databases in pharmacoepidemiologic studies has introduced biases in the studies, which may have led to erroneous conclusions. Methods to mitigate biases are available and must be considered in the design, analysis and interpretation phases of studies using these data sources.