Survival analysis using primary care electronic health record data: A systematic review of the literature

Hodgkins, Adam; Bonney, Andrew D; Mullan, Judy; Mayne, Darren J; Barnett, Stephen

doi:10.1177/1833358316687090

Cited by 7 publications

(9 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3,4 The time and cost of data collection can be reduced by using EHR data rather than other data collection methods. 3,5 The use of general practice EHRs as a research data source is expanding. 5 There are a large number of primary care EHR data collection projects and substantial research output from Australia, the USA, Canada and several European countries.…”

Section: Discussionmentioning

confidence: 99%

“…3,5 The use of general practice EHRs as a research data source is expanding. 5 There are a large number of primary care EHR data collection projects and substantial research output from Australia, the USA, Canada and several European countries. 6 The UK leads the way, with over 40% of this nation's research output using primary care EHR data.…”

Section: Discussionmentioning

confidence: 99%

“…AH has an interest in using EHR data for research and has published research in this area. 5,13 These details were disclosed to participants prior to obtaining consent.…”

Section: Interviewsmentioning

confidence: 99%

See 2 more Smart Citations

Australian general practitioners’ attitudes to the extraction of research data from electronic health records

Hodgkins

Mullan

Mayne³

et al. 2020

Aust J Gen Pract

Self Cite

View full text Add to dashboard Cite

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Australian general practitioners’ attitudes to the extraction of research data from electronic health records

Hodgkins

Mullan

Mayne³

et al. 2020

Aust J Gen Pract

Self Cite

View full text Add to dashboard Cite

“…Comprising heterogeneous clinical encounters including diagnostic and procedural billing codes, lab tests, prescriptions, and free text clinical notes for millions of patients, these rich data offer abundant opportunities for in silico epidemiological analysis. One application that has garnered recent interest is estimation of population disease risk within EHR patient cohorts, which can enable more powerful and precise estimation of real-world disease risks as well as comparative effectiveness analysis of alternative treatment strategies (Hodgkins and others , 2017; Dean and others , 2003; Liu and others , 2018; Panahiazar and others , 2015; Steele and others , 2018). Several studies have had success estimating time to death within rule-defined disease cohorts (Panahiazar and others , 2015; Steele and others , 2018).…”

Section: Introductionmentioning

confidence: 99%

Semi-supervised Calibration of Risk with Noisy Event Times (SCORNET) Using Electronic Health Record Data

Ahuja

Liang

Huang

et al. 2021

Preprint

View full text Add to dashboard Cite

Leveraging large-scale electronic health record (EHR) data to estimate survival curves for clinical events can enable more powerful risk estimation and comparative effectiveness research. However, use of EHR data is hindered by a lack of direct event times observations. Occurrence times of relevant diagnostic codes or target disease mentions in clinical notes are at best a good approximation of the true disease onset time. On the other hand, extracting precise information on the exact event time requires laborious manual chart review and is sometimes altogether infeasible due to a lack of detailed documentation. Current status labels – binary indicators of phenotype status during follow up – are significantly more efficient and feasible to compile, enabling more precise survival curve estimation given limited resources. Existing survival analysis methods using current status labels focus almost entirely on supervised estimation, and naive incorporation of unlabeled data into these methods may lead to biased results. In this paper we propose Semi-supervised Calibration of Risk with Noisy Event Times (SCORNET), which yields a consistent and efficient survival curve estimator by leveraging a small size of current status labels and a large size of imperfect surrogate features. In addition to providing theoretical justification of SCORNET, we demonstrate in both simulation and real-world EHR settings that SCORNET achieves efficiency akin to the parametric Weibull regression model, while also exhibiting non-parametric flexibility and relatively low empirical bias in a variety of generative settings.

show abstract

“…Data recorded contemporaneously can overcome recall bias [6]. By using EHR data, the time and cost of additional data collection can be reduced [4,7]. While some data in EHRs are of variable quality, other data are of high quality and have been benchmarked against accepted general practice standards [8].…”

Section: Introductionmentioning

confidence: 99%

The Use of Primary Care Electronic Health Records for Research: Lipid Medications and Mortality in Elderly Patients

et al. 2019

Self Cite

View full text Add to dashboard Cite

General practice electronic health record (EHR) data have significant potential for clinical research. This study demonstrates the feasibility of utilising longitudinal EHR data analysis to address clinically relevant outcomes and uses the relationship between lipid medication prescription and all-cause mortality in the elderly as an exemplar for the validity of this methodology. EHR data were analysed to describe the association of lipid medication use, non-use or cessation with all-cause mortality in patients aged ≥75 years. Survival analysis with Cox regression was used to calculate hazard ratios, which were adjusted for confounders. There was no significant difference in all-cause mortality among patients according to their use, non-use, or cessation of lipid medications. The outcomes of this study correlate well with the results of other research works. This single-practice study demonstrates the feasibility and potential of analysing EHR data to address important clinical issues such as the relationship between all-cause mortality and lipid medication prescription in the elderly.

show abstract

Survival analysis using primary care electronic health record data: A systematic review of the literature

Cited by 7 publications

References 51 publications

Australian general practitioners’ attitudes to the extraction of research data from electronic health records

Australian general practitioners’ attitudes to the extraction of research data from electronic health records

Semi-supervised Calibration of Risk with Noisy Event Times (SCORNET) Using Electronic Health Record Data

The Use of Primary Care Electronic Health Records for Research: Lipid Medications and Mortality in Elderly Patients

Contact Info

Product

Resources

About