Validity of algorithms for identifying five chronic conditions in MedicineInsight, an Australian national general practice database

Abstract

Background MedicineInsight is a database containing de-identified electronic health records (EHRs) from over 700 Australian general practices. It is one of the largest and most widely used primary health care EHR databases in Australia. This study examined the validity of algorithms that use information from various fields in the MedicineInsight data to indicate whether patients have specific health conditions. This study examined the validity of MedicineInsight algorithms for five common chronic conditions: anxiety, asthma, depression, osteoporosis and type 2 diabetes. Methods Patients’ disease status according to MedicineInsight algorithms was benchmarked against the recording of diagnoses in the original EHRs. Fifty general practices contributing data to MedicineInsight met the eligibility criteria regarding patient load and location. Five were randomly selected and four agreed to participate. Within each practice, 250 patients aged ≥ 40 years were randomly selected from the MedicineInsight database. Trained staff reviewed the original EHR for as many of the selected patients as possible within the time available for data collection in each practice. Results A total of 475 patients were included in the analysis. All the evaluated MedicineInsight algorithms had excellent specificity, positive predictive value, and negative predictive value (above 0.9) when benchmarked against the recording of diagnoses in the original EHR. The asthma and osteoporosis algorithms also had excellent sensitivity, while the algorithms for anxiety, depression and type 2 diabetes yielded sensitivities of 0.85, 0.89 and 0.89 respectively. Conclusions The MedicineInsight algorithms for asthma and osteoporosis have excellent accuracy and the algorithms for anxiety, depression and type 2 diabetes have good accuracy. This study provides support for the use of these algorithms when using MedicineInsight data for primary health care quality improvement activities, research and health system policymaking and planning.

Development and validation of algorithms to identify patients with chronic kidney disease and related chronic diseases across the Northern Territory, Australia

“…In contrast, most other CKD validation studies defined CKD as KDIGO stage 3a or higher (eGFR < 60 ml/min/1.73m2) -using this common definition of CKD, our algorithm had a sensitivity of 93% and specificity of 97%, and was comparable to existing studies with sensitivities ranging from 93 to 100% and specificities ranging from 0 to 99% [8,10,11]. Our algorithm sensitivity and specificity for diabetes [31][32][33], hypertension [34,35], and cardiovascular disease [36,37] also have comparable accuracy to that of previously published studies.…”

Diabetes Mellitus Diagnosis and Screening in Australian General Practice: A National Study

“…Moreover, the accuracy of the extracted information is another limitation. This limitation is mitigated by data checking: compared to the original EHRs available at the participating practices, data extracted from MedicineInsight had a sensitivity of 89% and specificity of 100% in identifying patients with diabetes [ 25 ].…”

“…Additional data extracted from the dataset included risk factors for diabetes (age 40+ years and overweight/obesity, AUSDRISK score ≥12 points, clinical history of CVD (including ischaemic heart disease and stroke), gestational diabetes, PCOS, or current use of antipsychotics (ATC N05A; 2018 only)) and other clinical conditions related to diabetes or prediabetes (hypertension, dyslipidaemia, CKD, atrial fibrillation, and heart failure) [ 14 ]. Data extraction was performed based on algorithms used in previous studies [ 25 , 30 , 33 ]. Overweight/obesity diagnosis used records of these terms as a “diagnosis,” “reason for encounter,” or “reason for prescription,” and body mass index data (i.e., ≥25.0 kg/m 2 ) recorded in the same fields or as a clinical measure in the “observation” field.…”

“…In this sense, EHRs generated by general practitioners (GPs) during medical appointments represent a unique data source for investigating the prevalence of diabetes and prediabetes diagnoses, screening activities, and management of these conditions. In addition, data extracted from EHR databases has been found a cost-effective method for exploring different health outcomes with appropriate accuracy [ 4 , 22 – 25 ].…”

Ethical considerations regarding the use of pooled data from electronic health records in general practice

“…10 It has since been used in a range of research and quality improvement projects, ranging from pharmaco-epidemiological studies to large cohort studies. 11,12 In another example, the Lumos program -a collaboration between NSW Health, Primary Health Networks and general practices -de-identified data from general practice EHRs that are linked to other New South Wales health and registry data. 13 From 1 August 2019, many general practices in Australia have been submitting de-identified data to Practice Incentives Program Eligible Data Sets with their local Primary Health Networks, 14,15 and may be familiar with the extraction process.…”