Data Resource Profile: MedicineInsight, an Australian national primary health care database

Implementing privacy preserving record linkage: Insights from Australian use cases

OBJECTIVE

To describe the use of privacy preserving linkage methods operationally in Australia, and to present insights and key learnings from their implementation.

METHODS

Privacy preserving record linkage (PPRL) utilising Bloom filters provides a unique practical mechanism that allows linkage to occur without the release of personally identifiable information (PII), while still ensuring high accuracy.

RESULTS

The methodology has received wide uptake within Australia, with four state linkage units with privacy preserving capability. It has enabled access to general practice and private pathology data amongst other, both much sought after datasets previous inaccessible for linkage.

CONCLUSION

The Australian experience suggests privacy preserving linkage is a practical solution for improving data access for policy, planning and population health research. It is hoped interest in this methodology internationally continues to grow.

Clinical activity in general practice before sarcoma diagnosis: an Australian cohort study

BACKGROUND

Increased time to diagnosis in sarcoma is associated with poor prognosis and patient outcomes. Research is needed to identify whether opportunities to expedite the diagnosis of sarcoma in general practice exist.

AIM

To examine pre-diagnostic GP clinical activity before sarcoma diagnosis.

DESIGN AND SETTING

An Australian retrospective cohort study using hospital registry data (Australian Comprehensive Cancer Outcomes and Research Database [ACCORD]) linked to two primary care datasets (Patron and MedicineInsight).

METHOD

The frequency of general practice healthcare utilisation events (general practice attendances, prescriptions, blood test, and imaging requests) were compared in 377 patients with soft tissue sarcoma (STS) and 64 patients with bone sarcoma (BS) in the year pre-diagnosis. Poisson regression models were used to calculate monthly incidence rate ratios (IRR) for the 24 months pre-diagnosis and estimate inflection points for when healthcare use started to increase from baseline.

RESULTS

In the 6 months pre-diagnosis, patients with sarcoma had a median of 3-4 general practice attendances, around one-third had a GP imaging request (33% [n = 21] BS and 36% [n = 134] STS), and approximately one in five had multiple imaging requests (19% [n = 12] BS and 21% [n = 80] STS). GP imaging requests progressively increased up to eight-fold from 6 months before sarcoma diagnosis (IRR 8.43, 95% confidence interval [CI] = 3.92 to 18.15, P<0.001) and general practice attendances increased from 3 months pre-diagnosis.

CONCLUSION

Patients with sarcoma have increased GP clinical activity from 6 months pre-diagnosis, indicating a diagnostic window where potential opportunities exist for earlier diagnosis. Interventions to help identify patients and promote appropriate use of imaging and direct specialist centre referrals could improve earlier diagnosis and patient outcomes.

Comparing secondary prevention for patients with coronary heart disease and stroke attending Australian general practices: a cross-sectional study using nationwide electronic database

OBJECTIVES

To compare secondary prevention care for patients with coronary heart disease (CHD) and stroke, exploring particularly the influences due to frequency and regularity of primary care visits.

SETTING

Secondary prevention for patients (≥18 years) in the National Prescription Service administrative electronic health record database collated from 458 Australian general practice sites across all states and territories.

DESIGN

Retrospective cross-sectional and panel study. Patient and care-level characteristics were compared for differing CHD/stroke diagnoses. Associations between the type of cardiovascular diagnosis and medication prescription as well as risk factor assessment were examined using multivariable logistic regression.

PARTICIPANTS

Patients with three or more general practice encounters within 2 years of their latest visit during 2016-2020.

OUTCOME MEASURES

Proportions and odds ratios (ORs) for (1) prescription of antihypertensives, antilipidaemics and antiplatelets and (2) assessment of blood pressure (BP) and low-density lipoprotein cholesterol (LDL-C) in patients with stroke only compared against those with CHD only and those with both conditions.

RESULTS

There were 111 892 patients with CHD only, 27 863 with stroke only and 9791 with both conditions. Relative to patients with CHD, patients with stroke were underprescribed antihypertensives (70.8% vs 82.8%), antilipidaemics (63.1% vs 78.7%) and antiplatelets (42.2% vs 45.7%). With sociodemographic factors, comorbidities and level of care considered as covariates, the odds of non-prescription of any recommended secondary prevention medications were higher in patients with stroke only (adjusted OR 1.37; 95% CI (1.31, 1.44)) compared with patients with CHD only. Patients with stroke only were also more likely to have neither BP nor LDL-C monitored (adjusted OR 1.26; 95% CI (1.18, 1.34)). Frequent and regular general practitioner encounters were independently associated with the prescription of secondary prevention medications (p<0.001).

CONCLUSIONS

Secondary prevention management is suboptimal in cardiovascular disease patients and worse post-stroke compared with post-CHD. More frequent and regular primary care encounters were associated with improved secondary prevention.

Herpes zoster vaccination and new diagnoses of dementia: A quasi-randomized study in Australia

Increasing evidence suggests that neurotropic herpesviruses could play a role in the development of dementia, possibly through a neuroinflammatory process. Herpes zoster (HZ) vaccination has been reported to lead to a reduced probability of being diagnosed with dementia in several correlational studies and in a prior analysis by our team in Wales. This present study constitutes the first investigation to use a quasi-randomized study design in an electronic health record dataset from a large and diverse nation (Australia) to aim to determine the effect of HZ vaccination on dementia. In Australia, starting on November 1 2016, live-attenuated HZ vaccination was provided for free to individuals aged 70 to 79 years of age through primary care providers. Thus, those whose 80th birthday was just a few days prior to November 1 2016 never became eligible, whereas those whose 80th birthday was just a few days later were eligible. The key advantage of our approach is that one would not expect that these population groups who differ in their age by only a minute degree would, on average, differ in any of their health characteristics and behaviors. We used detailed primary healthcare records with week-of-birth information from 65 general practices across Australia. We analyzed our data using a regression discontinuity approach. Our sample consisted of 101,219 patients. As expected, patients born just before versus shortly after the date-of-birth eligibility threshold (November 2 1936) for HZ vaccination were well-balanced in their past preventive health services uptake and chronic disease diagnoses. There was an abrupt increase of 15.7 (95% CI: [12.2 - 19.3], p < 0.001) percentage points in the probability of ever receiving HZ vaccination between patients born shortly before versus shortly after the eligibility threshold. The eligibility rules of the HZ vaccination program, thus, created comparison groups just on either side of the date-of-birth eligibility threshold who were similar to each other, except for a large difference in their probability of receiving the intervention (HZ vaccination) of interest. Eligibility for HZ vaccination (i.e., being born shortly before versus shortly after November 2 1936) decreased the probability of receiving a new dementia diagnosis over 7.4 years by 2.0 percentage points (95% CI: [0.3 - 3.7], p = 0.021). Being eligible for HZ vaccination did not affect the probability of taking up other preventive health services (including other vaccinations), nor the probability of being diagnosed with other common chronic conditions than dementia. This study provides important evidence on the potential benefits of HZ vaccination for dementia because its quasi-randomized design allows for conclusions that are more likely to be causal than those of the existing associational evidence.

Trends in clinical management of lactational mastitis among women attending Australian general practice: a national longitudinal study using MedicineInsight, 2011-2022

OBJECTIVE

To examine longitudinal trends in clinical management of lactational mastitis in women attending general practice.

DESIGN

Open cohort study.

SETTING

Australian general practice using data from MedicineInsight.

PARTICIPANTS

Women aged 18 to 44 years with one or more clinical encounters for lactational mastitis between January 2011 and July 2022.

PRIMARY AND SECONDARY OUTCOME MEASURES

The primary outcome measure was the proportion of prescribed oral antibiotics based on the antibiotic type. Secondary outcome measures were the proportion of women prescribed other medications (eg, antifungals, lactation suppressants) or ordered selected clinical investigations including breast ultrasound, blood test, breast milk culture, nipple swab culture or breast aspirate. Outcomes were examined based on the calendar year and individual- or clinical practice-level characteristics.

RESULTS

Among 25 002 women who had one or more clinical encounters related to mastitis, 90.9% were prescribed oral antibiotics. While the proportion of women prescribed an oral antibiotic remained consistent from 2011 to 2022 (91.1% vs 92.5%), there were changes in the proportion receiving prescriptions for di/flucloxacillin (46.1% vs 60.4%) and cefalexin (38.6% vs 26.5%). Fewer than 12% of women were clinically investigated for their mastitis encounter, most commonly a breast ultrasound (7.1%), followed by a selected blood test (3.8%). Requests for breast milk cultures, nipple swab cultures or breast aspirates occurred in less than 1.1% of individuals. Significant increases were evident with respect to ordering of all clinical investigations, with rates at least doubling between 2011 and 2022 (6.6% vs 14.7%). Large variability in clinical management was evident according to both individual- (eg, concessional status) and clinical practice-level characteristics (eg, remoteness).

CONCLUSIONS

Australian general practitioners commonly prescribe oral antibiotics to women with mastitis and largely in line with clinical guidelines. Their use of clinical investigations as part of mastitis management has increased over the last decade.

Uptake of pneumococcal vaccines in older Australian adults before and after universal public funding of PCV13

BACKGROUND

In 2020 Australia changed the funded universal older adult pneumococcal vaccination program from use of the 23-valent pneumococcal polysaccharide vaccine (PPV23) at age 65 to the 13-valent pneumococcal conjugate vaccine (PCV13) at age 70 years. We investigated uptake of both PCV13 and PPV23 in older adults before and after the program change.

METHODS

We analysed a national dataset of records of patients attending general practices (GPs). We included regular attendees aged 65 or above in 2020. Cumulative uptake of PCV13 and monthly uptake of PPV23 was compared for the two periods before (January 2019 to June 2020) and after (July 2020 to May 2021) the program change on 1 July 2020, by age groups and presence of comorbid conditions.

RESULTS

Our study included data from 192,508 patients (mean age in 2020: 75.1 years, 54.2 % female, 46.1 % with at least one comorbidity). Before July 2020, for all adults regardless of underlying comorbidities, the cumulative uptake of PCV13 was < 1 % but by May 2021, eleven months after the program changes, cumulative uptake of PCV13 had increased among those aged 70-79 years (without comorbidity: 16.3 %; with comorbidity: 21.1 %) and 80 + years (without comorbidity: 13.5 %; with comorbidity: 17.7 %), but not among those aged 65-69 years (without comorbidity: 1.3 %; with comorbidity: 3 %). Monthly uptake of PPV23 dropped following the program change across all age groups.

CONCLUSIONS

Changes in uptake of PCV13 and PPV23 among those aged 70 + years were consistent with program changes. However, PCV13 uptake was still substantially lower in individuals aged 65-69 years overall and in those with comorbidities.

Factors influencing blood pressure control in patients with atrial fibrillation and hypertension in Australian primary care

OBJECTIVE

This study explored factors that may influence blood pressure (BP) control in patients with atrial fibrillation (AF) with hypertension.

METHODS

Cross-sectional retrospective analysis of the MedicineInsight database which includes de-identified electronic health records from general practices (GPs) across Australia. BP control was assessed in patients with diagnosed AF and hypertension (controlled BP defined as <140/90 mm Hg). We explored BP control, factors influencing BP control and likelihood of receiving guideline-recommended treatment.

RESULTS

34 815 patients with AF and hypertension were included; mean age was 76.9 (10.2 SD) years and 46.2% were female. 38.0% had uncontrolled BP. Women (OR 0.72; 95% CI 0.68, 0.76; p<0.001) and adults ≥75 years (OR 0.78; 95% CI 0.70, 0.86; p<0.001) were less likely to have controlled BP. Greater continuity of care (CoC; that is, visits with the same clinician) and having frequent GP visits were associated with higher odds of controlled BP (model 1: CoC, OR 1.29; 95% CI 1.20, 1.40, p<0.001; GP visits, OR 1.71; 95% CI 1.58, 1.85, p<0.001) and a greater likelihood of being prescribed ≥2 types of BP-lowering medicines (model 2: CoC, OR 1.12; 95% CI 1.03, 1.23; p=0.011; GP visits, OR 1.80; 95% CI 1.63, 1.98; p<0.001).

CONCLUSIONS

Uncontrolled BP was more likely in women and adults ≥75 years. Patients who had frequent GP visits with the same clinician were more likely to have BP controlled and receive guideline-recommended antihypertensive treatment. This suggests that targeting these primary care factors could potentially improve BP control and subsequently reduce stroke risk in patients with AF.

Will a fee-for-service payment for a young people's health assessment in general practice increase the detection of health risk behaviours and health conditions? Protocol for a cluster randomised controlled trial (RAd Health Trial)

INTRODUCTION

Adolescence is a period of major transition in physical, cognitive, social and emotional development, and the peak time for the onset of mental health conditions, substance use disorders and sexual and reproductive health risks. Prevention and treatment during this time can improve health and well-being now and into the future. However, despite clinical guidelines recommending annual preventive health assessments for young people, health professionals cite lack of consultation time and adequate funding as key barriers. This trial aims to determine whether a specific fee-for-service ('rebate payment') for a young person's health assessment, is effective and cost-effective at increasing the detection and management of health risk behaviours and conditions among young people.

METHODS AND ANALYSIS

This cluster randomised controlled trial will be conducted in Australian general practice. 42 general practices (clusters) will be randomly allocated 1:1 to either an intervention arm where general practitioners receive a rebate payment for each annual health assessment undertaken for 14-24-year-olds during a 2 year study period, or a control arm (no rebate). The rebate amount will be based on the Medical Benefits Schedule (Australia's list of health professional services subsidised by the Australian Government) currently available for similar age-based assessments. Our primary outcome will be the annual rate of risk behaviours and health conditions recorded in the patient electronic health record (eg, alcohol/drug use, sexual activity and mental health issues). Secondary outcomes include the annual rate of patient management activities related to health risks and conditions identified (eg, contraception prescribed, sexually transmitted infection tests ordered). A process evaluation will assess acceptability, adoption, fidelity and sustainability of the rebate; an economic evaluation will assess its cost-effectiveness. Analyses will be intention-to-treat.

ETHICS AND DISSEMINATION

Ethics approval has been obtained from University of Melbourne Human and Research Ethics Committee (2022-23435-29990-3). Findings will be published in peer-reviewed journals.

TRIAL REGISTRATION NUMBER

ACTRN12622000114741.

Diabetes mellitus monitoring and control among adults in Australian general practice: a national retrospective cohort study

OBJECTIVES

This study investigated whether the monitoring and control of clinical parameters are better among patients with newly compared with past recorded diabetes diagnosis.

DESIGN

Retrospective cohort study.

SETTING

MedicineInsight, a national general practice database in Australia.

PARTICIPANTS

101 875 'regular' adults aged 18+ years with past recorded (2015-2016) and 9236 with newly recorded (2017) diabetes diagnosis.

MAIN OUTCOME MEASURES

Two different groups of outcomes were assessed in 2018. The first group of outcomes was the proportion of patients with clinical parameters (ie, glycated haemoglobin A1c (HbA1c), blood pressure (BP), total cholesterol, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol, triglycerides, estimated glomerular filtration rate and albumin-to-creatinine ratio) monitored at least once in 2018. The second group of outcomes were those related to diabetes control in 2018 (HbA1c ≤7.0%, (BP) ≤140/90 mm Hg, total cholesterol <4.0 mmol/L and LDL-C <2.0 mmol/L). Adjusted ORs (OR_adj) and adjusted probabilities (%) were obtained based on logistic regression models adjusted for practice variables and patients' socio-demographic and clinical characteristics.

RESULTS

The study included 111 111 patients (51.7% men; mean age 65.3±15.0 years) with recorded diabetes diagnosis (11.0% of all 1 007 714 adults in the database). HbA1c was monitored in 39.2% (95% CI 36.9% to 41.6%) of patients with newly recorded and 45.2% (95% CI 42.6% to 47.8%) with past recorded diabetes (OR_adj 0.78, 95% CI 0.73 to 0.82). HbA1c control was achieved by 78.4% (95% CI 76.7% to 80.0%) and 54.4% (95% CI 53.4% to 55.4%) of monitored patients with newly or past recorded diabetes, respectively (OR_adj 3.11, 95% CI 2.82 to 3.39). Less than 20% of patients with newly or past recorded diabetes had their HbA1c, BP and total cholesterol levels controlled (OR_adj 1.08, 95% CI 0.97 to 1.21).

CONCLUSIONS

The monitoring of clinical parameters was lower among patients with newly than past recorded diabetes. However, diabetes control was similarly low in both groups, with only one in five monitored patients achieving control of all clinical parameters.

The Feasibility of Deriving the Electronic Frailty Index from Australian General Practice Records

Purpose

Frailty is a prevalent condition in older adults. Identification of frailty using an electronic Frailty Index (eFI) has been successfully implemented across general practices in the United Kingdom. However, in Australia, the eFI remains understudied. Therefore, we aimed to (i) examine the feasibility of deriving an eFI from Australian general practice records and (ii) describe the prevalence of frailty as measured by the eFI and the prevalence with socioeconomic status and geographic remoteness.

Participants and Methods

This retrospective analysis included patients (≥70 years) attending any one of >700 general practices utilizing the Australian MedicineInsight data platform, 2017–2018. A 36-item eFI was derived using standard methodology, with frailty classified as mild (scores 0.13–0.24); moderate (0.25–0.36) or severe (≥0.37). Socioeconomic status (Socio-Economic Indexes for Areas (SEIFA) index)) and geographic remoteness (Australian Statistical Geography Standard (ASGC) remoteness areas) were also examined.

Results

In total, 79,251 patients (56% female) were included, mean age 80.0 years (SD 6.5); 37.4% (95% CI 37.0–37.7) were mildly frail, 16.7% (95% CI 16.4–16.9) moderately frail, 4.8% (95% CI 4.7–5.0) severely frail. Median eFI score was 0.14 (IQR 0.08 to 0.22); maximum eFI score was 0.69. Across all age groups, moderate and severe frailty was significantly more prevalent in females (P < 0.001). Frailty severity increased with increasing age (P < 0.001) and was strongly associated with socioeconomic disadvantage (P < 0.001) but not with geographic remoteness.

Conclusion

Frailty was identifiable from routinely collected general practice data. Frailty was more prevalent in socioeconomically disadvantaged groups, women and older patients and existed in all levels of remoteness. Routine implementation of an eFI could inform interventions to prevent or reduce frailty in all older adults, regardless of location.

Incidence of Osteoporosis in Primary Care Patients with Atrial Fibrillation Receiving Different Oral Anticoagulants

Background: Studies investigating the association between the use of oral anticoagulants (OACs) and osteoporosis are limited. We aimed to determine the risk of osteoporosis in patients with atrial fibrillation (AF) and receiving different OACs. Methods: We performed a population-based cohort study using a nationwide primary care dataset, MedicineInsight. Patients aged between 18 and 111 years with AF and newly recorded OAC prescriptions between 1 January 2013 and 31 December 2017 were included and followed until 31 December 2018. We applied propensity score matching to control for patients’ baseline characteristic differences before calculating adjusted hazard ratios (aHRs) for a new diagnosis of osteoporosis, using Cox proportional hazard models. Results: A total of 18,454 patients (1714 prescribed dabigatran, 5871 rivaroxaban, 5248 apixaban and 5621 warfarin) were included. Of these, 39.5% were females, and the overall mean age (standard deviation [SD] was 73.2(10.3) years. Over a mean follow-up of 841 days, 1627 patients (1028 receiving direct-acting oral anticoagulants (DOACs) and 599 warfarin) had a newly recorded diagnosis of osteoporosis. The weighted incidence rates (95% confidence interval; CI) per 100 person-years of treatment were 5.0 (4.7−5.2) for warfarin, 4.3 (3.8−4.8) for dabigatran, 3.6 (3.3−3.8) for rivaroxaban, and 4.4 (4.0−4.7) for apixaban. Overall, DOAC use was associated with a significantly lower risk of a new diagnosis of osteoporosis than warfarin use (aHR, 0.79, 95% confidence interval (CI) 0.74−0.85; p < 0.001). Use of each individual DOAC was associated with a significantly lower risk of osteoporosis compared with warfarin (aHRs, 0.75, 95% CI 0.69−0.82 for rivaroxaban; 0.78, 95% CI 0.71−0.86 for apixaban; 0.88, 95% CI 0.77−0.99 for dabigatran). Conclusion: Compared with warfarin, the use of DOACs was associated with a significantly lower risk of developing osteoporosis in patients with AF. This association remained significant when individual DOACs were compared with warfarin.

Prevalence of myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) in Australian primary care patients: only part of the story?

Background

ME/CFS is a disorder characterized by recurrent fatigue and intolerance to exertion which manifests as profound post-exertional malaise. Prevalence studies internationally have reported highly variable results due to the 20 + diagnostic criteria. For Australia, the prevalence of ME/CFS based on current case definitions is unknown.

Objectives

To report prevalence of ME/CFS in patients aged ≥ 13 years attending Australian primary care settings for years 2015–2019, and provide context for patterns of primary care attendance by people living with ME/CFS.

Methodology

Conducted in partnership with the Patient Advisory Group, this study adopted a mixed methods approach. De-identified primary care data from the national MedicineInsight program were analyzed. The cohort were regularly attending patients, i.e. 3 visits in the preceding 2 years. Crude prevalence rates were calculated for years 2015–2019, by sex, 10-year age groups, remoteness and socioeconomic status. Rates are presented per 100,000population (95% confidence intervals (CI)). Qualitative data was collected through focus groups and in-depth 1:1 interview.

Results

Qualitative evidence identified barriers to reaching diagnosis, and limited interactions with primary care due to a lack of available treatments/interventions, stigma and disbelief in ME/CFS as a condition.

In each year of interest, crude prevalence in the primary care setting ranged between 94.9/100,000 (95% CI: 91.5–98.5) and 103.9/100,000 population (95%CI: 100.3–107.7), equating to between 20,140 and 22,050 people living with ME/CFS in Australia in 2020. Higher rates were observed for age groups 50-59 years and 40-49 years. Rates were substantially higher in females (130.0–141.4/100,000) compared to males (50.9–57.5/100,000). In the context of the qualitative evidence, our prevalence rates likely represent an underestimate of the true prevalence of ME/CFS in the Australian primary care setting.

Conclusion

ME/CFS affects a substantial number of Australians. Whilst this study provides prevalence estimates for the Australian primary care setting, the qualitative evidence highlights the limitations of these. Future research should focus on using robust case ascertainment criteria in a community setting. Quantification of the burden of disease can be used to inform health policy and planning, for this understudied condition.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-022-13929-9.

Predicted cardiovascular disease risk and prescribing of antihypertensive therapy among patients with hypertension in Australia using MedicineInsight

Hypertension guidelines recommend that absolute cardiovascular disease (CVD) risk guide the management of hypertensive patients. This study aimed to assess the proportion of patients with diagnosed hypertension with sufficient data to calculate absolute CVD risk and determine whether CVD risk is associated with prescribing of antihypertensive therapies. This was a cross-sectional study using a large national database of electronic medical records of patients attending general practice in 2018 (MedicineInsight). Of 571,492 patients aged 45-74 years without a history of CVD, 251,733 [40.6% (95% CI: 39.8-41.2)] had a recorded hypertension diagnosis. The proportion of patients with sufficient recorded data available to calculate CVD risk was higher for patients diagnosed with hypertension [51.0% (95% CI: 48.0-53.9)] than for patients without a diagnosis of hypertension [38.7% (95% CI: 36.5-41.0)]. Of those patients with sufficient data to calculate CVD risk, 29.3% (95% CI: 28.1-30.6) were at high risk clinically, 6.0% (95% CI: 5.8-6.3) were at high risk based on their CVD risk score, 12.8% (95% CI: 12.5-13.2) at moderate risk and 51.8% (95% CI: 50.8-52.9) at low risk. The overall prevalence of antihypertensive therapy was 60.9% (95% CI: 59.3-62.5). Prescribing was slightly lower in patients at high risk based on their CVD risk score [57.4% (95% CI: 55.4-59.4)] compared with those at low [63.3% (95% CI: 61.9-64.8)] or moderate risk [61.8% (95% CI: 60.2-63.4)] or at high risk clinically [64.1% (95% CI: 61.9-66.3)]. Guideline adherence is suboptimal, and many patients miss out on treatments that may prevent future CVD events.

Comparing the renal outcomes in patients with atrial fibrillation receiving different oral anticoagulants

BACKGROUND

We aimed to compare renal function changes in patients with atrial fibrillation (AF) prescribed different oral anticoagulants (OACs).

RESEARCH DESIGN AND METHODS

We performed a retrospective analysis of Australian national primary care data. A total of 12,562 patients with AF and initiated OAC between 1 Jan 2013 and 31 Dec 2017 were included. Inverse probability of treatment weighting was used for balancing baseline characteristics and the risks of decline in estimated glomerular filtration rate (eGFR) in patients prescribed each OAC were compared using Cox proportional hazards regression model.

RESULTS

Compared with warfarin, prescribing of direct-acting oral anticoagulants (DOACs) was associated with a lower risk of renal function decline per 1000 person-years: hazard ratio (HR) 0.75, 95% confidence interval (CI) 0.68-0.81, p<0.001 for ≥30% decline in eGFR; HR 0.28, 95% CI 0.20-0.41, p<0.001 for eGFR decline to ≤30 mL/min/1.73m²; and HR 0.45, 95% CI 0.35-0.58, p<0.001 for serum creatinine doubling. Compared with dabigatran, rivaroxaban use had a significantly lowered risk of decline in eGFR to ≤30 mL/min/1.73m² (HR 0.29, 95% CI 0.13-0.66, p=0.003) and risk of doubling of serum creatinine (HR 0.62, 95% CI 0.40-0.95, p=0.030).

CONCLUSIONS

The risk of renal function decline appeared to be lower in patients prescribed DOACs versus warfarin.

Health Statistics in Australia: What We Know and Do Not Know

Australia is a federation of six states and two territories (the States). These eight governmental entities share responsibility for health and health services with the Australian Government. Mortality statistics, including causes of death, have been collected since the late 19th century, with national data produced by the (now) Australian Bureau of Statistics (ABS) from 1907. Each State introduced hospital in-patient statistics, assisted by State offices of the ABS. Beginning in the 1970s, the ABS conducts regular health surveys, including specific collections on Aboriginal and Torres Strait Islander peoples. Overall, Australia now has a comprehensive array of health statistics, published regularly without political or commercial interference. Privacy and confidentiality are guaranteed by legislation. Data linkage has grown and become widespread. However, there are gaps, as papers in this issue demonstrate. Most notably, data on primary care patients and encounters reveal stark gaps. This paper accompanies a range of papers from expert authors across the health statistics spectrum in Australia. It is hoped that the collection of papers will inform interested readers and stand as a comprehensive review of the strengths and weaknesses of Australian health statistics in the early 2020s.

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

Aims

To investigate the epidemiology of diabetes diagnosis and screening in Australian general practice.

Methods

Cross-sectional study using electronic health records of 1,522,622 patients aged 18+ years attending 544 Australian general practices (MedicineInsight database). The prevalence of diagnosed diabetes and diabetes screening was explored using all recorded diagnoses, laboratory results, and prescriptions between 2016 and 2018. Their relationship with patient sociodemographic and clinical characteristics was also investigated.

Results

Overall, 7.5% (95% CI 7.3, 7.8) of adults had diabetes diagnosis, 0.7% (95% CI 0.6, 0.7) prediabetes, and 0.3% (95% CI 0.3, 0.3) unrecorded diabetes/prediabetes (elevated glucose levels without a recorded diagnosis). Patients with unrecorded diabetes/prediabetes had clinical characteristics similar to those with recorded diabetes, except for a lower prevalence of overweight/obesity (55.5% and 69.9%, respectively). Dyslipidaemia was 1.8 times higher (36.2% vs. 19.7%), and hypertension was 15% more likely (38.6% vs. 33.8%) among patients with prediabetes than with diabetes. Diabetes screening (last three years) among people at high risk of diabetes was 55.2% (95% CI 52.7, 57.7), with lower rates among young or elderly males.

Conclusions

Unrecorded diabetes/prediabetes is infrequent in Australian general practice, but prediabetes diagnosis was also lower than expected. Diabetes screening among high-risk individuals can be improved, especially in men, to enhance earlier diabetes diagnosis and management.

Characteristics of patients aged 50-74 years with a request for an immunochemical faecal occult blood test in the Australian general practice setting

ObjectiveTo support improving participation in the National Bowel Cancer Screening Program (NBCSP), we aimed to identify Medicare-subsidised test requests for immunochemical faecal occult blood tests (FOBT) in Australian general practice for patients aged 50-74 years, eligible for the NBCSP, and describe sociodemographics, risk factors, indications and outcomes.MethodsA cross-sectional study was conducted using de-identified data from 441 Australian general practice sites in the MedicineInsight database, recorded from 1 January 2018 to 31 December 2019.ResultsOf the 683 625 eligible patients, 45 771 (6.7%) had a record of a general practitioner (GP)-requested FOBT, either to aid diagnosis in symptomatic patients, or for screening; 144 986 (21.2%) patients had only an NBCSP FOBT. A diagnosis of polyps, gastrointestinal inflammatory condition or haemorrhoids, or a referral to a gastroenterologist or general surgeon, was more commonly recorded in the 6 months after a GP-requested FOBT than after an NBCSP FOBT. Uptake of NBCSP FOBTs was lower among those with obesity, high alcohol consumption and current smokers, who are at higher risk of bowel cancer.ConclusionsThis study describes the patient characteristics, reasons and outcomes associated with GP-requested FOBTs, identifies under-screened population sub-groups, and suggests involvement of GPs to improve participation in the NBCSP.

General Practice Statistics in Australia: Pushing a Round Peg into a Square Hole

In Australia, general practice forms a core part of the health system, with general practitioners (GPs) having a gatekeeper role for patients to receive care from other health services. GPs manage the care of patients across their lifespan and have roles in preventive health care, chronic condition management, multimorbidity and population health. Most people in Australia see a GP once in any given year. Draft reforms have been released by the Australian Government that may change the model of general practice currently implemented in Australia. In order to quantify the impact and effectiveness of any implemented reforms in the future, reliable and valid data about general practice clinical activity over time, will be needed. In this context, this commentary outlines the historical and current approaches used to obtain general practice statistics in Australia and highlights the benefits and limitations of these approaches. The role of data generated from GP electronic health record extractions is discussed. A methodology to generate high quality statistics from Australian general practice in the future is presented.

Associations between regular GP contact, diabetes monitoring and glucose control: an observational study using general practice data

OBJECTIVE

Continuity and regularity of general practitioner (GP) contacts are associated with reduced hospitalisation in type 2 diabetes (T2DM). We assessed associations of these GP contact patterns with intermediate outcomes reflecting patient monitoring and health.

DESIGN

Observational longitudinal cohort study using general practice data 2011-2017.

SETTING

193 Australian general practices in Western Australia and New South Wales participating in the MedicineInsight programme run by NPS MedicineWise.

PARTICIPANTS

22 791 patients aged 18 and above with T2DM.

INTERVENTIONS

Regularity was assessed based on variation in the number of days between GP visits, with more regular contacts assumed to indicate planned, proactive care. Informational continuity (claims for care planning incentives) and relational continuity (usual provider of care index) were assessed separately.

OUTCOME MEASURES

Process of care indicators were glycosylated haemoglobin (HbA1c) test underuse (8 months without test), estimated glomerular filtration rate (eGFR) underuse (14 months) and HbA1c overuse (two tests within 80 days). The clinical indicator was T2DM control (HbA1c 6.5% (47.5 mmol/mol)-7.5% (58.5 mmol/mol)).

RESULTS

The quintile with most regular contact had reduced odds of HbA1c and eGFR underuse (OR 0.74, 95% CI 0.67 to 0.81 and OR 0.78, 95% CI 0.70 to 0.86, respectively), but increased odds of HbA1c overuse (OR 1.20, 95% CI 1.05 to 1.38). Informational continuity was associated with reduced odds of HbA1c underuse (OR 0.53, 95% CI 0.49 to 0.56), reduced eGFR underuse (OR 0.62, 95% CI 0.58 to 0.67) and higher odds of HbA1c overuse (OR 1.48, 95% CI 1.34 to 1.64). Neither had significant associations with HbA1c level. Results for relational continuity differed.

CONCLUSIONS

This study provides evidence that regularity and continuity influence processes of care in the management of patients with diabetes, though this did not result in the recording of HbA1c within target range. Research should capture these intermediate outcomes to better understand how GP contact patterns may influence health rather than solely assessing associations with hospitalisation outcomes.

Assessing the suitability of general practice electronic health records for clinical prediction model development: a data quality assessment

BACKGROUND

The use of general practice electronic health records (EHRs) for research purposes is in its infancy in Australia. Given these data were collected for clinical purposes, questions remain around data quality and whether these data are suitable for use in prediction model development. In this study we assess the quality of data recorded in 201,462 patient EHRs from 483 Australian general practices to determine its usefulness in the development of a clinical prediction model for total knee replacement (TKR) surgery in patients with osteoarthritis (OA).

METHODS

Variables to be used in model development were assessed for completeness and plausibility. Accuracy for the outcome and competing risk were assessed through record level linkage with two gold standard national registries, Australian Orthopaedic Association National Joint Replacement Registry (AOANJRR) and National Death Index (NDI). The validity of the EHR data was tested using participant characteristics from the 2014-15 Australian National Health Survey (NHS).

RESULTS

There were substantial missing data for body mass index and weight gain between early adulthood and middle age. TKR and death were recorded with good accuracy, however, year of TKR, year of death and side of TKR were poorly recorded. Patient characteristics recorded in the EHR were comparable to participant characteristics from the NHS, except for OA medication and metastatic solid tumour.

CONCLUSIONS

In this study, data relating to the outcome, competing risk and two predictors were unfit for prediction model development. This study highlights the need for more accurate and complete recording of patient data within EHRs if these data are to be used to develop clinical prediction models. Data linkage with other gold standard data sets/registries may in the meantime help overcome some of the current data quality challenges in general practice EHRs when developing prediction models.

Patterns of care for people presenting to Australian general practice with musculoskeletal complaints based on routinely collected data: protocol for an observational cohort study using the Population Level Analysis and Reporting (POLAR) database

INTRODUCTION

General practice is integral to the Australian healthcare system. Outcome Health's POpulation Level Analysis and Reporting (POLAR) database uses de-identified electronic health records to analyse general practice data in Australia. Previous studies using routinely collected health data for research have not consistently reported the codes and algorithms used to describe the population, exposures, interventions and outcomes in sufficient detail to allow replication. This paper reports a study protocol investigating patterns of care for people presenting with musculoskeletal conditions to general practice in Victoria, Australia. Its focus is on the systematic approach used to classify and select eligible records from the POLAR database to facilitate replication. This will be useful for other researchers using routinely collected health data for research.

METHODS AND ANALYSIS

This is a retrospective cohort study. Patient-related data will be obtained through electronic health records from a subset of general practices across three primary health networks (PHN) in southeastern Victoria. Data for patients with a low back, neck, shoulder and/or knee condition and who received at least one general practitioner (GP) face-to-face consultation between 1 January 2014 and 31 December 2018 will be included. Data quality checks will be conducted to exclude patients with poor data recording and/or non-continuous follow-up. Relational data files with eligible and valid records will be merged to select the study cohort and the GP care received (consultations, imaging requests, prescriptions and referrals) between diagnosis and 31 December 2018. Number and characteristics of patients and GPs, and number, type and timing of imaging requests, prescriptions for pain relief and referrals to other health providers will be investigated.

ETHICS AND DISSEMINATION

Ethics approval was obtained from the Cabrini and Monash University Human Research Ethics Committees (Reference Numbers 02-21-01-19 and 16975, respectively). Study findings will be reported to Outcome Health, participating PHNs, disseminated in academic journals and presented in conferences.

The changing characteristics of patients with chronic hepatitis C prescribed direct acting antiviral medicines in general practice since listing of the medicines on the Australian Pharmaceutical Benefits Scheme

Background and Aim

The primary objective of this study was to determine whether the characteristics of patients prescribed direct acting antiviral (DAA) medicines have changed since initial listing of the medicines on the Australian Pharmaceutical Benefits Scheme (PBS).

Methods

A cross‐sectional study was conducted using data from MedicineInsight, an Australian database of general practice electronic health records, from March 2016 to August 2018. We compared sociodemographic, comorbidity, and clinical characteristics of patients aged at least 18 years who were prescribed at least one DAA in the first 4 months of PBS listing in 2016 with those prescribed at least one DAA in 2018.

Results

There were 2251 eligible adult patients prescribed a DAA during the study period, 62% were men and 59% were aged 50 years and older. Patients prescribed DAA medicines initially were older (aged ≥50 years: 67.9% vs 49.3%; P < 0.001), and more likely to have liver cirrhosis (14.2% vs 8.4%; P = 0.01) and an aminotransferase to platelet ratio index (APRI) score >1 (20.4% vs 8.9%; P < 0.001) than those prescribed DAA medicines in 2018. A greater proportion of patients in regional/remote (46.5% vs 35.6%; P < 0.001) and socioeconomically disadvantaged areas (44.4% vs 34.5%; P = 0.003) accessed treatment in 2018 compared with 2016.

Conclusions

Despite evidence of decreasing uptake of DAA medicines across Australia, this study indicates broadened uptake among younger age groups and those residing in regional/remote and socioeconomically disadvantaged areas since 2016. While uptake of DAA medicines in some population subgroups appears to have improved, continuous efforts to improve uptake across the Australian population are essential.

Estimating pertussis incidence in general practice using a large Australian primary care database

BACKGROUND

While pertussis is notifiable in most countries, notifications typically underestimate the true pertussis burden. We explored the incidence of pertussis in general practice in Australia.

METHODS

Using MedicineInsight, a large longitudinal electronic medical record database of general practice (primary care) encounters which includes >1.5 million patients, we first defined a cohort of active patients and then used free-text search algorithms to identify patients with pertussis-related encounters. We defined and identified pertussis-related encounters in four patient categories: pertussis-associated (category 1), potential pertussis (category 2), epidemiologically-linked pertussis (category 3), and symptoms consistent with pertussis (category 4). Incident pertussis-related encounter rates per 100,000 active patients were calculated from Jan 2008 to Aug 2015.

RESULTS

Estimated mean annual pertussis incidence increased as definitions were expanded, from 94.3 (category 1 patients only) to 148.8 (categories 1+2+3 patients combined) per 100,000 active patients per year. Monthly time-series corresponding to the first three categories were highly correlated (Pearson's r > 90% for each pair), but each was poorly correlated with category 4. For categories 1+2+3, the highest incidence was among 0-4 and 5-9 year olds. Incidence was 30% higher in females than males (i.e. 184.5 vs 139.8 per 100,00 active patients for categories 1-3 patients combined). Pertussis-associated incidence (category 1) was similar to national pertussis notification rates. Categories 2 and 3 added 25% and 33%, respectively, on average relative to category 1 incidence. The estimated incidence from categories 1+2+3 together were on average 64% higher than national pertussis notification rates.

CONCLUSION

We provide comprehensive estimates of pertussis-related incidence in general practice (primary care), well in excess of notified pertussis incidence in Australia. This highlights the utility of MedicineInsight data in providing a greater understanding of the burden of medically-attended pertussis infections.

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

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.

Prevalence of inflammatory bowel disease in the Australian general practice population: A cross-sectional study

The burden of inflammatory bowel disease (IBD) in Australia is increasing but national data about the current prevalence are limited. We aimed to estimate the prevalence of IBD (including Crohn’s disease, ulcerative colitis and unspecified IBD) as well as Crohn’s disease and ulcerative colitis separately in a general practice population in Australia. We also assessed risk factors associated with Crohn’s disease and ulcerative colitis. We conducted a cross-sectional study using data from MedicineInsight, a national database of general practice electronic health records, from 1 July 2017 to 30 June 2019. The prevalence of IBD was calculated and stratified by sociodemographic characteristics. Logistic regression analysis was conducted to assess risk factors associated with Crohn’s disease and ulcerative colitis. The study comprised 2,428,461 regular patients from 481 practices. The estimated crude prevalence of IBD was 653 per 100,000 patients; Crohn’s disease was 306 per 100,000 and ulcerative colitis was 334 per 100,000. Males were independently associated with a lower risk of Crohn’s disease (OR: 0.86; 95% CI: 0.81, 0.90) but a greater risk of ulcerative colitis (OR: 1.12; 95% CI: 1.06, 1.17) than females. Compared to non-smokers, patients who were current smokers were associated with a greater risk of Crohn’s disease (OR: 1.13; 95% CI: 1.04, 1.23) but a lower risk of ulcerative colitis (OR: 0.52; 95% CI: 0.47, 0.57). Other factors positively associated with both Crohn’s disease and ulcerative colitis were age (≥ 25 years), non-Indigenous status and socioeconomic advantage. Our findings provide a current estimate of the prevalence of IBD, Crohn’s disease and ulcerative colitis in a large national general practice population in Australia and an assessment of the factors associated with Crohn’s disease and ulcerative colitis. These data can assist in estimating the health burden and costs, and planning for health services.

Trends in the prescription of drugs used for insomnia: an open-cohort study in Australian general practice, 2011-2018

BACKGROUND

Despite an increase in the prevalence of sleep problems, few studies have investigated changes in the prescribing of drugs that are often used to manage insomnia.

AIM

To explore changes in the pattern of benzodiazepine (BZD), Z-drug (zolpidem, zopiclone), and non-BZD prescriptions.

DESIGN AND SETTING

Open-cohort study comprising 1 773 525 patients (55 903 294 consultations) who attended one of 404 Australian general practices at least three times in two consecutive years between 2011 and 2018.

METHOD

Data were extracted from MedicineInsight, a database of 662 general practices in Australia. Prescription rates per 1000 consultations, the proportion of repeat prescriptions above recommendations, and the proportion of prescriptions for patients with a recent (within 2 years) recorded diagnosis of insomnia were analysed using adjusted regression models.

RESULTS

Rates of BZD, Z-drug, and non-BZD prescriptions were 56.6, 4.4, and 15.5 per 1000 consultations in 2011 and 41.8, 3.5, and 21.5 per 1000 consultations in 2018, respectively. Over the whole study period, temazepam represented 25.3% of the prescriptions and diazepam 21.9%. All BZD and zolpidem prescriptions declined over the whole study period (annual change varying from -1.4% to -10.8%), but non-BZD and zopiclone prescriptions increased in the same period (annual change 5.0% to 22.6%). Repeat prescriptions that exceeded recommended levels remained at <10% for all medications, except melatonin (64.5%), zolpidem (63.3%), zopiclone (31.4%), and alprazolam (13.3%). In 2018, >50% of Z-drug and melatonin prescriptions were for patients with insomnia. There was an annual increase of 0.8-5.9% in the proportion of prescriptions associated with a recently recorded diagnosis of insomnia.

CONCLUSION

Overall, BZD prescriptions in Australia declined between 2011 and 2018. However, the prescription of some of these drugs increased for patients with a recently recorded diagnosis of insomnia. This is concerning because of the potential adverse effects of these medications and the risk of dependence.

Trends in long-term opioid prescriptions for musculoskeletal conditions in Australian general practice: a national longitudinal study using MedicineInsight, 2012-2018

OBJECTIVE

Describe trends and patterns in long-term opioid prescriptions among adults with musculoskeletal conditions (MSK).

DESIGN

Interrupted time-series analysis based on an open cohort study.

SETTING

A representative sample of 402 Australian general practices contributing data to the MedicineInsight database.

PARTICIPANTS

811 174 patients aged 18+ years with an MSK diagnosis and three or more consultations in any two consecutive years between 2012 and 2018. Males represented 44.5% of the sample, 28.4% were 65+ years and 1.9% were Aboriginal or Torres Strait Islanders.

PRIMARY AND SECONDARY OUTCOME MEASURES

Annual prevalence and cumulative incidence (%) of long-term opioid prescribing (3+ prescriptions in 90 days) among patients with an MSK. Average duration of these episodes in each year between 2012 and 2018.

RESULTS

The prevalence of long-term opioid prescribing increased from 5.5% (95% CI 5.2 to 5.8) in 2012 to 9.1% (95% CI 8.8 to 9.7) in 2018 (annual change OR 1.09, 95% CI 1.08 to 1.09), but a slightly lower incidence was observed in 2018 (3.0% vs 3.6%-3.8% in other years; annual change OR 0.99, 95% CI 0.98 to 0.99). The incidence was between 37% and 52% higher among practices located in rural Australia or lower socioeconomic areas. Individual risk factors included increasing age (3.4 times higher among those aged 80+ years than the 18-34 years group in 2012, increasing to 4.8 times higher in 2018), identifying as Aboriginal or Torres Strait Islander (1.7-1.9 higher incidence than their peers), or living in disadvantaged areas (36%-57% more likely than among those living in wealthiest areas). Long-term opioid prescriptions lasted in average 287-301 days between 2012 and 2016, reducing to 229 days in 2017 and 140 days in 2018. A longer duration was observed in practices from more disadvantaged areas and females in all years, except in 2018.

CONCLUSIONS

The continued rise in the prevalence of long-term opioid prescribing is of concern, despite a recent reduction in the incidence and duration of opioid management.

Ten-Year Trends in the Use of Oral Anticoagulants in Australian General Practice Patients With Atrial Fibrillation

Objective: Appropriate use of oral anticoagulants (OACs) reduces the risk of stroke in patients with atrial fibrillation (AF). The study characterized the prescribing of OACs in people with AF in the Australian primary care setting over 10 years.

Design: Retrospective population study.

Setting and Participants: We performed 10 sequential cross-sectional analyses of patients with a recorded diagnosis of AF between 2009 and 2018 using national general practice data. The proportion of patients with AF who were prescribed an OAC based on their stroke risk was examined.

Primary and secondary outcomes: The primary outcome was the proportion of high stroke risk patients who were prescribed an OAC over a decade. The secondary outcome was variation in OAC prescribing among general practices.

Results: The sample size of patients with AF ranged from 9,874 in 2009 to 41,751 in 2018. The proportion who were prescribed an OAC increased from 39.5% (95% CI 38.6–40.5%) in 2009 to 52.0% (95% CI 51.5–52.4%) in 2018 (p for trend < 0.001). During this time, the proportion of patients with AF and high stroke risk who were prescribed an OAC rose from 41.7% (95% CI 40.7–42.8%) to 55.2% (95% CI 54.7–55.8%; p for trend < 0.001) with the direct-acting oral anticoagulants accounting for over three-quarters of usage by 2018. There was substantial variation in OAC prescribing between general practices. In 2018, the proportion of moderate to high stroke risk patients who were prescribed an OAC was 38.6% (95% CI 37.2–40.1%) in the lowest practice site quintiles and 65.6% (95% CI 64.5–66.7%) in the highest practice site quintiles.

Conclusions: Over the 10 years, OAC prescribing in high stroke risk patients with AF increased by one-third. There was considerable variation in OAC prescribing between general practices.

Cohort profile: St. Michael's Hospital Tuberculosis Database (SMH-TB), a retrospective cohort of electronic health record data and variables extracted using natural language processing

Background

Tuberculosis (TB) is a major cause of death worldwide. TB research draws heavily on clinical cohorts which can be generated using electronic health records (EHR), but granular information extracted from unstructured EHR data is limited. The St. Michael’s Hospital TB database (SMH-TB) was established to address gaps in EHR-derived TB clinical cohorts and provide researchers and clinicians with detailed, granular data related to TB management and treatment.

Methods

We collected and validated multiple layers of EHR data from the TB outpatient clinic at St. Michael’s Hospital, Toronto, Ontario, Canada to generate the SMH-TB database. SMH-TB contains structured data directly from the EHR, and variables generated using natural language processing (NLP) by extracting relevant information from free-text within clinic, radiology, and other notes. NLP performance was assessed using recall, precision and F₁ score averaged across variable labels. We present characteristics of the cohort population using binomial proportions and 95% confidence intervals (CI), with and without adjusting for NLP misclassification errors.

Results

SMH-TB currently contains retrospective patient data spanning 2011 to 2018, for a total of 3298 patients (N = 3237 with at least 1 associated dictation). Performance of TB diagnosis and medication NLP rulesets surpasses 93% in recall, precision and F₁ metrics, indicating good generalizability. We estimated 20% (95% CI: 18.4–21.2%) were diagnosed with active TB and 46% (95% CI: 43.8–47.2%) were diagnosed with latent TB. After adjusting for potential misclassification, the proportion of patients diagnosed with active and latent TB was 18% (95% CI: 16.8–19.7%) and 40% (95% CI: 37.8–41.6%) respectively

Conclusion

SMH-TB is a unique database that includes a breadth of structured data derived from structured and unstructured EHR data by using NLP rulesets. The data are available for a variety of research applications, such as clinical epidemiology, quality improvement and mathematical modeling studies.

Pregabalin prescribing patterns in Australian general practice, 2012-2018: a cross-sectional study

BACKGROUND

In 2013 pregabalin was subsidised by Australia's Pharmaceutical Benefits Scheme (PBS) for neuropathic pain. Since the subsidy, pregabalin prescribing has been increasing in Australia and so has related harm. There are concerns it is being prescribed for indications other than neuropathic pain, which have little evidence of efficacy.

AIM

To describe pregabalin prescribing in Australian general practice.

DESIGN & SETTING

A cross-sectional study of patients attending 445 general practice sites in the national MedicineInsight database from March 2012-February 2018.

METHOD

The following aspects were calculated: the proportion of prescriptions that were for pregabalin per year; the prevalence of pain conditions in patients prescribed pregabalin; and same-day prescribing of pregabalin with opioids or benzodiazepines.

RESULTS

Prescribing increased from 13 per 10 000 to 104 per 10 000 prescriptions between 2012-2013 and 2017-2018. A total of 1 891 623 patients were identified of whom 114 123 (6.0%) were prescribed pregabalin; 49.7% (n = 56 772) had a recorded diagnosis of neuropathic pain. Among people prescribed pregabalin without a recorded diagnosis of neuropathic pain, 43.5% (n = 24 927) had a diagnosis of back problems, 8.8% (n = 5073) chronic pain, and 26.4% (n = 30 146) had no pain diagnosis. Pregabalin was prescribed the same day as an opioid to 38.1% of patients (95% confidence interval [CI] = 37.1% to 39.1%) and a benzodiazepine to 13.1% of patients (95% CI = 12.5% to 13.7%). Patients with a diagnosis of chronic pain had the highest rate of same-day prescribing of pregabalin with an opioid (70.4%, 95% CI = 68.9% to 71.9%) or a benzodiazepine (25.8%, 95% CI = 24.2% to 27.4%) CONCLUSION: Substantial increases in pregabalin prescribing were identified in Australian general practice, but only half of patients had a neuropathic pain diagnosis recorded, the only approved indication for subsidy. High rates of same-day prescribing with opioids and benzodiazepines may put patients at increased risk of harm.

Associations between multimorbidity and glycaemia (HbA1c) in people with type 2 diabetes: cross-sectional study in Australian general practice

OBJECTIVES

To explore the prevalence of multimorbidity as well as individual and combinations of long-term conditions (LTCs) in people with type 2 diabetes (T2D) attending Australian general practice, using electronic health record (EHR) data. We also examine the association between multimorbidity condition count (total/concordant(T2D related)/discordant(unrelated)) and glycaemia (glycated haemoglobin, HbA1c).

DESIGN

Cross-sectional study.

SETTING

Australian general practice.

PARTICIPANTS

69 718 people with T2D with a general practice encounter between 2013 and 2015 captured in the MedicineInsight database (EHR Data from 557 general practices and >3.8 million Australian patients).

PRIMARY AND SECONDARY OUTCOME MEASURES

Prevalence of multimorbidity, individual and combinations of LTCs. Multivariable linear regression models used to examine associations between multimorbidity counts and HbA1c (%).

RESULTS

Mean (SD) age 66.42 (12.70) years, 46.1% female and mean (SD) HbA1c 7.1 (1.4)%. More than 90% of participants with T2D were living with multimorbidity. Discordant conditions were more prevalent (83.4%) than concordant conditions (69.9 %). The three most prevalent discordant conditions were: painful conditions (55.4%), dyspepsia (31.6%) and depression (22.8%). The three most prevalent concordant conditions were hypertension (61.4%), coronary heart disease (17.1%) and chronic kidney disease (8.5%). The three most common combinations of conditions were: painful conditions and hypertension (38.8%), painful conditions and dyspepsia (23.1%) and hypertension and dyspepsia (22.7%). We found no associations between any multimorbidity counts (total, concordant and discordant) or combinations and HbA1c.

CONCLUSIONS

Multimorbidity was common in our cohort of people with T2D attending Australian general practice, but was not associated with glycaemia. Although we did not explore mortality in this study, our results suggest that the increased mortality in those with multimorbidity and T2D observed in other studies may not be linked to glycaemia. Interestingly, discordant conditions were more prevalent than concordant conditions with painful conditions being the second most common comorbidity. Better understanding of the implications of different patterns of multimorbidity in people with T2D will allow more effective tailored care.

Five-Year Trends in Potential Drug Interactions with Direct-Acting Oral Anticoagulants in Patients with Atrial Fibrillation: An Australian-Wide Study

Background: Co-prescribing medications that can interact with direct-acting oral anticoagulants (DOACs) may decrease their safety and efficacy. The aim of this study was to examine the co-prescribing of such medications with DOACs using the Australian national general practice dataset, MedicineInsight, over a five-year period. Methods: We performed five sequential cross-sectional analyses in patients with atrial fibrillation (AF) and a recorded DOAC prescription. Patients were defined as having a drug interaction if they had a recorded prescription of an interacting medication while they had had a recorded prescription of DOAC in the previous six months. The sample size for the cross-sectional analyses ranged from 5333 in 2014 to 19,196 in 2018. Results: The proportion of patients who had potential drug interactions with a DOAC decreased from 45.9% (95% confidence interval (CI) 44.6%–47.4%) in 2014 to 39.9% (95% CI 39.2%–40.6%) in 2018, p for trend < 0.001. During this period, the most frequent interacting class of medication recorded as having been prescribed with DOACs was selective serotonin/serotonin and norepinephrine reuptake inhibitor (SSRI/SNRI) antidepressants, followed by non-steroidal anti-inflammatory drugs (NSAIDs), calcium channel blockers (CCBs) and amiodarone. Conclusions: Overall, potential drug interactions with DOACs have decreased slightly over the last five years; however, the rate of possible interaction with SSRIs/SNRIs has remained relatively unchanged and warrants awareness-raising amongst prescribers.

Microbiology testing and antibiotic treatment for urinary tract infections in general practice: a nationwide observational study

PURPOSE

Routine urine testing is recommended prior to antibiotic treatment for urinary tract infections (UTIs) among high-risk groups for complicated UTIs. This study aims to examine whether the proportion of UTI encounters where antibiotics are prescribed that have accompanying urine testing differs by patient groups.

METHODS

A retrospective analysis was conducted using records of general practice encounters for UTIs occurring between January 2013 and July 2018 in an Australian national database. We calculated the proportion of UTI encounters with antibiotics prescribed that had accompanying urine microbiology testing and the odds ratios for the likelihood of testing by patient groups using generalised estimating equations.

RESULTS

Of 132,688 UTI encounters with antibiotics prescribed, 95,800 (72.2%) were accompanied by urine testing. Among high-risk groups for complicated UTIs and expected to have a high likelihood of testing, we found pregnant women [82.6% vs. non-pregnant 72.3%, adjusted odds ratio (aOR) 1.82, 95% confidence intervals (CI) 1.55-2.12] and children aged 5-9 years (77.6% vs. 20-44 years 72.0%, aOR 1.33, 95% CI 1.22-1.45) had relatively high odds of testing. However, children aged < 5 years (68.7% vs. 20-44 years 72.0%, aOR 0.83, 95% CI 0.76-0.90), patients with recurrent UTIs (69.0% compared to first-onset UTIs 73.6%, aOR 0.81, 95% CI 0.79-0.83), and patients in residential aged care facilities (67.3% vs. not 72.3%, aOR 0.80, 95% CI 0.72-0.90) had relatively low odds of testing.

CONCLUSION

Our results suggest inconsistencies and potential underuse of urine testing when antibiotics were prescribed for high-risk groups in UTI management. Further antibiotic stewardship is needed to improve guideline-based antibiotic prescribing for UTIs.

Influenza-like illness in Australia: A comparison of general practice surveillance system with electronic medical records

Surveillance systems are fundamental to detect infectious disease outbreaks and guide public health responses. We compared influenza-like illness (ILI) rates for 2015-2017 using data from the Australian Sentinel Practice Research Network (ASPREN) and electronic medical records from 550 general practices across Australia (MedicineInsight). There was a high correlation between both sources (r = .84-.95) and a consistent higher ILI rate in 2017. Both sources also showed higher ILI rates among women and patients aged 20-49 years. The use of routinely collected electronic medical records like those in MedicineInsight could be used to complement active influenza surveillance systems in Australia.

Safety of live attenuated herpes zoster vaccine in adults 70-79 years: A self-controlled case series analysis using primary care data from Australia's MedicineInsight program

BACKGROUND

Australia introduced a funded shingles vaccination program for older adults in November 2016, administered predominantly in primary care clinics. MedicineInsight, a nationally representative primary care database, was used to investigate the risk of pre-specified outcomes following live attenuated herpes zoster vaccine (ZVL) in Australia.

METHODS

Individuals aged 70-79 years who received ZVL between 1 November 2016 and 31 July 2018 were identified from MedicineInsight. The self-controlled case series (SCCS) method was used to estimate the seasonally-adjusted relative incidence (RI) of seven pre-specified outcome events (injection site reaction (ISR) [positive control], burn [negative control], myocardial infarction (MI), stroke, rash, rash with an antiviral prescription, and clinical attendance) during a plausible post-vaccination at-risk window compared with times distant from vaccination. Sensitivity analyses examined the effect of common concomitant vaccinations and restriction to first outcome events.

RESULTS

A total of 332,988 vaccination encounters among 150,054 individuals were identified during the study period; over 2 million clinical attendances were observed. There was an increased RI of ISR in the seven days following ZVL (RI = 77.4, 95% CI 48.1-124.6); the RI of clinical attendance (RI = 0.94, 95% CI 0.94-0.95) and stroke (RI = 0.58, 95% CI 0.44-0.78) were lower in the 42 days following administration of ZVL compared to control periods. There was no evidence of a change in the RI of MI (RI = 0.74, 95% CI 0.41-1.33), rash (RI = 0.97, 95% CI 0.88-1.08), or rash with antiviral prescription (RI = 0.83, 95% CI 0.62-1.10) in the 42 days following ZVL compared to control periods.

CONCLUSION

No new safety concerns were identified for ZVL in this study based on a novel, Australian primary care data source. An expected increased risk of ISR was identified; findings in relation to cardiovascular disease were reassuring but require confirmation using additional data, including hospital records.

Prevalence, incidence and management of atopic dermatitis in Australian general practice using routinely collected data from MedicineInsight

Background/Objectives

The prevalence of atopic dermatitis (AD) has increased significantly in industrialised countries in recent decades but data about the incidence or prevalence of AD in Australia are sparse. We aimed to determine the prevalence and incidence of AD among patients seen in Australian general practice and the use of specified medicines.

Methods

This was a cross‐sectional study of 2.1 million patients attending 494 general practices in the MedicineInsight program from 1 January 2017 to 31 December 2018. We assessed the prevalence (lifetime and current), incidence, management and severity of AD.

Results

The lifetime (ever diagnosed) prevalence of AD in this general practice population was 16.4% and was greater in females (17.3%) than males (15.3%). One in five patients with AD were classified as having moderate‐to‐severe disease. Prevalence over the last two years was 6.3%. The incidence of AD in 2018 was 2.0% and was greater in females (2.2%) and for patients aged 0–4 years (3.9%). Patients with AD had an increased risk of insomnia, anxiety and depression, compared to those with no recorded AD. For AD patients, topical corticosteroids were the most commonly prescribed AD medication (36.5%) and topical calcineurin inhibitors the least (0.1%), with systemic corticosteroids (15.6%) more commonly prescribed than other immunosuppressants (0.9%).

Conclusions

Our findings provide important insights into the epidemiology of AD and its management in Australian general practice. This information is likely to be useful in planning effective interventions to support GPs in the optimal management of patients with AD.

Using general practice clinical information system data for research: the case in Australia

General practice is often a patient’s first point of contact with the health system and the gateway to specialist services. In Australia, different aspects of the health system are managed by the Commonwealth Government and individual state / territory governments. Although there is a long history of research using administrative data in Australia, this split in the management and funding of services has hindered whole-system research. Additionally, the administrative data typically available for research are often collected for reimbursement purposes and lack clinical information.

General practices collect a range of patient information including diagnoses, medications prescribed, results of pathology tests ordered and so on. Practices are increasingly using clinical information systems and data extraction tools to make use of this information. This paper describes approaches used on several research projects to access clinical, as opposed to administrative, general practice data which to date has seen little use as a resource for research.

This information was accessed in three ways. The first was by working directly with practices to access clinical and management data to support research. The second involved accessing general practice data through collaboration with Primary Health Networks, recently established in Australia to increase the efficiency and effectiveness of health services for patients. The third was via NPS MedicineWise’s MedicineInsight program, which collects data from consenting practices across Australia and makes these data available to researchers.

We describe each approach including data access requirements and the advantages and challenges of each method. All approaches provide the opportunity to better understand data previously unavailable for research in Australia. The challenge of linking general practice data to other sources, currently being explored for general practice data, is discussed.

Finally, we describe some general practice data collections used for research internationally and how these compare to collections available in Australia.

Pneumococcal vaccination uptake among patients aged 65 years or over in Australian general practice

In Australia, pneumococcal vaccine is provided free to all adults aged ≥65 years and Indigenous people aged 15–65 years, and is subsidized for non-Indigenous adults <65 years of age with risk factors. This study aimed to explore pneumococcal vaccination uptake in older patients attending 550 Australian general practices from 2010–2017 by patient sociodemographics, presence of comorbidities and practice characteristics. Study 1: a cross-sectional analysis of ‘active’ patients aged ≥65 years in each year was performed to calculate annual pneumococcal vaccination uptake. Study 2: a cohort of 58,589 ‘every year’ patients aged 60–65 years in 2010 was analyzed to identify the number of patients immunized during the study period. Logistic regression models assessed associations between vaccination, patient and practice characteristics. Annual pneumococcal vaccine uptake varied by patient’s age (65–74 or ≥75 years), presence of comorbidities and regularity of practice visits (range 36% to 76%), and it declined slowly from 2011–2016 amongst all groups. Cohort analyses showed that 69% of those aged 60–65 years in 2010 had a recorded pneumococcal vaccination by 2017 (peak age of vaccination = 66 years), and vaccination was more likely among those with comorbidities, ex-smokers and frequent attenders to practices. Findings demonstrate that the NPS MedicineInsight database provides estimates of vaccination uptake consistent with past surveys, reproducible every year and at low cost. It has the advantage of additional clinical information compared to the Australian Immunization Register. Whilst vaccination uptake was adequate among ‘every year’ patients, interventions are needed to improve pneumococcal vaccination for all older Australians.

Impact of funding influenza vaccination on coverage among Australian children: a national study using MedicineInsight, a large general practice database

Influenza contributes to morbidity and mortality worldwide. Children are at a higher risk of influenza-related complications and vaccination promotes direct protection and limits transmission. This study aimed to explore influenza vaccination coverage among children in Australian general practice from 2015 to 2018, and patterns in coverage before and after the implementation of state-funded immunization programs. Data from 196,520 'active' patients (3+ consultations in two consecutive years) aged <5 years from 542 Australian general practices were included (MedicineInsight database). Logistic regression models were used to identify associations between vaccination with patient and practice characteristics. The overall vaccination coverage increased more than five times from 2015 (3.9%) to 2018 (19.6%) and varied among states. Children attending practices located in the wealthiest areas were more likely to receive the vaccine and appeared to benefit most from the funding, as the increase in coverage from 2017 to 2018 was greater among them than those attending practices in the least advantaged areas (17 vs. 11 percentage points, respectively). This relationship was not evident when analyzing the patient's socioeconomic level. In conclusion, free influenza vaccinations increase coverage in at-risk populations. Promotional campaigns may be required to maintain higher coverage and target practices located in low-income areas.