Article Text
PDF
Abstract
Objective Long-acting reversible contraception (LARC) is the most effective form of contraception but use in Australia is low. Uptake of LARC prescribing by early-career general practitioners (GPs) has important implications for community reproductive health. We aimed to investigate the prevalence and associations of Australian GP registrars’ LARC prescribing.
Methods A cross-sectional analysis of the Registrar Clinical Encounters in Training (ReCEnT) cohort study 2010–2017. GP registrars collected data on 60 consecutive consultations on three occasions during their training. The outcome factor was prescription of LARC (compared with non-LARC). A secondary analysis was performed with problems involving prescription of LARC (compared with other problems). Associations with patient, practice, registrar and consultation independent variables were assessed by univariate and multivariable logistic regression.
Results 1737 registrars recorded 5382 problems/diagnoses involving women aged 12–55 years in which contraception was prescribed. 1356 (25%) involved LARC. Significant multivariable associations of prescribing LARC included patient age (OR 2.85, 95% CI 3.17 to 3.74, for age 36–45 years compared with age 12–18 years), practice rurality - inner-regional (OR 1.47, 95% CI 1.22 to 1.79) and outer-regional/remote/very remote (OR 1.47 95% CI 1.15 to 1.87) compared with major cities, practices in areas of lower socioeconomic status (SES) (OR 0.93, 95% CI 0.91 to 0.96 for SES by decile), generating learning goals (OR 1.37, 95% CI 1.04 to 1.79), in-consultation assistance-seeking (OR 1.58, 95% CI 1.24 to 2.01), and the registrar having reproductive health-related postgraduate qualifications (OR 1.33, 95% CI 1.01 to 1.76).
Conclusions The prevalence of LARC prescribing by Australian GP registrars is higher than has been previously estimated in established GPs. Postgraduate qualifications in reproductive health are associated with prescribing LARC. Prescribing practice differs according to rurality and relative socioeconomic disadvantage.
- education and training
- implants
- intrauterine devices
- intrauterine systems
- long-acting reversible contraception
- general practice
Statistics from Altmetric.com
- education and training
- implants
- intrauterine devices
- intrauterine systems
- long-acting reversible contraception
- general practice
Key messages
General practitioner (GP) registrars prescribe long-acting reversible contraception (LARC) methods more than has previously been estimated in established GPs in Australia.
GP registrars seek more help and knowledge when LARC is prescribed, and those with reproductive health qualifications are more likely to prescribe LARC.
Our findings suggest the role of GP education/training in increasing LARC use.
Introduction
Reproductive health choices are a human right and poor access to contraception is associated with poor health outcomes.1 Two-thirds of Australian women of reproductive age use contraception.2 However, over 50% of women will have an unplanned pregnancy3 and an estimated one in four pregnancies in Australia are terminated,4 which are among the highest rates in the developed world.5 Unintended pregnancies not resulting in termination are associated with poorer infant and maternal outcomes affecting women’s economic, physical, psychological and social outcomes.6
Long-acting reversible contraception (LARC) is defined as methods administered less frequently than monthly7 and includes implants, hormonal intrauterine devices (IUDs), non-hormonal IUDs and medroxyprogesterone injectables.7 8 LARC is the most effective form of contraception.9 A recent study showed that of women who experienced an unintended pregnancy while using contraception, 90% were using a non-LARC method.10
Despite well-documented advantages of LARC,8 oral contraception is the most commonly used method in Australia (33%).11 Uptake of LARC is poor, with only 13% of Australian women using these methods.11 In the UK, an estimated 12% of women aged 16–49 years use LARC methods,7 and in the USA this figure is 11.6% (although here injectables are not considered to be a LARC method).12 Increasing uptake of LARC is currently a health priority in the UK7 13 and USA14 but there is no clear policy in Australia.1
General practitioners (GPs) see 86.9% of the Australian population annually and play a critical role in contraception provision.15 Contraceptive problems are managed by Australian GPs at a rate of 6.1 per 100 consultations with reproductive-age women.8 Only 15% of contraceptive consultations in a 2011 Australian general practice study involved LARC, compared with 69% for the combined oral contraceptive pill (COCP).8 Little is known about GPs’ contraception management. The practice behaviours of early-career GPs are an important indicator of future primary care provision.
Australia’s health system is funded by both government and privately. ‘Medicare’ is a universal health insurance scheme which funds medical services, public hospitals and medicines (through the Pharmaceutical Benefits Scheme (PBS)). The Medicare Benefits Schedule (MBS) lists all the services for which doctors are remunerated. Australian General Practice Training (AGPT) involves 1 year in hospital followed by at least three 6-month terms in general practice. The AGPT includes registrars from both the Royal Australian College of General Practitioners (RACGP) and the Australian College of Rural and Remote Medicine (ACRRM). As postgraduate experience in obstetrics and gynaecology (O&G) is not a pre-requisite for training,16 some registrars may have little or no experience in reproductive health prior to entering training.
The prevalence and associations of Australian GP registrars’ prescribing of LARC has not previously been reported. In this study we sought to establish the prevalence of LARC prescribing, the associations of prescribing LARC versus non-LARC methods, and the overall associations of LARC prescribing.
Methods
Study design
This was a cross-sectional analysis of data from the Registrar Clinical Encounters in Training (ReCEnT) study.
ReCEnT – setting/participants/study materials
ReCEnT is an ongoing, multicentre, prospective cohort study of GP registrars. It was conducted in 2010–2015 in five of Australia’s 17 regional training providers (RTPs) in five of the six Australian states and, from 2016 (after a major reorganisation of GP vocational training), in three of Australia’s nine Regional Training Organisations (RTOs). RTPs and RTOs will hereafter be referred to as ‘regions’.
ReCEnT documents the nature and associations of registrars’ in-consultation clinical and educational experiences. Participation is a routine component of their educational programme.17 18 Registrars may also provide informed voluntary consent for their data to be used for research purposes. The study protocol is described in detail elsewhere.19 Registrars complete paper-based case report forms (CRFs) recording details of 60 consecutive consultations at approximately the midpoint of each of their three 6-month general practice training terms (part-time registrars participate 12-monthly). As data collection is intended to reflect a ‘normal’ week in general practice, consultations in specialised clinics (eg, vaccinations or cervical screening) are excluded. Only office-based (not home visits or nursing home visits) consultations are recorded. Registrar demographics and practice data are documented via questionnaires on training commencement or at the start of each collection period, as appropriate.
Outcome factor
The outcome factor for this analysis was prescription of LARC for problems/diagnoses related to contraception, as defined by relevant International Classification of Primary Care (ICPC-2) codes (see online supplementary appendix A). LARC was defined as the etonogestrel implant, levonorgestrel IUD, non-hormonal IUD, and medroxyprogesterone injection. Non-LARC was defined as all progesterone-only and combined contraceptive pills available in Australia and the vaginal ring (see online supplementary appendix B for Anatomic Therapeutic Chemical (ATC) codes). Barrier methods (condoms, female condoms, and diaphragm), and emergency contraception (oral or IUD) were not included. Barrier methods generally do not require GP-initiation and are not captured by our methodology. Emergency contraception is not considered as a regular prophylactic contraceptive method and was also excluded.
Supplemental material
Supplemental material
Independent variables
Independent variables related to registrar, practice, patient, consultation, or educational factors.
Registrar variables were age, gender, training term, whether in full-time or part-time (less than eight half-day clinical sessions per week) training, place of primary medical qualification (Australia or international), whether the registrar had previously worked at the practice, and reproductive health-related postgraduate qualifications (defined as completion of one or more of Certificate of Women’s Health (CWH), Diploma of Royal Australian and New Zealand College of Obstetrics & Gynaecology (DRANZCOG) or Family Planning Association Australia (FPAA) National Certificate in Reproductive & Sexual Health).
Practice variables were practice size (small <6 doctors vs large ≥6 doctors) and billing policy (whether the practice routinely bulk-bills, that is, government subsidy is accepted as full payment and there is no cost to the patient). Practice postcode was used to determine the Australian Standard Geographical Classification-Remoteness Area (ASGC-RA) to define the practice locations’ degree of rurality (very remote, remote, outer regional, inner regional or major city location) and Socioeconomic Index for Area (SEIFA) Index of Disadvantage (where low deciles represent lower level of disadvantage).
Patient variables (recorded for each patient) were age, gender, Aboriginal or Torres Strait Islander status, non-English speaking background status, the patient being new to the practice, or to the registrar.
Consultation variables were duration (in minutes) and if the problem/diagnosis was new or pre-existing.
Educational factors were whether the registrar sought advice or information in-consultation (from their supervisor or other sources, such as specialists, books or electronic resources) or generated learning goals.
Problems/diagnoses are coded according to ICPC-220 and medications according to the ATC classification.21
Statistical methods
This was a cross-sectional analysis of data from the longitudinal ReCEnT study. Analysis was at the level of problem/diagnosis and was confined to problems/diagnoses in female patients aged 12–55 years inclusive, for 16 rounds of data collection from 2010 to 2017.
The proportion of problems/diagnoses for which LARC was prescribed was calculated, with 95% confidence intervals.
The primary analysis was a comparison of prescription of LARC versus prescription of non-LARC methods. This assessed associations of registrars prescribing LARC as a contraceptive method and was chosen to provide important information for formulating measures to increase LARC utilisation by early-career GPs. The secondary analysis was a comparison of problems/diagnoses involving prescription of LARC versus all other problems/diagnoses. This is of importance in vocational training in establishing how often registrars are gaining experience in prescribing LARC (as well as which registrars and in which circumstances).
For both primary and secondary analyses, the frequencies of categorical variables were compared between outcome categories using Chi-squared tests or Fisher’s exact test, as appropriate. For continuous variables, means were compared using a t-test. Univariate and multivariable logistic regression was used within the generalised estimating equations (GEE) framework to account for clustering of repeated measures within registrars. An exchangeable working correlation structure was assumed. Covariates with a univariate p value <0.20 were considered for inclusion in the multiple regression model. Covariates with p values >0.20 in the multivariable model were removed from the final model if the covariate’s removal did not substantively change the resulting model.
Analyses were programmed using STATA 14.0 (StataCorp, College Station, TX, USA) and SAS V9.4 (SAS Institute Inc., Cary, NC, USA). Predictors were considered statistically significant if the p value <0.05.
Ethics approval for ReCEnT is from the University of Newcastle Human Research and Ethics Committee (Reference H-2009–0323).
Patient and public involvement
Patients were not involved in the design, recruitment or conduct of this study. Feedback from participating registrars is considered in ReCEnT study design, and results are disseminated to them through training provider ‘training updates’.
Results
A total of 1737 individual registrars contributed 4073 registrar-rounds of data (response rate 96.1%). Demographics of participating registrars, practices and patients are presented in table 1. These are comparable to the registrar population of Australia.22
Participating registrar, practice and patient characteristics 2010–2017
There were 84 821 consultations and 135 652 problems/diagnoses for female patients aged 12–55 years. Of these, 5382 problems/diagnoses (4.0%) involved contraceptive prescription, 1356 (25%, 95% CI 24.1% to 26.4%) of which were LARC prescriptions. For categories of LARC and non-LARC prescribed, see table 2.
Types of long-acting reversible contraception (LARC) and non-LARC prescribed: Australian general practitioner registrars 2010–2017
Primary analysis – association of prescribing LARC compared to non-LARC
Characteristics of LARC prescribing compared to non-LARC prescribing are presented in table 3.
Characteristics associated with Australian general practitioner registrars’ prescribing of long-acting reversible contraception (LARC) versus non-LARC 2010–2017
The univariate and multivariable associations of LARC prescribing are shown in table 4.
Associations of Australian general practitioner registrars' prescribing of long-acting reversible contraception (LARC) versus non-LARC 2010–2017: multivariable logistic regression
In the multivariable model adjusted for other variables, prescribing of LARC versus non-LARC was associated with older age groups (OR 1.60–2.86) compared with 12–18 years. Registrars were less likely to prescribe LARC if the patient was new to the practice (OR 0.42, 95% CI 0.31 to 0.57) or new to the registrar (OR 0.53, 95% CI 0.45 to 0.63). Prescribing LARC was more likely if the registrar had a reproductive health-related postgraduate qualification (OR 1.33, 95% CI 1.01 to 1.76). Compared with major cities, inner regional (OR 1.47, 95% CI 1.22 to 1.79) and outer regional/remote/very remote (OR 1.47, 95% CI 1.15 to 1.87) practice locations were associated with LARC prescription. LARC prescription was associated with lower practice-location SEIFA decile (OR 0.93, 95% CI 0.91 to 0.96). LARC prescription was associated with significantly longer consultation duration in minutes (OR 1.02, 95% CI 1.02 to 1.03), learning goals generated (OR 1.37, 95% CI 1.04 to 1.79) and information/assistance sought (OR 1.58, 95% CI 1.24 to 2.01). Of sources of information/assistance, 54% were electronic (see online supplementary appendix C), 25% were the registrars’ supervisor (or delegate), 11% were books and 2% were specialists or other health professionals.
Supplemental material
Secondary analysis – association of a problem/diagnosis involving prescription of LARC (compared to all other problems/diagnoses)
See online supplementary appendix D for characteristics and online supplementary appendix E for univariate and multivariable associations.
Supplemental material
Supplemental material
In the multivariable model, problems/diagnoses in women aged 46–55 years (OR 0.19, 95% CI 0.14 to 0.27) and 36–45 years (OR 0.73, 95% CI 0.60 to 0.90) compared with ages 12–18 years, and in non-English speaking background patients (OR 0.71, 95% CI 0.53 to 0.94) were less likely to involve prescription of LARC. The registrar was more likely female if LARC was prescribed (OR 1.34, 95% CI 1.15 to 1.55). LARC prescription was associated with postgraduate reproductive health qualification (OR 1.32, 95% CI 1.09 to 1.61). LARC was prescribed significantly more frequently in inner regional (OR 1.5, 95% CI 1.29 to 1.75,) and outer regional/remote/very remote areas (OR 1.71, 95% CI 1.47 to 1.98) compared with major cities. The higher the socioeconomic status of the area, the less likely LARC was prescribed (OR 0.97, 95% CI 0.95 to 0.99 for SEIFA deciles). If LARC was prescribed, registrars were more likely to have sought information/advice compared with other problems/diagnoses (OR 1.31, 95% CI 1.12 to 1.52).
Discussion
The prevalence of LARC prescribing by GP registrars in our study is higher than has previously been estimated for established GPs.8 Non-LARC methods, however, remain the most frequently prescribed contraception by Australian GP registrars. We identified multiple associations of LARC prescribing; most notably registrar completion of postgraduate qualifications (though the absolute number of these registrars was modest), increased generation of learning goals, increased seeking of information/advice and rurality of practice. We also found evidence supporting previous findings in established GPs that LARC is prescribed proportionately more in older women, and by female GPs compared with their male counterparts.8
These findings of relative underprescribing of LARC have important implications for GP training. It may be that formal training in LARC for registrars may lead to increased LARC prescribing. However, other barriers to increased LARC use such as patient perceptions, nurse training, MBS remuneration, insertion training, and maintaining skills all need to be addressed.5
The findings of associations with in-consultation information or advice-seeking, learning goal generation, and longer consultation, however, suggest that GP registrars find LARC prescribing more challenging than non-LARC methods. The need to refer for insertion may be a contributing factor. Training in LARC insertion is not universally included in GP registrar training and often requires additional training. Increased LARC prescribing by GP registrars who have completed postgraduate qualifications in reproductive health suggest that LARC prescribing increases with appropriate training.
The results also suggest differing patterns of contraception type prescribed according to geographical location and SEIFA index. This has implications for understanding contraception decision-making, and the training needs of registrars. It has previously been recognised that rural-located women are more likely to use LARC compared with city-living women both in Australia and the USA.23 24 A number of explanations have been postulated for this difference including access and travel distances, differing patient needs and differing GP skill sets.23 25 Our finding that LARC prescribing is associated with women managed in practices in lower SEIFA index areas reflects previous findings that LARC is used more frequently by women who do not have a university qualification, and those working in manual, trade or service occupations.23 However, further research is needed in understanding the factors at play, especially as living in a rural area and low SEIFA index is associated with higher rates of unintended pregnancy.26
Strengths and limitations
A strength of this study is the use of a large dataset (5382 problems/diagnoses involving contraceptive prescription) of GP registrar consultations with the contemporaneous recording of a large number of covariates. This has allowed us to adjust our findings for a wide range of potential confounding factors. The high response rate and inclusion of data from training organisations in five of Australia’s six states, including practices located from major cities to very remote classifications, are also strengths, providing good generalisability of findings to Australian GP registrars’ practice.
A limitation of this research is that the data provide only a ‘snap shot’ at the consultation level. While we have detailed data on individual consultations, we do not have data on contextual factors such as comorbidities that were not addressed within the index consultation but that may have influenced prescribing decisions, or on patient request. In addition, analysing the frequency of prescriptions does not give a true reflection of overall LARC use due to the varying prescribing intervals for the different methods (typically 3-yearly for implant, 5-yearly for IUD, 12 monthly for COCP, and 6-monthly for injection for PBS prescriptions). As such, our results will underestimate the true prevalence of LARC use compared with other methods. Furthermore, as our study is cross-sectional, we can hypothesise possible reasons for the associations found but cannot infer causality from our data.
Conclusions and implications
This research suggests that GP registrars prescribe non-LARC methods with greater frequency than LARC methods (though they prescribe LARC more frequently than established GPs) and that they find prescribing LARC challenging. Future research could explore whether introducing formal LARC training to GP registrars results in increased LARC prescribing, the barriers faced by GP registrars in prescribing LARC, and the reasons for geographical differences in LARC prescribing.
Acknowledgments
The authors would like to acknowledge the general practitioner (GP) registrars, GP supervisors and general practices that have participated in the ReCEnT project.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
Twitter @RachelHT, @SalSweeney
Contributors PM, SM, MvD and KH were investigators on the initial ReCEnT study. RT devised the design of the substudy. AT, AD, KH, PM, KF and NS oversaw data collection. AT, EH and JH analysed the data. RT drafted the manuscript. PM supervised the study progress. SS provided intellectual input to the manuscript. All authors contributed to critical revision of the manuscript. All authors read and approved the manuscript prior to submission.
Funding During the data collection period 2010 to 2015, funding of the ReCEnT study was by the participating educational organisations: General Practice Training Valley to Coast, the Victorian Metropolitan Alliance, General Practice Training Tasmania, Tropical Medicine Training, and Adelaide to Outback GP Training Program. These organisations were funded by the Australian Department of Health. Since 2016, the ReCEnT study is funded by an Australian Commonwealth Department of Health Commissioned Research Grant, and supported by GP Synergy, the general practice Regional Training Organization for New South Wales and the Australian Capital Territory. GP Synergy is funded by the Australian Department of Health. This particular project is supported by the Royal Australian College of General Practitioners with funding from the Australian Government under the Australian General Practice Training (AGPT) Program. The funders had no role in study design, collection, management, analysis, and interpretation of data; writing of the report; and the decision to submit the report for publication.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.