Young women with chronic disease were more likely to engage in the use of low efficacy contraception over the observation period compared to women in the general population, despite demonstrating similar rates of contraceptive use. When specific chronic diseases were examined, this pattern of use was evident for women with cardiac and autoinflammatory conditions but not those with diabetes or asthma. This study therefore provides much needed insight into how young Australian women with chronic disease ‘actually’ use contraception at the population level. The findings have important implications for the delivery of contraceptive counselling and reproductive life planning for young women with chronic disease, especially those with cardiac and autoinflammatory diseases.
Overall, contraceptive use among women meeting our chronic disease definitions was found to be high across the five-year observation period (above 85%). Although we focused exclusively on young women and included the ability to examine both hormonal and non-hormonal contraceptive methods, we found self-reported hormonal-based contraceptive use among women with chronic disease to be substantially higher than that reported by DeNoble et al. . Further, while our prevalence of contraceptive use among women with chronic disease was similar to that reported in a 2016 cross-sectional U.S. study, they found that the use of effective and highly effective contraception was lowest among women aged 18-24 years (<45%) with highly effective contraceptive use driven largely by sterilisation (even for young women) . In contrast, high efficacy contraceptive use in our study was attributed to the use of hormonal LARC. This finding is important given that LARC use has previously been reported as low among young women in Australia [10, 15, 22]. Our finding may be reflective of increased awareness in Australia around the acceptability of LARC across the reproductive life course and in particular the suitability of LARC for young nulliparous women. Increased LARC use is now also recognised as a key indicator in meeting the priorities of the current Australian National Women’s Health Strategy 2020-2030 .
Although our study findings demonstrate that LARC use is increasing among young women, when we examined patterns of contraceptive use, only women diagnosed with autoinflammatory disease were found to have 58% increased odds of hormonal LARC use compared to women without chronic disease using the pill. While this finding is promising given that LARC are recommended as first line options for women with autoinflammatory disease (including those on immunosuppressants) and provide the greatest protection against unintended pregnancy [9, 24], they contrast with a number of studies which have demonstrated low LARC uptake among this chronic disease population [25, 26]. Concerningly, while women with autoinflammatory conditions were more likely to use LARC than other women, they also had a 69% increase in odds of using low efficacy methods compared to women without chronic disease using the pill. Our finding is supported by research showing that women with SLE discontinue hormonal contraception (mainly the combined oral contraceptive pill) following diagnosis and take up lower efficacy methods despite being on potentially teratogenic medications (including methotrexate) . Use of low efficacy methods with high typical use failure rates, place these women at significant risk of unintended pregnancy. Importantly, withdrawal and condoms have been found to be the most prevalent forms of contraception used by women with SLE and RA including a substantial number with unintended pregnancy histories . Most concerning, however, is that young women diagnosed with cardiac disease had 220% and 54% increased odds of using low efficacy methods and of being non-users of contraception, compared to women without chronic disease using the pill, respectively. This finding is supported by previous cross-sectional research [6, 13]. Our findings, coupled with the emerging body of literature around contraceptive use among cardiac and autoinflammatory disease point to an unmet need regarding evidence-based contraceptive advice and support, particularly from rheumatologists and cardiologists who are responsible for prescribing (potentially teratogenic) medication and monitoring disease activity.
In our study no discernible difference in the contraceptive patterns between women with diabetes compared to women without chronic disease using the pill were found. While the international evidence around this topic has been equivocal, Australian research has found that although women with diabetes are high users of contraception (mostly condoms and the oral contraceptive pill), contraception is not consistently used . This is important given that the pill and condoms are the most prevalent forms of contraception used at the time of unintended pregnancy among young Australian women . Therefore, although the combined oral contraceptive pill is not contraindicated for young women with uncomplicated diabetes, increasing the use of highly effective contraception among this population is still warranted given the need for engagement in preconception contraceptive care to prevent adverse maternal and perinatal consequences associated with unintended pregnancy . International research has shown only 32% of teens and 18% of young adults with type 1 diabetes attain recommended glycaemic control, yet only one-quarter of adolescents are aware of the adverse impacts associated with poor glycaemic control in pregnancy .
While young women with chronic disease should be supported to choose and use a method of contraception that aligns with their reproductive and personal goals, our findings point to an underutilisation of highly effective LARC among most young women with chronic disease and suggest that gaps in the delivery of preconception contraceptive counselling may exist in Australia, particularly for those with cardiac and autoinflammatory conditions. Low rates of general and disease-specific contraceptive counselling among chronic disease populations have been demonstrated internationally, even in the presence of potentially teratogenic medication [33–35]. The reason for low contraceptive counselling among women with chronic disease however is not well understood. It has been postulated that lack of both contraceptive use and contraceptive counselling among women with chronic disease is attributed to misperceptions around fertility, knowledge regarding pregnancy risks [36, 37] and health system factors whereby the chronic condition takes up the healthcare providers’ time and focus during appointments . As a result, women with chronic disease often receive minimal contraceptive counselling from either general practitioners (GPs) or specialists. It has also been argued that healthcare providers are uncomfortable with prescribing contraception to women they perceive to be at higher risk of adverse events due to lack of familiarity with the safety of different methods for women with different medical conditions .
The contraceptive needs of women with cardiac disease are especially challenging to navigate due to variability in potential risks associated with both contraception type and the nature and severity of the cardiac disease. In general, however, medical eligibility guidelines for the provision of contraception , indicate the risks associated with the use of estrogen-containing hormonal contraception (e.g., combined oral contraceptive pill) outweigh the benefits given the increased risks of arterial and venous thrombosis for a number of cardiac conditions, including severe or poorly controlled hypertension (although in practice, the combined oral contraceptive pill is generally only advised against when there is a history of unprovoked arterial/venous thrombosis or a known genetic defect). By contrast, progestogen-only methods are not associated with a risk of venous or arterial thromboembolic disease and are safer options for most women with cardiac disease (including those with congenital heart disease) [40, 41]. Of the progestogen-only methods available, the progestogen-only implant followed by levonorgestrel-IUDs have the highest efficacy against pregnancy. Levonorgestrel IUDs are a suitable choice for women with cardiac conditions (including those with congenital heart disease), and their effect on reducing menstrual bleeding can be beneficial, including for those on anticoagulant therapy. However, women with major cardiac disease may require cardiology input before insertion . Similarly, in the presence of autoinflammatory conditions such as RA and IBD, there are concerns associated with estrogen-containing contraception in relation to disease exacerbation as well as thrombotic effects (particularly with women who have antiphospholipid syndrome and a history of IBD-related surgery or past biologic therapy use). Potential malabsorption issues for women with IBD will also limit the use of combined hormonal and progestogen-only oral methods.
Given the exposure to medications used in the treatment of autoinflammatory conditions, chronic hypertension and other cardiac diseases are associated with major congenital malformations [42–44], young women with chronic disease (particularly those with cardiac and autoinflammatory conditions) therefore require individualised contraceptive counselling and reproductive life planning based on their specific condition(s), with consideration around disease severity and medication use . As these women are already engaged within the healthcare system this provides an ideal opportunity to provide such care as part of a well-coordinated structured approach to chronic disease management involving GPs and specialists where contraceptive conversations are routine. While some key bodies such as the Australian Rheumatology Association provide guidance on prescribing medications during pregnancy and recommend that women of child-bearing age receive preconception counselling and discussions around contraception, Australia currently lacks formal guidelines for autoinflammatory diseases as well as other chronic conditions. Increased access to, and awareness of current therapeutic guidelines by peak medical associations and key bodies (e.g., Therapeutic Guidelines) as well as development of referral pathways are required alongside increasing medical education given there is a demonstrated lack of expertise and confidence regarding the provision of family planning among GPs and specialists in Australia and internationally [45–47]. This will ensure that young women with chronic disease receive access to clear and accurate information and regarding their contraceptive options. Increased education for young women with chronic disease about the risks of unintended pregnancy and the benefits of appropriate highly effective contraception for their specific condition, as well as evidence-based information to dispels myths around LARC including its impact on future fertility are also required . Importantly for young women who develop chronic disease early in life, greater acknowledgement, information provision and screening of paediatric populations in relation to sexual activity and contraceptive needs is needed . Such practices also need to be maintained as young women transition from paediatric to adult services to ensure these women do not fall through the gap.
A key strength of this study was the ability to examine contraceptive patterns for women diagnosed with seven key chronic diseases during early adulthood using longitudinal data. In addition, we were able to apply statistical techniques to accurately identify contraceptive combinations. This improves on previous research which has examined contraception within the context of chronic disease as either users or non-users or has used hierarchical approaches [12, 13]. Our analysis shows that for young women contraceptive use is complex and requires examination of all contraceptive combinations. A further strength of the study is the methods used to ascertain chronic disease cases with the inclusion of both survey and administrative data providing the ability to capture all forms of chronic disease . No studies have previously utilised these approaches for examining chronic disease among women of reproductive age. Given that contraceptive use and risk of an unintended pregnancy is dynamic across the reproductive life course , we excluded women not at risk of a future unintended pregnancy at each of the time points. Few studies have accounted for this in longitudinal research .
While we were able to examine contraceptive use at yearly intervals, the questions regarding contraceptive use were not substantive after survey 1. Unfortunately, we were not able to accurately ascertain the use of methods such as the depot injection or copper IUD. Due to minimal self-reports of their use, these items were included as ‘other’ contraception. As such, we have potentially over-estimated the use of low efficacy methods. Additionally, while we examined contraception over time using latent transition analysis, to examine time-varying covariates (including chronic disease status) we employed a classify-analyse approach. We acknowledge that this approach may induce some measurement error due to the uncertainty in latent status classification .