Demographic and pregnancy-related predictors of postnatal contraception uptake: A cross-sectional study

Objective: To examine the uptake of postnatal

feelings of ambivalence. 6A 2016 quality statement from the National Institute for Health and Care Excellence (NICE) recommends that 'women who give birth [should be] given information about, and offered a choice of, all contraceptive methods by their midwife within 7 days of delivery'. 7The Faculty of Sexual and Reproductive Health (FSRH) goes further, advising that 'maternity service providers should ensure that all women after pregnancy have access to the full range of contraceptives, including the most effective LARC [long-acting reversible contraception] methods, to start immediately after childbirth'. 8ore recently, the All Party Parliamentary Group (APPG) on Sexual and Reproductive Health in the UK has recommended that the full range of immediate postpregnancy contraception should be made available in abortion, maternity and early pregnancy settings. 9he North East and North Cumbria (NENC) Integrated Care System (ICS) is the largest ICS in England, with 3 million inhabitants and an annual birth population of 25 000.An ICS is a partnership of healthcare organisations that is responsible for planning and delivering healthcare services across geographical areas in England.The NENC ICS has the highest rate of conceptions to those aged under 18 years in England, and one of the highest rates of abortions occurring after a recent birth among individuals under 25 years of age. 10,11In 2022, the Women's Health Strategy For England identified access to contraception as a priority area for immediate action, including contraception after childbirth. 12In a fragmented commissionerprovider system, with multiple organisations responsible for delivering different elements of reproductive health care, providing comprehensive contraception care can be challenging.
The purpose of this cross-sectional study was to explore the current PNC offer in the NENC ICS by means of an online survey of a convenience sample of women who had completed a pregnancy in the preceding 3 years and had used regional healthcare services during and after their pregnancy.

| M ET HODS
During the period from 1 December 2022 to 3 April 2023, women aged 16 years or above who had completed a pregnancy in the preceding 3 years in the NENC region were invited to complete an anonymous online survey.The survey was developed by a multidisciplinary steering group of academic and practice partners and piloted by recently pregnant members of a public and patient involvement (PPI) panel, with changes made to the survey prior to launch in response to service-user feedback.The survey collected demographic data describing personal characteristics and individual-and area-level measures of socio-economic status, reproductive history, postnatal contraception use and access to preferred PNC methods during the respondents' most recent pregnancies.Several elements of the survey were modelled on questions and response options included in a validated reproductive health survey launched by Public Health England in 2021, including questions that listed the contraceptive methods available.The present study is part of a larger project -the North East and North Cumbria Postnatal Contraception (PoCo) Study -that also examined wider experiences of antenatal, intrapartum and postpartum care, in addition to its primary focus on PNC.Participants were required to respond positively to a series of consent statements before being able to proceed to the main survey questions.Respondents who completed the survey were given the option to be included in a prize draw, with the opportunity to win a £50 shopping voucher.The survey was hosted on the Jisc online surveys platform (https://app.onlinesurveys.jisc.ac.uk).
Recruitment to the survey was achieved via multiple routes.A link to the survey was shared on Facebook, Instagram and X (formerly Twitter) social media platforms and by means of targeted online advertising.Posters and business cards were placed in public spaces across the region.Although the aim of the survey was to reach a large convenience sample of participants, efforts were also made to recruit from population groups of particular interest by means of engagement with support organisations and special interest groups.The survey was available in non-English language versions and in paper versions for participants without digital access.With support from the National Institute for Health and Care Research (NIHR) clinical research network (CRN) NENC, ten GP practices in North Cumbria and rural North Northumberland responded to a call to participate as participant identification centres (PICs), running database searches on their practice lists and inviting eligible patients to consider completing the survey by means of a text, email or postal invitation.All eight National Health Service (NHS) Foundation Trusts in the NENC region also participated as PICs and provided current and former service users with written and verbal information about the study.
Statistical analyses were undertaken using SPSS 27 (IBM, Armonk, NY, USA).Descriptive statistics were generated to describe PNC uptake and the availability of preferred PNC method across the whole eligible sample by individual contraceptive type and by grouped methods of interest where applicable (Table 1).Proposed associations between demographic and pregnancy-related characteristics and PNC uptake/satisfaction were explored using odds ratios (ORs), and multivariate logistic regression was used to generate adjusted odds ratios (aORs) that controlled for the other demographic and pregnancy-related variables measured (see footnote to Tables 2-4).Potentially confounding variables used in the adjusted analyses were selected if a significant association was observed with that variable in the unadjusted analysis.Statistical significance was set at P < 0.05.Ethical approval for this study was granted by the Newcastle and North Tyneside 1 Research Ethics Committee (22/NE/0212).

| R E SU LTS
A total of 3088 survey responses were received, with 2509 responses eligible for inclusion in analyses after ineligible submissions were excluded.Most exclusions were made for participants submitting non-UK postcodes and/or returning internally inconsistent responses.Sample demographic

| Postnatal contraceptive uptake
In all, 47.1% (n = 1178) of survey respondents indicated that they resumed sexual activity within 8 weeks of completing their most recent pregnancy (Table 1).However, although 71.1% of respondents reported using one or more contraceptive methods during this period, only 38.7% (n = 969) used a more effective contraceptive method that was medically prescribed/administered, and only 15.5% (n = 389) used a LARC (for a description of method groupings, see the footnote to Table 1).Almost a third (29.7%, n = 742) of respondents reported using condoms during the postnatal period, 21.9% (n = 547) used oral contraceptives, 4.1% (n = 103) practiced the lactational amenorrhoea method (LAM) and 1.7% (n = 43) accessed permanent contraception methods (tubal ligation or male partner vasectomy).Notably, only 43.6% (n = 514) of the 1178 participants who indicated that they resumed sexual activity within 8 weeks of delivery reported using some form of medically prescribed or administered contraception during that period.Just over half (51.7%, n = 1238) of participants indicated that they were able to access their preferred contraceptive  method within weeks of completing their most recent pregnancy, but 18.8% (n = 451) said that they were unable to do so.The remaining 29.5% (n = 706) indicated that they did not want PNC or did not have a preferred method.

| Demographic predictors of PNC uptake
Younger women in this sample were significantly more likely to access any medically prescribed contraception or LARC during the postnatal period than women in older age categories (Tables 1 and 2).The uptake of any medically prescribed contraceptive method ranged from 71.4% (n = 25) in women aged 19 years or under to 19.4% (n = 7) among women aged 40 years or above, and the uptake of LARC ranged from 51.4% (n = 18) to 5.6% (n = 2) in the same age categories.These associations were statistically significant, followed a clear trend (with uptake decreasing with increasing age), and persisted after adjusting for other variables.
Across four measures of socio-economic status (SES; household income, educational attainment, employment status and home postcode IMD quintile), women in lower SES groups were consistently more likely to access any medically prescribed contraception and/or LARC postnatally than women in higher SES groups, with clear evidence of a trend in uptake decreasing with increasing SES.However, after adjusting for potential confounding factors, the only statistically significant association that persisted was in relation to the lower uptake of any medically prescribed contraception among women with an annual household income of £40,000-£69,000, compared with women with a household income of less than £40,000.
Patterns of uptake in relation to self-reported physical and mental health were, for the most part, not significant, but differences in some individual categories were significant without following any clear trend.

| Demographic predictors of accessing preferred PNC
Women who identified as lesbian/bisexual/queer were significantly less likely to have been able to access their preferred PNC method than women who described themselves as straight/heterosexual (63.1%, n = 53, vs 73.7%, n = 1179); this association persisted following adjustment for potential confounding factors (Table 4).Women who were non-white British were significantly more likely to indicate that they had been able to access their preferred PNC method than white British women (82.8%, n = 53, vs 72.9%, n = 1182).Some evidence emerged of lower access to preferred PNC method among respondents who self-reported their mental health as less than 'very good', but this was not consistent for all mental health categories.

| Pregnancy-related predictors of PNC uptake
Women who had had four or more viable pregnancies were found to be significantly more likely to access any medically prescribed PNC or LARC than women with lower parities (Tables 2 and 3).For any medically prescribed PNC, this association persisted after adjusting for potential confounding factors.Women who delivered their most recent pregnancy by caesarean section (planned and emergency) or with forceps/ventouse were also found to be significantly more likely to access any medically prescribed PNC method than women who had an unassisted vaginal delivery.
Pregnancy intention, wherein the respondent's most recent pregnancy was described as unplanned or ambivalent, was a significant predictor of PNC uptake, but only the higher uptake of postnatal LARC following an unplanned pregnancy remained significant after adjusting for other variables.
Women who did not breastfeed following their most recent pregnancy were more likely to access any medically prescribed PNC method or LARC than breastfeeding mothers (48.6%, n = 372, vs 34.4%, n = 596; 19.3%, n = 148, vs 13.9%, n = 241, respectively), but these associations were only robust to multivariate logistic regression in the case of any medically prescribed PNC.

| Pregnancy-related predictors of accessing preferred PNC
Pregnancy-related characteristics were not significant predictors of women accessing their preferred PNC method, with the exception of breastfeeding status: non-breastfeeding women were more likely to report being able to access their preferred PNC method compared with women who breastfed (Table 4).

| Main findings
This study found that the uptake of the most effective forms of PNC was low in the NENC ICS, and that almost one-fifth of respondents were unable to access their preferred method of PNC.Although many demographic and pregnancyrelated characteristics were not significantly associated with PNC uptake, women in this sample who were younger, had a lower household income, did not breastfeed, delivered by caesarean section, had three or more previous viable pregnancies and/or whose most recent pregnancy was unplanned were more likely to access reliable PNC methods.Women who identified as lesbian/bisexual/queer, were white British and who breastfed were more likely to say that they had been unable to access their preferred PNC.

| and limitations
The large sample size and extensive geographical reach of the survey (largely reflecting the demographic diversity of the background population in relation to age and SES, but not in relation to ethnicity) are key strengths of this study.As an online survey, the ability to describe sensitive topics anonymously may also have facilitated participation for some respondents.However, the small number of respondents from some subgroups is a weakness, and this precluded a more granular appraisal of the impact of ethnicity and gender identity on PNC uptake/satisfaction.Some of the associations in relation to these subgroups may have proven significant with a larger sample.That the data were selfreported is also a potential weakness: more than 500 ineligible responses were excluded and, given that women were asked to describe events that may have occurred up to 3 years ago, there was the potential for recall bias.

| Interpretation
The association between younger age and higher uptake of PNC in this cohort is noteworthy, as research in other settings has found that teenage mothers are less likely to access postnatal care following discharge from hospital. 13The finding that women in older age categories are significantly less likely to access PNC care is also important.There may be a perception among healthcare providers that PNC is less valued and desired among older, potentially perimenopausal women.However, a US study estimated that as many as 75% of pregnancies to women aged over 40 years are unplanned, and qualitative research has reported that the PNC care of women who have in vitro fertilisation (IVF) pregnancies is often suboptimal. 14,15he finding that women in lower SES groups were more likely to access PNC is in keeping with research in other high-income settings. 16Given that low income has been identified as an important risk factor for unplanned pregnancy, targeting PNC services at more socio-economically disadvantaged women may be beneficial. 17However, this approach may be ethically contentious, and it is incumbent on providers to avoid the risks of contraceptive coercion. 18,19lthough LARC methods may reduce the risk of rapid repeat and unplanned pregnancy, they will not, on their own at least, meaningfully address the social phenomena that make these outcomes more likely among women from lower SES groups.
The significant association between a small number of pregnancy-related/reproductive characteristics and PNC uptake highlights opportunities for maternity care providers to consider how they might look to deliver PNC care in response to patient profiles.In this sample, women with higher parities (parity 4+) were significantly more likely to access any medically prescribed PNC than women with lower parity.Women with higher parities are less likely to indicate a desire for further future pregnancies, and grand multiparity (parity 5+) may be associated with an increased risk of some adverse maternal outcomes. 20,21The observation that women in this sample who had a caesarean section were more likely to access any medically prescribed PNC method but not more likely to access LARC methods is intriguing, and suggests that opportunities to site intrauterine contraception at the time of caesarean section are not currently widely offered or accessed.Work remains to be done to utilise the opportunities associated with an operative delivery to deliver more comprehensive PNC options.
The finding that women who do not breastfeed are more likely to access more effective PNC is intuitive.However, although LAM may be an effective approach to family planning for up to 6 months postpartum for many women, it is user-dependent and does not provide longer-term protection of the type afforded by LARC methods.As such, FSRH recommends that PNC should be initiated by breastfeeding and non-breastfeeding mothers as soon as possible following delivery. 8omen who identified as gay/bisexual/queer were less likely to indicate that they were able to access their preferred PNC method than heterosexual women.Although there was not a significant difference between these two groups in terms of PNC uptake, this finding suggests that providers are falling short of meeting the needs of LGBTQ+ women in regard to their PNC preferences.Research in other settings has shown that women from a sexual minority group often experience higher rates of unintended pregnancy than heterosexual women, and has described queer-specific barriers to accessing effective contraception. 22,23The finding reported here suggests that these barriers persist in the postnatal period.The observation that respondents from ethnic minorities were more likely to have been able to access their preferred PNC method, but not more likely to access effective methods of PNC, is intriguing.This low PNC uptake but relatively high PNC 'satisfaction' seen in the non-white UK group may reflect diverse sociocultural attitudes towards contraception.Ultimately, qualitative research is required to understand the nuance of these findings.The finding that women whose self-reported mental health was less than 'very good' were less likely to be able to access their preferred PNC method is potentially noteworthy, but the inconsistency of this finding across mental health categories suggests that this finding should be interpreted with caution.][26] The relatively low uptake of PNC in this UK-based highincome cohort reflects some of the challenges of providing comprehensive postnatal family planning care in global settings.8][29] The LOWE (LARC fOrWard counsElling) trial in Sweden has identified the potential impact structured contraceptive counselling to increase LARC uptake and reduce the risk of short interpregnancy intervals. 30In global settings, where maternal mortality and morbidity remains a serious threat, the need for effective evidence-based approaches to PNC provision is more urgent still.

| CONCLUSION
This study sheds light on the characteristics of PNC users in a large English region/ICS and identifies unmet need in the provision of PNC care.However, there is evidence that some of the women at greatest risk of rapid repeat pregnancy are more likely to access the most effective PNC methods.This challenges the assumption of the inverse care law, and may signify targeted activities on the part of provider organisations to reach those with greatest need.Ultimately, a policy of proportionate universalism that delivers targeted activities with 'high risk' subgroups proportionate to their need, while also being universally accessible, is likely to be the most effective means of achieving the public health and patient benefits offered by comprehensive PNC coverage.

T A B L E 3
Significant predictors of uptake of postnatal long-acting reversible contraceptive (LARC) methods.

T A B L E 1
Survey respondents' reported contraceptive use during the 8-week postnatal period.
a Note that total is >100% because respondents were able to select more than one option.bIncludes all oral contraceptives, patches, rings, permanent contraceptive methods, intrauterine contraception, injections and implants.cIncludes all intrauterine contraception, implants and contraceptive injections.T A B L E 2 Significant predictors of uptake of medically prescribed or administered PNC methods.

356* (1.035-1.778) Emergency caesarean section 1.410* (1.082-1.837)
a Adjusted for age, relationship status, ethnicity, place of birth, postcode IMD quintile, education, household income, physical health, parity, pregnancy intention, breastfeeding status, gestational diabetes and postnatal depression.*P< 0.05.broadlyaligned in relation to age and IMD postcode quintile, it is noted that the survey sample did not fully reflect the ethnic diversity of the background population, or the disability or educational attainment population profile for the region.All logistic regression analyses are reported in TablesS2-S4.

Accessed preferred PNC method during the 8-week postnatal period Variable name/description aOR a (95% CI)
a Adjusted for age, relationship status, postcode IMD quintile, education, household income, employment status, physical health, mental health, disability status, parity, history of termination, pregnancy risk classification, pregnancy intention, breastfeeding status, gestational diabetes and postnatal depression.*P < 0.05.Bold values are statistically significant.T A B L E 4 Significant predictors of access to preferred postnatal contraception (PNC) method.a Adjusted for age, sexual orientation, household income, physical health, mental health, parity, pregnancy intention, mode of delivery and breastfeeding status.*P < 0.05.Bold values are statistically significant.