Article Text
PDF
Abstract
Objective To assess patient experiences of pain management during medical abortion up to 10 weeks’ gestation with opt-in versus universal codeine provision.
Methods We invited patients who underwent medical abortion up to 10 weeks of gestation to participate in an online, anonymous, English-language survey from November 2021 to March 2022. We performed ordinal regression analyses to compare satisfaction with pain management (5-point Likert scale) and maximum abortion pain score (11-point numerical rating scale) in the opt-in versus universal codeine provision groups.
Results Of 11 906 patients invited to participate, 1625 (13.6%) completed the survey. Participants reported a mean maximum pain score of 6.8±2.2. A total of 1149 participants (70.7%) reported using codeine for pain management during their abortion. Participants in the opt-in codeine provision group were significantly more likely to be satisfied with their pain management than those in the universal group (aOR 1.48, 95% CI 1.12 to 1.96, p<0.01). Maximum abortion pain scores were lower on average among the opt-in codeine provision group (OR 0.80, 95% CI 0.66 to 0.96, p=0.02); however, this association was not statistically significant in the model adjusted for covariates (aOR 0.85, 95% CI 0.70 to 1.03, p=0.09).
Conclusion Our findings suggest that patients have a better experience with pain management during medical abortion when able to opt-in to codeine provision following counselling versus receiving this medication routinely.
Data availability statement
Data are available upon reasonable request. Data are available upon reasonable request. Requests should be submitted to research@bpas.org for consideration.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Pain is often cited as one of the worst features of the medical abortion experience, and moderate or severe pain is common. However, existing evidence is lacking regarding optimum analgesic approach for medical abortion up to 10 weeks of gestation.
WHAT THIS STUDY ADDS
This evaluation assessed pain scores and satisfaction with pain management following a change from universal to opt-in provision of codeine during medical abortion up to 10 weeks of gestation. Opt-in provision of codeine was associated with patient satisfaction with pain management; this finding aligns with existing evidence that patients value choice as a marker of quality abortion care.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY.
Our findings suggest that patients have a better experience with pain management during medical abortion when they are able to opt-in to receiving codeine. There remains a need for robust research exploring strategies to improve the quality of pain counselling and management for patients undergoing medical abortion.
Introduction
Pain is often cited as one of the worst features of the medical abortion experience1; moderate or severe pain is common, with studies reporting average pain scores between 5 and 8 out of 10.2–6 Existing evidence is lacking regarding the optimum analgesic approach for medical abortion. In a recent Cochrane systematic review, the authors concluded that ibuprofen is superior to paracetamol or placebo for pain relief during medical abortion.7 However, patients often report moderate or severe pain during a medical abortion even with ibuprofen use, suggesting that it is unlikely to be sufficient for pain relief alone.3 While the use of opioids is commonplace, a randomised controlled trial published in 2019 found no difference in medical abortion pain scores between participants taking oral opioid (oxycodone) versus placebo.8 National and international guidelines do not make specific recommendations for pain relief beyond the use of non-steroidal anti-inflammatory drugs (NSAIDs),9 10 and prescribing practices vary significantly between providers.11 There is a need for more research on pain relief interventions during medical abortion to guide clinical practice.7
British Pregnancy Advisory Service (BPAS) is a national non-profit provider of abortion with clinics throughout England and Wales. Annually, BPAS provides approximately 80 000 medical abortions up to 10 weeks of gestation. Clinical policy at BPAS recommends that all patients without contraindications use NSAIDs for pain management during medical abortion up to 10 weeks of gestation. Historically, BPAS also provided a supply of codeine, a mild oral opioid, to all medical abortion patients unless contraindicated or refused. In March 2021, BPAS changed its clinical policy to recommend an informed consent discussion about pain management and offering codeine on an opt-in basis. This change from a universal to an opt-in approach was prompted by limited evidence for opioid analgesia during medical abortion, and concerns about the potential harms of routine opioid prescriptions following a steep rise in opioid-related morbidity and mortality in the UK.8 12 This evaluation aimed to assess experiences of pain management among patients undergoing medical abortion up to 10 weeks of gestation following the service change.
Methods
We performed a cross-sectional analysis of patients undergoing medical abortion up to 10 weeks of gestation at BPAS. BPAS provides medical abortion predominantly via telemedicine, wherein clinicians counsel and assess patients remotely for suitability and provide abortion medications via post or in-clinic collection. Current BPAS policy recommends that clinicians have an informed consent discussion of pain management options with patients, recommend NSAIDs as first line, and provide codeine to eligible patients who “opt-in” to take it. The medical abortion regimen consists of 200 mg oral mifepristone followed 1–2 days later by 800 µg buccal or vaginal misoprostol with an additional 400 µg misoprostol to use 3–4 hours after the first dose.10 13 Patients receive verbal and written instructions on medication use, including analgesia, and have access to 24-hour clinical support by telephone.
From 7 November 2021 to 3 March 2022, we sent a text or email invitation to all patients who had a medical abortion up to 10 weeks of gestation in the previous 7 to 21 days and had consented to be contacted for research or evaluation purposes. The invitation provided a brief introduction to the evaluation and a link to an anonymous online English-language survey via the ‘SurveyMonkey’ platform. We asked participants to report what their clinician told them about codeine for pain management. Those who reported that their clinician told them codeine would be provided automatically were defined as being in the “universal codeine provision group”. Those who reported being offered codeine following an informed consent discussion were defined as being in the “opt-in codeine provision group”. Our primary outcomes were satisfaction with pain management (5-point Likert scale, “very dissatisfied” to “very satisfied”) and maximum pain score during abortion (11-point numerical scale). The survey also included questions about medical history, abortion characteristics, use of medications, contact with healthcare providers, and some demographic information. Prior to disseminating, we piloted the survey with non-clinical staff at BPAS and made amendments accordingly. Our survey had a Flesch Reading Ease score of 75.3 out of 100 indicating that the survey was “fairly easy” to read.14 This project was not classified as research according to the Health Research Authority decision aid15 and thus BPAS Research and Ethics Committee granted it an exemption from ethical review as a service evaluation.
We analysed the data using R software version 4.1.2 with RMS package version 6.3.0. We used descriptive statistics to evaluate the health characteristics of participants. In accordance with STROBE guidelines, we have not used significance tests to assess the variability of health characteristics between groups.16 17 We performed ordinal regression to compare the outcomes of satisfaction with pain management and abortion pain scores among participants in the opt-in versus universal codeine provision groups. We fitted multivariable regression (Model 1) including variables (age, gestational age, parity, and average period pain score) that previous studies have identified as being associated with medical abortion pain.4 18 19 We also fitted multivariable regression including additional variables which we hypothesised to be associated with abortion pain (age, gestational, parity, period pain score, counselling that abortion could be painful, misoprostol route, misoprostol dose, prior abortion, regular use of analgesics in the year prior, history of chronic pelvic pain, history of fibromyalgia, history of endometriosis, history of anxiety and/or depression, and English as first language) and satisfaction with pain management (the aforementioned variables plus maximum abortion pain score, analgesia used during medical abortion, and need to contact a healthcare professional due to pain) using causal diagrams (Model 2). Causal diagrams are an established approach to support decision-making around which variables to adjust for in analyses and prevent inadvertently adjusting for mediators or inappropriately adjusting for colliders.20 21 In our primary analysis, we used multiple imputation for variables where the participant had provided a response to a survey question indicating that they could not recall or were unsure of the answer. This applied to the responses of between 170 and 179 participants in up to four questions. We fitted regression models with multiple imputation using chained equations (m=50, mice package v3.14). For assessing model fit, we calculated Nagelkerke R2 and adjusted for optimism (R2c) using the bootstrapping procedure.22
We performed sensitivity analyses to assess whether any of the assumptions in our models changed our conclusions. We repeated our analyses after: (1) excluding the responses of participants who could not recall how they were counselled about codeine, (2) including and imputing the responses of all participants who started the survey irrespective of whether they completed it and (3) defining satisfaction as an ordinal rather than categorical variable and repeating the analysis using a proportional odds model.
In a previous BPAS evaluation, 79.5% of medical abortion patients reported satisfaction with their pain management.23 For our sample size calculation, we used this value to determine that a sample size of 458 participants per group would provide 80% probability of detecting a 10% difference in the proportion of patients reporting satisfaction when comparing the universal versus opt-in codeine provision groups (α=0.05).
Patient and public involvement
We did not directly involve patients in the development, design or management of this evaluation as we designed it to assess the impact of change in clinical policy. We did not collect any identifiable data thus we cannot share results directly with participants. However, we will disseminate the results across BPAS’ website and social media in patient-friendly formats.
Results
During the evaluation period, we invited 11 906 patients to participate and 1625 (13.6% of those contacted) completed the survey (figure 1). Forty-nine percent (n=801) were in the opt-in codeine provision group and 37.0% (n=602) in the universal codeine provision group. Of the remaining participants, 1.7% (n=28) reported that codeine was not mentioned in their consultation, 2.2% (n=36) reported that the clinician said they were not eligible for codeine, and 9.7% (n=158) stated they could not remember how codeine was discussed. We describe the demographics and health characteristics of participants in table 1. More participants in the universal codeine provision group reported that English was not their first language compared with the opt-in codeine provision group (16.6%, n=100 vs 9.5%, n=76, respectively).
Recruitment to a cross-sectional evaluation of patients undergoing medical abortion up to 10 weeks of gestation at British Pregnancy Advisory Service (BPAS).
Characteristics of participants undergoing medical abortion up to 10 weeks of gestation at British Pregnancy Advisory Service (BPAS) by universal or opt-in model of codeine provision
In table 2, we describe participants' experience of clinician counselling on and experience of abortion pain. Almost all participants (95.1%, n=1546) reported that the clinician explained that abortion could be painful. However, 6.3% (n=38) of participants in the universal codeine provision group compared with 2.9% (n=23) in the opt-in group said that the clinician did not explain that abortion could be painful.
Pain outcomes of participants undergoing medical abortion up to 10 weeks of gestation at British Pregnancy Advisory Service (BPAS) by universal or opt-in model of codeine provision
Among all participants, 70.7% (n=1149) reported use of codeine. The frequency of reported codeine use was similar between the opt-in and universal groups (71.5%, n=573 vs 72.9%, n=439). Among all participants who received codeine, 76.7% (n=1149) said that they used it. More participants in the opt-in group reported using any of the codeine they received (78.1%, 569/729) compared with the universal provision group (74.8%, 439/587).
Using an 11-point numerical scale, participants rated maximum medical abortion pain score with a mean of 6.8 (±2.2). Over three-quarters of participants (75.9%, n=1233) reported that they were very satisfied or somewhat satisfied with their overall pain management. More participants in the universal codeine provision group reported that they experienced “a lot more pain than expected” than the opt in codeine provision group (33.4%, n=201 vs 26.8%, n=215) and fewer participants in the universal group reported that they experienced “about as much pain as expected” (21.1%, n=127 vs 27.2%, n=218) when compared with the opt-in codeine provision group.
We compare maximum abortion pain scores and satisfaction with pain management between opt-in and universal codeine provision groups in table 3. The responses of participants who were ineligible for codeine or reported that their clinician did not mention codeine are excluded in this analysis. Participants in the opt-in codeine provision group reported lower mean maximum abortion pain scores than those in the universal group; this association was statistically significant in the bivariate analysis and when adjusting for age, gestational age, parity, and average period pain score in Model 1 (adjusted odds ratio (aOR) 0.82, 95% CI 0.68 to 0.99, p=0.04). However, we found that this association was not statistically significant when we adjusted for additional covariates in Model 2 (aOR 0.85, 95% CI 0.70 to 1.03, p=0.09). Participants in the opt-in codeine provision group were more likely to report satisfaction with their pain management than those provided with codeine universally; this association was statistically significant in the bivariate and both multivariable regression analyses when adjusting for additional covariates (Model 2: aOR 1.48, 95% CI 1.12 to 1.96, p<0.01).
Unadjusted and adjusted odds ratios of opt-in compared with universal approach to codeine provision for pain score and satisfaction with pain management of participants undergoing mediation abortion up to 10 weeks of gestation from British Pregnancy Advisory Service (BPAS) versus opt-in model of codeine provision
Altering model parameters in sensitivity analyses did not change our conclusions about the association between codeine provision model (opt-in vs universal) and the primary outcomes of maximum abortion pain and satisfaction with pain management. The results of all sensitivity analyses are presented in the online supplemental material.
Supplemental material
Discussion
We found that an opt-in rather than universal approach to codeine provision during medical abortion up to 10 weeks of gestation was associated with greater satisfaction with pain management. While participants reported lower pain scores on average in the opt-in codeine group, the association did not remain significant after controlling for covariates. Our data are unable to explain why there is an association between opt-in codeine and greater satisfaction with pain management; however, we postulate that the opt-in approach prompted clinicians to have a more detailed discussion of pain, which better prepared patients for the experience. Our findings align with results from qualitative abortion research that suggest that patients value information, preparation and choice as markers of quality abortion care.24–26 Evidence also shows that pre-abortion information, particularly regarding pain and side effects, can reduce anxiety during the abortion and create a more positive experience overall.27
About three-quarters of participants who received codeine reported using it for pain management during their abortion. This finding correlated with other research demonstrating that 68–75% of patients use opioids during their medical abortion.4 28 A recent study by Friedlander et al,2 which describes real-time experience of pain during medical abortion, reported that only 50.0% of patients used their prescribed opioid; however, they also reported that 88.5% used their prescribed NSAID compared with 51.7% in our sample. This discrepancy in our findings may highlight the importance of NSAIDs in treatment of medical abortion pain.
Our data also suggest that there may be less unused codeine in the opt-in compared with the universal codeine provision group. Considering that recent randomised clinical trial data showing that opioids are not effective in reducing maximum pain during medical abortion, and concerns that routine opioid provision may contribute to addiction, medical abortion providers should consider taking a tailored approach to opioid provision. Further research should continue to explore optimal ways to support pain management during medical abortion.
Across the whole sample, almost half the participants reported experiencing more pain during the abortion than expected, and 4.9% reported that the clinician did not inform them that medical abortion could be painful. Differences between patient’s expectations of pain and their experienced symptoms have been found to be associated with dissatisfaction with medical abortion.29 This is also supported by research outside of the field of abortion care, indicating that pain expectations can influence experiences.30 Our findings highlight that although rates of satisfaction with pain management were high, there is room to improve the quality of counselling to better set expectations of the medical abortion experience.
A strength of this evaluation is the large sample size spanning multiple clinic locations and following a standardised clinical policy. Our evaluation has several limitations. The retrospective nature introduces recall bias on patient experience of pain and recollection of their counselling. However, we did not find evidence for any association between the number of days since the abortion and reported maximum pain scores (OR 1.00, 95% CI 0.99 to 1.00, p=0.37). Only 16.8% of the patients whom we invited actually started the survey; we acknowledge this low response rate introduces selection bias which may impact external validity. We did, however, compare the characteristics of participants in our evaluation to those of all patients who underwent medical abortions at BPAS in the evaluation period and found that there were no marked differences in distribution of characteristics such as age, parity and medications use. This data is provided in the online supplemental material. We only provided the survey online, in English, which reduces generalisability. We did not collect demographic data such as ethnicity or educational attainment so we cannot further comment on this aspect of our population. Finally, given the observational nature of this evaluation, we cannot draw conclusions regarding causality; even despite controlling for multiple covariates, there may be other explanations for the observed association between opt-in codeine provision and satisfaction with pain management. Nonetheless, our evaluation broadens understanding of pain experiences of those undergoing medical abortion and can be used to inform future research.
As more services worldwide shift to telemedicine abortion care, our evaluation offers a valuable insight into pain-related counselling practices and outcomes during medical abortion up to 10 weeks of gestation. Our findings demonstrate that opt-in provision of codeine for medical abortion, which involved more tailored counselling on pain management, was associated with higher rates of satisfaction. The results of this evaluation also highlight the need for rigorous research exploring strategies to improve the quality of pain counselling and pain management for medical abortion.
Data availability statement
Data are available upon reasonable request. Data are available upon reasonable request. Requests should be submitted to research@bpas.org for consideration.
Ethics statements
Patient consent for publication
References
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
KCW and PAL are joint senior authors.
X @drnedatnejadi, @krassowski_m, @Lohrpa
Contributors PAL conceived of this evaluation and the project was designed by NT, KCW and PAL. Analysis was performed by MK with input from NT, HM and PAL. Recruitment was completed by HM and AM-D. The manuscript was led by NT with input from HM, KCW, MK, AM-D and PAL. PAL and NT are responsible for the overall content as guarantors.
Funding The authors have not declared a specific grant for this evaluation from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests PAL is Medical Director of BPAS. KW is a paid consultant of Feral GMBH. The authors declare that they have no other known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.
Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Provenance and peer review Not commissioned; externally peer reviewed.
Author note NT is a National Institute for Health Research Academic Clinical Fellow in Community Sexual and Reproductive Health and is supported by the Oxford University Clinical Academic Graduate School (OUCAGS), University of Oxford. MK has been supported by the Scatcherd European Scholarship.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.