Article Text
PDF
Abstract
Introduction Contraceptive knowledge mediates access and use. We aimed to assess whether an online educational video describing all methods and their benefits, side effects and mode of action increased young women’s contraceptive knowledge and their long-acting reversible contraception (LARC) preference and uptake.
Method We used Facebook advertising to recruit young women aged 16–25 years. Participants completed the pre-video survey (S1), watched the 11-min video, then completed surveys immediately after (S2) and 6 months later (S3). Outcomes were analysed using McNemar tests and multivariate logistic regression (generalised estimating equations).
Results A total of 322 participants watched the video, completed S1 and S2, and 88% of those completed S3. At S1 only 6% rated their knowledge about every method as high. Knowledge improved at S2 for all methods (OR 10.0, 95% CI 5.9 to 17.1) and LARC (OR 4.2, 95% CI 3.1 to 5.7). LARC preference increased at S2 (OR 1.7, 95% CI 1.4 to 2.1) and S3 (OR 1.4, 95% CI 1.2 to 1.7), as did LARC uptake at S3 (OR 1.3, 95% CI 1.11 to 1.5). LARC uptake was driven by a 4.3% (n=12) absolute increase in intrauterine device (IUD) use, but there was no change in contraceptive implant use (p=0.8). The use of non-prescription methods such as condoms and withdrawal did not change (OR 0.92, 95% CI 0.76 to 1.11).
Conclusions Many young women in Australia do not feel well informed about their contraceptive options. The contraceptive education video, delivered via social media, increased their self-reported contraceptive knowledge and IUD preference immediately after viewing, and their IUD uptake 6 months later. Focus should be given to how young women navigate contraceptive access after internet-based education, and strategies to increase access to preferred methods.
- long-acting reversible contraception
- patient education as topic
- patient preference
- adolescent
- health education
- contraceptive agents
- female
Statistics from Altmetric.com
- long-acting reversible contraception
- patient education as topic
- patient preference
- adolescent
- health education
- contraceptive agents
- female
Key messages
What is already known on this topic?
Health information videos on contraception can influence contraceptive knowledge, method choice and reduce rates of unintended pregnancy. However, such videos have rarely been promoted to young people independent of a health provider interaction or outside healthcare settings (eg, via social media).
What this study adds?
We demonstrate that a video-based ‘contraceptive counselling session’, delivered over the internet, can improve young women’s self-rated contraceptive knowledge. We also observed an increase in preference for intrauterine devices (IUDs) immediately after the video and, contrary to the findings of previous online intervention studies, the increased use of IUDs 6 months later. The proportion of participants using no contraception, or a non-prescription or natural method to prevent pregnancy, did not change across timepoints, suggesting either an informed preference for these methods or a lack of access to contraception post-education.
How this study might affect research, practice or policy?
Video-based education can be used to complement practitioner contraceptive counselling. Ensuring access to contraception after online education is necessary for contraceptive preferences to be realised.
Introduction
Health literacy facilitates the navigation of health systems and people’s well-informed, correct contraceptive use.1 2 However, young people experience challenges in obtaining reliable contraceptive information.3–5 In particular, they report an interest in using long-acting reversible contraception (LARC) methods but require more information to consider LARC among their options.3–5
There have been no Australian studies examining the effect of education on young people’s contraceptive decisions, particularly in relation to LARC. Consequently, we lack evidence of the frequency that well-informed young people would use LARC if these methods were accessible. Gathering this evidence may leverage advocacy to address educational and other barriers to contraception overall, in addition to the inequitable promotion of LARC.
The internet is often used as a health information resource, with social media providing an opportunity to reach young people.6 7 They can selectively engage with health content that is algorithmically targeted to them, shared by their peers, and may be empowered to use evidence-based information sourced through social media to influence their health in positive ways.2 7 8
Videos in particular can be made accessible and relevant to specific populations by modifying video content, dissemination platforms and language use. Health-related videos can impact health behaviours such as treatment adherence,9 and can influence contraceptive knowledge, method choice and reduce rates of unintended pregnancy.10 11 However, video interventions for contraception have rarely been promoted to young people independent of a health provider interaction or outside healthcare settings.10 11
Internet-based educational videos could improve access to contraceptive information and potentially support young people’s access to their preferred contraceptive methods. We aimed to determine whether an educational video describing all the available contraceptive methods, advertised via social media, would increase young women’s contraceptive knowledge and their preference for and uptake of LARC.
Methods
This was a pre-post intervention study, undertaken from December 2017 to July 2018. We obtained ethics approval from Monash University Human Research Ethics Committee (Project Number: 10456) and followed the Template for Intervention Description and Replication (TIDieR) checklist to report the research.12
Patient and public involvement
The video and survey were adapted from the Australian Contraceptive Choice Project (ACCORd)13 based on evidence from prior research and expert opinion. Young women who were medical students at Monash University acted in the video and their input on the Facebook advertisement and video script was sought prior to these being finalised.
Participants
Young women aged 16–25 years, living in Australia, were eligible if: they reported being able to become pregnant; had no desire to be pregnant within a year; were sexually active with male partner/s within the past 6 months or anticipated near-future sexual activity; and where their male partners had not undergone vasectomy.
Sample size
Our sample size calculation determined a sample of 141 women as being sufficient to detect a 10% increase in LARC uptake with 80% power and a significance level of 0.05. This sample size was based on national data in which <5% of 16–24-year-olds used LARC14 and our expert stakeholders advising us that a 10% increase in LARC uptake would be a clinically meaningful result. We aimed to recruit 281 women to accommodate up to 50% potential attrition.
Intervention
In the 11-min video, five young women and a general practitioner provided a ‘contraceptive counselling session’, providing detailed information about contraceptive methods available in Australia, in the order of most to least effective. They described: presence and type of hormone, effectiveness, how the contraceptive is used and/or inserted and removed (and by whom), common side effects, non-contraceptive benefits and whether the contraceptive provides protection against sexually transmissible infections.
Study enrolment
A paid Facebook advertisement was created and posted online between December 2017 and January 2018.15 The Facebook Ads Manager fed the advertisement into News Feeds targeting by age (16–25 years), sex (female) and location (Australia).
Procedure
Those who clicked on the advertisement were taken to a page describing the study phases and linked to the explanatory statement. People who indicated an interest in participating completed the screening survey and, if eligible, were directed to the baseline survey 1 (S1). Consent was implied on survey completion. Participants then viewed the online video, completed the immediate post-intervention survey 2 (S2) and received a $A20 gift card. Six-months later they were contacted by email with a link to survey 3 (S3) and received a second $A20 gift card after completion.
Outcome measures
S1 collected: demographic characteristics; contraceptive methods ever used; contraceptive preference; method(s) currently used; awareness (yes/no) and knowledge (self-rated on a four-point Likert scale) about those contraceptives currently available in Australia (combined oral contraceptive pills (COCP), progestogen-only pill (POP), emergency contraceptive pill (ECP), intrauterine devices (IUDs) (including levonorgestrel (LNG)-releasing and copper-bearing devices), contraceptive injection, implant, ring and condoms); and satisfaction with methods used. S2 collected knowledge and preference. S3 collected knowledge, preference, methods used currently and in the past 6 months, and satisfaction. For analysis, participant self-rated knowledge was dichotomised into ‘low’ (‘know nothing’ and ‘know a little’) and ‘high’ (‘know a lot’ and ‘know everything’) categories. We also coded two summary variables: one summarising awareness across all methods and one summarising knowledge, where participants reporting awareness or having high knowledge about every one of the methods described in the video were coded as 1 and all others, 0.
The primary outcomes were: awareness and self-rated knowledge of contraception (S1 and S2), LARC preference (S1, S2 and S3) and LARC use (S1 and S3).
The secondary outcomes were: the use of non-prescription or natural primary methods (eg, condoms, withdrawal, natural family planning, diaphragm and/or ECP as primary methods) compared with prescription (‘effective') contraception (hormonal methods and IUDs) at S1 and S3, and participant-reported likelihood of contraceptive method switching (S2 and S3).
Statistical analysis
Analyses were undertaken in Stata Version 16 (StataCorp LP). We computed descriptive statistics reporting frequency (%) for all categorical variables. Age was dichotomised according to age-group size, with 50% of participants being under 22 years. Logistic regression analyses compared the demographic profiles and baseline contraceptive use and preferences of completing participants with those lost to follow-up. Complete cases were compared across timepoints using McNemar tests with dichotomous outcome data: contraceptive method awareness, preference and use (yes/no); knowledge (high/low) and likelihood of method switching (‘not’ or ‘slightly’ likely vs ‘quite’ or ‘extremely’ likely).
We used logistic regression with generalised estimating equations accounting for repeated measures to examine the population-estimated change in the proportion of participants who reported a high level of knowledge about LARC and about every method in the video, in addition to the proportional LARC uptake and preference, and the use of a non-prescription/natural primary method(s) post-video. We selected variables that could reasonably be confounders (age group, whether participants were born in Australia, lived in rural or metropolitan locations, held a Healthcare Card (available to people on very low incomes allowing access to cheaper medicines, healthcare services and other concessions) and received social security payments) a priori. We investigated each within univariable models with the outcome. When the univariable output p was <0.20, the variable was included in the multivariable model. Variable inflation factor (VIF) for each variable was inspected, and collinearity identified when VIF values were >10. As all VIF values were <10, collinearity was not detected. The limit of statistical significance was set at a two-sided p value of <0.05.
Results
The advertisement reached 130 129 Facebook users and received over 2000 clicks.15 The screening survey was completed by 462 individuals, 437 (95%) of whom were eligible to participate. A total of 322 women completed S1, S2 and watched the video (74% of 437 eligible respondents).15 The 88% (n=284) of those who completed S3 were similar to those lost to follow-up in terms of demographic characteristics, contraceptive use and preferences as reported by participants at S1 (all variables p>0.05; table 1).
Characteristics of participants at baseline, comparing those who completed survey 3 and those lost to follow-up
Knowledge
Only 17 (6%) participants at S1 reported a high level of knowledge about every method, increasing to 112 (35%) at S2 (p<0.0001; table 2). The multivariable model adjusted for age and social security payments status (table 3) confirmed 10-fold increased odds of reporting a high level of knowledge about every method at S2 (OR 10.0, 95% CI 5.9 to 17.0). Awareness of IUDs and the implant increased by 5% and 6%, respectively (both p<0.001); high knowledge increased by 40% and 29%, respectively (both p<0.0001). The multivariable model adjusted for participant age and whether they were born in Australia (table 3) confirmed the post-video increase in LARC knowledge (OR 4.3, 95% CI 3.1 to 5.6).
Comparison of participant-reported awareness and knowledge about the contraceptive methods discussed in the video, before watching the video (S1) and immediately after (S2)
Population-estimated odds of reporting: high knowledge of all methods compared with low or no knowledge; high self-rated knowledge of long-acting reversible contraception (LARC) compared with low and no knowledge; preference for LARC; current use of LARC; current use of natural/non-prescription primary method compared with that of hormonal or intrauterine methods
Preference
The most preferred methods were COCPs pre-video and IUDs immediately after (table 4). Preference for COCPs and condoms declined post-video (both p<0.01). At S3, preference remained lower than at baseline for condoms (p=0.01) but not COCPs (p=0.2). Preference was consistent throughout the study for all other methods (all p>0.05). The multivariable model adjusted for age (table 3) confirmed increased LARC preference at S2 (OR 1.7, 95% CI 1.4 to 2.1) and S3 (OR 1.4, 95% CI 1.1 to 1.6) although a decline in LARC preference was observed from S2 to S3, with 32 of 99 completing participants who preferred IUDs at S2 reporting a different preference at S3.
Contraceptive preference, use and satisfaction before and after viewing the video
Contraceptive use
At baseline, 9% (n=24) of completing participants reported current IUD use (table 4); all were followed up and still using an IUD at S3. The overall proportion of participants using IUDs increased to 13% at S3 (n=36, p<0.001). Of these S3 IUD users, one was not using contraception at baseline, four used natural/non-prescription primary methods and six used other hormonal methods. Implant use did not change significantly between S1 and S3 (p=0.8). The increased use of LARC overall was confirmed in the model adjusted for participants’ age (OR 1.3, 95% CI 1.1 to 1.5).
Fewer participants at S3 than S2 stated they were very likely to change their contraceptive method/behaviours (p=0.02; table 4). There was no change in the reporting of being unlikely to change contraceptive methods/behaviours between timepoints (p>0.1). High satisfaction among IUD users fell by 9% between S1 and S3, which was the largest decline in satisfaction across all effective contraception methods.
Dual method use declined between S1 and S3 (p<0.001; table 4). The proportion of those using no contraception did not change (p=0.3). Although the use of ECP, withdrawal (both p<0.05) and condoms (p<0.001) decreased, the overall rate of natural/non-prescription method use did not change (p=0.7). This was confirmed in the multivariable model adjusted for whether participants were born in Australia and received social security payments (OR 0.9, 95% CI 0.8 to 1.11).
Discussion
The online contraceptive education video improved young women’s awareness and self-rated knowledge of all contraception methods. Furthermore, IUDs became participants’ most preferred method immediately after the video, replacing their S1 preference for the COCP. We then observed a small but significant increase in IUD uptake 6 months after the intervention, with our adjusted analyses confirming the increased use of LARC overall.
This study contributes to a small and growing evidence base demonstrating the effectiveness of video-based education for improving contraceptive knowledge, preference and use, when delivered via the internet.10 Although many provider-independent interventions (including a previous IUD education video16) increased contraceptive knowledge,17 our provider-independent intervention was unique in that it provided comprehensive, plain-language information about all contraceptive options in the form of a video-based ‘contraceptive counselling session’. This may have influenced our finding of increased IUD use which is contrary to that of most prior online intervention studies18 19: only one previous evaluation of an online intervention (a contraceptive assessment tool) observed increased LARC use.20 Unlike our findings, however, the authors also observed the uptake of other effective methods at follow-up.20
Many of our participants who preferred IUDs at S2 preferred a different method or were not using IUDs at S3. There was also no change in contraceptive use among those not already using prescription methods at S1, but a reduced reported likelihood of future method switching at follow-up. Most previously evaluated, effective video interventions have been delivered in tandem with clinical care, providing facilitated access to contraception post-education10 11 21 and our findings indicate, as demonstrated in prior research,18 19 that the accessibility of contraception after provider-independent or online education may more reliably predict uptake than participant knowledge and preference. Inequities according to age may be inferred by the confounding effect of this variable on contraceptive use outcomes (table 3). Using multiple social media platforms, language translations and an in-depth co-design process22 23 may have improved the reach, relevance and effectiveness of the intervention for different population groups, including those born outside of Australia who were underrepresented among our participants. However, these strategies would not reasonably mitigate systemic barriers to healthcare.
The content of the educational video was based on that used in a training package delivered to general practitioners randomised to provide structured contraceptive counselling and optional rapid referrals to LARC insertion clinics, in previous Australian research.1 The intervention resulted in increased LARC uptake,1 and high satisfaction among LARC users at follow-up, compared with usual care.2 While these findings may infer the appropriateness of the video content to support contraceptive choice, we observed a decline in satisfaction among IUD users across timepoints. LARC generally have high rates of continuation and attributes that make them desirable to many young people including their high effectiveness and generally acceptable side effect profiles.24–26 While videos that emphasise the benefits of LARC (primarily relating to their high effectiveness compared with other reversible methods) may increase LARC uptake,21 we note that the strongest desire to prevent pregnancy does not always correspond to a preference for the most effective methods.27–29 Contraceptive preference immediately after the video may have been biased towards LARC due to the presentation of contraceptive methods in the order of most to least effective. Alternatively, our participants’ increased IUD preference may have occurred because they had never discussed the option of an IUD with their doctor. Misconceptions about IUDs not being suitable for nulliparous women abound not only in the community but also among health professionals in Australia.30
The observed decline in ‘current’ condom use at follow-up may have been due to the phrasing and interpretation of the contraceptive use question in the survey (eg, “Are you currently using condoms?”), and a longer follow-up timeframe would be necessary to verify these findings. Furthermore, participants were required to select their preferred contraceptive method from a choice of the contraceptives described in the video only. These participants whose preferences were not listed may have been disproportionately represented among those lost-to-follow-up.
There are many potential reasons for the non-completion of baseline surveys among the 22% of all those who were deemed eligible during screening but did not participate, including the time commitment required, or a potential moderating effect of literacy on participation. Although we were unable to assess self-selection bias, we obtained our required sample size, participant retention was high and there was no difference between participants who completed and did not complete the study, suggesting a high level of validity. Our findings, however, have limited generalisability to young people of immigrant backgrounds, those who have experienced pregnancy, and those who live in rural areas and Australian states other than Victoria, and this raises additional questions about the appropriateness of Facebook recruitment in population-based research, given the narrow demographic profile of users.
Conclusions
Promoting educational videos on social media is an effective way to increase young women’s awareness and self-rated knowledge about contraception in the short term, in addition to their long-term preference and use of IUDs, and may complement practitioner contraceptive counselling. However, ensuring access to contraception after online education is necessary for contraceptive preferences to be realised.
Ethics statements
Patient consent for publication
Acknowledgments
The authors thank Professor Deborah Bateson for reviewing the manuscript, and staff and students at Monash University for their assistance with the development of the video and surveys.
References
Footnotes
Twitter @PipBuckingham
Contributors DM: guarantor; conceptualisation (equal); methodology (supporting); writing – review and editing (equal). PB: methodology (supporting); writing – original draft (lead); EM: conceptualisation (equal); methodology (lead); writing – original draft (supporting); writing – review and editing (equal). JE: conceptualisation (supporting); methodology (supporting); writing – review and editing (equal).
Funding This work was supported by Bayer HealthCare.
Competing interests DM has received funding and honoraria from Bayer Healthcare. The other authors have no conflicts of interest to declare.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.