If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
RTI International, Women's Global Health Imperative, Berkeley, CA, United StatesCenter for AIDS Pervention Studies, Department of Medicine, University of California San Francisco, San Francisco, CA, United States
RTI International, Women's Global Health Imperative, Berkeley, CA, United StatesCenter for AIDS Pervention Studies, Department of Medicine, University of California San Francisco, San Francisco, CA, United States
Department of Obstetrics, Gynecology & Reproductive Sciences, University of Pittsburgh School of Medicine, Pittsburgh, PA, United StatesMagee-Womens Research Institute, Pittsburgh, PA, USA
The vaginal ring (ring) is a female-initiated, long-acting drug delivery system for different indications, including HIV prevention. Our aim was to provide evidence for acceptability of the vaginal ring across indications to support dapivirine and multipurpose prevention technology ring introduction and roll out.
Study design
This systematic review and meta-analysis followed PRISMA guidelines. We searched PubMed, Web of Science, Embase, and grey literature for publications reporting favorable ring acceptability and secondary outcomes involving actual ring use (comfort, ease of ring use, ring comfort during sex, expulsions, and vaginal symptoms) or hypothetical acceptability for any indication published January 1, 1970−June 15, 2021. We estimated random-effects pooled prevalence, assessing between-study variation using meta-regression.
Results
Of 2,234 records, we included 123 studies with 40,434 actual and hypothetical ring users. The primary outcome assessment included 50 studies with 60 ring subgroups totaling 19,271 ring users. The favorable acceptability pooled prevalence was 85.6% (95%CI 81.3, 89.0), while hypothetical acceptability among non-ring users was 27.6% (95%CI 17.5, 40.5). In meta-regression, acceptability was higher in menopause (95.4%; 95%CI 88.4, 98.2) compared to contraceptive rings (83.7%; 95%CI 75.6, 89.5). Acceptability was lower in pharmacokinetic studies (50%; 95%CI 22.1, 77.9) compared to RCTs (89.5%; 95%CI 85.8.92.4) and in studies assessing acceptability at ≥12 months (78.5%; 95%CI 66.5, 87.1) versus studies assessing acceptability at <3 months (91.9%; 95%CI 83.7, 96.1). European (90.6%; 95%CI 83.9, 94.7), Asian (97.1%; 95%CI 92.0, 99.0), and multi-region studies (93.5%; 95%CI 84.6, 97.4) reported more favorable acceptability compared to African studies (59.4%; 95%CI 38.3, 77.5). Secondary outcomes were similarly favorable, including ring comfort (92.9%; 95%CI 89.2, 95.4), ease of use (90.9%; 95%CI 86.5, 94.0), and comfort during sex (82.7%; 95%CI 76.4, 87.6). Limitations include inconsistent outcome definitions and unmeasured factors affecting acceptability.
Conclusions
Women who used vaginal rings reported they were acceptable across indications geographic regions and indications. Policy makers should consider the ring as an important option for pregnancy and HIV prevention drug development.
Implications
This review found favorable acceptability among vaginal ring users across indications and geographic areas, in contrast to low hypothetical acceptability among non-users. Vaginal rings are an important drug delivery system for pregnancy and HIV preventions, and scale-up should plan to address initial hesitancy among new users.
The vaginal ring (ring) is a drug delivery system providing long-acting, controlled release of an active pharmaceutical ingredient (API) for systemic or localized effects [
]. In March 2021, the World Health Organization released guidance on the dapivirine ring for HIV prevention, after including the dapivirine ring on the prequalification list of medicines in November 2020 [
IPM’s New Drug Application for Dapivirine Vaginal Ring to Reduce HIV Risk in Women Accepted for Filing by US Food and Drug Administration [press release].
]. Rings are also in clinical trials as multipurpose prevention technologies (MPTs), protecting against multiple sexual and reproductive health risks, such as unintended pregnancy and HIV [
]. Interest in rings has grown as research has demonstrated increased contraceptive options and access to longer-acting methods are associated with improved sexual and reproductive health outcomes [
]. Novel, woman-controlled technologies, including rings, provide options for long-acting protection from sexual and reproductive health risks and, unlike implants or intrauterine devices (IUDs), allow women to initiate or discontinue use without involving a healthcare provider.
Ring users have expressed preferences for long-acting, partner-approved methods preventing both HIV and pregnancy, with few side effects, and the potential for use without partner knowledge [
]. Qualitative systematic reviews suggest most women are satisfied with rings, find them easy to use, and report infrequent expulsions and other adverse events [
Given the importance of providing options for women to protect themselves against multiple sexual and reproductive health risks, we conducted a meta-analysis to summarize current knowledge about acceptability of the vaginal ring as a drug-delivery system, agnostic of the API. We estimated the proportion of women finding the ring acceptable and examined related acceptability constructs such as ease of use, comfort, and ring expulsions [
]. We examine rings used globally and for all indications; address the need for quantitative estimates of ring acceptability; and, investigate the association of study and ring characteristics with primary and secondary acceptability outcomes via meta-regression. Our aim is to examine vaginal ring acceptability across indications to provide evidence for the potential acceptability of the dapivirine ring and other rings under development as multipurpose prevention technologies.
]. We included studies reporting primary quantitative data on relevant acceptability outcomes; with women of any age; assessing active, placebo, or hypothetical rings for any indication; and, published in English between January 1, 1970 and June 15, 2021. The review was completed July 5, 2021. Due to our interest in ring acceptability as a drug-delivery platform, we excluded references only reporting effectiveness or API-related complications.
We searched PubMed, Web of Science, and Embase using a predetermined search strategy and searched grey literature including conference proceedings and reports. We identified additional references via a hand-search of bibliographies, a web search, and through a predetermined list of experts. We developed a search strategy with an RTI International librarian and can be found in the appendix (S2 File). Two authors (KR and JG) conducted screening and data extraction independently in Covidence. The same authors (KR and JG) resolved disagreements via consensus and contacting publication authors where necessary. This review utilized summary estimates.
One primary (KR) and one secondary reviewer (JG) assessed risk of bias for all studies, using a 10-item risk of bias tool; [
] each item was scored as "low" or "high” risk (1 or 0 points, respectively). We summed the four items pertaining to internal validity and six items pertaining to external validity; and, due to little variation in the internal validity score, combined them to create a single continuous score with higher values indicating greater validity. We did not exclude studies based on risk of bias and included the validity score as a covariate in meta-regression models. Due to the wide variety of study designs and the assessment of multiple outcomes, we did not judge the overall body of evidence using an approach such as GRADE.
We extracted descriptive data (proportions, means, and standard deviations) for our primary acceptability outcome, “general acceptability”; for secondary acceptability outcomes, including study completion, ease of use, comfort, comfort during sex; and, for secondary unfavorable ring outcomes likely to be associated with lack of acceptability including ring expulsions and vaginal symptoms. Acceptability outcome measures are summarized below. We extracted study-level data, including country, study design, population and setting, and sample size, ring indication, and ring attributes. We classified progestogen-only rings for postpartum contraception separately from other contraception rings, given their unique indication and population. To account for within-study variation in ring characteristics, the unit of analysis is unique ring groups within studies.
We defined the primary outcome of general acceptability as “the extent to which people…receiving a healthcare intervention [vaginal ring] consider it to be appropriate, based on anticipated or experiential cognitive and emotional responses to the intervention” [
]. During data extraction, we recorded the proportion of users selecting a favorable response regarding acceptability (“very” or “somewhat acceptable”; “acceptable”); satisfaction (“extremely,” “very,” or “somewhat satisfied”; “satisfied”); would recommend (“would definitely” or “would probably recommend”; “would recommend”); would use in the future (“would like to” or “intended to” continue use after study); or, liked (liked “very much” or “somewhat”; “liked”). Secondary acceptability outcomes included: (1) hypothetical acceptability, defined as acceptability of a product described to potential users, lacking the context and preferences of real-life use [
]. The proportion of potential users selecting, following description of a hypothetical ring product, that they were “likely” or “very likely” to use; “wanted to use” or “were willing to use”; or “would be interested in using”; (2) comfort, defined as the proportion of users selecting a favorable response regarding comfort (“comfortable” or “very comfortable;” “agreeing” or “strongly agreeing” the ring was comfortable; “never” or “only occasionally” felt the ring was uncomfortable; no vaginal discomfort); and, not feeling the ring (“never” or “rarely” felt; did not feel; no foreign-body sensation); (3) ease of use, defined as the proportion of users selecting a favorable response regarding ease of insertion (insertion “easy,” “always easy,” or “mostly/fairly easy”; insertion was “not at all a problem”; no difficulty with insertion); and, ease of use (insertion and removal easy or unproblematic; “very easy to use”; no difficulty inserting and/or removing); (4) comfort during sex, defined as the proportion of users selecting a favorable response regarding not feeling during sex (“never” or “rarely” felt during sex); comfort during sex (“comfortable” during sex; no discomfort during sex); partner not feeling during sex (partner did not feel during sex; partner did not experience discomfort during sex); and, no interference with sex (did not interfere with sex; not bothersome during sex). Secondary unfavorable ring outcomes were defined as: (1) expulsions defined as the cumulative proportion of participants reporting a ring expulsion during the study period; and (2) vaginal symptoms defined as the proportion of participants reporting discharge (discharge; leukorrhea); vaginal irritation; and, vaginitis (vaginitis; colpitis).
We extracted data on cumulative product continuation and product-related discontinuation over the entire study period to assess the correlation of each outcome with general acceptability. Product continuation was defined as the proportion of users who initially agreed to participate continuing to use the study product at the last study assessment. Product-related discontinuation was defined as the proportion of users not completing the study due to discontinuations related to the ring (i.e. adverse events, discomfort) vs discontinuations due to change in need (i.e. pregnancy intention) or client situation (i.e. moved, personal reasons).
We used Stata v.15 metapreg package to conduct random-effects meta-analysis for primary and secondary acceptability outcomes [
]. Meta-analysis derives an overall mean score from individual studies (here, ring groups), weighting each group by its relative precision. The results for our primary outcome are summarized in a forest plot showing the reported proportions and 95% confidence intervals by group. For each outcome, heterogeneity between estimates was measured using the I² statistic, with 50% and 75% indicating substantial and considerable heterogeneity, respectively [
]. Studies with conceptually similar outcomes were combined for meta-analysis. If studies reported the same outcome over multiple timepoints (e.g. at 3 months, 6 months, and 12 months), we included the first timepoint after initiation of ring use, as results from multiple timepoints cannot be analyzed in standard meta-analysis; the earliest sample is the most representative as it captures ring users prior to discontinuation or loss to follow up; and, ring users have reported having the most difficulty with the ring early in use. For product continuation and product-related discontinuation, we reported cumulative continuation and discontinuation over the entire study period.
Between-study variation was investigated using random-effects meta-regression, which estimates the effects of ring group-level covariates on outcome proportions. The coefficients estimate the impact of a covariate on the logit of the outcome (S3 File). We conducted meta-regression of ring- and study-level covariates for all outcomes except for hypothetical acceptability (n=12 studies); as such, the pooled prevalence of hypothetical acceptability was compared descriptively to that of general acceptability among ring users. Analyses examined associations between ring group-specific outcomes and the study-level covariates (S6 File). As this research intends to inform research and implementation for the dapivirine ring and other rings for HIV prevention, we used North and sub-Saharan Africa, a high-priority setting, as the referent group for geographic region. We captured data on several physical attributes of rings and included ring diameter in our analyses since it was least frequently missing. We calculated the relative reduction in between-study variance using τ2 in meta-regression with all covariates compared to τ2 in a model with no covariates [
]. We assessed whether data collection method (face-to-face vs self-reported) influenced outcomes but excluded it from meta-regression due to lack of change in R2 values. We gathered data on shore hardness (a measure of ring material resistance to indentation) and cross-sectional diameter but excluded these from meta-regression due to high correlations between both features and ring outer diameter. We assessed publication bias across studies using funnel plots and Egger's test for small-study effects (S8 File) [
Our search identified 2,234 records, of which 90 were duplicates. Following title and abstract screening, we excluded 1,869 records; and, in full text screening, we excluded 133 records: 71 records were excluded due to reporting irrelevant outcome data, 16 for not reporting acceptability outcomes, 16 for reporting partial data captured fully elsewhere, and 30 for other reasons such as reporting only qualitative data. Following screening, 151 publications describing 123 studies with 34,740 ring users and 5,694 hypothetical users were included. Twenty-five studies were excluded from meta-analysis due to reporting product continuation without also reporting general acceptability [
Local treatment of urogenital atrophy with an estradiol-releasing vaginal ring: a comparative and a placebo-controlled multicenter study. Vaginal Ring Study Group.
Effects on cycle control and bodyweight of the combined contraceptive ring, NuvaRing, versus an oral contraceptive containing 30 microg ethinyl estradiol and 3 mg drospirenone.
Superior cycle control with a contraceptive vaginal ring compared with an oral contraceptive containing 30 microg ethinylestradiol and 150 microg levonorgestrel: a randomized trial.
] (Fig. 1). Some secondary manuscripts of the included studies assessed acceptability across ring groups with different characteristics (e.g. active and placebo rings) and were not included in meta-analysis.
Fig 1PRISMA flowchart of publication screening and inclusion for systematic review and meta-analysis.
Local treatment of urogenital atrophy with an estradiol-releasing vaginal ring: a comparative and a placebo-controlled multicenter study. Vaginal Ring Study Group.
Effects on cycle control and bodyweight of the combined contraceptive ring, NuvaRing, versus an oral contraceptive containing 30 microg ethinyl estradiol and 3 mg drospirenone.
Efficacy, acceptability and tolerability of the combined contraceptive ring, NuvaRing, compared with an oral contraceptive containing 30 microg of ethinyl estradiol and 3 mg of drospirenone.
Transdermal estrogen with a levonorgestrel-releasing intrauterine device for climacteric complaints versus estradiol-releasing vaginal ring with a vaginal progesterone suppository: clinical and endometrial responses.
A comparative multicenter study of the effects of continuous low-dose estradiol released from a new vaginal ring versus estriol vaginal pessaries in postmenopausal women with symptoms and signs of urogenital atrophy.
A one-year multicenter study of efficacy and safety of a continuous, low-dose, estradiol-releasing vaginal ring (Estring) in postmenopausal women with symptoms and signs of urogenital aging.
Pharmacokinetic bioequivalence, safety and acceptability of Ornibel((R)), a new polymer composition contraceptive vaginal ring (etonogestrel/ethinylestradiol 11.00/3.474 mg) compared with Nuvaring((R)) (etonogestrel/ethinylestradiol 11.7/2.7 mg).
A comparative study of safety and efficacy of continuous low dose oestradiol released from a vaginal ring compared with conjugated equine oestrogen vaginal cream in the treatment of postmenopausal urogenital atrophy.
Comparison of combined hormonal vaginal ring with ultralow-dose combined oral contraceptive pills in the management of heavy menstrual bleeding: A pilot study.
End-user preference for and choice of four vaginally delivered HIV prevention methods among young women in South Africa and Zimbabwe: the Quatro Clinical Crossover Study.
An open-label, randomized crossover study to evaluate the acceptability and preference for contraceptive options in female adolescents, 15 to 19 years of age in Cape Town, as a proxy for HIV prevention methods (UChoose).
HIV prevention at drug shops: awareness and attitudes among shop dispensers and young women about oral pre-exposure prophylaxis and the dapivirine ring in Shinyanga, Tanzania.
Superior cycle control with a contraceptive vaginal ring compared with an oral contraceptive containing 30 microg ethinylestradiol and 150 microg levonorgestrel: a randomized trial.
Clinical performance and menstrual bleeding patterns with three dosage combinations of a Nestorone progestogen/ethinyl estradiol contraceptive vaginal ring used on a bleeding-signaled regimen.
]. Sixty-three studies evaluated rings used for contraception, 21 for menopause, 21 for HIV prevention, and five for MPT; of the 123 studies, 58 were RCTs, 58 were observational studies, and 7 were pharmacokinetic studies. The median assessment time was 4 months (interquartile range [IQR]: 3, 12) (Table 1 and S4A and B File). The median internal threat to validity score was 6 (IQR: 6, 6); the median external validity score was 1 (IQR: 1, 2). Threats to external validity included lacking a sampling frame representative of the target population, not using random selection for recruitment, and having high non-response bias; threats to internal validity included not using validated measures (S4C File).
Table 1Studies of vaginal ring acceptability included in meta-analysis
Ishaku, Salisu Mohammed, Afolabi Kayode, Godwin Unumeri, Ayodeji Oginni, Adekunle Adeyemi, Deepa Rajamani, Heather Clark, Naomi Rijo, and Saumya RamaRao. 2015. Progesterone vaginal ring: Results of an acceptability study in Nigeria. New York: Population Council.
Efficacy, acceptability and tolerability of the combined contraceptive ring, NuvaRing, compared with an oral contraceptive containing 30 microg of ethinyl estradiol and 3 mg of drospirenone.
Pharmacokinetic bioequivalence, safety and acceptability of Ornibel((R)), a new polymer composition contraceptive vaginal ring (etonogestrel/ethinylestradiol 11.00/3.474 mg) compared with Nuvaring((R)) (etonogestrel/ethinylestradiol 11.7/2.7 mg).
An open-label, randomized, multi-center trial to evaluate continuation rates, side effects and acceptability of the contraceptive ring vs. the contraceptive patch in women previously using combined oral contraceptives.
Single-dose pharmacokinetics and pharmacodynamics assessment of oestriol and trimegestone containing vaginal rings in healthy women with childbearing potential.
Efficacy and safety of the contraceptive vaginal ring (NuvaRing) compared with a combined oral contraceptive in Chinese women: a 1-year randomised trial.
Comparison of efficacy, compliance, acceptability and tolerability of hormonal contraceptive vaginal ring (NuvaRing) versus combined oral contraceptives (COCs) - a prospective randomized controlled study in a district level hospital in East Delhi.
Microdose intravaginal levonorgestrel contraception: a multicentre clinical trial. I. Contraceptive efficacy and side effects. World Health Organization. Task Force on Long-Acting Systemic Agents for Fertility Regulation.
Efficacy of the 1-year (13-cycle) segesterone acetate and ethinylestradiol contraceptive vaginal system: results of two multicentre, open-label, single-arm, phase 3 trials.
Clinical performance and menstrual bleeding patterns with three dosage combinations of a Nestorone progestogen/ethinyl estradiol contraceptive vaginal ring used on a bleeding-signaled regimen.
An open-label, randomized crossover study to evaluate the acceptability and preference for contraceptive options in female adolescents, 15 to 19 years of age in Cape Town, as a proxy for HIV prevention methods (UChoose).
End-user preference for and choice of four vaginally delivered HIV prevention methods among young women in South Africa and Zimbabwe: the Quatro Clinical Crossover Study.
HIV prevention at drug shops: awareness and attitudes among shop dispensers and young women about oral pre-exposure prophylaxis and the dapivirine ring in Shinyanga, Tanzania.
Preferred Product Attributes of Potential Multipurpose Prevention Technologies for Unintended Pregnancy and Sexually Transmitted Infections or HIV Among U.S. Women.
Tenofovir disoproxil fumarate intravaginal ring for HIV pre-exposure prophylaxis in sexually active women: a phase 1, single-blind, randomised, controlled trial.
Acceptability of microbicidal vaginal rings and oral pre-exposure prophylaxis for HIV prevention among female sex workers in a high-prevalence US city.
Randomized, placebo controlled phase I trial of safety, pharmacokinetics, pharmacodynamics and acceptability of tenofovir and tenofovir plus levonorgestrel vaginal rings in women.
Safety and pharmacokinetics of single, dual, and triple antiretroviral drug formulations delivered by pod-intravaginal rings designed for HIV-1 prevention: A Phase I trial.
Transdermal estrogen with a levonorgestrel-releasing intrauterine device for climacteric complaints versus estradiol-releasing vaginal ring with a vaginal progesterone suppository: clinical and endometrial responses.
A comparative multicenter study of the effects of continuous low-dose estradiol released from a new vaginal ring versus estriol vaginal pessaries in postmenopausal women with symptoms and signs of urogenital atrophy.
A one-year multicenter study of efficacy and safety of a continuous, low-dose, estradiol-releasing vaginal ring (Estring) in postmenopausal women with symptoms and signs of urogenital aging.
A comparative study of safety and efficacy of continuous low dose oestradiol released from a vaginal ring compared with conjugated equine oestrogen vaginal cream in the treatment of postmenopausal urogenital atrophy.
Patient experience in a randomized trial of a weekly progesterone vaginal ring versus a daily progesterone gel for luteal support after in vitro fertilization.
Comparison of combined hormonal vaginal ring with ultralow-dose combined oral contraceptive pills in the management of heavy menstrual bleeding: A pilot study.
A Randomised Controlled Trial Comparing the Efficacy and Side-Effects of Intravaginal Ring (Nuvaring((R))) With Combined Oral Hormonal Preparation in Dysfunctional Uterine Bleeding.
Comparison of combined hormonal vaginal ring and low dose combined oral hormonal pill for the treatment of idiopathic chronic pelvic pain: a randomised trial.
Progesterone vaginal ring as a new contraceptive option for lactating mothers: Evidence from a multicenter non-randomized comparative clinical trial in India.
aOutcomes for vaginal ring acceptability are reported as percentages. For studies reporting outcomes for more than one vaginal ring, the outcome percentages are reported separately for each vaginal ring (e.g. VR1, VR2, etc.).
b Nuvaring (Merck, USA) attributes: Ethylene vinyl acetate copolymers and magnesium stearate; 54 × 4mm; 0.120 mg/d etonogestrel and 0.015 mg/d ethinyl estradiol; continuous use for 21days., 7 ring-free day.
c Ornibel (Exeltis Healthcare, Spain) attributes: Ethylene vinyl acetate; 54 × 4mm; 0.120 mg/d. etonogestrel and 0.015 mg/d. ethinyl estradiol; continuous use for 21days, 7 ring-free days.
d Ring-004 (International Partnership for Microbicides, USA) attributes: Platinum-catalyzed silicone; 56 × 7.7mm; 25 mg DPV; continuous use for 1 month.
e MK-2048A (MTN-28) attributes: Ethylene vinyl acetate; 54 × 4mm; 182 mg of vicriviroc, 30 mg of MK-2048; continuous use for 28 days.
f Femring/Menoring (Galen Holdings, Northern Ireland) attributes: Silicone elastomer, 56 × 7.6mm, 5 or 10 mcg/d estradiol, continuous use for 90 days.
gEstring (Pfizer, USA) attributes: Silicone elastomer, 55 × 9mm, 7.5 mcg/d estradiol, continuous use for 90 days.
In meta-analysis, the crude pooled prevalence of general acceptability was 85.6% (95%CI 81.3, 89.0) from 50 studies (60 ring groups) with 19,271 ring users (Fig. 2). In contrast to the high general acceptability found among ring users, hypothetical studies not based on actual ring experience reported low acceptability (27.6%; 95%CI 17.5, 40.5) (S5A Figure). Secondary outcomes were similarly high, including ring comfort (92.9%; 95%CI 89.2, 95.4), ease of use (90.9%; 95%CI 86.5, 94.0), and comfort during sex (82.7%; 95%CI 76.4, 87.6) (S5B-F Figure).
Fig 2Forest plot of vaginal ring acceptability from meta-analysis.
Ring expulsions (9.1%; 95% CI 6.8, 12.3) and vaginal symptoms (11.3%; 95%CI 8.3, 15.1) were relatively infrequent. We observed substantial to considerable heterogeneity for all outcomes (I2: 54.1%−94.5%). Adjusted estimates of pooled prevalence from meta-regression fell within the confidence limits of the unadjusted estimates for all outcomes (Table 2).
Table 2Unadjusted and adjusted pooled prevalence estimates for each vaginal ring (ring) acceptability outcome
Adjusted pooled prevalence from meta-regression model adjusting for region, outcome type (coding dependent on outcomes reported, see Methods section for details), validity score, study design, publication year, outcome measurement timepoint, ring indication, presence of an API, and ring diameter.
CI,confidence interval; NA,not applicable; ring,vaginal ring
a Unadjusted pooled prevalence estimated from meta-analysis.
b Adjusted pooled prevalence from meta-regression model adjusting for region, outcome type (coding dependent on outcomes reported, see Methods section for details), validity score, study design, publication year, outcome measurement timepoint, ring indication, presence of an API, and ring diameter.
We conducted meta-regression to identify characteristics associated with the primary outcome, general acceptability (Table 3), secondary acceptability outcomes (Table 4), and unfavorable ring outcomes (Table 5). Full results for meta-regression models can be found in the appendix (S6 File) along with summary figures (S7 File). The proportion of total between-study variance explained in the adjusted meta-regression model ranged from 52% to 75%.
Table 3Results of meta-regression models for the primary outcome of general vaginal ring (ring) acceptability, reporting values of the adjusted proportion, the 95% confidence interval of the adjusted proportion, the beta coefficient (β), the standard error (SE) of the beta coefficient, and the p-value for each study-level covariate
Proportion (95% CI)
β (SE)
p-value
Overall acceptability outcome type
Satisfaction
85.7 (76.8, 91.6)
REF
REF
Acceptable
72.2 (52.6, 85.8)
-0.84 (0.57)
0.139
Recommend
94.9 (89.6, 97.6)
1.14 (0.47)
0.016
Would continue
84.6 (66.2, 93.9)
-0.09 (0.69)
0.892
Like
84.5 (61.3, 94.9)
-0.10 (0.79)
0.900
Region
Africa
59.4 (38.3, 77.5)
REF
REF
Europe
90.6 (83.9, 94.7)
1.89 (0.58)
0.001
Americas and Australia
79.5 (71.4, 85.7)
0.98 (0.53)
0.067
Asia
97.1 (92.0, 99.0)
3.13 (0.66)
<0.001
Multi-region
93.5 (84.6, 97.4)
2.28 (0.67)
<0.001
Validity score
NA
0.15 (0.15)
0.308
Study design
RCT
89.5 (85.8, 92.4)
REF
REF
Observational
83.9 (74.8, 90.1)
-0.50 (0.36)
0.169
Pharmacokinetic
50.0 (22.1, 77.9)
-2.15 (0.69)
0.002
Publication year
NA
0.04 (0.03)
0.145
Assessment time
<3 mo
91.9 (83.7, 96.1)
REF
REF
3−11 mo
84.9 (79.8, 89.0)
-0.69 (0.47)
0.140
12 mo or more
78.5 (66.5, 87.1)
-1.12 (0.56)
0.044
Indication
Contraception
83.7 (75.6, 89.5)
REF
REF
HIV/MPT
83.7 (65.6, 93.2)
-0.01 (0.68)
0.994
Menopause
95.4 (88.4, 98.2)
1.38 (0.60)
0.020
Other hormonal
67.2 (44.6, 84.0)
-0.92 (0.52)
0.077
Ring API
Ring with API
86.8 (83.5, 89.5)
REF
REF
Ring with no API
68.7 (42.1, 86.9)
-1.1 (0.60)
0.066
Ring diameter
≤55mm
82.3 (77.3, 86.3)
REF
REF
>55 mm
92.4 (86.6, 95.8)
0.96 (0.39)
0.015
Reduction in R2
0.523
Likelihood ratio Χ2, p-value
404.98, p<0.01
API, active pharmaceutical ingredient; MPT, multipurpose prevention technology; RCT, randomized controlled trial; Ring, vaginal ring; REF, reference.
Significant associations (p<0.05) are indicated in bold typeface
Table 4Results of meta-regression models for the secondary vaginal ring (ring) acceptability outcomes, reporting values of the adjusted proportion, the 95% confidence interval of the adjusted proportion, the beta coefficient (β), the standard error (SE) of the beta coefficient, and the p-value for each study-level covariate
Comfort
Ease of use
Comfort During Sex
Proportion (95% CI)
β (SE)
p-value
Proportion (95% CI)
β (SE)
p-value
Proportion (95% CI)
β (SE)
p-value
Outcome Type
Comfort: 96.7 (94.9, 97.8) Not feeling ring: 73.6 (62.9, 82.1)
Comfort: REF Not feeling ring: -2.34 (0.40)
Comfort: REF Not feeling ring: <0.001
Ease of insertion: 92.3 (89.6, 94.3) Ease of use: 71.8 (51.3, 86.0)
Ease of insertion: REF Ease of use: -1.54 (0.49)
Ease of insertion: REF Ease of use: 0.002
Did not feel: 87.1 (83.9, 89.6) Comfort: 87.8 (80.3, 92.7) No sex interference: 88.6 (85.8, 91.0) Partner did not feel: 53.9 (43.4, 64.1)
Did not feel: REF Comfort: 0.06 (0.32) No sex interference: 0.15 (0.18) Partner did not feel: -1.75 (0.29)
Did not feel: REF Comfort: 0.841 No sex interference: 0.415 Partner did not feel: <0.001
Table 5Results of meta-regression models for the unfavorable vaginal ring (ring) outcomes, reporting values of the adjusted proportion, the 95% confidence interval of the adjusted proportion, the beta coefficient (β), the standard error (SE) of the beta coefficient, and the p-value for each study-level covariate
In meta-regression, the primary outcome of general acceptability was associated with study and ring characteristics (Table 3). Outcome proportions varied by operationalization of general acceptability, with "would recommend" higher than "satisfaction.” General acceptability was higher in the Europe, Asia Pacific and multi-region studies compared to African studies; in menopausal rings compared to contraceptive rings; in API rings compared to placebo rings; in studies with earlier assessment of acceptability (<3 months) compared to later acceptability assessment (12 months or later); and, in larger rings (>55 mm) compared to smaller rings (≤55 mm). General acceptability was significantly lower in pharmacokinetic studies compared to RCTs.
3.2. Secondary acceptability outcomes
Secondary acceptability outcomes, including comfort, ease of use, and comfort during sex, were also associated with study and ring characteristics in meta-regression (Table 4). Outcome proportions varied by operationalization of comfort, ease of use, and comfort during sex. Secondary acceptability outcome proportions varied by region, with higher comfort in the Europe, Asia, and multi-region studies; lower ease of use in European, Asian and multi-region studies; and lower comfort during sex in the Americas/Australia and in multi-region studies, compared to African studies. Later assessment times were associated with higher ease of use and comfort during sex, but the relationship between comfort and assessment time was unclear. Compared to contraceptive rings, HIV/MPT rings had higher ease of use and menopause rings had higher comfort during sex and lower overall comfort. Rings with no API had higher comfort compared to rings with an API. Rings with a diameter >55 mm were less comfortable and more difficult to use compared rings with a diameter ≤55 mm.
3.3. Secondary unfavorable ring outcomes
Unfavorable ring outcomes, including expulsions and vaginal symptoms, were associated with study and ring characteristics in meta-regression (Table 5). The validity score was negatively associated with expulsions. Vaginal symptoms were higher in Europe, the Americas and Australia, Asia, and in multi-region studies compared to Africa. Expulsions were higher for observational studies compared to RCTs. Observational studies and pharmacokinetic studies had lower vaginal symptoms compared to RCTs. HIV/MPT rings had higher vaginal symptoms, and other hormonal rings had lower vaginal symptoms compared to contraceptive rings. Rings with larger diameters had higher expulsions and vaginal symptoms compared to rings with smaller diameters.
3.4. Sensitivity analysis
Visual inspection and Egger's tests demonstrated evidence of publication bias in the comfort during sex and expulsions models, which were explored in sensitivity analysis. Egger's tests remained statistically significant after excluding one outlier for comfort during sex, and 2 outliers for expulsions (S8 File).
4. Discussion
This study assessed global acceptability of the vaginal ring across indications, including general acceptability and related constructs, such as reported comfort, ease of use, and expulsions. A fundamental expression of acceptability among end-users is essential to the success of the ring for any health indication. This analysis comes in the context of the November 2020 inclusion of the dapivirine ring for HIV prevention on the prequalification list of medicines by the WHO [
IPM’s New Drug Application for Dapivirine Vaginal Ring to Reduce HIV Risk in Women Accepted for Filing by US Food and Drug Administration [press release].
]. Our analysis offers several critical insights about ring acceptability and use that can inform future research and ring scale-up activities globally.
Most women who used the ring liked it, irrespective of geographic location, indication, or other factors. However, women who had not actually used the ring reported low hypothetical acceptability; and, it is likely that women with low hypothetical acceptability do not enter ring trials. Although the ring has been used as a drug delivery system for decades, it constitutes a novel technology for many due to limited global distribution. Lack of awareness, familiarity, and experience likely contribute to the lower hypothetical acceptability of the ring [
]. We found increased duration of ring experience was associated with higher ease of use and comfort during sex. This is supported by studies indicating acceptability increases following ring use [
Young Women's Ratings of Three Placebo Multipurpose Prevention Technologies for HIV and Pregnancy Prevention in a Randomized, Cross-Over Study in Kenya and South Africa.
End-user preference for and choice of four vaginally delivered HIV prevention methods among young women in South Africa and Zimbabwe: the Quatro Clinical Crossover Study.
], and the existence of a “learning curve”, where ring users were initially apprehensive and required peer and provider support to become more consistent and proficient users [
]. Consequently, scale up of the dapivirine or other rings must include recognition of, and planning for, hesitancy among women around uptake and initiation. Investments in overcoming initial hesitancy may result in greater adherence and persistence once initiated, a conclusion that is supported by recent positive findings of high continued use of the dapivirine vaginal ring in the DREAM and HOPE open-label extension studies and the REACH trial among adolescent girls and young women [
] and was operationalized differently across studies. In the models exploring overall acceptability, ease of use, and comfort during sex, there were different results depending on outcome operationalization. The measurement of acceptability constructs may have important implications for data consistency and interpretation [
] that use standardized, self-reported measures of acceptability and attitudes regarding the intervention across multiple acceptability domains and include both anticipated and experienced acceptability. Careful translation and cognitive pretesting are necessary to ensure studies are adequately capturing acceptability constructs, particularly in multinational or multicultural studies [
There was some regional variation in key outcomes, as most regions had higher general acceptability compared to Africa. The Americas and Australia were associated with less favorable comfort during sex and vaginal symptoms, but less frequent expulsions. Studies conducted in Asia reported less favorable ease of use and more frequent vaginal symptoms. Multi-region studies were associated less favorable comfort during sex and more frequent vaginal symptoms. This variation could be related to the proportion of users with previous experiences with vaginal products [
]. Ring acceptability across regions and cultures should continue to be assessed with ring scale-up, particularly as regulatory bodies in sub-Saharan Africa review the dapivirine ring and prepare for introduction and roll-out of the dapivirine ring and future multipurpose prevention technologies.
General acceptability and several other outcomes varied by indication. Indication-specific variations could be related to the purpose of the ring (e.g. whether it was for treatment or prevention), perceived risk, and/or the availability of alternative dosing forms. For example, general acceptability was highest for menopausal indications, where rings are used for symptom management and are indicative of a situation where dosing alternatives are limited compared to contraceptive indications. The interplay between the purpose of the ring, perceived risk, and the availability of alternative options, will change over time as new delivery forms are developed and approved for use.
Comfort was higher for placebo rings than for API rings, potentially indicating that API plays a role in user perceptions of ring comfort. While APIs may influence ring acceptability, and this review did not assess API-related adverse events, it is worth considering that vaginal insertion presents less systemic drug exposure to potent drugs like those for HIV PrEP, minimizing systemic side effects [
]. Of note, reported vaginal symptoms were not higher among the API subgroups, and the pooled prevalence of vaginal symptoms was relatively low (11.3%).
For some acceptability outcomes, the ring's API may favorably influence user perceptions of acceptability, making it challenging to separate the effects of the ring's API from the delivery system itself. For example, the finding that comfort during sex was more favorable for menopausal rings (vs contraceptive rings) may reflect these biases, as menopausal rings may be prescribed to treat menopause-related vaginal dryness which is largely evident during sex. Rings with an outer diameter >55mm were associated with higher general acceptability, but lower ease of use, lower comfort, and higher expulsions. This finding suggests physical properties of the ring matter, with the enhanced risk of expulsions suggesting a biophysical component. Ring properties are an important consideration for product developers, and this finding offers insight into an attribute that might still be optimized for rings under development. We found several ring characteristics were highly colinear, including outer diameter, cross-sectional diameter, and shore hardness; thus, in our analysis, we only included dichotomized outer diameter in meta-regression models. Further research is needed to elucidate which ring characteristics influence acceptability and should consider features not included in this analysis such as tensile strength and shore hardness. To date there are limited studies reporting a comparison of acceptability and use experiences with rings having different properties in the same population [
This analysis has several limitations. We prespecified covariates investigated as potential sources of statistical heterogeneity; however, factors not amenable to meta-regression such as user age, education level, and cultural norms could also be important predictors of acceptability. The small number of studies reporting hypothetical acceptability (n=12) precluded our ability to assess sources of between-study variation in meta-regression. There were issues with assessing fit statistics for two meta-regression models, limiting our ability to assess the extent to which heterogeneity was explained. Longitudinal studies of ring acceptability may be prone to bias as dissatisfied users drop out over time; however, we did not find statistically significant correlations either between acceptability and study completion among or between acceptability and product-related discontinuation among studies reporting both outcomes (S9 File). While we included unpublished material in our search strategy, we found evidence of publication bias for two outcomes. Meta-regression identifies factors associated with the outcome of interest but does not establish causality; therefore, these findings should be interpreted as hypothesis-generating.
In conclusion, our review provides strong support for the acceptability and favorable use experience of rings used for diverse indications in women globally. Interventions with high acceptability are more likely to lead to persistent and adherent use [
], having the potential to have an important public health impact.
Declaration of Competing Interest
SA has received consulting fees from Mayne Pharma and research grants from Mithra, Evofem, and Merck that are managed by Magee-Womens Research Institute. The other authors declare no competing interests.
Funding
This review was funded by the Microbicide Trials Network (MTN). The MTN is funded by the National Institute of Allergy and Infectious Diseases (UM1AI068633, UM1AI068615, UM1AI106707), with co-funding from the Eunice Kennedy Shriver National Institute of Child Health and Human Development and the National Institute of Mental Health, all components of the U.S. National Institutes of Health. This review builds upon a previous systematic review supported by the OPTIONS Consortium, a program made possible by the generous assistance from the American people through the U.S. Agency for International Development (USAID) and the U.S. President's Emergency Plan for AIDS Relief (PEPFAR). Financial assistance was provided by USAID (https://www.usaid.gov/) to FHI 360 under the terms of Cooperative Agreement No. AID-OAA-A-15-00035. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health, USAID, PEPFAR, or the United States Government. The funders did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Acknowledgments
Jennifer Griffin was affiliated with Center for Global Health at RTI International at the time of this analysis and is currently affiliated with Acute Communicable Disease Control at Los Angeles County Department of Public Health. Ariane van der Straten was affiliated with the Women's Global Health Imperative at RTI International, at the time of this analysis and is currently an independent consultant with ASTRA consulting, Kensington, CA. Sharon Achilles was affiliated with Magee-Womens Research Institute at the time of this analysis and is currently affiliated with the Bill & Melinda Gates Foundation, Seattle, WA.
WHO recommends the dapivirine vaginal ring as a new choice for HIV prevention for women at substantial risk of HIV infection [press release].
World Health Organization,
Geneva, Switzerland2021 (Available from https://www.who.int/news/item/26-01-2021-who-recommends-the-dapivirine-vaginal-ring-as-a-new-choice-for-hiv-prevention-for-women-at-substantial-risk-of-hiv-infection, [Accessed 1 August 2021])
Vaginal ring to reduce the risk of HIV infection for women in non-EU countries with high disease burden [press release].
EMA,
Amsterdam, Netherlands2020 (Available from https://www.ema.europa.eu/en/news/vaginal-ring-reduce-risk-hiv-infection-women-non-eu-countries-high-disease-burden, [Accessed 1 August 2021])
#BREAKING: Dapivirine, Vaginal Ring Approved for Use in Zimbabwe.
HealthTimes,
2021 (Available from https://healthtimes.co.zw/2021/07/14/breaking-dapivirine-vaginal-ring-approved-for-use-in-zimbabwe/, [Accessed 1 August 2021])
IPM’s New Drug Application for Dapivirine Vaginal Ring to Reduce HIV Risk in Women Accepted for Filing by US Food and Drug Administration [press release].
IPM,
Silver Spring, MD, USA2021 (Available from https://www.ipmglobal.org/content/ipm%E2%80%99s-new-drug-application-dapivirine-vaginal-ring-reduce-hiv-risk-women-accepted-filing-us, [Accessed 1 August 2021])
Local treatment of urogenital atrophy with an estradiol-releasing vaginal ring: a comparative and a placebo-controlled multicenter study. Vaginal Ring Study Group.
Effects on cycle control and bodyweight of the combined contraceptive ring, NuvaRing, versus an oral contraceptive containing 30 microg ethinyl estradiol and 3 mg drospirenone.
Superior cycle control with a contraceptive vaginal ring compared with an oral contraceptive containing 30 microg ethinylestradiol and 150 microg levonorgestrel: a randomized trial.
Efficacy, acceptability and tolerability of the combined contraceptive ring, NuvaRing, compared with an oral contraceptive containing 30 microg of ethinyl estradiol and 3 mg of drospirenone.
Transdermal estrogen with a levonorgestrel-releasing intrauterine device for climacteric complaints versus estradiol-releasing vaginal ring with a vaginal progesterone suppository: clinical and endometrial responses.
A comparative multicenter study of the effects of continuous low-dose estradiol released from a new vaginal ring versus estriol vaginal pessaries in postmenopausal women with symptoms and signs of urogenital atrophy.
A one-year multicenter study of efficacy and safety of a continuous, low-dose, estradiol-releasing vaginal ring (Estring) in postmenopausal women with symptoms and signs of urogenital aging.
Pharmacokinetic bioequivalence, safety and acceptability of Ornibel((R)), a new polymer composition contraceptive vaginal ring (etonogestrel/ethinylestradiol 11.00/3.474 mg) compared with Nuvaring((R)) (etonogestrel/ethinylestradiol 11.7/2.7 mg).
A comparative study of safety and efficacy of continuous low dose oestradiol released from a vaginal ring compared with conjugated equine oestrogen vaginal cream in the treatment of postmenopausal urogenital atrophy.
An open-label, randomized, multi-center trial to evaluate continuation rates, side effects and acceptability of the contraceptive ring vs. the contraceptive patch in women previously using combined oral contraceptives.
Patient experience in a randomized trial of a weekly progesterone vaginal ring versus a daily progesterone gel for luteal support after in vitro fertilization.
Preferred Product Attributes of Potential Multipurpose Prevention Technologies for Unintended Pregnancy and Sexually Transmitted Infections or HIV Among U.S. Women.
Tenofovir disoproxil fumarate intravaginal ring for HIV pre-exposure prophylaxis in sexually active women: a phase 1, single-blind, randomised, controlled trial.
Acceptability of microbicidal vaginal rings and oral pre-exposure prophylaxis for HIV prevention among female sex workers in a high-prevalence US city.
Safety and pharmacokinetics of single, dual, and triple antiretroviral drug formulations delivered by pod-intravaginal rings designed for HIV-1 prevention: A Phase I trial.
Single-dose pharmacokinetics and pharmacodynamics assessment of oestriol and trimegestone containing vaginal rings in healthy women with childbearing potential.
Comparison of combined hormonal vaginal ring with ultralow-dose combined oral contraceptive pills in the management of heavy menstrual bleeding: A pilot study.
Efficacy and safety of the contraceptive vaginal ring (NuvaRing) compared with a combined oral contraceptive in Chinese women: a 1-year randomised trial.
A Randomised Controlled Trial Comparing the Efficacy and Side-Effects of Intravaginal Ring (Nuvaring((R))) With Combined Oral Hormonal Preparation in Dysfunctional Uterine Bleeding.
Comparison of combined hormonal vaginal ring and low dose combined oral hormonal pill for the treatment of idiopathic chronic pelvic pain: a randomised trial.
Comparison of efficacy, compliance, acceptability and tolerability of hormonal contraceptive vaginal ring (NuvaRing) versus combined oral contraceptives (COCs) - a prospective randomized controlled study in a district level hospital in East Delhi.
Assessing the potential of MPTs in South Africa, Uganda and Nigeria.
Bill and Melinda Gates Foundation,
Seattle, WA2014 (Available from https://theimpt.org/assessing-the-potential-of-mpts-in-south-africa-uganda-and-nigeria/, [Accessed 1 August 2021].)
End-user preference for and choice of four vaginally delivered HIV prevention methods among young women in South Africa and Zimbabwe: the Quatro Clinical Crossover Study.
An open-label, randomized crossover study to evaluate the acceptability and preference for contraceptive options in female adolescents, 15 to 19 years of age in Cape Town, as a proxy for HIV prevention methods (UChoose).
HIV prevention at drug shops: awareness and attitudes among shop dispensers and young women about oral pre-exposure prophylaxis and the dapivirine ring in Shinyanga, Tanzania.
00410-8&id=gr1.jpg){kind=link}
00410-8&id=gr2.jpg){kind=link}