Description of the condition
Dual-method contraception refers to using condoms as well as another modern method of contraception. The latter (usually non-barrier) method is commonly hormonal (e.g., oral contraceptives) or a non-hormonal intrauterine device (IUD). Use of two methods can better prevent pregnancy and the transmission of HIV and other sexually transmitted infections (STIs). In contrast, 'dual protection' may refer to using condoms to prevent both pregnancy and disease (Lopez 2013a). Unprotected sex increases risk for disease, disability, and mortality in many areas due to the prevalence and incidence of HIV/STI (Warner 2012). STIs are associated with increased risk for infertility (Higgins 2012). In 2011, young people aged 15 to 24 years accounted for two-thirds of the gonorrhea and chlamydia cases reported in the USA (CDC 2011b). An estimated 24,000 women become infertile each year due to undiagnosed STIs. In many countries, women account for the majority of people age 15 or older who are living with HIV (WHO 2013b).
Millions of women, especially in lower-resource areas, also have an unmet need for pregnancy prevention (Thurman 2011). Unplanned pregnancy increases risk for morbidity and mortality. Even in higher-resource areas such as the USA, nearly 60% of pregnancies are unintended, i.e., either unwanted (23%) or mistimed (37%) (Mosher 2012). The teen birth rate in the USA declined by a third since 1991 (Martin 2010; Chin 2012). This may be attributed partly to a decrease since 1991 in the proportion of teenagers reporting sexual activity (CDC 2011a). The proportion that used condoms increased until 2003 but has not changed since then (CDC 2011a).
Multipurpose prevention technologies are being developed or improved to prevent disease and pregnancy (Friend 2010; Thurman 2011). These include physical and chemical barriers alone or in combination. Male and female condoms are existing physical barriers that can provide dual protection against both pregnancy and HIV/STI. Chemical barriers refer to drugs that may have contraceptive properties and can work against specific STIs (Friend 2010; Thurman 2011). Dual protection rings are being developed to provide both contraception and protection against HIV/STI (Sitruk-Ware 2013). Meanwhile, the use of dual methods is often recommended to prevent pregnancy and disease.
Adherence to one contraceptive method can be challenging (Halpern 2011; Lopez 2013b). Using two methods requires multiple actions and the processing of multiple messages about risk, i.e., for pregnancy and HIV/STI (O'Leary 2011). Consequently, dual-method use has been relatively low. In the USA, rates may range from 7% to 23% (Brown 2011; Eisenberg 2012). In a national survey of reproductive-age women, 7% of those who were sexually active used dual methods in the past year (Eisenberg 2012). Rates were highest among women 15 to 20 years old (23%), those who were never married (19%), and women with two sexual partners in the past year (19%). In Europe, dual-method use is estimated to be 15% to 30% (Higgins 2012). In Australia, contraceptive use and dual protection were estimated from a national household survey (Parr 2007). Most of those reporting use of dual-methods were women 18- to 24-years old, of whom 21% currently used oral contraceptives (OCs) and condoms. Among the reproductive-age women, 39% used OCs of which 28% also used condoms. Rates were lower among women using an injectable contraceptive or subdermal implant. Within South Africa, a 1998 national survey showed 10% of reproductive age women reported using a condom at last intercourse (Kleinschmidt 2003). This includes 13.5% of women currently using injectables, 13% of those using OCs, and 7% of women with an IUD.
Much research on dual-method use has focused on adolescents. Contraceptive use among 15-year olds was analyzed with data from 24 countries (Godeau 2008). Areas included Eastern and Western Europe as well as Canada, Israel, and the Ukraine. Overall use of OCs plus condoms at last intercourse was about 16% in 2002. The highest rates were in Canada (29%), Flemish Belgium (30%), and the Netherlands (31%). Three countries had rates of 22% to 23%: France, Switzerland, and Wales. Overall, 58% of the adolescents reported using condoms only (Godeau 2008). In a USA study, 35% of African American adolescents in urban clinics reported condoms-only as the most frequently used contraceptive method (Brown 2011). From a 2004 study of urban adolescents, contraception knowledge and attitudes were examined for dual-method users versus users of condoms only or hormonal contraception only (Pack 2011). Of the sexually-active adolescents, 47% reported using dual methods at last intercourse, i.e., condoms and hormonal contraception. Those who used dual methods were more likely to report having negative attitudes about sex and pregnancy.
Description of the intervention
Use of condoms as the sole protection may be appropriate when exposure to infection is the major issue due to the prevalence of HIV/STI or the individual's risk behavior (Cates 2002). When used correctly and consistently, condoms can provide dual protection, i.e., against both pregnancy and disease (Steiner 1999; CDC 2010; O'Leary 2011). For the male condom, the estimated first-year pregnancy rate is 18% for typical use and 2% for perfect use (Trussell 2011). However, if unplanned pregnancy is the primary concern, dual-method use could be more helpful, given the greater effectiveness of hormonal contraception for preventing pregnancy (Cates 2002). Oral contraceptives have first-year pregnancy rates of 9% with typical use and 0.3% with perfect use (Trussell 2011). In contrast, long-acting reversible contraception (LARC), i.e., IUD or implant, does not require regular user action. A large cohort study reported a pregnancy rate of 0.27 per 100 participant-years for LARC users compared to 4.55 for those using pills, the transdermal patch, or the intravaginal ring (Winner 2012). However, some evidence suggests users of LARC users may be less likely to use dual methods (Eisenberg 2012).
Contraceptives that are more effective for preventing pregnancy do not protect against HIV/STI. Consistent use of latex condoms can reduce risk of HIV transmission from an infected partner by 80% to 90% (USAID 2005). However, failure to use condoms correctly with every sex act can increase risk. Incorrect use includes donning the condom after starting the sex act and removing it before ejaculation (Warner 2012). Furthermore, users can experience condom malfunctioning, such as breakage or slippage. Consistent and correct use of condoms decreases the risk of many STIs transmitted via the urethra, notably gonorrhea and chlamydia (Warner 2012). Some risk reduction has been shown for herpes simplex virus 2 (HSV-2) and human papillomavirus (HPV), which are transmitted through contact with the skin or mucosal surface (Crosby 2012; Warner 2012). Inconsistent condom use provides little or no protection (USAID 2005). Educational interventions focused on reducing condom errors may reduce the typical failure rate (for pregnancy) of 18%.
How the intervention might work
Behavioral interventions to improve dual-method use typically involve counseling or educating individuals or groups. Programs may be based on direct oral communication and written materials. Broader educational programs and communication campaigns may also be included. Social marketing can have greater reach than direct interpersonal interventions and may increase awareness and promote use of contraceptives (Chapman 2012). However, person-to-person interaction can be useful in communicating complex information about method use. Projects also utilize technology, such as computer-assisted interviews and mobile phone reminders, while others engage community workers or peer educators. Program descriptions may be easier to locate than evidence of intervention effectiveness. Evaluations of programs to promote dual-method use often depend on self-report. Assessment of condom use, in particular, could benefit from using valid and reliable outcome measures rather than self-reports alone (Crosby 2012).
Relevant interventions include family planning counseling, education, or information dissemination. Such interventions may increase contraceptive uptake as well as improve use and continuation of the chosen method. Counseling can help women meet their fertility goals of avoiding or delaying childbearing or of spacing children. The counseling should be appropriate for the woman's fertility intentions, lifestyle, preferences, and socioeconomic situation. Successful interventions may change beliefs about risk for pregnancy or HIV/STI transmission, attitudes and knowledge about prevention, and use of effective contraceptive methods including condoms to prevent disease.
Why it is important to do this review
A recent review examined condom-only use for dual protection (Lopez 2013a), and excluded dual-method use. Many reviews focus on preventing HIV/STI and do not address pregnancy prevention. Examples are condom promotion (Carvalho 2011; Sweat 2012), peer education (Medley 2009), STI prevention among young people (Picot 2012), and HIV prevention among people who use drugs (Meader 2013). A systematic review examined preventing pregnancy and HIV/STI among adolescents (Chin 2012). The investigators included comprehensive risk-reduction and abstinence programs; the latter will not be included here.
Behavioral interventions for this review have to address contraception as well as preventing HIV/STI. Motivation may differ for preventing disease versus pregnancy. We aim to identify behavioral interventions associated with improved use of dual contraceptive methods. Many interventions promote dual protection through using condoms only, and HIV/STI prevention programs may even have current use of hormonal contraceptives as an eligibility criterion (Roye 2007). We will search for studies with participants of any child-bearing age, recognizing that much research may be limited to adolescents. We will assess the evidence by population of focus, geographic location, and setting of the tested intervention (clinic or community). Promotional or educational programs with some evidence of effectiveness may be adaptable to other situations.
We will examine comparative studies of behavioral interventions for improving use of dual methods of contraception. Dual-method use refers to using condoms as well as another modern contraceptive method. Our intent is to identify effective interventions for preventing pregnancy as well as HIV/STI transmission.
Criteria for considering studies for this review
Types of studies
Studies must be comparative and may be randomized or non-randomized. They should have examined a behavioral intervention for improving dual-method use. The comparison may be another behavioral intervention, usual care, or no intervention.
Types of participants
Due to the focus on preventing pregnancy as well as HIV/STI transmission, participants would be heterosexual and could be men or women. Participants may be at risk for pregnancy or HIV/STI. We will not include studies that focus on people with a chronic health condition, such as lupus or diabetes, or those with a known psychiatric or substance abuse disorder.
Types of interventions
The behavioral intervention must have an educational or counseling component to encourage or improve the use of dual methods of contraception. The intervention must address preventing pregnancy as well as the transmission of HIV/STI. Therefore, it included the use of condoms plus another modern contraceptive (WHO 2013a), most commonly a hormonal contraceptive or an intrauterine system. The program or service may be targeted to individuals, couples, or communities. The comparison condition could be another behavioral intervention to improve contraceptive use, usual care, other health education, or no intervention.
The report must describe the content or process of the behavioral intervention. Counseling described as 'standard' or 'routine' will not be sufficient, nor will standard contraception counseling that covers a range of contraceptive methods without a specific component on dual-method use. Standard contraceptive counseling would not be informative about how to improve dual-method use. However, routine services will be acceptable for the comparison or control condition.
Types of outcome measures
Studies must provide reported use of dual methods, i.e., condoms plus another modern contraceptive method. Contraceptive use may be assessed in various ways, e.g., consistent use or improved adherence. When we find multiple measures within a study, we will focus on the investigator’s assessment of consistent use or use at last sex. If we do not find one of those preferred measures, we will accept the method used by the investigator.
Outcomes must be measured at least three months after the behavioral intervention began, to provide evidence of protected sex for a minimum duration. For assessing evidence quality, six months or more will be the criterion for a more meaningful outcome measure.
Outcomes may include biological measures, e.g., test results for pregnancy, HIV/STI, or presence of semen as assessed with a biological marker such as prostate-specific antigen (Gallo 2013).
For studies that meet the criteria for Primary outcomes, we will also examine outcome data on knowledge or attitudes about dual-method use.
Search methods for identification of studies
We will conduct searches of MEDLINE via PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), POPLINE, EMBASE, COPAC, and Open Grey. In addition, we will search ClinicalTrials.gov and ICTRP for current trials and for trials with relevant reports or data. Proposed search strategies are given in Appendix 1.
Searching other resources
We will examine reference lists of pertinent papers, including review articles, for additional citations. In addition, we will contact investigators in the field for other relevant studies and further clarification.
Data collection and analysis
Selection of studies
We will assess for inclusion all titles and abstracts identified during the literature search. Two authors will examine the search results for potentially eligible studies. Any discrepancies will be resolved by discussion or with a third author if necessary. For studies that appear to be eligible for this review, we will obtain and examine the full text articles.
Data extraction and management
Two authors will extract the data. One author will enter the data into RevMan, and a second author will check accuracy. These data include the study characteristics, risk of bias (quality assessment), and outcome data. Any discrepancies will be resolved by discussion or with a third author if necessary.
Assessment of risk of bias in included studies
We will use the framework in Borrelli 2011 to assess the quality of the intervention. Domains of treatment fidelity are study design, training of providers, delivery of treatment (intervention), receipt of treatment, and enactment of treatment skills. The framework was intended for assessing current trials. Criteria of interest for our review include the following:
- Study design - had a curriculum or treatment manual.
- specified provider credentials;
- provided standardized training for the intervention.
- Delivery - assessed adherence to the protocol.
- Receipt - assessed participants' understanding and skills regarding the intervention.
For the assessment of evidence quality, studies will be downgraded if three or fewer criteria were met.
For randomized controlled trials (RCTs), we will evaluate methodological quality according to recommended principles (Higgins 2011). That is, we will examine the information on randomization method, allocation concealment, blinding, and losses to follow up and early discontinuation. For RCTs in which individuals were randomized, adequate methods for allocation concealment include a centralized telephone system and the use of sequentially-numbered, opaque, sealed envelopes (Schulz 2002). In cluster RCTs, the randomization is usually done all at once, making allocation concealment less of an issue at the cluster level (Campbell 2012; Higgins 2011). However, selection bias may be introduced when individuals are approached for consent after the cluster has been randomized.
For non-randomized studies, we will again follow recommended principles for assessing the evidence quality (Higgins 2011). The Newcastle-Ottawa Scale (NOS) was developed for use with non-randomized studies (Wells 2013). According to the NOS researchers, the content validity and inter-rater reliability of this scale have been established, and criterion validity and intra-rater reliability are being examined. The investigators are developing a plan to assess construct validity. The scale does not have an overall scoring or cutoff for a 'good' or 'poor' quality study. We will adapt the NOS items for the interventions and outcomes in this review as per the developers' suggestions (Wells 2013). The scale has two versions, i.e., for case-control and for cohort studies. The latter is most likely to be useful for this review (Appendix 2). The NOS for cohort studies is consistent in content with a tool the Cochrane Methods Group is developing to assess risk of bias tool for non-randomized studies (Sterne 2013). The new tool is more complex than the NOS, having multiple questions per domain, but has not been disseminated for general use.
The NOS has eight items within the three domains of selection (representativeness), comparability (due to design or analysis), and outcomes (assessment and follow up). A study can receive one star (✸) for meeting each NOS criterion. The exception is comparability, for which a study can receive a maximum of two stars (for design and analysis). Assessment of analysis will include adjustment for potential confounding factors. For one star under comparability, the study would have controlled for age and gender, if appropriate. For two stars, adjustment would have included at least two of the following: parity, marital status or living with partner, sexual activity, HIV status, or socioeconomic status. We will create 'Additional tables' to present the quality assessment of non-randomized studies.
Quality assessment will include the length of follow up. While three months is the minimum for inclusion, six months of follow up provides more meaningful outcome measures. Losses to follow up will be examined for all types of studies. Losses of 20% or more will be considered high.
Measures of treatment effect
For RCTs with dichotomous outcomes, the Mantel-Haenszel odds ratio (OR) with 95% confidence interval (CI) will be calculated using a fixed-effect model. An example is the incidence of an STI. Fixed and random effects give the same result if no heterogeneity exists, as when a comparison includes only one study. When available, we will extract life-table rates for pregnancy and use the rate difference as the effect measure.
Cluster RCTs may use a variety of adjustment strategies. We will use adjusted measures that the investigators considered as the primary effect measures. Odds ratio is commonly provided when adjusted analyses are obtained using logistic regression models. However, if an appropriate adjusted OR is not available, we will consider other effect measures, e.g., rate ratio, hazard ratio, or incidence difference. Where multivariate models are used, we will not analyze the treatment effect as that would usually require individual participant data. Instead, we will present the results from adjusted models as reported by the investigators.
With non-randomized studies, investigators need to control for confounding factors and may use a variety of adjustment strategies. We will specify whether potential confounding was considered in the design (e.g., matching or stratification) or analysis and will note the specific confounding factors addressed. When investigators used multivariate models to adjust for potential confounding, we will present the results from adjusted models as noted above for cluster RCTs. If no adjusted measure was given as part of the primary analysis, we will use unadjusted measures. For such dichotomous outcomes, we will calculate the Mantel-Haenszel OR as noted above for RCTs.
Unit of analysis issues
If clustering is part of the design, we will assess whether the estimates were properly adjusted to account for clustering effects. Investigators may use various methods of accounting for the clustering, such as multilevel modeling. We will give the specific methods in the results for each study.
Dealing with missing data
If reports are missing data needed for analysis, we will write to the study investigators. However, we will limit our data requests to studies less than 10 years old, unless an older study had a report in the past five years. Investigators are unlikely to have access to data from older studies.
Assessment of heterogeneity
Given the diversity of design features with non-randomized studies, we might not conduct meta-analysis for pooled estimates. In addition, behavioral interventions can vary widely in design. In those cases, we would assess sources of heterogeneity without pooling the data. We will address heterogeneity due to differences in study design, analysis strategies, and confounding adjustment. To interpret the intervention results, we will examine the intervention location, i.e., country and setting (clinic, school, community); participant characteristics; and the intervention content, implementation, and fidelity information.
We will apply principles from GRADE to assess the evidence quality and address confidence in the effect estimates (Balshem 2011). However, when a meta-analysis is not viable due to varied interventions, a summary of findings table is not feasible. Therefore, we do not expect to conduct a formal GRADE assessment with an evidence profile and summary of findings table (Guyatt 2011).
Our assessment of the body of evidence will be based on the quality of evidence from the included studies. We will include intervention fidelity in the overall quality assessment. Evidence quality could be high, moderate, low, or very low. We will consider the evidence from RCTs to be high quality initially, then downgrade for each of the following: a) randomization sequence generation and allocation concealment: no information on either, or one was inadequate; b) intervention fidelity information reported for fewer than four of five categories; c) follow up less than six months; d) losses of 20% or more.
For non-randomized studies, we will assess the quality of evidence with the Newcastle-Ottawa Quality Assessment Scale (NOS). The evidence from non-randomized studies will be considered moderate quality initially and then downgraded for meeting fewer than six NOS criteria or not controlling for any confounding (i.e., not having any stars for comparability).
From FHI 360, Carol Manion provided input on the search strategy.
Appendix 1. Search strategies
MEDLINE via PubMed
(("Contraception"[Mesh] OR "Contraception Behavior"[Mesh] OR "Contraceptive Agents"[Mesh] OR "Contraceptive Devices"[Mesh] OR contraceptive OR contraception OR "family planning" OR dual-method[tiab] OR dual[tiab] OR "dual protection"[tiab]) AND (HIV[tiab] OR STI[tiab] OR "sexually transmitted"[tiab] or condom*[tiab] OR protected[tiab] OR unprotected[tiab]) AND (educat*[tiab] OR counsel*[tiab] OR development [tiab] OR communicat*[tiab] OR behavioral[tiab] OR behavioural[tiab] OR psychosocial[tiab] OR use[tiab] OR uptake[tiab] OR continuation[tiab] OR adherence[tiab])) NOT ("men who have sex with men"[ti] OR MSM[ti] OR microbicide[ti])
Filters activated: Clinical Trial, Comparative Study, Evaluation Studies
Title, Abstract or Keywords: contracept* OR "family planning" OR dual-method OR dual method OR dual protection
AND Title, Abstract or Keywords: HIV OR STI OR "sexually transmitted" OR pregnan*
AND Title, Abstract or Keywords: educat* OR counsel* OR development OR communicat* OR behavioral OR behavioural OR psychosocial OR use OR uptake OR continuation OR adherence
NOT Title: "men who have sex with men" OR MSM OR microbicide OR emergency
Any field: (contracept* OR dual-method OR dual)
AND (education OR counseling OR communication OR behavior OR psychosocial OR development)
AND (HIV OR STI OR "sexually transmitted" OR pregnan* OR condom* OR protected OR unprotected)
Title: NOT ("men having sex with men" OR microbicides OR emergency OR GnRH OR intrauterine OR survey OR trends)
Filter by keywords: Research report, Women
(contracept*:ab OR 'dual method':ab) AND (HIV:ab OR STI:ab OR 'sexually transmitted':ab OR condom*:ab OR protected:ab OR unprotected:ab) AND (educat*:ab OR counsel*:ab OR communicat*:ab OR behavioral:ab OR development:ab OR communicat*:ab OR adherence:ab OR continuation:ab) NOT (intrauterine:ti OR emergency:ti OR 'men who have sex with men':ti OR msm:ti OR microbicide*:ti)
AND [obstetrics and gynecology]/lim OR [public health]/lim AND [humans]/lim
Search terms: (contraception OR contraceptive) AND (dual OR condom)
Study type: interventional studies
Conditions: NOT (lupus OR menorrhagia OR anorexia OR polycystic OR depression OR diabetes)
Interventions: NOT (microbicide OR abortion OR misoprostol)
Outcome measures: NOT bone
1) contracept* AND dual
2) contracept* AND condom
contracept* AND (dual OR condom OR protect* OR unprotect*)
Subject: (condom* OR HIV OR STI OR "sexually transmitted") AND (contracept* OR pregnan*)
Material type: Electronic resources, computer programs, etc.; Journals and other periodicals; Theses
Appendix 2. Newcastle - Ottawa Quality Assessment Scale for cohort studies
Note: A study can be awarded a maximum of one star (✸) for each numbered item within the Selection and Outcome categories. A maximum of two stars can be given for Comparability.
1) Representativeness of the exposed cohort
a) truly representative of the average _______________ (describe) in the community ✸
b) somewhat representative of the average ______________ in the community ✸
c) selected group of users eg nurses, volunteers
d) no description of the derivation of the cohort
2) Selection of the non exposed cohort
a) drawn from the same community as the exposed cohort ✸
b) drawn from a different source
c) no description of the derivation of the non exposed cohort
3) Ascertainment of exposure
a) secure record (eg surgical records) ✸
b) structured interview ✸
c) written self report
d) no description
4) Demonstration that outcome of interest was not present at start of study
a) yes ✸
1) Comparability of cohorts on the basis of the design or analysis
a) study controls for _____________ (select the most important factor) ✸
b) study controls for any additional factor ✸ (This criteria could be modified to indicate specific control for a second important factor.)
1) Assessment of outcome
a) independent blind assessment ✸
b) record linkage ✸
c) self report
d) no description
2) Was follow-up long enough for outcomes to occur
a) yes (select an adequate follow up period for outcome of interest) ✸
3) Adequacy of follow up of cohorts
a) complete follow up - all subjects accounted for ✸
b) subjects lost to follow up unlikely to introduce bias - small number lost - > ____ % (select an adequate %) follow up, or description provided of those lost) ✸
c) follow up rate < ____% (select an adequate %) and no description of those lost
d) no statement
Contributions of authors
LM Lopez drafted the Protocol and developed the searches. MG Gallo and M Steiner provided input on the inclusion criteria as well as content expertise. M Chen contributed to the Methods section. All authors reviewed and commented on the Protocol.
Declarations of interest
No known conflicts.
Sources of support
- No sources of support supplied
- National Institute of Child Health and Human Development, USA.For conducting the review (FHI 360 staff)