Educación de las mujeres después del parto en el uso de los anticonceptivos
Resumen
Antecedentes
La educación en materia de anticonceptivos suele ser un componente estándar de la atención posparto, aunque pocas veces se examina su efectividad. Las suposiciones que forman la base de tales programas incluyen que las mujeres en el postparto están motivadas para usar anticonceptivos y que no volverán a un proveedor de salud para recibir consejos sobre planificación familiar. Las mujeres pueden querer discutir la anticoncepción antes y después del alta hospitalaria. No obstante, las dos terceras partes de las puérperas pueden tener necesidades no atendidas de anticoncepción. En los EE.UU. muchas adolescentes tienen embarazos repetidos dentro del año de haber dado a luz.
Objetivos
Evaluar la efectividad de las intervenciones educativas sobre el uso de anticonceptivos para las mujeres en el posparto
Métodos de búsqueda
Se buscaron ensayos hasta junio 2015 en PubMed, CENTRAL, CINAHL, POPLINE y Web of Science. Para los ensayos actuales se buscó en ClinicalTrials.gov y en ICTRP. Las búsquedas anteriores también incluyeron EMBASE y PsycInfo. También se examinaron las listas de referencia de los artículos pertinentes. Para las versiones anteriores se estableció contacto con los investigadores para identificar informes adicionales.
Criterios de selección
Se consideraron los ensayos controlados aleatorizados (ECA) que examinaron la educación posparto acerca del uso de anticonceptivos administrada a mujeres individuales o a grupos de mujeres. Los estudios en los que se asignaron al azar grupos en lugar de individuos fueron elegibles si los investigadores tuvieron en cuenta el grupo en el análisis. La intervención debe haber comenzado dentro del mes siguiente al parto.
Obtención y análisis de los datos
Se evaluaron todos los títulos y resúmenes identificados a través de las búsquedas bibliográficas. Los datos se extrajeron y se introdujeron en Review Manager. Se examinó la calidad metodológica de los estudios. Para los resultados dicotómicos se calculó el odds ratio (OR) de Mantel‐Haenszel, con los intervalos de confianza (IC) del 95%. Cuando los datos fueron para variables continuas, se calculó la diferencia de medias (DM), con los IC del 95%. No se realizó un metanálisis debido a la variabilidad en las intervenciones y las medidas de resultado.
Resultados principales
Doce ensayos cumplieron los criterios de elegibilidad, incluidos los tres agregados en esta actualización. Los estudios incluyeron 4145 personas. Ocho ensayos se realizaron en EE.UU., y los otros en Australia, Nepal, Pakistán y Siria. Cuatro estudios proporcionaron una sesión antes del alta hospitalaria; tres tenían un asesoramiento estructurado de intensidad variable y uno incluyó asesoramiento informal. De ocho intervenciones con más de un contacto, cinco se centraron en adolescentes. Tres de los cinco participaron en visitas domiciliarias, uno proporcionó múltiples servicios clínicos, y uno tuvo contacto en persona y seguimiento telefónico. De los tres restantes para mujeres de diferentes edades, dos implicaron visitas domiciliarias y uno hizo el seguimiento por vía telefónica.
El análisis de sensibilidad incluyó seis ensayos con evidencia de calidad moderada o alta. En un estudio con adolescentes, el grupo con tutoría domiciliaria tuvo menos segundos partos en dos años, en comparación con el grupo control (OR 0,41; IC del 95%: 0,17 a 1,00). Las otras cinco intervenciones no tuvieron efecto. De los ensayos con evidencia de menor calidad, dos mostraron cierta efectividad. En Nepal, las mujeres con una sesión educativa inmediatamente después del parto tuvieron más probabilidades de usar anticonceptivos a los seis meses que las que recibieron una sesión posterior o ninguna sesión (OR 1,62; IC del 95%: 1,06 a 2,50). En un estudio australiano, las adolescentes que participaron en un programa estructurado de visitas domiciliarias tuvieron más probabilidades de utilizar anticonceptivos efectivos a los seis meses que las que recibieron visitas domiciliarias estándar (OR 3,24; IC del 95%: 1,35 a 7,79).
Conclusiones de los autores
Este resumen de resultados se centró en ensayos con evidencia de calidad moderada o alta. En general, la calidad general de la evidencia de esta revisión fue moderada a baja y la evidencia de efectividad fue en su mayoría de calidad baja. Las intervenciones se podrían mejorar al fortalece el diseño y la implementación del programa. Algunos estudios no informaron sobre el adiestramiento de los proveedores en el programa, el cumplimiento del protocolo de intervención o la medición de los conocimientos y aptitudes de los participantes. Muchos ensayos no tuvieron una medida de resultado objetiva, es decir, una prueba de embarazo o un cuestionario estructurado para el uso de anticonceptivos. Se necesitan medidas de resultado válidas y fiables para obtener resultados significativos. Aún así, debido a los costes asociados y a la logística necesaria, algunos programas no serían factibles en muchos contextos.
PICOs
Resumen en términos sencillos
Educación acerca de la planificación familiar para las mujeres con un parto reciente
El asesoramiento sobre planificación familiar es estándar para la mayoría de las mujeres que acaban de dar a luz. Algunos proveedores e investigadores han analizado cuán bien funciona el asesoramiento. No se conoce si las mujeres en el posparto desean utilizar la planificación familiar o si volverán a un proveedor de servicios de salud para recibir asesoramiento sobre la anticoncepción. Las encuestas indican que las mujeres pueden desear tratar la planificación familiar antes de tener a su recién nacido y después que se van del hospital. Es posible que las mujeres también prefieran hablar sobre la anticoncepción junto con otros temas de salud. En esta revisión se analizaron los efectos de los programas educativos sobre planificación familiar para las mujeres que acaban de tener un recién nacido.
Hasta junio 2015 se buscaron los ensayos de educación sobre planificación familiar después de tener un recién nacido. También se escribió a los investigadores para identificar otros ensayos. Los ensayos tenían que estudiar en qué medida el programa afectó el uso de la planificación familiar. El programa debía haber ocurrido en el mes posterior al parto. Los datos se introdujeron en RevMan y se utilizó el odds ratio para examinar el efecto. También se examinó la calidad de los métodos de investigación.
Se encontraron 12 ensayos con 4145 mujeres. Ocho estudios se realizaron en EE.UU. y los otros en Australia, Nepal, Pakistán y Siria. Cuatro ensayos proporcionaron una sesión de asesoramiento antes del alta hospitalaria. De ocho estudios con más de un contacto, cinco se centraron en adolescentes. Tres de los cinco tuvieron visitas a domicilio, uno utilizó los servicios del consultorio, y uno tuvo contactos personales y telefónicos. De tres estudios con mujeres y adolescentes, dos tuvieron visitas a domicilio y uno utilizó el contacto telefónico.
Seis ensayos tuvieron resultados de calidad moderada. En un estudio con adolescentes, el grupo con tutoría domiciliaria tuvo menos segundos partos en dos años en comparación con el grupo control. De los ensayos con evidencia de menor calidad, dos mostraron algún efecto. En Nepal, la mayoría de las mujeres que reciben algún tipo de asesoramiento inmediatamente después del parto pueden utilizar anticonceptivos a los seis meses, en comparación con las que tienen una sesión más tarde o ninguna. En Australia, más adolescentes en un programa especial de visitas domiciliarias utilizaron correctamente los anticonceptivos a los seis meses, en comparación con las visitas domiciliarias estándar.
Se encontraron resultados de calidad moderada a baja en general. La mayoría de los que tuvieron algún efecto fueron de calidad baja. Un mejor diseño y realización de los programas podría hacerlos más fuertes. Aún así, algunos programas pueden costar demasiado para algunos contextos.
Authors' conclusions
Summary of findings
| Special education compared with routine or delayed education for contraceptive use | ||||
| Patient or population: women with postpartum status Settings: clinic or community Intervention: special contraception education Comparison: see Comments | ||||
| Outcomes | Relative effect | No of Participants | Quality of the evidence | Comments |
| Second birth (by 24 months) | OR 0.41 (95% CI 0.17 to 1.00) | Participants = 149 | ⊕⊕⊕⊝ | Home mentoring vs usual care |
| Contraception use (at 6 months): immediate session vs no immediate session | OR 1.62 (95% CI 1.06 to 2.50) | Participants = 393 | ⊕⊕⊝⊝ | Education: immediate vs later |
| Effective contraception use (at 6 months) | OR 3.24 (95% CI 1.35 to 7.79) | Participants = 124 (Quinlivan 2003) | ⊕⊕⊝⊝ | Home visiting: structured vs routine |
| GRADE Working Group grades of evidence | ||||
Background
Description of the condition
Nearly two‐thirds of women in their first postpartum year have an unmet need for family planning (Ross 2001; USAID 2014). Data from 17 countries show that return to sexual activity is associated with the return of menses, breastfeeding status, and postpartum duration but not generally associated with contraceptive use (Borda 2010). Millions of women, especially in lower‐resource areas, are at risk for unplanned pregnancy and its consequent 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). First‐time adolescent mothers who do not start contraception before hospital discharge are more likely to have a repeat pregnancy within two years (Damle 2015). While adolescents may start using contraception during the postpartum period, they often discontinue due to lack of information or support (Wilson 2011).
Description of the intervention
Contraceptive education is generally considered a standard component of postpartum care. The counseling is frequently part of discharge planning but may also begin antepartum (Glasier 1996; Glazer 2011). Postpartum contraception counseling is often limited to one encounter, which is unlikely to affect behavior. Decisions about contraception made right after counseling may differ considerably from contraceptive use postpartum (Engin‐Üstün 2007; Glazer 2011). In some locations, postpartum care may be limited and undocumented, and may focus on the infant rather than the mother (Do 2013).
Postpartum women may wish to discuss contraception prenatally or after hospital discharge, preferably in the context of general education about maternal and child health (Glasier 1996; Ozvaris 1997). Many women are comfortable with advice and a prescription from their physician during well‐baby visits (Fagan 2009). In a rural area in Bangladesh, intensive provision of maternal and child health and family planning programs resulted in increased uptake of contraceptives (Koenig 1992). A recent trial examined the effect on contraceptive use from integrating family planning into infant immunization services (FHI 360 2013). However, the women were 6 to 12 months postpartum by that time.
Educational interventions provided to individuals or groups may increase contraceptive uptake as well as improve use and continuation of the chosen method. Counseling may involve a single contact or multiple sessions. Personal interaction can help women choose an appropriate method and obtain detailed information about method use. Social marketing can have greater reach and may increase awareness and promote use of contraceptives (Chapman 2012). Educational projects may also utilize technology, such as mobile phone reminders for appointments and medication use. Some programs engage peer educators or community workers to provide reminders and encourage method continuation.
Why it is important to do this review
In the past, midwifery and obstetric texts rarely questioned the effectiveness of postpartum contraceptive education, though it was considered a responsibility in postpartum care (Keith 1980; Semeraro 1996). Research evaluating such education is still sparse. More is known about the appropriateness of specific contraceptive methods for postpartum women than about how to help women use certain contraceptives (Shaw 2007). This review examines randomized controlled trials of postpartum interventions to educate women about contraceptive use. In this update we sought additional evidence on the effectiveness of such efforts.
Objectives
We assessed the effectiveness of educational interventions for postpartum women on contraceptive use.
Methods
Criteria for considering studies for this review
Types of studies
We considered randomized controlled trials (RCTs) that examined postpartum education about contraceptive use, whether delivered to individuals or to groups of women. Studies that randomized clusters rather than individuals were eligible if the investigators accounted for the clustering in the analysis.
Types of participants
Women gave birth at 20 weeks of gestation or more. We excluded trials focused on the needs of women with alcohol or drug problems and trials focused on women with chronic health conditions such as HIV or diabetes.
Types of interventions
Trials were included if they evaluated postpartum education provided to influence uptake of contraception including lactational amenorrhea. Educational interventions may have been based on written materials, video or audio recordings, or individual or group counseling. The intervention must have started postpartum and begun within one month of delivery.
Types of outcome measures
Primary outcomes
To be included, trials had to have data on unplanned pregnancies or contraceptive choice or use. The trials may have had other primary outcomes.
Secondary outcomes
Additional outcomes were knowledge and attitudes about contraception and satisfaction with postnatal care.
Search methods for identification of studies
Electronic searches
Through June 2015, we conducted searches of MEDLINE, Cochrane Central Register of Controlled Trials (CENTRAL), POPLINE, CINAHL, and Web of Science. We also searched for current trials via ClinicalTrials.gov and ICTRP. The search strategies are given in Appendix 1. Previous search strategies also included EMBASE and PsycINFO; they are shown in Appendix 2.
Searching other resources
Reference lists of relevant papers were examined for additional citations. For previous versions, we also contacted investigators in the field to seek unpublished trials or published trials we may have missed in our searches.
For the initial review in 1999, the authors contacted the following organizations for advice about relevant research: Guttmacher Institute, California Family Health Council, Contraceptive Research and Development, Couple to Couple League, Engender Health, European Commission, Health, Family Planning and AIDS Unit, Family Planning Association of Queensland, Family Planning Councils of America, Family Planning International Assistance, Family Planning Management Development, Healthy Women, Johns Hopkins University Center for Communication Programs, Marie Stopes International, National Family Planning and Reproductive Health Association, Planned Parenthood Global Partners, Population and Community Development Association, Population Reference Bureau, Prime II, Program of Appropriate Technology in Health. The authors also searched databases listing publications by the Population Council, Family Health International (now FHI 360), and the World Health Organization.
Data collection and analysis
Selection of studies
One author reviewed the search results and a second author examined the reports identified for appropriate categorization. We excluded studies that randomized clusters rather than individuals and did not account for the clustering in the analysis. For the initial review, the three authors independently assessed the studies to determine which were suitable for inclusion. The authors resolved any differences by discussion.
Data extraction and management
Two authors conducted the data extraction. One author entered the data into Review Manager (RevMan 2014), and a second author checked accuracy (Contributions of authors). These data included the study characteristics, risk of bias, and outcome data.
Assessment of risk of bias in included studies
Intervention fidelity
We used 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 are shown below.
-
Study design: had a curriculum or treatment manual
-
Prior training of providers: specified providers' credentials
-
Project‐specific training: provided standardized training for the intervention
-
Delivery: assessed adherence to the protocol
-
Receipt: assessed participants' understanding and skills regarding the intervention.
In this update, we revised the terminology for the first four criteria, but the concepts are the same. We added the fifth criterion of 'Receipt' in this version of this review, based on newer information in Borrelli 2011. For the assessment of evidence quality, we downgraded the studies that met fewer than four of the five criteria.
Research design
Included trials were evaluated for methodological quality in accordance with recommended principles (Higgins 2011). Factors considered included randomization method, allocation concealment, blinding, and losses to follow‐up and early discontinuation. This information was entered into the Risk of bias tables (Characteristics of included studies).
Measures of treatment effect
For the dichotomous outcomes, the Mantel‐Haenszel odds ratio (OR) with 95% confidence interval (CI) was calculated. An example is the proportion of women who initiated use of a particular contraceptive method. Fixed and random effects give the same result if no heterogeneity exists, as when a comparison includes only one study. Where data were sparse, we used the Peto OR. For continuous variables, the mean difference (MD) was computed with 95% CI. Review Manager uses the inverse variance approach.
Assessment of heterogeneity
Due to varied study designs, we were unable to conduct meta‐analysis. Therefore, we did not need to assess statistical heterogeneity. However, we address heterogeneity due to differences in interventions, study design, and populations in the Discussion.
Data synthesis
We applied principles from GRADE to assess the quality of evidence and address confidence in the effect estimates (Balshem 2011; Higgins 2011). Our assessment of the body of evidence is based on the quality of evidence from the studies. When a meta‐analysis is not viable because of varied interventions or outcome measures, a summary of findings table is not feasible. Therefore, we did not conduct a formal GRADE assessment, i.e., with an evidence profile and summary of findings table, for all outcomes (Guyatt 2011).
Our assessment of evidence quality included the design, implementation, and reporting of both the intervention and the trial. We incorporated the quality of intervention evidence into the overall assessment of evidence quality. We considered RCTs to be high quality then downgraded for the following:
-
intervention fidelity information for fewer than four criteria;
-
no information on randomization sequence generation or allocation concealment, or risk of bias was high for one;
-
outcome assessment lacked an objective measure, e.g., pregnancy test or structured questionnaire for contraceptive use;
-
follow‐up less than three months for contraceptive use or less than six months for pregnancy;
-
loss to follow‐up greater than 20%.
Sensitivity analysis
The sensitivity analysis included evidence of moderate or high quality.
Results
Description of studies
Results of the search
The 2012 search produced 217 new references. After reviewing the titles and abstracts, 212 were discarded due to not meeting the eligibility criteria. We examined the full text of five reports; two were later excluded. Three reports from two trials were added to the eight trials included in earlier versions of this review. In addition, we included two ongoing trials identified in clinical trials databases.
In 2015, searches of databases yielded 277 unduplicated citations. A total of 101 duplicates were removed electronically or by hand. With one identified from other sources, the total of unduplicated references was 278. We reviewed the full text of eight items. We included three primary reports; one was a conference abstract for which we had the ClinicalTrials.gov listing. We also included one secondary article and two abstracts related to full reports. We excluded two primary reports. From recent clinical trial listings, we obtained 40 unduplicated trials. We categorized one as ongoing and updated one listing from the earlier review.
Included studies
We included 12 trials that met our eligibility criteria after adding 3 in this update (Simmons 2013; Tang 2014; Torres 2014). The studies included a total of 4145 women; the mean was 319 and the median was 240. One trial had a preliminary report published as a conference abstract (Torres 2014).
Eight studies were conducted in the USA; the other four were from Australia (Quinlivan 2003), Nepal (Bolam 1998), Pakistan (Saeed 2008), and Syria (Bashour 2008). Six trials focused on adolescents (O'Sullivan 1992; Quinlivan 2003; Black 2006; Barnet 2009; Katz 2011) or young women (Gilliam 2004).
The studies varied in the content and format of the education provided (Table).
-
Four provided one educational session before hospital discharge, with the content focused on contraception. Three had routine or alternative care as the comparison (Gilliam 2004; Tang 2014; Torres 2014); in the fourth, the control group had no intervention (Saeed 2008).
-
Eight trials provided more than one contact. In addition to contraception, some addressed broader health education or parenting issues while others provided logistical support or case management. The interventions involved one or more home visits (Quinlivan 2003; Black 2006; Bashour 2008; Bolam 1998; Barnet 2009), clinic contacts (O'Sullivan 1992), or phone sessions (Katz 2011; Simmons 2013).
All reports had some information regarding intervention fidelity (Table 2). All reports had information on the intervention content or its development. In most trials, clinicians provided the education and had some intervention training although the intensity of training varied. In a few studies, the women who provided the intervention had demographics similar to those of the participants.
| Study | Curriculum or manual | Provider credentials | Training for intervention | Assessed adherence to protocol | Assessed intervention | Fidelity |
| 4 goals and specific services identified; which professionals provide each component | Directed by nurse practitioner; providers included social worker, pediatrician and nurse practitioner, volunteers | Volunteers were 'trained' | _ | _ | 3 | |
| Format and content identified for sessions including key messages | 3 health educators, 2 midwives, 1 community health worker | Providers were 'trained' to give the health education | Investigators monitored weekly and gave feedback | Knowledge of infant health issues (primary outcome) | 5 | |
| Structured home visits outlined in report | Certified midwives | _ | _ | Knowledge of contraception (primary outcome) | 3 | |
| Developed counseling program, video, pamphlet; development described | Resident physicians, nurses for additional counseling | Training session for resident physicians and nurses | _ | Knowledge of OCs (secondary outcome) | 4 | |
| Curriculum with 19 lessons; order could vary after 2 sessions | 2 Black women, college‐educated, in their 20s, single mothers, living independently | 'Extensive' training provided | Weekly supervisory sessions | _ | 4 | |
| Objectives for each visit; breastfeeding in visits 2 and 3, family planning in visit 4 | Registered midwives | 5 days of special training | _ | _ | 3 | |
| Counseling leaflet used | Physicians | Providers had 40‐minute training on leaflet and interview methods. | _ | _ | 3 | |
| Structured software (computer‐assisted motivational intervention (CAMI)); 20‐min stage‐matched motivational interviewing (MI); parenting curriculum | African American paraprofessional women from participants' communities; had empathetic qualities, rapport with adolescents, community knowledge | 2.5 days on transtheoretical model, MI, and CAMI | First 4 months, counselors met biweekly with MI supervisor (discussed audiotapes and gave feedback) | _ | 4 | |
| Curriculum with standardized format and session structure; teen workbooks with visual material on topics | Masters‐level young women of similar racial‐ethnic background as teens | _ | Process evaluation included delivery | _ | 3 | |
| _ | 2 personal assistants with contraceptive counseling experience (OBGYN resident, 4th‐year medical student) | _ | Contacts planned; additional counseling as needed | _ | 2 | |
| 1‐minute script | Research assistants | _ | _ | _ | 2 | |
| Written structured script on all contraceptive methods by effectiveness | _ | _ | _ | Knowledge assessed with 'widely accepted instrument' | 2 |
aAssessed participants' understanding and skills regarding the intervention.
Outcomes included contraceptive use, pregnancy, and contraceptive knowledge. All trials assessed contraceptive use; six reported on repeat pregnancy or second birth. In Characteristics of included studies, we focus on the primary and secondary outcomes for this review. Studies may have had additional outcomes.
Risk of bias in included studies
Details for each study can be found in Characteristics of included studies. The trial with only a conference abstract had limited design information (Torres 2014).
Intervention quality
All reports provided some documentation of intervention content and most had implementation information (Table 2). A few did not provide much or any information on training of providers for the specific intervention. Only five reported on how the investigators assessed delivery adherence, i.e., whether the intervention was provided as intended.
Outcome assessment was limited in many studies (Table 3). That is, three of six trials that assessed pregnancies or repeat births had some objective validation, such as pregnancy test or vital statistics. Some studies assessed or reported contraceptive use with one dichotomous item, while others had structured questionnaires. Three studies had limited follow‐up periods. The short‐term nature of several interventions may also have limited the usefulness of the effectiveness measures.
| Study | Intervention | Randomization; | Outcome | Follow‐up period | Loss > 20% | Evidence | Evidence of effectiveness |
| _ | _ | _ | _ | _ | High | _ | |
| _ | _ | ‐1 | _ | _ | Moderate | Fewer repeat births (by 24 months) | |
| _ | _ | _ | _ | ‐1 | Moderate | _ | |
| ‐1 | _ | _ | _ | _ | Moderate | _ | |
| ‐1 | _ | _ | _ | _ | Moderate | _ | |
| ‐1 | _ | _ | ‐1 | _ | Moderate to lowc | _ | |
| _ | _ | ‐1 | _ | ‐1 | Low | Use of contraception (at 6 months) | |
| ‐1 | _ | ‐1 | _ | _ | Low | Effective use of contraception (at 6 months) | |
| ‐1 | _ | _ | ‐1 | _ | Low | _ | |
| ‐1 | ‐1 | ‐1 | _ | _ | Very low | Fewer repeat pregnancies (by 18 months) | |
| ‐1 | _ | ‐1 | ‐1 | _ | Very low | Use of contraception (at 8 to 12 weeks) | |
| ‐1 | ‐1 | ‐1d | _ | ‐1d | Very low | Use of highly effective contraception (at 3 months) |
aRCTs considered high quality then downgraded: 1) intervention fidelity < 4 criteria; 2) no information on randomization sequence generation or allocation concealment, or risk of bias was high for one; 3) outcome assessment lacked an objective measure (e.g., pregnancy test or structured questionnaire for contraceptive use); 4) follow‐up < 3 months for contraceptive use or < 6 months for pregnancy; 5) loss to follow‐up > 20%
bSensitivity analysis included evidence of moderate or high quality
cModerate for contraceptive use; low for pregnancy
dUnknown; preliminary report
Allocation
Five trials provided information on sequence generation and used sealed envelopes to conceal the allocation (Bolam 1998; Quinlivan 2003; Gilliam 2004; Bashour 2008; Tang 2014). Four had information on the randomization procedure but nothing on allocation concealment (Saeed 2008; Barnet 2009; Katz 2011; Simmons 2013). Three did not have information on sequence generation or concealment (O'Sullivan 1992; Black 2006; Torres 2014).
Blinding
Blinding of assignment was not possible in many trials, given the nature of the interventions. However, the outcome assessors were blind to group of allocation in three trials (Bolam 1998; Bashour 2008; Saeed 2008). Research team members were blinded in three studies (Gilliam 2004; Simmons 2013; Tang 2014). One trial was open (Torres 2014).
Incomplete outcome data
Losses to follow‐up were greater than 20% in two trials: Bolam 1998 (25% at three months and 27% at six months) and Gilliam 2004 (52% by one year). Losses were not as high overall as for some trials in contraceptive education (Lopez 2013).
Other potential sources of bias
Black 2006 excluded participants after randomization due to missing data. Barnet 2009 excluded one participant who had a stillborn infant and one whose two‐month‐old infant died.
Effects of interventions
See: Summary of findings for the main comparison
Counseling (one contact)
Four studies provided a single counseling session focused on contraception (Gilliam 2004; Saeed 2008; Tang 2014; Torres 2014). Torres 2014 also reportedly facilitated access to the woman's desired contraceptive method, but we had limited information from the conference abstract. Three trials were conducted in the USA and one in Pakistan (Saeed 2008). Gilliam 2004 focused on young women (<= 25 years old), two trials included participants from age 14 up to 45 years (Tang 2014) or up to 50 years (Torres 2014), and Saeed 2008 did not have any age specifications.
-
Gilliam 2004 (N = 33) provided a multi‐component intervention of counseling, videotape about oral contraceptives (OC), and written material. The comparison group had usual care. The experimental and comparison groups were not significantly different in the proportions that continued oral contraceptive use at one year (Analysis 1.1), those who switched the type of contraceptive used (Analysis 1.2), or known pregnancies (Analysis 1.3). The investigators noted that the sample size was not sufficient to detect a 20% difference between groups, due to resource limitations.
-
In Saeed 2008 (N = 648), the experimental group received informal counseling on contraception plus a pamphlet. The control group had no intervention. At 8 to 12 weeks postpartum in Saeed 2008, women in the counseling group were more likely to report using contraception (OR 19.56; 95% CI 11.65 to 32.83) (Analysis 2.1). For choice of contraceptive, all women in the counseling group planned to use a modern contraceptive method by six months postpartum compared to a third of the control group (Peto OR 18.53, 95% CI 13.15 to 26.12) (Analysis 2.2). The physician‐assessor was reportedly blinded to study arm.
-
Tang 2014 (N = 800) used a one‐minute script on long‐acting, reversible contraception (LARC). Long‐acting methods include intrauterine contraception and the subdermal implant. Women were followed up by phone after the six‐week visit. The comparison group received routine counseling at the hospital, which was not standardized. Despite the large sample size, the study arms did not differ significantly for LARC use (Analysis 3.1), interest in but not using LARC (Analysis 3.2), or use of any contraceptive method (Analysis 3.3).
-
For Torres 2014 (N = 121), we had preliminary information from the conference abstract. Results were from three months of follow‐up. The trial plans to follow 362 participants for 36 months (12 months for the primary outcome) with completion in 2016. Women with a preterm birth received structured counseling on the relative effectiveness of contraceptive methods, while the comparison group had usual care. The experimental group also reportedly had some assistance with access to the desired contraceptive method, but details were not available. The group with structured counseling was more likely than the routine care group to be using a highly effective contraceptive method at three months (OR 2.23, 95% CI 1.03 to 4.83) (Analysis 4.1). Mean increase in contraceptive knowledge was also higher for structured counseling compared to routine care (MD 2.30, 95% CI 2.00 to 2.60) (Analysis 4.2). No detail was available regarding the knowledge assessment.
Programs with two or more contacts (home, clinic, or phone)
Eight trials provided interventions that covered a range of health and lifestyle issues, including contraception. Five targeted adolescents; four were conducted in the eastern USA (O'Sullivan 1992; Black 2006; Barnet 2009; Katz 2011) and one in Australia (Quinlivan 2003). The three that included adult women were conducted in Syria (Bashour 2008), Nepal (Bolam 1998), and the northwestern USA (Simmons 2013).
Adolescents
Five trials focused on adolescents, three of which involved home visiting.
-
For Quinlivan 2003 (N = 139), young women in the experimental group had a structured home‐visiting program as opposed to standard home visits. Girls in the experimental group were more likely to have effective contraceptive use at six months than those in the comparison group (OR 3.24; 95% CI 1.35 to 7.79) (Analysis 5.1). We did not have sufficient data to analyze contraceptive knowledge in this review. Reportedly, the mean difference in contraceptive knowledge at six months favored the experimental group (reported MD 0.92; 95% CI 0.32 to 1.52).
-
Black 2006 (N = 181) evaluated second births during home visits. The experimental group had multiple home visits over two years, while the controls had usual care. The mean number of intervention visits was 6.63 (standard deviation 6.58). The adolescents in the treatment group were less likely to have had a second birth within two years than the usual care group (OR 0.41; 95% CI 0.17 to 1.00) (Analysis 6.1).
-
For Barnet 2009 (N = 237), the experimental groups received a computer‐assisted motivational intervention (CAMI) plus a parenting curriculum (CAMI+) and case management. The comparison groups had the CAMI or usual care. The study arms did not differ significantly for repeat births by 24 months from index birth (Analysis 7.1). Births were assessed through Vital Statistics; 100% of the index births were located. The repeat birth rates for both CAMI groups were lower than, but not significantly different from, the rate for the usual care group. The figures were 13.8% for CAMI plus parenting curriculum (CAMI+), 17.2% for CAMI‐only, and 25% for usual care. Abortion information was obtained at the follow‐up interview. The investigators provided the percentages for reported abortions: CAMI+ 22%, CAMI‐only 20%, and usual care 21%.
The other two trials for adolescents provided clinic‐based services plus reminders or cell‐phone counseling.
-
In O'Sullivan 1992 (N = 243), the experimental group had special services provided within the well‐baby clinic, including reminder contacts. The comparison group had the usual well‐baby care. The teenagers in the experimental group were less likely to have a repeat pregnancy (self‐reported) by 18 months compared to the control group (OR 0.35; 95% CI 0.17 to 0.70) (Analysis 8.1). The difference in pregnancies was largely within the subgroup of clinic dropouts. Of the control group, 29/91 had a repeat pregnancy versus 9/60 in the experimental group (data not shown).
-
Katz 2011 (N = 249) provided a cell‐phone counseling intervention, supplemented by quarterly group sessions. Subsequent pregnancy was assessed via cell phone calls at 3, 9, 15, and 21 months. Pregnancy status was confirmed by urine pregnancy tests at 6, 12, 18 and 24 months. Pregnancy rates did not differ significantly between the study groups during the two‐year follow‐up (Analysis 9.1). The rates were 31% for the group with the cell phone intervention and 36% for the usual care group.
Adult and younger women
Of three interventions for women of various ages, one provided phone follow‐up and two involved home visits.
-
In Simmons 2013 (N = 50), the special intervention group received reminders and assistance with follow‐up visits as well as counseling by phone. The comparison group had usual care. The groups did not differ significantly for LARC placement by three months in this small trial (Analysis 9.2).
-
In Bashour 2008 (N = 903), the experimental group had up to four home visits, with the last one focusing on family planning. One study group had a single visit without family planning and the control group had usual care, which did not include a home visit. At four months, the study groups did not differ significantly in contraceptive use (Analysis 10.1; Analysis 10.2) or in self‐reported pregnancy (Analysis 10.3; Analysis 10.4).
-
For Bolam 1998 (N = 540), two of the four study groups had health education during their postpartum hospital stay. One of those had a second session at three months that included family planning. A third group had the educational session at three months, and the fourth received no health education (Types of interventions). We grouped those with a health education session during their postpartum hospital stay (with or without a later session) and those with a later session or no session. The two groups did not differ significantly for contraceptive use at three months (Analysis 11.1). However, at six months, the group with an immediate postpartum session was more likely to use contraception than the group with a later or no session (OR 1.62; 95% CI 1.06 to 2.50) (Analysis 11.2). We also compared the arms within those groups at six months (three‐month data were not available). Women with an immediate and later session did not differ significantly in contraceptive use from those with one immediate session (Analysis 11.3). Also, contraceptive use among women with only the later session was not significantly different from those with no educational session (Analysis 11.3). Exclusive breastfeeding was emphasized in the immediate postpartum session. The study arms did not differ significantly in exclusive breastfeeding at three months or for more than five months (Analysis 11.4 to Analysis 11.6).
Discussion
Summary of main results
Sensitivity analysis
We focus here on the six trials with evidence of moderate or high quality (Table 3). One USA study had evidence of intervention effectiveness (Black 2006). In that trial, the experimental group had enhanced services through a home‐visiting program versus usual care for the controls. The experimental group had fewer repeat pregnancies or second births within two years compared to the control group. Second births were assessed during home visits.
The other five trials with evidence of moderate or high quality showed no effect of the intervention. All assessed contraceptive use and three examined pregnancies (a fourth had low quality evidence for pregnancy). The three trials with adolescents or young women provided the following: one multi‐component session (without sufficient power to detect differences) (Gilliam 2004), computer‐assisted motivational interviewing plus case management (Barnet 2009), and a cell‐phone counseling intervention with multiple contacts (Katz 2011). Of the two trials that included adult women, one had up to four home visits with one visit focused on family planning (Bashour 2008) and the other provided reminders and assistance with follow‐up visits as well as phone counseling (Simmons 2013).
Other results
Six trials provided evidence of low or very low quality in our assessment. Of three studies with low quality evidence, two showed evidence of effect on contraceptive use (Bolam 1998; Quinlivan 2003). An Australian trial was based on a home‐visiting program for adolescents, and compared structured versus standard home visits (Quinlivan 2003). The group with structured visits was more likely to use effective contraception at six months than the group with routine care. From Nepal, a four‐arm study provided health education for women of varying ages, including family planning for two groups at different times (Bolam 1998). Three trials with very low quality evidence also showed some effect on pregnancy or contraceptive use (Table 3). A recent study was downgraded largely because of limited information in the preliminary report (Torres 2014). This trial may be rated higher when more information is available on design and implementation.
Of nine studies with low, moderate, or high quality evidence, three showed any significant difference between comparison groups (summary of findings Table for the main comparison). With so few studies overall, we could not detect any pattern based on number of contacts or population of focus. Half of the programs showed some evidence of effect, regardless of the number of contacts.
-
Of six trials that focused on adolescents or young women, three showed some effect of the intervention. Two had fewer repeat pregnancies or births within the experimental group; both involved home‐visiting programs. The third provided clinic services and reminders, and showed more effective use of contraception in the special‐intervention group.
-
Of six trials that included women of various ages, three had some effect on contraceptive use. They provided one counseling session versus none, structured counseling versus usual care, and health education including family planning.
Overall completeness and applicability of evidence
We did not combine any trials in meta‐analysis due to varying interventions and outcome measures. The included trials represented various types of postpartum education: one contact or multiple‐session programs in five countries. Several interventions were provided during the postpartum hospital stay, while others began two or three weeks later. In some trials, family planning education was integrated with other health education or health services. The six trials showing positive effects were conducted in Australia, Nepal, Pakistan, and the USA. However, five provided low quality evidence. We did not examine interventions for postpartum contraception that began during the prenatal period. For example, Adanikin 2013 assigned women to either three sessions in the last trimester or the usual single session at the six‐week postpartum visit. Women with the prenatal sessions were more likely to use modern contraceptives at six months compared with the group that only had one postnatal session.
Clinics are likely to vary in the type and amount of usual services. In an urban teaching hospital in the USA, a retrospective study examined factors associated with repeat pregnancy among first‐time adolescent mothers (Damle 2015). Lower risk of rapid repeat pregnancy was found among those with a postpartum visit by eight weeks and who initiated use of long‐acting reversible contraception (LARC) by eight weeks. Special populations may also benefit from targeted programs. In Shanghai, migrant women had an unintended pregnancy rate about four times that of permanent residents (Huang 2014). A contraceptive services program provided counseling and a choice of contraceptive methods for migrant women. The education was provided in the maternity ward prior to delivery, with a second session if time was too limited. Long‐acting methods available prior to discharge included tubal ligation, IUD insertion immediately after delivery, and injection of depot medroxyprogesterone acetate (DMPA). Condoms were provided to those who did not desire a long‐acting method. Contraceptive use at 12 months was 97%, with IUD and condom use at about 39% each followed by tubal ligation at 8% and DMPA at 7%.
Costs for home‐visiting or multiple‐contact programs could be a limiting factor for many settings. Two studies in this review addressed costs. In O'Sullivan 1992, both groups received well‐baby care in the clinic. The hospital estimated the cost per visit to be lower for special care than routine care. The savings were attributed to several factors, such as combining services and not using medical residents who would need training and faculty supervision. In a 2010 article from Barnet 2009, the weighted mean costs for any CAMI were US $2064 per teen. Costs per teen were US $1449 for CAMI‐only and US $2635 for CAMI + parenting curriculum.
Quality of the evidence
Six studies had evidence of moderate or high quality (Table 3), but the overall quality of evidence was moderate to low. Most evidence of effectiveness came from trials of low or very low quality. Our assessment considers the intervention design and implementation as well as the basic trial design. For the latter, Figure 1 summarizes the risk of bias for the review overall. Risk of bias for individual trials is shown in Figure 2. Most trials were published since the first CONSORT guidelines, which have been updated (Schulz 2010), so the trials would be expected to have adequate reporting.
Risk of bias graph: authors' judgements about risk of bias as percentages across all included studies
Risk of bias summary: authors' judgements about risk of bias item for each included study
Risk of bias graph: authors' judgements about risk of bias as percentages across all included studies
Risk of bias summary: authors' judgements about risk of bias item for each included study
Comparison 1 Oral contraceptive education program (1 session) versus routine counseling, Outcome 1 Continuation of oral contraceptives at one year.
Comparison 1 Oral contraceptive education program (1 session) versus routine counseling, Outcome 2 Switched contraceptives by one year.
Comparison 1 Oral contraceptive education program (1 session) versus routine counseling, Outcome 3 Known pregnancy by one year.
Comparison 2 Contraceptive counseling (1 session) versus no counseling, Outcome 1 Use of any contraceptive at 8 to 12 weeks postpartum.
Comparison 2 Contraceptive counseling (1 session) versus no counseling, Outcome 2 Choice of modern contraceptive (using or plan to use) at 8 to 12 weeks postpartum.
Comparison 3 LARC script (1‐minute) + routine counseling versus routine counseling, Outcome 1 LARC use after 6 weeks.
Comparison 3 LARC script (1‐minute) + routine counseling versus routine counseling, Outcome 2 Interested in but not using LARC after 6 weeks.
Comparison 3 LARC script (1‐minute) + routine counseling versus routine counseling, Outcome 3 Using any contraceptive after 6 weeks.
Comparison 4 Focused contraceptive counseling (1 session) versus usual care, Outcome 1 Use of highly effective contraceptive method at 3 months.
Comparison 4 Focused contraceptive counseling (1 session) versus usual care, Outcome 2 Increase in contraceptive knowledge by 3 months.
Comparison 5 Home visiting: structured versus routine, Outcome 1 Effective use of contraception at 6 months.
Comparison 6 Home‐based mentoring (multiple visits) versus usual care, Outcome 1 Second birth by 24 months.
Comparison 7 Computer‐assisted motivational interviewing (CAMI) with parenting curriculum versus CAMI‐only versus usual care, Outcome 1 Repeat birth by 24 months.
Comparison 8 Special postpartum care (including contraception) versus routine services (multiple well‐baby contacts), Outcome 1 Repeat pregnancy (self‐report) by 18 months.
Comparison 9 Phone counseling + follow‐up versus usual services, Outcome 1 Repeat pregnancy by 24 months.
Comparison 9 Phone counseling + follow‐up versus usual services, Outcome 2 LARC use at 3 months.
Comparison 10 Home visiting: 4 visits versus 1 visit versus usual care, Outcome 1 Contraception use at 4 months: 4 visits versus 1 visit.
Comparison 10 Home visiting: 4 visits versus 1 visit versus usual care, Outcome 2 Contraception use at 4 months: 1 visit versus usual care.
Comparison 10 Home visiting: 4 visits versus 1 visit versus usual care, Outcome 3 Pregnancy (self‐report) at 4 months postpartum: 4 visits versus 1 visit.
Comparison 10 Home visiting: 4 visits versus 1 visit versus usual care, Outcome 4 Pregnancy (self‐report) at 4 months postpartum: 1 visit versus usual care.
Comparison 11 Health education (including contraception): immediate, later, or no session, Outcome 1 Contraception use at 3 months: immediate session versus no immediate session.
Comparison 11 Health education (including contraception): immediate, later, or no session, Outcome 2 Contraception use at 6 months: immediate session versus no immediate session.
Comparison 11 Health education (including contraception): immediate, later, or no session, Outcome 3 Contraception use at 6 months.
Comparison 11 Health education (including contraception): immediate, later, or no session, Outcome 4 Exclusive breastfeeding at 3 months: immediate session versus no immediate session.
Comparison 11 Health education (including contraception): immediate, later, or no session, Outcome 5 Exclusive breastfeeding >= 5 months: immediate session versus no immediate session.
Comparison 11 Health education (including contraception): immediate, later, or no session, Outcome 6 Exclusive breastfeeding >= 5 months.
| Special education compared with routine or delayed education for contraceptive use | ||||
| Patient or population: women with postpartum status Settings: clinic or community Intervention: special contraception education Comparison: see Comments | ||||
| Outcomes | Relative effect | No of Participants | Quality of the evidence | Comments |
| Second birth (by 24 months) | OR 0.41 (95% CI 0.17 to 1.00) | Participants = 149 | ⊕⊕⊕⊝ | Home mentoring vs usual care |
| Contraception use (at 6 months): immediate session vs no immediate session | OR 1.62 (95% CI 1.06 to 2.50) | Participants = 393 | ⊕⊕⊝⊝ | Education: immediate vs later |
| Effective contraception use (at 6 months) | OR 3.24 (95% CI 1.35 to 7.79) | Participants = 124 (Quinlivan 2003) | ⊕⊕⊝⊝ | Home visiting: structured vs routine |
| GRADE Working Group grades of evidence | ||||
| Study | Curriculum or manual | Provider credentials | Training for intervention | Assessed adherence to protocol | Assessed intervention | Fidelity |
| 4 goals and specific services identified; which professionals provide each component | Directed by nurse practitioner; providers included social worker, pediatrician and nurse practitioner, volunteers | Volunteers were 'trained' | _ | _ | 3 | |
| Format and content identified for sessions including key messages | 3 health educators, 2 midwives, 1 community health worker | Providers were 'trained' to give the health education | Investigators monitored weekly and gave feedback | Knowledge of infant health issues (primary outcome) | 5 | |
| Structured home visits outlined in report | Certified midwives | _ | _ | Knowledge of contraception (primary outcome) | 3 | |
| Developed counseling program, video, pamphlet; development described | Resident physicians, nurses for additional counseling | Training session for resident physicians and nurses | _ | Knowledge of OCs (secondary outcome) | 4 | |
| Curriculum with 19 lessons; order could vary after 2 sessions | 2 Black women, college‐educated, in their 20s, single mothers, living independently | 'Extensive' training provided | Weekly supervisory sessions | _ | 4 | |
| Objectives for each visit; breastfeeding in visits 2 and 3, family planning in visit 4 | Registered midwives | 5 days of special training | _ | _ | 3 | |
| Counseling leaflet used | Physicians | Providers had 40‐minute training on leaflet and interview methods. | _ | _ | 3 | |
| Structured software (computer‐assisted motivational intervention (CAMI)); 20‐min stage‐matched motivational interviewing (MI); parenting curriculum | African American paraprofessional women from participants' communities; had empathetic qualities, rapport with adolescents, community knowledge | 2.5 days on transtheoretical model, MI, and CAMI | First 4 months, counselors met biweekly with MI supervisor (discussed audiotapes and gave feedback) | _ | 4 | |
| Curriculum with standardized format and session structure; teen workbooks with visual material on topics | Masters‐level young women of similar racial‐ethnic background as teens | _ | Process evaluation included delivery | _ | 3 | |
| _ | 2 personal assistants with contraceptive counseling experience (OBGYN resident, 4th‐year medical student) | _ | Contacts planned; additional counseling as needed | _ | 2 | |
| 1‐minute script | Research assistants | _ | _ | _ | 2 | |
| Written structured script on all contraceptive methods by effectiveness | _ | _ | _ | Knowledge assessed with 'widely accepted instrument' | 2 | |
| aAssessed participants' understanding and skills regarding the intervention. | ||||||
| Study | Intervention | Randomization; | Outcome | Follow‐up period | Loss > 20% | Evidence | Evidence of effectiveness |
| _ | _ | _ | _ | _ | High | _ | |
| _ | _ | ‐1 | _ | _ | Moderate | Fewer repeat births (by 24 months) | |
| _ | _ | _ | _ | ‐1 | Moderate | _ | |
| ‐1 | _ | _ | _ | _ | Moderate | _ | |
| ‐1 | _ | _ | _ | _ | Moderate | _ | |
| ‐1 | _ | _ | ‐1 | _ | Moderate to lowc | _ | |
| _ | _ | ‐1 | _ | ‐1 | Low | Use of contraception (at 6 months) | |
| ‐1 | _ | ‐1 | _ | _ | Low | Effective use of contraception (at 6 months) | |
| ‐1 | _ | _ | ‐1 | _ | Low | _ | |
| ‐1 | ‐1 | ‐1 | _ | _ | Very low | Fewer repeat pregnancies (by 18 months) | |
| ‐1 | _ | ‐1 | ‐1 | _ | Very low | Use of contraception (at 8 to 12 weeks) | |
| ‐1 | ‐1 | ‐1d | _ | ‐1d | Very low | Use of highly effective contraception (at 3 months) | |
| aRCTs considered high quality then downgraded: 1) intervention fidelity < 4 criteria; 2) no information on randomization sequence generation or allocation concealment, or risk of bias was high for one; 3) outcome assessment lacked an objective measure (e.g., pregnancy test or structured questionnaire for contraceptive use); 4) follow‐up < 3 months for contraceptive use or < 6 months for pregnancy; 5) loss to follow‐up > 20% | |||||||
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Continuation of oral contraceptives at one year Show forest plot | 1 | 25 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.67 [0.11, 3.99] |
| 2 Switched contraceptives by one year Show forest plot | 1 | 25 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.0 [0.37, 10.92] |
| 3 Known pregnancy by one year Show forest plot | 1 | 25 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.81 [0.11, 6.04] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Use of any contraceptive at 8 to 12 weeks postpartum Show forest plot | 1 | 600 | Odds Ratio (M‐H, Fixed, 95% CI) | 19.56 [11.65, 32.83] |
| 2 Choice of modern contraceptive (using or plan to use) at 8 to 12 weeks postpartum Show forest plot | 1 | 600 | Peto Odds Ratio (Peto, Fixed, 95% CI) | 18.53 [13.15, 26.12] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 LARC use after 6 weeks Show forest plot | 1 | 738 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.40 [0.93, 2.09] |
| 2 Interested in but not using LARC after 6 weeks Show forest plot | 1 | 738 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.31 [0.95, 1.80] |
| 3 Using any contraceptive after 6 weeks Show forest plot | 1 | 734 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.28 [0.79, 2.08] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Use of highly effective contraceptive method at 3 months Show forest plot | 1 | 121 | Odds Ratio (M‐H, Fixed, 95% CI) | 2.23 [1.03, 4.83] |
| 2 Increase in contraceptive knowledge by 3 months Show forest plot | 1 | 121 | Mean Difference (IV, Fixed, 95% CI) | 2.30 [2.00, 2.60] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Effective use of contraception at 6 months Show forest plot | 1 | 124 | Odds Ratio (M‐H, Fixed, 95% CI) | 3.24 [1.35, 7.79] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Second birth by 24 months Show forest plot | 1 | 149 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.41 [0.17, 1.00] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Repeat birth by 24 months Show forest plot | 1 | Odds Ratio (M‐H, Fixed, 95% CI) | Subtotals only | |
| 1.1 CAMI + parenting curriculum versus usual care | 1 | 148 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.48 [0.21, 1.11] |
| 1.2 CAMI‐only versus usual care | 1 | 155 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.63 [0.29, 1.37] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Repeat pregnancy (self‐report) by 18 months Show forest plot | 1 | 221 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.35 [0.17, 0.70] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Repeat pregnancy by 24 months Show forest plot | 1 | 249 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.80 [0.47, 1.35] |
| 2 LARC use at 3 months Show forest plot | 1 | 49 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.29 [0.38, 4.34] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Contraception use at 4 months: 4 visits versus 1 visit Show forest plot | 1 | 565 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.25 [0.89, 1.75] |
| 2 Contraception use at 4 months: 1 visit versus usual care Show forest plot | 1 | 580 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.86 [0.62, 1.20] |
| 3 Pregnancy (self‐report) at 4 months postpartum: 4 visits versus 1 visit Show forest plot | 1 | 554 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.49 [0.33, 6.74] |
| 4 Pregnancy (self‐report) at 4 months postpartum: 1 visit versus usual care Show forest plot | 1 | 585 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.00 [0.20, 5.01] |
| Outcome or subgroup title | No. of studies | No. of participants | Statistical method | Effect size |
| 1 Contraception use at 3 months: immediate session versus no immediate session Show forest plot | 1 | 402 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.50 [0.88, 2.54] |
| 2 Contraception use at 6 months: immediate session versus no immediate session Show forest plot | 1 | 393 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.62 [1.06, 2.50] |
| 3 Contraception use at 6 months Show forest plot | 1 | 393 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.89 [0.58, 1.37] |
| 3.1 Immediate + later sessions vs immediate session | 1 | 199 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.85 [0.48, 1.52] |
| 3.2 Later session vs no session | 1 | 194 | Odds Ratio (M‐H, Fixed, 95% CI) | 0.95 [0.50, 1.80] |
| 4 Exclusive breastfeeding at 3 months: immediate session versus no immediate session Show forest plot | 1 | 403 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.00 [0.67, 1.49] |
| 5 Exclusive breastfeeding >= 5 months: immediate session versus no immediate session Show forest plot | 1 | 390 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.01 [0.65, 1.57] |
| 6 Exclusive breastfeeding >= 5 months Show forest plot | 1 | 390 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.29 [0.83, 2.00] |
| 6.1 Immediate + later sessions vs immediate session | 1 | 198 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.55 [0.83, 2.90] |
| 6.2 Later session vs no session | 1 | 192 | Odds Ratio (M‐H, Fixed, 95% CI) | 1.06 [0.56, 1.99] |
