Abstract
Objectives
The objectives of this analysis were to 1) estimate prevalence of contraceptive use among women at risk for unintended pregnancy and 2) identify correlates of contraceptive use among women with ongoing or potential need for contraceptive services in Puerto Rico during the 2016 Zika virus (ZIKV) outbreak.
Study Design
We conducted a cell-phone survey July–November, 2016. Women aged 18–49 years living in Puerto Rico were eligible. We completed 3,059 interviews; the overall response rate was 69.2%. After weighting, the data provide population-based estimates. For this analysis, we included women at risk for unintended pregnancy, and assessed ongoing or potential need for contraceptive services in this group, excluding women using permanent contraceptive methods.
Results
Most women reported using contraception (82.8%), and use increased with age. Female sterilization and male condoms were most frequently reported (40.8% and 17.1%, respectively). Among women with ongoing or potential need for contraceptive services, 24.7% talked to a healthcare provider about ZIKV, and 31.2% reported a change in childbearing intentions due to ZIKV. Most women were at least a little worried about getting infected with ZIKV (74.3%) or having a baby with a birth defect (80.9%). Being very worried about getting infected with ZIKV and already having Zika were significantly associated with use of any contraception (adjusted prevalence ratio: 1.19, 95% CI: 1.03–1.38 and 1.32, 95% CI: 1.01–1.72, respectively).
Conclusions
These findings underscore the need for regular contraceptive prevalence studies to inform programs about contraceptive needs, especially during public health emergencies.
Implications:
When the 2016 Zika virus outbreak began in Puerto Rico there were no recent population-based data available on contraceptive prevalence. To fill this information gap, we conducted a population-based survey. Our findings provided baseline contraceptive prevalence estimates to support response planning and allocation of health resources.
Introduction
The Puerto Rico Department of Health (PRDH) reported the first locally acquired infection with Zika virus (ZIKV) on December 31, 2015 [1]. Reported cases remained low until April 2016 and then quickly increased, peaking in August 2016. Among states and territories in the United States, Puerto Rico was the most severely affected by the ZIKV outbreak [2-4]. ZIKV infection during pregnancy can result in severe birth defects including microcephaly and structural abnormalities of the brain and eye [5-13]. In 2016, before we understood the range of brain and eye defects that could be caused by ZIKV infection during pregnancy, it was estimated that 5,900 to 10,300 pregnant women would be infected during the initial ZIKV outbreak in Puerto Rico. As a result, an estimated 100 to 270 infants would be born with microcephaly in the absence of prevention measures [14]. As in the continental United States, abortion is legally permitted in Puerto Rico [15].
Because neither a vaccine nor prophylactic medication is available to prevent ZIKV infection, effective contraception for women who want to avoid pregnancy is a key countermeasure for preventing ZIKV-associated fetal and infant outcomes in areas with ZIKV. In 2016, the Centers for Disease Control and Prevention (CDC) recommended that women of reproductive age who have had or anticipate ZIKV exposure and who do not want to become pregnant should use the most effective contraceptive method that can be used correctly and consistently. Healthcare providers should counsel women accordingly [16].
The 2016 ZIKV outbreak in Puerto Rico prompted an examination of contraceptive access and use. To understand access issues, Tepper et al. carried out an assessment using key informant interviews during the first few months of the outbreak. Key informants reported limited access to long-acting reversible contraceptive (LARC) methods due to lack of availability, inadequate training and reimbursement of providers, incomplete insurance coverage, and high costs [3]. In addition to access issues, there was insufficient understanding about the prevalence of contraceptive use, which had last been measured during the 2002 Behavioral Risk Factor Surveillance System (BRFSS) survey [17]. To address this information gap, the PRDH and CDC conducted a population-based survey in 2016 among women of reproductive age (18–49 years) residing in Puerto Rico to collect information on contraceptive use and ZIKV-related attitudes and prevention behaviors. The objectives of this analysis were to 1) estimate the prevalence of contraceptive use among women at risk for unintended pregnancy and 2) identify correlates of contraceptive use among women with ongoing or potential need for contraceptive services in Puerto Rico.
Methods
2.1. Data Collection
We conducted a cell phone survey from July–November, 2016. The survey consisted of 48 questions covering ZIKV-prevention behaviors, contraception, and childbearing. All survey questions underwent cognitive testing with the intended population before implementation. The survey was available in Spanish and English and predominantly completed in Spanish (>99%).
Random samples of Puerto Rico cell phone numbers were purchased from Marketing Systems Group, Inc., for the survey. Interviewers called sampled cell phone numbers repeatedly until someone answered the phone or until the number was tried at least six times. Days of the week and times of day were varied, including evening and weekend calls. The survey was administered in WINCATI, a computer-assisted telephone interview software.
2.2. Ethics
After reading a brief script describing the survey, interviewers obtained verbal consent from respondents before proceeding. The CDC determined the survey to be a non-research, public health practice activity during an emergency response [18].
2.3. Measures
To determine contraceptive use prevalence, women were asked what they or their partners were currently doing to prevent pregnancy. Typical-use effectiveness of contraceptive methods is classified according to the estimated percentage of users who experience pregnancy during the first year of typical use: most effective (<1%), moderately effective (4%–7%), and less effective (>13%) [19]. We calculated prevalence of the following most effective methods separately: female sterilization, male sterilization, and long-acting reversible contraception (LARC) (intrauterine devices [IUDs] and contraceptive implants). Moderately effective contraceptive methods were combined and included hormone injections, contraceptive pills, transdermal contraceptive patches, and vaginal rings. Prevalence of male condoms as a primary contraceptive method was calculated separately, and other less effective methods were combined: diaphragm, female condoms, cervical cap, sponge, withdrawal, spermicide, fertility awareness-based methods, and emergency contraception. Among women who reported using more than one contraceptive method, the most effective method reported was used for classification.
The following factors were examined to determine their association with contraceptive use among women with ongoing need: 1) having had a discussion with a healthcare provider about ZIKV (yes/no), 2) being unable to visit a health care provider for contraceptive services in the past 12 months (yes/no), 3) having had a change in childbearing intentions due to ZIKV (yes/no), 4) degree of worry about being infected with ZIKV, and 5) degree of worry about having a baby with a birth defect.
We included the following covariates in our analysis: age (18–24 years, 25–34 years, 35–49 years); education (less than high school, high school graduate, more than high school); health insurance (private, Medicaid/public, no insurance); relationship status (married/long-term relationship with a man, married/long-term relationship with a woman, divorced/widowed/separated, never married and not in a long-term relationship); and having a previous pregnancy (yes/no).
2.4. Participants
To be eligible for the survey, women had to be aged 18–49 and living in Puerto Rico at the time of the interview. For overall contraceptive prevalence estimates, the sample was limited to women at risk for unintended pregnancy, defined as sexually active women (reporting sex with a man in the last 3 months or reporting being married or in a long-term relationship with a man), excluding women who were currently pregnant, desiring or ambivalent about pregnancy, or noncontraceptively sterile. To assess factors associated with any contraceptive use, we excluded women using permanent contraceptive methods (defined as female and male sterilization).
2.5. Weighting
To produce estimates that were representative of the population, the data were weighted using raking methodology [20]. The raking margins used were based on population estimates available by age and sex from the U.S. Census Bureau’s American Community Survey in Puerto Rico for 2015 [21]. The data were raked to the population estimates for women aged 18 to 49 using seven age categories (18–19, 20–24, 25–29, 30–34, 35–39, 40–44, 45–49) and five marital status categories (married, separated, divorced, widowed, never married).
2.6. Data Analysis
Cooperation and response rates were calculated using standards set by the American Association of Public Opinion Research. The cooperation rate is the percentage of contacted and eligible participants who completed the survey. The response rate is calculated using completed interviews in the numerator and an estimate of all eligible participants in the denominator. The response rate calculation assumes that unresolved cell phone numbers contain the same percentage of eligible participants as the cell phone numbers for which eligibility or ineligibility were determined [22].
Weighted prevalence estimates and 95% confidence intervals (CIs) were calculated for overall contraceptive use among women at risk for unintended pregnancy and by contraceptive method type and age. For women at risk for unintended pregnancy who were not using contraception, we collected the primary reason for non-use. Bivariate analyses were conducted to examine prevalence of any contraceptive use among women with ongoing need by factors of interest and demographic characteristics, using the Rao-Scott chi-square test. We estimated prevalence ratios from log-binomial regression models to examine the association between each factor and any contraception use, controlling for age, education, health insurance, relationship status, and previous pregnancy. All data analyses were conducted in SAS version 9.4, using weighted survey methods to account for sampling weights.
Results
As shown in the Figure, a total of 38,701 cell phone numbers were sampled, and 16,311 calls were answered. Of the calls answered, 3,169 respondents met eligibility criteria, and of those, 3,059 agreed to participate (cooperation rate = 96.5% and overall response rate 69.2%). An estimated 58.0% (95% CI: 56.1%–59.9%) of women aged 18–49 were at risk for unintended pregnancy at the time of the survey (Table 1).
Figure.
Sampling outcomes for a reproductive health survey in Puerto Rico, 2016
Table 1.
Demographic characteristics of women age 18-49 completing a reproductive health survey during the Zika Virus Outbreak, Puerto Rico, 2016
| N (Unweighted) |
% (Weighted) |
95% CI | |
|---|---|---|---|
| Overall | 3,059 | 100.0 | (–) |
| Age group | |||
| 18–24 years | 586 | 23.1 | (21.4–24.8) |
| 25–34 years | 1,026 | 30.3 | (28.6–32.1) |
| 35–49 years | 1,447 | 46.6 | (44.7–48.5) |
| Ethnicity 1 | |||
| Hispanic/Latina | 3,045 | 99.7 | (99.5–99.9) |
| Non-Hispanic/Latina | 11 | 0.3 | (0.1–0.5) |
| Place of birth 2 | |||
| Puerto Rico | 2,757 | 90.5 | (89.4–91.6) |
| Other (United States) | 211 | 6.7 | (5.7–7.6) |
| Other | 90 | 2.8 | (2.2–3.5) |
| Education 2 | |||
| Less than high school | 179 | 5.7 | (4.8–6.5) |
| High school graduate | 712 | 24.3 | (22.6–25.9) |
| More than high school | 2,167 | 70.1 | (68.3–71.8) |
| Health insurance 3 | |||
| Private/through employer | 1,214 | 38.2 | (36.4–40.1) |
| Medicaid or other public insurance | 1,686 | 57.4 | (55.5–59.3) |
| No insurance | 129 | 4.4 | (3.6–5.2) |
| Relationship status 4 | |||
| Married/long-term relationship with a man | 1,506 | 31.8 | (30.2–33.3) |
| Married/long-term relationship with a woman | 30 | 0.6 | (0.4–0.8) |
| Divorced/Widowed/Separated | 443 | 18.6 | (17.0–20.3) |
| Never married and not in a long-term relationship | 1,069 | 49.0 | (47.0–51.9) |
| Municipality of residence 5 | |||
| Aguadilla | 183 | 5.9 | (5.0–6.8) |
| Arecibo | 353 | 11.6 | (10.4–12.9) |
| Bayamon | 523 | 17.0 | (15.6–18.4) |
| Caguas | 546 | 17.4 | (16.0–18.9) |
| Fajardo | 99 | 3.3 | (2.6–4.0) |
| Mayaguez | 206 | 6.5 | (5.6–7.4) |
| Ponce | 493 | 16.2 | (14.8–17.6) |
| San Juan Metro | 648 | 22.0 | (20.4–23.6) |
| Not at risk for unintended pregnancy 6 | 1,060 | 42.0 | (40.1–43.9) |
| Never had sex | 127 | 5.8 | (4.9–6.8) |
| Not sexually active/not in a long-term relationship7 | 857 | 36.3 | (34.3–38.2) |
| Noncontraceptively sterile | 97 | 3.0 | (2.3–3.6) |
| Pregnant/ Seeking Pregnancy/Ambivalent | 242 | 6.6 | (5.7–7.4) |
| At risk for unintended pregnancy 8 | 1,962 | 58.0 | (56.1–59.9) |
Three missing responses.
One missing response.
Thirty missing responses.
Eleven missing responses.
Eight missing responses.
37 Women could not be classified due to missing/refused responses. Some women reported multiple reasons for not at risk for unintended pregnancy; percentages sum to >100.
Defined as not sexually active in the last 3 months and not in a long-term relationship with a man.
Defined as sexually active women (reporting sex with a man in the last 3 months or reporting being married or in a long-term relationship with a man), excluding women who were currently pregnant, desiring or ambivalent about pregnancy, or noncontraceptively sterile.
Most women at risk for unintended pregnancy reported using contraception (82.8%, 95% CI: 81.0%–84.7%), and use of contraception increased with age (Table 2). Female sterilization was the most frequently reported contraceptive method overall (40.8%, 95% CI: 38.5%–43.1%) and among women aged 25–34 and 35–49. Male condoms were the second most frequently reported contraceptive method overall (17.1%, 95% CI: 15.2%–19.0%) and the most frequently reported method for women aged 18–24 (37.5%, 95% CI: 31.8%–43.3%). Overall, 9.1% (95% CI: 7.7%–10.5%) of women reported using a moderately effective contraceptive method (pills, patch, shot, or ring), and this proportion was highest among women aged 18–24 (16.7%) (Table 2). Among all moderately effective methods, the pill was the most frequently reported (7.9%, 95% CI: 6.6%–9.2%, data not shown). Male partner sterilization was reported by 6.7% (95% CI: 5.6%–7.9%) of women at risk for unintended pregnancy and by less than 1% of women aged 18–24. LARC methods were reported by 3.9%, (95% CI: 3.0%–4.8%) of women. The only LARC methods reported were copper and hormonal IUDs; no women reported using a contraceptive implant. Among the estimated 453,223 women in Puerto Rico at risk for unintended pregnancy, an estimated 179,084 (39.6%, 95% CI: 37.2%–41.9%) were using a less effective or no method (data not shown).
Table 2.
Prevalence of contraceptive use by method type and age group among women aged 18-49 at risk for unintended pregnancy1 during the 2016 Zika Virus Outbreak, Puerto Rico
| Contraceptive Method | Overall Weighted N=453,223 |
18–24 years | 25–34 years | 35–49 years | ||||
|---|---|---|---|---|---|---|---|---|
| % | 95% CI | % | 95% CI | % | 95% CI | % | 95% CI | |
| Any Method | 82.8 | (81.0–84.7) | 68.1 | (62.7–73.6) | 80.6 | (77.3–84.0) | 91.0 | (89.0–92.9) |
| Female sterilization | 40.8 | (38.5–43.1) | 4.2 | (2.2–6.2) | 35.1 | (31.4–38.9) | 61.2 | (57.8–64.6) |
| Male sterilization | 6.7 | (5.6–7.9) | 0.4 | (0.0–1.2) | 5.2 | (3.4–6.9) | 10.7 | (8.6–12.8) |
| Intrauterine contraception2 | 3.9 | (3.0–4.8) | 4.7 | (2.4–7.1) | 5.7 | (3.8–7.5) | 2.2 | (1.2–3.2) |
| Moderately effective methods3 | 9.1 | (7.7–10.5) | 16.7 | (12.4–21.0) | 10.7 | (8.2–13.2) | 4.4 | (3.0–5.9) |
| Male condom | 17.1 | (15.2–19.0) | 37.5 | (31.8–43.3) | 17.8 | (14.5–21.0) | 7.6 | (5.6–9.5) |
| Other less effective methods4 | 5.3 | (4.2–6.3) | 4.6 | (2.2–7.0) | 6.2 | (4.3–8.1) | 4.9 | (3.4–6.4) |
| None | 17.2 | (15.3–19.0) | 31.9 | (26.4–37.3) | 19.4 | (16.1–22.7) | 9.0 | (7.1–11.0) |
Defined as sexually active women (reporting sex with a man in the last 3 months or reporting being married or in a long-term relationship with a man), excluding women who were currently pregnant, desiring or ambivalent about pregnancy, or noncontraceptively sterile.
Intrauterine devices were the only reported long-acting reversible contraceptive method; no women reported use of the contraceptive implant.
Includes pills, patch, ring, or injection.
Includes rhythm/fertility–based awareness, withdrawal, female condom, and diaphragm.
Among women at risk for unintended pregnancy who were not using a contraceptive method when interviewed, 50.8% did not specify why. The next most frequently reported reasons were that they didn’t think about it/forgot (22.6%) or they did not currently have a male sex partner/were not sexually active (10.1%) (Table 3).
Table 3 ––
Percentage distribution of primary reason for not using contraception among women aged 18-49 at risk for unintended pregnancy1 by age group, during the 2016 Zika Virus Outbreak, Puerto Rico
| Primary Reason for Not Using Contraception | Respondent Age | |||||||
|---|---|---|---|---|---|---|---|---|
| All Ages | 18–24 years | 25–34 years | 35–49 years | |||||
| % | 95% CI | % | 95% CI | % | 95% CI | % | 95% CI | |
| Didn't think about it, forgot, or don't know | 22.6 | (17.3–27.9) | 32.6 | (22.3–42.9) | 18.8 | (11.0–26.5) | 13.2 | (4.5–21.9) |
| No male partner/ not sexually active | 10.1 | (6.2–14.0) | 4.5 | (0.0–9.1) | 17.7 | (9.7–25.6) | 6.4 | (0.0–13.0) |
| Don't think I can get pregnant | 6.5 | (3.6–9.4) | 1.9 | (0.0–4.7) | 3.5 | (0.0–7.2) | 18.3 | (9.1–27.4) |
| Worried about side effects | 3.7 | (1.3–6.1) | 4.1 | (0.0–8.3) | 5.1 | (0.6–9.7) | 1.0 | (0.0–3.0) |
| Partner objects | 2.4 | (0.6–4.1) | 2.0 | (0.0–5.0) | 2.4 | (0.0–5.3) | 2.9 | (0.0–6.1) |
| Was breastfeeding or recently postpartum | 2.1 | (0.8–3.5) | 1.8 | (0.0–3.8) | 3.3 | (0.6–5.9) | 0.9 | (0.0–2.7) |
| Can't pay for contraception | 1.1 | (0–2.4) | 1.9 | (0.0–4.8) | 1.1 | (0.0–2.6) | 0.0 | –– |
| Religious reasons | 0.7 | (0–1.7) | 0.0 | –– | 1.2 | (0.0–3.4) | 1.0 | (0.0–3.0) |
| Other | 50.8 | (44.6–57.0) | 51.2 | (40.3–62.1) | 47.0 | (37.1–56.9) | 56.3 | (44.5–68.2) |
Defined as sexually active women (reporting sex with a man in the last 3 months or reporting being married or in a long-term relationship with a man), excluding women who were currently pregnant, desiring or ambivalent about pregnancy, or noncontraceptively sterile.
Among women with ongoing or potential need for contraceptive services, 67.3% reported using any contraceptive method (Table 4). Women with private health insurance were more likely to use contraception, and women who reported a previous pregnancy were less likely to use contraception. Approximately one-quarter (24.7%) of women had ever talked to a healthcare provider about ZIKV, and 31.2% of women reported a change in childbearing intentions due to ZIKV; these factors did not differ for women using versus not using contraception. Most women were at least a little worried about getting infected with ZIKV (74.3%) and 4.3% reported they already had Zika. Most women also reporting being at least a little worried about having a baby with a birth defect (80.9%), and this did not differ by contraception use status. Women who reported being unable to visit a healthcare provider for contraception in the past 12 months were significantly less likely to use contraception (20.7% vs 14.5%, p=0.03) (Table 4).
Table 4.
Use of any contraceptive method versus no method among women aged 18-49 with ongoing or potential need for contraceptive services1, during the 2016 Zika Virus Outbreak, Puerto Rico
| Total Weighted N=238,179 |
Any Method N=160,402 |
No Method N=77,777 |
P– Value |
||||
|---|---|---|---|---|---|---|---|
| Characteristic | % | 95% CI | % | 95% CI | % | 95% CI | |
| Total | 67.3 | (64.1–70.6) | 32.7 | (29.4–35.9) | |||
| Age group | |||||||
| 18–24 years | 36.9 | (33.5–40.4) | 36.5 | (32.4–40.7) | 37.8 | (31.8–43.8) | 0.93 |
| 25–34 years | 38.4 | (35.1–41.8) | 38.6 | (34.6–42.6) | 38.1 | (32.3–44.0) | |
| 35–49 years | 24.6 | (21.7–27.5) | 24.9 | (21.3–28.5) | 24.1 | (19.2–29.0) | |
| Education | |||||||
| Less than high school | 4.2 | (2.8–5.5) | 3.6 | (2.1–5.1) | 5.3 | (2.7–7.9) | 0.14 |
| High school graduate | 21.9 | (19.0–24.8) | 20.4 | (17.0–23.9) | 24.9 | (19.7–30.2) | |
| More than high school | 73.9 | (70.9–77.0) | 76.0 | (72.3–79.6) | 69.7 | (64.2–75.3) | |
| Health insurance | |||||||
| Private/through employer | 43.5 | (40.1–47.0) | 48.1 | (43.8–52.3) | 34.2 | (28.5–39.8) | <0.01 |
| Medicaid or other public insurance | 51.0 | (47.5–54.4) | 46.3 | (42.1–50.6) | 60.5 | (54.7–66.4) | |
| No insurance | 5.5 | (3.9–7.1) | 5.6 | (3.6–7.6) | 5.3 | (2.6–8.0) | |
| Relationship status | |||||||
| Married/long-term relationship with a man | 32.3 | (29.4–35.1) | 32.1 | (28.7–35.6) | 32.6 | (27.5–37.5) | 0.65 |
| Divorced/Widowed/Separated | 10.1 | (7.9–12.4) | 10.8 | (7.8–13.6) | 8.8 | (5.2–12.3) | |
| Never married and not in a long-term relationship | 57.6 | (54.3–60.9) | 57.1 | (53.1–61.1) | 58.7 | (53.1–64.4) | |
| Ever been pregnant | |||||||
| Yes | 54.9 | (51.4–58.3) | 51.8 | (47.6–56.1) | 61.1 | (55.1–67.1) | 0.01 |
| No | 45.1 | (41.7–48.6) | 48.2 | (43.9–52.4) | 38.9 | (32.9–44.9) | |
| Talked to a healthcare provider about Zika | |||||||
| Yes | 24.7 | (21.8–27.6) | 26.0 | (22.4–29.6) | 22.1 | (17.3–26.9) | 0.21 |
| No | 75.3 | (72.4–78.2) | 74.0 | (70.4–77.6) | 77.9 | (73.1–82.7) | |
| Unable to see a provider in the last 12 months for contraception | |||||||
| Yes | 16.5 | (13.9–19.1) | 14.5 | (11.5–17.5) | 20.7 | (15.7–25.6) | 0.03 |
| No | 83.5 | (80.9–86.1) | 85.5 | (82.5–88.5) | 79.3 | (74.4–84.3) | |
| Reported a change in childbearing intentions due to Zika | |||||||
| Yes | 31.2 | (28.1–34.4) | 29.9 | (26.1–33.7) | 33.9 | (28.2–39.6) | 0.25 |
| No | 68.8 | (65.6–71.9) | 70.1 | (66.3–73.9) | 66.1 | (60.4–71.8) | |
| Worried about getting infected with Zika | |||||||
| Very worried | 27.6 | (24.6–30.7) | 28.5 | (24.7–32.3) | 25.8 | (20.7–31.0) | 0.15 |
| A little/somewhat worried | 46.7 | (43.3–50.2) | 47.8 | (43.5.52.0) | 44.5 | (38.5–50.6) | |
| Already had Zika | 4.3 | (3.0–5.6) | 4.6 | (3.0–6.3) | 3.6 | (1.4–5.9) | |
| Not worried | 21.3 | (18.5–24.2) | 19.1 | (15.7–22.5) | 26.0 | (20.6–31.3) | |
| Worried about having a baby with a birth defect | |||||||
| Very worried | 68.7 | (65.5–72.0) | 68.9 | (64.9–72.8) | 68.4 | (62.7–74.1) | 0.99 |
| A little/somewhat worried | 12.2 | (9.9–14.5) | 12.1 | (9.4–14.9) | 12.5 | (8.3–16.6) | |
| Not worried | 19.1 | (16.3–21.8) | 19.0 | (15.6–22.4) | 19.2 | (14.3–24.0) | |
Defined as women at risk for unintended pregnancy [sexually active women (reporting sex with a man in the last 3 months or reporting being married or in a long-term relationship with a man), excluding women who were currently pregnant, desiring or ambivalent about pregnancy, or noncontraceptively sterile] And Excludes women who reported a permanent contraceptive method.
Table 5 shows the results of the multivariable log-binomial regression modeling of factors associated with any use of contraception among women with potential or ongoing need for contraception. Being very worried about getting infected with ZIKV (adjusted prevalence ratio (APR): 1.19, 95% CI: 1.03–1.38) or reporting already having Zika (APR: 1.32, 95% CI 1.01–1.72) were the only factors significantly associated with any contraception use compared to not being worried. Women who were unable to visit a provider in the last 12 months for contraception were less likely to use any contraception, but this association was not statistically significant (APR: 0.90, 95% CI: 0.78-1.05).
Table 5.
Factors associated with use of any contraception among women aged 18-49 with ongoing or potential need for contraceptive services1, during the 2016 Zika Virus Outbreak, Puerto Rico
| Crude Prevalence Ratio |
Adjusted Prevalence Ratio2 |
||
|---|---|---|---|
| Covariates | Estimate | Estimate | 95% CI |
| Talked to a healthcare provider about Zika | |||
| Yes | 1.07 | 1.04 | (0.94–1.16) |
| No | Ref | ref | –– |
| Unable to see a provider in the last 12 months for contraception | |||
| Yes | 0.86 | 0.90 | (0.78–1.05) |
| No | ref | ref | –– |
| Reported a change in childbearing intentions due to Zika | |||
| Yes | 0.94 | 0.92 | (0.83–1.04) |
| No | ref | ref | –– |
| Worried about getting infected with Zika | |||
| Very worried | 1.15 | 1.19 | (1.03–1.38) |
| A little/somewhat worried | 1.14 | 1.11 | (0.96–1.27) |
| Already had Zika | 1.32 | 1.32 | (1.01–1.72) |
| Not worried | ref | ref | –– |
| Worried about having a baby with a birth defect | |||
| Very worried | 1.00 | 1.01 | (0.88–1.14) |
| A little/somewhat worried | 0.99 | 1.02 | (0.87–1.21) |
| Not worried | ref | ref | –– |
Defined as women at risk for unintended pregnancy [sexually active women (reporting sex with a man in the last 3 months or reporting being married or in a long-term relationship with a man), excluding women who were currently pregnant, desiring or ambivalent about pregnancy, or noncontraceptively sterile] And Excludes women who reported a permanent contraceptive method.
Adjusted for age group, education, health insurance, relationship status, and past pregnancy.
Discussion
The ZIKV outbreak in Puerto Rico highlighted the need for up-to-date data on contraception use during a public health emergency. We leveraged the BRFSS platform in Puerto Rico to rapidly collect data needed for response planning. We found that while most women reported use of a contraceptive method, nearly 2 in 5 used either a less effective or no method of contraception when interviewed in mid-late 2016, which coincided with the peak of the ZIKV outbreak. LARC use was reported by 4% of women, substantially less than the estimated 14% of women in the 50 U.S. states using a LARC method [23]. Additionally, no women in the survey reported using a contraceptive implant.
In adjusted analysis of women with ongoing or potential need for contraceptive services, women who were very worried about getting infected with ZIKV and women who already had Zika were 19% and 32% more likely to use any contraception compared to women who were not worried, respectively. Additionally, women who were not able to visit a provider in the last 12 months for contraception were 10% less likely to use any contraception, but this association was not statistically significant.
In the 2002 BRFSS survey, 84% of women aged 18–44 at risk for unintended pregnancy in Puerto Rico were using a contraceptive method, and among contraceptive users, the most common methods were female sterilization, pill, condoms, and the rhythm method [17]. Our 2016 survey resulted in a similar prevalence of any contraception use among women aged 18–49 (83%) with female sterilization, male condom, moderately effective methods, and male sterilization as the most common methods reported.
To our knowledge, this is the only population-based survey among women of reproductive age that collected data on contraceptive prevalence and ZIKV and attitudes in Puerto Rico during the 2016 ZIKV outbreak. The analysis is strengthened by the large sample size and high response and cooperation rates. This analysis is also subject to several limitations. Data were self-reported and may be subject to social desirability bias and recall bias, particularly for reporting of highly sensitive questions, such as those about sexual experience and contraceptive use. We were unable to capture the primary reason for not using contraception for half of respondents who reported no contraceptive use because free text was not captured for this survey question. In addition, some data were contradictory, since 10% of women categorized as sexually active who were not using contraception said the reason was that they did not have a male partner or were not sexually active. Additionally, the data are cross-sectional, so it is not possible to establish a temporal relationship between the factors examined and initiation of contraceptive method. Lastly, we were unable to collect data from adolescents younger than 18 who may be at increased risk for unintended pregnancy.
The findings from this analysis provide baseline contraceptive prevalence estimates to support reponse planning and allocation of health resources, especially for the Zika Contraception Access Network (Z-CAN) in Puerto Rico [3, 24]. Z-CAN was a network of trained physician providers that provided women with access to client-centered contraceptive counseling and a full range of contraceptive methods at no cost during the ZIKV outbreak in Puerto Rico. The program served more than 20,000 women, over two-thirds of whom chose a LARC method. The program launched in May 2016, but rollout was initially slow, and most women were served in 2017, after the peak of the ZIKV outbreak in Puerto Rico [24]. Ongoing contraceptive prevalence studies can help decision makers understand the magnitude of contraceptive use and identify gaps in contraceptive access, particularly during public health emergencies that impact pregnancy and pregnancy outcomes.
Footnotes
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.
References
- [1].Thomas DL, Sharp TM, Torres J, Armstrong PA, Munoz-Jordan J, Ryff KR, et al. Local transmission of Zika virus—Puerto Rico, November 23, 2015–January 28, 2016. Morbidity and Mortality Weekly Report. 2016. February 19;65(6):154–8. [DOI] [PubMed] [Google Scholar]
- [2].Dirlikov E, Ryff KR, Torres-Aponte J, Thomas DL, Perez-Padilla J, Munoz-Jordan J, et al. Update: ongoing Zika virus transmission—Puerto Rico, November 1, 2015–April 14, 2016. Morbidity and Mortality Weekly Report. 2016. May 6;65(17):451–5. [DOI] [PubMed] [Google Scholar]
- [3].Tepper NK, Goldberg HI, Bernal MI, Rivera B, Frey MT, Malave C, et al. Estimating contraceptive needs and increasing access to contraception in response to the Zika virus disease outbreak—Puerto Rico, 2016. Morbidity and Mortality Weekly Report. 2016. April 1;65(12):311–4. [DOI] [PubMed] [Google Scholar]
- [4].Estado Libre Asociado De Puerto Rico Departamento de Salud. Informe Semanal de Enfermedades Arbovirales. 2016. http://www.salud.gov.pr/Estadisticas-Registros-y-Publicaciones/Pages/Informe-Arboviral.aspx. Published 2016. Accessed January 4, 2018. [Google Scholar]
- [5].Rasmussen SA, Jamieson DJ, Honein MA, Petersen LR. Zika virus and birth defects—reviewing the evidence for causality. New England Journal of Medicine. 2016. May 19;374(20):1981–7. [DOI] [PubMed] [Google Scholar]
- [6].Brasil P, Pereira JP Jr, Moreira ME, Ribeiro Nogueira RM, Damasceno L, et al. Zika virus infection in pregnant women in Rio de Janeiro. New England Journal of Medicine. 2016. December 15;375(24):2321–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Calvet G, Aguiar RS, Melo AS, Sampaio SA, De Filippis I, Fabri A, et al. Detection and sequencing of Zika virus from amniotic fluid of fetuses with microcephaly in Brazil: a case study. The Lancet infectious diseases. 2016. June 1;16(6):653–60. [DOI] [PubMed] [Google Scholar]
- [8].de Paula Freitas B, de Oliveira Dias JR, Prazeres J, Sacramento GA, Ko AI, Maia M, et al. Ocular findings in infants with microcephaly associated with presumed Zika virus congenital infection in Salvador, Brazil. JAMA ophthalmology. 2016. May 1;134(5):529–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Driggers RW, Ho CY, Korhonen EM, Kuivanen S, Jääskeläinen AJ, Smura T, et al. Zika virus infection with prolonged maternal viremia and fetal brain abnormalities. New England Journal of Medicine. 2016. June 2;374(22):2142–51. [DOI] [PubMed] [Google Scholar]
- [10].Hazin AN, Poretti A, Di Cavalcanti Souza Cruz D, Tenorio M, van der Linden A, Pena LJ, et al. Computed tomographic findings in microcephaly associated with Zika virus. New England Journal of Medicine. 2016. June 2;374(22):2193–5. [DOI] [PubMed] [Google Scholar]
- [11].Mlakar J, Korva M, Tul N, Popović M, Poljšak-Prijatelj M, Mraz J, et al. Zika virus associated with microcephaly. New England Journal of Medicine. 2016. March 10;374(10):951–8. [DOI] [PubMed] [Google Scholar]
- [12].Oliveira MA, Malinger G, Ximenes R, Szejnfeld PO, Alves SS, Bispo DF. Zika virus intrauterine infection causes fetal brain abnormality and microcephaly: tip of the iceberg?. Ultrasound in obstetrics & gynecology: the official journal of the International Society of Ultrasound in Obstetrics and Gynecology. 2016. January;47(1):6. [DOI] [PubMed] [Google Scholar]
- [13].Ventura CV, Maia M, Bravo-Filho V, Góis AL, Belfort R. Zika virus in Brazil and macular atrophy in a child with microcephaly. The Lancet. 2016. January 16;387(10015):228. [DOI] [PubMed] [Google Scholar]
- [14].Ellington SR, Devine O, Bertolli J, Quiñones AM, Shapiro-Mendoza CK, Perez-Padilla J, et al. Estimating the number of pregnant women infected with Zika virus and expected infants with microcephaly following the Zika virus outbreak in Puerto Rico, 2016. JAMA pediatrics. 2016. October 1;170(10):940–5. [DOI] [PubMed] [Google Scholar]
- [15].Guttmacher Institute. Abortion in Latin America and the Caribbean. Guttmacher Institute; 2017. [Cited 2020 March 01]. Available from: https://www.guttmacher.org/sites/default/files/factsheet/ib_aww-latin-america.pdf. [Google Scholar]
- [16].Petersen EE, Meaney-Delman D, Neblett-Fanfair R, Havers F, Oduyebo T, Hills SL, et al. Update: interim guidance for preconception counseling and prevention of sexual transmission of Zika virus for persons with possible Zika virus exposure—United States, September 2016. Morbidity and Mortality Weekly Report. 2016. October 7;65(39):1077–81. [DOI] [PubMed] [Google Scholar]
- [17].Bensyl DM, Iuliano AD, Carter M, Santelli J, Gilbert BC. Contraceptive use—United States and territories, behavioral risk factor surveillance system, 2002. Morbidity and Mortality Weekly Report: Surveillance Summaries. 2005. November 18;54(6):1–72. [PubMed] [Google Scholar]
- [18].Centers for Disease Control and Prevention. Distinguishing Public Health Research and Public Health Nonresearch. 2010. https://www.cdc.gov/os/integrity/docs/cdc-policy-distinguishing-public-health-research-nonresearch.pdf [Google Scholar]
- [19].Hatcher R, Nelson A, Trussell J, et al. Contraceptive Technology. 21st ed. Ayer Company Publishers, Inc, New York, NY: (2018). [Google Scholar]
- [20].Behavioral Risk Factor Surveiallance System: Weighting BRFSS Data, BRFSS 2016. https://www.cdc.gov/brfss/annual_data/2016/pdf/weighting_the-data_webpage_content.pdf
- [21].U.S. Census Buerau American Fact Finder: Annual Estimates of the Resident Population: 2015 Population Estimates. https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?pid=ACS_16_5YR_S0101&prodType=table
- [22].Behavioral Risk Factor Surveillance System: 2016 Summary Data Quality Report. https://www.cdc.gov/brfss/annual_data/2016/pdf/2016-sdqr.pdf
- [23].Kavanaugh ML, Jerman J. Contraceptive method use in the United States: trends and characteristics between 2008, 2012 and 2014. Contraception. 2018. January 1;97(1):14–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [24].Lathrop E, Romero L, Hurst S, Bracero N, Zapata LB, Frey MT, et al. The Zika Contraception Access Network: a feasibility programme to increase access to contraception in Puerto Rico during the 2016–17 Zika virus outbreak. The Lancet Public Health. 2018. February 1;3(2):e91–9. [DOI] [PMC free article] [PubMed] [Google Scholar]