Missed Opportunities for Repeat HIV Testing in Pregnancy: Implications for Elimination of Mother-to-Child Transmission in the United States

Abstract

HIV testing is an effective intervention that is used for reducing perinatal HIV transmission. Centers for Disease Control and Prevention recommends a second HIV test during the third trimester of pregnancy for women in settings with an elevated HIV incidence (≥17 cases per 100,000 person-years). We conducted a retrospective cohort study at a single hospital in Baltimore, Maryland, to determine whether a second HIV test was done and to compare HIV retesting with mandated syphilis retesting. Of women who delivered at this hospital, 98.8% received prenatal care. Descriptive, bivariate, and multivariable analyses were performed. Among 1632 women, mean age was 27.6 years (standard deviation: 6.3), 59.6% were black, and 55.5% were single. HIV retesting was done in 28.4% of women, which was significantly less often compared with the state-mandated syphilis retesting (78.7%, p < 0.001). The odds of having an HIV retest were 15 times higher among women who received prenatal care at a teaching clinic [adjusted odds ratio (aOR): 15.58; 95% confidence interval (CI): 11.12–21.81], and they were lower among women with private insurance (aOR: 0.54, 95% CI: 0.34–0.86). The odds of having a syphilis retest were twice as high among women who received prenatal care at a faculty practice (aOR: 2.17; 95% CI: 1.53–3.09), and they were lower among women with private insurance (aOR: 0.61, 95% CI: 0.43–0.88). Emphasizing an “opt-out” HIV retesting approach through state laws may minimize risk perception, and this is one strategy that can be considered in areas of high HIV incidence to reach the goal of eliminating perinatal HIV transmission in the United States.

Introduction

Perinatal HIV transmission in the United States has been reduced drastically through testing and use of highly active antiretroviral therapy.^1
–3 In 2013, there were 69 infants born with HIV in the United States, which is a 55% decrease since 2008.⁴ This success has ignited discussions about the real possibility of eliminating perinatal HIV transmission. The US National Organizations' Collaborative to Eliminate Perinatal HIV recommended three strategies to sustain the success in prevention of mother-to-child HIV transmission (PMTCT) and to eliminate mother-to-child HIV transmission (EMTCT): (1) institutionalize medical interventions and policy changes that support PMTCT; (2) standardize and solidify HIV screening to support early diagnosis of HIV infections in women; and (3) direct resources to enhance primary prevention of HIV among young women.⁵ In recent years, delayed diagnosis of maternal HIV infection has been associated with the majority of perinatal HIV transmission.^6,7 Among the perinatal HIV transmissions reported to the Centers for Disease Control and Prevention (CDC) during 2008–2012, 23% of maternal HIV infections remained undiagnosed until the intrapartum and postpartum periods,⁸ which was higher than the estimated 13% of people living with HIV who were unaware of their infections in 2013.⁹ In addition, of all live births to HIV-positive mothers in 2010 in the states participating in CDC's enhanced perinatal surveillance system, 1.7% seroconverted during pregnancy, representing a 25% increase annually over a 5 year period.¹⁰

To achieve the goal of EMTCT, the World Health Organization, the American Academy of Pediatrics and American College of Obstetricians and Gynecologists (ACOG) have recommended universal HIV testing during pregnancy by using an opt-out approach where allowed.^11
–13 Women at increased risk of HIV infection, such as those who reside in areas with high HIV incidence, are recommended to undergo HIV retesting in the third trimester of pregnancy before 36 gestational weeks.¹⁴ Further, ACOG recommends third-trimester HIV retesting in women who: are diagnosed with a sexually transmitted infection (STI) within the past year; are injection drug users, or whose sex partners are injection drug users; exchange sex for money or drugs; have a new or more than one sexual partner during the pregnancy; or have a sexual partner who is known to be HIV positive.¹³ To increase compliance with these guidelines, several states passed legislation that requires healthcare providers to offer third-trimester HIV retesting for women at increased HIV risk. However, in states without legislation, repeat HIV testing is based on clinical risk factors or the subjective risk perception by healthcare providers, which has been proved inaccurate.^15
–17

Given that a major step to achieving the goal of EMTCT includes maternal HIV diagnosis, we sought to assess the proportion of women who underwent antepartum HIV retesting in a high HIV incidence geographic area. We examined independent associations between clinical and demographic factors with an antepartum HIV retest. Previous national studies have shown that 96–98% of pregnant women are tested for syphilis at least once during prenatal care, whereas only 57–82% receive any prenatal HIV testing.^18,19 Therefore, we hypothesized that the frequency of prenatal syphilis retesting would be greater than prenatal HIV retesting, and it would likely result from the Maryland state mandate for syphilis retesting.

Methods

We conducted a retrospective cohort study among all women who delivered an infant at the Johns Hopkins Hospital (JHH) in Baltimore, Maryland, from January 1 to December 31, 2012. During this year, among individuals 13 years and older, the HIV incidence was 30 out of 100,000 individuals in Maryland and 72.3 out of 100,000 individuals in the city of Baltimore.²⁰ Maryland had the second highest rate of HIV diagnoses nationwide, and the Baltimore Eligibility Metropolitan Area ranked third.²¹ CDC recommends a second HIV test during the third trimester of pregnancy for women in settings with an elevated HIV incidence (≥17 cases per 100,000 person-years).¹⁴

Women were excluded if they were determined to be HIV positive at the time of the first prenatal visit, delivered before third trimester testing was typically ordered (i.e., before 26–28 gestational weeks), underwent elective termination of pregnancy, had zero prenatal or other hospital visits, or transferred prenatal care from an outside clinic to JHH. Demographic and clinical data were abstracted from the electronic medical record by trained research personnel and were double checked by a second abstractor to minimize data entry errors. Self-identified race/ethnicity was collected. The Johns Hopkins University School of Medicine Institutional Review Board approved the study.

Measures

Three distinct clinical practices provided prenatal care at JHH, including a community-based clinic, a resident (or teaching) clinic, and a faculty clinic. Advanced practice providers and general obstetrician/gynecologists provided services at the community-based clinic, advanced practice providers and postgraduate trainees provided services at the resident clinic, and maternal-fetal medicine specialists provided services at the faculty practice. All of the resident patients and the majority of faculty patients were seen at clinics located within JHH, whereas the community-based clinics were located in five different locations both within the city and in the greater Baltimore community. Some of the providers in the faculty clinic also supervised the residents in the resident clinic. Prenatal laboratory tests were ordered based on ACOG guidelines, which included ordering gonorrhea and chlamydia screening tests at the first prenatal visit, and repeated testing for symptomatic women or high-risk women in the third trimester.^22,23 Syphilis tests were ordered at the first prenatal care visit, in the third trimester, and at delivery for all pregnant women, as mandated by the Maryland Department of Health and Mental Hygiene as a public health measure to decrease congenital syphilis.²⁴

Previous studies have shown an association between STIs and mental illness with HIV acquisition.^25
–27 Therefore, data on a prior STI (e.g., gonorrhea, chlamydia, syphilis, trichomoniasis, genital herpes, hepatitis B, hepatitis C) and mental illness (e.g., bipolar disorder, schizophrenia, depression, anxiety, developmental delay, post-traumatic disorder, attention deficit disorder, oppositional defiant disorder) were abstracted from the electronic medical record. Information on HIV risk behaviors was routinely collected at the first prenatal visit for all patients, including sexual relationship with an HIV-positive man, current or previous alcohol abuse (>3 glasses of alcoholic beverage per night), current tobacco use, and current or previous use of marijuana or other illicit substances, including injection drug use.

The following laboratory tests were performed per ACOG guidelines during the first prenatal care visit or based on clinical indications: Neisseria gonorrhoeae (GC) nucleic acid amplification tests (NAAT); Chlamydia trachomatis NAAT; Treponema pallidum (syphilis treponemal test, which is a chemiluminescence immunoassay, with reflex rapid plasma reagin and fluorescent treponemal antibody test, if needed); Trichomonas vaginalis (microscopy); herpes simplex virus [(HSV) types 1 and 2, culture of genital lesion or serology]; hepatitis viruses B (HBV) and C (HCV) enzyme-linked immunosorbent assays (ELISA); HIV third-generation antibody test (ELISA) or rapid test; and urine toxicology screen [(UTOX), including barbiturate, benzodiazepine, cannabinoid, cocaine, codeine or morphine, methadone, and phencyclidine]. A UTOX was performed at the first prenatal visit and on admission to the labor floor.

The primary outcome was completion of an antepartum HIV retest. Antepartum HIV retest was defined as having at least two HIV tests before admission to the labor floor, with at least one of them performed after 26 0/7 weeks. Based on CDC guidelines, all pregnant women in our cohort should have been offered an HIV retest, given the high incidence of HIV infection in Maryland. Labor floor HIV test was defined as any HIV test that was done within 3 days of delivery (including postpartum). Besides geographic HIV risk factors, we also examined clinical risk factors that should prompt clinicians to repeat an HIV test, such as a new STI diagnosis during pregnancy or a positive UTOX. Completion of intake, repeat antepartum, and delivery screening for syphilis were also evaluated. We selected syphilis testing as a comparison to determine whether the lack of HIV retesting was due to the process of obtaining a laboratory test (i.e., provider ordering, patient compliance with undergoing a blood draw) rather than other factors such as patient or provider risk assessment.

Statistical analyses

We summarized descriptive statistics for the demographic characteristics, obstetric histories, medical comorbidities, and behavioral risk factors stratified by presence or absence of antepartum HIV retesting. We compared normally distributed variables with Student's t-test and skewed variables with Wilcoxon rank-sum test. We also compared frequency distributions of binary and categorical variables with Chi-square or Fisher's exact tests. Variables that showed statistical significance (p < 0.05) in univariable comparisons were entered into forward and backward stepwise selections to identify multiple logistic regression (MLR) models. The results of regression analyses were expressed as unadjusted and adjusted odds ratios (uOR and aOR) and 95% confidence intervals (CIs). We used Hosmer–Lemeshow Goodness-of-Fit (GOF) test and area under the receiver-operator characteristics curves (C-statistics) to assess model fit and used variance inflation factor (VIF) to assess statistical collinearity. We used model-wise deletion for missing data in the regression analyses. Reported p-values were based on two-sided tests, and p < 0.05 was considered statistically significant. Data analyses were performed with Stata 13.1 (StataCorp LP, College Station, TX).

Results

During 2012, 1988 women delivered at JHH and the majority of them had received prenatal care (1964/1988, 98.8%). After excluding women with elective terminations of pregnancy (n = 14), pre-existing HIV infection (n = 36), delivery before 26 0/7 gestational weeks (n = 25), no prenatal care (defined as no prenatal or other hospital visits, n = 24), and transfer of care (n = 257), 1632 women remained. All 1632 charts were maintained electronically, and all were reviewed. Ten (0.6%) charts were missing maternal education level, 25 (1.5%) charts were missing insurance coverage, and one chart was missing the date of STI diagnosis. Otherwise, the data were complete.

The mean [standard deviation (SD)] age of the cohort was 27.6 (6.3) years. Among the cohort, 59.6% were black, and most women were single, never married (55.5%). The community practice delivered slightly more patients (38.5%) compared with the resident practice (26.4%) and the faculty practice (35.0%). Patients with private insurance were more likely to receive prenatal care at the faculty practice compared with the community practice and the resident practice (63.8%, 31.7%, and 4.5%, respectively, p < 0.001). Additional demographics are shown in Table 1.

Table 1.

Comparisons of the Demographic and Clinical Characteristics By Completion of HIV Retest

Variable ^a	HIV retest completed	HIV retest not completed	All	Unadjusted OR (95% CI) ^b	p	Adjusted OR (95% CI) ^c	p
Total	464 (28.4)	1168 (71.6)	1632 (100)
Age, year, mean (SD)^b	25.3 (5.7)	28.5 (6.2)	27.6 (6.3)	0.92 (0.90–0.93)	<0.001	1.00 (0.96–1.03)	0.83
Gravidity, median (IQR)^d	2 (1–3)	1 (0–3)	1 (0–3)	1.12 (1.06–1.17)	<0.001	0.96 (0.89–1.05)	0.36
Race^e					<0.001
Black	379 (81.7)	594 (50.9)	973 (59.6)	Reference		Reference
White	63 (13.6)	444 (38.0)	507 (31.1)	0.22 (0.17–0.30)		0.80 (0.49–1.31)	0.38
Hispanic	7 (1.5)	15 (1.3)	22 (1.3)	0.73 (0.30–1.81)		0.79 (0.22–2.83)	0.72
Other (mainly Asian)	15 (3.2)	115 (9.8)	130 (8.0)	0.20 (0.12–0.36)		0.82 (0.37–1.83)	0.63
Practice group					<0.001
Community	99 (21.3)	530 (45.4)	629 (38.5)	Reference		Reference
Resident	330 (71.1)	101 (8.6)	431 (26.4)	17.49 (12.84–23.83)		15.58 (11.12–21.81)	<0.001
Faculty	35 (7.5)	537 (46.0)	572 (35.0)	0.35 (0.23–0.52)		0.69 (0.42–1.12)	0.13
Marital status					<0.001
Single (never married)	368 (79.3)	537 (46.0)	905 (55.5)	Reference		Reference
Married	82 (17.7)	617 (52.8)	699 (42.8)	0.19 (0.15–0.25)		0.75 (0.47–1.21)	0.24
Separated/divorced/widowed	14 (3.0)	14 (1.2)	28 (1.7)	1.46 (0.69–3.10)		0.98 (0.34–2.84)	0.97
Education					<0.001
<High school	138 (29.7)	178 (15.2)	316 (19.4)	Reference		Reference
High school graduate	197 (42.5)	268 (22.9)	465 (28.4)	0.95 (0.71–1.27)		1.31 (0.89–1.93)	0.17
<4 years in college	86 (18.5)	182 (15.6)	268 (16.4)	0.61 (0.43–0.86)		1.26 (0.77–2.05)	0.35
Bachelor or graduate degree	40 (8.6)	534 (45.7)	574 (35.2)	0.10 (0.07–0.14)		0.78 (0.38–1.59)	0.5
Private insurance coverage	59 (12.7)	670 (57.4)	729 (44.7)	0.10 (0.08–0.14)	<0.001	0.54 (0.34–0.86)	0.009
History of mental illness	96 (20.7)	165 (14.1)	261 (16.0)	1.59 (1.20–2.10)	0.001	1.17 (0.77–1.79)	0.46
Ever used illicit substance	108 (23.3)	135 (11.6)	243 (16.0)	2.02 (1.57–2.60)	<0.001	0.92 (0.63–1.35)	0.66
Positive UTOX	91 (19.6)	116 (9.9)	207 (12.7)	2.21 (1.64–2.98)	<0.001
Current or previous alcohol abuse (>3 glasses/night)^e	8 (1.7)	7 (0.6)	15 (0.9)	2.91 (1.05–8.07)	0.04	1.67 (0.41–6.76)	0.47
Current tobacco use	71 (15.3)	85 (7.3)	156 (9.6)	2.30 (1.65–3.22)	<0.001	1.25 (0.75–2.08)	0.4
Delivered before 37 0/7 weeks	41 (8.8)	115 (9.8)	156 (9.6)	0.89 (0.61–1.29)	0.53
Prior or prevalent STI	241 (51.9)	370 (31.7)	611 (37.4)	2.33 (1.87–2.90)	<0.001
Incident STI	47 (10.1)	38 (3.3)	85 (5.2)	3.35 (2.15–5.21)	<0.001
Any STI ever	265 (57.1)	379 (32.4)	644 (39.5)	2.77 (2.22–3.46)	<0.001	1.39 (1.00–1.93)	0.05
Registered to prenatal care <13 0/7 weeks	256 (55.2)	753 (64.5)	1009 (61.8)	0.68 (0.55–0.84)	0.001	1.16 (0.85–1.58)	0.35
Registered to prenatal care ≥20 0/7 weeks	103 (22.2)	238 (20.4)	341 (20.9)	1.11 (0.86–1.45)	0.41
≤5 prenatal visits	53 (11.2)	150 (12.8)	202 (12.4)	1.17 (0.83–1.63)	0.36
Syphilis tested at intake, repeated third trimester	415 (89.4)	870 (74.5)	1285 (78.4)	2.90 (2.10–4.01)	<0.001

Data are presented as n (%), unless otherwise specified. Data were missing for some study participants on education (9, 0.6%), private insurance coverage (23, 1.4%), or date of STI diagnosis (1, 0%). Percentages for variables with missingness would not add up to 1. All the other data are complete.

Adjusted odds ratios shown only for those variables that were included as predictors in the final multivariable logistic regression model.

Compared with Student's t-test.

In the multivariable logistic regression model, 1600 observations were retained, and 32 were dropped due to missing data. p for Hosmer–Lemeshow Goodness-of-Fit test (10 groups) = 0.18; C-statistic = 0.88; mean VIF = 1.79. None of the covariates had VIF larger than 10.

Compared with Wilcoxon rank-sum test.

Compared with Fisher's exact test, all the other binary or categorical variables were compared with Chi-squared test.

CI, confidence interval; IQR, interquartile range; OR, odds ratio; SD, standard deviation; STI, sexually transmitted infection; UTOX, urinary toxicology; VIF, variance inflation factor; gravidity refers to the total number of prior pregnancies; parity refers to the total number of prior term and preterm deliveries.

Missed opportunities for antepartum HIV retesting based on HIV incidence

An antepartum HIV retest was done in 464 (28.4%) women. Of those who did not have a retest (n = 1168), 69 (5.9%) were tested multiple times yet did not meet the retesting definition (e.g., did not have a third-trimester HIV test), 62 (5.3%) were tested for HIV on the labor floor, and 19 (1.6%) were never tested for HIV either antepartum or on admission to the labor floor. The majority of women (n = 1613, 98.4%) had at least one HIV test during pregnancy or postpartum. A labor floor HIV test was performed on 86 (5.3%) women and was the only HIV test for 19 of these women. Among the 1291 women who registered for prenatal care before 20 0/7 gestational weeks, and hence had several opportunities for testing, 361 (28.0%) underwent antepartum HIV retesting. None of the women in our cohort who were retested (n = 464) seroconverted during pregnancy.

More than 91% of the women who missed an opportunity for HIV retesting received prenatal care at either the community-based or faculty practices (p < 0.001). Univariable analyses showed that women who missed the antepartum HIV retest were more likely to be older (mean age [SD] 28.5 [6.2] years vs. 25.3 [5.7] years, p < 0.001), married (52.8% vs. 17.7%, p < 0.001), and white or Asian (47.8% vs. 16.8%, p < 0.001), and to hold bachelor or postgraduate degrees (45.7% vs. 8.6%, p < 0.001), compared with women who completed HIV retesting. In an MLR model adjusting for age, practice group, gravidity, race, marital status, education, insurance status, mental illness history, current tobacco smoking, current alcohol abuse, illicit drug use ever, any STI ever, and registry to prenatal care before 13 0/7 weeks, the odds of having an HIV retest were 15 times higher among women who received prenatal care at the resident practice (aOR: 15.58; 95% CI: 11.12–21.81, p < 0.001) (Table 1). Model fit was supported by GOF test (p = 0.18) and VIF (1.79). The C-statistics for this MLR model was 0.88, which means that one can correctly predict whether a participant in this sample would complete an antepartum HIV retest 88% of the time.

Missed opportunities for antepartum retesting based on clinical risk factors

In total, 164 (10.1%) women were diagnosed with an STI during the pregnancy. The most common STI in this cohort was chlamydia (109/1632, 6.7%), followed by trichomoniasis (35/1632, 2.1%) and gonorrhea (26/1632, 1.6%). Syphilis (n = 6), genital herpes (n = 13), hepatitis B (n = 2), and hepatitis C (n = 2) were diagnosed in less than 1% of the cohort. Of the 163 women for whom the dates of STI diagnoses were available, 161 (98.2%) tested positive during prenatal care, 2 (1.2%) tested positive on admission to the labor floor, and 1 (0.6%) tested positive for two different STIs at both time points. A positive UTOX was documented among 207 (12.7%) women, and 194 (93.7%) of them occurred during prenatal care. Thirty-two women tested positive for both STIs and UTOX before delivery. Collectively, an antepartum HIV retest was indicated in 323 (19.8%) women out of the entire cohort because of clinical risk factors (i.e., STI or UTOX). Of these, 160 (49.5%) women completed an antepartum HIV retest, 178 (55.1%) women had at least a third-trimester HIV test, and 46 (14.2%) women underwent labor floor HIV testing. Most of the clinical risk factors were identified during the antepartum course. Hence, if an HIV retest was routinely ordered at 32 0/7 or 35 0/7 weeks, 91.0% (294/323) and 91.3% (295/323), respectively, of these high-risk women (who were still pregnant) would have been retested.

Comparison to syphilis testing

There were 1285 (78.7%) women who had both an intake and a third-trimester syphilis test, which was significantly higher than the HIV retest completion (p < 0.001). More than 75% of women (n = 1239) completed all three syphilis tests, and 1604 (98.3%) had a labor floor syphilis test. All of the women in the cohort had at least one syphilis test during pregnancy or postpartum. In an MLR model adjusting for practice group, race, marital status, insurance status, any STI ever, and registry to prenatal care before 13 0/7 weeks (data not shown), the odds of having a syphilis retest were higher among women who received prenatal care at the faculty practice (aOR: 2.17; 95% CI: 1.53–3.09, p < 0.001), were Asian or white (aOR: 2.12; 95% CI: 1.13–4.00, p = 0.02 and aOR: 1.46; 95% CI: 1.01–2.12, p = 0.04, respectively), or began prenatal care in the first trimester (aOR: 2.45; 95% CI: 1.88–3.18, p < 0.001). The odds of a syphilis retest were lower among women who had private insurance (aOR: 0.61; 95% CI: 0.43–0.88, p = 0.008). There were 6 (0.36%) women who were diagnosed with and treated for syphilis. Five were diagnosed based on the first prenatal test results, whereas one woman seroconverted and was detected at 29 2/7 gestational weeks.

Discussion

Early diagnosis of maternal HIV infection is the first and crucial step for timely initiation of antiretroviral treatment for women who seroconvert during pregnancy. Timely identification of HIV infection during pregnancy also provides the opportunity to address medical, psychological, and social issues that may compromise adherence to antepartum and postpartum care.²⁸ Previous studies have shown increased risk of heterosexual HIV transmission during pregnancy, and acute HIV during pregnancy has been associated with higher risk of MTCT.^29,30 Among our cohort, only 28% of pregnant women underwent antepartum HIV retesting, which was surprising given the high incidence of HIV in Maryland and a high frequency of clinical risk factors. Our MLR model, which reflects patients' demographic characteristics, socioeconomic status, and HIV risk behaviors, was remarkably accurate in identifying predictors of antepartum HIV retesting. Women who received prenatal care delivered by the faculty practice, who mostly had private insurance, were more likely to lack antepartum HIV retests. There were other significant patient sociodemographic differences between clinics, and providers from clinics located in the same hospital provided care at both the faculty and resident clinics. However, prenatal HIV retesting uptake at these clinics differed. These findings suggest that providers' and patients' risk perceptions may be a major reason for inadequate repeat HIV testing.

The poor predictive value of a provider's perception of HIV risk has been previously reported,^15
–17 and prenatal care providers cannot rely on their patients' own risk perception either. In the CDC-sponsored Mother-Infant Rapid Intervention at Delivery study conducted in six US cities (Atlanta, Baton Rouge, Chicago, Miami, New Orleans, and New York City) between 2001 and 2005, all women who seroconverted during pregnancy denied new sexual partners, alcohol or illicit drug use.³¹ Hence, universal opt-out antepartum HIV retesting for all pregnant women living in areas of high HIV incidence, regardless of perceived risks, should continue to be recommended and efforts should be made to operationalize the approach. Lessons can be learned from resource poor settings, where MTCT represents a much larger issue. In Kenya, where there are widely accessible and free PMTCT services, failure to provide HIV education (OR: 3.57 95%; CI: 1.36–9.33) and HIV counseling (OR: 3.95; 95% CI: 1.28–12.21) at health facilities was associated with greater odds of perinatal transmission.³² Hence, providers' adherence to recommendations and guidelines played a crucial role in PMTCT. In the United States, therefore, efforts should be made to improve providers' adherence to our national HIV retesting recommendations.

In contrast to the poor completion of antepartum HIV retesting, the prenatal syphilis testing protocol was more consistently followed, especially in the faculty practice. Part of the reason for greater syphilis testing might be because patients in the faculty practice tended to have earlier registration into prenatal care, more visits, and opportunities to perform testing. However, the differences between clinics remained even after controlling for early registration into prenatal care, suggesting that the state mandate for syphilis retesting could be influencing uptake. Because the providers at the faculty practices retested their patients for syphilis quite frequently, this suggests that lack of the HIV retest is not simply a process issue, but rather involves an HIV risk assessment by the provider and/or patient. Further, private insurance coverage was associated with both missed HIV and syphilis testing, which might imply that women with private insurance coverage were believed to be at lower risk for either HIV or syphilis infections. This represents an opportunity for private health insurers to take adequate steps to encourage that HIV and syphilis retesting are offered to their members who live in high-incidence areas. Prior studies have shown that providers' persistence in offering prenatal HIV testing, regardless of perceived risk, is associated with universal test uptake.¹⁷ We thereby postulate that a state mandate for HIV retesting may motivate prenatal care providers to recommend third-trimester HIV retesting more consistently, which may, in turn, improve HIV retest uptake.

This study has many strengths. The higher frequency of HIV risk factors in this cohort, including STIs and substance use, has provided a better chance to examine antepartum HIV retesting when clearly clinically indicated. We illustrated by regression analyses the differential mechanisms leading to antepartum HIV retesting and syphilis testing. There are also limitations. First, information bias was likely due to the retrospective nature of this study. We were unable to accurately estimate the percentage of women who were offered but declined HIV retesting. Our prenatal charts easily allow providers to document patients who decline testing, but this was noted in very few patient charts. Regardless, instead of selective risk-based offering of antepartum retesting, we support universal retesting with an opt-out approach, which has been associated with a lower decline rate.^33
–35 Second, the study results are from three provider practices associated with a major academic center that are located in HIV high-incidence areas; hence, our findings may not be generalizable to other clinical settings or regions. However, antepartum HIV retesting was still missed, which raises concern for other regions with lower HIV incidence where providers may be less aware of the indications to retest for HIV. Third, because of the low percentage of patients with antepartum HIV retests, we could not estimate seroconversions. Therefore, we were not able to speculate on cost-effectiveness of antepartum HIV retesting. Nonetheless, HIV retesting in high HIV cost-effectiveness has been shown in other studies.³⁶

Since the debut of the revised CDC recommendations for prenatal HIV screening 10 years ago, baseline HIV screening has been widely implemented as part of routine prenatal care in the United States.³⁷ This is further supported by our data, which showed that a majority of women (>98%) had at least one HIV test performed. However, it will be impossible to capture high-risk women who seroconvert with one-time screening; and to eliminate perinatal HIV transmission, opt-out universal antepartum retesting is needed. Given the possible increased sexual HIV acquisition risk during pregnancy and the high perinatal transmission risk during HIV seroconversion in pregnancy,^29,30 there needs to be a stronger emphasis on retesting. Moreover, syphilis and HIV seroconversion in pregnancy should be equally valued. Increasing congenital syphilis rates resulted in an impressive public health effort to minimize the preventable illness, and legislation was drafted to support the efforts. Thus, the different test completion rates between HIV and syphilis suggest that legislation could improve antepartum HIV retesting. A recent law was passed in Maryland that required prenatal providers to offer repeat HIV testing in the third trimester for all pregnant women.³⁸ However, it will be years before we know whether the legislation increased HIV retesting. Regardless, emphasizing HIV retesting in state laws by using the “opt-out” approach can minimize the dependence on providers' and patients' risk perception, and this will be a pivotal and strategic step to eliminate perinatal HIV transmission in the United States.

Footnotes

Acknowledgments

The authors would like to thank Amy Lee, DNP, ARNP, WHNP-BC, Maybel Wahab, MSN CNM, and Lynn McDonald, DNP, RN for their assistance with data collection. They would also like to thank all patients and obstetric providers at Johns Hopkins Hospital.

Author Disclosure Statement

No competing financial interests exist.

References

Connor

, Sperling

, Gelber

, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. Pediatric AIDS Clinical Trials Group Protocol 076 Study Group. N Engl J Med, 1994; 331:1173–1180.

Fowler

, Gable

, Lampe

, Etima

, Owor

. Perinatal HIV and its prevention: Progress toward an HIV-free generation. Clin Perinatol, 2010; 37:699–719, vii.

Centers for Disease Control and Prevention. Twenty-five years of HIV/AIDS—United States, 1981–2006. MMWR Morb Mortal Wkly Rep, 2006; 55:585–589.

Prevention CfDCa. Monitoring selected national HIV prevention and care objectives by using HIV surveillance data—United States and 6 dependent areas, 2014. HIV Surveill Suppl Rep, 2016; 21:47.

Burr

, Lampe

, Corle

, Margolin

, Abresh

, Clark

. An end to perinatal HIV: Success in the US requires ongoing and innovative efforts that should expand globally. J Public Health Policy, 2007; 28:249–260.

Nesheim

, Taylor

, Lampe

, et al. A framework for elimination of perinatal transmission of HIV in the United States. Pediatrics, 2012; 130:738–744.

Nesheim

, Harris

, Lampe

. Elimination of perinatal HIV infection in the USA and other high-income countries: Achievements and challenges. Curr Opin HIV AIDS, 2013; 8:447–456.

Centers for Disease Control and Prevention.. Pediatric HIV Surveillance (through 2014). Available at: www.cdc.gov/hiv/library/slideSets/index.html (Last accessed October 9, 2016 ).

Centers for Disease Control and Prevention.. Selected National HIV Prevention and Care Outcomes. July 2016. Available at: www.cdc.gov/hiv/pdf/library/factsheets/cdc-hiv-national-hiv-care-outcomes.pdf (Last accessed October 9, 2016 ).

10.

Singh

, Lampe

, Surendera Babu

, Rao

, Borkowf

, Nesheim

. HIV seroconversion during pregnancy and mother-to-child HIV transmission: Data from the Enhanced Perinatal Surveillance Project, United States, 2005–2010. Paper presented at: 19th International AIDS Conference 2012; Washington DC, USA.

11.

UNAIDS Global Reference Group on HIV/AIDS and Human Rights.. UNAIDS/WHO Policy Statement on HIV testing. 2004. Available at: http://www.who.int/hiv/pub/vct/en/hivtestingpolicy04.pdf (Last accessed November 10, 2016 ).

12.

American Academy of Pediatrics and the American College of Obstetricians and Gynecologists. Human immunodeficiency virus screening. Joint statement of the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists. Pediatrics, 1999; 104(1 Pt 1):128.

13.

Committee opinion no: 635: Prenatal and perinatal human immunodeficiency virus testing: Expanded recommendations. Obstet Gynecol, 2015; 125:1544–1547.

14.

Branson

, Handsfield

, Lampe

, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR. MMWR Recomm Rep, 2006; 55(RR-14):1–17; quiz CE11-14.

15.

Hightow-Weidman

, Golin

, Green

, Shaw

, MacDonald

, Leone

. Identifying people with acute HIV infection: Demographic features, risk factors, and use of health care among individuals with AHI in North Carolina. AIDS Behav, 2009; 13:1075–1083.

16.

Lopez-Quintero

, Rojas

, Dillon

, Varga

, De La Rosa

. HIV testing practices among Latina women at risk of getting infected: A five-year follow-up of a community sample in South Florida. AIDS Care, 2016; 28:137–146.

17.

Anderson

, Koenig

, Lampe

, Wright

, Leiss

, Saul

. Achieving universal HIV screening in prenatal care in the United States: Provider persistence pays off. AIDS Patient Care STDS, 2005; 19:247–252.

18.

Schrag

, Arnold

, Mohle-Boetani

, et al. Prenatal screening for infectious diseases and opportunities for prevention. Obstet Gynecol, 2003; 102:753–760.

19.

Ross

, Tao

, Patton

, Hoover

. Screening for human immunodeficiency virus and other sexually transmitted diseases among U.S. women with prenatal care. Obstet Gynecol, 2015; 125:1211–1216.

20.

Maryland Department of Health and Mental Hygiene.. Maryland HIV/AIDS Quarterly Update, Third Quarter 2014. Center for HIV Surveillance, Epidemiology, and Evaluation, Maryland Department of Health and Mental Hygiene. Baltimore, Maryland, 2014.

21.

Flynn

, Maryland HIV Planning Group. HIV in Maryland. Center for HIV Surveillance, Epidemiology, and Evaluation. Maryland Department of Health and Mental Hygiene. Baltimore, Maryland, 2015.

22.

Workowski

, Berman

. Sexually transmitted diseases treatment guidelines, 2010. MMWR Recomm Rep, 2010; 59:1–110.

23.

American Academy of Pediatrics, American College of Obstetricians and Gynecologists. Guidelines for prenatal care. Seventh ed. In: Riley

, Stark

, eds. Washington, DC: American Academy of Pediatrics, American College of Obstetricians and Gynecologists, 2012, p. 99.

24.

Maryland Department of Health and Mental Hygiene.. Provider Alert: Increasing Rates of Syphilis in Maryland. Infectious Disease and Environmental Health Administration, Maryland Department of Health and Mental Hygiene, Baltimore, Maryland, 2012.

25.

Siriwasin

, Shaffer

, Roongpisuthipong

, et al. HIV prevalence, risk, and partner serodiscordance among pregnant women in Bangkok. Bangkok Collaborative Perinatal HIV Transmission Study Group. JAMA, 1998; 280:49–54.

26.

Van Tieu

, Koblin

. HIV, alcohol, and noninjection drug use. Curr Opin HIV AIDS, 2009; 4:314–318.

27.

Wand

, Ramjee

. Combined impact of sexual risk behaviors for HIV seroconversion among women in Durban, South Africa: Implications for prevention policy and planning. AIDS Behav, 2011; 15:479–486.

28.

Buchberg

, Fletcher

, Vidrine

, et al. A mixed-methods approach to understanding barriers to postpartum retention in care among low-income, HIV-infected women. AIDS Patient Care STDS, 2015; 29:126–132.

29.

Gray

, Li

, Kigozi

, et al. Increased risk of incident HIV during pregnancy in Rakai, Uganda: A prospective study. Lancet, 2005; 366:1182–1188.

30.

Birkhead

, Pulver

, Warren

, Hackel

, Rodriguez

, Smith

. Acquiring human immunodeficiency virus during pregnancy and mother-to-child transmission in New York: 2002–2006. Obstet Gynecol, 2010; 115:1247–1255.

31.

Nesheim

, Jamieson

, Danner

, et al. Primary human immunodeficiency virus infection during pregnancy detected by repeat testing. Am J Obstet Gynecol, 2007; 197:149.e141–e145.

32.

Onono

, Owuor

, Turan

, Bukusi

, Gray

, Cohen

. The role of maternal, health system, and psychosocial factors in prevention of mother-to-child transmission failure in the era of programmatic scale up in western Kenya: A case control study. AIDS Patient Care STDS, 2015; 29:204–211.

33.

Williams

, Costello

, McHugh

, et al. Repeat antenatal HIV testing in the third trimester: A study of feasibility and maternal uptake rates. HIV Med, 2014; 15:362–366.

34.

Coleman

, Morgan

, Carlson

, Hawks

, Schulkin

. Patient perceptions of obstetrician-gynecologists' practices related to HIV testing. Matern Child Health J, 2009; 13:355–363.

35.

Wettstein

, Mugglin

, Egger

, et al. Missed opportunities to prevent mother-to-child-transmission: Systematic review and meta-analysis. AIDS, 2012; 26:2361–2373.

36.

Sansom

, Jamieson

, Farnham

, Bulterys

, Fowler

. Human immunodeficiency virus retesting during pregnancy: Costs and effectiveness in preventing perinatal transmission. Obstet Gynecol, 2003; 102:782–790.

37.

Fitz Harris

, Taylor

, Zhang

, et al. Factors associated with human immunodeficiency virus screening of women during pregnancy, labor and delivery, United States, 2005–2006. Matern Child Health J, 2014; 18:648–656.

38.

General Assembly of Maryland. HB0180 Public Health–HIV Testing during Pregnancy.. 2016. Available at: http://mgaleg.maryland.gov/2016RS/bills/hb/hb0180T.pdf (Last accessed November 10, 2016 ).