Rapid human immunodeficiency virus testing in obstetric outpatient settings: the MIRIAD study

Article Outline

• Abstract
• Materials and Methods
• Results
• Comment
• Acknowledgments
• References
• Copyright

Objective

To evaluate the acceptability and feasibility of rapid human immunodeficiency virus testing in obstetric outpatient settings.

Study Design

The Mother-Infant Rapid Intervention at Delivery (MIRIAD) study was a prospective, multicenter study. Women were offered rapid and conventional human immunodeficiency virus testing if they presented to outpatient settings late in pregnancy with undocumented human immunodeficiency virus status. We compared median times between conventional and rapid testing and between rapid point-of-care and rapid laboratory-based testing.

Results

Among eligible women who were offered participation, 90% accepted testing. The median time from blood draw to result available was faster for rapid testing (25 minutes) than conventional testing (23 hours; P < .0001). For rapid tests, point-of-care testing was faster than laboratory-based testing (24 minutes vs 35 minutes; P < .0001). Almost 96% of rapid test results were available within 1 hour.

Conclusion

Rapid human immunodeficiency virus testing is acceptable, feasible, and provides results far sooner than conventional testing in obstetric outpatient settings.

Key words: human immunodeficiency virus, outpatient settings, rapid testing

Human immunodeficiency virus (HIV) infection is increasingly affecting women. More than one-fourth of newly diagnosed cases of HIV in the United States occur among women, who represented an estimated 15,000 new cases in 2006.¹, ², ³ Approximately 25% of HIV-infected individuals in the United States are unaware of their status, thereby significantly reducing opportunities for intervention.⁴ Recently revised guidelines from the American College of Obstetricians and Gynecologists (ACOG)⁵ recommend routine HIV screening for patients in all health care settings and are consistent with guidelines from the US Centers for Disease Control and Prevention (CDC).⁶ Because obstetrician-gynecologists are often the sole contact women have with the health care system, they have a unique opportunity to promote HIV screening. ACOG suggests that rapid testing may be particularly useful in outpatient obstetric-gynecologic settings⁷ by providing more timely results and reducing the resources needed to follow up with patients and inform them of results.⁵ Despite the potential use of rapid testing in advancing universal HIV screening, many clinics still do not offer rapid testing.⁸ Although rapid testing has been described in other outpatient settings, there has been little published regarding the use of rapid testing in outpatient obstetric settings.

The Mother-Infant Rapid Intervention at Delivery (MIRIAD) study was a prospective, multicenter study funded by CDC to evaluate rapid HIV testing of pregnant women in inpatient and outpatient settings. As previously reported, the primary component of the study focused on women who presented to labor and delivery units with undocumented HIV status.⁹, ¹⁰ In this article, we describe women who presented late to prenatal care, and we compare rapid HIV testing with conventional HIV testing. We hypothesize that rapid testing will be acceptable and feasible and provide results faster than conventional testing in outpatient obstetric settings. These results may be helpful to obstetrician-gynecologists as they work to implement ACOG's new HIV screening guidelines.⁵

Back to Article Outline

Materials and Methods 

The MIRIAD study was conducted at 17 hospitals in 6 cities (Atlanta, GA; Baton Rouge, LA; Chicago, IL; Miami, FL; New Orleans, LA; and New York, NY). Full study methods have been described in prior publications.⁹, ¹⁰ The study offered voluntary, rapid HIV testing to women in late pregnancy whose current HIV status was undocumented. Previous papers have reported results from women who presented in labor⁹, ¹⁰; in this study, we report results from 285 women tested in obstetric outpatient settings.

Six prenatal clinic sites, 1 from each of the cities listed previously, participated in the clinic component of the study from April 2002-January 2005. All clinics were affiliated with a hospital and located on that hospital's premises. All women presenting to these prenatal clinics were screened by clinic staff for MIRIAD eligibility. Women were considered eligible if they were at least 34 weeks' gestation, were not in labor and not possessing an indication for delivery, and their HIV status was undocumented during the current pregnancy. Labor was defined as regular, painful contractions with cervical dilation. Documentation of HIV status required a written, electronic, or verbal report by a health care provider or documentation of previous antiretroviral therapy. An initial MIRIAD requirement that study participants must have had no prenatal care visits before enrollment was removed in September 2003.

Three clinic sites (Atlanta, Baton Rouge, and New Orleans) recorded information on all women who presented to the clinic, including the date and time of their presentation, the time when eligible women were offered participation in MIRIAD, the time when the woman accepted or declined participation in MIRIAD, the reasons for declining or not being offered MIRIAD, and selected demographic and delivery information. The other 3 clinic sites collected this information only for women who enrolled in the study. At most of the clinics, when an eligible woman arrived at the clinic, clinic staff notified MIRIAD study personnel, who immediately came to the clinic and initiated an informed consent process. Written informed consent was obtained from all study participants. The study protocol was approved by the institutional review boards at CDC and all participating clinics. A prestudy sample size calculation was not performed.

All study participants underwent parallel testing, with both rapid and conventional HIV tests. In most cases, a blood specimen was collected via venipuncture, and staff performed both tests on this specimen. In 50 cases, fingerstick specimens were used for the rapid test. Rapid testing was performed with the use of the OraQuick rapid HIV-1 antibody test (OraSure Technologies Inc, Bethlehem, PA). The US Food and Drug Administration (FDA) formally approved this test in November 2002; before this, the FDA allowed its use by the MIRIAD study under an investigational device exemption. The OraQuick test has a Clinical Laboratory Improvement Amendment (CLIA) waiver for testing outside of the laboratory setting.¹¹ At 4 of the clinic sites, rapid tests were performed in the clinic by trained MIRIAD staff (point-of-care testing). At the 2 other sites, rapid tests were performed in their laboratories by trained laboratory staff (laboratory-based testing).

Conventional testing consisted of an enzyme immunoassay (EIA) with Western blot (WB) confirmation. Three clinics used the Abbott HIV-1/HIV-2 EIA (Abbott Laboratories, Abbott Park, IL), 2 clinics used the Genetic Systems HIV-1/HIV-2 peptide EIA (BioRad Laboratories, Hercules, CA), and 1 clinic used the bioMerieux Vironostika HIV-1 enzyme-linked immunosorbent assay (ELISA) kit (BioMerieux, Durham, NC).

Times were recorded for each step of the testing process, including the time of blood draw, the time results were available to the provider, and the time the woman was informed of her result. The time when the test was completed was not recorded, so the duration of the testing process was calculated as the time from blood draw to the time when the result was obtained by the clinic providers. Staff was trained to manually record all times on data collection forms. Women were excluded for unrealistic durations of time from blood draw to result availability, defined as less than 20 minutes for a rapid test and less than 3.5 hours for a conventional test.

The date that conventional test results were available was recorded for all women included in the analysis; however, for 150 (52.6%) women, the exact time that conventional test results were available was not recorded. If test results with no recorded time were available on the same day that blood was drawn (n = 17), we assumed that 3.5 hours elapsed between the 2 events (the minimum time for conventional testing). If the blood draw and the availability of test results occurred on different days (n = 133), we assumed that the test result was available to the provider at 9:00 am. These imputed times are the most conservative estimates of duration of conventional testing. Calculations were also performed excluding these 150 women.

SAS statistical software version 9.1 (SAS Institute, Cary, NC) was used for all statistical analyses. Median time intervals were calculated for each step of the testing process. The Wilcoxon rank-sum test was used to compare the median time interval for conventional tests with that for rapid tests and for rapid point-of-care tests with that for rapid laboratory-based tests.

Back to Article Outline

Results 

There were a total of 285 women at the 6 clinics who had conventional and rapid test results available: 229 (80%) of these women attended the 3 clinics that recorded demographic information on all women who presented for care (Atlanta, Baton Rouge, and New Orleans), and 56 women (20%) attended the 3 clinics that recorded information only for those enrolled (Chicago, Miami, and New York). During the study period, there were 12,886 women who presented to the 3 clinics that recorded information on all women (Figure). Of these, 364 women (2.8%) were deemed eligible for the MIRIAD study. Approximately 73% of eligible women were asked to participate in the MIRIAD study. The most frequently reported reasons for not asking eligible women to participate in the MIRIAD study was that they were overlooked for study participation or the MIRIAD staff was not notified of their eligibility (n = 24). Other reasons cited were women being unable to wait or having left without being seen (n = 16), the staff's inability to locate them (n = 9), the staff being too busy or unavailable (n = 6), an interpreter not being available (n = 3), and other reasons (n = 9). No reason was given for 31 women.

• 
  • View Large Image
  • Download to PowerPoint

• FIGURE. 

  Women who presented to prenatal care clinics in Atlanta, Baton Rouge, or New Orleans: MIRIAD Study, 2002-2005

• ^aEligibility criteria: gestational age ≥ 34 weeks, not in labor, and undocumented HIV status during the current pregnancy.

• HIV, human immunodeficiency virus; MIRIAD, Mother-Infant Rapid Intervention at Delivery.

• Tepper. Rapid HIV testing in obstetric outpatient settings. Am J Obstet Gynecol 2009.

Of the 266 women asked to participate at these 3 clinics, 240 (90%) accepted enrollment and testing (Figure). Among the 26 women who declined participation, 10 reported previous testing (4 during the current pregnancy, 1 before the current pregnancy, and 5 had missing data on when testing occurred). Other reasons for declining participation included: the woman did not want to know her status (n = 4); refused to be part of research (n = 4); perceived herself not at risk for HIV (n = 2); or was in too much pain (n = 1). The reason for declining participation was unknown for 4 women. Some women gave more than 1 reason for declining participation. Of these 240 women, 239 received rapid and conventional HIV tests (99.6%), and both results were available for 229 (95.8%).

Among all 6 clinics that participated in the study, there were a total of 307 eligible women who underwent testing. Women were excluded for unrealistic durations of time from blood draw to result availability (n = 15 or 4.9%), unknown date or time of rapid test results (n = 3 or 1.0%), and times recorded that were incompatible with clinic hours (n = 4 or 1.3%). Therefore, 285 (92.8%) women were included in the final analyses. The characteristics of these women are shown in Table 1. Nearly 60% were under 25 years of age, 76% were black, 78% had 12 years or less of education, and 80% had no prenatal care visits before the day of study enrollment.

TABLE 1. Characteristics of 285 women who presented late to prenatal care with undocumented HIV status and received rapid and conventional HIV testing: MIRIAD study; all 6 clinics, 2002-2005

Characteristic	Number	Percent
Age (y)
< 20	65	23.5
20-24	95	34.3
25-29	51	18.4
≥ 30	66	23.8
Unknown/missing	8
Race/ethnicity
Non-Hispanic white	15	5.4
Non-Hispanic black	211	75.9
Hispanic	40	14.4
Other	12	4.3
Unknown/missing	7
Years of education
≤ 11	86	42.6
12	72	35.6
> 12	44	21.8
Unknown/missing	82
Number of prenatal care visitsa
0	200	80.0
1-5	35	14.0
>5	15	6.0
Unknown/missing	35
Gestational age (wk)
34-36	167	60.1
>36	111	39.9
Unknown/missingb	7
Location of rapid testing
Point-of-carec	204	71.6
Laboratory-basedd	81	28.4

HIV, human immunodeficiency virus; MIRIAD, Mother-Infant Rapid Intervention at Delivery.

Tepper. Rapid HIV testing in obstetric outpatient settings. Am J Obstet Gynecol 2009.

As shown in Table 2, for the conventional HIV tests, the median time between blood draw and result available to provider was 23 hours (range, 3.5 hours to 43 days). For the rapid HIV tests, the overall median time between blood draw and result available to provider was 25 minutes (range, 20-110 minutes). Most results (n = 273 or 96%) were available within 1 hour. Rapid test results were available sooner than conventional test results (P < .0001). Excluding the 150 women whose time for availability of conventional test result was imputed did not affect overall results (median = 24 hours; P < .0001 compared with rapid testing). All rapid test results were available on the same day as the blood draw, whereas only 9% of conventional test results were available on the same day as the blood draw.

TABLE 2. Time durations for rapid and conventional HIV tests for women who presented late to prenatal care with undocumented HIV status: MIRIAD study; all clinics, 2002-2005

Time duration	n	Median time	Range	P value
Blood draw to availability of test result
Rapid testa	285	25 min	20-110 min
Conventional testb	285	23 h	3.5 h to 43 d	<.0001
Point-of-care rapid testing	204	24 min	20-80 min
Laboratory-based rapid testing	81	35 min	20-110 min	<.0001
Availability of test result to delivery
Rapid test	190	21 d	0-60 d
Conventional testc	190	20 d	-8 to 59 d	.15

HIV, human immunodeficiency virus; MIRIAD, Mother-Infant Rapid Intervention at Delivery.

Tepper. Rapid HIV testing in obstetric outpatient settings. Am J Obstet Gynecol 2009.

Among rapid tests, median test duration was shorter for point-of-care (24 minutes) than for laboratory-based testing (35 minutes; P < .0001). The median time between blood draws and availability of rapid test results at each of the 6 clinics ranged from 20-43 minutes and did not differ by volume of enrolled subjects, although numbers were too small for statistical comparison.

For the 283 women who received their rapid HIV test result, the median time from blood draw to informing the woman of her result was 28 minutes (range, 20-261 minutes). This duration was also significantly shorter for point-of-care (25 minutes) than for laboratory-based testing (43 minutes; P < .0001).

None of the 285 women tested had a positive rapid test result. One woman had a positive EIA result, followed by a second positive EIA; however, the confirmatory Western blot test result, available 3 days after the blood was drawn, was negative.

Among the 190 women for whom an accurate delivery date was available, the median interval between the availability of their test results and their delivery was 21 days for rapid testing and 20 days for conventional testing (P = .15). Seven women delivered before their conventional HIV test result was available.

Back to Article Outline

Comment 

Despite the potential utility of rapid testing in advancing universal HIV screening, many clinics still do not offer rapid testing.⁸ Our results demonstrate that rapid testing in the outpatient obstetric setting is acceptable, feasible, and provides results far sooner than conventional testing. More than 90% of women who were approached accepted rapid testing. Although all women in this study underwent both rapid and conventional testing, acceptance of rapid testing alone may be even higher if parallel testing were not required. In addition, it is unknown whether the 10% of women who declined study participation underwent rapid testing for clinical purposes. Therefore, our data provide a conservative estimate of acceptance of rapid testing in a clinic setting. Previous US studies, which have focused on testing in labor and delivery units, have found similarly high acceptance rates of rapid HIV testing, ranging from 74-97%.⁹, ¹², ¹³ Other US studies in emergency departments and outpatient settings, such as sexually transmitted disease clinics, have also shown that a high percentage of patients accepted rapid testing and received their test results.⁴, ¹⁴, ¹⁵, ¹⁶ These populations may differ from obstetric-gynecologic patients with respect to patient and provider HIV risk perception. The current study is unique in reporting results from an outpatient obstetric setting, where feasibility and acceptability have not been previously described.

Our study also demonstrates that results were available for rapid tests far sooner than for conventional tests. All women in our study had a rapid test result available to the provider within 2 hours, and almost 96% were available within 1 hour. Studies have shown that the quicker test results are available, the more likely patients will receive those results.¹⁶ It has been demonstrated among different populations that with conventional testing 10-59% of patients do not receive their test result, because of the lack of follow-up.¹² In this study, only 9% of conventional test results were available on the same day, which, at a minimum, would have required 91% of women to return for a second visit for their result. Rapid testing, therefore, has the potential to provide a significant advantage in increasing patient and provider knowledge of HIV status and, if positive, allowing for immediate entry into HIV care. For women who only undergo conventional testing, lack of follow-up would mean a loss of opportunity for antenatal and peripartum interventions in the event of a positive result. Even with good follow-up, the delay in getting conventional HIV test results could prevent women testing positive from receiving timely treatment. In this study, 7 women delivered before their conventional test result was available, emphasizing the critical need for rapid testing among women who present for prenatal care late in pregnancy.

The performance of rapid tests is comparable to that of the conventional EIA and WB tests. The OraQuick test is reported to have a sensitivity of 99.6% and a specificity of 100%.¹⁷ Although our study was too small to reliably calculate the sensitivity and specificity of the tests used, the rapid test performed with high accuracy and the only false-positive result was an EIA test. A decrease in positive predictive value is anticipated in low HIV prevalence settings. However, this should be balanced with the increase in awareness of HIV status among the population that would result from universal rapid testing.

Some considerations should be made before generalizing these results to all prenatal clinics in the United States. First, some of the study activities, such as point-of-care testing and reporting of results, were performed by trained MIRIAD personnel, thereby limiting generalizations of feasibility to regular clinic staff. However, once established, rapid testing requires less resources than conventional testing. Previous studies have shown that nursing staff and other personnel can be easily trained to perform the OraQuick test.¹¹ The OraQuick test may be performed by nonlaboratory personnel because of the CLIA waiver, which may enhance ease of implementation in the clinic setting, particularly clinics not located on hospital premises. Furthermore, an HIV counselor is not required onsite. Health care providers in the clinic can convey positive rapid test results as preliminary and explain the need for confirmatory testing. Referral to appropriate counseling should occur in the event of a positive confirmatory test.⁵ Second, because demographic information was not collected from the majority of women not enrolled in the study, we were unable to determine whether they differed from those who were enrolled. The demographic profile of women who enrolled is consistent with women who delay prenatal care¹⁸ and may be more likely to decline testing; women more compliant with care may have even higher rates of acceptance of rapid testing. Time durations for testing did not differ by whether women had any prenatal visits before enrollment, demonstrating that rapid testing can perform equivalently regardless of use of care. Third, our results were demonstrated in pregnant women. However, we believe they can be extrapolated to all outpatient gynecologic settings, where feasibility and timeliness among nonpregnant women would be comparable. Acceptability can be enhanced by the “opt-out” testing approach recommended by CDC and ACOG (in which HIV testing is performed as part of routine care unless patients decline).⁶, ¹⁹ Fourth, because each site performed the rapid test at only 1 location (point-of-care or laboratory-based), direct comparisons within sites were not possible. However, the difference in test duration for point-of-care vs laboratory-based rapid testing was minimal.

Despite these limitations, our results showed that rapid HIV testing is feasible in outpatient obstetric settings. Obstetrician-gynecologists have become increasingly aware of the importance of prenatal HIV testing and have become accustomed to offering HIV screening to all pregnant patients.²⁰ With the recent publication of ACOG guidelines recommending HIV testing for all women of reproductive age, obstetrician-gynecologists will need to expand HIV testing opportunities in their outpatient practices. The MIRIAD study can serve as a model for how rapid HIV testing can be incorporated into outpatient obstetric-gynecologic practices, and the results of our analyses demonstrate that testing in this setting is acceptable and feasible.

Back to Article Outline

Acknowledgments 

We thank Sivakumar Rangarajan, PMP (Business Analyst, CDC), and Sanjyot Shinde, PhD (Scientific Data Analyst, CDC), for their assistance; Jeffrey Wiener, MS (Mathematical Statistician, CDC), for his contribution to statistical analyses; and Dmitry Kissin, MD, MPH (Senior Service Fellow, CDC), for his careful review of the manuscript. At the study sites, we especially thank Rosalind Carter, PhD (Project Director, New York City, NY), Renata Dennis, RN, MPH (Instructor/Coordinator, Emory University, Atlanta, GA), Yolanda Olszewski, MPH, MSc (PRTII Project Director, Chicago, IL), and Angela Bradley-Byers, RN, BSN (Study Coordinator, New Orleans, LA).

Back to Article Outline

References 

1. US Preventive Services Task Force. Screening for HIV: recommendation statement. Ann Intern Med. 2005;143:32–37
   • View In Article
2. 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
   • View In Article
3. Hall HI, Song R, Rhodes P, et al. Estimation of HIV incidence in the United States. JAMA. 2008;300:520–529
   • View In Article
   • CrossRef
4. Centers for Disease Control and Prevention. Rapid HIV testing in outreach and other community settings—United States, 2004-2006. MMWR Morb Mortal Wkly Rep. 2007;56:1233–1237
   • View In Article
5. ACOG committee opinion no. 411: routine human immunodeficiency virus screening. Obstet Gynecol. 2008;112:401–403
   • View In Article
   • CrossRef
6. Centers for Disease Control and Prevention. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Recomm Rep. 2006;55(RR-14):1–17quiz CE1-4
   • View In Article
7. ACOG committee opinion no. 418: prenatal and perinatal human immunodeficiency virus testing: expanded recommendations. Obstet Gynecol. 2008;112:739–742
   • View In Article
   • CrossRef
8. Bogart LM, Howerton D, Lange J, Becker K, Setodji CM, Asch SM. Scope of rapid HIV testing in private nonprofit urban community health settings in the United States. Am J Public Health. 2008;98:736–742
   • View In Article
   • CrossRef
9. Bulterys M, Jamieson DJ, O'Sullivan MJ, et al. Rapid HIV-1 testing during labor: a multicenter study. JAMA. 2004;292:219–223
   • View In Article
   • CrossRef
10. Jamieson DJ, Cohen MH, Maupin R, et al. Rapid human immunodeficiency virus-1 testing on labor and delivery in 17 US hospitals: the MIRIAD experience. Am J Obstet Gynecol. 2007;197(suppl 3):S72–S82
    • View In Article
    • Full Text
    • Full-Text PDF (47 KB)
    • CrossRef
11. Dennis RL, Negron TJ, Lindsay M, Nesheim SR, Lee FK, Jamieson DJ. Rapid human immunodeficiency virus testing in labor and delivery: a comparison of implementation models between 2 hospitals. J Perinat Neonatal Nurs. 2007;21:298–306
    • View In Article
12. Roberts KJ, Grusky O, Swanson AN. Outcomes of blood and oral fluid rapid HIV testing: a literature review, 2000-2006. AIDS Patient Care STDS. 2007;21:621–637
    • View In Article
    • CrossRef
13. Pai NP, Tulsky JP, Cohan D, Colford JM, Reingold AL. Rapid point-of-care HIV testing in pregnant women: a systematic review and meta-analysis. Trop Med Int Health. 2007;12:162–173
    • View In Article
    • MEDLINE
    • CrossRef
14. Kelen GD, Shahan JB, Quinn TC. Emergency department-based HIV screening and counseling: experience with rapid and standard serologic testing. Ann Emerg Med. 1999;33:147–155
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (97 KB)
    • CrossRef
15. Kendrick SR, Kroc KA, Couture E, Weinstein RA. Comparison of point-of-care rapid HIV testing in three clinical venues. AIDS. 2004;18:2208–2210
    • View In Article
    • MEDLINE
    • CrossRef
16. Kendrick SR, Kroc KA, Withum D, Rydman RJ, Branson BM, Weinstein RA. Outcomes of offering rapid point-of-care HIV testing in a sexually transmitted disease clinic. J Acquir Immune Defic Syndr. 2005;38:142–146
    • View In Article
    • MEDLINE
    • CrossRef
17. Franco-Paredes C, Tellez I, del Rio C. Rapid HIV testing: a review of the literature and implications for the clinician. Curr HIV/AIDS Rep. 2006;3:169–175
    • View In Article
    • MEDLINE
    • CrossRef
18. Mayer JP. Unintended childbearing, maternal beliefs, and delay of prenatal care. Birth. 1997;24:247–252
    • View In Article
    • MEDLINE
    • CrossRef
19. ACOG committee opinion no. 304: prenatal and perinatal human immunodeficiency virus testing: expanded recommendations. Obstet Gynecol. 2004;104:1119–1124
    • View In Article
    • MEDLINE
    • CrossRef
20. Gray AD, Carlson R, Morgan MA, Hawks D, Schulkin J. Obstetrician gynecologists' knowledge and practice regarding human immunodeficiency virus screening. Obstet Gynecol. 2007;110:1019–1026
    • View In Article
    • CrossRef