- Email: [email protected]
Prenatal HIV Counseling, Testing, and Antiretroviral Prophylaxis by Obstetric and Family Medicine Providers in Alabama
Prenatal HIV Counseling, Testing, and Antiretroviral Prophylaxis by Obstetric and Family Medicine Providers in Alabama STACY A. NICHOLS, MD, MPH; MADHAV P. BHATTA, MPH; JENNIFER LEWIS, MPH; STEN H. VERMUND, MD, PHD
ABSTRACT: Background: The study reported here assessed the implementation of recommendations for routine universal prenatal counseling and voluntary HIV testing among Alabama physicians who provide prenatal care and determined factors associated with noncompliance. Methods: Voluntary, confidential mailed survey of obstetrics and family medicine practitioners. Results: Of the 138 physicians who responded to a mailed survey in Alabama, 17 (12.3%) indicated that they did not offer universal HIV counseling and testing to pregnant women. Factors associated with failure to offer universal HIV counseling and testing included having more than 50% of patients refuse HIV counseling and testing when offered and never knowingly having given prenatal/perinatal care to women with HIV. Low/me-
dium familiarity with the US Public Health Service recommendations for perinatal zidovudine use to reduce HIV transmission and physician specialty are also suggested as predictors of not offering universal testing. Conclusions: Despite the well-established benefits of antiretroviral prophylaxis to prevent vertical transmission of HIV, some physicians in Alabama have been slow to adopt universal testing of their pregnant patients for HIV in the prenatal period. Practitioner education is as important as patient education in eliminating pediatric HIV in the Deep South. KEY INDEXING TERMS: Perinatal HIV transmission; Antiretroviral agents; Alabama; Prevention. [Am J Med Sci 2002;324(6):305– 309.]
V
lished in 1998.6,7 Furthermore, more aggressive combination drug regimens that can suppress viral replication are clearly beneficial for mother and, indirectly, for the HIV-exposed fetus.6 A 1998 Institute of Medicine (IOM) report1 carries this recommendation further: to minimize the mother-to-infant transmission of HIV, the United States should adopt “a national policy of universal HIV testing, with patient notification, as a routine component of prenatal care.” Thus, the IOM is suggesting explicitly that the benefits of HIV prevention outweigh the harm of testing with or without counseling. This recommendation has been supported by the American College of Obstetricians and Gynecologists8 and the American Academy of Pediatrics.8 Results of short-course ZDV prophylaxis trials,9 –11 observational data,12 and a nevirapine trial13 point to the significant benefits of short-course chemoprophylaxis on vertical transmission, even when administered to the woman late in pregnancy and/or to the infant immediately after birth. Because prenatal testing and advance planning are desirable to ensure the availability of antiretroviral chemotherapies intrapartum and in the immediate neonatal period, physicians who provide prenatal care
ertical transmission of HIV occurs in about 25% of non– breast-fed infants in the United States without perinatal chemotherapeutic intervention.1 Use of zidovudine (ZDV) in the AIDS Clinical Trial Group protocol 076 was found to reduce mother-tochild transmission by about two-thirds among women who were not breastfeeding.2 The US Public Health Service (USPHS) published recommendations for the use of ZDV to reduce perinatal HIV transmission in 1994.3 Recommendations urging universal prenatal HIV counseling and testing originally were made in 1995.4 Significant reductions in perinatal HIV transmission seen in the United States since 1994 are attributed, at least in part, to physicians’ implementation of HIV counseling, testing, and antiretroviral prophylaxis.1,5 Recommendations for universal screening in pregnancy were pub-
From the Department of Epidemiology & International Health, University of Alabama at Birmingham School of Public Health, Birmingham, Alabama. Submitted April 2, 2002; accepted August 15, 2002. Correspondence: Sten H. Vermund, M.D., Ph.D., University of Alabama at Birmingham, 845 19th Street South, BBRB 206, Birmingham, AL 35294 –2170 (E-mail: [email protected]). THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
305
Prenatal HIV Healthcare in Alabama
must be made aware of the recommendations regarding universal counseling and voluntary testing of pregnant women and the perinatal use of antiretroviral chemotherapy. In 1997, we surveyed obstetricians and family practice physicians who provide prenatal services in Alabama to assess compliance with the 1994 and 1995 USPHS3,4 recommendations and factors associated with noncompliance. Although pediatric HIV/ AIDS has declined from 11 cases in Alabama in 1996 to 4 cases in 2000,14,15 the fact that there are any cases at all suggests that physician attitudes to HIV screening are worthy of consideration. Methods The Alabama State Perinatal Advisory Committee of the Medical Association of the State of Alabama (MASA) provided a mailing list of 515 obstetrician/gynecologists (OB/GYN) and 1383 family practice/general practice physicians (FP). Ross Pediatrics, a formula supplier that markets to physicians who deliver perinatal health care, supplied a second, supplementary mailing list. The Ross list contained names that were also included on the MASA list, as well as 76 additional names of Alabama physicians. When the 2 lists differed slightly in the address for the same physician, the Ross address was used, because that list was more recently compiled. To address the problem of uncertainty regarding which physicians in the database were prenatal caregivers, an “opt-out” question appeared on the first page of the survey, allowing those persons to identify themselves. They were asked to return the survey in the preaddressed stamped envelope to ensure accurate response rate determination. A cover letter stating the purpose of the survey and asking the physicians to complete the questionnaire was included. No efforts were made to survey nonresponders by telephone or otherwise. The Institutional Review Board of the University of Alabama at Birmingham approved the protocol and survey. Confidentiality was maintained by establishing a separate database for responses to survey items. A person who did not have access to the address database entered data. The address database was updated with the response identification number assigned to each completed survey to determine the response rate. The survey instrument included 20 questions in 5 categories: demographics; experience and current practice regarding perinatal HIV care; practice characteristics/systems issues potentially impacting implementation and compliance of USPHS recommendations; educational tools used for HIV counseling of pregnant women; and information sources for physicians regarding antiretroviral prophylaxis against perinatal HIV transmission. We neglected to ask about the size of the practice group in which a physician worked. Data analyses were performed using the SAS software (SAS Institute, Cary, NC). Logistic regression analysis was conducted using Stepwise Selection procedure with the outcome variable dichotomized as “offer counseling and testing to all” versus “offer counseling and testing to some or no” pregnant patients.
Results Of 1974 original names in the merged (MASA and Ross) address database, 1911 surveys were believed to have reached the target population of OB/GYNs (499) and FPs (1412) in Alabama—the other 63 were returned marked “retired,” “deceased,” “out-ofstate,” or “not OB/GYN” (survey unopened). Of 1911 surveys mailed, 522 were returned in the envelope provided (159 OB/GYNs and 357 FPs), for an overall 306
response rate of 27%. Of the 522 surveys mailed back, 378 marked the “opt-out/no prenatal care” question on the first page, representing 90% of responding FPs and 30% of OB/GYNs. Six responses were omitted because surveys were not fully completed with regard to key questions, leaving 138 respondents who had practices relevant to this “obstetrics provider” survey. Looking at the response rate another way, assuming that 70% of 499 OB/ GYNs actually practice some obstetrics, 10% of the 1412 FPs practice OB/GYN, and all 76 extra names from the Ross list practice obstetrics, our 138 respondents represented a 24% response rate. Of the 138 physicians, 78.6% were OB/GYNs and 76.6% were male. Forty-four percent practiced in communities with fewer than 100,000 people; 6% of the physicians practiced in communities with fewer than 10,000 people. One third of the physicians indicated that they had never, as far as they knew, given prenatal or perinatal care to a woman with HIV. Fifteen percent of the physicians surveyed indicated that more than half of their patients refused HIV counseling and testing when offered. In terms of familiarity with USPHS recommendations for prenatal HIV counseling and testing, 21.3% indicated low, 41.8% medium, and 36.9% high familiarity. A majority (70.2%) of the physicians reported being comfortable in discussing HIV risk factors and other HIV counseling issues with their patients; 29.8% reported feeling uncomfortable or neutral. Thirty-eight percent of the physicians in the survey indicated that more than half of their patients were covered by Medicaid for prenatal care. Forty-six percent of the respondents indicated that their practice did not have personnel trained to meet the medical needs of HIV-positive pregnant women; 28.1% of physicians indicated that their practice did not have personnel and/or referral site to meet psychosocial needs of women with HIV. Thirteen percent of the respondents indicated that a patient’s lack of insurance coverage to pay for HIV testing and/or ZDV prophylaxis would be a factor that might hinder HIV counseling /testing/ZDV prophylaxis in their practice. Seventeen (12.3%) of the respondents indicated that they did not offer universal HIV-testing and counseling to pregnant women. Factors in univariable analysis that were significantly associated with not offering HIV counseling and testing to some or no patients include low/medium familiarity with CDC recommendations for perinatal HIV testing and counseling, having no capacity/system for administering intravenous ZDV intrapartum when indicated, having never knowingly given prenatal/ perinatal care to women with HIV, and having greater than 50% of patients refuse HIV testing and counseling when offered. (Table 1). The factors that remained statistically significant as correlates of not providing universal testing and counseling to all December 2002 Volume 324 Number 6
Nichols et al
Table 1. Univariable Analysis of Factors Associated with Partial or No HIV Testing and Counseling among Alabama Physicians Providing Obstetric Care Offer HIV Counseling and Testing to Some or No Patients (%)
To All Patients (%)
Odds Ratio
1 (2) 16 (19)
51 (98) 70 (81)
11.6
0.02*
9 (9) 8 (21)
90 (91) 31 (79)
2.6
0.07
12 (11) 5 (18)
98 (89) 23 (82)
1.8
0.29
16 (12) 1 (17)
116 (88) 5 (83)
1.5
0.74
6 (7) 11 (26)
88 (93) 31 (73)
5.2
0.003*
10 (13) 7 (12)
68 (87) 53 (88)
0.9
0.84
12 (11) 5 (17)
96 (89) 24 (83)
1.7
0.38
10 (15) 7 (10)
58 (85) 62 (90)
0.7
0.65
6 (7) 11 (25)
88 (93) 33 (75)
4.9
0.004*
7 (6) 4 (21)
103 (94) 15 (79)
3.9
0.05*
Familiarity with USPHS recommendation for perinatal HIV testing and counseling High Low/Medium Level of comfort in discussing HIV risk factors and other HIV counseling issues Comfortable Uncomfortable/Neutral Availability of staff for psychological counseling and/or trained personnel for the care of HIV patients Either available Neither available Availability of lab facilities or referral lab for HIV testing Yes No Capacity/system for administering IV ZDV intrapartun when indicated Yes No Population size of the community of practice ⱖ 100,000 ⬍ 100,000 Specialty of the participants Obstetrics/gynecology Family practice Percent of population in Medicaid ⬍50 ⱖ50 Number of HIV ⫹ women ever given prenatal/perinatal care One or more None Percent of patients refusing HIV testing/counseling when offered ⱕ35 ⱖ50
P value
* Statistically significant at ␣ ⫽ 0.05 level
pregnant women in a multivariable analysis included: having never knowingly given prenatal/perinatal care to women with HIV, and having more than 50% of patients refuse HIV counseling and testing when offered (Table 2). Discussion This Alabama physician survey gives insight into the failure of some physicians to offer HIV counseling
and testing to all pregnant women. A physician’s expectation or experience of patient refusal of HIV testing and a physician’s having never cared for (or having never recognized caring for) a woman with HIV were the significant predictors of not offering universal HIV testing among obstetricians and family medicine physicians in Alabama. Relative unfamiliarity with published guidelines and specialty of physicians were also suggested strongly as a predictors of not offering HIV
Table 2. Multivariable Analysis of Factors Associated with Partial or No HIV Testing and Counseling Offered by Alabama Physicians Providing Obstetric Care Odds Ratio Low/medium familiarity with USPHS recommendation of perinatal HIV testing and counseling Family practice physician ⱖ 50% of patients refusing HIV testing/counseling when offered Having never given prenatal/perinatal care to a HIV ⫹ woman
6.1 4.3 12.2 7.3
P value 0.11 0.08 0.001* 0.01*
* Statistically significant at ␣ ⫽ 0.05 level THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
307
Prenatal HIV Healthcare in Alabama
testing, but this did not reach statistical significance. These factors were among many cited during a 1999 CDC teleconference as possible hindrances to implementation of ZDV prophylaxis against perinatal HIV transmission.16 Of note was that although only 12% of physicians reported a lack of full compliance with recommendations, much larger numbers of physicians reported a lack of adequately trained personnel to meet the needs of HIV-positive patients and/or a lack of a system of delivery of intrapartum intravenous ZDV. This calls into question the quality of HIV counseling/prevention received by some women who have been offered testing. These findings support the IOM recommendation that HIV testing become a universal, routine part of prenatal care, with patient notification included to preserve their autonomy in the decision to test. With such a system, patient refusal might be expected to decrease if testing were perceived as recommended for all women and nonjudgmental/nonstigmatizing. This is especially important in Alabama, the home of the Tuskegee syphilis study, which continues to cast mistrust on the medical community.17 Routine testing suggests that physicians not assess risk but rather screen for HIV in all pregnant women. Further, routine testing would allow for individualized counseling for patients who desire more information if they are considering refusal of the test; such counseling could be provided via regionalized experts by phone or in central locations, thus limiting the potential reluctance to test if local personnel are not available to meet medical and psychosocial needs of those women who test positive. With regard to physician knowledge, this study shows that in 1997, one fifth of physicians reported a low level of familiarity with recommendations for ZDV prophylaxis. Certainly, it would be expected that this proportion has diminished subsequently. However, with new (1998) guidelines and even newer drug regimens—including short courses and combination therapies, continuing medical education to clarify for prenatal care givers the increasingly complex decision matrix to be employed in efforts to prevent perinatal HIV transmission.18,19 It should be emphasized that, although safety issues remain unresolved with regard to some antiretroviral agents, data to date show that antiretroviral prophylaxis against perinatal HIV transmission has not been shown to have adverse effects on the HIVexposed fetus.20 –22 Limitations of the study include the relatively small sample size, low response rate, and the expansion in chemotherapeutic options since the survey was conducted. “Volunteer” bias might suggest that our respondents reflect the activities of somewhat more motivated physicians. Thus, our findings may be an overly optimistic assessment of the true state of physician management of pregnant women in 1997. This study predated the findings that shorter 308
course ZDV can be effective,10,11 that immediate postpartum treatment showed promise,11 and the IOM recommendation for making perinatal screening universal and routine, with patient notification, to minimize the HIV mother-to child transmission rate in the U.S.1 A follow-up study conducted with more optimal survey methods would be informative, although such a survey might better follow an intensive physician education effort. Such a study might use the current survey’s findings as a point for comparison with current knowledge, attitudes, and practices. If universal testing of pregnant women is promulgated in obstetric practices that have very low seroprevalence of HIV, then most enzyme-linked immunosorbent assay (ELISA) positive tests will be false positives. Hence, it is incumbent upon clinicians to confirm ELISA positives with the more specific Western blot test before sharing HIV status data with a patient. Commercial and health department laboratories will do confirmatory testing as a matter of course, but this point is nonetheless worth emphasizing as HIV testing becomes more routine in antenatal care. In summary, our study reports that a substantial minority (12%) of Alabama obstetric providers responding to a mail survey did not implement universal HIV counseling and testing in 1997. Similar surveys from Thailand suggests that Alabama findings may well be true in quite diverse obstetric provider settings.23,24 Furthermore, a large national survey of 1362 parturient women from 7 hospitals in North Carolina, Connecticut, New York, and Florida noted that 11.1% of women who were not tested for HIV stated that the test had not been offered or recommended by the obstetric providers.25 Physician education regarding current recommendations— universal routine testing with patient notification, regionalization of care for pregnant women with HIV, with consultation support available to local physicians, and steps to ensure adequate systems of intrapartum ZDV delivery— could enhance achievement of the ongoing continuing goal of radical reduction in mother-to-child transmission of HIV. The attitude that 1 affected child is too many must motivate continued and expanded implementation of antiretroviral chemoprophylaxis against perinatal HIV transmission. Acknowledgments We thank Drs. LeaVonne Pulley and Leslie Clark from the UAB Department of Health Behavior for their assistance in the development of the survey instrument. References 1. Stoto MA, Almario DA, McCormick MC, editors. Reducing the odds: preventing perinatal transmission of HIV in the
December 2002 Volume 324 Number 6
Nichols et al
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
United States. Washington (DC): Institute of Medicine, National Academy Press; 1998. p. 397. Available at: URL: http:// www.nap.edu/books/0309062861/html/index.html Conner EM, Sperling RS, Gelber R, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994; 331:1173– 80. Anonymous. Recommendations of the U.S. Public Health Service Task Force on the use of zidovudine to reduce perinatal transmission of human immunodeficiency virus. MMWR Recomm Rep 1994;43:1–20. Anonymous. U.S. Public Health Service recommendations for human immunodeficiency virus counseling and voluntary testing for pregnant women. MMWR Recomm Rep 1995;44: 1–15. Centers for Disease Control and Prevention. Table 26, Estimated persons living with AIDS, by region of residence and year. In: HIV/AIDS surveillance report. 1999;11(2):37. Available at: URL: http://www.cdc.gov/hiv/stats/hasr1102.htm Anonymous. Public Health Service Task Force recommendations for the use of antiretroviral drugs in pregnant women infected with HIV-1 for maternal health and for reducing perinatal HIV-1 transmission in the United States. Centers for Disease Control and Prevention. MMWR Recomm Rep 1998;47:1–30. Matheson PB, Abrams EJ, Thomas PA, et al. Efficacy of antenatal zidovudine in reducing perinatal transmission of human immunodeficiency virus type 1. J Infect Dis 1995;172: 353– 8. Anonymous. Human immunodeficiency virus screening. Joint statement of the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists. Pediatrics 1999;104:128. Shaffer N, Chuachoowong R, Mock PA, et al. Shortcourse zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial. Lancet 1999;353:773– 80. Wiktor SZ, Ekpini E, Karon JM, et al. Short-course oral zidovudine for prevention of mother-to-child transmission of HIV-1 in Abidjan, Côte d’Ivoire: a randomised trial. Lancet 1999;353:781–5. Dabis F, Msellati P, Meda N, et al. 6-month efficacy, tolerance, and acceptability of a short regimen of oral zidovudine to reduce vertical transmission of HIV in breastfed children in Côte d’Ivoire and Burkina Faso: a double-blind placebo-controlled multicentre trial. Lancet 1999;353:786 – 92. Wade NA, Birkhead GS, Warren BL, et al. Abbreviated regimens of zidovudine prophylaxis and perinatal transmission of human immunodeficiency virus. N Engl J Med 1998; 339:1409 –14.
THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
13. Guay LA, Musoke P, Fleming T, et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet 1999;354:795– 802. 14. Centers for Disease Control and Prevention. Figure 5, Pediatric AIDS cases. In: HIV/AIDS surveillance report 2000; 12(2):26. Available at: URL: http://www.cdc.gov/hiv/stats/ hasr1202.htm 15. Centers for Disease Control and Prevention. Table 16, Pediatric HIV infection cases by exposure category and race/ ethnicity. In: HIV/AIDS surveillance report 1999;12(2):25. Available at: URL: http://www.cdc.gov/hiv/stats/hasr1202.htm 16. Centers for Disease Control and Prevention. Update on preventing perinatal transmission of HIV. Satellite broadcast, 1999 Apr 29. CDC Public Health Training Network (PHTN), CDC National Center for HIV, STD, and TB Prevention (NCHSTP), and the CDC National Prevention Information Network (NPIN). Transcript available at: URL: http:// www.cdcnpin.org/broadcast/past/1999/0429/Transcript.pdf 17. Thomas SB, Curran JW. Tuskegee: from science to conspiracy to metaphor. Am J Med Sci 1999;317:1– 4. 18. Montaner JSG, Reiss P, Cooper D, et al. A randomized, double-blind trial comparing combinations of nevirapine, didanosine, and zidovudine for HIV-infected patients. JAMA 1998;270:930 –7. 19. Stringer JSA, Vermund S. Prevention of mother-to-child transmission of HIV-1. Curr Opin Obstet Gynecol 1999;11: 427–34. 20. Rouse DJ, Owen J, Goldenberg RL, et al. Zidovudine for the prevention of vertical HIV transmission: a decision analytic approach. J Acquir Immun Defic Syndr Hum Retrovirol 1995;9:401–7. 21. Culnane M, Fowler M, Lee SS, et al. Lack of long-term effects of in utero exposure to zidovudine among uninfected children born to HIV-infected women. Pediatric AIDS Clinical Trials Group Protocol 219/076 Teams. JAMA 1999;281: 151–7. 22. Stringer JSA, Sinkala M, Rouse DJ, et al. Effect of nevirapine toxicity on choice of perinatal HIV prevention strategies. Am J Public Health 2002;92:365– 6. 23. Stringer JSA, Stringer EM, Phanuphak P, et al. Prevention of mother-to-child transmission of HIV in Thailand: physicians’ attitudes on zidovudine use, pregnancy termination, and willingness to provide care. J Acquir Immune Defic Syndr 1999;21:217–22. 24. Bhatta MP, Stringer JSA, Phanuphak P, et al. Motherto-child HIV transmission prevention in Thailand: physician zidovudine use and willingness to provide care. Int J STD AIDS, in press. 25. Royce RA, Walter EB, Fernandez MI, et al. Barriers to universal prenatal HIV testing in 4 US locations in 1997. Am J Public Health 2001;91:727–33.
309
ABSTRACT: Background: The study reported here assessed the implementation of recommendations for routine universal prenatal counseling and voluntary HIV testing among Alabama physicians who provide prenatal care and determined factors associated with noncompliance. Methods: Voluntary, confidential mailed survey of obstetrics and family medicine practitioners. Results: Of the 138 physicians who responded to a mailed survey in Alabama, 17 (12.3%) indicated that they did not offer universal HIV counseling and testing to pregnant women. Factors associated with failure to offer universal HIV counseling and testing included having more than 50% of patients refuse HIV counseling and testing when offered and never knowingly having given prenatal/perinatal care to women with HIV. Low/me-
dium familiarity with the US Public Health Service recommendations for perinatal zidovudine use to reduce HIV transmission and physician specialty are also suggested as predictors of not offering universal testing. Conclusions: Despite the well-established benefits of antiretroviral prophylaxis to prevent vertical transmission of HIV, some physicians in Alabama have been slow to adopt universal testing of their pregnant patients for HIV in the prenatal period. Practitioner education is as important as patient education in eliminating pediatric HIV in the Deep South. KEY INDEXING TERMS: Perinatal HIV transmission; Antiretroviral agents; Alabama; Prevention. [Am J Med Sci 2002;324(6):305– 309.]
V
lished in 1998.6,7 Furthermore, more aggressive combination drug regimens that can suppress viral replication are clearly beneficial for mother and, indirectly, for the HIV-exposed fetus.6 A 1998 Institute of Medicine (IOM) report1 carries this recommendation further: to minimize the mother-to-infant transmission of HIV, the United States should adopt “a national policy of universal HIV testing, with patient notification, as a routine component of prenatal care.” Thus, the IOM is suggesting explicitly that the benefits of HIV prevention outweigh the harm of testing with or without counseling. This recommendation has been supported by the American College of Obstetricians and Gynecologists8 and the American Academy of Pediatrics.8 Results of short-course ZDV prophylaxis trials,9 –11 observational data,12 and a nevirapine trial13 point to the significant benefits of short-course chemoprophylaxis on vertical transmission, even when administered to the woman late in pregnancy and/or to the infant immediately after birth. Because prenatal testing and advance planning are desirable to ensure the availability of antiretroviral chemotherapies intrapartum and in the immediate neonatal period, physicians who provide prenatal care
ertical transmission of HIV occurs in about 25% of non– breast-fed infants in the United States without perinatal chemotherapeutic intervention.1 Use of zidovudine (ZDV) in the AIDS Clinical Trial Group protocol 076 was found to reduce mother-tochild transmission by about two-thirds among women who were not breastfeeding.2 The US Public Health Service (USPHS) published recommendations for the use of ZDV to reduce perinatal HIV transmission in 1994.3 Recommendations urging universal prenatal HIV counseling and testing originally were made in 1995.4 Significant reductions in perinatal HIV transmission seen in the United States since 1994 are attributed, at least in part, to physicians’ implementation of HIV counseling, testing, and antiretroviral prophylaxis.1,5 Recommendations for universal screening in pregnancy were pub-
From the Department of Epidemiology & International Health, University of Alabama at Birmingham School of Public Health, Birmingham, Alabama. Submitted April 2, 2002; accepted August 15, 2002. Correspondence: Sten H. Vermund, M.D., Ph.D., University of Alabama at Birmingham, 845 19th Street South, BBRB 206, Birmingham, AL 35294 –2170 (E-mail: [email protected]). THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
305
Prenatal HIV Healthcare in Alabama
must be made aware of the recommendations regarding universal counseling and voluntary testing of pregnant women and the perinatal use of antiretroviral chemotherapy. In 1997, we surveyed obstetricians and family practice physicians who provide prenatal services in Alabama to assess compliance with the 1994 and 1995 USPHS3,4 recommendations and factors associated with noncompliance. Although pediatric HIV/ AIDS has declined from 11 cases in Alabama in 1996 to 4 cases in 2000,14,15 the fact that there are any cases at all suggests that physician attitudes to HIV screening are worthy of consideration. Methods The Alabama State Perinatal Advisory Committee of the Medical Association of the State of Alabama (MASA) provided a mailing list of 515 obstetrician/gynecologists (OB/GYN) and 1383 family practice/general practice physicians (FP). Ross Pediatrics, a formula supplier that markets to physicians who deliver perinatal health care, supplied a second, supplementary mailing list. The Ross list contained names that were also included on the MASA list, as well as 76 additional names of Alabama physicians. When the 2 lists differed slightly in the address for the same physician, the Ross address was used, because that list was more recently compiled. To address the problem of uncertainty regarding which physicians in the database were prenatal caregivers, an “opt-out” question appeared on the first page of the survey, allowing those persons to identify themselves. They were asked to return the survey in the preaddressed stamped envelope to ensure accurate response rate determination. A cover letter stating the purpose of the survey and asking the physicians to complete the questionnaire was included. No efforts were made to survey nonresponders by telephone or otherwise. The Institutional Review Board of the University of Alabama at Birmingham approved the protocol and survey. Confidentiality was maintained by establishing a separate database for responses to survey items. A person who did not have access to the address database entered data. The address database was updated with the response identification number assigned to each completed survey to determine the response rate. The survey instrument included 20 questions in 5 categories: demographics; experience and current practice regarding perinatal HIV care; practice characteristics/systems issues potentially impacting implementation and compliance of USPHS recommendations; educational tools used for HIV counseling of pregnant women; and information sources for physicians regarding antiretroviral prophylaxis against perinatal HIV transmission. We neglected to ask about the size of the practice group in which a physician worked. Data analyses were performed using the SAS software (SAS Institute, Cary, NC). Logistic regression analysis was conducted using Stepwise Selection procedure with the outcome variable dichotomized as “offer counseling and testing to all” versus “offer counseling and testing to some or no” pregnant patients.
Results Of 1974 original names in the merged (MASA and Ross) address database, 1911 surveys were believed to have reached the target population of OB/GYNs (499) and FPs (1412) in Alabama—the other 63 were returned marked “retired,” “deceased,” “out-ofstate,” or “not OB/GYN” (survey unopened). Of 1911 surveys mailed, 522 were returned in the envelope provided (159 OB/GYNs and 357 FPs), for an overall 306
response rate of 27%. Of the 522 surveys mailed back, 378 marked the “opt-out/no prenatal care” question on the first page, representing 90% of responding FPs and 30% of OB/GYNs. Six responses were omitted because surveys were not fully completed with regard to key questions, leaving 138 respondents who had practices relevant to this “obstetrics provider” survey. Looking at the response rate another way, assuming that 70% of 499 OB/ GYNs actually practice some obstetrics, 10% of the 1412 FPs practice OB/GYN, and all 76 extra names from the Ross list practice obstetrics, our 138 respondents represented a 24% response rate. Of the 138 physicians, 78.6% were OB/GYNs and 76.6% were male. Forty-four percent practiced in communities with fewer than 100,000 people; 6% of the physicians practiced in communities with fewer than 10,000 people. One third of the physicians indicated that they had never, as far as they knew, given prenatal or perinatal care to a woman with HIV. Fifteen percent of the physicians surveyed indicated that more than half of their patients refused HIV counseling and testing when offered. In terms of familiarity with USPHS recommendations for prenatal HIV counseling and testing, 21.3% indicated low, 41.8% medium, and 36.9% high familiarity. A majority (70.2%) of the physicians reported being comfortable in discussing HIV risk factors and other HIV counseling issues with their patients; 29.8% reported feeling uncomfortable or neutral. Thirty-eight percent of the physicians in the survey indicated that more than half of their patients were covered by Medicaid for prenatal care. Forty-six percent of the respondents indicated that their practice did not have personnel trained to meet the medical needs of HIV-positive pregnant women; 28.1% of physicians indicated that their practice did not have personnel and/or referral site to meet psychosocial needs of women with HIV. Thirteen percent of the respondents indicated that a patient’s lack of insurance coverage to pay for HIV testing and/or ZDV prophylaxis would be a factor that might hinder HIV counseling /testing/ZDV prophylaxis in their practice. Seventeen (12.3%) of the respondents indicated that they did not offer universal HIV-testing and counseling to pregnant women. Factors in univariable analysis that were significantly associated with not offering HIV counseling and testing to some or no patients include low/medium familiarity with CDC recommendations for perinatal HIV testing and counseling, having no capacity/system for administering intravenous ZDV intrapartum when indicated, having never knowingly given prenatal/ perinatal care to women with HIV, and having greater than 50% of patients refuse HIV testing and counseling when offered. (Table 1). The factors that remained statistically significant as correlates of not providing universal testing and counseling to all December 2002 Volume 324 Number 6
Nichols et al
Table 1. Univariable Analysis of Factors Associated with Partial or No HIV Testing and Counseling among Alabama Physicians Providing Obstetric Care Offer HIV Counseling and Testing to Some or No Patients (%)
To All Patients (%)
Odds Ratio
1 (2) 16 (19)
51 (98) 70 (81)
11.6
0.02*
9 (9) 8 (21)
90 (91) 31 (79)
2.6
0.07
12 (11) 5 (18)
98 (89) 23 (82)
1.8
0.29
16 (12) 1 (17)
116 (88) 5 (83)
1.5
0.74
6 (7) 11 (26)
88 (93) 31 (73)
5.2
0.003*
10 (13) 7 (12)
68 (87) 53 (88)
0.9
0.84
12 (11) 5 (17)
96 (89) 24 (83)
1.7
0.38
10 (15) 7 (10)
58 (85) 62 (90)
0.7
0.65
6 (7) 11 (25)
88 (93) 33 (75)
4.9
0.004*
7 (6) 4 (21)
103 (94) 15 (79)
3.9
0.05*
Familiarity with USPHS recommendation for perinatal HIV testing and counseling High Low/Medium Level of comfort in discussing HIV risk factors and other HIV counseling issues Comfortable Uncomfortable/Neutral Availability of staff for psychological counseling and/or trained personnel for the care of HIV patients Either available Neither available Availability of lab facilities or referral lab for HIV testing Yes No Capacity/system for administering IV ZDV intrapartun when indicated Yes No Population size of the community of practice ⱖ 100,000 ⬍ 100,000 Specialty of the participants Obstetrics/gynecology Family practice Percent of population in Medicaid ⬍50 ⱖ50 Number of HIV ⫹ women ever given prenatal/perinatal care One or more None Percent of patients refusing HIV testing/counseling when offered ⱕ35 ⱖ50
P value
* Statistically significant at ␣ ⫽ 0.05 level
pregnant women in a multivariable analysis included: having never knowingly given prenatal/perinatal care to women with HIV, and having more than 50% of patients refuse HIV counseling and testing when offered (Table 2). Discussion This Alabama physician survey gives insight into the failure of some physicians to offer HIV counseling
and testing to all pregnant women. A physician’s expectation or experience of patient refusal of HIV testing and a physician’s having never cared for (or having never recognized caring for) a woman with HIV were the significant predictors of not offering universal HIV testing among obstetricians and family medicine physicians in Alabama. Relative unfamiliarity with published guidelines and specialty of physicians were also suggested strongly as a predictors of not offering HIV
Table 2. Multivariable Analysis of Factors Associated with Partial or No HIV Testing and Counseling Offered by Alabama Physicians Providing Obstetric Care Odds Ratio Low/medium familiarity with USPHS recommendation of perinatal HIV testing and counseling Family practice physician ⱖ 50% of patients refusing HIV testing/counseling when offered Having never given prenatal/perinatal care to a HIV ⫹ woman
6.1 4.3 12.2 7.3
P value 0.11 0.08 0.001* 0.01*
* Statistically significant at ␣ ⫽ 0.05 level THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
307
Prenatal HIV Healthcare in Alabama
testing, but this did not reach statistical significance. These factors were among many cited during a 1999 CDC teleconference as possible hindrances to implementation of ZDV prophylaxis against perinatal HIV transmission.16 Of note was that although only 12% of physicians reported a lack of full compliance with recommendations, much larger numbers of physicians reported a lack of adequately trained personnel to meet the needs of HIV-positive patients and/or a lack of a system of delivery of intrapartum intravenous ZDV. This calls into question the quality of HIV counseling/prevention received by some women who have been offered testing. These findings support the IOM recommendation that HIV testing become a universal, routine part of prenatal care, with patient notification included to preserve their autonomy in the decision to test. With such a system, patient refusal might be expected to decrease if testing were perceived as recommended for all women and nonjudgmental/nonstigmatizing. This is especially important in Alabama, the home of the Tuskegee syphilis study, which continues to cast mistrust on the medical community.17 Routine testing suggests that physicians not assess risk but rather screen for HIV in all pregnant women. Further, routine testing would allow for individualized counseling for patients who desire more information if they are considering refusal of the test; such counseling could be provided via regionalized experts by phone or in central locations, thus limiting the potential reluctance to test if local personnel are not available to meet medical and psychosocial needs of those women who test positive. With regard to physician knowledge, this study shows that in 1997, one fifth of physicians reported a low level of familiarity with recommendations for ZDV prophylaxis. Certainly, it would be expected that this proportion has diminished subsequently. However, with new (1998) guidelines and even newer drug regimens—including short courses and combination therapies, continuing medical education to clarify for prenatal care givers the increasingly complex decision matrix to be employed in efforts to prevent perinatal HIV transmission.18,19 It should be emphasized that, although safety issues remain unresolved with regard to some antiretroviral agents, data to date show that antiretroviral prophylaxis against perinatal HIV transmission has not been shown to have adverse effects on the HIVexposed fetus.20 –22 Limitations of the study include the relatively small sample size, low response rate, and the expansion in chemotherapeutic options since the survey was conducted. “Volunteer” bias might suggest that our respondents reflect the activities of somewhat more motivated physicians. Thus, our findings may be an overly optimistic assessment of the true state of physician management of pregnant women in 1997. This study predated the findings that shorter 308
course ZDV can be effective,10,11 that immediate postpartum treatment showed promise,11 and the IOM recommendation for making perinatal screening universal and routine, with patient notification, to minimize the HIV mother-to child transmission rate in the U.S.1 A follow-up study conducted with more optimal survey methods would be informative, although such a survey might better follow an intensive physician education effort. Such a study might use the current survey’s findings as a point for comparison with current knowledge, attitudes, and practices. If universal testing of pregnant women is promulgated in obstetric practices that have very low seroprevalence of HIV, then most enzyme-linked immunosorbent assay (ELISA) positive tests will be false positives. Hence, it is incumbent upon clinicians to confirm ELISA positives with the more specific Western blot test before sharing HIV status data with a patient. Commercial and health department laboratories will do confirmatory testing as a matter of course, but this point is nonetheless worth emphasizing as HIV testing becomes more routine in antenatal care. In summary, our study reports that a substantial minority (12%) of Alabama obstetric providers responding to a mail survey did not implement universal HIV counseling and testing in 1997. Similar surveys from Thailand suggests that Alabama findings may well be true in quite diverse obstetric provider settings.23,24 Furthermore, a large national survey of 1362 parturient women from 7 hospitals in North Carolina, Connecticut, New York, and Florida noted that 11.1% of women who were not tested for HIV stated that the test had not been offered or recommended by the obstetric providers.25 Physician education regarding current recommendations— universal routine testing with patient notification, regionalization of care for pregnant women with HIV, with consultation support available to local physicians, and steps to ensure adequate systems of intrapartum ZDV delivery— could enhance achievement of the ongoing continuing goal of radical reduction in mother-to-child transmission of HIV. The attitude that 1 affected child is too many must motivate continued and expanded implementation of antiretroviral chemoprophylaxis against perinatal HIV transmission. Acknowledgments We thank Drs. LeaVonne Pulley and Leslie Clark from the UAB Department of Health Behavior for their assistance in the development of the survey instrument. References 1. Stoto MA, Almario DA, McCormick MC, editors. Reducing the odds: preventing perinatal transmission of HIV in the
December 2002 Volume 324 Number 6
Nichols et al
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
United States. Washington (DC): Institute of Medicine, National Academy Press; 1998. p. 397. Available at: URL: http:// www.nap.edu/books/0309062861/html/index.html Conner EM, Sperling RS, Gelber R, et al. Reduction of maternal-infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994; 331:1173– 80. Anonymous. Recommendations of the U.S. Public Health Service Task Force on the use of zidovudine to reduce perinatal transmission of human immunodeficiency virus. MMWR Recomm Rep 1994;43:1–20. Anonymous. U.S. Public Health Service recommendations for human immunodeficiency virus counseling and voluntary testing for pregnant women. MMWR Recomm Rep 1995;44: 1–15. Centers for Disease Control and Prevention. Table 26, Estimated persons living with AIDS, by region of residence and year. In: HIV/AIDS surveillance report. 1999;11(2):37. Available at: URL: http://www.cdc.gov/hiv/stats/hasr1102.htm Anonymous. Public Health Service Task Force recommendations for the use of antiretroviral drugs in pregnant women infected with HIV-1 for maternal health and for reducing perinatal HIV-1 transmission in the United States. Centers for Disease Control and Prevention. MMWR Recomm Rep 1998;47:1–30. Matheson PB, Abrams EJ, Thomas PA, et al. Efficacy of antenatal zidovudine in reducing perinatal transmission of human immunodeficiency virus type 1. J Infect Dis 1995;172: 353– 8. Anonymous. Human immunodeficiency virus screening. Joint statement of the American Academy of Pediatrics and the American College of Obstetricians and Gynecologists. Pediatrics 1999;104:128. Shaffer N, Chuachoowong R, Mock PA, et al. Shortcourse zidovudine for perinatal HIV-1 transmission in Bangkok, Thailand: a randomised controlled trial. Lancet 1999;353:773– 80. Wiktor SZ, Ekpini E, Karon JM, et al. Short-course oral zidovudine for prevention of mother-to-child transmission of HIV-1 in Abidjan, Côte d’Ivoire: a randomised trial. Lancet 1999;353:781–5. Dabis F, Msellati P, Meda N, et al. 6-month efficacy, tolerance, and acceptability of a short regimen of oral zidovudine to reduce vertical transmission of HIV in breastfed children in Côte d’Ivoire and Burkina Faso: a double-blind placebo-controlled multicentre trial. Lancet 1999;353:786 – 92. Wade NA, Birkhead GS, Warren BL, et al. Abbreviated regimens of zidovudine prophylaxis and perinatal transmission of human immunodeficiency virus. N Engl J Med 1998; 339:1409 –14.
THE AMERICAN JOURNAL OF THE MEDICAL SCIENCES
13. Guay LA, Musoke P, Fleming T, et al. Intrapartum and neonatal single-dose nevirapine compared with zidovudine for prevention of mother-to-child transmission of HIV-1 in Kampala, Uganda: HIVNET 012 randomised trial. Lancet 1999;354:795– 802. 14. Centers for Disease Control and Prevention. Figure 5, Pediatric AIDS cases. In: HIV/AIDS surveillance report 2000; 12(2):26. Available at: URL: http://www.cdc.gov/hiv/stats/ hasr1202.htm 15. Centers for Disease Control and Prevention. Table 16, Pediatric HIV infection cases by exposure category and race/ ethnicity. In: HIV/AIDS surveillance report 1999;12(2):25. Available at: URL: http://www.cdc.gov/hiv/stats/hasr1202.htm 16. Centers for Disease Control and Prevention. Update on preventing perinatal transmission of HIV. Satellite broadcast, 1999 Apr 29. CDC Public Health Training Network (PHTN), CDC National Center for HIV, STD, and TB Prevention (NCHSTP), and the CDC National Prevention Information Network (NPIN). Transcript available at: URL: http:// www.cdcnpin.org/broadcast/past/1999/0429/Transcript.pdf 17. Thomas SB, Curran JW. Tuskegee: from science to conspiracy to metaphor. Am J Med Sci 1999;317:1– 4. 18. Montaner JSG, Reiss P, Cooper D, et al. A randomized, double-blind trial comparing combinations of nevirapine, didanosine, and zidovudine for HIV-infected patients. JAMA 1998;270:930 –7. 19. Stringer JSA, Vermund S. Prevention of mother-to-child transmission of HIV-1. Curr Opin Obstet Gynecol 1999;11: 427–34. 20. Rouse DJ, Owen J, Goldenberg RL, et al. Zidovudine for the prevention of vertical HIV transmission: a decision analytic approach. J Acquir Immun Defic Syndr Hum Retrovirol 1995;9:401–7. 21. Culnane M, Fowler M, Lee SS, et al. Lack of long-term effects of in utero exposure to zidovudine among uninfected children born to HIV-infected women. Pediatric AIDS Clinical Trials Group Protocol 219/076 Teams. JAMA 1999;281: 151–7. 22. Stringer JSA, Sinkala M, Rouse DJ, et al. Effect of nevirapine toxicity on choice of perinatal HIV prevention strategies. Am J Public Health 2002;92:365– 6. 23. Stringer JSA, Stringer EM, Phanuphak P, et al. Prevention of mother-to-child transmission of HIV in Thailand: physicians’ attitudes on zidovudine use, pregnancy termination, and willingness to provide care. J Acquir Immune Defic Syndr 1999;21:217–22. 24. Bhatta MP, Stringer JSA, Phanuphak P, et al. Motherto-child HIV transmission prevention in Thailand: physician zidovudine use and willingness to provide care. Int J STD AIDS, in press. 25. Royce RA, Walter EB, Fernandez MI, et al. Barriers to universal prenatal HIV testing in 4 US locations in 1997. Am J Public Health 2001;91:727–33.
309