Volume 59, Issue 7 pp. 2180-2183
EDITORIAL
Free Access

Do high-risk behavior deferrals work? How to make it better?

Suchitra Pandey

Corresponding Author

Suchitra Pandey

Department of Pathology, Stanford University, Palo Alto, California

Stanford Blood Center and Transfusion Service, Stanford Health Care, Palo Alto, California

Search for more papers by this author
Hua Shan

Hua Shan

Department of Pathology, Stanford University, Palo Alto, California

Stanford Blood Center and Transfusion Service, Stanford Health Care, Palo Alto, California

Search for more papers by this author
First published: 03 July 2019
Citations: 1

High-risk behavior deferrals are one of several mechanisms employed by blood centers to reduce transfusion-transmitted infection (TTI) risk. Predonation donor screening includes a donor history questionnaire (DHQ) interview before eligible donors undergo TTI screening tests. Even though the implementation of nucleic acid testing (NAT) has enhanced the sensitivity of donor TTI testing in some countries, a residual risk for TTI still remains. Most of the residual risk is due to false negative results from window period donations. Human errors and mutant/aberrant strain infections are also theoretically possible. Also contributing to the residual TTI risk are unknown or emerging infections for which routine donor testing may not be available. On the current AABB full length DHQ (Version 2.0 May 2016), 19/43 questions (44%) are related to the risks of TTIs that are currently being tested for as a part of routine donor screening.1 These questions and resulting deferrals are intended as an added layer of safety to identify donors at higher risk for TTI including those with recent infection. The level of residual risk varies for different countries and different TTIs. While the current estimated residual risk for major known TTIs including HIV, HBV, and HCV has been reduced to very low levels in many developed countries (e.g., 0.68 per 1 million donations in the US for HIV),2-5 higher residual risk exists in other countries which may not have implemented NAT and have higher population TTI incidence and/or prevalence. In Brazil, the estimated residual risk of HIV is 11.3 per 1 million donations, and even after NAT is implemented nationwide the residual risk is still expected to be 6.8 per 1 million.6, 7

It has been difficult to obtain evidence to support the effectiveness of high-risk behavior deferral in reducing TTI risk. One study using mathematical modeling estimated that there would be a four-fold increase of HIV window period donations in the US if all risk assessment questions were removed from donor screening.8 Many studies have shown that the prevalence of HIV, HCV, and HBV in the donor population is significantly lower than the general population.9, 10 However, the observed decreased prevalence among donors is thought to largely reflect donor self-deferral; it is difficult to separately assess the contribution from high-risk behavior deferral by the donor center. One American Red Cross study that evaluated eligible donors who returned to donate after temporary high-risk behavior deferral did not find a higher TTI marker rate compared to first-time/repeat donors, but the authors cautioned that there may have been a selection bias for returning donors.11 Subsequently, this group reported that hepatitis marker rates were higher among potential donors deferred for a risk of viral hepatitis or intravenous drug use compared with donors who were not deferred, but other high risk deferrals or markers could not be assessed fully due to the small sample size.12 Overall, there have been few studies evaluating the impact of specific donor deferrals on blood safety, and one recent review reported that the available evidence for deferral reasons was of low quality, and for 60% of the top 30 reasons for excluding donors, no evidence was found.13

The new report by Goncalez and colleagues14 in this issue of TRANSFUSION describes the largest study to date aiming to demonstrate the effectiveness of high-risk behavior deferral by comparing TTI marker rates in deferred individuals to those of eligible donors. Over a 7-month period, potential donors who were deferred for high-risk behavior by routine DHQ interview at four large Brazilian blood centers were recruited on-site for a study that involved completing an audio computer-assisted structured interview (ACASI) and providing a blood sample for TTI testing. During the study period, 22.3% (66,870/299,848) of all potential donors were deferred. Among the deferred individuals, 15.6% (10,453/66,870) were deferred for high-risk behavior. The study design of recruiting deferred potential donors on-site contributed to its recruitment success: of 4,860 eligible subjects who were approached, 4,013 (82.5%) consented to participate.

The Brazil Ministry of Health has protocols for donor screening based on the AABB DHQ and WHO guidelines for blood donor selection.1, 15 In Brazil, face-to-face interviews are used to review the DHQ with potential donors and assess donor eligibility. The four participating blood centers differ in how DHQ interviews are conducted, the contents of the DHQ, and donor eligibility criteria. This study used ACASI to further evaluate deferred individuals, standardizing the collection of high-risk information. High-risk behaviors disclosed included male-to-male sex (MSM), exchanging money or drugs for sex, injectable drug use, high-infection-risk sexual exposure (sex with anyone who has tested positive for HIV/HCV), sexual partner risk (e.g., sex with an IV drug user), multiple sexual partners (≥4 partners in the last 12 months), unprotected sex (with unknown or high-risk partner), test seeking, and other high-risk exposures (e.g., tattoos, piercing, medical procedure, blood exposure). This study found that the most common high-risk behavior deferrals in Brazil were other high-risk behaviors (75%), unprotected sex (48%), multiple sex partners (30%), and test seeking (21%). Less than 15% of deferred donors reported sexual partner risk (14.7%), MSM (11.5% of males), injectable drug use (8.4%), exchanging money or drugs for sex (7.6%), and high-infection-risk sexual exposure (2.2%).

Samples from all study participants were tested for TTI with the same tests used for routine donor screening. Of the 4,013 individuals deferred for high-risk behavior, 3.7% had a repeat reactive result for any TTI marker. A follow up sample for confirmatory testing was obtained from about 60% of participants. Using weighted regression modeling, authors estimated that 3.5% (142/4,013) of all individuals deferred for high-risk behavior would have confirmed positive for HIV, HCV, HBV, or syphilis if all donors had returned for confirmatory testing. More telling is the finding that HIV prevalence in individuals reporting high-risk behavior was approximately 3.75 times higher than in eligible first-time donors during the study period (0.35% vs. 0.092%) at the same blood centers. Syphilis was also significantly more prevalent among potential donors reporting high-risk behavior. While the rates of confirmed HBV and HCV infections between deferred individuals and eligible first-time donors were not substantially different, the significantly increased prevalence rates for HIV and syphilis provide evidence that high-risk behavior deferral serves the intended purpose of helping to screen out potential donors with high TTI risks at these blood centers.

The study also found that different high-risk behaviors are associated with different TTI prevalence rates. For HIV, individuals who reported high-infection-risk sexual exposure, MSM, or sexual partner risk had the highest frequency of confirmed HIV infection: 1.2%, 0.7%, and 0.7%, respectively. Another Brazilian study reported that in donors who confirmed positive for HIV, MSM was the sexual behavior most strongly associated with being HIV-positive following a high-infection-risk sexual exposure.7 This evidence supports maintaining a deferral period for donors with these risk behaviors in Brazil. On the other hand, some of the more common high-risk behaviors, such as unprotected sex, multiple sexual partners, test seeking, and other behaviors did not strongly correlate with increased HIV infection (confirmed positive infection frequency was ≤0.2%). In contrast, another Brazilian study showed an association between HIV infection and ≥4 sexual partners.16 The confirmed positive rate for syphilis among donors deferred for high-risk behavior was significantly higher than for eligible first-time donors, suggesting that donor screening questions are effective in identifying donors with syphilis.

Understanding the effectiveness of high-risk behavior deferral is important not only for blood safety but also to prevent unnecessary donor loss. In this study, 15.6% of deferrals were due to high-risk behavior, and 3.4% of all individuals who presented to donate were deferred for high-risk behavior. Data from six US blood centers showed that deferral for high-risk behavior (including exposures) occurred in 0.6% of donor visits and accounted for 4.5% of deferrals.17 Although the overall rate of high-risk behavior deferral may be low, studies have shown that donors who are temporarily deferred for high-risk behavior are unlikely to return. In one US study, only 6.5% to 9.5% of first-time donors who were temporarily deferred for high-risk behavior returned to donate.18 Thus, the use of high-risk behavior deferral to improve blood safety must be balanced with the potential short-term and long-term impact to the overall blood supply in a region. To maximize blood safety while minimizing unnecessary donor loss, both the sensitivity and specificity of high-risk behavior deferral need to be optimized. An optimized deferral process can be developed only after careful evaluation of several key elements. First, it must be pathogen-specific. It is critical to understand the unique epidemiological features and risk factor profiles for each pathogen. Take as an example the difference between HIV/syphilis and HBV/HCV from this study. Likely, the questions aimed at HBV/HCV risk in the current DHQ do not have enough sensitivity nor specificity in Brazil. Second, regional TTI epidemiological features are important to consider. For example, in some countries, such as Australia, where there is a highly MSM-focused epidemiology of the HIV epidemic, using ungendered high-risk behavioral criteria for donor selection may overestimate the risk in heterosexual donors and result in unnecessary deferrals.19 This is in contrast to Brazil, where heterosexual transmission of HIV also plays a significant role in the HIV epidemic and accounts for more than 60% of new HIV cases in the general population.20 Third, cultural context must be considered. In some cultures, direct questions pertaining to sexual behavior may be viewed as too intrusive and unacceptable, especially if potential donors do not feel that protection of confidentiality is guaranteed.

Improving blood safety goes beyond asking questions about high-risk behaviors, and the overall donor screening process must be evaluated. Goncalez et al pointed out the need for improved predonation educational material and educational programs on TTI prevention in Brazil given their finding of a large percentage of deferred heterosexual females who reported no condom use and low rates of being tested for HIV. In Brazil it has been noted that test-seeking is a major motivation for many potential blood donors, and the rate of acknowledged test seekers was 7% at one Brazilian blood center.21, 22 In the US, 19% of blood donors with sexually transmittable viral infections reported test seeking and in France, 56% of blood donors with HIV were test seeking.23, 24 Test seeking individuals are less likely to be truthful about their risk behaviors and need to be discouraged from joining the potential donor pool. Information about alternative venues of free and anonymous TTI testing should be emphasized to potential donors. Adding this information to traditional predonation materials may not be sufficient, and other educational methods should be evaluated.20, 25 Finally, creating a more trusting and confidential donor screening experience will encourage potential donors to report high-risk behaviors. Multiple studies evaluating CASI showed a higher incidence of donors reporting risky or “stigmatized” behaviors.26 In Brazil, 13% of eligible HIV-negative donors declared a deferring risk factor on a subsequent ACASI that was not disclosed during their face-to-face interview,20 and it was reported that approximately 50% of donors who confirmed positive for HIV in Sao Paulo, Brazil were aware of their HIV risk behaviors but did not disclose them during their face-to-face interview.27 The implementation of CASI for all donors in Brazil will likely lead to greater disclosure of behavioral risk factors as compared to face-to-face interview and improve the screening process.

The ultimate direct determination of the effectiveness of high-risk behavior deferral would require demonstrating the removal from the donor pool of TTI-infected individuals whose infections would otherwise be undetected by routine donor screening testing. This approach would require testing a huge number of high-risk individuals with tests that are not readily available such as individual NAT. By demonstrating significantly increased HIV and syphilis prevalence rates in potential donors deferred for high-risk behavior compared to first-time eligible donors, Goncalez et al provided indirect yet compelling evidence to support the effectiveness of high-risk behavior deferral in Brazil. It is reasonable to postulate that individuals deferred for high-risk behavior likely have higher rates of recent TTI infections and possible exposure to emerging TTI that are not currently routinely screened for, thus posing greater risk for being missed by the routine donor screening testing process. Especially for many parts of the world where high-quality, highly sensitive donor screening tests are not consistently available and for countries with high prevalence and incidence of TTI, high-risk behavior deferrals are likely to be an even more valuable tool in safeguarding the blood supply than in developed countries.

CONFLICTS OF INTEREST

The authors have disclosed no conflicts of interest.

    The full text of this article hosted at iucr.org is unavailable due to technical difficulties.