Abstract
An HIV transmission prevention program incorporating universal voluntary counseling and testing (UVCT) was conducted in a general practice (GP) clinic of a Thai hospital. Of the 494 participating patients, 356 (72%) accepted HIV-UVCT. Independent factors associated with HIV-UVCT acceptance included participating in the program after office hours (4-8
Introduction
HIV infection and AIDS have been major public health problems in Thailand. Among Thai adults aged 15 to 49 years, the prevalence of HIV infection was estimated to be 1.2%, and in 2011, 9700 individuals were newly diagnosed with HIV infection (15 per 100 000 persons). 1 The Ministry of Public Health of Thailand reported that the rates of new HIV infection were 31.9% among injection drug users, 3.2% among commercial sex workers, and 1.9% among male attendees of sexually transmitted infection (STI) clinics, while the rate among general Thai population was 0.2% in 2011. 2 The annual rates of new HIV infection have not declined dramatically despite efforts and several campaigns to prevent HIV transmission in the country. 1 One of the possible reasons is thought to be the unawareness of HIV status among asymptomatic infected individuals who engage in risk behaviors and fuel the ongoing HIV transmission in the communities. 3
HIV universal testing has been recommended by the Centers for Disease Control and Prevention (CDC) of the United States since 2006 to increase the awareness of HIV status in the U.S. communities. 4 The recommendations state that universal HIV testing should be offered to everyone aged 13 to 64 years who present to health care facilities regardless of their complaining and presenting symptoms or HIV risks in areas where HIV prevalence is 0.1% or more. The universal HIV testing provides advantages of early HIV detection, including early engagement of the infected patients into care to prevent subsequent immunosuppression and HIV-related morbidity and mortality and increase in awareness of HIV status, which results in actions to prevent HIV transmission. 3 –6 In addition, assessment of risk perception and targeted education to improve risk perception and reduce risk behaviors can be incorporated into this HIV testing program. 1
In Thailand, HIV infection screening and diagnosis are usually done by a standard blood test for HIV antibody. All Thai people are allowed to have this test performed free of charge every 6 months as a part of a campaign conducted by the Ministry of Public Health to reduce new HIV infection and transmission. Pretest and posttest counseling and written consent forms are mandatory for the HIV testing in accordance with Thai laws. Generally, population groups at high risk for HIV infection include men who have sex with men, commercial sex workers, injection drug users, and attendees of STI clinics 7 and are targets for this HIV universal voluntary counseling and testing (HIV-UVCT) in Thailand. However, there has not been any official recommendation about how to implement the HIV-UVCT in settings with populations at lower HIV risk, such as general outpatient clinics or emergency departments. We conducted this study to evaluate the feasibility of the in-house HIV transmission prevention program that incorporates HIV-UVCT among patients visiting a general practice (GP) clinic in Thailand.
Methods
Study Population, Setting, and Design
An HIV transmission prevention program incorporating HIV-UVCT was conducted in a GP clinic of Thammasat University Hospital (TUH) in Pathumthani, Thailand, during the period from June 1, 2012, to July 31, 2012. The GP clinic provides primary care services to adult patients (age ≥15 years) in Pathumthani and the nearby provinces. The HIV transmission prevention program was initiated by the HIV care team of TUH, which consists of 3 infectious disease physicians, 2 HIV nurses, 2 nonphysician medical assistants, and 2 HIV-infected volunteers, to promote HIV transmission prevention and provide relevant education for patients visiting GP clinic. The program was set up specifically and not included in the regular patient care at the GP clinic. Due to the limited resources and staff of the GP clinic to do HIV-UVCT, the HIV care team worked on the program independently of the GP clinic staff and as an extra service. This study was a prospective study conducted among GP clinic patients who participated in this HIV transmission prevention program and was approved by the Faculty of Medicine, Thammasat University Ethics Committee.
Study Procedures
The HIV transmission prevention program was conducted in the GP clinic from 8:00
Study Definitions
Participants self-identified their sexual orientation and HIV risk perception during the counseling. To identify their own HIV risks, the participants answered “No risk at all,” “A little risk (low risk),” “More than a little (moderate risk),” and “A lot of risk (high risk)” to the HIV counselors. The investigators of this study subsequently evaluated the participants’ risk as “low risk,” “moderate risk,” and “high risk” based on the prespecified risk characteristics and behaviors reported in the counseling form and according to previous studies’ criteria. 8,9 This risk categorization tool was validated in the previous studies for use in differentiating participants with different levels of risk behavior. 8,9 Participants who were categorized as moderate or high risk but perceived their risks as no or low risk were classified as having low HIV risk perception.
HIV Testing and Result Notification
The HIV counselors obtained the participants’ consent for HIV testing in written documentation. Participants’ blood was collected at the GP clinic and was sent to the TUH serology laboratory. HIV testing was initially performed using ARCHITECT HIV Ag/Ab Combo (Abbott, Germany), which is a chemiluminescent microparticle immunoassay. If the HIV antibody was detected by this test, it was then confirmed by the other 2 HIV antibody detection assays, Determine HIV-1/2 (Alere Medical, Japan), which is an immunochromatographic test, and SERODIA HIV-1/2 (FUJIREBIO, Japan), which is a passive particle agglutination test. If all 3 assays indicated a positive result, the test result would be reported as HIV positive. Generally, it takes about 24 to 72 hours before the official result comes out. Notification of the test results was done via telephone. The HIV counselors called all participants to inform the negative results. However, if the test result was positive, they would inform the participants via telephone to come to TUH for posttest counseling and result notification, in accordance with Thai laws. A total of 3 calls (1 week apart) were attempted to notify the participants of their results, and additional 3-week duration was allowed after that for the participants to call back before contact failure was considered. Only the participants who could identify the identification codes correctly were informed of the negative result or were requested to come to TUH. Among participants with negative HIV testing, further HIV transmission prevention strategies were recommended over the phone along with posttest counseling, while HIV-infected participants were inquired about the plans for their HIV continuity care in regard to medical coverage and chosen health care facilities to establish care. HIV transmission prevention and health maintenance strategies were recommended at the end of the posttest counseling.
Data Analyses
All statistical analyses were performed using SPSS version 15.0 (SPSS, Chicago, Illinois). Categorical variables were compared using Pearson χ 2 or Fisher exact test as appropriate. Continuous variables were compared using Mann-Whitney U test. All P values were 2 tailed; P values less than .05 were considered statistically significant. Variables that were present at a significance level of P < .20 in univariate analysis or had prior significance for HIV testing acceptance (age, sex, level of education, and marital status), 10,11 HIV positivity (age, sexual orientation, and risk behaviors), 7,12 and low HIV risk perception (age, sex, sexual orientation, and education) 13 –15 were entered into logistic regression models. These variables were subsequently removed from the models in backward stepwise fashion if their P values were >.05 until the final model had reached. Adjusted odds ratios and 95% confidence intervals were determined for risk factors associated with HIV testing acceptance, HIV positivity, and low HIV risk perception.
Results
Characteristic of the Study Participants
A total of 821 patients attending the GP clinic were approached, of which 494 (60%) patients participated in the HIV transmission prevention program. The median age was 34 years (range 15-68 years). Three hundred seven (62%) patients participated in the program during office hours (8
Characteristics of the Study Participants.a
Abbreviatons: UVCT, universal voluntary counseling and testing.
a Data are in numbers (%) unless otherwise indicated.
b Comparison between participants accepted and declined universal voluntary counseling and testing (UVCT).
c Any kind of infections including HIV infection. *significant P values (<0.05).
Reasons and Characteristics Associated with Declining or Accepting HIV-UVCT
A total of 138 (28%) patients participated in the HIV knowledge assessment questionnaire and education session but declined to undergo HIV-UVCT. Of these 138 participants, 60 (44%) did not provide the reason for HIV-UVCT declining. The 3 most common reasons provided by the remaining 78 participants were time constraint (30 [38%]), perceiving self as no risk for HIV infection (27 [35%]), and having HIV test with negative result within a year (14 [18%]). Other reasons included being afraid of pain or of needles (5%), being healthy (3%), and being afraid of having a positive HIV test result (1%). When comparing between participants who accepted and declined HIV-UVCT (Table 1), the declining participants were more likely to participate in the program during office hours, be a merchant, have a college or a bachelor degree education, have higher monthly household income, and present to the GP clinic for regular visit for health problems and health checkup. Among the 356 HIV-UVCT accepting participants, the reasons for undergoing testing included desire to know HIV infection status (321 [90%]), suspect of having HIV infection (11 [3%]), preparing for marriage and pregnancy (10 (3%)], interest in free test (8 [2%]), following friends who want to be tested (4 [1%]), and recent high-risk exposure (2 [0.6%]). By multivariable logistic regression analysis (Table 2), characteristics independently associated with HIV-UVCT acceptance included participating in program during nonoffice hours (P < .001), living with a domestic partner without marriage (P = .01), and having primary school education or less (P = .02).
Multivariable Logistic Regression Analysis for Factors Associated with HIV-UVCT Acceptance, HIV-Positive Test Result, and Low HIV Risk Perception among the Study Participants.
Abbreviation: UVCT, universal voluntary counseling and testing.
Characteristics and HIV Risks among Participants Undergoing HIV Test
Of the 356 participants undergoing HIV-UVCT, 254 (71%) were self-willing to be tested, while 102 (29%) were advised by accompanying friends or family members before they decided to get tested, 118 (33%) had prior HIV test, all of which had negative results and 344 (97%) were heterosexual. The majority of the 356 participants reported having vaginal sex (299 [84%]) while having oral sex and anal sex were reported in 78 (22%) and 44 (12%), respectively. The reported rates of consistent condom use were all low with each type of sex (13% with vaginal sex, 9% with oral sex, and 5% with anal sex). Other reported risks for HIV acquisition included having tattoo or piercing (24%), having STIs within the past year (3%), having sexual partner who exchanged sex for money or drugs within the past 30 days (1%), injection drug use (0.3%), and having sexual partner who had STIs within the past year (0.3%). Based on the risk categorization tool, 80 (23%) participants had moderate to high HIV risk. Of these 356 participants, 345 (97%) were reachable for HIV result notification within the median time of 5 days (range 1-36 days), while 11 (3%) had contact failure. Among the 11 participants with contact failure, the median age was 29 years (range 15-67 years), 6 (55%) were male, 6 (55%) were married, 7 (64%) had secondary school education or less, 8 (73%) participated in the program during nonoffice hours, 11 (100%) came to the GP clinic for health checkup, 1 (9%) had prior HIV test, 11 (100%) were heterosexual, 3 (27%) were categorized into HIV high-risk group, and 1 (9%) had low perception of HIV risk.
Characteristics Associated with HIV Positivity
Among the 356 participants undergoing HIV test, 4 (1.1%) were HIV positive. The diagnosis of HIV infection was new for these 4 participants. Of these, 4 (100%) of the newly HIV diagnosed participants were male, 3 (75%) were homosexual; potential route of HIV acquisition included sexual contact 3 (75%) and tattooing and piercing 2 (50%; 1 person can have multiple potential routes of acquisition). Three (75%) had moderate to high HIV risk and 2 (50%) had low-risk perception. All of the 4 HIV-infected participants were contacted and informed the result within the median time of 7 days (range 3-14 days). Comparing between HIV-positive and HIV-negative participants (Table 3), HIV-positive participants were more likely to be homosexual, have anal and oral sex, have moderate to high HIV risk, and be uncertain about HIV risks in sexual partner. By multivariable logistic regression analysis, characteristics independently associated with HIV positivity were being homosexual men (P < .001; Table 2).
Characteristics of the 356 Participants Undergoing HIV Universal Voluntary Counseling and Testing Categorized by the Test Result.a
Abbreviation: STI, sexually transmitted infection.
a Data are in numbers (%) unless otherwise indicated.
b Comparison between participants with HIV-positive and HIV-negative test results.
HIV Risk Perception and Characteristics Associated with Low Risk Perception
Among the 356 participants undergoing HIV-UVCT, 36 (10%) had low HIV risk perception. Comparing between participants with low risk perception (having moderate or high risk for HIV but perceiving self as no or low risk) and correct risk perception, participants with low risk perception were more likely to be men, younger, homosexual or bisexual, uncertain about HIV risks in sexual partners and have STIs within the past year (Table 4). By multivariable logistic regression analysis, having moderate to high HIV risk and male sex were independently associated with low HIV risk perception (Table 2).
Characteristics of the 356 Participants Undergoing HIV Universal Voluntary Counseling and Testing Categorized by their HIV Risk Perception.a
Abbreviation: STI, sexually transmitted infection.
a Data are in numbers (%) unless otherwise indicated.
b Comparison between participants with low and correct HIV risk perception.
Discussion
Our study demonstrates feasible implementation of HIV-UVCT that was incorporated into an HIV transmission prevention program in a GP clinic of a Thai university hospital with the high rate of acceptance (72%). Given that the program was conducted by the hospital’s HIV care team and utilized its laboratory resource and the HIV antibody tests were totally supported by the Ministry of Public Health of Thailand, the extra cost incurred from the program was modest.
Health care provider-initiated HIV-UVCT has been recommended in the 2006 CDC guidelines to increase access to HIV testing. 4 However, several barriers including insufficient time, burdensome consent process, lack of knowledge/training, lack of patient acceptance, pretest counseling requirements, competing priorities, and inadequate reimbursement limit effectiveness of this testing approach. 16 Setting up the extra HIV transmission prevention program like ours in a GP clinic has some advantages over the HIV-UVCT offering by GP physicians. First, the program provides specific knowledge-guided education along with detailed HIV risk assessment and counseling. This type of education could shorten the time used for pretest counseling for about 10 minutes and made the counseling more efficient. Second, HIV care staff in the program has more dedicated time to talk, discuss, and educate the participants about HIV-UVCT and prevention strategies compared to a GP physician. Finally, the active approach of the program can potentially increase the likelihood of program participation and HIV-UVCT acceptance as evident in our study.
Notification of the HIV test result is critical for further counseling and engaging new HIV-infected individuals into care. 4 However, previous studies demonstrated that 10% to 55% of individuals at risk for HIV infection did not return for HIV results after the standard HIV testing. 17,18 HIV rapid testing that uses blood-based tests with or without oral fluid-based test has thus become an ideal strategy that can inform the result within about an hour. 19 A randomized study from the United States revealed that a significant higher proportion of the participants receiving HIV rapid tests were informed about their status than were those receiving standard HIV tests (95% versus 43%). 20 HIV rapid testing was shown to be associated with high rate (97%) of linkage to care and successful virological suppression within the median time of 8 months after HIV diagnosis in another study. 21 Based on a model of HIV transmission, increased HIV testing will modestly reduce HIV incidence among gay men in Australia and would be acceptable if HIV testing becomes convenient. 22 Nonetheless, there have not been randomized trials comparing between rapid and standard HIV tests in regard to linkage to care, treatment outcomes, and HIV transmission reduction. Two studies have shown that the incidences of posttest risk behaviors among participants undergoing both types of test were not different. 23,24 In addition, cost and availability may limit utility of the rapid tests in resource-constrained settings. We demonstrated that use of the standard conventional blood-based tests in this study was associated with high rate of acceptance (72%). The rates of result notification were 97% for all participants and 100% for HIV-infected participants. The difference in the rates of result notification in our study and the previous study was due to the use of HIV counselor-initiated confidential telephone contact in our study compared to having the participants calling in for follow-up appointments to be notified of the results in another study. 20 Altogether, HIV rapid testing provides some benefits over the standard testing and is recommended for use. However, in settings where the rapid testing is not available, the standard testing can be used.
In a general outpatient setting, common reasons for declining HIV-UVCT include having a previous/recent HIV test, perceiving oneself as no risk for HIV, and unwillingness to know the status. 20,25 Our results concurred with these findings but also identified time constraint as the most common reported reason. In addition, we found that participation in the program during nonoffice hours was independently associated with HIV-UVCT acceptance. These results suggest that operating time of the program in relation to availability of the participants is an important factor to be considered to improve the HIV-UVCT program. Participants who had recent HIV testing or declined HIV testing should be informed about the benefits of regular testing when risks are present and frequency of testing they can receive based on the national policies and medical coverage. During the HIV-UVCT offering, assessment of risk perception could help identify additional participants who should receive HIV-UVCT. It should be noted that only a few or none of the participants declined HIV-UVCT because of being afraid of pain or of needles in this study and previous studies. 20,25 This suggests that blood-based tests remain feasible and acceptable in most settings especially when standard, non-blood-based and painless tests are not available.
In high-risk population, HIV-UVCT acceptance was reported to be associated with injection drug use, not having steady partner, currently having STIs, and history of STIs, while in the low-risk general outpatient setting, younger age and low education were significantly associated with HIV-UVCT acceptance. 25,26 Our study results revealed that living with domestic partner with no marriage and low education were independent factors associated with HIV-UVCT acceptance. These may reflect the participants’ uncertainty about behaviors and HIV status of their unmarried partners. However, the reason why less-educated participants were more likely to accept HIV-UVCT than were well-educated participants was unclear and requires further studies.
Our findings indicated that being homosexual men was independently associated with HIV positivity. This corresponds with the reported high incidence of HIV infection in this population in Thailand. 7,27 In our program, assessment of risk perception was performed during the pretest counseling. Although only 10% of the participants in this study had low HIV risk perception, which was significantly less than those in a high-risk population (84%), 8 the group that risk perception should be assessed was moderate or high-risk men. These participants should be informed about their actual risks and educated on behavior modification and HIV prevention during the HIV-UVCT. In a GP clinic setting, persons at risk for low HIV risk perception may be identified through a self-assessment questionnaire that can be completed while they are waiting to see a doctor in the clinic. A screening nurse in the clinic can use the questionnaire to guide further educational interventions for each individual which are provided by available health care workers in the clinic or an HIV care team of the hospital. Studies are needed to determine a campaign that can effectively establish correct awareness of HIV risk and to create skills for self-assessment of HIV risk in the community.
There were limitations in this study. First, risks and risk behaviors of participants declining HIV-UVCT were not collected in this study and might be different from the accepting participants. Second, the results may have limited generalizability, given that our study population consisted of primarily heterosexuals with low to moderate income in Thailand. Finally, the face-to-face interview with the HIV counselors might impact the disclosure of HIV risks and risk behaviors of the participants. However, this limitation was minimized because the pretest counseling was done in a private room and the HIV counselors were well trained in conversations to make the participants trust them and open up to tell the truth. In addition, the counselors expressed their willingness to educate and support the participants.
In conclusion, implementation of HIV-UVCT as a part of HIV transmission prevention program in our GP clinic was feasible with high acceptance rate. The strategies used in this study were able to identify a number of new HIV-infected individuals among the generally low-risk population. Reasons for and characteristics associated with accepting or declining HIV-UVCT should be considered to improve the HIV-UVCT program. Opening an HIV testing facility during nonoffice hours and using the knowledge-based education strategy could potentially attract more individuals accepting HIV testing and shorten the time spent on the process, respectively. HIV risks and risk behavior assessment and emphasis on consistent condom use are important and should be incorporated into the HIV-UCVT program. Health care system strengthening and policies to make rapid HIV testing and/or point-of-care CD4 testing available along with educational interventions in both health care and community settings are needed to increase access to HIV testing and successful linkage to care and treatment. Given the advantages of early HIV diagnosis, future research is needed to determine appropriate strategies to implement HIV-UVCT in emergency departments, private physician offices, and settings outside health care facilities, such as entertainment venues, colleges, government offices, company offices, and factories in HIV-endemic resource-limited countries.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Pathumthani Province Public Health Office, Pathumthani, Thailand.