The effectiveness of HIV post-test counselling in determining healthcare-seeking behavior

To examine the relationship between HIV post-test counselling and healthcare-seeking behaviors among HIV-positive individuals, we conducted patient interviews and medical chart reviews on 100 HIV-positive patients at a public hospital. When performed, HIV post-test counselling effectively encourages HIV-positive individuals to seek medical care, especially when performed by trained personnel. The consistent use of post-test counselling may improve care for HIV-positive individuals and should be encouraged at all healthcare facilities that conduct HIV testing. Increasing the number of HIV-infected individuals who access care and prevention services is one of the goals of the recently published National HIV Prevention Strategic Plan [1]. To accomplish this goal HIV post-test counselling plays a critical role as it may encourage and facilitate entry into care for those newly diagnosed with HIV infection. In 1994 the Centers for Disease Control and Prevention published guidelines for HIV pre- and post-test counselling [2] that were revised in 2000 [3]. One objective of these documents and post-test counselling is to get more HIV-positive individuals into care [3]. However, since the guidelines were published there has been limited research to evaluate the effectiveness of counselling in producing the desired outcomes such as seeking medical care [3,4]. The purpose of this study was to evaluate whether post-test counselling for newly diagnosed HIV-positive individuals has an effect on healthcare-seeking behaviors. A total of 100 HIV-positive inpatients at a public hospital in Atlanta, Georgia, USA were interviewed in June and July 2000. The survey consisted of 38 open-ended questions and a medical chart review. The patients’ names were obtained from a list, compiled daily by social services, of all known HIV-positive patients admitted to the hospital. All consecutive patients were approached until a sample of 100 was obtained; the response rate was 90%. Written informed consent was obtained, and patients were compensated five US dollars. The Emory University Human Investigations Committee approved this study. The sample consisted of 100 HIV-positive individuals. Sample demographics are shown in Table 1. A total of 54% of the sample reported receiving post-test counselling. Of the 54 individuals who had counselling, 49 (91%) received medical referrals, 39 (72%) received prevention services referrals, and 39 (72%) received psychological/support services referrals. Furthermore, the majority of the sample reported following up with medical (86%), psychological (59%), and preventative care (77%) referrals. Most patients reported that counselling contained pertinent HIV information, including coping with HIV, the importance of taking medication correctly and practising safer sex, and patients found this information helpful. There were no statistically significant differences between individuals who did and did not receive counselling with respect to the following variables: race, sex, housing status, insurance status, place of HIV diagnosis, likelihood of ever taking HIV medication, likelihood of stopping medication, and ever having had an influenza or pneumococcal vaccine (all P ≥ 0.06). Although 50% of the sample was diagnosed before the Centers for Disease Control and Prevention guidelines were published in 1994, there were no differences in the likelihood of receiving counselling, or the content of counselling between those diagnosed before or after 1994.Table 1: Sample demographics. The median time to seek medical care after HIV diagnosis was 7 months earlier for those who received counselling (n = 47) than those who did not (n = 39) (0 versus 7 months, P = 0.009). The median time to seek medical care after HIV testing was 7 months earlier for those who were offered medical referrals (n = 43) compared with those who were not (n = 43) (0 versus 7 months, P = 0.01). Individuals who received counselling were 2.7 [95% confidence interval (CI) 1.04, 7.43] times more likely to seek care within 3 months of HIV diagnosis than those who did not receive counselling (P = 0.023). Of the 54 individuals who received counselling, 30 (56%) were counselled by a post-test counsellor or social worker, and 14 (26%) were counselled by a physician or nurse. Individuals counselled by a post-test counsellor or social worker were 11.6 (95% CI 1.0, 314.2) times more likely to follow-up with medical care than those counselled by a physician or nurse (P = 0.029) and were 8.7 (95% CI 1.6, 53.1) times more likely to take antiretroviral therapy (P = 0.004). In examining the percentage of individuals who received counselling over time, we did not find an increase in the number of HIV-positive patients who received counselling when comparing three time periods. Among patients diagnosed in the past 2 years 62.5% received post-test counselling, among those diagnosed between 2 and 10 years ago, 58% received counselling, and among those diagnosed more than 10 years ago 38% received counselling (P = 0.08 for trend). Our data suggest that post-test counselling and medical referrals effectively encourage newly diagnosed HIV-positive individuals to seek medical care. Therefore, post-test counselling for HIV-infected individuals succeeds in its mission; post-test counselling can aid in getting HIV-positive individuals into care. This is increasingly important in order to allow patients to benefit from available medical care, which can decrease morbidity and mortality rates [5], and also serves as a prevention strategy [6]. However, it is also clear that there is room for improving HIV post-test counselling for those newly diagnosed with HIV infection. In addition, physicians and nurses appear to be less effective in conveying effective post-test counselling. This is particularly important because now the majority of HIV testing occurs in various healthcare settings, not at publicly funded testing sites [7], but medical professionals are ill-prepared for HIV counselling [8]. The revised guidelines for HIV counselling, testing, and referrals acknowledges this trend [3], but appropriate training in counselling or the availability of post-test counselling services is lacking at many healthcare facilities. There are strengths and limitations associated with this research design. One primary limitation is recall bias. Although most patients vividly recalled their HIV testing and counselling experiences, some of the participants were very ill, which may hinder their ability accurately to remember the past. In addition, this sampling design was not random; therefore the counselling results observed with this sample may not apply to all HIV-positive individuals. A prospective study that tracks newly diagnosed individuals from counselling and testing through medical care and treatment is necessary to evaluate fully the effectiveness of counselling. However, the variety of HIV testing and counselling experiences in this sample is one of our strengths. In conclusion, HIV counselling is cost effective even for busy clinics [9–12], and is efficient in encouraging harm reduction and encouraging patients to act upon referrals to medical care. However, although counselling is effective it is not routine or standardized. An increased emphasis on the consistent use of counselling, rather than changing the current standard of care, may thus be sufficient to improve care for HIV-positive individuals. It is imperative that counselling and referral services be expanded to a variety of settings in both the public and private sectors, and that individuals at every testing site are trained to conduct effective counselling sessions. HIV post-test counselling works, it just needs to be done. Michelle R. Eichler Susan M. Ray Carlos del Rio Acknowledgements The Emory University Human Investigations Committee approved this study, and written informed consent was obtained from all participants. The authors would like to thank Christopher Summers-Bean and Jennifer Shevlin for their assistance in designing the study and analysing the results.