Factors associated with failure to return for HIV test results in a free and anonymous screening centre

Abstract

Free and anonymous screening centres (CDAG: Centres de Depistage Anononyme et Gratuit) are public facilities set up for HIV infection diagnosis in France. Some people visiting CDAG fail to return for test results and are not informed of their serology. This study aimed to assess factors associated with failure to return for HIV test results. Patients visiting the Fernand-Widal CDAG (Paris) for an HIV test in January–February 2011 were eligible to take part in the study. Data were collected with an anonymous self-administered questionnaire. Factors associated with failure to return were assessed using logistic regression models. Of the 710 participants (participation rate 88%), 46 patients failed to return. Not specifying birthplace and not living in the region of Paris were associated with failure to return. Those who perceived no risk of HIV infection and those who felt they were more at risk than other people were both statistically associated with failure to return. Self-perceived risk seemed to be of chief concern for failure to return for HIV test results and should be considered during pre-test counselling.

Keywords

HIV screening secondary prevention sexual risk behaviours AIDS serodiagnosis health behaviour failure to return results

Introduction

Throughout France, since 1988, free and anonymous screening centres (CDAG: Centres de Dépistage Anonyme et Gratuit) have been set up for the diagnosis of human immunodeficiency virus (HIV) infections.¹ Thus, French people have two main ways of being screened for HIV: they may see their usual doctor, receive an HIV testing prescription and go to a medical laboratory, or they may be tested anonymously in a CDAG. Participation of the CDAG in all HIV screening activity in 2006 was estimated to be 8% in France but reached 18% in the Paris region.² Differences between Paris and other regions may be explained by population characteristics. Younger and more at-risk people attend the CDAG, particularly men who have sex with men (MSM) and Sub-Saharan African migrants.² These subpopulations are over-represented in the Parisian population, which is younger, with a greater proportion of MSM and Sub-Saharan migrants than the rest of France. In CDAG, patients meet a physician who assesses the risks the person takes and gives prevention advice at two different times: pre-test counselling (before blood sampling) and post-test counselling (the day when test results are given to the patient). Pre- and post-test counselling is part of CDAG activity, unlike other facilities. CDAG patients must return in person to obtain their test results. To preserve anonymity, no contact details (phone number, email address, etc.) are collected. The identification information is limited to the patient’s anonymity number that is written on the card given to him or her during the first visit and which the patient has to present in order to receive the test results.

However, not all patients return for their test results, and thus receive no post-test counselling. This failure to return (FTR) for test results represents a loss of effectiveness of CDAG’s preventive approach. Notably, HIV-positive patients who fail to return are not aware of their positive serology. Consequently, they are not given medical attention and cannot begin early antiretroviral therapy. Moreover, because of their lack of awareness of their own HIV status, they may not change their at-risk behaviours³ and may continue to transmit the virus.

Some studies have explored factors associated with FTR for HIV test results, but with mixed results. For example, some Australian studies found that women failed to return for their test results more often than men,^4,5 whereas some US studies found the opposite, i.e. that men failed more often to return.^6,7 There are studies that suggest that perceived risk has an important impact on FTR, but this seems to vary with geographical location. In an African study in Tanzania, FTR was associated with a lower perceived risk of infection,⁸ whereas in another study conducted in the US, it was associated with a higher perceived risk.⁹ Some studies showed that FTR was associated with less education and lack of information about HIV infection.^8,10,11 In regions where financial or geographic barriers, or social stigma hamper access to health care, FTR could be higher. Conversely, availability of the most up-to-date treatment may explain in part a lower FTR rate. Repeating studies on FTR may help to identify factors impacting HIV test result attendance. It was found that awareness of one’s own infection may help to change his or her behaviour and reduce HIV spread.³ To our knowledge, there is no published study conducted in France until now on factors associated with FTR for HIV test results. This study aimed to assess factors associated with FTR for HIV test results in a free and anonymous screening centre (CDAG) in Paris.

Methods

Setting

The study was conducted in one of the 11 CDAG of Paris, in Fernand-Widal hospital, which is located in the northern part of Paris. The Paris region has the highest rate of confirmed positive HIV screening tests in France (0.4% vs. 0.2% in France).¹² In the Fernand-Widal CDAG, HIV testing activity is performed by six physicians on five afternoons and one morning a week. HIV diagnosis is based on antibody tests (enzyme-linked immunosorbent assay and Western blot) and the p24 antigen test, and results are available three days after the blood sampling. Data on test results are routinely and anonymously registered in the CDAG consultations database, including number of visits, yearly rates of HIV-positive tests and of FTR. About 5000 patients visit the CDAG each year, and 15% of them fail to return for HIV test results. Based on these results, a two-month study period was chosen, in order to include at least 100 patients who failed to return for HIV test results.

Study population and data collection

The study population was restricted to new patients visiting the Fernand-Widal CDAG for an HIV screening test between 1 January 2011 and 28 February 2011. Thus, patients who were not prescribed an HIV test, who came for a confirmation test, or for post-exposure prophylaxis were not included in the present study.

During the two-month study period, an anonymous self-administered questionnaire on risk-taking developed by the CDAG physicians was given to all new patients visiting CDAG before the consultation. Patients were informed that refusing to answer the questionnaire would not affect care. Those who agreed to complete it did so during the waiting time before seeing the doctor and then handed the questionnaire to him. Each questionnaire was identified with the patient’s anonymity number.

Data

Information collected in the questionnaires was matched to the CDAG consultation database using the anonymity number. The questionnaire included socio-demographic factors (age, sex, place of residence and birth, educational level, health insurance coverage), sexual behaviour (sexual orientation, number of sex partners), injecting drug use (IDU), reason for visiting and self-perceived risk of HIV infection. Health insurance coverage was categorised as follows: none, statutory health insurance with or without complementary voluntary health insurance and other coverage linked to low income or illegal immigration (universal health coverage, complementary universal health coverage or state medical help).¹³ Quantitative data were categorised on the basis of their distribution in the sample, or according to the categorisations used in previous studies: patient age (younger or older than 30 years of age),^6,14 number of sex partners during the last 12 months (fewer than two; two to five, more than five or unknown). Reason for visiting was categorised as follows: clinical signs, risk-taking, check-up as part of a relationship (beginning or end of a relationship), others or unknown.

Statistical analysis

FTR for HIV test results was defined as not returning for test results within 60 days after the blood sampling. FTR and no-FTR patients’ characteristics were compared using the χ² test or Fisher’s exact test. Univariate and multivariate analyses were conducted to assess factors associated with FTR using logistic regression models. The following factors were included in the multivariate model: age, sexual orientation, factors associated with FTR in the literature (educational level, result of HIV testing, perceived risk of HIV infection) and variables associated with FTR in univariate analysis with a p value <0.20. Statistical analysis was performed using STATA/SE 11.0 (Stata Corporation, College Station, TX, USA).

Results

Of the 805 eligible patients who visited the CDAG during the study period for an HIV test, 710 patients participated in the study (participation rate 88%), of whom 46 did not return for their test results (FTR rate 6.5%) within 60 days after blood sampling. FTR rate did not differ significantly between participants and non-participants (6.5% vs. 4.2%, p = 0.39).

Characteristics of the study population are presented in Table 1. The median age was 27 years (min–max: 15–65 years) and 54.8% of participants were men. MSM represented 10% of the total study population. Nine percent of participants were covered by health insurance that was linked to low income or illegal immigration; 3% had no health insurance coverage. Nine percent of participants were born in Sub-Saharan Africa, two-thirds in metropolitan France. Insurance coverage was closely linked to birthplace: 59% of people declaring health insurance coverage linked to low income or illegal immigration, and 78% of those declaring no health insurance were born in Overseas France or abroad (data not shown). Half of the participants were living in Paris. About 1% had used injecting drugs at least once. Among the five HIV-positive patients, there were two women, two heterosexual men and one MSM; three of them were born abroad (data not shown).

Table 1.

Baseline characteristics of study population (n = 710).

Characteristics	Distribution of study population (%)	n
Age (years)
<25	36.6	260
25–29	23.4	166
30–34	17.0	121
≥35	23.0	163
Sexual orientation
MSM (men who have sex with men)^a	10.1	72
Heterosexual men	44.6	317
Women	45.2	321
Health insurance coverage
Statutory health insurance	72.7	516
Other insurance^b	8.6	61
None	3.2	23
Missing	15.5	110
Place of residence
Paris	52.1	370
Paris region	33.4	237
Outside of the Paris region	2.5	18
Missing	12.0	85
Birthplace
Metropolitan France	65.8	467
Overseas France	2.7	19
Sub-Saharan Africa	9.0	64
North Africa	3.9	28
Other regions	7.7	55
Abroad (without further indication)	4.5	32
Missing	6.3	45
IDU (injecting drug use)
At least once in life	0.6	4
Regular use	0.3	2
Never	90.9	645
Missing	8.3	59

including one transgender patient, with current male phenotypic sex at the visit.

linked to low income or illegal immigration.

Table 2 presents factors associated with FTR. The proportion of FTR varied according to population characteristics, from 2.5% for patients reporting six sex partners or more during the last 12 months to 22.2% for patients living outside of the Paris region. The FTR rate reached 20.0% among HIV-positive patients and 16.7% among IDU patients.

Table 2.

Factors associated with FTR for HIV test results (n = 710).

Variable	n of FTR/N of participants		Univariate analysis (n = 710)			Multivariate analysis (n = 710)
Variable	n of FTR/N of participants	FTR (%)	OR	95% CI	p	Adjusted OR	95% CI	p
Age					0.619			0.916
Younger than 30	26/426	6.1	1.00			1.00
30 or older	20/284	7.0	1.05	0.86–1.29		0.99	0.79–1.24
Education					0.180			0.556
Graduated high school or more	33/582	5.7	1.00			1.00
Not graduated high school	8/76	10.5	1.96	0.87–4.41		1.66	0.66–4.12
Unknown^a	5/52	9.6	1.77	0.66–4.75		1.17	0.34–4.09
Birthplace					0.023			0.073
Metropolitan France	24/467	5.1	1.00			1.00
Overseas France or abroad	15/199	7.5	1.50	0.77–2.93		1.72	0.79–3.74
Unknown^a	7/44	15.9	3.49	1.41–8.64		3.44	1.15–10.30
Place of residence					0.004			0.006
Paris region	32/607	5.3	1.00			1.00
Outside of the Paris region	4/18	22.2	5.13	1.60–16.49		6.87	1.93–24.48
Unknown^a	10/85	11.8	2.40	1.13–5.07		1.88	0.77–4.62
Sexual orientation					0.434			0.190
Heterosexual men	16/317	5.0	1.00			1.00
MSM (Men who have Sex with Men)^b	4/72	5.6	1.11	0.36–3.41		1.42	0.39–5.15
Women	26/321	8.1	1.66	0.87–3.15		1.96	0.95–4.02
Number of sex partners last 12 months					0.045			0.035
0–1	15/127	11.8	1.00			1.00
2–5	15/285	5.3	0.41	0.20–0.88		0.45	0.20–1.00
6 and over	2/81	2.5	0.19	0.04–0.85		0.16	0.03–0.87
Unknown^a	14/217	6.5	0.51	0.24–1.11		0.34	0.14–0.83
IDU (Injecting Drug Use) ^c					0.298			–
Never	39/645	6.1	1.00			–	–
At least once in life	1/6	16.7	3.11	0.35–27.25		–	–
Unknown^a	6/59	10.2	1.76	0.71–4.34		–	–
HIV test result					0.285			0.291
HIV-negative	45/705	6.4	1.00			1.00
HIV-positive	1/5	20.0	3.67	0.40–33.5		3.51	0.29–42.41
Reported reason for visit					0.194			0.045
Check-up as part of a relationship^d	11/224	4.9	1.00			1.00
Clinical symptoms	3/18	16.7	3.87	0.97–15.4		8.42	1.86–38.22
Risk-taking	24/381	6.3	1.30	0.63–2.71		1.95	0.85–4.47
Unknown^a	8/87	9.2	1.12	0.76–5.05		2.26	0.73–6.95
Self-perceived risk of HIV infection					0.125			0.047
Absolutely absent	9/77	11.7	2.50	1.04–6.01		2.89	1.07–7.81
Lower than others	6/113	5.3	1.06	0.40–2.82		1.27	0.45–3.58
As much risk as others	14/278	5.0	1.00			1.00
Higher than others	3/18	16.7	3.77	0.98–14.56		5.26	1.14–24.26
Unknown	14/224	6.3	1.26	0.59–2.70		0.90	0.36–2.22

FTR: failure to return for HIV test results; OR: odds ratio; CI: confidence interval.

including patients who did not want to answer.

including one transgender patient, with current male phenotypic sex at the visit.

excluded from the multivariate model.

including the desire to stop using condoms with the partner, the end of a relationship, etc. or a simple check-up.

In multivariate analysis, people who did not specify their birthplace were more likely not to return than those who specified they were born in France (adjusted odds ratio [aOR]: 3.44, 95% confidence interval [CI] [1.15–10.31]). Place of residence was also significantly associated with FTR: people who were living outside of the Paris region were more likely to fail to return compared to those living in Paris region (aOR: 6.88, 95% CI [1.93–24.48]). Among behavioural factors, the number of sex partners during the last 12 months was statistically associated with FTR. The higher the number of sex partners that was declared by participants, the less likely participants were not to return for their test results: the aOR associated with six or more sex partners, compared to fewer than two, was 0.16 (95% CI [0.03–0.87]). People who reported visiting for clinical symptoms were significantly more likely to fail to return than those who came for a check-up as part of a relationship (aOR: 8.43, 95% CI [1.86–38.22]). Finally, self-perception of HIV infection risk was statistically associated with FTR. Participants who had absolutely no self-perceived risk were more likely to fail to return for test results than those who felt they were as much at risk as other people (aOR: 2.89, 95% CI [1.07–7.81]). A higher FTR rate was also observed for those who had a higher self-perceived risk than others (aOR: 5.26, 95% CI [1.14–24.26]). When the HIV screening test was positive (n = 5), participants seemed to be more likely to fail to return for test results, but the association was not significant (aOR: 3.51, 95% CI [0.29–42.41]).

Discussion

Of the 710 study participants who had an HIV test in January or February 2011, 6.5% failed to return for test results within 60 days after the screening. Factors significantly associated with FTR were: living outside of the Paris region, not reporting birthplace, having had fewer than two sex partners within one year, reporting visiting for clinical symptoms and perceiving his or her self-risk of HIV infection to be absolutely absent, or higher than other people. Moreover, even if non-significant in our sample, participants with a positive HIV test failed to return more often than participants with negative results (20% vs. 6%). Previous studies have reported mixed results concerning this association: two US studies found that a negative HIV test was associated with FTR,^7,15 whereas an African study found FTR associated with positive HIV test results.⁸ The public health impact of FTR is particularly important among HIV-positive patients. These patients are not able to begin early treatment, even though more and more national HIV treatment guidelines recommend earlier treatment.^16,17 Furthermore, they may keep transmitting the virus as they are not aware of their HIV status and may not change their at-risk behaviours.³

Socio-demographic factors associated with FTR in our sample were birthplace and place of residence. An Australian study found that patients born overseas were more likely to return for test results than Australian-born patients.⁴ Our results, based on self-administered questionnaires, showed that people who did not want to answer the question about their birthplace were less likely to return. A portion of patients at the CDAG may be illegal immigrants, and fear of breach of confidentiality regarding birthplace and the possibility of deportation could lead them to not return for results. Association between place of residence and FTR appears understandable: the farther away people live, the less likely they are to return for the results. In some studies, FTR has also been reported to be associated with a younger age^6,7,14 and with a lower educational level.^8,10 Such associations were not found in our study.

With regard to behavioural factors, a lower number of sex partners was associated with FTR in our sample, and injected drug use tended to be associated. Results of prior studies do not appear constant for these two factors. One Australian study found that fewer than six sex partners within a year was associated with FTR,⁴ whereas an African study showed that two sex partners or more within one month was associated with FTR.⁸ The association between FTR and IDU was already reported by two previous studies.^4,14 In our study, the low number of IDUs (n = 6) probably explains the lack of significant association, even if it was included in the multivariate analysis (p = 0.194, data not shown). Other previous results showed that MSM were more likely to return for test results than other people.^4,7,14 In our study, no association was found between sexual orientation and FTR. In two previous US studies, FTR was associated with visiting for a reason other than an HIV test.^6,7 Our findings show that visiting with clinical symptoms was associated with higher FTR, compared with visiting for a check-up as part of a relationship (for example, to stop using condoms with their regular partner). These findings are consistent with those of an African study in Ethiopia.¹¹ We hypothesise that people are less likely to return for results when the clinical symptoms that initially motivated the visit disappear.

Finally, self-perceived risk of HIV infection was associated with FTR: patients perceiving themselves not to be at risk and those perceiving themselves to be at higher risk than other people failed to return significantly more often than those perceiving themselves to be as likely at risk as other people. In our study, perceived risk is assessed in the questionnaire, before pre-test counselling. It is possible that after the pre-test counselling, people who had perceived themselves at higher risk may realise that they had not in fact been engaging in HIV risk behaviour. These results are in agreement with two studies that reported an association between FTR and perceived risk: one in the US for a higher perceived risk⁹ and one in Tanzania for an absence of perceived risk.⁸

To explore factors associated with FTR, self-administered questionnaires were used, not face-to-face interviews. This method risks misunderstanding of the questions by participants and a non-response bias.¹⁸ However, in the ANRS-PRIMO cohort, reported frequency of unsafe sex rose sharply after confidential self-reporting replaced clinician-administered questionnaires.¹⁹ Moreover, self-administered questionnaires allow for better standardisation of data collection and eliminate the interviewer effects. Nevertheless, the questionnaire covers sensitive issues on sexual and addictive behaviours and is based on patients’ self-reporting, so an over- or under-reporting bias remained possible. The questionnaire was only available in French. Therefore, we cannot rule out that a share of non-participation was due to language barrier, leading to a selection bias. We defined FTR as not returning for test results within 60 days after blood sampling, as the period chosen in most studies published on FTR was between 30 and 60 days.^{4–6,8,11,14} In our sample, 46 patients failed to return within 60 days, and 42 of them never came back during the 2011 year, and thus were never informed of their serology. The relatively low number of people in our sample who failed to return (n = 46) is an encouraging result but led to a lack of statistical power in the study. This might explain in part the absence of significant association between HIV test results and FTR found during the study period.

An at-risk group for FTR for HIV test results is difficult to define. FTR for HIV test results has been studied less in France and Europe than in other parts of the world. However, most of our results are in agreement with previous studies conducted in other countries. As socio-cultural factors may impact returning for HIV test results, studies have to be repeated in different settings. Self-perceived risk seems to be of chief concern. Adapted policies have to be established to decrease FTR rates and to raise the impact of prevention. During pre-test counselling, a patient’s self-perceived risk could be assessed, for example, by the use of a visual analog scale. People with higher or lower self-perceived risk of HIV infection could, therefore, get a focused pre-test counselling. Furthermore, policies could also include focused pre-test counselling for people who have few sex partners or who live far from the CDAG. Providing rapid HIV tests could be another option. Previous studies have shown the efficiency of this strategy.^20,21 A sociological study suggested client-friendly policies to decrease FTR rates by, for example, matching the post-test counselling with the doctor who had done the pre-test, or by having more flexible opening hours at the centre, for example, opening later one night per week, or during the week-end.²² Multi-lingual staff or culturally appropriate counselling techniques can also be other client-friendly policies that may decrease FTR rates. Efficacy of these client-friendly policies remains to be assessed by comparing FTR rates before and after their implementation.

Footnotes

Acknowledgements

The authors are grateful to Fabienne Vereecke, reception agent at the CDAG, to nurses and physicians at CDAG for their help during the study, and to CDAG patients. The authors would also like to thank Judith Armbruster and Diego Ramallo for their skillful assistance in reviewing the language of this manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Bouvet

Le Vu

. [Free anonymous screening consultation and clinical management of HIV infection]. Médecine Sci MS 2004; 20: 1145–1148.

Le Vu

Semaille

. [Free and anonymous HIV testing in France, 2006]. Bull Epidemiol Hebd 2008; 7–8: 49–52.

Marks

Crepaz

Senterfitt

. Meta-analysis of high-risk sexual behavior in persons aware and unaware they are infected with HIV in the United States: implications for HIV prevention programs. J Acquir Immune Defic Syndr 2005; 39: 446–453.

Chan

McNulty

Tribe

. Who returns for HIV screening test results? Int J STD AIDS 2007; 18: 171–174.

Healey

O’Connor

Templeton

. HIV result giving.. Is it time to change our thinking? Sex Health 2010; 7: 8–10.

Erbelding

Chung

Zenilman

. Following-up for HIV test results: what limits return in an STD clinic population? Int J STD AIDS 2004; 15: 29–32.

Valdiserri

Moore

Gerber

. A study of clients returning for counseling after HIV testing: implications for improving rates of return. Public Health Rep 1993; 108: 12–18.

Mmbaga

Leyna

Mnyika

. Prevalence and predictors of failure to return for HIV-1 post-test counseling in the era of antiretroviral therapy in rural Kilimanjaro, Tanzania: challenges and opportunities. AIDS Care 2009; 21: 160–167.

Sullivan

Lansky

Drake

. Failure to return for HIV test results among persons at high risk for HIV infection: results from a multistate interview project. J Acquir Immune Defic Syndr 2004; 35: 511–518.

10.

Catania

Kegeles

Coates

. Psychosocial predictors of people who fail to return for their HIV test results. AIDS Lond Engl 1990; 4: 261–262.

11.

Sahlu

Kassa

Agonafer

. Sexual behaviours, perception of risk of HIV infection, and factors associated with attending HIV post-test counselling in Ethiopia. AIDS Lond Engl 1999; 13: 1263–1272.

12.

Cazein

Le Strat

Le Vu

. [HIV testing in France, 2003-2011]. Bull Epidémiologique Hebd 2012; 46–47: 529–533.

13.

Chevreul

Durand-Zaleski

Bahrami

. France: health system review. Health Syst Transit 2010; 12: 1–291, xxi–xxii.

14.

Slutsker

Klockner

Fleming

. Factors associated with failure to return for HIV post-test counseling. AIDS Lond Engl 1992; 6: 1226–1227.

15.

Wiley

Frerichs

Ford

. Failure to learn human immunodeficiency virus test results in Los Angeles public sexually transmitted disease clinics. Sex Transm Dis 1998; 25: 342–345.

16.

Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Panel on antiretroviral guidelines for adults and adolescents, http://aidsinfo.nih.gov/contentfiles/lvguidelines/adultandadolescentgl.pdf (2014, accessed 16 July 2014).

17.

Morlat

. Prise en charge médicale des personnes vivant avec le VIH, Paris: Ministère des Affaires sociales et de la Santé, 2013.

18.

Catania

McDermott

Pollack

. Questionnaire response bias and face-to-face interview sample bias in sexuality research. J Sex Res 1986; 22: 52–72.

19.

Seng

Rolland

Beck-Wirth

. Trends in unsafe sex and influence of viral load among patients followed since primary HIV infection, 2000–2009. AIDS Lond Engl 2011; 25: 977–988.

20.

Cabié

Bissuel

Abel

. [Use of HIV rapid tests in free and anonymous screening consultations in the French West Indies]. Médecine Mal Infect 2009; 39: 370–374.

21.

Thornton

Delpech

Kall

. HIV testing in community settings in resource-rich countries: a systematic review of the evidence. HIV Med 2012; 13: 416–426.

22.

Grusky

Roberts

Swanson

A-N

. Failure to return for HIV test results: a pilot study of three community testing sites. J Int Assoc Physicians AIDS Care (Chic) 2007; 6: 47–55.