Characteristics of the study population
A total of 100 PWID and 100 MSM/TG were enrolled in the study between December 2019 and June 2020. The overall participation rate was 98.5%. Baseline characteristics of the study populations are summarized in Table 1.
Table 1
Sociodemographic characteristics of study participants
Population group
|
PWID (n = 100)
|
|
MSM/TG
(n = 100)*
|
P value
|
Characteristics
|
No.
|
%
|
|
No.
|
%
|
|
Median age, years (IQR)
|
35 (28–41)
|
|
24 (21–29)
|
< 0.001
|
Gender
|
|
|
|
|
|
0.004
|
Male
|
98
|
98.0
|
|
91
|
91.0
|
|
Female
|
2
|
2.0
|
|
0
|
0.0
|
|
Transgender
|
0
|
0.0
|
|
9
|
9.0
|
|
Educational level
|
|
|
|
|
|
< 0.001
|
Below high school
|
2
|
2.0
|
|
2
|
2.0
|
|
High school
|
58
|
58.0
|
|
19
|
19.0
|
|
College
|
10
|
10.0
|
|
2
|
2.0
|
|
Undergraduate
|
27
|
27.0
|
|
59
|
59.0
|
|
Post graduate
|
3
|
3.0
|
|
18
|
18.0
|
|
Employment
|
|
|
|
|
|
< 0.001
|
Employed
|
41
|
41.0
|
|
74
|
74.0
|
|
Unemployed
|
59
|
59.0
|
|
25
|
25.0
|
|
Missing data
|
0
|
0.0
|
|
1
|
1.0
|
|
Marital status
|
|
|
|
|
|
< 0.001
|
Married or living with a partner
|
28
|
28.0
|
|
5
|
5.0
|
|
Single
|
53
|
53.0
|
|
88
|
88.0
|
|
Divorced or widowed
|
19
|
19.0
|
|
7
|
7.0
|
|
Self-reported HCV risk factors
|
|
|
|
|
|
|
Dental procedure(s)
|
75
|
75.0
|
|
58
|
58.0
|
0.011
|
Surgical procedure(s)
|
36
|
36.0
|
|
30
|
30.0
|
0.340
|
Sharing shaving tools / toothbrushes
|
12
|
12.0
|
|
17
|
17.0
|
0.315
|
Injecting unprescribed drugs
|
100
|
100.0
|
|
8
|
8.0
|
< 0.001
|
Sharing needles
|
33
|
33.0
|
|
1
|
1.0
|
< 0.001
|
Condomless anal sex
|
0
|
0.0
|
|
79
|
79.0
|
< 0.001
|
Has a tattoo
|
40
|
40.0
|
|
31
|
31.0
|
0.184
|
None reported
|
0
|
0.0
|
|
4
|
4.0
|
0.043
|
Frequency of routine health check
|
|
|
|
|
|
< 0.001
|
More than once per year
|
3
|
3.0
|
|
57
|
57.0
|
|
Once per year
|
16
|
16.0
|
|
24
|
24.0
|
|
Rarely
|
66
|
66.0
|
|
18
|
18.0
|
|
Never
|
15
|
15.0
|
|
1
|
1.0
|
|
Awareness about self-testing
|
|
|
|
|
|
|
Aware certain tests can be done at home
|
32
|
32.0
|
|
96
|
96.0
|
< 0.001
|
*Nine participants self-identified as transgender people |
PWID were older than MSM/TG (median age: 35 vs 24 year, p = < 0.001). The PWID population was predominantly male with only 2 women participating in the study. Nine of the MSM/TG individuals self-identified as transgender. There were some sociodemographic differences between MSM/TG and PWIDs that were statistically significant, namely: the MSM/TG population had a higher level of education (59.0% vs 27.0%, p = < 0.001), a higher proportion had a job (74.0% vs 41.0%, p = < 0.001), and most were aware of self-testing (96.0% vs 32.0%, p = < 0.001) compared to PWID. Eight participants among MSM/TG also reported injecting unprescribed drugs (8.0%).
For the sub-sample of 40 participants who underwent cognitive interview or in-depth interviews, the main sociodemographic characteristics between groups are presented in (see Additional file 8). Overall, MSM/TG were younger, more educated, tended to be employed and single compared to PWID. This subsample was representative of the entire sample because the sociodemographic characteristics in this subset of participants was consistent with that of the larger group reported in Table 1.
Qualitative findings from cognitive interviews
There were three main areas where participants struggled with the test which were observed by the research assistants and then raised in the cognitive interviews: collecting the oral specimen; the interpretation of the test result; and timing and reading window of the test. These errors may have a negative impact on the accuracy of the test result.
Collecting the oral specimen
Five of the 20 participants had difficulties collecting the oral specimen. These were all from the PWID group and all of them had high school level education or lower. Two participants were unemployed and one was in a manual occupation. One participant only rubbed the pad on the bottom gum, another switched the side of the pad for each gum, and another participant only sampled halfway long each gum.
Interpretation of the result
More than half of the participants (11/20) struggled with some part of the interpretation of results. Many participants had difficulty understanding that a positive result meant they had antibodies to HCV and conversely that a negative result meant the antibodies were not detected. The word “positive” in the IFU was associated with a favourable outcome. In this case, some participants believed that a positive result meant they did not have HCV infection. Conversely, the word “negative” was associated with an unfavourable outcome, in this case, having HCV infection. Therefore, how HCV positive or negative results were translated in Georgian or worded in the IFU led to confusion on the true interpretation of the result by the participants.
“What does“positive” mean? It means that you don’t have virus, yes? … I think – I am confused, because I went for analyses, it did not show the virus and they say positive. This time I thought that I had virus, but it seems it was not and still they say positive. According to this, I think positive means that I don’t have (HCV). [112, PWID].
Some participants had difficulty understanding the difference between the control line and test line. They interpreted any single line as negative and therefore missing a possible invalid result. It was noted by some participants that there was no picture in the IFUs depicting an invalid test that had a test line only, which may have affected some individual’s comprehension of the text. Some participants were not able to link the pictorial instructions of test device with results to the references of control and test lines in the text.
“Yes, it does not reflect the instruction, because there is different words in instruction, ok there is no line along C but there should be line along T, here is written that there is line along T. the red background if shown but there is not red line along T.” [104, PWID].
In the MSM/TG group, participants who struggled to interpret the different test results in the IFUs were unemployed and only had a high school education, whereas those who successfully interpreted the test were mostly employed and had a university-level of education. In the PWID group, individuals who struggled with interpretation of test results also had a high school education or lower and were unemployed. Only one participant had a university-level of education and was able to interpret the different test results correctly.
Time requirement to ensure the correct reading window for the test
Half of participants (10/20) struggled to understand the instructions regarding the requirement to time the test and/or the need to read the results within the 20-40-minute reading window. Some participants (6/20) did not understand why they needed a timer for the test and were confused by this:
“I don’t need them, I don’t need these devices, there are for measuring time, how long I will need to do the test, I don’t need them.” [105, PWID].
Often it was thought that a timer was required to measure how long the participant took to perform the test rather than how long to wait before reading the result. This was observed more commonly among PWID.
Thirteen participants did not understand the instruction regarding the manufacturer allowed time window that is required to read the test result. They were unclear as to what they were being asked to do, with the term “don’t read” being highlighted as difficult to understand, and not being viewed as connected to the interpretation of the test results. Many participants who did understand this instruction had to read this instruction carefully to understand the true intended meaning and identified it as hard to interpret.
“From the beginning it was difficult for me. …see, not read the results it is also needed to be mentioned, but if you read it, there is no sense, you cannot read them, if it does not become visible. If you change it and write that read your results after 20 minutes, it will be less confusing… And if it is directly written that put it inside and read it after 20 minutes, it will be more… in any case you will understand but still it is little difficult…” [108, PWID].
After inserting the test device in the developer fluid, nine out of 13 participants did not set the timer to time the test and had to be prompted by the researcher.
“… it is unclear, first they say read then don’t read, I cannot understand what they really want. Do they mean don’t look at it? … I don’t have watch or even the phone.” [107, PWID].
Qualitative findings from in-depth interviews
The self-test was viewed by the majority of participants as easy to use. The test was seen as very convenient, especially the ability to test in a comfortable environment. Some participants trusted the result of a blood test more, as they were felt to be more accurate. There was massive support for HCVST to be universally available rather than targeted at certain key populations. Pharmacies were seen as the best route of distribution, though other routes such as through peers and workplace distribution were highlighted. The cost of the test was seen as important to uptake with an acceptable price point being 10 Georgian Lari ($3.50 USD).
Feelings related to stigmatisation of HCV such as fear, shame, and denial were cited by participants as psychological barriers to individuals accessing testing. This was linked to a lack of trust in the confidentiality of testing at both public and private facilities, especially in the MSM/TG group. This would indicate that HCVST would be a useful tool for this group to determine their status, but may then cause more distress if the individual finds themselves to be reactive for HCV, and therefore will not totally remove the effect of stigma on accessing confirmatory testing services.
Most participants knew that HCV was a virus that affects the liver. Most participants understood the main routes of transmission, such as sexual contact and sharing needles, though a minority also had incorrect knowledge such as transmission via saliva and food. This indicates that knowledge campaigns are still required to both allow people to manage risk and decrease stigma.
Most participants knew about the availability of treatment for HCV but many had anecdotal evidence of poor treatment outcomes or major side effects to treatment. Some acknowledged that this may be due to old treatment regimens, but it does indicate that undertaking treatment for HCV may be viewed as carrying some element of risk. Further details about emerging topics and some quotations from participants participating in in-depth interviews can be found in Additional file 9.