Use of the Internet for home testing for Chlamydia trachomatis in Sweden: who are the users?

Abstract

Sweden was the first country to introduce online Chlamydia trachomatis testing for both men and women and this article provides information about the user characteristics and which risk groups are reached with this method. During the years 2005–2007 a questionnaire was administered with a response rate of 86% (n = 6025) (62.5% women, 37.5% men). Over 60% of respondents were aged below 26 years and single. The average years of education for men and women was 14 (range 9.0–26.0). The mean Internet usage time was 22.4 hours/week among men and 15.5 hours/week among women. Only 22% men and 34% women had tested themselves for C. trachomatis at clinics previously. Respondents reported a higher frequency of previous sexually transmitted infections (STIs) compared with other Internet users (i.e. C. trachomatis: men 19%, women 24%). Eighty percent of online test users had engaged in sex with ≥2 partners without a condom during the previous year. Online C. trachomatis test users are mostly young people with high sexual risk behaviour.

Keywords

sexually transmitted infection STI screening Internet home testing gender Sweden

INTRODUCTION

Home testing is part of the new strategy to increase the testing coverage of sexually transmitted infections (STIs). It has been proven to be a feasible and acceptable testing method in both low-¹ and high-income settings.² Home screening for Chlamydia trachomatis has been considered to be cost-effective and outcome measures such as the incidence of pelvic inflammatory disease (PID) has been shown to be half as common among high-risk women offered home testing for C. trachomatis versus conventional clinical testing.^2,3

The Internet, being a new health communication platform, has facilitated home testing and the development of new health services. The Internet is utilized to test for STIs^4,5 and also serve as a complement to ordinary clinical testing. The overall Internet usage is increasing yearly and currently the average time a Swede spends on the Internet is 15.5 hours/week.⁶ Sweden was the first country to introduce home testing for C. trachomatis with the help of the Internet for both men and women in the year 2004.^7,8 Since the introduction in the county of Västerbotten it is now available for free to over half the Swedish population. Home testing by means of the Internet may be seen as a cost-effective, convenient and easily accessible complement to ordinary clinical testing and a service that may attract people who are embarrassed or not willing to visit a clinic for testing.^9–11 However, since this was the first online STI testing in the world for both men and women, and this is a medical service that most likely will expand in the future, this article aims at providing information about the characteristics of these online C. trachomatis test users and which risk groups are reached with this method.

METHODS

Swedish Internet users ordering a C. trachomatis home test from the largest publicly funded online testing service www.klamydia.se in Sweden answered a questionnaire. The Internet testing method is described in detail elsewhere.⁸ Briefly, citizens in the county of Västerbotten need to enter their personal civic number and postal address on the website to make an order. Orders are stored in a database and the test package is sent containing instructions, a personal coded urine specimen container, laboratory requisition slip and a paper questionnaire. Women also receive a vaginal swab for self-collection that they thereafter insert into their urine sample. Test samples together with laboratory requisition slip and the questionnaire are sent by mail in pre-addressed envelopes to the laboratory and test results are retrieved from the Internet site with their personal codes. Infected individuals seek treatment independently by printing out a physician referral sheet and the physician confirms that treatment has been given by sending a letter to us.

The questionnaire that accompanied all ordered C. trachomatis tests contained questions regarding sexual activity, previous STIs and Internet usage along with sociodemographic background questions. Internet usage was evaluated by the question ‘how many hours per week do you use the Internet?’, which is believed to include both work and recreational usage. The variable sex with number of unknown partners without condom during the previous year was defined from a question which reads ‘during the previous year, how many times did you have sex without using a condom with a person who was unknown to you at the time of intercourse?’ The questionnaires were sent out during three consecutive years, 2005–07. Respondents were informed that answering the questionnaires was voluntary and that their decision would not alter the testing procedure. The local medical ethics committee approved the study.

RESULTS

A total of 9360 orders were made during the three-year period. Of these orders 6978 urine samples were returned for analysis. A total of 6025 questionnaires were received together with the urine samples. The total questionnaire response rate was 86% (6025/6978): response rate among men was 77% (2256/2923) and 93% (3769/4055) among women (Table 1). Incomplete questionnaire data for the selected variables in the study ranged from 0.4–2.6%. The average and median ages of male testers were 25 and 24 years, respectively (range 15–64 years). The average and median ages of female testers were 24 and 23 years, respectively (range 15–61 years). The average and median ages of male respondents were similar to the testers but with a range from 15 to 58 years. The average and median age of female respondents were also similar to the testers, but with a range from 15 to 54 years. However, the vast majority of the respondents were below the age of 26 years (63% men and 77% women) (Table 1). The majority of the testers were single (≥66%). The average years in education were 13.8 (range 9.0–26.0) for women and 13.9 (range 9.0–28) for men. About 22% of the men and 34% of the women had been tested for C. trachomatis at a clinic previously. Symptoms were the most common reason for Internet testing among women followed by believing to be infected. Among men the most common reason was believing to be infected, followed by partner suggestion. Eighty percent (80%) of women and men in the study had sex with two or more partners without using a condom during the previous year. Men spent an average of 22.4 hours on the Internet per week while women spent an average of 15.5 hours. C. trachomatis was the most common previously reported STI among testers (Figure 1). The average weekly Internet usage time among the online STI test users was comparable to the general Internet usage in Sweden (Figure 2).

Table 1

Background characteristics of the klamydia.se users by gender

Characteristics	All (n = 6025)	Men (n = 2256)	Women (n = 3769)	P value (men versus women)
Age group (years)
15–25	71.8 (4256)	63.1 (1388)	76.9 (2866)	NS
26–35	23.5 (1391)	31.3 (689)	18.8 (702)
36–45	3.9 (231)	4.2 (92)	3.7 (139)
46–55	0.8 (48)	1.3 (29)	0.5 (19)
≥56	0.1 (3)	0.1 (3)	–
Average education in years (range)	13.8 (9.0–28.0)	13.8 (9.0–26.0)	13.9 (9.0–28.0)	NS
Civil status
Single	66.7 (3984)	67.7 (1510)	66.0 (2474)	NS
Not single	20.3 (1213)	19.7 (440)	20.6 (773)
Cohabiting	9.6 (575)	9.8 (219)	9.5 (356)
Engaged	1.7 (103)	1.7 (37)	1.8 (66)
Married	0.7 (42)	0.4 (8)	0.9 (34)
Divorced	1.0 (60)	0.8 (17)	1.1 (43)
Previously tested for Chlamydia other than at Klamydia.se
Youth health clinic	15.5 (931)	9.4 (212)	19.1 (719)	0.05
Hospital clinics (STI clinic/dermatologist, gynaecologist)	6.4 (385)	6.1 (137)	6.6 (248)	NS
General practitioner	6.5 (391)	5.4 (122)	7.1 (269)	NS
Other	1.3 (80)	1.3 (29)	1.4 (51)	NS
Total	29.7 (1787)	22.2 (500)	34.1 (1287)	0.05
Reason for testing
Believed to be infected	21.7 (1308)	22.2 (501)	21.4 (807)	NS
Pain when urinating	8.5 (515)	9.0 (204)	8.3 (311)	NS
Discharge	10.4 (628)	3.0 (67)	14.9 (561)	0.05
Bleeding between periods	3.7 (220)	–	5.8 (220)	–
Partner suggestion	10.3 (622)	17.9 (403)	5.8 (220)	0.05
Other	48.5 (2921)	44.4 (1001)	50.9 (1920)	NS
Number of hours spent on Internet per week, mean (median) (range)
Chatting	6.0 (3.0) (0–95.0)	5.8 (3.0) (0–95.0)	6.2 (3.0) (0–95.0)
Gaming	2.3 (0.0) (0–100.0)	3.8 (1.0) (0–100.0)	1.2 (0.0) (0–54.0)
Porn surfing	0.7 (0.0) (0–30.0)	1.5 (1.0) (0–30.0)	0.1 (0.00) (0.0–6.0)
Film/music	3.1 (2.0) (0–216.0)	4.8 (2.0) (0–168.0)	2.0 (1.0) (0–216.0)
Other	5.8 (4.0) (0–115.0)	6.5 (5.0) (0–100.0)	6.0 (4.0) (0–115.0)	NS
Total	17.9 (3.0) (0–216.0)	22.4 (3.0) (0–168.0)	15.5 (3.0) (0–216.0)
Number of SMS sent per day: mean (median) (range)	5.0 (3.0) (0–40)	4.5 (3.0) (0.0–40.0)	5.2 (3.0) (0–40)	NS
Number of emails sent per week: mean (median) (range)	6.8 ((3.0) (0–120)	7.7 (3.0) (0.0–120.0)	6.2 (3.0) (0–100)	NS
Number of sexual partners during previous year
0	1.0 (57)	1.3 (29)	0.7 (28)
1	22.6 (1358)	21.4 (481)	23.4 (877)
2–4	52.2 (3134)	49.0 (1101)	54.2 (2033)
5+	24.2 (1450)	28.3 (635)	21.7 (815)	NS
Number of sexual partners without a condom during the previous year
0	4.5 (272)	5.3 (118)	4.1 (154)
1	15.3 (917)	14.7 (330)	15.7 (587)
2+	80.2 (4796)	80.0 (1791)	80.2 (3005)	NS
Number of unknown sexual partners without a condom during the previous year
0	55.7 (3258)	49.2 (1075)	59.5 (2183)
1	24.1 (1413)	25.0 (547)	23.6 (866)
2+	20.2 (1183)	25.8 (562)	16.9 (621)	NS

SMS = short message service (text message); STI = sexually transmitted infection; NS = non-significant

Note: The values presented in this table are percentages and number of subjects in parentheses unless stated otherwise

Figure 1

Frequency (%) of having sexually transmitted infections previously among the users of www.klamydia.se by gender

Figure 2

Average hours of Internet use per week among the users of www.klamydia.se versus World Internet Projects by age groups. Source: Findahl O²⁰

DISCUSSION

New online STI testing services are quickly emerging worldwide; some are publicly funded and offer free testing while others may be expensive and not regulated.^5,12 There are risks and disadvantages with online STI testing such as the lack of individualized care including lack of individualized information, counselling, and appropriate STI testing and management. Home testing for STIs by means of the Internet may seem like a convenient and appealing alternative to young adults, but there is yet little knowledge about the characteristics of these users. Since online STI testing is a new medical service that most likely will expand in the future, it is vital to understand who the users are and what groups are reached by this method and these are the key issues addressed by the present study.

Sexual risk behaviour and previous STIs

The vast majority of respondents in our study were aged below 26 years (63% men and 77% women), which correlates to the general testing trend of www.klamydia.se, where 57% of the men and 65% of the women are below the age of 26 years. The online STI testing site was used by persons with high-risk behaviour for C. trachomatis, where 80% of respondents reported having sex with two or more partners without a condom during the previous year. In the large Natsal 2000 study, which evaluated sexual behaviour in Britain related to self-reported C. trachomatis infection, men having sex with two or more partners without a condom during the past year had an adjusted odds ratio (OR) of 15.62 (P = 0.0001) for C. trachomatis infection and the corresponding adjusted OR for women was 4.15 (P = 0.001).¹³ The testing distribution regarded to marital status was also correlated to the risk level of each civil state. Being married, engaged or cohabitating had the lowest risk for C. trachomatis ¹³ and they were also least likely to order an Internet C. trachomatis test. The age and marital status was similar in a male Internet-based STI screening study in the USA.¹⁴

Compared with other Swedish Internet users, our participants reported an overall higher frequency of previous STIs. In a recent study, Ross et al. investigated the STI history of 904 men and 931 women who visited one of the top Swedish Internet portals. A banner invited visitors over the age of 18 years to answer an Internet-based questionnaire and found that 8.0% of the men and 12.0% of the women had previously been infected with C. trachomatis, 3.9% of the men and 7.6% of the women had previously been infected with human papillomavirus (HPV), 3.1% of the men and 3.4% of the women had previously been infected with genital herpes, and 3.3% of men and 7.6% of women reported a previous gonorrhoea infection.¹⁵ Compared with the Ross et al. study the users of www.klamydia.se who answered the questionnaire had a higher frequency of previous STIs except for herpes and gonorrhoea in men. A plausible explanation could be that gonorrhoea is still mostly prevalent in bigger Swedish cities and the participants in our study live in a sparsely populated county in northern Sweden, while Ross et al.'s¹⁶ participants had a predominantly urban background. In a further comparison the previous STIs among the users of www.klamydia.se who answered the questionnaire was similar to a male Internet-based STI screening study in the USA where 34% had a prior history of any STI.¹⁴

Study limitations

The response rate of 86% could have been higher if the online STI test users were resent questionnaires or reminded by telephone. The characteristics of the non-respondents could also have been evaluated with a follow-up questionnaire. Only one questionnaire was sent without any further action in order to minimize the risk of invading the users’ privacy. We believed that if a user experienced a privacy invasion, he/she might be less interested in future STI testing. A further limitation is that it is not known if condoms were used with a stable or temporary partner. The question was formulated to ask if a condom was used with a person who was known or unknown during intercourse. Internet usage was self-reported which may include a recall bias limitation with a possibility of an over- or underestimation of the Internet usage. Almost one quarter (25%) of all test orders were never returned (2382/9360). This is a limitation of the service itself but also a study limitation since we do not know the characteristics of these non-testers.

Gender aspects

Women answered more questionnaires than men (93% versus 77%), which is consistent with the overall Internet testing rate at www.klamydia.se where more women than men tested themselves (60% women and 40% men). In comparison, regular C. trachomatis testing is done in youth health clinics, STI clinics or general health clinics and most of the young sexually active visitors are women.¹⁷ In Sweden only 28% of chlamydia tests are undertaken by men.¹⁸ The online STI service offers an alternative to these otherwise difficult-to-reach men, which also might make them more willing to answer a questionnaire aiming at improving the online STI testing method.

Previous testing for C. trachomatis was gender-related in the younger age group where fewer men than women mentioned the youth health clinic as a clinical testing alternative to the Internet. It is known that only 11% of the visits to the youth health clinics in Sweden are by men, thereby making Internet testing a comfortable substitute for this group.¹⁷ This discrepancy in clinical testing among men and women in the younger age group is worrisome. Future studies need to evaluate if online STI testing services will completely replace the first clinical sexual health visit by young men, which would be unsatisfactory. A first supportive clinical visit for a young person becoming sexually active is extremely important from a preventive point of view.

Reasons for Internet testing varied between men and women. It was much more common for men to test themselves after recommendations from their partner than vice versa (18% versus 5.8%). From a heterosexual, gender, and preventive aspect this is an unequal distribution of sexual health responsibilities. The optimal scenario would be that both genders were recommended equally to test themselves; this was not achieved, possibly due to behaviour differences but also due to fewer sexual health-care visits for men than women. In Sweden, women seek care in youth and STI clinics more often than men¹⁷ because of regular birth control measures and Pap smears. This is also shown in our study where previous clinical testing for C. trachomatis was much more common among women than men (34.1% versus 22.2%). Partner recommendation as a preventive measure would be more effective if it had an equal gender distribution since STIs most often involve both members of a relationship. Whatever the reason for the unequal distribution of partner testing recommendations, it would be improved by more sexual health visits and/or testing by men. Believing themselves to be infected was the most common reason for testing themselves on the Internet and our previous study showed that this was the strongest variable associated with chlamydia infection among men when performing an Internet chlamydia test, OR 9.10 (confidence interval [CI]: 2.80–29.5).⁸ Therefore, easily accessible health clinics with a focus on attracting both women and men are recommended. On the other hand, the most common reason for testing themselves among women was having symptoms, which was also shown to be strongly associated with chlamydia infection in one of our previous studies, OR 2.94 (CI:1.70–7.38).⁸

Internet usage and Internet activity

Sweden has one of the highest levels of Internet access in the world, second only to Iceland, where 87.7% of the Swedish population had access to the Internet during the year 2008.¹⁹ While Internet access is high, Sweden does not have the highest Internet usage. The overall average Internet usage is only 15.5 hours per week compared with the USA at 19 hours per week.⁶ The general Internet usage of our participants is best compared with the Swedish part of the World Internet Project, which investigates the Internet usage by a revolving panel design consisting of 2000 individuals, evenly distributed into different age groups, interviewed by telephone.²⁰ Figure 2 shows that average weekly Internet usage is approximately the same in the young age group as in the Swedish part of the World Internet Project, but slightly higher in the age group 26–35 years. Older age groups have too few participants whereby no conclusion can be drawn. The distribution of Internet activities of our participants is comparable to the general Internet usage in Sweden where the most popular activity among both young men and women is chatting (Instant Messaging) followed by watching movies and playing games.²⁰ The interest of playing games decreases with age and is replaced by other activities, especially chatting. Chatting becomes more important with age and among young Swedish adults aged 21–30 years almost half of the young Swedish population chat daily.²⁰ The respondents in our study show a similar trend where playing games decreases with age and chatting is the single most common Internet activity for both genders. Chatting and instant message services therefore act as a valuable daily meeting platform where new social and sexual networks are formed for young adults in Sweden. Among Swedish adults using the Internet, flirting by means of chatting was the most popular online sexual activity (OSA) among women and the second most popular OSA among men.²¹ Viewing erotica was the most popular OSA among Swedish men in all age groups²¹ and viewing erotica was also much more common among men than women in our study. Chat rooms and online communities should therefore serve as good platforms for STI preventive messages.

CONCLUSIONS

The users of the Swedish online C. trachomatis testing service were mostly below the age of 26 years, with no educational differences between gender, and they do not spend more time on the Internet than other general Swedish people. However, eight out of ten respondents had sex with two or more partners without using a condom during the previous year. Few have previously tested themselves for C. trachomatis at a health clinic and they also reported a higher frequency of previous STIs than other Internet users.

Footnotes

ACKNOWLEDGEMENTS

Gratitude to Professor Roger Karlsson, the co-founder of the online STI testing service, and for supplying the data material. Thank you also the virology laboratory at Norrlands University Hospital for collecting the questionnaires.

References

Jones

, Altini

, de Kock

, Home-based versus clinic-based self-sampling and testing for sexually transmitted infections in Gugulethu, South Africa: randomised controlled trial. Sex Transm Infect 2007;83:552–57

Ostergaard

, Andersen

, Moller

, Olesen

. Home sampling versus conventional swab sampling for screening of Chlamydia trachomatis in women: a cluster-randomized 1-year follow-up study. Clin Infect Dis 2000;31:951–57

Novak

, Lindholm

, Jonsson

, Karlsson

. A Swedish cost-effectiveness analysis of community-based Chlamydia trachomatis PCR testing of postal urine specimens obtained at home. Scand J Public Health 2004;32:324–32

Gaydos

, Dwyer

, Barnes

, Internet-based screening for Chlamydia trachomatis to reach non-clinic populations with mailed self-administered vaginal swabs. Sex Transm Dis 2006;33:451–57

Owens

, Arora

, Quinn

, Utilising the internet to test for sexually transmitted infections: results of a survey and accuracy testing. Sex Transm Infect 2010;86:112–16

Findahl

. The Swedes and the Internet 2010. Stockholm: SE (The Internet Infrastructure Foundation), 2010

Novak

, Edman

, Jonsson

, Karlsson

. The internet, a simple and convenient tool in Chlamydia trachomatis screening of young people. Euro Surveill 2003;8:171–76

Novak

, Karlsson

. Simplifying chlamydia testing: an innovative Chlamydia trachomatis testing approach using the internet and a home sampling strategy: population based study. Sex Transm Infect 2006;82:142–47; discussion 52–53

Gaydos

, Rizzo-Price

, Barnes

, The use of focus groups to design an internet-based program for chlamydia screening with self-administered vaginal swabs: what women want. Sex Health 2006;3:209–15

10.

Lane

, McCright

, Garrett

, Features of sexually transmitted disease services important to African American adolescents. Arch Pediatr Adolesc Med 1999;153:829–33

11.

Millstein

, Adler

, Irwin

Jr . Sources of anxiety about pelvic examinations among adolescent females. J Adolesc Health Care 1984;5:105–11

12.

Peeling

. Testing for sexually transmitted infections: a brave new world? Sex Transm Infect 2006;82:425–30

13.

Fenton

, Korovessis

, Johnson

, Sexual behaviour in Britain: reported sexually transmitted infections and prevalent genital Chlamydia trachomatis infection. Lancet 2001;358:1851–54

14.

Chai

, Aumakhan

, Barnes

, Internet-based screening for sexually transmitted infections to reach nonclinic populations in the community: risk factors for infection in men. Sex Transm Dis 2010;37:756–63

15.

Ross

, Daneback

, Mansson

, Reported sexually transmitted infections in Swedish Internet-using men and women. J Eur Acad Dermatol Venereol 2008;22:696–703

16.

Ross

, Mansson

, Daneback

, Biases in internet sexual health samples: comparison of an internet sexuality survey and a national sexual health survey in Sweden. Soc Sci Med 2005;61:245–52

17.

FSUM. The Forum for Swedish Youth Centres Statistics. Statistics . 2010. See http://www.fsum.org (last accessed 2 January 2012)

18.

SIDDC. Statistics on Sexually Transmitted Infections in Sweden. Stockholm: Swedish Institute for Infectious Disease Control, 2005. See www.smittskyddsinstitutet.se [Swedish English summaries]

19.

ITU. International Telecommunication Union: World Telecommunication Development Report and database . 2008. See http://data.worldbank.org/indicator/IT.NET.USER (last accessed 2 January 2012)

20.

Findahl

. The Young Swedes and the Internet 2009. Stockholm: The Swedish Internet Infrastructural Foundation, 2009

21.

Cooper

A, Masson SA

, Daneback

, Predicting the future of Internet sex: online sexual activities in Sweden. Sex Relationship Ther 2003;18:277–91