Secondary Logo

Journal Logo

Original Study

Computer-Assisted Self Interviewing in Sexual Health Clinics

Fairley, Christopher K. MBBS, PhD, FRACP, FAFPHM*†; Sze, Jun Kit Grad Dip Bus Sys, BComp*; Vodstrcil, Lenka A. BBioMedSci, BSc (Hons), PhD*†; Chen, Marcus Y. MRCP, FAChSHM, PhD*†

Author Information
doi: 10.1097/OLQ.0b013e3181f7d505
  • Free

Health care costs are rising in almost every developed country. In the United States, for example, spending on health care in 2007 was $2.4 trillion or $7900 per person and is rising at double the rate of inflation.1

Access to health is critically important for the control of sexually transmitted infections (STI) because bacterial STIs can be treated, preventing further onward transmission. If access to sexual health care is to be improved or even maintained then health services must improve the efficiency of the services they provide.

A number of innovative changes have recently occurred in sexual health services including the following: the use of email and text messaging for communication with clients, use of the internet for enhancing partner notification, and the advent of video- and computer-based counseling.2–5 Another example is the use of computer-based sexual history taking which has the potential to shorten the duration of consultations, or even potentially, in lower risk individuals, of negating the need for a full clinical consultation.6 Computer-assisted self interviews (CASI) have been used in research settings and allow sensitive personal information such as a sexual history to be obtained without face-to-face interviews and to increase condom use.7 To our knowledge, however, CASI to date has rarely been used routinely in the clinical management of clients attending sexually transmitted disease (STD) clinics.

Implementing CASI as part of routine care within a clinical service requires considerable planning and investment. This review aimed to describe the published articles on CASI compared to clinician-obtained histories in STD clinics. We also review the development of a core sexual history in STD clinics: questions which should form the basis for a sexual history if CASI was to be implemented.

METHODS

We searched OVID Medline using a series of search terms relevant to CASI since 1990 up to February 2010. To identify published articles on CASI in STD clinics, we used the terms “computer assisted interviewing” and “sex,” and to identify published articles on a core sexual history, we used the term “core sexual history.” The abstracts of these articles were then reviewed to identify studies that dealt with these 2 topics in the setting of STD clinics. References in the identified articles were also reviewed. For information on a core sexual history for STD clinics, we searched the web sites and contacted the key individuals in the National Coalition of STD Directors in the United States and British Association for Sexual Health and human immunodeficiency virus (HIV) in the United Kingdom.

RESULTS

Development of a Core Sexual History

Since 1990, there were 104 published articles on OVID Medline with the search term “core sexual history” of which 3 published articles and 1 web reference were identified from the abstracts and relevant cited references.8–10

The process for deciding what questions form the core components of a sexual history and therefore the ones that should be considered for inclusion in CASI have been approached a number of ways. In the United States, investigators contacted 1 STD clinic or STI care facility that was randomly chosen from each of the 65 cities with a population of more than 200,000 and requested a copy of the clinic's sexual history forms. Forty-eight clinics (74% response rate) responded and 72% were public STD clinics. Over 80% included the number and sex of the partners and condom use on the proforma. The CDC 2006 STD guidelines provide 1 suggested approach to questions for a sexual history for clinicians based on “The 5 Ps: Partners, Prevention of Pregnancy, Protection from STDs, Practices, and Past History of STDs.”11 In the United States, the National Coalition of STD Directors undertakes HIV/STD behavioral surveillance as part of its core components of work; however, these questions are not primarily developed for clinical care.12

In the United Kingdom, the British Association for Sexual Health and HIV used a committee of 6 eminent members of the profession (Sexual History Working Party). This committee developed a set of questions they felt constituted reasonable best practice with feedback from relevant stake holders.9

In Australia and New Zealand, sexual health physicians were surveyed and asked to estimate the proportion of new clients who were asked specific sexual history questions.13 The Delphi technique was employed with a group of eminent members of the profession to create a list of questions that were thought to represent best practice for the core sexual history.10

There are similarities in the suggested sexual history across the United States, United Kingdom, Australia, and New Zealand. For example, all suggesting asking about the sex of the partners, the number of partners (although the time period varies), the types of sex (oral, anal, and vaginal) and condom use, pregnancy intent, and contraceptive methods.

Studies Comparing CASI to Interview or Clinician Collected Sexual Histories

Between 1990 and 2010, there were 254 published articles identified on OVID Medline using the search term computer-assisted interviewing of which 68 included the term sex. Of these, 5 were studies that compared computer-assisted interviewing to another form of interviewing and were conducted in a STD clinic. These 5 studies were undertaken in the United States, United Kingdom, and Australia.

In the US study, 609 clinic attendees at the Seattle STD clinic first answered questions using an audio CASI, and then saw a clinician.14 Women reported drug use, sex work, and oral sex, and both men and women reported same sex partners more commonly to CASI. The correlation between a clinician history and CASI was lower for socially sensitive and socially rewarded variables than for neutral variables where the correlation was almost perfect. Individuals reported the presence of symptoms less commonly to CASI (55%) than to a clinician (65%). There was also less missing data in CASI than clinician completed histories. In this study, 82% of clinic attendees said that CASI allowed more honest reporting and 89% found it acceptable.

In the second US study, at the Baltimore STD clinic, 671 attendees answered an audio CASI, and then saw a clinician.15 Individuals reported anal sex, oral sex, same sex partners, and more partners to CASI than to clinicians. In both studies, women were less likely than men to report socially undesirable practices to a clinician than to CASI.

In a third US study, 1351 individuals at an urban STD clinic were nonrandomly assigned to audio CASI or a interview administered questionnaire.16 Individuals undertaking CASI reported anal sex, oral sex, unprotected sex, paying for sex, and alcohol consumption more frequently than those interviewed in person. However, those undertaking CASI reported some STD symptoms less frequently.

Like the other US studies, there were bigger differences between CASI and face to face interviews for some questions for women compared to men, particularly for some socially sensitive variables.

These 3 US studies suggest that socially desirability bias may be operating in some clinics and that CASI may enable more accurate reporting of some risks. However, questions relating to symptoms of STI may be more sensitive when asked by a clinician.

In an Australian study, 611 individuals attending an STI centre were randomized to either a clinician or CASI. In this study, the only question about STI risk that was answered differently was where women reported a higher number of male sexual partners to CASI than to a clinician.17 The study also found that past hepatitis B vaccination was more commonly reported to clinicians, presumably because clinicians were able to prompt an individual's memory or clarify the question. In this study, 95% found CASI easy to do and 87% were comfortable answering CASI.

In a UK study, 2351 STD clinic attendees were randomized to 1 of 3 arms: CASI, a clinician doing computer interviews and a clinician with a paper history.18 This study showed that those randomized to CASI or to a clinician doing computer interviews more commonly reported concurrent relationships, anal sex, and a higher number of partners than to a clinician undertaking a paper history. Interestingly, the responses to CASI and a clinician doing computer interviews were very similar. The study also looked at whether individuals in the 3 arms had different tests ordered or diagnoses performed. There was no difference between the 3 arms with regards to STI diagnoses. Testing for any STI was different between the 3 groups (CASI, 92.6%; clinician doing computer interviews, 95.3%; and clinician with paper history, 92.6%). The finding that HIV testing was less common in those doing CASI (62.6%) was less common than clinicians with paper histories (68.8%).18

These 5 studies demonstrate the consistent finding that CASI identifies higher risk behavior more commonly than is the case with clinician interview. Currently, there are insufficient data to determine whether the higher risk reported through CASI translates into differences in testing, diagnosis, or treatment, but this is an important issue that needs further study.

The completeness of the different questions was reported in 3 of the 5 published clinical studies. Kurth et al reported that up to 12% of questions had responded with either “don't know” or “refused to answer” to certain questions about sexual behavior but in general most were refused by only a small percentage.14 Ghanem et al reported that 16% of those undertaking CASI had left questions on sexual behavior blank while this information was only missing in 0.3% of those seen face to face.15 Finally, Tideman et al reported that between 1% and 5% of questions were answered as either don't know or refused to answer.17

We could not identify any published articles reporting completion rates of CASI operating in routine practice in STD clinics but can report on our first year.19 The proportion of 8195 new clients answering CASI in 2009 who chose the option “declined to answer” was highest for women asked about working as a sex worker (2.45%) and relatively uncommon for questions about sex with someone of the same sex (women, 1.6%; men, 1.1%), and 0.95% for injecting drug use. It was lower for other questions (e.g., ever past STI, 0.39%) (unpublished data).

Time to complete CASI was reported in 3 published studies.14,16,17 The average time to complete the questionnaires depended on the number of items. Kurth et al used a questionnaire with at least 162 items which took an average of 16 minutes to complete.14 Rogers et al used a questionnaire with at least 80 items, but up to 128 items which took an average of 19 minutes to complete.16 Tideman et al used a much shorter questionnaire of up to 21 items that took 5 minutes to complete.17 In 2009 in our service where women receive up to 32 questions and men up to 35 questions, the average time for completing CASI for the 4934 men was 2.7 minutes (range: 44 seconds–24 minutes) and for the 3261 women was 3.2 minutes (range, 53 seconds–16 minutes) (unpublished data).

DISCUSSION

In this review, there was remarkable consistency between countries in a number of the key core sexual health questions either recommended by key organizations or routinely asked in STI clinics. Studies of CASI compared to in person interviews demonstrated that CASI identifies higher risk behavior more commonly than clinician interviews although there appear to be substantial differences between countries. STI symptoms are generally less commonly identified by CASI. CASI is generally acceptable to patients and relatively quick if confined to core sexual history questions. The consistency of questions together with the high acceptability of CASI is encouraging for its potential use in national or even international STI surveillance.

Our review had a number of weaknesses. First, we were only able to access published studies of CASI, and it is possible that a number of clinics have these systems operating but have not published the data although we understand from the national organizations approached that systems operating as part of routine care are rare. Second, the published studies of CASI involved study participants and not all clinic attendees and therefore may not be generalizable to entire clinic populations, although most studies had high participation rates.

In addition to the published information on CASI that we have presented in the results, there are a number of additional issues that need to be addressed if implementing CASI routinely. It should provide a similar sense of privacy for sensitive questions that in person consultations,20 should use secure hardware, and user-friendly software. The hardware must only allow inputting of data without permitting the user to access the main computer system or other clients' details. This can be achieved using a virtual keyboard as a substitute to the real keyboard that prevents exposing “system keys” (e.g., CTRL + ALT + DEL). Depending on the literacy of the clients who attend the centre, consideration should be given to the use of audio equipment with head phones, which provide a high degree of sound security because of the nature of the questions asked. The software needs to have a number of features including the ability to skip questions based on the answers to one or more previous questions and the ability to support different languages if CASI needs to be translated. In addition, when designing the screen format and questions for use in CASI, users should consider the large number of published information on human-computer interaction.21,22

There are a number of potential uses of routinely collected CASI data including its incorporation into decision support software to alert clinicians about recommended vaccines, overdue cervical smears or the need to address inadequate contraception. Decision support software has been shown to often significantly improve clinical practice.23 High-risk individuals could be offered online counseling programs for STI and HIV prevention or reminders for STI checks.24 During 2009, STI reminders were offered to 3287 MSM clients and 967 (29.5%) accepted.

Some services operate with CASI and self-collected samples only, without individuals going on to see a clinician. One UK clinic operates such a service for those who just need condoms, pregnancy tests, or tests for chlamydia or gonorrhea. At this clinic, 10% of chlamydia and gonorrhea tests are now self collected.25

There is the potential if all clinics used the same questions for CASI that it might form the basis of a detailed behavioral surveillance program. This could be particularly effective in a country like the United Kingdom, which has a network of over 300 genitourinary medicine clinics that see the majority of the population with STIs. Linking this extensive behavioral data with specific disease notifications could allow a detailed analysis of time trends in prevalence adjusted for risk factors.

CONCLUSIONS

Initial research has shown that CASI results in less social desirability bias than clinician interview, is more complete and is acceptable to the both clients and clinicians. For many clinics to devote the significant resources that CASI requires, it will need to improve the quality of sexual health care or improve its efficiency. Although these appear likely, there is currently no direct-published evidence to support them, although it is reasonable to expect that it will. Considering the critical funding issues for sexual health care in a number of countries at present, a broader research agenda for determining the clinical utility and health systems efficiencies of CASI-collected sex histories and their use is imperative.26 One example of allowing low-risk individuals (as assessed by CASI) to self-collected samples without seeing a clinician which would free up clinician time to manage individuals with more complex needs, while also potentially saving considerable cost.

REFERENCES

1. Keehan S, Sisko A, Truffer C, et al. Health spending projections through 2017: The baby-boom generation is coming to Medicare. Health Aff (Millwood) 2008; 27:w145–w155.
2. Bilardi JE, Fairley C, Hopkins CA, et al. Let them know: Evaluation of an online partner notification service for chlamydia that offers email and SMS messaging. Sex Transm Dis 2010; 37:563–565.
3. Levine DK, Woodriff AJ, Mocello AR, et al. inSPOT: The first online STD partner notification system using electronic postcards. PLoS Med 2008; 5:1428–1431.
4. Calderon Y, Leider J, Hailpern S, et al. A randomized control trial evaluating the educational effectiveness of a rapid HIV posttest counseling video. Sex Transm Dis 2009; 36:207–210.
5. Menon-Johansson AS, McNaught F, Mandalia S, et al. Texting decreases the time to treatment for genital Chlamydia trachomatis infection. Sex Transm Infect 2006; 82:49–51.
6. Camberwell Sexual Health Centre, Denmark Hill, London. King's College Hospital NHS Fundation Trust. Available at: http://www.kch.nhs.uk/services/specialist-medicine/sexual-health/camberwell-sexual-health-centre/. Accessed September 23, 2009.
7. Noar SM, Black HG, Pierce LB. Efficacy of computer technology-based HIV prevention interventions: A meta-analysis. AIDS 2009; 23:107–115.
8. Kurth AE, Holmes KK, Hawkins R, et al. A national survey of clinic sexual histories for sexually transmitted infection and HIV screening. Sex Transm Dis 2005; 32:370–376.
9. French P. BASHH 2006 National Guidelines—consultations requiring sexual history-taking. Int J STD AIDS 2007; 18:17–22.
10. Tideman RL, Pitts M, Fairley CK. Use of Delphi sorting technique to establish a core sexual history. Int J STD AIDS 2006; 17:170–172.
11. Sexually Transmitted Diseases Treatment Guidelines. Atlanta, GA: Centers for Disease Control and Prevention, 2006.
12. National coalition of STD directors. Available at: http://www.ncsddc.org/about/index.xml. Cited February 24, 2010.
13. Tideman RL, Pitts MK, Fairley CK. Current practice of sexual history taking by sexual health physicians in Australia and New Zealand. Sex Transm Infect 2004; 80:389–391.
14. Kurth AE, Martin DP, Golden MR, et al. A comparison between audio computer-assisted self-interviews and clinician interviews for obtaining the sexual history. Sex Transm Dis 2004; 31:719–726.
15. Ghanem KG, Hutton HE, Zenilman JM, et al. Audio computer assisted self interview and face to face interview modes in assessing response bias among STD clinic patients. Sex Transm Infect 2005; 81:421–425.
16. Rogers SM, Willis G, Al-Tayyib A, et al. Audio computer assisted interviewing to measure HIV risk behaviours in a clinic population. Sex Transm Infect 2005; 81:501–507.
17. Tideman RL, Chen MY, Pitts MK, et al. A randomised controlled trial comparing computer-assisted with face-to-face sexual history taking in a clinical setting. Sex Transm Infect 2007; 83:52–56.
18. Richens J, Copas A, Sadiq ST, et al. A randomised controlled trial of computer-assisted interviewing in sexual health clinics. Sex Transm Infect 2010; 86:310–314.
19. Fairley CK. Communication technologies and STI care. 18th International Society for STD Research (ISSTDR) in conjunction with British Association for Sexual Health and HIV (BASHH) 2009. Available at: http://www.stdpreventiononline.org/index.php/resources/resources/. Accessed September 24, 2009.
20. Centre for Disease Control and Prevention—Department of Health and Human Services. Program operations guidelines for STD prevention medical and laboratory services. Available at: http://www.cdc.gov/std/program/medlab/TOC-PGmedlab.htm. Cited February 9, 2010.
21. Olson GM, Olson JS. Human-computer interaction: Psychological aspects of the human use of computing. Annu Rev Psychol 2003; 54:491–516.
22. Shneiderman B. Designing the User Interface: Strategies for Effective Human-Computer Interaction. 3rd ed. Reading, MA: Addison-Wesley, Publishing, 1998.
23. Garg AX, Adhikari NK, McDonald H, et al. Effects of computerized clinical decision support systems on practitioner performance and patient outcomes: A systematic review. JAMA 2005; 293:1223–1238.
24. Gray RT, Hoare TA, Prestage GP, et al. Frequent testing of highly sexually active gay men is required to control syphilis. Sex Transm Dis 2010; 37:298–305.
25. Brady M, Kumar U, Baritser P, et al. Is one-stop-shop possible? The development of integrated sexual and reproductive health services. In: The British Association for Sexual Health and HIV and the American STD Association (BASHH ASTDA) Meeting; 2008; New York, NY. Abstract 0–26.
26. Golden MR, Kerndt PR. Improving clinical operations: Can we and should we save our std clinics? Sex Transm Dis 2010; 37:264–265.
© Copyright 2010 American Sexually Transmitted Diseases Association