National audit of asymptomatic screening in UK genitourinary medicine clinics: case-notes audit

INTRODUCTION

This paper accompanies the report on the audit of clinic policies on asymptomatic screening, also conducted in 2009 by the National Audit Group (NAG) of the British Association for Sexual Health and HIV (BASHH) in the UK. ¹

OBJECTIVE AND GOAL

The objective was to measure the level of adherence to national standards and guidelines for asymptomatic screening in UK genitourinary medicine (GU) clinics, to allow comparison against average UK performances, between regions and between services within regions. The goal of the audit was to improve asymptomatic screening in the UK.

METHODS

The NAG, comprising the chairs of Regional Audit Groups in the UK, a representative from the British HIV Association (BHIVA), and two NHS Trust clinical audit experts, planned the audit methodology and organization. The guidelines audited against were the BASHH Sexually Transmitted Infections: UK National Screening and Testing Guidelines. ² These guidelines recommend screening for chlamydial, gonorrhoeal, syphilis and HIV infections in all heterosexual men, men who have sex with men (MSM) and women (standard 100%). For chlamydial screening, the following is recommended: urethral or urine nucleic acid amplification testing (NAAT) in heterosexual men and MSM; cervical, vulvovaginal or urine NAAT (if a urethral specimen is not available) in women; and rectal chlamydial NAAT in MSM, if indicated by sexual history. For gonorrhoeal screening, the following is recommended: urethral culture or urine NAAT (if a urethral specimen is not available) in heterosexual men and MSM; cervical culture in women; and rectal and pharyngeal cultures in MSM, if indicated by sexual history. Use of NAATs for gonorrhoea screening is not recommended for invasively collected samples, although may be used for urine and non-invasively collected samples when direct sampling of mucosal surfaces is not possible. Blood is the recommended sample for syphilis (EIA [enzyme immunoassay] or TPPA [Treponema pallidum particle agglutination] or cardiolipin test plus TPHA [Treponema pallidum haemagglutination]) and HIV testing (EIA). Hepatitis B screening is also recommended for all MSM, although testing may include all of hepatitis B surface antigen, anti-hepatitis B core antibody and anti-hepatitis B surface antibody, or different algorithms for testing. Two questionnaires were designed by the NAG to audit practice in asymptomatic screening: one on management recorded in case-notes, and the other on clinic policy; both were translated into web-based forms. Design of the questionnaires was led by a subgroup of Regional Audit Chairs, who constructed questions to test practice against the guidelines. This included having the questions tested against actual case-notes material by clinic staff who reported on findings to the Regional Chairs. This process was re-iterated until there was agreement that the questions were useful. Finally, the questions were re-tested on a prototype web form by the subgroup, until final versions of the questions were agreed. Both questionnaires are available on the BASHH website. ³

An asymptomatic case was defined as a patient attending a new or re-book episode, who did not offer any symptoms on presentation, either on a triage form, or similar form, or on direct questioning by a clinician. For the case-notes audit, the sample number for each centre was the first 30 patients seen during 1 January 2009 to 31 March 2009. Data were submitted electronically between May and July 2009 using Feedback Server software ⁴ and downloaded for analysis using Microsoft Access and Excel. Pseudonymized (non-identifiable) data were submitted with an audit number, which clinics were asked to keep matched with their own record of the case clinic number (to keep count of cases submitted and to check on data, if necessary). Data were collated and aggregated by region, and by clinics within regions. This study reports on overall national performances, as well as ranges of regional performances expressed as percentage values. Detailed data on regional performance for both case-notes and the clinic policy part of the audit, allowing comparisons between regions, are posted on the BASHH website (see below). Individual clinics have been provided with performance data included within regional aggregates of data, allowing comparisons between clinics within regions.

RESULTS

The figures given are national totals and percentages, and regional ranges expressed as percentages.

DISCUSSION

The interpretation of the results of this audit should be viewed in the light of several limitations. With regard to questionnaire design, no formal test of reliability, such as test–retest reliability with clinic staff, was carried out; this is an area of methodology that the audit group should seek to improve. An important number of clinics (around 40%) did not contribute data, with less contribution of cases from regions with a smaller number of clinics; however, participation compares well with previous national audits. More work is needed to explore the lack of involvement of clinics that manage smaller numbers of patients. Collection of more detailed sexual history data may have enabled a better understanding of the appropriateness of some reported testing practice. Little is known about patient, clinician or system preferences that might have determined the choice of cervical gonorrhoea NAATs over culture, or the factors influencing the choice between urine or vulvovaginal NAATs. It is also not known whether NAAT specimens testing positive for gonorrhoea were confirmed with culture, which allows sensitivity testing and guides appropriate choice of an antibiotic, as recommended in the guidelines, as well as supplementary testing with a different nucleic acid target to provide a positive predictive value of >90%, as recommended in the more recent guidance for gonorrhoea testing in England and Wales ⁵ (although the latter was not available at the time the audit was carried out). Also, our case definition did not refer to factors that may have led to microscopy being performed in asymptomatic patients. Similarly, no data were collected that might have helped understand the appropriateness of some of the other tests performed, e.g. hepatitis B testing in groups other than MSM. These limitations are discussed further below. Other possible explanations for cases not having recommended screening tests documented include recent screening by other agencies, problems with specimen and results processing, and patient factors, such as patient preference for not having tests and needle aversion.

Compliance with the guidelines was good with regard to offering screening tests for chlamydia and gonorrhoea, with 96% of men and 98% of women tested for chlamydia, and 93% of heterosexual men, 94% MSM and 93% of women tested for gonorrhoea, respectively. However, with regard to testing method, there is significant use of NAATs for gonorrhoea. Although culture of samples collected from mucosal surfaces is the first-line recommended test for gonorrhoea, four regions used urine NAATs more frequently in MSM compared with urethral culture. However, this audit did not collect data that would have helped understand the determinants of this departure from the guideline; for example, it is possible that these were frequently tested groups for whom repeated urethral sampling is less acceptable. Five regions (including the four above) performed cervical or vulvovaginal NAATs more often than cervical culture. Again, the reasons for not doing culture testing is not known, and further work is needed to determine whether factors existed, such as lack of transport, clinic equipment or laboratory support for doing culture; patient preference or clinic preference owing to increased sensitivity; and convenience of NAATs. Better acceptability for patients, in particular, may be an important factor, ⁶ and although the findings in this audit suggest that obtaining non-invasive specimens for NAATs for gonorrhoea screening is often practised, little evidence exists in the UK as to whether screening based on this methodology would be more acceptable and lead to more confirmed infection. In any case, the preponderance of gonorrhoea NAATs in some regions should be accompanied by gonorrhoea culture testing, especially in view of the emerging resistance to third-generation cephalosporins reported recently in The Gonococcal Resistance to Antimicrobials Surveillance Programme. ⁷ Culture confirmation of positive NAATs for gonorrhoea had been recommended, where supplementary NAATs were not available, ⁸ and this was reflected in the Guidelines. ² However, more recent guidance ⁵ on gonorrhoea testing (not available at the time of the audit) now recommends supplementary testing of positive gonorrhoea NAATs with a different nucleic acid target, and not culture (because of low sensitivity), although culture for gonorrhoea remains important for antibiotic sensitivity testing. There was a surprisingly large number of cervical chlamydial culture tests performed, and the reasons for this should be investigated.

HIV screening for all groups is well above the target in the 2001 National Strategy to increase the uptake of the test to 60% by the end of 2007. ⁹ However, there is considerable inter-regional variation, with up to 35% and 26% difference in testing rates for asymptomatic women and heterosexual men, respectively, contrasting with much more consistent testing rates for MSM. Although not directly comparable, the rates of HIV screening in this audit suggest improvement compared with the rates of testing in the BASHH 2007 audit of 5032 people with chlamydial infection: ¹⁰ 83% versus 67%, 94% versus 73% and 81% versus 65% for heterosexual men, MSM and women, respectively.

It is not possible to report on the order in which tests for syphilis were carried out, since initial positive screening tests may have given rise to other tests being performed. However, the vast majority of tests were EIA tests. Again, although not directly comparable, the rates of syphilis screening in this audit suggest improvement compared with those of the BASHH 2007 audit of chlamydial infection: ¹⁰ 84% versus 71%, 97% versus 87% and 81% versus 67% for heterosexual men, MSM and women, respectively. Similar variation in testing for syphilis is seen as for HIV testing, again with more consistent testing for syphilis in MSM between regions. Routine screening for hepatitis B is recommended for all MSM, but in six regions the screening rate was less than 80% for MSM not known to be immune. The relative predominance of hepatitis C testing in MSM may result from awareness of increased risk of hepatitis C transmission in MSM who have HIV infection, ¹¹ although this audit did not ask about other risk factors for hepatitis C transmission.

Non-compliance with the guidelines with respect to non-recommended testing was observed, and an important minority of cases had urethral and/or cervical microscopy performed. However, this audit does not provide information about the determinants of performing microscopy in asymptomatic patients, including cases requesting examination that were found to have signs suggesting infection. Most men having urethral microscopy, and most women having cervical microscopy, also had gonorrhoea culture performed from these sites, raising the possibility that microscopy may not have been a primary test in some cases. The issue of offering male urethral microscopy has been particularly controversial ¹² and a recent regional audit ¹³ also showed that the majority of clinics were not complying with the guidelines by offering urethral microscopy to asymptomatic men. Rectal microscopy is seldom performed, in keeping with the guidelines. ² Also, almost 40% of asymptomatic women had tests performed for vaginal trichomonal infection, although screening for this infection is not recommended in the guidelines. ² A large number of women had urethral sampling for gonorrhoea and, although this is not recommended, it is possible that this practice is influenced by the expectation of detecting isolated urethral gonorrhoeal infection. ¹⁴ This site is suggested by the guidelines for hysterectomized women. It is possible that management of contacts of infections may play a role in some of the testing outside the guidelines discussed above.

Finally, the poor recording of information about oral and anal sex concurs with the findings of the 2008 BASHH audit on sexual history taking. ¹⁵ Tests from the throat and rectum for both chlamydia and gonorrhoea were carried out more often in MSM and women when receptive oral and anal sex was documented, with particularly high rates of throat testing for MSM who were documented as having practised receptive oral sex. BASHH guidelines on sexual history taking recommend asking about the anatomical site of sexual contact to guide which sites need to be sampled. ¹⁶ There is further discussion of oropharyngeal and rectal screening in MSM in the clinic policies part of the audit.

CONCLUSIONS

SUGGESTED AREAS FOR PRACTICE IMPROVEMENT/INTERVENTIONS

The following areas of practice are suggested as areas for interventions to improve practice:

Increased documentation of discussion about oral and anal sex, as recommended in the BASHH recommendations on sexual history taking, to identify which anatomical sites need to be sampled for infection; ¹⁶