Table 1 Data items collected by the survey tool by source and service type and rationale for collection.

Questionnaire and clinical data extract

Gender, age (date of birth and date of attendance used to calculate age in years from clinical data extract)

To find out the demographic profile served by the clinic/LESSH practice. Prevalence of STIs varies by age and gender.

Collect from both the questionnaire and clinical data extract in order to check the correct match has been made.

Questionnaire

Reason for attendance

Comparing reason for attendance with the patient's care pathway (duration, any other services used) can be informative. For instance, patients attending because of their own symptoms, or because a partner has been diagnosed with an STI, should ideally be accessing services fast.

This also informs the proportion of patients attending due to clinic- or patient-led partner notification (i.e. people attending for new episodes of care who report that they were called in by the clinic, or that their partner has been diagnosed with an infection, as reasons for attendance).

Duration of care-seeking

This can help identify groups of patients for whom access is difficult (i.e. lengthy care-seeking).

If the patient has an STI, delays in care-seeking (and therefore time spent infectious) can influence the progress of the disease, and the likelihood of transmission (depending on sexual behaviour).

Use of other services (and if relevant, type/name of service, how long ago and what happened)

We can use this information to assess whether there has been duplication of effort in a patient's care pathway, and/or whether their STI could have been detected and treated by the previous service(s) they contacted. We will also be able to see which services refer patients on (and assess whether this is likely to be justifiable).

(Proviso: obviously we will not have data from patients who abandoned care-seeking and so were never surveyed).

Number of male/female sexual partners in the last year;

Number of these who were new partners;

Number of partners in the last 3 months;

Number and gender of recent partners influences STI risk. It is also informative to find out whether people with many partners, and men who have sex with men, are more likely to attend particular services.

Regarding each partner in the last 3 months (up to a max. of 3):

- how long ago first had sex,

- how long ago most recently had sex,

- condom use,

- expectation of having sex again with this person;

How long ago each sexual partnership began and ended informs measurement of concurrency (partnership overlap), which influences transmission risk.

Expectation of having sex with a person again informs measurement of concurrency (above) and is an indicator of the likelihood of successful partner notification.

Condom use informs the extent to which the partnership was 'protected' and (if low) suggests in which groups within the service's users further health promotion on condom use might be needed

Whether the respondent had sex since recognising a need to seek care (for the reason attending the service), and if so:

- number of partners,

- number of new partners,

- number of occasions of sex,

- condom use

Together with STI diagnosis this informs the likelihood of transmission since the patient recognised the need to seek care.

This information can be compared with patients' reason(s) for attendance, presence of symptoms, and STIs diagnosed.

It may indicate a need for health promotion messages about abstaining from sex once the need to seek testing/care is recognised.

Whether the respondent has ever been diagnosed with an STI

Past STI diagnosis may affect future care-seeking.

Whether the respondent has ever had a Chlamydia test, and if so in what setting

A measure of past contact with Chlamydia screening services.

The care-seeking and demographics of patients who have tested for Chlamydia before, and specifically at a sexual health clinic, can be compared to those who have not.

Whether the respondent has symptoms now, and if so, duration of symptoms

Although many infections are asymptomatic, it is useful to know how patients with symptoms differ in their care-seeking to those without.

Whether the respondent is registered with a GP

Patients registered with a GP may have more opportunity to seek sexual health care from primary care (including LESSH) than those not registered.

LESSH only:

- whether the patient attended for a booked appointment or as a 'walk-in' patient

- whether the LESSH GP is the patient's own GP - and if not, the name of the patient's GP surgery

The type of appointment can influence the length of time a patient waits (in contrast we know that once they contact a clinic, most GUM patients are seen within 48 hours)

Patients registered with a practice providing a LESSH may be more likely to attend the LESSH. It is also important to know whether patients not registered at a LESSH practice benefit from the service, and if so how many.

Clinical data extract

STIs tested for;

STIs diagnosed;

Knowing which STIs were tested for informs which STIs could have been diagnosed.

Patients with STIs can be compared to patients without STIs, by demographics, reason(s) for attendance, care pathway duration, etc.

Whether patient is already known to be HIV positive

As HIV testing is not relevant for patients known to be HIV positive

Partner notification outcomes for patients diagnosed with Chlamydia or gonorrhoea (at least one partner tested; at least one partner treated)

These indicators of partner notification assess the extent to which the service is managing to notify the partners of patients with common STIs.

LESSH only:

- whether microscopy was performed;

- treatment received

- referral to other services

We know that GUM services have facilities for microscopy but some LESSH may not have this facility. Similarly, we wished to measure the appropriateness of treatment received by doctors whose main work is not sexual health. We also wanted to measure how commonly patients were referred on for problems that could have been dealt with by the LESSH (for instance if practices or contracts changed), particularly referral to GUM.

GUM only:

- PCT of residence

As LESSH are advertised within a PCT, we assumed the majority of patients lived in the same PCT. GUM clinic users may travel from further afield; data on PCT of residence is routinely collected in GUM.

Source differed by setting

Collected in clinical data extract in GUM but questionnaire in LESSH:

- ethnicity;

- new/follow-up attendance status;

- past attendance at the service (in LESSH services patients were asked to state whether this was for a sexual health or other primary care reason, or both);

- name of GP surgery

Ethnicity is routinely by GUM clinics, but not necessarily by LESSH services. STI prevalence varies by ethnicity, and ethnicity can be used to assess whether any groups are underserved, attend for different reasons, or have longer care pathways.

GUM clinics routinely code whether patients are attending as new patients, rebook (i.e. patient attending for a new episode of care, but who has attended before), and follow-up (subsequent visits in the same episode of care). We gained this information from the questionnaire for LESSH patients, and additionally asked whether the patient had attended the practice for a reason not related to sexual health.

GUM clinics routinely ask patients their GP (though patients are not required to provide it), whereas this data may not be routinely collected by LESSH services. We could assess from this what PCT the patient's GP was in (which sometimes differed from PCT of residence), and whether the practice provided a LESSH. It is interesting to know whether patients whose own GP surgery provides a LESSH chose to attend this service or GUM.