Abstract
In recent years, the sexual health of the nation has risen in profile. We face increasing demands and targets, in particular the 48-hour waiting time directive, and as a result clinic access has become a priority. eTriage is a novel, secure, web-based service designed specifically to increase access to our clinics. It has proved a popular booking method, providing access to 10% of all appointments across the Directorate within six months of introduction. KC60 analyses revealed that the majority of users (58%) underwent asymptomatic screening with the remainder having some degree of pathology. There was a greater percentage prevalence of human papilloma virus, chlamydia, non-specific urethritis, gonorrhoea, herpes and trichomonas in the eTriage population when compared with the general clinic population. A notes review illustrated a high degree of concordance between data entered on eTriage registration and clinical review (97%). A patient survey revealed high levels of patient satisfaction with the service. As an adjunct to our existing booking services, eTriage has served to increase patient choice and has proved itself to be a safe, efficient and effective means of improving patient access.
INTRODUCTION
Identified as one of the Selbie six, the sexual health of the nation has been highlighted as an area in need of central and local investment and modernization. 1 Patients requiring sexual health review are often limited by difficulties with clinic access. 2–4 As a result, the Department of Health has introduced the 48-hour waiting time target where, by April 2008, any individual requiring an appointment with a sexual health service should be offered an appointment within two working days. Genitourinary (GU) medicine clinics were audited by the health protection agency (HPA) to ensure these targets are met. In order to meet government targets, the National Health Service (NHS) has recommended, in accordance with its Plan, 5 that novel use of information technology (IT) should be employed.
Chelsea and Westminster NHS Foundation Trust is a major provider of sexual health services. The sexual health department is comprised of three GU medicine clinics, which offer both appointment and emergency walk-in services. In 2006, the number of local clinic attendances rose by 8%. 3 Historically, individuals made appointments with our clinical services in person or on the telephone, via our call centre. From patient feedback, barriers to making these appointments have been identified as the amount of time spent on the telephone, finding the call centre opening times inconvenient, and not wishing or being unable to vocalize symptoms over the telephone. eTriage was developed in an attempt to address and remedy some of these issues and assist with the increased demand for our services.
System design
eTriage is a novel, web-based triage and booking service designed specifically to increase access to our clinics. Developed in conjunction with the UK-based IT company Mikkom Ltd (Windsor, UK) clinical, reception, administrative and patient input, this software facilitates direct and appropriate access to GU medicine appointments. Patients enter the system via the Internet or the web function of their mobile telephones. They provide basic personal information and answer simple questions; prompts are provided in order to help individuals describe more complex symptomatology. A medical algorithm analyses the data entered by the patient and triages them so that they are offered an appropriate appointment: either routine, urgent or within 48 hours. A patient profile is generated along with a triage rating and this is combined with the patient's preference of clinic and mode of communication. The patient's demographic and sexual health data are stored on a secure server behind an NHS firewall. A text or email message containing appointment details is sent to them within one working day with instructions on how to cancel their appointment if necessary.
The service is available to patients aged 16 and above. If the eTriage responses indicate a potentially serious or time-sensitive condition, e.g. ectopic pregnancy, testicular torsion, severe pain, pelvic inflammatory disease, a need for emergency contraception or post-exposure prophylaxis following HIV exposure, these patients are advised to attend one of our clinics urgently or the Accident and Emergency Department, if out of hours. In these situations the system defaults to the relevant advice page and the patient is unable to progress further through the system. Patients indicating a history of recent sexual assault are given appropriate information and service contact details.
The eTriage system was evaluated for accuracy, ease of adoption, appointment outcomes, user characteristics and acceptability following database interrogation and a patient survey.
Testing and governance
Following the principles of the Caldicott committee report, 6 steps were taken to ensure the protection of patient information and security. The eTriage database was designed to collect the minimum amount of required data, and only specific individuals have access to the records.
METHODS
System safety analysis
Prior to introducing the eTriage system, an imitation database was created, using staff members as patients to ensure that the system triaged patients appropriately and to provide a platform for future monitoring and quality control following any system changes. Staff were asked to provide details of their eTriage experience. This served as a mechanism for ensuring that time sensitive or more urgent medical problems were directed appropriately and that individuals received notification of their appointment at the clinic of their choice, by text or email as requested, within one working day. It also ensured that the reception staff had sufficient experience of the system prior to ‘going live’.
Thirty members of staff were enrolled. Each was provided with a GU problem and registered on the database. They were asked to provide feedback once they had been texted or emailed an appointment or directed to attend services more urgently. Each case was reviewed to ensure that individuals were provided with appropriate advice/appointments. When changes were made to the eTriage system, it was interrogated using the imitation database.
Activity and outcomes
The eTriage database was interrogated to reveal the total number of hits on the website over the four-month period following its introduction in October 2006. Specifically, we detailed:
Web hits: in and out of hours; Number of appointments sent; Demographic data; Time between logging on to the system and clinic appointment; Chosen method of appointment notification (email or text); Type of appointment requested e.g. asymptomatic or symptomatic review.
For comparison, the website activity in July 2007, nine month's following the eTriage introduction was also documented.
The outcome of attendance using KC60 diagnosis data was also recorded and compared with the KC60 returns of the general clinic population.
Audit
A review of the case-notes pertaining to the first 175 individuals who had booked their appointment via eTriage and subsequently attended the John Hunter Clinic for Sexual Health was performed and compared with data captured by eTriage. This was used to investigate concordance between the details given on eTriage and those recorded during clinical consultation.
Patient questionnaire
For those patients booked by the eTriage system, we collected feedback using a questionnaire-based survey completed on attendance. We recorded the following:
Demographic data; Time between registration and appointment notification (patient perception); How patients had heard of eTriage; Whether patients liked the website; If they found it easy to understand; If they would recommend the site; If they would use it again.
RESULTS
Activity and outcomes
System testing results
Launched in October 2006, with no supplementary promotion, the eTriage website sustained 1516 hits during its initial four-month period. The mean number of hits per month was 379 (range 293–423). This included 909 appointment requests (60%), of which 460 (51%) were made out-of-hours (see Table 1). In comparison, in July 2007, there was a total of 604 hits on the website (159% increase compared with mean hits per month Oct 06–Jan 07), of which 205 (34%) were made out-of-hours. Of the 604 hits, 469 appointments were made (78%). When reviewing the overall increase in eTriage uptake, appointments made via eTriage in October 2006 accounted for 3.5% of all appointments, this rose to 10% by July 2007.
Percentage prevalence comparison of the KC60 diagnoses in eTriage attendees with the general clinic population during the period of October 2006–January 2007 inclusive
HPV = human papillomavirus; NSU = non-specific urethritis; HSV = herpes simplex virus; TV = Trichomonas vaginalis
From the initial four-month data, there were 909 appointment requests representing 847 patients. In total, 352 appointments were cancelled, of which 264 were rebooked. This provided a total of 591 patients seen via the eTriage system in the initial four-month period.
Of the initial 909 appointment requests, 510 (56%) were made by women, a significantly different population when compared with the regular clinic attendees where 51% (9041) were men (P < 0.01). The mean age of the eTriage population was 28.5 years (17–68), compared with 40 years (12–98) in the regular clinic population for the same period. The average waiting time from logging on to appointment was 2.6 days (inclusive of weekends). For notification and subsequent appointment reminder, 592 (65%) favoured a text message. Of those requesting an appointment, 301 (33%) described GU-related symptoms, 46 (5%) had been in contact with a sexually transmitted infection (STI) and the remainder requested a routine screen with or without family planning review.
KC60 returns were available for 563 clinic appointments (69% of total number of appointments). The KC60 returns showed that 58% (326) of eTriage attendees underwent asymptomatic screening compared with 23.8% of the general clinic population. However, many diagnoses of pathology were also made, including:
23 wart presentations (4%); 14 cases of chlamydia (3%); 18 episodes of non-specific urethritis (NSU) (3%); Five cases of gonorrhoea (1%); Four new presentations of herpes infection (1%); Four cases of pelvic inflammatory disease (1%); Two cases of trichomonas infection (0.5%); One transfer of HIV care (0.2%).
Diagnoses of urinary tract infection, bacterial vaginosis and STI contacts comprised the remainder. Comparison of KC60 diagnoses in the eTriage population with the general clinic population is shown in Table 1.
Notes review
The notes review revealed no patient with potentially serious conditions or time-sensitive needs was inappropriately given an appointment. There was a high concordance between the data entered on eTriage and the clinical details recorded. Of 175 notes reviewed, it appeared initially that 150 (86%) of patient histories positively correlated with eTriage records. Of the 25 individuals where the clinical documentation did not appear to correlate with eTriage data, the notes review highlighted that 19 individuals (11%) had symptoms or signs in clinic which they had not divulged on registration. Some of these individuals, however, were unaware of their infection, e.g. genital warts (2), Molluscum contagiousum (1), bacterial vaginosis (5), NSU (3), candida (2), balanitis (1) and urinary tract infection (1). Four of the 19 (2% of the 175 patients undergoing notes review) were STI contacts which they had not disclosed on registration.
The remaining six patients (3%) were found to be asymptomatic in clinic, having described symptoms on registration. Four (2%) of these individuals stated they had ulcers on the eTriage system but on attendance described rashes that had resolved. There is no facility to state that one has a ‘rash’ on the eTriage form; thus, one might think that ‘ulcer’ would be the most helpful description to highlight. Of the remaining two, one had used the ‘rectal pain’ symptom to describe a history of chronic pyelonephritis which was quiescent on review. Another described testicular pain but denied any symptoms on review. Thus, the overall concordance rate was likely to be much higher −97% (170).
Patient survey
Data from the patient survey are shown in Table 2.
Results from patient questionnaire
DISCUSSION
During the first four months over 900 appointments were made via eTriage. More than half were requested out-of-hours, when no other means of contacting the appointment system is available. The ‘triage’ ability of the system worked well, ensuring that all individuals were seen quickly where clinically indicated, with the majority offered appointments within 48 hours. There was a high rate of cancellations and rebooks initially. This, in part, was due to the requirement to meet the HPA waiting time target of offering all patients an appointment within 48 hours.
There had been an initial concern that the system would be used by asymptomatic individuals only, rather than those requiring more urgent review. However, the KC60 data showed that 42% of eTriage users harboured pathology, with over one-third of these patients describing symptoms at the time of registration. Interestingly, when comparing the most common STIs, there was a higher prevalence of human papilloma virus, chlamydia, NSU, gonorrhoea, herpes simplex virus and Trichomonas vaginalis in the eTriage population than in the general clinic population during the October 2006 to January 2007 period.
All individuals requiring urgent medical review were directed to the appropriate service and those with more concerning GU symptoms were seen within 24 hours where necessary. Thus, eTriage has proved to be safe, consisting of questions posed in such a way that the system could extract accurate data from users and triage appropriately. The rate of correlation between the eTriage registration forms and clinical review was higher than in other reported studies. 7,8
The patient satisfaction questionnaire proved a useful tool to illustrate that the majority of users liked the website, found the system easy to use, would use it again and would recommend it. The free text section highlighted the fact that individuals would like a varied choice of appointment. However, as with all GU services, we are constrained by the 48-hour waiting time target. Plans are in place to develop the system to allow greater patient choice within this constraint.
eTriage may not be a suitable booking method for all patients. It is not ideal for individuals with restricted Internet access or for those with language or reading difficulties. There is potential to alter the system to make it language specific, bettering access for certain communities.
In providing an alternative booking method, we would hope that demand on the call centre should reduce, improving the waiting time for individuals using this method of clinic access. In May 2006, prior to eTriage introduction, the percentage of patients offered an appointment in 48 hours within our unit was 48%, this increased to 100% by May 2008. Historically, patients telephoning with symptoms requesting urgent review were transferred to the clinic floor where doctors, nurses and health advisers would be responsible for booking relevant appointments. Thus, the introduction of eTriage has impacted positively upon clinical and call centre staff time in releasing clinic capacity.
CONCLUSION
Complementing the existing booking systems, eTriage has served to increase patient choice and has proved itself to be a safe, efficient, effective means of improving patient access, with high levels of patient satisfaction. It serves as an excellent example of the collaboration between IT and health providers, in accordance with the targets laid out in the NHS Plan.
Footnotes
Acknowledgements