January 2013
Outpatient Services: A literature search of evidence from the UK
Introduction
The aim of this report is to summarise evidence from the UK that relates to the following questions:
1. What evidence based interventions, or improvements in hospitals, primary care, GP Practices, community teams, or patients homes have been shown to reduce outpatient referrals and appointments?
2. What interventions have been shown to reduce any variation in outpatient referrals?
3. What interventions or improvements have been shown to increase efficiency or productivity in outpatient care or services.
4. What interventions have been shown to increase the efficiency, quality and patient experience of management and booking of outpatient appointments.
Methods
A systematic search of the secondary literature was carried out between the 3 and 10 December 2012. Key secondary resources were searched. Terms used for these searches included:
· Outpatient(s) or ambulatory care
AND
· Reduced referrals/appointments
· Reduced variation
· Increased efficiency/productivity
· Increased efficiency, quality and patient experience of management and booking of outpatient appointments
A focussed search of the following databases was conducted and not limited by study design:
· Medline
· Medline in process
· Embase
· HMIC
All databases were searched used the OVID SP platform. Results were limited to studies published from 2000-2012 in English.
A full list of resources searched and primary literature search strategies are available on request.
The literature was limited to studies that had a UK-focus.
Summary of Findings
The literature search identified a comprehensive scoping report1, a health technology assessment (HTA)2, a systematic review3 and 10 primary studies4-13. These can be classified under two broad headings:
- Those that discussed alternative approaches to traditional outpatient services.
- Those that discussed strategies that were used to improve productivity and efficiency of existing services.
The literature has been summarised under these headings.
Reorganisation and alternatives to outpatient treatment
The evidence for this section comes from a comprehensive scoping report1 and one primary study9.
The scoping report, published in 2006, focused on evidence relevant to the UK context1. The authors state that there are two approaches for reducing waiting times for specialist care. The first is to increase hospital capacity, to allow greater throughput of patients. The second is to reduce demand for specialist care by finding alternatives to outpatient treatment. The focus of the report was the second approach, in particular strategies that involved primary care. The review was based around four broad strategies:
- Transfer: the substitution of services delivered by hospital clinicians for services delivered by primary care clinicians.
- Relocation: shifting the venue of specialist care from outpatient clinics to primary care without changing the people who deliver the service.
- Liaison: joint working between specialists and primary care practitioners.
- Professional behaviour change: interventions intended to change the referral behaviour of primary care practitioners.
The authors of the scoping report highlight the lack of high-quality research in this area. Overall, they concluded that the literature broadly suggests that transfer and professional behaviour change strategies can be effective for reducing outpatient demand, but relocation and liaison are largely ineffective.
With regards to the ‘transfer’ model, the scoping report highlights the following two ways as effective in transferring outpatient care to primary care:
- Discharge of patients to (i) no follow-up (ii) patient-initiated follow-up or (iii) general practice follow-up as alternatives to routine follow-up in hospital outpatient clinics.
- Direct access for GPs to (i) hospital-based diagnostic tests and investigations or (ii) hospital-provided treatments, without the prior approval of a specialist in an outpatient clinic.
In addition, they highlight two further approaches that ‘merit further investigation’:
- General practitioners with special interests acting as substitutes for outpatient specialists. Impact on cost and effectiveness appears to be context dependent and merits further investigation.
- Transfer of medical care for common chronic conditions, like diabetes, from secondary to primary care. The potential of moving management of a wider range of chronic conditions from hospitals to primary care warrants further investigation.
With regards to the ‘professional behaviour change model’, the authors identified two strategies as effective in reducing inappropriate referral to outpatient clinics:
- Structured referral sheets that prompt GPs to conduct any necessary pre-referral tests or treatments.
- Educational outreach by specialists.
In addition, they identified a further approach that they considered ‘merits further investigation’:
- In house second opinion prior to referral was shown to reduce outpatient referral without adversely affecting the quality of care in one study. Further studies are needed to assess the reproducibility of this finding.
Finally, the authors summarised interventions that the published evidence has suggested are ineffective:
- Relocating specialist services to primary care was generally associated with improved access for patients. Greater equity in care provision may be achieved by relocating specialists to communities with poor access to secondary care services (eg remote rural areas). Locating specialists to well served communities was associated with reduced outpatient effectiveness and efficiency. Specialist attachment to primary care teams was shown to reduce outpatient attendance for only one of three specialities evaluated (physiotherapy).
- Joint working between primary and secondary care clinicians may improve the quality of care, but appears to have little impact on outpatient attendance.
- Certain professional behaviour changes are ineffective in changing the referral behaviour of primary care clinicians ie passive dissemination of referral guidelines; audit-and-feedback of referral rates; and discussion of referral behaviour with an independent medical advisor.
The primary study was retrospective and ran from January to July 2003 in Glasgow9. The aim was to compare patient management and outcomes from a newly developed one-stop post-menopausal bleeding (PMB) clinic with four traditional consultant-led outpatient gynaecology clinics also seeing women with PMB. The one-stop clinic was run by a single consultant and nursing assistant.
In the study period, 95 women were referred to the one-stop clinic and 51 to the general clinics. There was no difference in patient demographics but the time from referral to first consultation was shorter in the PMB clinic (p<0.001) and women had fewer visits (p<0.001)9. The mean time from first consultation to definitive treatment or discharge was also shorter (p<0.001). Fewer hysteroscopies were generated from the PMB clinic (p<0.0001) and yet there was no difference in the rates of abnormal histology between the two groups.
Based on these findings, the authors concluded that one-stop investigation of PMB, compared to traditional outpatient clinics, reduced waiting times and theatre costs by reducing the number of hysteroscopies9. While this study was retrospective and therefore susceptible to inherent bias, the authors argued that the data represented a true reflection of current management.
Improving quality, efficiency and productivity of outpatient services
Telephone and SMS reminders
A recent systematic review (2011)3 aimed to evaluate whether telephone and short message service (SMS) reminders improved attendance at hospital appointments. The authors present a summary of 29 studies (including nine randomised controlled trials (RCTs)), with 33 intervention arms. Eighteen of the interventions were based on manual reminders (ie phonecalls made by health staff) and 15 were based on automated reminders (ie automated phone messages or SMS).
The median did not attend (DNA) rate reported at baseline was 23%3. All studies but one reported that the intervention improved DNA rate. The synthesis suggested that the weighted mean relative change in non-attendance was 34% of the baseline non-attendance rate (the authors used the relative change in DNA, rather than the absolute change, to compensate for the different baseline DNA rates in the different studies). Automated reminders were less effective than manual phonecalls (29% versus 39%; relative change of baseline value). There appeared to be no difference in non-attendance rate, whether the reminder was sent the day before the appointment or the week before. Further, nearly half of the studies included cost estimates, which suggested that reminders cost less than €0.50 per patient for SMS or automated reminders.
While this review is of reasonable quality, there are a few limitations that should be noted. The literature search was limited to Medline, and so some studies may have been missed. The quality of the included studies was also low, with a median score of 7 (out of 14). The authors included studies from any country, and only eight were from the UK. However, the results are still likely to be generalisable to the UK context.
Published after the systematic review, a further small study (n=55) from Ireland considered the use of text messaging to improve outpatient service efficiency13. Text messages were sent to outpatients two weeks after discharge. Patients who replied that they were well were discharged. If there were no replies after two attempts, patients were scheduled for the next outpatient clinic. Further, the mobile phone was carried by different members of the surgical team, and patients in the study were able to contact them if they had any concerns or queries.
Over a 4-month period, 74.5% of patients were discharged from follow-up using text message surveillance. A survey of patients suggested that this approach was the preferred method of surveillance. Overall, there was a 13.6% decrease in outpatient visits. Also, allowing patients to phone the surgical team directly resulted in the avoidance of six emergency department visits.
The authors of the study note that the sample size is small, and at these numbers it may be difficult to detect a low adverse outcome rate among participating patients.
Patient booking systems
Two papers were identified that evaluated changes to patient booking systems7,12.
The first was a retrospective comparative audit from an audiological medicine clinic in Manchester12. It examined the effect of a ‘choose and book’ system on outpatient non-attendance.
‘Choose and book’ is a national electronic referral and booking system which allows patients a choice of place, date and time for their first outpatient appointment12. This study examined referrals from an audiological medicine department that came from both traditional and ‘choose and book’ sources.
During the 7-month period analysed, a total of 416 new patients had appointments to attend the clinic12. Bookings were made traditionally in 225 (54.1%) of patients, and with ‘choose and book’ in 191 (45.9%). ‘Choose and book’ patients had a significantly better rate of attendance (82.2%) than traditionally booked patients (69.8%) (mean difference 12.4%. 95% confidence interval (CI) 4.3 to 20.5; p<0.001). There was no significant difference between the two groups in terms of sex (p>0.1). The ‘choose and book’ patients, however, were significantly older than the traditional appointment patients (p<0.001). Other than age and sex, the author did not compare the demographics of the ‘choose and book’ and ‘traditionally booked’ populations.
The second article described the implementation of a new patient booking system in an orthopaedic department in Wales7. The system was introduced in July 2001, and involved the patient telephoning to make an outpatient appointment rather than automatically receiving a date through the post. The patients received a letter asking them to make a booking. If there was no reply after 10-days, a second letter was sent and followed by a phonecall 7-days later if there had still been no reply. The patient was then either allocated an appointment or removed from the waiting list, as appropriate.
Over the 6-month period before the new system was in operation, the nonattendance rate was 17.6% (21 out of 119 new patients)7. Between July 2001 and January 2002, following implementation of the patient conformation system, 176 (out of 214) new patients telephoned to ‘confirm’ their appointment and only seven of these failed to attend the clinic. This brought the nonattendance rate down to 4%. Thirty-eight (out of 214) patients were removed from the waiting list: eighteen felt that their condition had improved or were seen privately, 18 were not contactable, and two had died. The authors stated that the telephone booking system is now being rolled out to include outpatient clinics of other specialities and orthopaedic day case lists.
Patient Letters
Three studies were identified that related to patient letters5,6,8.
The first was a small study, which considered whether a tool (the Sheffield Assessment Instrument for Letters – SAIL) improved the quality of outpatient letters as communication between hospital and primary care doctors8. On two occasions, 15 unselected outpatient letters written by each of seven hospital practitioners were rated by another doctor and a GP using SAIL. Individualised feedback was provided following the rating of the first set of letters. The audit cycle was completed 3 months later without forewarning to see if there was any improvement in the correspondence. The mean quality scores, derived for each letter from the summation of a 20-point checklist and a global score, improved from 23.3 out of 30 (95% CI 22.1 to 24.4) to 26.6 (95% CI 25.8 to 27.4) p=0.001. The study did not evaluate whether measured improvement in letter quality persists in the long term, or how the improvement influenced patient outcomes.
Normally after an outpatient appointment, a letter is sent to the patient’s GP. The second study considered whether sending this letter to the patients as well improved their satisfaction with the consultation6. Two hundred ear, nose and throat (ENT) patients were randomly assigned to receive or not receive the letter, and their satisfaction was assessed using a postal questionnaire. Those who did not receive a copy letter had a median overall satisfaction score of 7.75 out of 10, whilst those who did had a median score of 9.0 (p=0.014). The study was not able to establish which aspect of receiving a letter contributed to the overall improvement in satisfaction. The authors suggested that it might be the result of the patients feeling more included in their care, and better communication and understanding.
Finally a single centre, non-randomised study with retrospective controls aimed to evaluate the effect of information given to patients before appointments on non-attendance rates5. The study was conducted at a diabetes clinic in a district general hospital (Merseyside) run by a consultant, one or two diabetes nurse specialists, a dietitian, and a podiatrist.
The intervention being tested was an information pack being sent to patients two-weeks prior to their appointment. This detailed when and where to go, where to park, what to bring, who they will see and what to expect. One week before the appointment they received a supplementary phonecall. DNA rates in 325 new patients who attended after the intervention was compared with 1336 historical controls from the same clinic in the three years before the scheme.