Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Measuring access to primary care appointments: a review of methods

Measuring access to primary care appointments: a review of methods Background: Patient access to primary care appointments is not routinely measured despite the increasing interest in this aspect of practice activity. The generation of standardised data (or benchmarks) for access could inform developments within primary care organisations and act as a quality marker for clinical governance. Logically the setting of targets should be based on a sound system of measurement. The practicalities of developing appropriate measures need debate. Therefore we aimed to search for and compare methods that have been published or are being developed to measure patient access to primary care appointments, with particular focus on finding methods using appointment system data. Method: A search and review was made of the primary care literature from 1990 to 2001, which included an assessment of online resources (websites) and communication with recognised experts. The identified methods were assessed. Results: The published literature in this specific area was not extensive but revealed emerging interest in the late 1990s. Two broad approaches to the measurement of waiting times to GP appointments were identified. Firstly, appointment systems in primary care organisations were analysed in differing ways to provide numerical data and, secondly, patient perceptions (reports) of access were evaluated using survey techniques. Six different methods were found which were based on appointment systems data. Conclusion: The two approaches of either using patient questionnaires or appointment system data are methods that represent entirely different aims. The latter method when used to represent patient waiting times for 'routine' elective appointments seems to hold promise as a useful tool and this avoids the definitional problems that surround 'urgent' appointments. The purpose for which the data is being collected needs to be borne in mind and will determine the chosen methods of data retrieval and representation. Page 1 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 scales or other instruments) would be compared, with Background Primary care is under scrutiny along with other public specific attention given to the type and levels of access services to improve access to its users. Access in primary they aimed to assess. care is typically conceptualised as the achievable access to appointments with clinical professionals, although it is Method not routinely measured in most practices. It is widely rec- Preliminary searches indicated that this area did not have ognised that it represents an important dimension in an extensive or long-standing research literature; most determining the quality of care [1,2]. Nevertheless, it is articles had been published after 1998. A broad but sys- also known that waiting times can vary widely in differing tematic search process was designed to allow for a poorly localities and countries, from patients being seen the same indexed publication pool. Medline, PubMed, Clin Psyc day to a wait of several weeks. Most primary care organi- and ASSIA were searched for relevant publications sations have no more than a perception of variable between 1990 and 2001. The following MeSH terms were demand and no method of comparing fluctuating levels used: family practice, health service accessibility (organisa- of access to appointments within or between practices. tion and administration, statistics and numerical data, Measuring patient access could generate useful informa- standards, trends, methods, manpower), appointments and tion for patients, clinicians and practice managers. schedules (waiting lists), research design, health service needs Demand management initiatives and ways to optimise and demand, weights and measures, quality of healthcare, access could then be audited bearing in mind the impact management audit, patient satisfaction, health service needs of such initiatives on opportunity costs [3–5]. and demand. In addition the following keywords were used: general practice, access, appointments (same day, There are potential difficulties, recognised by a recent dis- urgent, routine) appointment systems, measurement, measures, cussion document [6], that have been accentuated by pol- tools, scales, demand, availability, audit and waiting times. icy and political influences. The NHS Plan in the UK [7] Terms were used both singly and in combination. Title suggested that patients should have access to primary care searches were used to increase the sensitivity. All citations services within 48 hours, but the concept lacked sufficient and abstracts were appraised for relevance and full articles definition. The proposed new GMS contract includes an selected for examination by two researchers independ- optional 48 hour target for access to GP appointments. ently (GE and WJ). Key authors were contacted directly But what exactly should we measure? Access can be meas- [1,12,16,18,21,25,27] and searches conducted on their ured at many different interfaces, from the wait for tradi- previous work. Departments of General Practice in Uni- tional services such as appointments with a clinician to versities in the United Kingdom were also asked to send the alternative solutions of nurse triage, nurse led clinics, details of any relevant research. Conference literature [21] telephone advice (including NHS Direct) or electronic and non-peer reviewed literature obtained from websites mail responses. Differing interpretation of terms can also was also appraised. Relevant websites were identified cause confusion. How soon for example should a prob- using http://omni.ac.uk and the search terms health service lem that is defined by a patient as urgent be seen in general delivery, access to primary care and general practice. The fol- practice? The perception of urgent differs between patients lowing sites were reviewed: the Royal College of General themselves [8] and between doctors and patients. The Practitioners [6], the National Primary Care Research and concept of routine appointments is easier to define and Development Centre [1], The National Primary Care quantify. A proposed measure is the waiting time for the Development Team [10] and the Centre for Innovation in next available routine appointment but this provokes debate Primary Care [11]. about whether this should be practice-based or specified for individual clinicians. If the latter, factors such as part- Studies or articles were included in the review if they time working, practitioner popularity, the creation of described tools, scales, questionnaires or other methods multiple review appointments are likely to rapidly dimin- of measuring actual patient access to appointments. We ish a clinician's accessibility. How transparent would a also included descriptions of methods that were currently practice want to be about such data and how useful or being developed in this field, provided they had under- acceptable would it be to publish information at the clini- taken pilot studies and had completed one data collection cian level? [9] exercise. Articles were excluded if they were purely editorial. Recognising the complex nature of this issue, we set out to review the literature. Our main aim was to search the Results international primary care literature for methods that had A total of 1763 citations were initially identified and 38 been, or were being developed to measure access to GP articles retrieved for detailed assessment from the Pubmed appointments, focusing on measures using appointment and the Medline searches. Clin Psych and ASSIA searches system data. Once identified, existing methods (tools, provided some overlap but no new relevant material. The Page 2 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 Table 1: Comparison of methods based on appointments systems to measure access to primary care Third appointment NEMAS [15] Ledlow [14] Access Response Campbell [12] Kendrick [13] [21] Index AROS [22] rd Measurement The 3 available routine Date of patient call Appointments Number of days Number of appoint- Number of appoint- appointment at 12 midday, and appointment pro- demanded but not until next available ments provided at the ments available at one day per week, for vided. GP requested available in US style pri- routine appoint- beginning of the day the start of the day every clinician and GP allocated. mary care clinics com- ment (with any clini- and the number still and the number of pared to community cian) at 4 pm, every available. Total patients seen as clinics. Demand versus working day number of patients 'extras' at the end availability gap coded seen during the day, of the day into 4 categories. noting the number of 'extras' Frequency of data Once a week Continuous Daily Once a day Twice a day Twice a day collection Weighted for part Yes No No No No No time staff Named clinician Yes Yes No No No No access measured Data analysis Weekly median score and Computerised Demand versus availa- Computer to work Data related to prac- Daily tally monthly average bility gap out 5 day moving tice list size, with average rates given per 1000 patients Results Weekly snapshot of Complete computer- Feedback reports gen- Trends across Bar charts represent Graphical display of patient access profile ised analysis of prac- erated to clinic staff weekly schedules. number of appoint- extras versus tice appointment ments offered versus number of free system number of patients appointments dur- seen. Start of day ing the day appointment availabil- ity categorised as low, medium and high Extent of and rea- Primary Care Collabora- 145 teaching prac- US Military Clinic Study 10 practices To 19 practices Research 1 practice Research son for use tive in England. To inform tices Audit inform Study Study implementation of improvement advanced access Co-ordination National Primary Care Department of Gen- Healthcare Programs University depart- University Depart- Department of Pri- Development Team eral Practice, Univer- Central Michigan ment of General ment of General mary Health Care, sity of Glasgow. University. Practice Practice, Edinburgh. University of Southampton. most helpful pointers to relevant publications were tem for recording differing levels of unmet daily demand obtained from the website searches and personal commu- in a military medical service [14] nications rather than the traditional search engines. Two broad approaches to the measurement of patient access Three methods measure access as days waited by repre- were identified. Firstly, appointment systems in organisa- senting appointment system data. A computerised pack- tions were analysed in differing ways to provide numerical age, NEMAS [15,16] enables practices to audit four areas: data and, secondly, patient perceptions (reports) of access practice service provision (including 'appointment availa- were evaluated using survey techniques. bility' and patient satisfaction), chronic disease manage- ment, drug monitoring and significant event analysis. The Methods using appointment system data method calculates the mean time waited in days plus the Table 1 summarises the six identified methods that were minimum and maximum patient waits. Data can be pre- based on appointment system data. sented for the whole practice or for individual GPs. Data entry involves using electronic forms to record the date of Three methods determine appointment availability and/ appointment request, the date the patient was seen, or the satisfaction of demand on a daily basis but do not whether it was an elective, forced (i.e. next available clini- measure the days wait for appointments. Campbell meas- cian) or urgent appointment, the clinician requested and ured clinician availability by recording the number of pro- the clinician actually consulted. The costs of collecting vided appointments at the beginning of each day plus the data has varied considerably depending on which staff number of these un-booked at this time [12]. At the end member is employed for the task [17]. Transfer of data of the day numbers seen and numbers of extras were from the practice system can be automatic with compati- counted, and adjusted for practice list size. Kendrick and ble systems (personal communication). Kerry recorded the number of available appointments at the beginning of the day and the number of extras seen at The National Primary Care Development Team is measur- the end [13]. Ledlow also suggested a categoriations sys- ing access as part of the adoption of the system of 'Advanced Access', a system developed in the USA [18,19], Page 3 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 1 5 9 4 2 3 10 7 6 11 8 Timeline (days 61-66 represent Christmas period 2000) A Figure 1 ROS scores for routine appointment availability (data from 11 practices) AROS scores for routine appointment availability (data from 11 practices) as a response to patient waiting times of 4 – 6 weeks for oped [21] for part time workers to enable their scores to routine appointments in primary care [20]. This method be incorporated. collects data on one day per week (which can vary) at 12 noon. The number of days to the third available routine The Access Response Index (AROS) was developed as a appointment for each clinician is recorded and a median rapidly calculable measure of organisational access [22]. figure calculated to represent an access score for the speci- This index is derived by counting the number of days until fied week. Over a month, the average of four median val- the next available routine appointment, with any clini- ues is taken to represent a monthly access score. The third cian, once during every normal working day. The data is appointment is chosen, rather than the first, in order to recorded at four pm – a time chosen to avoid the influence negate the effect of sudden cancellations which otherwise of embargoed appointments that many organisations use could give false impressions of availability if at the to maintain urgent same day availability. The results are moment of measuring there is one sudden cancellation in plotted on a graph, the daily fluctuation represented by a schedule that is otherwise booked for several days the raw data is smoothed to a demonstrate trends by cal- ahead. The third appointment has been found by trial and culating a 5-day moving average (a data point which is the improvement to best represent the actual waits involved mean of every successive 5-day group). An example of the (personal communication). Embargoed appointments data produced by this measure is shown in Figure 1. are not included. A system of weighting has been devel- Page 4 of 7 (page number not for citation purposes) AROS Index (days to next available appointment) 125 BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 Table 2: Patient survey instruments: items used to determine access perceptions Survey items Response ratings GPAS [23] 6. Thinking of times when you want to see a particular a) 5 point scale, 1 = same day, 5 = more than 5 days doctor: a) How quickly do you get an appointment? b) How do you rate this? b) Range from 1 = very poor, 6 = excellent 7. Thinking of times when you are willing to see any doctor: a) How quickly do you get an appointment? 8. Yes / No / Not applicable / Don't know b). How do you rate this? 8. If you need an urgent appointment to see your GP can you normally get one on the same day? EUROPEP [25,26] What is your opinion of the general practitioner and/or the practice over the past 12 months 5 point scale (poor to excellent) with respect to: 19) Getting an appointment to suit you? 23) Providing quick services for urgent health problems? Baker [28] 10) It can sometimes be difficult to get an appointment 5 point agreement scale with my doctor at this surgery. 14) It can be hard to get an appointment for medical care right away. Grogan [29] 33) Getting an appointment at a convenient time is easy. 5 point agreement scale 34) Appointments are easy to make whenever I need them. 35) It is often difficult to get an appointment with a doctor. 36) It is easy to see a doctor of my choice. Methods using patient questionnaires be given 'appointments'). Appointments for healthcare Four patient experience questionnaires were found that could be categorised into urgent, soon, and elective. Urgent contained access assessments and the relevant items are appointments are typically seen as requests for same-day outlined in Table 2. The General Practice Assessment Sur- consultations. The soon category would fit problems that vey (GPAS) [23] was developed by the National Primary should be seen within two or three days to prevent escala- Care Research and Development Centre, by adapting the tion or symptom prolongation. Finally, routine or elective Family Practice Assessment Survey (FPAS)[24]. A second appointments suit individuals who value an agreed time survey named Europep [25,26] has been validated in 10 window over other factors. The methods that represent European countries. Thirdly Baker describes the use of a patient waiting times for 'routine', i.e. elective appoint- validated surgery satisfaction questionnaire (SSQ) that he ments seem to hold promise as they avoid the definitional developed [27,28] and finally Grogan's patient Satisfac- problems that surround 'urgent' appointments, and the tion Questionnaire contains two sections that ask about different views that patients, clinicians and others have access and appointment availability[29]. about 'urgency'. The task of deciding whether to represent access profiles for organisations or for individual clini- cians also needs careful consideration. It may be more Discussion Principal findings feasible (and less threatening) to routinely measure This review of access measurement reveals the heterogene- organisational access, especially if the data is to be used ous nature of the methods and the lack of any widely for benchmarking purposes. accepted conceptualisation of patient access. Identified measures are either practice centred using appointment To attempt to measure access means to obtain meaningful data or patient orientated via surgery satisfaction ques- data from a dynamic system that is not always in equilib- tionnaires. It is clear that these two methods represent rium. The access experienced by an individual varies entirely different aims. It is not possible for episodic determined by demand, adequate appointment provi- patient surveys to provide data that has enough currency sion, sudden cancellations and block release of held or accuracy to inform organisational responses to patient appointments. We have distinguished two approaches to demand. overcome this, either for a full statistical analysis using compatible systems or data retrieval software, or to use a Also demonstrated are inherent problems over definition simple snapshot method, deciding whether data-smooth- of terms – defining what is to be measured and setting of ing methods such as daily moving averages or aggregated targets. There appear to be three appointment categories weekly median scores are the best portrayal of overall over and above emergencies (which by definition cannot access patterns. Page 5 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 Strengths and weaknesses of the study measurable [35] will, if a consistent standard can be Multiple search methods were used to ensure that the agreed, provide benchmark data, and mark an important breadth of literature and online resources were examined step towards a compendium of methods to assess quality as systematically as possible. The searches proved difficult in primary care. and reflect the emergent status and the diversity of terms used in this area. We may have overlooked methods Conclusion developed in other healthcare systems. The two approaches of either using patient questionnaires or appointment system data to measure access are meth- Implications of the findings to healthcare services and ods that represent entirely different aims. The latter research method when used to represent patient waiting times for The lack of a widely agreed measurement method to rep- 'routine' elective appointments seems to hold promise as resent patient access to primary care services will make it a useful tool and this avoids the definitional problems impossible for practices to compare their response to that surround 'urgent' appointments. The purpose for patient demand with any degree of certainty. The 'third which the data is being collected needs to be borne in appointment' system is the most widely used method and mind and will determine the chosen methods of data is currently supported by the National Primary Care retrieval and representation Development Team in England (but has no equivalent support in Scotland, Wales or Northern Ireland). It is Competing Interests however a relatively complex manipulation of appoint- The authors of this article have piloted the AROS Index. ment system data, and incorporates individual clinician availability. It seems from first principles that the impor- Authors' Contributions tant requirements of a tool designed to measure a Peter Edwards, Melody Emmerson and Glyn Elwyn were dynamic concept such as patient access is simplicity and involved in the development of the Aros index. Wendy ease of regular data collection, so that longitudinal data Jones and Glyn Elwyn conducted the Aros pilot study and patterns capable of indicating trends in organisations can the literature review with contributions from Adrian be generated rather than data on individual clinician Edwards and Richard Hibbs. availability. Acknowledgements The partners, manager and receptionists at Ely Bridge Surgery. The AROS A recent survey compared mechanisms used to manage Pilot Study Group: New Park Surgery, Talbot Green; Porthcawl Group requests for same day appointments[30]. Murray's pro- Practice; Woodlands Medical Centre, Ely, Cardiff; Old School Surgery, Pon- posal of doing today's work today (Advanced Access) tyclun; Greenmount Surgery, Ely, Cardiff; Salisbury Rd Surgery, Barry; Llynfi eliminates appointment categories and the work involved Surgery, Maesteg; Oldcastle Surgery, Bridgend; Park Lane Surgery, Tonyre- in negotiating urgency by dealing with virtually all fail; Meddygfa Teilo, Llandeilo; The Surgery, Overton-on-Dee, Wrexham. demand on the day it arises[31]. Access is an important determinant of healthcare quality but what are the impli- This work was sponsored by CAPRICORN which receives funding from cations of this approach to the balance of overall quality? the Welsh Assembly Government Continuity of care, whilst not important for some prob- References lems is desirable for others [32,33]. Too drastic a shift in 1. Roland M, Holden J and Campbell S: Quality Assessment for Gen- favour of access is likely to be at the cost of reduced conti- eral Practice: supporting clinical governance in primary care nuity and a diminution of other services, such as screening groups Manchester: National Primary Care Research and Development Centre 1998 [http://www.npcrdc.man.ac.uk/]. and chronic disease management. Measures of organisa- 2. Campbell SM and Roland MO: Defining quality of care Soc Sci Med tional quality need to be aware of the dangers of focusing 2000, 51:1611-25. too much on one dimension, and should work towards 3. Rogers A, Entwistle V and Pencheon D: A patient led NHS: man- aging demand at the interface between lay and primary care the creation of measures that balance scores across inter- BMJ 1998, 316:1816-9. nal and external requirements [34]. 4. Pencheon D: Matching demand and supply fairly and efficiently BMJ 1998, 316:1665-7. 5. Gillam S and Pencheon D: Managing demand in general practice An effective consultation with a well trained clinician who BMJ 1998, 316:1895-8. knows the patient and who has access to a well structured 6. Royal College of General Practitioners: Access to general prac- tice based primary care 2001 [http://www.rcgp.org.uk/]. Accessed longitudinal record will probably remain a gold standard March 2001. London and the issue of immediate 'access' should not be elevated 7. The Stationery Office: The NHS Plan London: The Stationery Office above all the other components in this equation. Never- 2000. 8. Campbell JL: Patient's perceptions of medical urgency, does theless, it seems prudent to develop measures that provide deprivation matter? Fam Pract 1999, 16:28-32. a better understanding of patient access to organisations 9. Marshall MN, Shekelle PG, Leatherman S and Brook RH: The public release of performance data: what do we expect to gain? with similar resources. Measuring the interval to the next JAMA 2000, 283:1866-74. available routine appointment, whilst measuring the Page 6 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 10. The National Primary Care Development Team [http:// http://www.biomedcentral.com/1471-2296/4/8/prepub www.npdt.org/] 11. Centre for Innovation in Primary Care [http://www.inno- vate.org.uk] 12. Campbell JL: General practitioner appointment systems, patient satisfaction, and use of accident and emergency serv- ices: a study in one geographical area Fam Pract 1994, 11:438- 13. Kendrick T and Kerry S: How many surgery appointments should be offered to avoid undesirable number of "extras"? BJGP 1999, 49:273-6. 14. Ledlow GR, Bradshaw DM and Shockley C: Primary care access improvement: an empowerment interaction model Military Medicine 2000, 165:390-4. 15. NEMAS [program]: Glasgow: West of Scotland Postgraduate Medical Education Board and MI Technology Ltd 2000. 16. Lough M and Wilmot M: Combining audit and information tech- nology in the west of Scotland Practice Computing 1996, Autumn:26-8. 17. Lough JRM, Wilmot M and Murray TS: Supporting practice-based audit: a price to be paid for collecting data BJGP 1999, 49:793-5. 18. Murray M and Tantau C: Same-day appointments: exploding the access paradigm Family Practice Management 2000, 7:45-50. 19. Murray M and Tantau C: Redefining open access to primary care Managed Care Quarterly 1999, 7:45-55. 20. Murray M and Tantau C: Must patients wait? Journal of Quality Improvement 1998, 24:423-5. 21. Oldham J: Advanced Access in Primary Care Manchester: National Primary Care Development Team 2001. 22. Elwyn G, Jones W, Emmerson M, Edwards P and Edwards A: Devel- oping a measure for benchmarking access in primary care: Access Response (AROS) Journal of Clinical Evaluation in Clinical Practice 9,1:33-37. 23. Ramsay J, Campbell JL, Schroter S, Green J and Roland M: The Gen- eral Practice Assessment Survey (GPAS) tests of data qual- ity and Measurement properties Fam Pract 2000, 17:372-9. 24. Safran DG, Kosinski M, Tarlov AR, Rogers WH, Taira DH, Lieberman N and Ware JE: The Primary Care Asessment Survey:tests of data quality and measurement performance Med Care 1998, 36:728-39. 25. Grol R, Wensing M, Mainz J, Jung HP, Ferreira P, Hearnshaw H, Hjort- dahl P, Olesen F, Reis S and Ribacke M et al.: Patients in Europe evaluate general practice care: an international comparision Br J Gen Pract 2000, 50:882-7. 26. Grol R and Wensing M: Patients evaluate family pracitce The Europep instrument Nijmegen: Centre for Health Care Quality; 2000. 27. Baker R: The reliability and criterion validity of a measure of patients satisfaction with their general practice Family Practice 1991, 8:171-7. 28. Baker R and Streatfield J: What type of general practice do patients prefer? Exploration of practice characteristics influ- encing patient satisfaction Br J Gen Pract 1995, 45:654-659. 29. Grogan S, Conner M, Willits D and Norman P: Development of a questionnaire to measure patients' satisfaction with general practitioners' services BJGP 1995, 45:525-9. 30. Luthra M and Marshall MN: How do general practices manage requests from patients for "same-day" appointments? A questionnaire survey BJGP 2001, 51:39-41. 31. Murray M: Patient care: access BMJ 2000, 320:1594-6. 32. Stott NCH, Kinnersley P and Elwyn G: Measuring general prac- tice-based primary care: generic outcomes Family Practice 1997, Publish with Bio Med Central and every 14:486-491. scientist can read your work free of charge 33. Guthrie B and Wyke S: Does continuity in general practice really matter? BMJ 2000, 321:734-6. "BioMed Central will be the most significant development for 34. Kaplan RS and Norton DP: Using the balanced scorecard as a disseminating the results of biomedical researc h in our lifetime." strategic management system Harvard Business Review 1996:75- Sir Paul Nurse, Cancer Research UK 35. Campbell J and Proctor S: A developmental performance Your research papers will be: framework for primary care International Journal of Healthcare available free of charge to the entire biomedical community Quality Assurance 1999, 12:279-286. peer reviewed and published immediately upon acceptance Pre-publication history cited in PubMed and archived on PubMed Central The pre-publication history for this paper can be accessed yours — you keep the copyright here: BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 7 of 7 (page number not for citation purposes) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Family Practice Springer Journals

Measuring access to primary care appointments: a review of methods

Loading next page...
 
/lp/springer-journals/measuring-access-to-primary-care-appointments-a-review-of-methods-0PN3PM83Rs

References (32)

Publisher
Springer Journals
Copyright
Copyright © 2003 by Jones et al; licensee BioMed Central Ltd.
Subject
Medicine & Public Health; General Practice / Family Medicine; Primary Care Medicine
eISSN
1471-2296
DOI
10.1186/1471-2296-4-8
pmid
12846934
Publisher site
See Article on Publisher Site

Abstract

Background: Patient access to primary care appointments is not routinely measured despite the increasing interest in this aspect of practice activity. The generation of standardised data (or benchmarks) for access could inform developments within primary care organisations and act as a quality marker for clinical governance. Logically the setting of targets should be based on a sound system of measurement. The practicalities of developing appropriate measures need debate. Therefore we aimed to search for and compare methods that have been published or are being developed to measure patient access to primary care appointments, with particular focus on finding methods using appointment system data. Method: A search and review was made of the primary care literature from 1990 to 2001, which included an assessment of online resources (websites) and communication with recognised experts. The identified methods were assessed. Results: The published literature in this specific area was not extensive but revealed emerging interest in the late 1990s. Two broad approaches to the measurement of waiting times to GP appointments were identified. Firstly, appointment systems in primary care organisations were analysed in differing ways to provide numerical data and, secondly, patient perceptions (reports) of access were evaluated using survey techniques. Six different methods were found which were based on appointment systems data. Conclusion: The two approaches of either using patient questionnaires or appointment system data are methods that represent entirely different aims. The latter method when used to represent patient waiting times for 'routine' elective appointments seems to hold promise as a useful tool and this avoids the definitional problems that surround 'urgent' appointments. The purpose for which the data is being collected needs to be borne in mind and will determine the chosen methods of data retrieval and representation. Page 1 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 scales or other instruments) would be compared, with Background Primary care is under scrutiny along with other public specific attention given to the type and levels of access services to improve access to its users. Access in primary they aimed to assess. care is typically conceptualised as the achievable access to appointments with clinical professionals, although it is Method not routinely measured in most practices. It is widely rec- Preliminary searches indicated that this area did not have ognised that it represents an important dimension in an extensive or long-standing research literature; most determining the quality of care [1,2]. Nevertheless, it is articles had been published after 1998. A broad but sys- also known that waiting times can vary widely in differing tematic search process was designed to allow for a poorly localities and countries, from patients being seen the same indexed publication pool. Medline, PubMed, Clin Psyc day to a wait of several weeks. Most primary care organi- and ASSIA were searched for relevant publications sations have no more than a perception of variable between 1990 and 2001. The following MeSH terms were demand and no method of comparing fluctuating levels used: family practice, health service accessibility (organisa- of access to appointments within or between practices. tion and administration, statistics and numerical data, Measuring patient access could generate useful informa- standards, trends, methods, manpower), appointments and tion for patients, clinicians and practice managers. schedules (waiting lists), research design, health service needs Demand management initiatives and ways to optimise and demand, weights and measures, quality of healthcare, access could then be audited bearing in mind the impact management audit, patient satisfaction, health service needs of such initiatives on opportunity costs [3–5]. and demand. In addition the following keywords were used: general practice, access, appointments (same day, There are potential difficulties, recognised by a recent dis- urgent, routine) appointment systems, measurement, measures, cussion document [6], that have been accentuated by pol- tools, scales, demand, availability, audit and waiting times. icy and political influences. The NHS Plan in the UK [7] Terms were used both singly and in combination. Title suggested that patients should have access to primary care searches were used to increase the sensitivity. All citations services within 48 hours, but the concept lacked sufficient and abstracts were appraised for relevance and full articles definition. The proposed new GMS contract includes an selected for examination by two researchers independ- optional 48 hour target for access to GP appointments. ently (GE and WJ). Key authors were contacted directly But what exactly should we measure? Access can be meas- [1,12,16,18,21,25,27] and searches conducted on their ured at many different interfaces, from the wait for tradi- previous work. Departments of General Practice in Uni- tional services such as appointments with a clinician to versities in the United Kingdom were also asked to send the alternative solutions of nurse triage, nurse led clinics, details of any relevant research. Conference literature [21] telephone advice (including NHS Direct) or electronic and non-peer reviewed literature obtained from websites mail responses. Differing interpretation of terms can also was also appraised. Relevant websites were identified cause confusion. How soon for example should a prob- using http://omni.ac.uk and the search terms health service lem that is defined by a patient as urgent be seen in general delivery, access to primary care and general practice. The fol- practice? The perception of urgent differs between patients lowing sites were reviewed: the Royal College of General themselves [8] and between doctors and patients. The Practitioners [6], the National Primary Care Research and concept of routine appointments is easier to define and Development Centre [1], The National Primary Care quantify. A proposed measure is the waiting time for the Development Team [10] and the Centre for Innovation in next available routine appointment but this provokes debate Primary Care [11]. about whether this should be practice-based or specified for individual clinicians. If the latter, factors such as part- Studies or articles were included in the review if they time working, practitioner popularity, the creation of described tools, scales, questionnaires or other methods multiple review appointments are likely to rapidly dimin- of measuring actual patient access to appointments. We ish a clinician's accessibility. How transparent would a also included descriptions of methods that were currently practice want to be about such data and how useful or being developed in this field, provided they had under- acceptable would it be to publish information at the clini- taken pilot studies and had completed one data collection cian level? [9] exercise. Articles were excluded if they were purely editorial. Recognising the complex nature of this issue, we set out to review the literature. Our main aim was to search the Results international primary care literature for methods that had A total of 1763 citations were initially identified and 38 been, or were being developed to measure access to GP articles retrieved for detailed assessment from the Pubmed appointments, focusing on measures using appointment and the Medline searches. Clin Psych and ASSIA searches system data. Once identified, existing methods (tools, provided some overlap but no new relevant material. The Page 2 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 Table 1: Comparison of methods based on appointments systems to measure access to primary care Third appointment NEMAS [15] Ledlow [14] Access Response Campbell [12] Kendrick [13] [21] Index AROS [22] rd Measurement The 3 available routine Date of patient call Appointments Number of days Number of appoint- Number of appoint- appointment at 12 midday, and appointment pro- demanded but not until next available ments provided at the ments available at one day per week, for vided. GP requested available in US style pri- routine appoint- beginning of the day the start of the day every clinician and GP allocated. mary care clinics com- ment (with any clini- and the number still and the number of pared to community cian) at 4 pm, every available. Total patients seen as clinics. Demand versus working day number of patients 'extras' at the end availability gap coded seen during the day, of the day into 4 categories. noting the number of 'extras' Frequency of data Once a week Continuous Daily Once a day Twice a day Twice a day collection Weighted for part Yes No No No No No time staff Named clinician Yes Yes No No No No access measured Data analysis Weekly median score and Computerised Demand versus availa- Computer to work Data related to prac- Daily tally monthly average bility gap out 5 day moving tice list size, with average rates given per 1000 patients Results Weekly snapshot of Complete computer- Feedback reports gen- Trends across Bar charts represent Graphical display of patient access profile ised analysis of prac- erated to clinic staff weekly schedules. number of appoint- extras versus tice appointment ments offered versus number of free system number of patients appointments dur- seen. Start of day ing the day appointment availabil- ity categorised as low, medium and high Extent of and rea- Primary Care Collabora- 145 teaching prac- US Military Clinic Study 10 practices To 19 practices Research 1 practice Research son for use tive in England. To inform tices Audit inform Study Study implementation of improvement advanced access Co-ordination National Primary Care Department of Gen- Healthcare Programs University depart- University Depart- Department of Pri- Development Team eral Practice, Univer- Central Michigan ment of General ment of General mary Health Care, sity of Glasgow. University. Practice Practice, Edinburgh. University of Southampton. most helpful pointers to relevant publications were tem for recording differing levels of unmet daily demand obtained from the website searches and personal commu- in a military medical service [14] nications rather than the traditional search engines. Two broad approaches to the measurement of patient access Three methods measure access as days waited by repre- were identified. Firstly, appointment systems in organisa- senting appointment system data. A computerised pack- tions were analysed in differing ways to provide numerical age, NEMAS [15,16] enables practices to audit four areas: data and, secondly, patient perceptions (reports) of access practice service provision (including 'appointment availa- were evaluated using survey techniques. bility' and patient satisfaction), chronic disease manage- ment, drug monitoring and significant event analysis. The Methods using appointment system data method calculates the mean time waited in days plus the Table 1 summarises the six identified methods that were minimum and maximum patient waits. Data can be pre- based on appointment system data. sented for the whole practice or for individual GPs. Data entry involves using electronic forms to record the date of Three methods determine appointment availability and/ appointment request, the date the patient was seen, or the satisfaction of demand on a daily basis but do not whether it was an elective, forced (i.e. next available clini- measure the days wait for appointments. Campbell meas- cian) or urgent appointment, the clinician requested and ured clinician availability by recording the number of pro- the clinician actually consulted. The costs of collecting vided appointments at the beginning of each day plus the data has varied considerably depending on which staff number of these un-booked at this time [12]. At the end member is employed for the task [17]. Transfer of data of the day numbers seen and numbers of extras were from the practice system can be automatic with compati- counted, and adjusted for practice list size. Kendrick and ble systems (personal communication). Kerry recorded the number of available appointments at the beginning of the day and the number of extras seen at The National Primary Care Development Team is measur- the end [13]. Ledlow also suggested a categoriations sys- ing access as part of the adoption of the system of 'Advanced Access', a system developed in the USA [18,19], Page 3 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 1 5 9 4 2 3 10 7 6 11 8 Timeline (days 61-66 represent Christmas period 2000) A Figure 1 ROS scores for routine appointment availability (data from 11 practices) AROS scores for routine appointment availability (data from 11 practices) as a response to patient waiting times of 4 – 6 weeks for oped [21] for part time workers to enable their scores to routine appointments in primary care [20]. This method be incorporated. collects data on one day per week (which can vary) at 12 noon. The number of days to the third available routine The Access Response Index (AROS) was developed as a appointment for each clinician is recorded and a median rapidly calculable measure of organisational access [22]. figure calculated to represent an access score for the speci- This index is derived by counting the number of days until fied week. Over a month, the average of four median val- the next available routine appointment, with any clini- ues is taken to represent a monthly access score. The third cian, once during every normal working day. The data is appointment is chosen, rather than the first, in order to recorded at four pm – a time chosen to avoid the influence negate the effect of sudden cancellations which otherwise of embargoed appointments that many organisations use could give false impressions of availability if at the to maintain urgent same day availability. The results are moment of measuring there is one sudden cancellation in plotted on a graph, the daily fluctuation represented by a schedule that is otherwise booked for several days the raw data is smoothed to a demonstrate trends by cal- ahead. The third appointment has been found by trial and culating a 5-day moving average (a data point which is the improvement to best represent the actual waits involved mean of every successive 5-day group). An example of the (personal communication). Embargoed appointments data produced by this measure is shown in Figure 1. are not included. A system of weighting has been devel- Page 4 of 7 (page number not for citation purposes) AROS Index (days to next available appointment) 125 BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 Table 2: Patient survey instruments: items used to determine access perceptions Survey items Response ratings GPAS [23] 6. Thinking of times when you want to see a particular a) 5 point scale, 1 = same day, 5 = more than 5 days doctor: a) How quickly do you get an appointment? b) How do you rate this? b) Range from 1 = very poor, 6 = excellent 7. Thinking of times when you are willing to see any doctor: a) How quickly do you get an appointment? 8. Yes / No / Not applicable / Don't know b). How do you rate this? 8. If you need an urgent appointment to see your GP can you normally get one on the same day? EUROPEP [25,26] What is your opinion of the general practitioner and/or the practice over the past 12 months 5 point scale (poor to excellent) with respect to: 19) Getting an appointment to suit you? 23) Providing quick services for urgent health problems? Baker [28] 10) It can sometimes be difficult to get an appointment 5 point agreement scale with my doctor at this surgery. 14) It can be hard to get an appointment for medical care right away. Grogan [29] 33) Getting an appointment at a convenient time is easy. 5 point agreement scale 34) Appointments are easy to make whenever I need them. 35) It is often difficult to get an appointment with a doctor. 36) It is easy to see a doctor of my choice. Methods using patient questionnaires be given 'appointments'). Appointments for healthcare Four patient experience questionnaires were found that could be categorised into urgent, soon, and elective. Urgent contained access assessments and the relevant items are appointments are typically seen as requests for same-day outlined in Table 2. The General Practice Assessment Sur- consultations. The soon category would fit problems that vey (GPAS) [23] was developed by the National Primary should be seen within two or three days to prevent escala- Care Research and Development Centre, by adapting the tion or symptom prolongation. Finally, routine or elective Family Practice Assessment Survey (FPAS)[24]. A second appointments suit individuals who value an agreed time survey named Europep [25,26] has been validated in 10 window over other factors. The methods that represent European countries. Thirdly Baker describes the use of a patient waiting times for 'routine', i.e. elective appoint- validated surgery satisfaction questionnaire (SSQ) that he ments seem to hold promise as they avoid the definitional developed [27,28] and finally Grogan's patient Satisfac- problems that surround 'urgent' appointments, and the tion Questionnaire contains two sections that ask about different views that patients, clinicians and others have access and appointment availability[29]. about 'urgency'. The task of deciding whether to represent access profiles for organisations or for individual clini- cians also needs careful consideration. It may be more Discussion Principal findings feasible (and less threatening) to routinely measure This review of access measurement reveals the heterogene- organisational access, especially if the data is to be used ous nature of the methods and the lack of any widely for benchmarking purposes. accepted conceptualisation of patient access. Identified measures are either practice centred using appointment To attempt to measure access means to obtain meaningful data or patient orientated via surgery satisfaction ques- data from a dynamic system that is not always in equilib- tionnaires. It is clear that these two methods represent rium. The access experienced by an individual varies entirely different aims. It is not possible for episodic determined by demand, adequate appointment provi- patient surveys to provide data that has enough currency sion, sudden cancellations and block release of held or accuracy to inform organisational responses to patient appointments. We have distinguished two approaches to demand. overcome this, either for a full statistical analysis using compatible systems or data retrieval software, or to use a Also demonstrated are inherent problems over definition simple snapshot method, deciding whether data-smooth- of terms – defining what is to be measured and setting of ing methods such as daily moving averages or aggregated targets. There appear to be three appointment categories weekly median scores are the best portrayal of overall over and above emergencies (which by definition cannot access patterns. Page 5 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 Strengths and weaknesses of the study measurable [35] will, if a consistent standard can be Multiple search methods were used to ensure that the agreed, provide benchmark data, and mark an important breadth of literature and online resources were examined step towards a compendium of methods to assess quality as systematically as possible. The searches proved difficult in primary care. and reflect the emergent status and the diversity of terms used in this area. We may have overlooked methods Conclusion developed in other healthcare systems. The two approaches of either using patient questionnaires or appointment system data to measure access are meth- Implications of the findings to healthcare services and ods that represent entirely different aims. The latter research method when used to represent patient waiting times for The lack of a widely agreed measurement method to rep- 'routine' elective appointments seems to hold promise as resent patient access to primary care services will make it a useful tool and this avoids the definitional problems impossible for practices to compare their response to that surround 'urgent' appointments. The purpose for patient demand with any degree of certainty. The 'third which the data is being collected needs to be borne in appointment' system is the most widely used method and mind and will determine the chosen methods of data is currently supported by the National Primary Care retrieval and representation Development Team in England (but has no equivalent support in Scotland, Wales or Northern Ireland). It is Competing Interests however a relatively complex manipulation of appoint- The authors of this article have piloted the AROS Index. ment system data, and incorporates individual clinician availability. It seems from first principles that the impor- Authors' Contributions tant requirements of a tool designed to measure a Peter Edwards, Melody Emmerson and Glyn Elwyn were dynamic concept such as patient access is simplicity and involved in the development of the Aros index. Wendy ease of regular data collection, so that longitudinal data Jones and Glyn Elwyn conducted the Aros pilot study and patterns capable of indicating trends in organisations can the literature review with contributions from Adrian be generated rather than data on individual clinician Edwards and Richard Hibbs. availability. Acknowledgements The partners, manager and receptionists at Ely Bridge Surgery. The AROS A recent survey compared mechanisms used to manage Pilot Study Group: New Park Surgery, Talbot Green; Porthcawl Group requests for same day appointments[30]. Murray's pro- Practice; Woodlands Medical Centre, Ely, Cardiff; Old School Surgery, Pon- posal of doing today's work today (Advanced Access) tyclun; Greenmount Surgery, Ely, Cardiff; Salisbury Rd Surgery, Barry; Llynfi eliminates appointment categories and the work involved Surgery, Maesteg; Oldcastle Surgery, Bridgend; Park Lane Surgery, Tonyre- in negotiating urgency by dealing with virtually all fail; Meddygfa Teilo, Llandeilo; The Surgery, Overton-on-Dee, Wrexham. demand on the day it arises[31]. Access is an important determinant of healthcare quality but what are the impli- This work was sponsored by CAPRICORN which receives funding from cations of this approach to the balance of overall quality? the Welsh Assembly Government Continuity of care, whilst not important for some prob- References lems is desirable for others [32,33]. Too drastic a shift in 1. Roland M, Holden J and Campbell S: Quality Assessment for Gen- favour of access is likely to be at the cost of reduced conti- eral Practice: supporting clinical governance in primary care nuity and a diminution of other services, such as screening groups Manchester: National Primary Care Research and Development Centre 1998 [http://www.npcrdc.man.ac.uk/]. and chronic disease management. Measures of organisa- 2. Campbell SM and Roland MO: Defining quality of care Soc Sci Med tional quality need to be aware of the dangers of focusing 2000, 51:1611-25. too much on one dimension, and should work towards 3. Rogers A, Entwistle V and Pencheon D: A patient led NHS: man- aging demand at the interface between lay and primary care the creation of measures that balance scores across inter- BMJ 1998, 316:1816-9. nal and external requirements [34]. 4. Pencheon D: Matching demand and supply fairly and efficiently BMJ 1998, 316:1665-7. 5. Gillam S and Pencheon D: Managing demand in general practice An effective consultation with a well trained clinician who BMJ 1998, 316:1895-8. knows the patient and who has access to a well structured 6. Royal College of General Practitioners: Access to general prac- tice based primary care 2001 [http://www.rcgp.org.uk/]. Accessed longitudinal record will probably remain a gold standard March 2001. London and the issue of immediate 'access' should not be elevated 7. The Stationery Office: The NHS Plan London: The Stationery Office above all the other components in this equation. Never- 2000. 8. Campbell JL: Patient's perceptions of medical urgency, does theless, it seems prudent to develop measures that provide deprivation matter? Fam Pract 1999, 16:28-32. a better understanding of patient access to organisations 9. Marshall MN, Shekelle PG, Leatherman S and Brook RH: The public release of performance data: what do we expect to gain? with similar resources. Measuring the interval to the next JAMA 2000, 283:1866-74. available routine appointment, whilst measuring the Page 6 of 7 (page number not for citation purposes) BMC Family Practice 2003, 4 http://www.biomedcentral.com/1471-2296/4/8 10. The National Primary Care Development Team [http:// http://www.biomedcentral.com/1471-2296/4/8/prepub www.npdt.org/] 11. Centre for Innovation in Primary Care [http://www.inno- vate.org.uk] 12. Campbell JL: General practitioner appointment systems, patient satisfaction, and use of accident and emergency serv- ices: a study in one geographical area Fam Pract 1994, 11:438- 13. Kendrick T and Kerry S: How many surgery appointments should be offered to avoid undesirable number of "extras"? BJGP 1999, 49:273-6. 14. Ledlow GR, Bradshaw DM and Shockley C: Primary care access improvement: an empowerment interaction model Military Medicine 2000, 165:390-4. 15. NEMAS [program]: Glasgow: West of Scotland Postgraduate Medical Education Board and MI Technology Ltd 2000. 16. Lough M and Wilmot M: Combining audit and information tech- nology in the west of Scotland Practice Computing 1996, Autumn:26-8. 17. Lough JRM, Wilmot M and Murray TS: Supporting practice-based audit: a price to be paid for collecting data BJGP 1999, 49:793-5. 18. Murray M and Tantau C: Same-day appointments: exploding the access paradigm Family Practice Management 2000, 7:45-50. 19. Murray M and Tantau C: Redefining open access to primary care Managed Care Quarterly 1999, 7:45-55. 20. Murray M and Tantau C: Must patients wait? Journal of Quality Improvement 1998, 24:423-5. 21. Oldham J: Advanced Access in Primary Care Manchester: National Primary Care Development Team 2001. 22. Elwyn G, Jones W, Emmerson M, Edwards P and Edwards A: Devel- oping a measure for benchmarking access in primary care: Access Response (AROS) Journal of Clinical Evaluation in Clinical Practice 9,1:33-37. 23. Ramsay J, Campbell JL, Schroter S, Green J and Roland M: The Gen- eral Practice Assessment Survey (GPAS) tests of data qual- ity and Measurement properties Fam Pract 2000, 17:372-9. 24. Safran DG, Kosinski M, Tarlov AR, Rogers WH, Taira DH, Lieberman N and Ware JE: The Primary Care Asessment Survey:tests of data quality and measurement performance Med Care 1998, 36:728-39. 25. Grol R, Wensing M, Mainz J, Jung HP, Ferreira P, Hearnshaw H, Hjort- dahl P, Olesen F, Reis S and Ribacke M et al.: Patients in Europe evaluate general practice care: an international comparision Br J Gen Pract 2000, 50:882-7. 26. Grol R and Wensing M: Patients evaluate family pracitce The Europep instrument Nijmegen: Centre for Health Care Quality; 2000. 27. Baker R: The reliability and criterion validity of a measure of patients satisfaction with their general practice Family Practice 1991, 8:171-7. 28. Baker R and Streatfield J: What type of general practice do patients prefer? Exploration of practice characteristics influ- encing patient satisfaction Br J Gen Pract 1995, 45:654-659. 29. Grogan S, Conner M, Willits D and Norman P: Development of a questionnaire to measure patients' satisfaction with general practitioners' services BJGP 1995, 45:525-9. 30. Luthra M and Marshall MN: How do general practices manage requests from patients for "same-day" appointments? A questionnaire survey BJGP 2001, 51:39-41. 31. Murray M: Patient care: access BMJ 2000, 320:1594-6. 32. Stott NCH, Kinnersley P and Elwyn G: Measuring general prac- tice-based primary care: generic outcomes Family Practice 1997, Publish with Bio Med Central and every 14:486-491. scientist can read your work free of charge 33. Guthrie B and Wyke S: Does continuity in general practice really matter? BMJ 2000, 321:734-6. "BioMed Central will be the most significant development for 34. Kaplan RS and Norton DP: Using the balanced scorecard as a disseminating the results of biomedical researc h in our lifetime." strategic management system Harvard Business Review 1996:75- Sir Paul Nurse, Cancer Research UK 35. Campbell J and Proctor S: A developmental performance Your research papers will be: framework for primary care International Journal of Healthcare available free of charge to the entire biomedical community Quality Assurance 1999, 12:279-286. peer reviewed and published immediately upon acceptance Pre-publication history cited in PubMed and archived on PubMed Central The pre-publication history for this paper can be accessed yours — you keep the copyright here: BioMedcentral Submit your manuscript here: http://www.biomedcentral.com/info/publishing_adv.asp Page 7 of 7 (page number not for citation purposes)

Journal

BMC Family PracticeSpringer Journals

Published: Jul 7, 2003

There are no references for this article.