Evaluation of US-trained Physician Assistants working in the NHS in England

Transcription

1 Evaluation of US-trained Physician Assistants working in the NHS in England Interim report The introduction of US-trained Physician Assistants to primary care in Tipton: first impressions Juliet Woodin Hugh McLeod Richard McManus Health Services Management Centre Department of Primary Care and General Practice University of Birmingham 2004

2 Published by: Health Services Management Centre, University of Birmingham, Park House, 40 Edgbaston Park Road, Birmingham B15 2RT. University of Birmingham 2004 First Published 2004 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any other means, electronic or mechanical, photocopying, recording and/or otherwise without the prior written permission of the publishers. This book may not be lent, resold, hired out or otherwise disposed of by way of trade in any form, binding or cover than that in which it is published. ISBN ii

3 Acknowledgements The authors would like to thank all those who gave their time to be interviewed, and in particular, Rachel Catanzaro and Lynn Tyrer, Physician Assistants, and their colleagues for supplying activity data. In addition, the authors wish to acknowledge the work of Nicola Walsh, Fellow at the Health Services Management Centre until December 2003, who undertook the first stages of the evaluation in Material from her interviews and initial research has been incorporated into this report. iii

4 iv

5 Contents Executive Summary 1 Introduction 4 Background 5 What is a physician assistant? 5 Primary care in Tipton 6 Great Bridge Partnerships for Health 6 Swanpool Medical Practice 7 Evaluation methodology 7 Introduction of PAs to Tipton 9 PA recruitment 9 PA practice in the UK 10 PA induction 11 Activity and workload: who is doing what? 12 Clinical resources and the management of patient consultation 13 Patient characteristics 15 Measures of activity content 17 Consultation patterns recurring patients and problems 25 Contact between the PA and supervising GP about patient treatment 26 Activity and workload: themes and issues 28 Selection of patients 28 Signing of prescriptions 30 Home visits 30 Appointment times 30 Role comparisons 31 GPs and PAs 31 Nurses and PAs 32 Re-profiling of work 34 Knowledge and training 35 Regulation and independent practice 36 Individual competencies and interests 37 Assessing impact: benefits and drawbacks 38 Additional staffing 38 Access and waiting times 38 Managing increasing workload 39 Job satisfaction in wider team 39 Development of special interests and service developments 40 Patient satisfaction 40 Factors assisting development of the PA role and its integration into 41 primary care in the UK Strategic 41 Inter-personal skills 42 Factors impeding development of the PA role and its integration into 43 primary care in the UK Prescribing powers 44 Requesting of radiological investigations 44 Absence of regulatory body and registration processes 45 Systems and processes for assuring clinical quality 46 v

6 Language and culture 47 Colleagues awareness of the PA role 48 Work permits 48 Implications for UK commissioning of PA education 49 Competencies 49 Entry points, recruitment and selection 49 Conclusions 50 References 53 Appendices 54 Tables Table 1 Summary of GP requirements in the Sandwell/ Heart of 6 Birmingham Health Economy Table 2 Consultation activity: Swanpool and Great Bridge practices 13 Table 3 Consultation duration measures: Swanpool and Great Bridge 14 practices Table 4 Patient age and sex: Swanpool and Great Bridge practices 16 Table 5 Consultations and ICPC codes: Swanpool and Great Bridge 18 practices Table 6 ICPC codes by practice and consultations reported in table 2 19 Table 7 Psychological ICPC codes in the first 75 consultations for 23 selected clinicians Table 8 Musculoskeletal ICPC codes in the first 75 consultations for 24 selected clinicians Table 9 Respiratory ICPC codes in the first 75 consultations for selected 24 clinicians Table 10 PA consultation characteristics: previous patient contact at both 26 practices Table 11 Contact between the PA and supervising GP about patient 26 treatment at both practices Table 12 Consultations resulting in GP contact initiated by one PA other than for prescription or brief review of treatment plan 27 Figures Figure 1 Swanpool practice: consultation duration for the PA, 15 supervising GP and Practice Nurse Figure 2 Great Bridge practice: consultation duration for the PA, 15 supervising GP, Nurse Practitioner and Practice Nurse Figure 3 Distribution of patient age at consultation: Swanpool practice 17 Figure 4 Distribution of patient age at consultation: Great Bridge practice 17 Figure 5 Activity by ICPC chapter for the PA and supervising GP at the 20 Swanpool practice Figure 6 Activity by ICPC chapter and clinician at the Great Bridge practice 21 vi

7 Executive Summary A Physician Assistant (PA) in the United States of America (USA) is a healthcare professional licensed to practice medicine with physician supervision. PAs are dependent practitioners, delivering care in partnership with a supervising physician. In response to local recruitment difficulties for General Practitioners (GPs), two general practices in Tipton, Sandwell, (the Swanpool practice and the Great Bridge practice) have each had a USA-trained PA working with them since April and September 2003, respectively. This report presents the initial findings of an evaluation commissioned by the Department of Health (NHS Modernisation Agency) Changing Workforce Programme (CWP) from the Health Services Management Centre. The evaluation provides evidence about the impact of the PA role, drawing on qualitative interviews and activity data about consultations collected from the two practices. This interim report considers just two PAs, and their activities and experiences over a limited period of time. Conclusions drawn from the data are therefore necessarily tentative and may not be generalisable to other settings and staff. Nevertheless, a number of themes have emerged. The study suggests that the PA role has made the transition from the USA healthcare system and is contributing successfully to primary care in Tipton. The PAs have, in effect, to a great extent replaced GP roles in an under-doctored area. The PAs were reported to have made a positive impact on the workload carried by other members of the practice teams and contributed to improvements in access for patients to primary care services. While there have been some issues connected with the transition from the USA, including language and culture, public health approaches, and differences in prescribing and diagnostic practice, these have been limited and have been addressed and overcome fairly quickly. The PAs are undertaking a wide range of clinical work, covering a similar spread of presenting problems to their supervising GPs. The volume of activity carried out by the two PAs is different, reflecting both the number of consultations undertaken in each 1

8 session (due to a difference in appointment and actual consultation times), and the number of sessions undertaken per week, with one PA being part-time. The Nurse Practitioner at the Great Bridge practice also undertook a wide range of activity, in particular sharing high volumes of respiratory, skin and general & unspecific work, but there were some differences between her role and that of the PA. Psychological problems did not feature in her workload, and this is notable given the high proportion of overall clinical activity comprised by psychological conditions. Both PAs recorded undertaking psychological activity but interview evidence suggests that this is an area in which certain complex ongoing patients tend to be managed by the GP. This may, however, be more a function of the newness of the PAs than of their scope of practice. PAs are unable to independently issue prescriptions in the UK. Apart from the signing of prescriptions, the PAs reported initiating contact with the supervising GPs about patient treatment for only a small proportion of patient consultations. Brief and infrequent reviews of the care plan took place, and contact for other reasons such as confirming diagnosis or checking on local procedures, was limited. The PA role has been well received by colleagues. Factors facilitating this integration included the team working ethos which is a fundamental characteristic of the PA role and training, and the induction period with the supervising physician, which continued until the point at which both GP and PA had established clinical rapport, that is understanding of, and confidence in, each other s practice. A further facilitating factor was the flexibility of the PA role. Evidence of this was provided not only by the range of conditions seen, but also by the reports of PAs taking on some work normally undertaken by nursing staff to help with cover during vacancy periods. There is also evidence of PAs undertaking home visits, running clinics, and providing a service to a residential home, and discussion of potential for involvement in out-of-hours work, and minor surgery. Staff reported that patients appear to have responded well to the role, with no adverse reactions and evidence of satisfaction from repeat visits and anecdotal comments. The only complaints have related to having to wait for a prescription, and few such complaints have been reported. 2

9 Some barriers exist which limit the current practice of PAs in the UK. Some of these need to be addressed nationally, such as the limitation on prescribing which derives from, and links to, the non-registered status of the PA profession in the UK. Some local measures to mitigate the effects of this are being developed, such as the OTC voucher scheme. However, these are not a substitute for resolving the registration and regulation issues at national level. Other barriers, such as limitations on the requesting of radiological investigations are being tackled locally. To date there has been limited consideration of the inclusion of PAs within the systems and processes being developed to support clinical quality in primary care, such as clinical audit, and personal appraisal systems. These areas need to be addressed. The quality and outcomes framework within the new GMS contract provides a stimulus and an opportunity to do this. There is little evidence in this study that the introduction of the PA role has resulted in redefinition of GP or other professional roles, re-profiling of work or professional boundaries, or significant service development or service improvement, other than improved access times. The main impact to date has been to increase the capacity of the primary care workforce to deal with routine work, in an area where GPs are difficult to recruit. 3

10 Introduction This is the interim report of an evaluation commissioned from the Health Services Management Centre by the Department of Health (NHS Modernisation Agency) Changing Workforce Programme (CWP). CWP has a substantial programme of work in the development and design of new roles and ways of working in the NHS in England. While not initiated by the CWP, the introduction of US-trained Physician Assistants (PAs) into primary care in the Tipton area is a development relevant to the CWP work programme. The initial plan for the evaluation was to focus on three PAs who were due to commence working in primary care in Tipton (in the West Midlands) in early One PA commenced work in April 2003 but another did not take up the offered post. The third PA returned to the US shortly after arriving, but was replaced by another PA in September This interim report focuses on the two PAs who have been in post since April and September The former works in a traditional practice located in the Swanpool Medical Centre, which is hereafter referred to as the Swanpool practice. The latter works in a nurse-led practice known as Great Bridge Partnerships for Health, which is hereafter referred to as the Great Bridge practice. In January 2004, the evaluation was extended to include a further 12 US-trained PAs who have been recently recruited to work in the wider Sandwell area. This recruitment exercise marks the beginning of a major initiative led by the Rowley Regis and Tipton Primary Care Trust, with considerable input from the West Birmingham and Sandwell NHS Trust. Most of the 12 further PAs have commenced work in Sandwell, in a variety of settings spanning primary, intermediate and secondary care. The phase of the evaluation will cover the introduction of these new PAs to the NHS, in addition to the two who have been in post since This interim report focuses on the primary care setting and the experiences of the Swanpool and Great Bridge practices. The majority of Great Bridge data relates to the PA who has been in post since September However, some limited interview data from practice staff relating to the PA who temporarily worked in the practice are also reported. The overall aim of this evaluation is to provide evidence about the impact of the PA role. The following specific research questions will be addressed: 4

11 1. What are the potential benefits and drawbacks (to patients, professionals and the NHS) of introducing the role of physician assistant into healthcare provision? 2. What impact does the introduction of the physician assistant role have on service quality and service improvement? 3. What factors improve and impede the integration of physician assistants with other professional groups, in particular doctors and nurses? 4. To what extent does the introduction of the role of physician assistant re-define professional boundaries and produce new lines of accountability? 5. Does the introduction of the role of physician assistant result in a re-profiling of work between doctors, nurses and physician assistants? 6. What issues does the experience of introducing the physician assistant role raise in respect of the commissioning of programmes to train PAs in England, and for regulation? Background What is a physician assistant? A physician assistant (PA) in the United States of America (USA) is a healthcare professional licensed to practice medicine with physician supervision (Frary, 1996). The profession developed in the 1960s to expand capacity in primary care where doctors were often over-worked in rural and medically under-served areas (Mittman, Cawley and Fenn, 2002). PAs are not independent practitioners, but are described as delivering care in partnership with physicians, in a role described as negotiated performance autonomy (Mittman, Cawley and Fenn, 2002, p485). Today, the profession is an established part of the medical workforce in the USA, with more than 44,000 physician assistants in practice. The training of a physician assistant consists of, on average, 25 months studying an intensive core curriculum, emphasising a primary care, generalist approach. Most entrants have a prior first degree, and in many cases have clinical experience of some sort as well. On graduation from accredited training, PAs must pass a national certifying examination of the National Commission on Certification of Physician Assistants. Subsequently, they are required to complete 100 hours of continuing medical education every two years and to pass a generalist medical recertification examination every six years, regardless of their clinical specialty. 5

12 Primary Care in Tipton Tipton is one of the six towns within the metropolitan borough of Sandwell, in the West Midlands. It has a population of 41,372 (2001 census). The main statutory bodies with health responsibilities for the area are Rowley Regis and Tipton PCT, and Sandwell and West Birmingham Hospitals NHS Trust, which operates Sandwell General Hospital and City Hospital. The wider locality is associated with under-provision for General Practitioners (GPs) and a need to increase recruitment in order to replace GPs due to retire in the next few years (table 1). Table 1 Summary of GP requirements in the Sandwell/Heart of Birmingham Health Economy PCT GPs aged 65+ Number of GP required due to GPs aged Local Delivery Plan requirements vacancies Tipton and Rowley Regis Oldbury and Smethwick Wednesbury and West Bromwich total Heart of Birmingham Total Source: adapted from Rowley Regis and Tipton PCT (not dated, p4) Tipton Care Organisation (TCO) is a first-wave Personal Medical Services (PMS) pilot project, created in 1998 and originally involving eight practices. The project attracted additional resources which were used to fund salaried GPs, additional services, and a central office to support the practices. The TCO now covers a total practice population of 37,000. One of the TCO practices, the Swanpool practice, and a former TCO practice, the Great Bridge practice, were the first to have US-trained PAs working within them. The two practices provide different models of primary care. Great Bridge Partnerships for Health In April 2001 one of the TCO practices, Great Bridge practice, formed a separate thirdwave PMS practice on the retirement of the practice s GP. The Great Bridge practice has an innovative structure. The partners are the Nurse/Case Manager and the 6

13 Business Manager, and the practice is formed as a limited company. The practice has employed a salaried GP since September 2002 who works four days per week and is female. A male GP employed by the TCO also undertakes two sessions per week at the practice. The practice employs a nurse practitioner (from January 2004), a practice nurse, a practice manager, and administrative staff. The Great Bridge practice employs their PA, who is full-time (nine sessions per week). The practice had an advanced nurse practitioner until January The practice s partners initiated the introduction of US-trained PAs to the NHS in response to their difficulty in GP recruitment. The practice has recently expanded to take on a practice in West Bromwich. Swanpool Medical Centre The Swanpool practice has a traditional practice structure, with two full-time GPs (one female and one male), who are the partners, and two part-time practice nurses (although one has been on study leave for the period of the evaluation). In addition there is a practice manager and administrative staff. The PA at Swanpool practice is employed by the Tipton Care Organisation and was allocated to the practice because of its comparatively large list size. The Swanpool PA is part-time (six sessions per week) in the practice. This initiative is viewed by the TCO as contributing to its general efforts in response to difficulties in recruiting GPs. Evaluation methodology This evaluation draws on the conceptual framework developed by Pawson and Tilley (1997) in which differential outcomes are explained in terms of the interplay of the mechanism with the context in which that mechanism is applied. In this case, the mechanism is the PA, and the outcomes from the introduction of the PA may be expected to be influenced not only by national contextual factors (such as NHS workforce development policies, and structures for professional regulation), but also by the local context in which they are working. As described above, the two practices to which the PAs are attached are different in character. Initially, the researchers made exploratory visits to both practices, and to the PCT, in order to introduce themselves, and establish what data might be available, and who should be interviewed. 7

14 Face-to-face interviews were conducted to elicit information about activities and roles, and to collect views and perceptions of practice staff and stakeholders of the PA role. Eleven interviews had been conducted in July/August 2003 by another researcher from HSMC as part of an initial phase of the evaluation, and these were incorporated in the analysis. A further 18 were conducted in March and April 2004, making a total of 29 interviews. The total number of individuals interviewed was 21, (eight people were interviewed twice, both in summer 2003 and spring 2004). A schedule of interviewees is given in Appendix C. The interviews were semi-structured, tape recorded and transcribed. Second interviews attempted to capture developments since the summer, and any changes in perception since that time. In addition to the interviews, activity data were collected from the two practices. The data include summary information relating to consultations including patient age and sex, consultation durations, presenting problems, and consultation outcomes. These data cover PAs, their supervising GPs, and the nursing staff working in the two practices. Additional activity data relate to the consultations for which the PAs contacted the supervising GP about patient treatment. All data were collected by practice staff. The data collection forms used are shown in Appendix D. The activity data were requested for one week for all clinicians in order to provide a snap shot picture of the activity undertaken within each practice. Both PAs have provided data for longer periods. The results of the data collection exercise will be used to inform more substantial data collection in the autumn of Data on the presenting problem(s) recorded for each consultation were coded using the International Classification of Primary Care (ICPC) (World Organization of National Colleges, Academies and Academic Associations of General Practitioners/Family Physicians, 1998). The data were coded by the GP member of the research team who was blind to the identity of the consulting clinician. Where more than one presenting problem was recorded, each was coded separately. Where non-specific presenting problems were recorded, they were coded as general and unspecified (ICPC chapter A). Patient focus groups will be held in the summer of

15 A literature review was undertaken and relevant documentation was reviewed from the PCT, and Department of Health. Ethical approval was given to the study by the Sandwell Local Research Ethics Committee. Introduction of PAs to Tipton PA Recruitment Local interest in PAs was prompted by one of the nurse partners at Great Bridge practice reading Mittman, Cawley and Fenn (2002) which summarises the role of PAs in the USA. Following exploratory discussions with colleagues in the Department of Health, she and her practice partner, with the clinical director from the local PCT, made a fact-finding visit to the USA. They visited two different PA schemes, one in rural midwest Michigan where they looked at the role of PAs working in primary care and family medicine and another, to a specialist renal centre in Brooklyn New York. They returned enthused by what they had seen and began to hold discussions about the possibility of bringing PAs from the USA to work in Tipton. With the support of the PCT and individuals at the Department of Health, an advertisement was placed in The New York Times and on the Internet for three PA positions. Over 20 applications were received and with the help of Professor Bill Fenn from the University of Western Michigan and one of his colleagues, eight candidates were short-listed for interview. Interviews were held in New York in September The interview panel consisted of Professor Fenn, the Clinical Director from the PCT, a GP from TCO, and two of the partners from Great Bridge practice. Three PAs were appointed to start work in Tipton from January They were given two-year fixed term contracts and a salary of 39,500 per annum plus relocation costs. A copy of their job description is given in Appendix B. In the event, only two of the successful candidates took up their appointments, the third deciding to stay in the USA. 9

16 These two PAs were attached to Great Bridge practice and the Swanpool Practice. In May 2003 the PA attached to the Great Bridge practice returned to the USA, leaving her post vacant. She was replaced in September 2003 by another PA, who had made contact in March 2003 while on holiday in the UK. Although the PA position had already been filled at that time, her details were kept on file by the practice and she was then contacted in June when the vacancy arose. PA practice in the UK Initially it was planned that the PAs would undertake a paramedic training course at the University of Hertfordshire, and register as paramedics in order to be able to practice in this country as registered professionals. However, it soon became evident that this would not be an appropriate arrangement because the skills of a PA exceed those of a paramedic, and this training would include skills which they did not need, such as driving skills. Legal advice was obtained which identified that the PA could practice under a delegation and referral clause of the General Medical Council s Good Medical Practice guidance (General Medical Council, 2001, p15): Delegation involves asking a nurse, doctor, medical student or other health care worker to provide treatment or care on your behalf. When you delegate care or treatment you must be sure that the person to whom you delegate is competent to carry out the procedure or provide the therapy involved. You must always pass on enough information about the patient and the treatment needed. You will still be responsible for the overall management of the patient. The advice stated that the standard to be achieved is one of competence. As long as the professional practice of the PA is under the direct supervision of the GP and the GP is confident that the PA is competent it is legal for the GP to delegate in this way. I see no difficulty in promoting the use of PAs by GPs. Though the issues around delegation and referral is a key one and each GP taking advantage of the resource will have to be confident that individual provided is confident (sic) to discharge the obligation of care which has been delegated. [Solicitor s letter dated 7 April 2003] Both PAs were allocated a supervising GP. These GPs had responsibility for assessing their competence, and for supervising their practice on an ongoing basis. 10

17 One supervising GP is a GP trainer. The other trained and worked as a salaried GP in New Zealand before coming to the UK in 1996 and registering to practice here. The Medical Protection Society has provided insurance cover on the basis of the arrangements for delegation and supervision described above. PA Induction Both PAs followed a locally-designed induction programme for the first three months. This enabled the PAs to familiarise themselves with the structure and organisation of the NHS. They also met staff from the local practices, the PCT and social services. They visited Sandwell General Hospital and met the medical director to discuss procedures for referring patients. Most important, in the view of the PAs and their supervising GPs, was the induction period spent with the GP. This consisted initially of shadowing the GP as she saw patients, and discussing the consultation. The PAs then began to see patients with the GP present, before seeing patients on their own. In the early stages, the number of patients seen was limited, and the PA discussed every patient with the GP to check the management plan. Gradually, both PA and GP felt confident enough to discuss fewer and fewer patients until a point was reached at which very few patients (perhaps one or two a week) required discussion. This was considered the end of the induction period: the relationship between the GP and PA had reached a stable position and trust was fully established. Pretty much after each session at the beginning it was every patient, whether it was complex or not. We would talk about every patient and that s very important to me and I think it was important to her as well to get comfortable with each other. So that s what we did initially. Now it s just the more complicated or I guess this system and things I am not familiar with [Interviewee E, PA] Then she started to sit in with me and she sat in for several sessions and I was able to explain to her as we were going along with the patients what I was doing and why I was doing it, why I was prescribing things, why I was doing tests and then we started to very gradually see patients but with me initially supervising with each patient and talking through surgeries with her at the end. She gradually built up the number of 11

18 patients she was seeing until I was able to feel that I could trust her and trust the investigation she was doing and the prescriptions she was doing and things like that. [Interviewee D, GP] That is something we are building up as we go along because obviously [the PA] comes in and discusses patient with me, if it is something she is happy with. Or if it is something she is not happy with she will call me in, for example there was a child yesterday she was very worried about. She actually contacted the hospital, arranged the admission, I signed for it [Interviewee N, GP] In both practices, the period of induction during which the PAs were closely supervised, allowed mutual trust to develop between the supervising GP and PA. The length of time to build up this relationship of trust, and stable pattern of practice, varies from relationship to relationship. I felt comfortable with her skills right away I felt that she felt comfortable with me probably about 6 8 weeks into it. [Interviewee E, PA] probably 2 3 months I would say [Interviewee D, GP] I think when the person first comes you have got to monitor them very closely because you have got somebody who s totally unknown to the team, you know their qualifications because they come with that as said but I would think for the first 2 to 3 months - I mean [the PA] shadowed me to start with and then she came and discussed every patient. It was very hard work the first 2 or 3 months and I think it was very frustrating for [the PA] but I think you have got to really work closely together to build up that mutual trust, awareness of the PA s knowledge [Interviewee N, GP] Activity and workload: who is doing what? Information about the activities of the PAs and their clinical colleagues working in the Swanpool and Great Bridge practices was derived from the interviews and activity data supplied by the clinicians and practice staff. 12

19 Clinical resources and the management of patient consultations The two practices are very different in terms of the overall mixture of clinical resources available and the way in which the PAs contribute to the delivery of healthcare. As noted above, in addition to a PA, the Swanpool practice has two full-time GPs and a part-time practice nurse. Hence, the majority of care is provided by the GPs. This balance of resources is further emphasised by the PA working four days per week in the practice and having longer, and therefore fewer, consultations than the GPs in each session. In contrast, the PA at the Great Bridge practice is working with two GPs (one four days per week and one providing two sessions per week), a nurse practitioner, and a part-time practice nurse. Here, the majority of care is provided by the PA and nurse practitioner. The comparison between the PA and supervising GP is characterised by the PA having 10 minute consultation slots compared to 15 minutes for the GP, the opposite arrangement to that at the Swanpool practice. These differences between the practices are illustrated in table 2, which summarises a snap shot picture of activity in April Table 2 Consultation activity: Swanpool and Great Bridge practices 1 number of sessions per week number of consultations per session number of consultations per week DNA rate 6 (DNAs as a % of consultations) number of days worked per week PA 2 Swanpool practice supervising GP nonsupervising GP 3 Practice PA Supervising Nurse 4 GP Great Bridge practice nonsupervising GP 5 Nurse Practitioner Practice Nurse n/a n/a 12.6 n/a number of days for which data were supplied Unless otherwise stated, the data relate to days in April 2004 and were collected by clinicians working in the practices. 2 These data include 5 days in April and 7 days in May. 3 These data are from electronic appointment records. 4 These data are from June. 5 These data were reported by a member of practice staff. 6 The Swanpool practice DNA data are from electronic appointment records. 13

20 Table 2 shows that in the Swanpool practice the mean number of consultations per session was 7.5 for the PA and 11.4 for the supervising GP, compared to 10.4 for the PA and 6.8 for the supervising GP in the Great Bridge practice. The DNA rates associated with this brief snap shot picture warrant further study. At the Swanpool practice, the median consultation duration for the PA was 14 minutes compared to 9 for the supervising GP (table 3), and the difference was statistically significant (p<0.001). At the Great Bridge practice, the median consultation duration for the PA was 11 minutes compared to 13 for the supervising GP, and the difference was also statistically significant (p=0.015). The data for the practice nurse at the Swanpool practice appear to relate to the length of consultation slots rather than the duration of consultations. Table 3 Consultation duration measures: Swanpool and Great Bridge practices median consultation duration PA 2 Swanpool practice 1 Great Bridge practice 4 Supervising GP 2 Practice PA Supervising Nurse Nurse 3 GP Practice Practitioner Nurse mean consultation duration consultation duration standard deviation Number of consultations for which data were available Number of days for which data were available Data relating to the non-supervising GP were not available. 2 These data are from electronic appointment records. 3 These data appear to relate to the length of consultation slots rather than the duration of consultations. The duration of the consultation slots for the non-supervising GP was reported to be 10 minutes. The range of consultation times represented in the data shown in table 3 are illustrated for the Swanpool and Great Bridge practices in figures 1 and 2 respectively. 14

21 Figure 1 Swanpool practice: consultation duration for the PA, supervising GP and Practice Nurse percentage of consultations or less 6 to to to to to or longer consultation duration (minutes) PA Supervising GP Practice Nurse Figure 2 Great Bridge practice: consultation duration for the PA, supervising GP, Nurse Practitioner and Practice Nurse percentage of consultations or less 6 to to to to to or longer consultation duration (minutes) PA Supervising GP Practice Nurse Nurse Practitioner Patient characteristics Additional context for the following analysis of activity is provided by comparison of patient age and sex, based on consultation-level data (table 4). At the Swanpool practice, the median patient age for the PA was 33 years compared to 46 years for the supervising GP. The difference in median age was statistically significant (p<0.001). 15

22 At the Great Bridge practice, the median patient age for the PA was 41 years and 45 years for the supervising GP. This difference in median age was not statistically significant (p=0.337). The differences in median patient age between the nurse practitioner and the PA or GP at the Great Bridge practice were not statistically significant. Table 4 Patient age and sex: Swanpool and Great Bridge practices PA 2 Swanpool practice 1 Supervising GP 2 Great Bridge practice Practice PA Supervising Nurse Nurse 3 GP Practice Practitioner Nurse median patient age (years) mean patient age (years) percentage of female patients number of consultations for which patient age was recorded number of consultations for which patient sex was recorded Frequency distributions for patient age are shown below for the Swanpool practice (figure 3) and the Great Bridge practice (figure 4). Table 4 also shows that the proportion of consultations with female patients was higher for both the PA and supervising GP at the Swanpool practice compared to the PA, supervising GP and NP at the Great Bridge practice. One possible explanation for this finding is that the non-supervising GP at the Swanpool practice is male and may see comparatively more male patients (the practice was not able to readily supply data on this issue). 16

23 Figure 3 Distribution of patient age at consultation: Swanpool practice percentage of patients in age band to to to to to to to to 80 >80 age band (years) PA Supervising GP Practice Nurse Figure 4 Distribution of patient age at consultation: Great Bridge practice percentage of patients in age band to to to to to to to to 80 >80 age band (years) PA Supervising GP Practice Nurse Nurse Practitioner Measures of activity content The different patterns of activity (tables 2 and 3, figures 1 and 2) and patient characteristics (table 4, figures 3 and 4) provide context for considering the content of the activity undertaken by the PAs and their colleagues in each practice. One measure of the range of clinical activity undertaken is provided by categorising the content of the presenting problem(s) at consultation using the ICPC classification codes. Some consultations are associated with more than one presenting problem. The number of 17

24 ICPC codes used to classify the total consultation data made available by the practices is shown in table 5. These data represent from 4 to 12 days activity for the clinicians, as shown in the final row of table 2. Table 5 Consultations and ICPC codes: Swanpool and Great Bridge practices PA Swanpool practice Supervising GP Practice Nurse PA Great Bridge practice Supervising GP Nurse Practitioner Practice Nurse number of consultations no of ICPC codes % of consultations with 1 ICPC code % of consultations with 2 ICPC codes % of consultations with 3 ICPC codes The ICPC classification codes are grouped into the 17 chapters shown in table 6. Table 5 shows the percentage and number of ICPC codes recorded in table 5 in each chapter for the PA, supervising GP and Practice Nurse from both practices and the Nurse Practitioner working in the Great Bridge practice. Table 6 provides a broad brush picture of the range of activity undertaken by each clinician. The table does not provide a basis for directly comparing the overall patterns of activity at the two practices because the data for the non-supervising GP at the Swanpool practice are missing, and, as table 2 shows, this GP undertakes more consultations than the other clinicians in the practice. Nevertheless, comparisons are possible. The part-time Practice Nurses, for example, illustrate different patterns of activity over the days they recorded data. Sixty-two percent of the activity of the Practice Nurse at the Great Bridge practice was coded as General and unspecified (chapter A), and was largely due to consultations for which the presenting problem was recorded as BP (which was coded as A39 physical function test for measuring blood pressure, in the absence of any additional information). In contrast, the greatest proportion of activity for the Practice Nurse at the Swanpool practice was coded as Endocrine, metabolic and nutritional (chapter T; 30% of codes). The most common recorded presenting problem recorded here was diabetes (which was coded as T90 Diabetes non-insulin dependent ). 18

25 Table 6 ICPC codes by practice and consultations reported in table 2 PA Swanpool practice Supervising GP Practice Nurse PA Great Bridge practice Supervising GP Nurse Practitioner Practice Nurse % (n) % (n) % (n) % (n) % (n) % (n) % (n) A General and unspecified 6 (9) 8 (7) 14 (7) 11 (19) 10 (10) 21 (20) 62 (48) B Blood, blood-forming organs, lymphatics 4 (7) 4 (4) 0 (0) 2 (4) 2 (2) 1 (1) 1 (1) D F H K L N P R S Digestive 8 (13) 9 (8) 0 (0) 7 (13) 8 (8) 2 (2) 0 (0) Eye 2 (3) 0 (0) 0 (0) 2 (3) 0 (0) 2 (2) 0 (0) Ear 4 (6) 2 (2) 4 (2) 3 (5) 3 (3) 4 (4) 3 (2) Circulatory 5 (8) 15 (14) 10 (5) 6 (11) 2 (2) 6 (6) 9 (7) Musculoskeletal 11 (17) 11 (10) 4 (2) 14 (24) 12 (12) 2 (2) 1 (1) Neurological 1 (2) 3 (3) 0 (0) 5 (8) 3 (3) 1 (1) 0 (0) Psychological 14 (23) 12 (11) 0 (0) 13 (23) 22 (22) 0 (0) 3 (2) Respiratory 16 (25) 22 (20) 20 (10) 8 (15) 15 (15) 20 (19) 1 (1) Skin 7 (11) 5 (5) 0 (0) 12 (21) 7 (7) 13 (13) 5 (4) T Endocrine, metabolic and nutritional 2 (4) 2 (2) 30 (15) 5 (9) 1 (1) 15 (15) 8 (6) U Urological 1 (1) 2 (2) 0 (0) 1 (2) 4 (4) 2 (2) 0 (0) W Pregnancy, child-bearing, family plan. 6 (9) 2 (2) 2 (1) 1 (2) 6 (6) 1 (1) 0 (0) X Y Z Female genital 14 (23) 3 (3) 16 (8) 5 (8) 4 (4) 9 (9) 6 (5) Male genital 0 (0) 0 (0) 0 (0) 2 (3) 1 (1) 0 (0) 0 (0) Social problems 0 (0) 0 (0) 0 (0) 4 (7) 1 (1) 0 (0) 0 (0) Total 100 (161) 100 (93) 100 (50) 100 (177) 100 (101) 100 (97) 100 (77) 19

26 Excluding the Practice Nurses, the analysis shown in table 6 is illustrated graphically in figures 5 and 6. Figure 5 shows that the proportion of activity by ICPC chapter is broadly similar for the PA and supervising GP. More than 10% of both clinicians activity was coded in each of three ICPC chapters; Psychological (P), Respiratory (R) and Musculoskeletal (L). The PA s activity coded as Female genital (X) represented more than 10% of her activity and compared to a smaller proportion of activity for the supervising GP (14% and 3% of total activity respectively). This area of work by the PA was reported to be of particular interest to the PA, and builds on an earlier stage of her career in which the PA gained experience in this field. Figure 5 also shows that 15% of the supervising GP s activity was coded as Circulatory (K), compared to 5% for the PA. These areas of activity are explored further below. Figure 5 Activity by ICPC chapter for the PA and supervising GP at the Swanpool practice 25 percentage of all ICPC codes A B D F H K L N P R S T U W X Y Z ICPC Chapter (see key below) PA Supervising GP A General and unspecified L Musculoskeletal U Urological B Blood, blood-forming organs, lymphatics N Neurological W Pregnancy, child-bearing, family planning D Digestive P Psychological X Female genital F Eye R Respiratory Y Male genital H Ear S Skin Z Social problems K Circulatory T Endocrine, metabolic and nutritional Given that in the Swanpool practice the PA sees fewer patients and has longer consultations than the supervising GP (table 2), figure 5 shows that the PA and GP share a broadly similar range of work across the ICPC chapters which represent the 20

27 bulk of the day-to-day workload over the time period examined. A more detailed comparison of the most common activities undertaken by the clinicians at the level of individual ICPC codes is presented below, but at this point the figure 5 snap shot picture suggests that the PA is seeing a range of issues which is similar in breadth to those seen by her supervising GP. The workload of the PA and supervising GP at the Great Bridge practice is not readily comparable to that at the Swanpool practice, because of the presence of the Nurse Practitioner at the Great Bridge practice. Figure 6 shows that more than 10% of both PA and supervising GP s activity was coded in each of two ICPC chapters; Psychological (P), and Musculoskeletal (L). In these two chapters, the work was divided between the PA and GP exclusively or almost exclusively. Activity coded as Respiratory (R), Skin (S) and General and unspecified (A) was both comparatively high volume and shared between the PA, supervising GP and Nurse Practitioner. Figure 6 Activity by ICPC chapter and clinician at the Great Bridge practice 25 percentage of all ICPC codes A B D F H K L N P R S T U W X Y Z ICPC Chapter (see key below) PA Supervising GP Nurse Practitioner A General and unspecified L Musculoskeletal U Urological B Blood, blood-forming organs, lymphatics N Neurological W Pregnancy, child-bearing, family planning D Digestive P Psychological X Female genital F Eye R Respiratory Y Male genital H Ear S Skin Z Social problems K Circulatory T Endocrine, metabolic and nutritional In contrast to the Swanpool practice, in the Great Bridge practice, the PA sees more patients and has shorter consultations than the supervising GP (table 2). 21

28 Nevertheless, as at the Swanpool practice, figure 6 shows that in the Great Bridge practice, the PA and GP share a broadly similar range of work across the ICPC chapters, representing the bulk of the day-to-day workload over the time period examined. The three chapters (Psychological, Musculoskeletal and Respiratory) represented more than 10% of total activity for each of the PAs and their supervising GPs (with one exception) and are considered further below. In order to aid comparison between the PAs, their supervising GPs and the Nurse Practitioner, the following analysis focuses on data for the first 75 consultations reported. Seventy-five consultations were chosen here because data for at least this number of consultations were available for all five clinicians. The comparison of activity based on the same number of consultations is helpful, but not entirely satisfactory because, as noted above, the number and duration of consultations varied substantially across the clinicians. Psychological activity In both practices, psychological issues were addressed by both PA and supervising GP (table 7). The nurse practitioner at the Great Bridge practice did not undertake any activity in this area. The most common issues in both practices related to depression and anxiety (table 7). Depression was most commonly reported as a sole presenting problem (11 consultations) or with anxiety (10 consultations). Focusing on the consultations coded with both depression and anxiety during the period of the first 75 consultations for each clinician, the data suggest that each consultation was with a different patient. All 10 consultations were with female patients, with a median age of 32 years (mean 34, SD 8.7, range 22 to 49). The PA at the Swanpool practice had four of the consultations coded as both depression and anxiety. In all four cases, the patient had previously been seen by the GP for the same problems, and in three of the consultations, the PA had both seen the patient before for the same problems and had contact with a GP about the patient for the same problems. In two consultations the PA initiated a prescription. In one consultation the PA advised continuation with the current medication, and in the other consultation the PA advised an increase in the current medication dose. The supervising GP at the Swanpool practice had one consultation coded as both depression and anxiety, and the GP had seen the patient before for the same 22

29 problems. The GP advised an increase in the current medication dose at this consultation. The PA at the Great Bridge practice had one consultation coded as both depression and anxiety, and in this case the patient had not seen the PA or GP before for the same problems. The supervising GP at the Great Bridge practice had four consultations coded as both depression and anxiety, and in two cases the GP had seen the patient before for the same problems. The GP initiated a prescription in all four consultations. Table 7 Psychological ICPC codes in the first 75 consultations for selected clinicians ICPC code Swanpool practice Great Bridge practice total PA Supervising GP PA Supervising GP Nurse Practitioner Depressive disorder P Anxiety disorder/ anxiety state P Sleep disturbance P Feeling anxious/ nervous/tense P Acute stress reaction P Chronic alcohol abuse P Others Total Musculoskeletal activity A wide range of low volume musculoskeletal issues were addressed by both PA and supervising GP in both practices (Table 8). The most common issue overall related to back pain. The Great Bridge practice provides an opportunity to compare the pattern of PA and GP activity with that of the nurse practitioner. Table 8 suggests that the nurse practitioner s work in the area of musculoskeletal issues was comparatively limited. 23

30 Table 8 Musculoskeletal ICPC codes in the first 75 consultations for selected clinicians ICPC codes Swanpool practice Great Bridge practice total back symptom, complaint or syndrome L03, L02, L84, L86 Joint symptom/ complaint L15, L08, L17, L20 Chest symptom/ complaint L04 PA Supervising GP PA Supervising GP Nurse Practitioner Other Total Respiratory activity In both practices, acute upper respiratory tract symptoms or infections were the most common presenting problems within respiratory activity during the brief period for which data were collected (table 9). The snap shot picture represented in table 9 shows all the clinicians working in this area. Asthma related activity (R96) was also recorded for the Practice Nurse at the Swanpool practice, but not for the Practice Nurse at the Great Bridge practice. Table 9 Respiratory ICPC codes in the first 75 consultations for selected clinicians ICPC codes Swanpool practice Great Bridge practice total Acute upper respiratory tract symptom/infection R05, R07, R21, R74, R75, R76, R83 Chronic obstructive pulmonary disease or Asthma R95, R96 Other respiratory symptoms R01, R02, R06, R45 PA Supervising GP PA Supervising GP Nurse Practitioner Total Most frequent presenting problems In order to facilitate broadly comparable measures of activity within each practice in the analysis reported above, the volume of data used was limited to 75 consultations for each PA and supervising GP. However, both PAs supplied additional data relating to their own activity, giving a total of 129 consultations in 17 sessions for the PA at the 24