A new quality assurance package for hospital palliative care teams: the Trent Hospice Audit Group model

Cancer patients in hospitals are increasingly cared for jointly by palliative care teams, as well as oncologists and surgeons. There has been a considerable growth in the number and range of hospital palliative care teams (HPCTs) in the United Kingdom. HPCTs can include specialist doctors and nurses, social workers, chaplains, allied health professionals and pharmacists. Some teams work closely with existing cancer multidisciplinary teams (MDTs) while others are less well integrated. Quality assurance and clinical governance requirements have an impact on the monitoring of such teams, but so far there is no standardised way of measuring the amount and quality of HPCTs' workload. Trent Hospice Audit Group (THAG) is a multiprofessional research group, which has been developing standards and audit tools for palliative care since the 1990s. These follow a format of structure–process–outcome for standards and measures. We describe a collaborative programme of work with HPCTs that has led to a new set of standards and audit tools. Nine HPCTs participated in three rounds of consultation, piloting and modification of standard statements and tools. The final pack of HPCT quality assurance tools covers: policies and documentation; medical notes review; questionnaires for ward-based staff. The tools measure the HPCT workload and casemix; the views of ward-based staff on the supportive role of the HPCT and the effectiveness of HPCT education programmes, particularly in changing practice. The THAG HPCT quality assurance pack is now available for use in cancer peer review.

Specialist palliative care has, since the 1980s, expanded its services in order to offer direct care and support to people living at home and to those in acute hospital beds and outpatient clinics.
The Calman-Hine report (1995) emphasised the requirement for the provision of palliative care throughout cancer care: 'there should be a smooth progression of [palliative] care between home, hospital and hospice' (p 16). The National Council for Hospice and Specialist Palliative Care Services (NCHSPCS) Occasional Paper 10 'Palliative Care in the Hospital Setting ' (1996) stated that A hospital palliative care team provides specialist palliative care within the acute hospital setting. The team has an advisory and educational role and may also provide direct care to patients and their families requiring a high level of palliative care skills (p 7).
A later NCHSPCS document 'Palliative Care 2000' (1999 identified the core and extended members of a hospital palliative care team as One or more nurses who hold or who are working towards a specialist practitioner recordable qualification in palliative care; a consultant in palliative medicine supported by other medical staff including junior staff who may be on rotations. Secretarial/administrative support. The extended team should include chaplaincy, social work, psychology and pharmacy expertise and access to specialist pain management. In addition, there should be access to physiotherapy, occupational therapy and dietetics.' (p 41) HPCTs have since grown to include dedicated social workers, chaplains, allied health professionals and pharmacists. The emergence of cancer site-specific multidisciplinary core and extended teams has seen a greater specialisation for many palliative care clinical nurse specialists (CNSs) and a further involvement in joint clinic activity for medical members of the team.
The Clinical Governance agenda for the NHS (adapted, but with the same component requirements, for independent hospice services) has impacted upon HPCTs, like all other palliative care service providers. This has included a rigorous review of systems, policies and procedures, documentation and the adoption of appropriate quality measures to assist the evidence base for clinical effectiveness (NCHSPCS, 2000).
Measures used to date have focussed on workload/casemix and the effectiveness of interventions.

Workload
The 'levels of intervention' first described by Webber (1994) range from telephone advice to regular visiting of a patient to manage, monitor and reassess. This has become the standard method of allocating and recording workload among teams.

Effectiveness of interventions
Quality measures have been used by HPCTs to monitor the effectiveness of interventions. The Support Team Assessment -STAS (Higginson, 1993) (modified for hospital use from a tool developed for use in the community) and its 'daughter' measure, the Palliative care Outcome Scale -POS (Hearn and Higginson, 1999) have both figured as tools. A modification of STAS, E-STAS has also been established 'as a useful tool to evaluate interventions by a hospital palliative care team ' (Edmonds et al., 1998). Ellershaw et al. (1995) developed the patient care assessment form (PACA) specifically to measure the effectiveness of a hospital palliative care team 'in the provision of symptom control, patients' and relatives' awareness of the diagnosis, and outcome regarding the patients' placements'.
To date the only standards developed for HPCTs are in the NHS Manual of Cancer Standards (2000). These are limited to a defined membership of the core and extended team (with qualifications for CNS), requirements for regular clinical meetings to discuss patients, regular meetings to discuss operational issues, the availability of the team 'out of hours', the production of a service directory, the existence of a palliative care strategy group and for clinical audits to be carried out by the team.
The aim of the work described in this paper was to develop (and validate) a comprehensive set of standards for HPCTs together with audit tools to monitor compliance.

TRENT HOSPICE AUDIT GROUP MODEL
The Trent Hospice Audit Group (THAG) was formed in 1990 as a working group of senior physicians and nurses from specialist palliative care in the former Trent Region of the UK. It has produced 'Palliative Care Core Standards' for inpatient palliative care services (Ahmedzai et al, 1998) based on the Donabedian framework of Structure -Process -Outcome (S -P -O) criteria, with audit tools (Box 1).
Since 2001, THAG has concentrated on broadening its quality measures to incorporate palliative care services in the community and acute hospitals. These developments have incorporated the views of a wide range of palliative care service providers, the quality measure 'users'. This paper describes the development of a Hospital Palliative Care Team standard and audit package using the THAG approach.

METHODS
The THAG Hospital Palliative Care Team Standard was initially developed in 2001 by an ad hoc panel of consultants in palliative medicine and clinical nurse specialists, responding to a call for those interested in producing a quality measure. The panel used policy and planning documents produced by the NCHSPCS as reference.
The main aims of the standard development group were: To produce a quality measure for specialist hospital teams.
To audit not only the effectiveness of the HPCTs' systems, but also its impact, on patient needs, on ward team staff and on the development of knowledge, skills and practice within an acute hospital setting.
To enable palliative care teams and services to develop an acute hospital service using the standard as guidance.
As described earlier, criteria were developed using the S -P -O format. Box 2 contains the standard statement and examples of the Structure, Process and Outcome criteria.
The first draft package included: A documentation audit -designed to ensure that the HPCT had dated, signed operational policies and procedures; that it performed its resource function, with documented guidelines on pain and symptom control measures and evidence of educational programmes and that it recorded, in its own documentation, key aspects of intervention activity. (The auditor noted each recorded visit, the discipline of the HPCT member who made the visit and the nature of the visit; assessment, monitoring, symptom control advice, psychological support to patient and/or informal carer, discharge planning discussions or liaison with other professionals.) A questionnaire for ward staff -designed to elicit staff's views on the availability and effectiveness of the HPCT in supporting patients and families, in advising on symptom management and in delivering staff support. It also sought to identify the level of attendance at HPCT educational events during the previous 2 years. An education evaluation questionnaire -designed to enable ward-based staff to indicate the value of HPCT teaching in terms of new knowledge, new skills and change in practice.
Circulation for consultation followed. Teams were targeted using two criteria: Hospital based teams in Trent -including the consultants in palliative medicine and clinical nurse specialists who had participated in the development. Teams who had registered with THAG as users of other existing standards or interested services.
Four acute hospital trust HPCTs -two in Trent, one each from the North West and the South East -were identified as pilot sites Box 1 Donabedian's structure -process -outcome criteria Structure, process and outcome Definitions Structure The inputs to care: staff and resources, including the physical and organisational settings in which the service is provided. Structure is relatively easy to measure, for example, staff numbers, beds, size of budget, but its relationship to quality of outcome (see below) is uncertain Process The activities of providing care, which uses resources and produces outcomes. Measuring process can include both what is done (e.g. counting staff -patient contacts) and how it is done (e.g. what procedures are followed) Outcome The result of the clinical intervention, as reflected in the patient's health status and quality of life for the revised draft (Table 1). First pilot audits took place between January and March 2002. The same auditor (JH) undertook all pilot site visits.
The main purpose of the pilot audits was to test the usability, relevance and coverage of the audit tools. The HPCTs received a written report within 2 weeks of the audit exercise. The reports were consistently divided into three sections: The response to the evidence provided relating to the audit questions. Comments or suggestions, which the team might find useful, relating to some of the evidence material. Recommendations for changes or additions to the standard criteria or audit tool questions.
Following a pilot audit, further revisions were made and a second round of auditing took place during July and August 2002 using version 6. Five Trust HPCTs were visited. None was the same as in the first phase of piloting. The sites were in London ( Â 3), the North West and Trent (Table 1).
Guidance was offered to participating teams to assist their preparation. Each team identified a coordinator. In most cases this was a CNS but a Specialist Registrar in Palliative Medicine and an Administrator also undertook that role. The documentary evidence was collected into a folders or folders, titled or divided into S -P -O sections. The coordinator facilitated visits to the wards (selected by the site's HPCT) and introduced the auditor. Recruitment of staff for interview was undertaken by the ward manager. Visit times were at the convenience of the ward. The coordinator enabled access to patient's medical records. Five sets of records, of most recently discharged or deceased patients in whose care the HPCT had participated, were made available, either on the ward or in the HPCT office. Coordinators made themselves available to answer auditor queries/questions at times throughout the visit.
Teams were asked to comment on the reports that they received following the audits.
Following the second pilot audit, the final draft consultation document was completed and disseminated to a total of 23 hospital palliative care team services in October 2002. The services consisted of the participating teams, those registered as THAG Users (interested in participating in the development of the standard/audit tools) and other units selected from the Hospice Information directory (2002).
The process of consultation, testing and review is designed to establish two aspects of validity: face validity (i.e. is the tool clear, logical and relevant?) and content validity (i.e. does the measure omit anything important?).

RESULTS
The auditor identified 20 changes to the standard criteria, audit tools and auditing documentation from the first round of testing. Some common themes for change were identified from the majority of reports. Table 2 illustrates the value of testing the tools, highlighting the gaps in criteria or the absence of specific audit questions. It also offers an insight into the process of standard development, showing how specific criteria change over a span of time, testing and review. There was substantial redrafting and reformatting of the standard package, with a further consultation round following on.
A further 10 changes to the standard criteria, audit tools and documentation were identified (Table 3 contains the details). Table 4 illustrates the range of comments from the participating pilot sites on the format and content of their individual audit reports.
An indication of the relevance and acceptability of the audit tool is confirmed by the observation that when the pilot HPCTs were asked whether they would like to accept THAG as their quality measure of choice, the majority (seven of nine) stated that they would, while the remaining two services indicated 'probably' and 'possibly' to the question.
The sample, though a small proportion of HPCT services, covered a wide geographical spread, from the North-West to the South-East of England. Responses to the preparation for testing, and the feedback following audit reporting, strongly suggest that Box 2 THAG hospital palliative care team -standard statement and examples of structure, process and outcome criteria Standard statement: The Hospital Palliative Care Team is an effective resource to referring hospital specialties, patients, their carers/families and staff Structural criteria: (examples only) S2: The team has an operational policy relating to its clinical role and function S9: The team has written guidelines for the control of pain and other distressing symptoms, for ward-based teams S11: The HPCT has a system for maintaining records of referrals, discharges/deaths, details of counselling and bereavement support to families/carers and quick reference on medication, pain control and other key clinical information  Although the aim of the pilot audits was to validate and refine the audit tools, the results of the audit were made available to teams for their use. Some examples are included here to illustrate the types of information that can be obtained:

Workload/casemix
Of 87 sets of HPCT documentation reviewed, the levels of intervention were recorded in 49% (43/87); 82 sets of medical records were reviewed overall. The range of HPCT recorded visits to patients, across the pilot sites, was 1 -32 per patient.
Reasons for referral and continued support were consistent across the pilot sites and covered the range of HPCT activity. However, there were some developments in activity, which reflected pressure on ward staff, rather than a natural extension of the role of the HPCT, for example, liaison with other professionals and discharge planning. Here, HPCT staff appeared to have taken on total responsibility for these aspects of care, with ward-based representatives indicating that this was what they expected from the specialist resource.

Ward staff views
'[They are a] good coordinator of care, organising family meetings and appropriate discharge plans', and '[They help] patients and families come to terms with what's happening, we don't have the time to do that stuff, can't organise complex discharges'.
Ward staff comments on the supportive role of the HPCT for patients, families and professionals were consistent across the pilot sites.  and add a note to the auditor to ensure that the correct phrase is used S9/10: create tick boxes for evidence relating to service details (S9) and aspects of management (S10)

Effectiveness of education programmes
Of the 36 ward staff interviewed overall, 78% had not attended an HPCT educational programme in the previous 2 years.
Three pilot sites presented completed modified THAG Education Evaluation Questionnaires as evidence. The questionnaire asks ward staff (predominantly nurses) to identify the education programme, offered by the HPCT, that they have attended within the previous 2 years.
Staff were asked to comment on the improvements in their knowledge and skills in palliative care as a result of attending the study event. They were also asked to state whether any area of their practice had changed following the educational programme. The response rates in the three areas were 13% (four out of 30); 69% (34 out of 49) and 25% (six out of 24) respectively. Combining the results from the three areas 61% (27 out of 44) respondents identified an improvement in knowledge, 59% (26 out of 44) an improvement in skills, and 43% (19 out of 44) a change in palliative care practice.
Two pilot sites were using a PACA record in their documentation. However, there was either a single entry record or the frequency of recording did not reflect the number of visits to the patient.

DISCUSSION
The principal aim of testing the audit tools was achieved. It enabled us to refine them, particularly the documentation for completion by the auditor. Comments from services on the criterion statements were incorporated into subsequent revisions.
This has established face and content validity of the standards and audit tools.
The example results from the pilot audits helped to describe the workload and casemix of an HPCT, ward staff views of the team and the effectiveness of education programmes provided by the team. For individual teams these results not only provide a measure of quality assurance but also facilitate a review of the service and inform future service developments.
The recent consultation has also raised a number of areas that require further consideration, including telephone advice activity and its recording as workload, and the need for an HPCT to actively engage with its palliative care network. A more significant issue is that of assessing whether an HPCT is adequately resourced to support its hospital population, acknowledging the differing dependency of patients.
Clearly, it is important to continue the development of a range of measures to monitor the effectiveness of hospital-based palliative care. A recent systematic literature review, to determine whether hospital-based palliative care teams improve the process or outcomes of care for patients and families at the end of life (Higginson et al, 2002), recommended that future evaluations should: 'Compare different models of hospital based team, namely those giving a more intensive intervention vs those that are more advisory and concentrate on education and have lower contact with patients. Use standardised outcome measures assessing patient pain, symptoms, carer outcomes and, where possible, the effect on professionals or the overall hospital service' (p 104).
The revised version of the national Manual of Cancer Standards is due to be published in early 2004. The new specialist palliative care standards are likely to reflect the recommendations of the NICE Guidance on Supportive and Palliative Care for Cancer, which is also due to be published in early 2004. They will probably be of similar style to those from the first version of the Manual, that is, will largely focus on structure with some process standards.
The standards and audit tools reported here cover a wide range of aspects of HPCT work and we feel provide a useful framework for HPCTs to quality assure, review and develop their services. It is likely that they will complement the new National Standards when they become available.