Audit. The implementation of an end-of-life integrated care pathway in a Chinese population

Transcription

1 The implementation of an end-of-life integrated care pathway in a Chinese population S-H Lo, C-Y Chan, C-H Chan, W-k Sze, K-K Yuen, C-S Wong, T-Y Ng, Y Tung Abstract The integrated care pathway is used in end-of-life care to improve quality of care; the Liverpool Care Pathway (LCP) has been used in Europe and North America. Tuen Mun Hospital is a regional hospital in Hong Kong, China. The End-of-life Care Pathway (ECP) based on the concepts used in the Liverpool Care Pathway, was developed, with modifi cation to suit the local condition. Criteria for entry onto the ECP were that the multidisciplinary team agreed the patient was dying, and was at least two of the following: bedbound; semi-comatose; only able to take sips of fl uid; no longer able to take tablets. The ECP template replaced all other inpatient documents. The ECP was implemented in the palliative care unit for terminal cancer patients. An audit was performed to review the result. Fifty-one Chinese patients were included in the audit with mean age 64. The median duration of ECP use was 24 hours. All patients had current medication assessed and non-essential drugs were discontinued. The audit result suggested integrated care pathway in end-of-life care could be implemented successfully in an Oriental culture. The acceptance of using the ECP as a standard clinical practice takes time and education. Appropriate template design and supervision are the keys to success. Key words: End-of-life care Integrated care pathway Chinese Nurse education Sing-hung Lo, Kwokkeung Yuen, Chi-sing Wong, Ting-ying Ng are Associate Consultants; Chau-ying Chan and Chun-hung Chan are Advance Practice Nurses; Wing-kin Sze is Consultant; and Yuk Tung is Chief of Service, Department of Clinical Oncology, Tuen Mun Hospital, Hong Kong Correspondence to: Sing-hung Lo losinghung@ yahoo.com.hk End-of-life care is an essential part of palliative care. The Liverpool Care Pathway (LCP) has been developed in the United Kingdom (UK) to standardize the quality of care for dying patients in hospice and non-hospice settings (Ellershaw et al, 1997; Ellershaw, 2003). It is based on the principle of the integrated care pathway, which aims to facilitate distinct care through integrated multidisciplinary cooperation (Kitchiner et al, 1996). The LCP sets up 18 goals in the care of the dying patient, involves a multidisciplinary team in diagnosing and caring for the dying patient, and uses a template to replace all other patient care documents (LCP Central Team UK, 2009). It emphasizes comfort measures, anticipatory prescribing of medicines and discontinuation of inappropriate interventions in the care of the dying patient. A variance analysis system is incorporated in the LCP for continuous improvement. Through years of development, the LCP has been used in the UK and Netherlands (Veerbeek et al, 2006). In Australia an audit of the care of the dying based on the goals of the LCP has been published (Hardy et al, 2007). The use of an integrated care pathway in end-of-life care in a Chinese or Asian population is not available in literature. Tuen Mun Hospital (TMH) is a regional hospital in Hong Kong serving a population of 1 million. The Palliative Care Unit in TMH is an integral part of the Department of Clinical Oncology. The multidisciplinary team consists of palliative medicine specialists, clinical oncologists, nurse, clinical psychologists, medical social workers, occupational therapists, physiotherapists and chaplain/pastoral care workers. The service covers breaking bad news, psychosocial and spiritual support, symptom control and bereavement. The palliative care unit has a new caseload of 1000 per year. In order improve the service to the unit s terminal cancer patients, the team assigned a workgroup (consisting of one doctor and two nurses) to review the concepts in LCP, with modification according to local situation, and develop a clinical care pathway: End-of-life Care Pathway (ECP). Preparation The workgroup recommended that in order to implement the pathway as soon as possible, it was better to keep the goals simple and minimize administrative work. The workgroup suggested starting the ECP in two phases. In the first phase of ECP, seven goals were defined for the optimal care of the dying (Table 1). Some goals in the LCP were not implemented in the ECP. Communication with general practitioner (goal 9 and 12 in the LCP) was not thought appropriate in Hong Kong as most patients do not have a regular primary doctor. A majority of patients were Chinese, and a language assessment (goal 4 in the LCP) was not 384 International Journal of Palliative Nursing 2009, Vol 15, No 8

2 considered necessary at this point. In a majority of cases the patient s next of kin had stated that they would like to be informed of the patient s death as soon as possible, as they may have a ritual procedure to perform. Therefore, identify how family/other are to be informed of the patient s impending death (goal 7 in the LCP) was considered not necessary. Standardized protocol in mortality management (goal in the LCP) were already available, therefore the workgroup recommended to review these procedures in phase two, to minimize the administrative impact to frontline staff during implementation of the new pathway. Providing families with a leaflet on hospital information and bereavement (goals 8 and 18 in LCP) would also be implemented in phase two. The recommendations were approved by the multidisciplinary team. Criteria for entry onto the ECP were that the multidisciplinary team agreed the patient was dying, and was at least two of the following: bed-bound; semi-comatose; only able to take sips of fluid; no longer able to take tablets. The ECP template replaced all other inpatient documents for this specific stage of care. The template consisted of initial assessment and ongoing assessment. Before implementation, the template was distributed to the frontline staff for feedback, and revisions were made. Two education sessions were organized to brief the staff on the objectives, entry criteria and procedures of ECP. Staff anxiety on the withdrawal of medical routine such as blood pressure measurement was reassured. A nursing member of the workgroup was assigned in-ward to provide onsite supervision to frontline staff. Guidelines were available in-ward for reference. The ECP program began in November 2007 in a designated palliative care ward as a pilot project. The team performed an evaluation of the pilot project to investigate the effects of using the ECP in a Chinese population. Methods The hospital records of patients served with ECP in the Palliative Care Unit in TMH between November 2007 to August 2008 were reviewed. The definitions of standards used in the initial assessment are: current medication assessed and non-essential drugs were discontinued; as required subcutaneous medications were written up (prescribed); inappropriate interventions were discontinued; patient awareness of his/her dying condition was assessed; family awareness of patient dying condition was assessed; religious or Table 1. Goals of the end-of-life care pathway Goal Definition Sucess criteria 1 Current medication Current medication assessed and assessed and non- non-essential drugs were essentials discontinued discontinued 2 As-required subcutaneous As-required subcutaneous medication written up medications were written up 3 Discontinue inappropriate Inappropriate interventions were interventions discontinued 4 Insight into condition is Patient awareness of his/her dying assessed condition was assessed 5 Plan of care explained and Family awareness of patient dying discussed with patient and condition was assessed and family/other 6 Religious/spiritual needs Religious or spiritual needs of assessed patient and caregiver were assessed 7 Symptom assessment and Patient has good symptom control treatment given appropriately in his/her last 24 hours Table 2. Patient characteristics Characteristic Tuen Mun Liverpool* Rotterdam* Hospital Number of patients Gender (male) 60% 50% 50% Mean age (range) 64 (47 95) 61 (40 76) 61 (40 76) Median hours of 24 (8 460) 29 (3 213) 28 (2 218) Clinical Care Pathway use (range): 0 24 hours 30 (57%) 19 (48%) 18 (46%) hours 4 (8%) 4 (10%) 11 (27%) >48 hours 18 (35%) 17 (42%) 11 (27%) Table 3. Patient characteristics: site of primary tumour Primary tumour site Number of patients (%) Lung 13 (25.5%) Liver 7 (13.8%) Rectum 5 (9.8%) Colon 4 (7.8%) Stomach 3 (5.9%) Breast 3 (5.9%) Obstetrics and gynaecology tumour 4 (7.8%) Pancreas 2 (3.9%) Head and neck tumour 2 (3.9%) Unknown primary with metastasis 3 (5.9%) Other 5 (9.8%) spiritual needs of patient and care giver was assessed. These success criteria (Table 1) were used by another audit published in 2006 (Veerbeek et al, 2006). For ongoing assessment International Journal of Palliative Nursing 2009, Vol 15, No 8 385

3 Table 4. Initial assessments Success criteria Status Tuen Mun Hospital (n=51) Liverpool (n=40)* Rotterdam (n=40)* Current medication assessed 100% (51) Not available Not available and non-essential drugs were discontinued As-needed subcutaneous Achieved 92.2% (47) 87% 95% medications were written up Not achieved 7.8% (4) 0 5% Missing 0 13% 0 Inappropriate interventions Achieved 94.1% (48) 85% 100% were discontinued Not achieved 5.9% (3) 0 0 Missing 0 15% 0 Patient awareness of his/her Achieved 35.3% (18) 37% 60% dying condition was assessed Not achieved 0 5% 5% Missing 25.5% (13) 33% 0 Cannot be done 39.2% (20) 25% 35% Family awareness of patient Achieved 74.5% (38) 87% 100% dying condition was assessed Not achieved Missing 25.5% (13) 13% 0 Religious or spiritual needs of Achieved 71% (36) 84% 78% patient and caregiver were Not achieved 2% (1) 3% 17% assessed Missing 27% (14) 13% 5% Table 5. Symptom control in the last 24 hours Success criteria Status Tuen Mun Hospital (n=51) Liverpool (n=40)* Rotterdam (n=40)* Number of patients per shift Patient was pain free Achieved 87% 82% 86% Not achieved 13% 14% 8% Missing 0 4% 6% Patient was not agitated Achieved 95% 74% 83% Not achieved 5% 22% 12% Missing 0 4% 5% Patient was free from Achieved 84% 84% 87% excessive secretion Not achieved 16% 13% 8% Missing 0 3% 5% the team looked at the control of three common symptoms in the last 24 hours of life: pain, agitation and respiratory tract secretion. As the symptoms were observed in every nursing shift (am, pm and night), the total number of observations made in the 24 hours before death in each setting were calculated. The proportion of observations for each symptom that was documented as under control was then calculated. The results were compared with those patients who died in the Marie Curie Hospice of Liverpool, between April 2002 and July 2003 and Erasmus Medical Centre of Rotterdam, between October 2001 to January 2003, using the LCP (Veerbeek et al, 2006). All patients were Chinese and 18 years or older when they died. The data were analyzed descriptively. Results Altogether 51 patient s hospital records were included in this audit. The mean age at death of the patients was 64 years, compared to 61 in the other two institutes (Table 2); 60% were men. The most common primary tumour was lung, followed by liver and rectum (Table 3). The median duration of clinical care pathway use was 24 hours (range: hours), compared to 29 hours in Liverpool (range: hours) and 28 hours in Rotterdam (range: hours). For the initial assessment, all patients had current medication assessed and non-essential drugs were discontinued (Table 4). As required, subcutaneous medications were completely written up (prescribed) in 92.2 of patients. 386 International Journal of Palliative Nursing 2009, Vol 15, No 8

4 Inappropriate interventions were discontinued in 94.1% of patients. Assessment of patient and family awareness of dying condition was done in 35.3% and 74.5% respectively. Religious or spiritual needs of the patient and caregiver were assessed in 71%. The compliance was affected by 25% missing data in the assessment of awareness of dying condition and religious or spiritual needs. For symptom control in the last 24-hours of life, 87% of patients were free from pain, 95% were free from agitation and 84% were free from excessive secretions. These figures were comparable to that of Liverpool and Rotterdam (Table 5). Discussion The use of the end-of-life clinical care pathway has been developing since the 1990s (Ellershaw et al, 1997). Today end-of-life pathways are used widely: in UK and Europe as LCP; and Palliative Care for Advanced Disease in the USA (Bookbinder et al, 2005; Luhr et al, 2005). However, as the studies on the use of these integrated care pathways are descriptive, and randomized studies are not practical, their benefit has been criticized (Shah, 2005). The continuous sedation used in the protocol also raised concern (Murray, 2008; Treloar, 2008). Up to now there is general consensus in literature that integrated pathways in end-oflife care improve symptom control in patients, as well as standardization, continuity and collaboration among the interdisciplinary team (Luhr, 2007). As there was no literature available on the use of the LCP in Asia, the goals listed in the LCP were reviewed and some of them were selected as the goals of the pilot project (Table 1) to suit the local situation. In this review, most of the care goals for dying patients were achieved in the majority of patients. Patients received an initial assessment of symptoms and appropriate adjustment of intervention in the ongoing assessment. The result is comparable to the other two institutes. The result suggests that the integrated care pathway is feasible in providing quality care during the end-of-life phase in an Oriental cultural population (as all patients and staff involved were Chinese). Only 35.2% of patients had been assessed for their awareness of dying condition, which is the lowest among the three institutes. This is possibly due to the relatively late initiation of the ECP, as suggested by the shorter duration of service with the ECP in our patient compared to the others. The 25% missing data in the assessment of awareness of dying condition and religious or spiritual needs may also lower the compliance rate. There are points worth noting during comparison. Firstly the cancer incidence in the European population is different from the Chinese population. In Hong Kong we have a much higher incidence of hepatocellular carcinoma (Yuen et al, 2009), which was the second most common primary tumour in this cohort. Therefore the spectrum of symptoms may be different during the dying phase. Secondly, the assessment interval of ongoing assessment is 4 hours in the LCP but 8 hours in the ECP, due to the limitation in manpower. ECP was new to the unit. Training sessions had been provided to equip the staff before implementation. In this review, most of the template record was completed, except in the psychological and spiritual assessment part, which had 25% missing data. The error occurred mainly in the early phase of the implementation. The condition was steadily improved after further on-site supervision. The design of the template may also play a role in this issue. In the second phase of the implementation of the ECP after this review, a check box has been incorporated to the revised ongoing assessment form to confirm the completion of the initial assessment. New goals on care after death and information leaflets have been added to further improve our quality of care. Further education sessions have also been conducted to enhance staff competence in using ECP. During the early phase of the introduction of ECP, only 10% of the patients who died in the Palliative Care Unit were served with ECP. It reflected the uncertainty that doctors and nurses still have in committing to a diagnosis of a dying patient, even when clear criteria had been suggested. After one year of experience, 40% of the patients who have died in the unit have received ECP, and the trend is still rising. Conclusion The use of the ECP provided a foundation for good symptom control, psychological, social and spiritual support. As suggested by our evaluation, good quality of care could be achieved, in a cohort of patients and staff with an Oriental cultural background using integrated care pathway. The acceptance of using the ECP as a standard clinical practice takes time and education. Appropriate template The integrated care pathway may be feasible in providing end-of-life care to an Oriental population International Journal of Palliative Nursing 2009, Vol 15, No 8 387

5 design and supervision are the keys to success. Ethics This work was based on a retrospective review; therefore, separate ethic approval was not considered necessary. All characteristics that might have identified place or person have been removed. IJPN Acknowledgement The authors would like to thank all members of the palliative care team, especially the staff in H1 ward of Tuen Mun Hospital for their support with this program and Miss Ka-sin Yu for her effort in data analysis. Hardy J, Haberecht J, Maresco-Pennisi D, Yate P, Australian Best Care of the Dying Network, Queensland (2007) Audit of the are of the dying in a network of hospitals and institutions in Queensland. Int Med J 37(5): Bookbinder M, Blank A, Amey E et al (2005) Improving end of life care: development and pilot test of a clinical pathway. J Pain Symptom Manage 29(6): Ellershaw J (2003) Introduction. In: Ellershaw J, Wilkinson S, eds. Care of the Dying: a pathway to excellence. Oxford University Press Ellershaw J, Foster A, Murphy D, Shea T, Overill S (1997). Developing an integrated care pathway for the dying patient. Eur J Palliat Care 4(6): Kitchiner D, Davidson C, Bundred P (1996). Integrated care pathways: effective tools for continuous evaluation of clinical practice. J Eval Clin Pract 2(1): 65 9 LCP Central Team UK MCPCIL. 10 Step Continuous Quality Improvement Programme (CQIP) for Care of the Dying using the LCP Framework. The Marie Curie Palliative Care Institute, Liverpool www. mcpcil.org.uk (accessed 13 August 2009) Luhr C, Meghani S, Homel P et al (2005) Pilot of a pathway to improve the care of imminently dying oncology inpatients in a Veterans Affairs Medical Center. J Pain Symptom Manage 29(6): Luhrs C, Penrod J (2007) End-of-life care pathways. Curr Opin Support Palliat Care 1(3): Murray SA (2008). Continuous deep sedation in patients nearing death. BMJ 336(7648): Shah SH (2005) The Liveropool Care Pathway: its impact on improving the care of the dying. Age Aging 34: Treloar AJ (2008) Dutch research reflects problems with the Liverpool care pathway. BMJ 336(7650): 905 Veerbeek L, van Zuylen L, Gambles M et al (2006). Audit of the Liverpool Care Pathway for the Dying Patient in a Dutch Cancer Hospital. J Palliat Care 22(4): Yuen MF, Hou JL, Chutaputti A (2009) Hepatocellular carcinoma in the Asia pacific region. J Gastroenterol Hepatol 24(3): International Journal of Palliative Nursing 2009, Vol 15, No 8