Palliative care is the care of individuals with life limiting illnesses; people dying from active progressive diseases or other conditions that are not responsive to curative treatment. Such care aims to optimise an individuals quality of life and support the individuals family/whnau and other caregivers through the illness and after death.1Palliative care endeavours to anticipate needs and provide symptom control to maximise patients quality of life. It has been recognised that a busy, noisy and often overcrowded Emergency Department (ED) may not be the optimal treatment environment for patients with terminal illness.2,3Consequently a palliative care patient who needs to visit the ED may be considered as a failure of the health care system and some have suggested reduced use of the ED to be a quality measure.4Research has shown that a common cause for visits to the ED is the need for symptom control, for example, increasing dyspnoea or for pain relief.5 It is acknowledged that Mori access to, utilisation and needs for palliative care services differs to non-Mori.1,6,7 There may also be a lack of understanding on behalf of some mainstream providers of the whnau ora model of health and illness, and Mori cultural beliefs in providing care for the dying.8,9Strategies identified for developing culturally appropriate palliative care services for Mori have included promoting informal links between service providers in local areas such as between palliative care providers and Mori development organisations; and employing and training staff to enable them to meet the needs of Mori. However the increased availability of culturally appropriate services does not seem to have increased access and utilisation by Mori.10Waikato District Health Board covers a population of 369,000 of whom, 129,000 (35%) live in Hamilton . Twenty percent of the District Health Board population are Mori compared with 14% for all of NZ (although only 8% of those aged over 60 years in Waikato are Mori). The Waikato District Health Board operates five EDs: one in a major tertiary hospital; one that covers the Thames/Coromandel region and three smaller centres south of Hamilton in Te Kuiti, Taumarunui and Tokoroa (known locally as the Southern T Hospitals).The palliative care service at Waikato District Health Board support patients through a multidisciplinary team of health professionals, including nurses, doctors and social workers. It is based at Waikato Hospital in Hamilton and has an established staff of three specialists in palliative medicine, two advanced trainees as well as six specialist palliative care nurses. Patients maybe admitted to Waikato Hospitals under the care of a specialist physician or oncologist, or may be admitted directly under the care of a palliative care physician.The service also visits the rural hospitals and provides Community Liaison to Waikato Hospice consulting on Hospice inpatients as well as providing domiciliary visits. Patients referred by a general practitioner may be admitted directly to a ward or may be seen in the Emergency Department. There is a 24-hour service by consultants and registrars. After hours the Palliative Care phone is switched to the Oncology ward where the calls are triaged.Health Care professionals at Hospice Waikato and in Community have direct access 24hrs a day to consultants. All newly referred patients whether from general practice or referrals from other hospital specialties are registered on the Palliative Care database. Registered patients are advised of the services available to them in the community and how to contact the palliative care service.Once registered subsequent referrals from a general practitioner will usually be directed to the ward and so only patient initiated visits to ED will be recorded subsequently. It is believed that there should ideally be few circumstances where patients who are under the care of the specialist palliative care service present to ED where they may be seen by a generalist who is not familiar with their case. Thus such presentations can be deemed a failure in patient management or in communication.The aim of this study was to examine attendances by palliative care patients at Waikato EDs. We aimed to identify the number, illness and palliative care needs of people presenting at ED in relation to patient age, gender, ethnicity and domicile. We also wished to examine the outcome of their visits.Method This was an observational study using record linkage. The study population was palliative care patients aged 20 years of age and older who were on the Waikato DHB palliative care register between 1 November 2010 and 31 October 2011 and were domiciled in the Waikato DHB catchment zone. We used hospital attendance records to identify all visits to any one of the five EDs by patients who were on the palliative care register in the defined time period. Patients were identified through their unique National Health Index (NHI) number. A spreadsheet was used to enter data from the two patient data systems. For each patient their age, gender, ethnicity, domicile (Hamilton or other), hospital visited and number of visits to ED were recorded from the hospital records. Diagnosis and date registered with the Waikato DHB palliative care service were identified from the palliative care register. Ethical approval was granted by the local Health and Disability Ethics Committee (Northern Y Ethics NTY 11/11/111). The key outcome of interest in our analyses was a visit to the ED. The attendance rate was analysed by age in the groups; 20-49 years, 50-59, 60-69, 70-79 and 80+, gender and ethnicity (categorised as Mori, Pacific or other). We also looked at attendance rate by diagnosis this was categorised into 9 groups: breast cancer, colorectal cancer, gynaecological cancer (cervical, uterine, ovarian etc), lung cancer, skin cancer, upper GI cancers, urogenital cancers (prostate, bladder, renal etc) and other cancers (e.g. lymphoma, sarcoma etc). A single category of non-cancer was included for diseases such as chronic obstructive pulmonary disease, congestive heart failure and neurological disorders. We undertook univariate analyses of attendance at ED by age category, gender, ethnicity and illness cause. We then undertook a multivariate analysis looking at first attendance at ED adjusted for age category, gender, ethnicity and illness cause. We also looked at reasons for attendance at ED and the outcome of the visit e.g. admission to hospital or discharge to the community. These outcomes were adjusted for age and ethnicity. Data were analysed using Microsoft Excel (Microsoft Corp, 2007) and STATA v11.2 (StataCorp, 2011) software. Results 1185 palliative care patients residing in the Waikato DHB catchment area were identified from the palliative care register from 1 November 2010 to 31 October 2011. There were 645 men (54.4%) and 540 women. 197 Mori (16.6%), 18 Pacific (1.5%) and 973 (81.9%) other /non-Mori were identified. The age ranged from 20-101 years (interquartile range 63-81). The mean age overall was 70.8 years: non-Mori 72.5 years, Mori 63.3 years and Pacific 64.1 years. The mean length of time on the register during the year of interest was 120.7 (median 66) days: women 129 days (67.5), men 113 days (66), Mori 123 days (62), Pacific Peoples 68 days (31) and for non-Mori 121 days (69). 449 (37.9%) of the study population visited ED at least once during the study period. The distribution of visits to ED is presented in Figure 1. The percentage of Mori who visited ED at least once had a similar likelihood of attendance as non-Mori (38.6% vs 37.4% respectively) and was not statistically different (Chi-squared=0.048, p=0.827). Figure 1. Distribution of visits to the ED The maximum number of presentations by any one patient was 12. Of the 449 patients who visited the ED at least once, 202 patients (45%) went on to have more than one visit. This was similar for Mori and non-Mori (15.2% vs 17.4%) and was not statistically different (Chi-squared=1.19, p<0.552). 38% of visits to ED were within a month of a previous visit to ED. This was also similar between Mori and non-Mori. When looking at the diagnoses responsible for patients referral to palliative care categorised into 9 major groups (Table 1), it was established that there was an excess of lung cancer cases for Mori compared with non-Mori (27.9% cf 17.3%) whilst urogenital cancers (principally prostate cancer) were more common in non-Mori (8.4% cf 4.6% Mori). Table 1. Illness categories for Mori and non-Mori palliative care patients Illness Mori Non-Mori Non-cancer 35 (17.8%) 182 (18.4%) Breast cancer 12 (6.1%) 60 (6.1%) Colorectal 21 (10.7%) 106 (10.8%) Gynaecological 7 (3.6%) 34 (3.4%) Lung 55 (27.9%) 171 (17.3%) Skin 0 (0%) 47 (4.8%) Upper GI 28 (14.2%) 135 (13.7%) Urogenital 9 (4.6%) 83 (8.4%) Other cancers 30 (15.2%) 170 (17.2%) All skin cancer cases on the register were for non-Mori. These are in keeping with the reported national incidences of these cancers.11 Otherwise the diagnosis between Mori and non-Mori were similar. Apart from gender-specific cancers, colorectal, lung, skin, upper GI and other cancers were more prevalent in men than women (60-70% vs 30-40%). The percentage of the study population with non-cancer causes of palliative care were also higher in men than women (Table 2). Table 2. Likelihood of palliative care patients attending ED within study period Variable Univariate analysis Multivariate analysis Gender * Female 33.7% 1.00 (reference) Male 41.4% 1.64 (1.25-2.14)** Ethnic group Other 37.4% 1.00 (reference) Maori 38.6% 0.99 (0.71-1.39) Pacific 55.6% (not included) Illness category Non Cancer 35.9% 1.00 (reference) Breast 38.9% 1.49 (0.83-2.67) Colorectal 37.8% 1.07 (0.67-1.71) Gynaecological 56.1% 3.33 (1.62-6.84)** Lung 39.8% 1.08 (0.73-1.60) Skin 40.4% 1.11 (0.58-2.15) Upper GI 35.0% 0.93 (0.61-1.44) Urogenital 37.0% 0.81 (0.48-1.38) Other Cancers 36.0% 0.93 (0.62-1.40) Age group 20-49yrs 37.8% 1.00 (reference) 50-59yrs 38.9% 1.00 (0.58-1.75) 60-69yrs 40.4% 1.09 (0.66-1.80) 70-79yrs 41.3% 1.15 (0.71-1.86) 80+yrs 32.1% 0.79 (0.48-1.30) *test for homogeneity (Pr>chi2) Gender=0.007 **P>|z| </=0.001 Women with gynaecological cancer were most likely to visit ED, otherwise cause of illness did not seem to influence the likelihood of visiting ED. A multivariate analysis of likelihood of a palliative care patient visiting ED and examining key variables including age (by age groupings), gender, ethnicity (Mori, Pacific and Other) and cause of illness revealed that being male increased the likelihood of attending ED by 60% (p=0.001) and having a gynaecological cancer increased the odds of attending ED by 230% (p<0.001). No other factor including ethnicity was associated with the risk of visiting ED. Of the 449 patients who visited an ED, 78.4% attended Waikato Hospital, 169/1185, 8.69% visited Thames Hospital and 12.9% vis

Palliative care is the care of individuals with life limiting illnesses; people dying from active progressive diseases or other conditions that are not responsive to curative treatment. Such care aims to optimise an individuals quality of life and support the individuals family/whnau and other caregivers through the illness and after death.1Palliative care endeavours to anticipate needs and provide symptom control to maximise patients quality of life. It has been recognised that a busy, noisy and often overcrowded Emergency Department (ED) may not be the optimal treatment environment for patients with terminal illness.2,3Consequently a palliative care patient who needs to visit the ED may be considered as a failure of the health care system and some have suggested reduced use of the ED to be a quality measure.4Research has shown that a common cause for visits to the ED is the need for symptom control, for example, increasing dyspnoea or for pain relief.5 It is acknowledged that Mori access to, utilisation and needs for palliative care services differs to non-Mori.1,6,7 There may also be a lack of understanding on behalf of some mainstream providers of the whnau ora model of health and illness, and Mori cultural beliefs in providing care for the dying.8,9Strategies identified for developing culturally appropriate palliative care services for Mori have included promoting informal links between service providers in local areas such as between palliative care providers and Mori development organisations; and employing and training staff to enable them to meet the needs of Mori. However the increased availability of culturally appropriate services does not seem to have increased access and utilisation by Mori.10Waikato District Health Board covers a population of 369,000 of whom, 129,000 (35%) live in Hamilton . Twenty percent of the District Health Board population are Mori compared with 14% for all of NZ (although only 8% of those aged over 60 years in Waikato are Mori). The Waikato District Health Board operates five EDs: one in a major tertiary hospital; one that covers the Thames/Coromandel region and three smaller centres south of Hamilton in Te Kuiti, Taumarunui and Tokoroa (known locally as the Southern T Hospitals).The palliative care service at Waikato District Health Board support patients through a multidisciplinary team of health professionals, including nurses, doctors and social workers. It is based at Waikato Hospital in Hamilton and has an established staff of three specialists in palliative medicine, two advanced trainees as well as six specialist palliative care nurses. Patients maybe admitted to Waikato Hospitals under the care of a specialist physician or oncologist, or may be admitted directly under the care of a palliative care physician.The service also visits the rural hospitals and provides Community Liaison to Waikato Hospice consulting on Hospice inpatients as well as providing domiciliary visits. Patients referred by a general practitioner may be admitted directly to a ward or may be seen in the Emergency Department. There is a 24-hour service by consultants and registrars. After hours the Palliative Care phone is switched to the Oncology ward where the calls are triaged.Health Care professionals at Hospice Waikato and in Community have direct access 24hrs a day to consultants. All newly referred patients whether from general practice or referrals from other hospital specialties are registered on the Palliative Care database. Registered patients are advised of the services available to them in the community and how to contact the palliative care service.Once registered subsequent referrals from a general practitioner will usually be directed to the ward and so only patient initiated visits to ED will be recorded subsequently. It is believed that there should ideally be few circumstances where patients who are under the care of the specialist palliative care service present to ED where they may be seen by a generalist who is not familiar with their case. Thus such presentations can be deemed a failure in patient management or in communication.The aim of this study was to examine attendances by palliative care patients at Waikato EDs. We aimed to identify the number, illness and palliative care needs of people presenting at ED in relation to patient age, gender, ethnicity and domicile. We also wished to examine the outcome of their visits.Method This was an observational study using record linkage. The study population was palliative care patients aged 20 years of age and older who were on the Waikato DHB palliative care register between 1 November 2010 and 31 October 2011 and were domiciled in the Waikato DHB catchment zone. We used hospital attendance records to identify all visits to any one of the five EDs by patients who were on the palliative care register in the defined time period. Patients were identified through their unique National Health Index (NHI) number. A spreadsheet was used to enter data from the two patient data systems. For each patient their age, gender, ethnicity, domicile (Hamilton or other), hospital visited and number of visits to ED were recorded from the hospital records. Diagnosis and date registered with the Waikato DHB palliative care service were identified from the palliative care register. Ethical approval was granted by the local Health and Disability Ethics Committee (Northern Y Ethics NTY 11/11/111). The key outcome of interest in our analyses was a visit to the ED. The attendance rate was analysed by age in the groups; 20-49 years, 50-59, 60-69, 70-79 and 80+, gender and ethnicity (categorised as Mori, Pacific or other). We also looked at attendance rate by diagnosis this was categorised into 9 groups: breast cancer, colorectal cancer, gynaecological cancer (cervical, uterine, ovarian etc), lung cancer, skin cancer, upper GI cancers, urogenital cancers (prostate, bladder, renal etc) and other cancers (e.g. lymphoma, sarcoma etc). A single category of non-cancer was included for diseases such as chronic obstructive pulmonary disease, congestive heart failure and neurological disorders. We undertook univariate analyses of attendance at ED by age category, gender, ethnicity and illness cause. We then undertook a multivariate analysis looking at first attendance at ED adjusted for age category, gender, ethnicity and illness cause. We also looked at reasons for attendance at ED and the outcome of the visit e.g. admission to hospital or discharge to the community. These outcomes were adjusted for age and ethnicity. Data were analysed using Microsoft Excel (Microsoft Corp, 2007) and STATA v11.2 (StataCorp, 2011) software. Results 1185 palliative care patients residing in the Waikato DHB catchment area were identified from the palliative care register from 1 November 2010 to 31 October 2011. There were 645 men (54.4%) and 540 women. 197 Mori (16.6%), 18 Pacific (1.5%) and 973 (81.9%) other /non-Mori were identified. The age ranged from 20-101 years (interquartile range 63-81). The mean age overall was 70.8 years: non-Mori 72.5 years, Mori 63.3 years and Pacific 64.1 years. The mean length of time on the register during the year of interest was 120.7 (median 66) days: women 129 days (67.5), men 113 days (66), Mori 123 days (62), Pacific Peoples 68 days (31) and for non-Mori 121 days (69). 449 (37.9%) of the study population visited ED at least once during the study period. The distribution of visits to ED is presented in Figure 1. The percentage of Mori who visited ED at least once had a similar likelihood of attendance as non-Mori (38.6% vs 37.4% respectively) and was not statistically different (Chi-squared=0.048, p=0.827). Figure 1. Distribution of visits to the ED The maximum number of presentations by any one patient was 12. Of the 449 patients who visited the ED at least once, 202 patients (45%) went on to have more than one visit. This was similar for Mori and non-Mori (15.2% vs 17.4%) and was not statistically different (Chi-squared=1.19, p<0.552). 38% of visits to ED were within a month of a previous visit to ED. This was also similar between Mori and non-Mori. When looking at the diagnoses responsible for patients referral to palliative care categorised into 9 major groups (Table 1), it was established that there was an excess of lung cancer cases for Mori compared with non-Mori (27.9% cf 17.3%) whilst urogenital cancers (principally prostate cancer) were more common in non-Mori (8.4% cf 4.6% Mori). Table 1. Illness categories for Mori and non-Mori palliative care patients Illness Mori Non-Mori Non-cancer 35 (17.8%) 182 (18.4%) Breast cancer 12 (6.1%) 60 (6.1%) Colorectal 21 (10.7%) 106 (10.8%) Gynaecological 7 (3.6%) 34 (3.4%) Lung 55 (27.9%) 171 (17.3%) Skin 0 (0%) 47 (4.8%) Upper GI 28 (14.2%) 135 (13.7%) Urogenital 9 (4.6%) 83 (8.4%) Other cancers 30 (15.2%) 170 (17.2%) All skin cancer cases on the register were for non-Mori. These are in keeping with the reported national incidences of these cancers.11 Otherwise the diagnosis between Mori and non-Mori were similar. Apart from gender-specific cancers, colorectal, lung, skin, upper GI and other cancers were more prevalent in men than women (60-70% vs 30-40%). The percentage of the study population with non-cancer causes of palliative care were also higher in men than women (Table 2). Table 2. Likelihood of palliative care patients attending ED within study period Variable Univariate analysis Multivariate analysis Gender * Female 33.7% 1.00 (reference) Male 41.4% 1.64 (1.25-2.14)** Ethnic group Other 37.4% 1.00 (reference) Maori 38.6% 0.99 (0.71-1.39) Pacific 55.6% (not included) Illness category Non Cancer 35.9% 1.00 (reference) Breast 38.9% 1.49 (0.83-2.67) Colorectal 37.8% 1.07 (0.67-1.71) Gynaecological 56.1% 3.33 (1.62-6.84)** Lung 39.8% 1.08 (0.73-1.60) Skin 40.4% 1.11 (0.58-2.15) Upper GI 35.0% 0.93 (0.61-1.44) Urogenital 37.0% 0.81 (0.48-1.38) Other Cancers 36.0% 0.93 (0.62-1.40) Age group 20-49yrs 37.8% 1.00 (reference) 50-59yrs 38.9% 1.00 (0.58-1.75) 60-69yrs 40.4% 1.09 (0.66-1.80) 70-79yrs 41.3% 1.15 (0.71-1.86) 80+yrs 32.1% 0.79 (0.48-1.30) *test for homogeneity (Pr>chi2) Gender=0.007 **P>|z| </=0.001 Women with gynaecological cancer were most likely to visit ED, otherwise cause of illness did not seem to influence the likelihood of visiting ED. A multivariate analysis of likelihood of a palliative care patient visiting ED and examining key variables including age (by age groupings), gender, ethnicity (Mori, Pacific and Other) and cause of illness revealed that being male increased the likelihood of attending ED by 60% (p=0.001) and having a gynaecological cancer increased the odds of attending ED by 230% (p<0.001). No other factor including ethnicity was associated with the risk of visiting ED. Of the 449 patients who visited an ED, 78.4% attended Waikato Hospital, 169/1185, 8.69% visited Thames Hospital and 12.9% vis