Intensity of Integrated Primary and Specialist Home-Based Palliative Care for Chronic Diseases in Northeast Italy and Its Impact on End-of-Life Hospital Access

Abstract

Background:

Hospital admissions at the end of life (EOL) represent an established indicator of poor quality of care.

Objective:

To examine the impact of intensity of integrated primary and specialist home-based palliative care for chronic diseases (HPCCD) plans of care on EOL hospital access.

Methods:

Retrospective population-based study using linked mortality, hospitalization, and home care data. Intensity of HPCCD was measured 90–31 days before death; outcomes were hospital death and prolonged hospital stay for medical reasons in the last month of life. Outcomes were modeled through Poisson and quartile regressions. Adults aged 65–84 years with at least an ordinary hospitalization and a drug treatment in the year before death, who died from nononcological chronic diseases in the Veneto Region, January 2012–December 2013, were included.

Results:

Among 2087 patients, 1016 (48.7%) did not receive any HPCCD homecare visit; 860 (41.2%), 152 (7.3%), and 59 (2.8%) had <2, 2–4, and 4–7 homecare visits/week, respectively. Hospital death occurred for 1310 patients (62.8%) and the median hospital stay in the last month of life was five days (interquartile range 0–14). In multivariate analysis, a higher intensity of HPCCD was associated with lower rates of prolonged (≥14 days) EOL hospitalization and hospital death with a dose–response relationship. When no access to HPCCD was compared with 2–4 visits/week, adjusted percentage of hospital death decreased by −18.4% (95% confidence interval [CI] −5.4% to −29.7%) and the length of hospital stay decreased by 37.9% (95% CI 16.7%–56.0%).

Conclusions:

The intensity of integrated HPCCD plans of care was associated with a reduction in EOL hospital stay and in hospital death.

Introduction

During the last decade, palliative care has been emphasized for all life-threatening diseases, strengthening the need to widen the access to palliative care for all elderly patients with chronic progressive illnesses.^1–3 Some of these diseases, such as heart failure, chronic obstructive pulmonary disease, and multiple sclerosis, have already been evaluated by experimental studies to assess the impact of palliative care programs on relevant health outcomes such as quality of life, symptoms, and end-of-life (EOL) hospitalization.⁴ In the last months of life, many chronic disease patients present with distressing symptoms such as pain, dyspnea, anorexia, mental confusion, constipation, insomnia, low mood, and problems with bladder and bowel control. This results in a high risk of experiencing repeated EOL hospitalizations when navigating a fragmented medical care system.^5,6

To avoid inappropriate health service utilization and, at the same time, improve quality of life in terminally ill patients, a multidisciplinary team approach centered in primary homecare is mandatory, with a strict coordination between generalist and specialist palliative care.^3,7 When appropriate specialist support and facilities are available, general practitioners (GPs) have already been shown to deliver effective palliative care.^8–11 However, GPs may have uneven experience, training, and professional interest in this field and, as a result, there is still limited evidence to gather an on-going debate regarding the impact of integrated primary and specialist palliative care on EOL hospital utilization and quality of care.¹²

The purpose of the present study is to evaluate the impact of intensity of a Home-based Palliative Care for Chronic Diseases (HPCCD) plan of care on place of death and length of hospital stay for medical reasons over the last 30 days of life.

Methods

Setting

In Italy, several regulatory measures have recently strengthened palliative care services.^13,14 In particular, the approach of the Veneto Region is characterized by an emphasis on the integration among health professionals involved in HPCCD: GPs, nurses, out-of-hours physicians, and palliative care physicians.^15,16 When a chronic disease patient needs to receive home-based palliative care, he/she is referred to the health district to undergo a multiprofessional and multidimensional evaluation, leading to a tailored integrated homecare plan. Less intensive integrated HPCCD plans of care are implemented by health district nurses (24 hours/7 days) and the GP under the supervision of the palliative care specialist. The GPs are generally readily available on call 12 hours/5 days with on-call physicians providing coverage for the rest of time. More complex integrated HPCCD plans of care are directly carried out by a dedicated palliative care unit (24 hours/7 days palliative care nurses and palliative care specialists readily available on call 12 hours/7 days).

Study population

This study included patients residing in the Veneto Region, aged 65–84 years, who died from nononcological chronic diseases between January 1, 2012, and December 31, 2013. The chronic diseases indicative of palliative care need were selected according to death certificates and the list proposed by the Global Atlas of Palliative Care at the End of Life.³ Nononcological chronic diseases as causes of death of patients included in the study are reported in Table 1. The following exclusion criteria were applied: (1) patients who died at 85+ years of age (as they often live in nursing homes, thus not eligible for HPCCD plan of care); (2) patients who died by accidents, tumors, acute myocardial infarction, stroke (as they do not generally show the typical EOL illness trajectory and pattern of care); (3) patients not hospitalized and not undergoing drug treatment between 365 and 91 days before death (as they were probably not affected by a severe chronic disease); and (4) patients deceased in a nursing home or hospice or discharged in a nursing home 90–31 days before death (Fig. 1).

FIG. 1.

Definition of the study population. Deaths from selected chronic diseases in patients aged 65–84 years (13 local health units of the Veneto Region, 2012–2013). HPCCD, home-based palliative care for chronic diseases.

Table 1.

Chronic Diseases Selected: ICD10 Codes, Absolute Frequencies, and Distribution by Chronic Disease Group

Chronic disease group	ICD10 codes	n	%
Chronic ischemic heart disease	I20, I25	423	34.3
Hypertensive diseases	I10–I13	169	13.7
Diabetes	E10–E14	182	14.7
Arrhythmias	I44–I45, I47–I49	63	5.1
Cardiomyopathy, heart failure	I42, I50	117	9.5
Cardiac valve disorders	I05–I09, I34–I38	137	11.1
Other heart diseases	I27, I30, I31, I51	144	11.7
Cardiovascular diseases and diabetes		1235	100
Dementia, Alzheimer's	F01, F03, G30	130	39.0
Chronic/unspecified cerebrovascular diseases	I67, I69	119	35.7
Selected degenerative neurological disorders	G10, G12, G20, G31, G35	84	25.2
Neurological diseases and dementia		333	100
COPD	J40-J44, J47	190	88.0
Other respiratory diseases	J45, J96, J98	26	12.0
Respiratory diseases		216	100
Chronic liver diseases	B18, K70, K73, K74	167	68.4
Chronic renal failure	N18	40	16.4
Other liver and kidney diseases	B16-B19, K72, K75, K76, N04, N05, N12, N13, N15, N17, N19	37	15.2
Liver and kidney diseases		244	100
Frailty markers, cutaneous, and rheumatological diseases	E46, E86, R54, R64, L^, M^	59	100

L^*, all codes of the ‘Diseases of the skin and subcutaneous tissue’ chapter in International Statistical Classification of Diseases and Releated Health Problems, 10th Revision. (ICD) M^*, all codes of the ‘Diseases of the musculoskeletal system and connective tissue' chapter in ICD-10.

According to the Italian law on privacy (DL 196/2003), the study is exempt from Ethics Committee approval and no specific written consent is needed as data were anonymized by the regional statistical office by means of a unique identifier.

The analysis was limited to 13 local health units (of 21 within the Veneto Region), which reported an accurate registration of the dates of HPCCD visits made by GPs, for a total population of about 2.9 million inhabitants. Data on healthcare utilization in the EOL of all selected decedents were retrieved from Regional Hospital Discharge records (2011–2013), homecare visit records (2012–2013), and hospice records (2013). Death in hospital was defined as a death occurring in the hospital according to Hospital Discharge records. Prolonged EOL hospital stay for medical reasons, considering only medical diagnosis-related groups, was defined as a stay exceeding the 75th percentile of stay that occurred during the last 30 days of life.

Statistical analyses

Two distinct time windows were defined: an exposure window between 90 and 31 days before death and an outcome window corresponding to the last 30 days of life. Intensity of HPCCD plans of care was classified into four categories by grouping patients receiving no homecare access, <2, 2–4, and 4–7 visits per week, the denominator being days spent at home. The association between the four HPCCD categories and the study outcomes was assessed by quartile regression (prolonged EOL hospital stay) and robust Poisson regression (death in hospital). Covariates included in the regression models were gender, marital status, Charlson Comorbidity Index (derived from hospital discharge diagnoses, four classes: 0, 1, 2, 3+), number of hospitalizations occurring 365–91 days before death (two classes: 1, 2+), number of hospitalizations that occurred 90–31 days before death (two classes: 0,1+), underlying cause of death (five classes: Cardiovascular Diseases and Diabetes [taken as reference]; Neurological Disease and Dementia; Respiratory Diseases; Liver and Kidney Diseases; and Frailty Markers, Cutaneous, Subcutaneous, and Rheumatological Diseases). All statistical analyses were performed using Stata, version 13; a value of p < 0.05 was considered statistically significant.

Results

The linkage of the three sources identified 2087 patients who died from chronic diseases in the study period. Their main characteristics are described in Table 2. The majority of patients were male, older than 75 years, and married. The Charlson Comorbidity Index was greater than zero in 83.9% of subjects; 41.3% had two or more hospitalizations between 365 and 91 days before death and 43.1% had a further hospitalization 90–31 days before death.

Table 2.

Patients Died from Chronic Disease at 65–84 Years of Age and Eligible for Home-Based Palliative Care for Chronic Diseases (13 Local Health Units of the Veneto Region: 2012–2013)

	Intensity of HPCC involvement (%)
	No homecare visits (N = 1.016)	<2 visits/week (N = 860)	2–4 visits/week (N = 152)	4–7 visits/week (N = 59)	Total (%)
Gender
Man	62.3	55.5	63.2	52.5	59.3
Woman	37.7	44.5	36.8	47.5	40.7
Age class
65–74 years old	32.7	19.9	21.7	22.0	26.3
75–84 years old	67.3	80.1	78.3	78.0	73.7
Marital status
Not married	43.7	43.5	32.2	42.4	42.7
Married	56.3	56.5	67.8	57.6	57.3
Charlson Comorbidity Index at 90 days before death
Charlson score 0	18.7	13.7	11.2	18.6	16.1
Charlson score 1	19.8	17.8	20.4	18.6	19.0
Charlson score 2	19.4	19.2	13.8	18.6	18.9
Charlson score 3	42.1	49.3	54.6	44.1	46.0
Hospitalizations 365–91 days before death
1 hospitalization 365–91 days before death	64.7	53.7	48.0	55.9	58.7
2+ hospitalizations 365–91 days before death	35.3	46.3	52.0	44.1	41.3
Hospitalizations 90–31 days before death
No hospitalization 90–31 days before death	60.6	55.7	44.1	44.1	56.9
1+ hospitalizations 90–31 days before death	39.4	44.3	55.9	55.9	43.1
Underlying cause of death
Cardiovascular diseases and diabetes	64.4	57.0	46.1	35.6	59.2
Neurological disease and dementia	10.0	19.8	27.0	33.9	16.0
Respiratory diseases	9.9	11.6	4.6	13.6	10.3
Liver and kidney diseases	13.6	8.8	13.8	15.3	11.7
Frailty markers, cutaneous, and rheumatological diseases	2.1	2.8	8.6	1.7	2.8

Almost half of the patients (n = 1016; 48.7%) were not included in any HPCCD plan of care, with remarkable differences among local health units (range 39.2%–65.4%). On average, 860 (41.2%), 152 (7.3%), and 59 (2.8%) patients received, respectively, <2, 2–4, and 4–7 homecare visits/week during the exposure period, while 130 (6.2%) received the first homecare visit 90 days or less before death. Among patients entering an HPCCD plan of care, the median number of homecare visits during the observation period was 7 (interquartile range [IQR] 3–14); visits were made mostly by nurses (72.6%), followed by GPs (21.1%). Palliative care physicians made 2.0% of the homecare visits (4.1% among those receiving 4+ visits/week during the exposure period).

The overall percentage of patients who died in hospital was 62.8% (n = 1310, range among local health units 52.9%–71.4%), while the median duration of EOL hospital stay for medical reasons was 5 days (IQR 0–14, with range of 2–7 days among local health units); 956 (45.8%) and 502 patients (24.1%) had an EOL hospital stay of 7+ and 15+ days, respectively.

Risk factors for death in hospital and prolonged EOL hospital stay (≥14 days) are shown in Table 3. After adjustment for other known factors, patients with previous or recent hospitalization had an increased risk of hospital death and prolonged EOL hospital stay for medical reasons. An increased risk was also observed for patients who died from liver and kidney disease as well as from respiratory diseases (only for hospital death).

Table 3.

Risk Factors for Death in Hospital and Prolonged (≥14 Days) End-of-Life Hospital Stay for Medical Reasons during the Last Month of Life among Patients Who Died from Chronic Disease at 65–84 Years of Age (13 Local Health Units of the Veneto Region: 2012–2013)

		Prolonged EOL hospital stay for medical reasons (days) ^a		Death in hospital (relative risk) ^b
	N	Coeff.	95% CI	RR	95% CI
Man	1237	0.00		1.00
Woman	850	−0.41	−2.34–1.51	0.99	0.92–1.06
65–74 years old	549	0.00		1.00
75–84 years old	1538	0.90	−1.14–2.93	1.03	0.96–1.11
Not married	892	0.00		1.00
Married	1195	1.41	−0.49–3.32	0.98	0.92–1.05
Charlson score 0	336	0.00		1.00
Charlson score 1	396	0.45	−2.53–3.43	1.09	0.97–1.23
Charlson score 2	394	0.38	−2.63–3.39	1.03	0.91–1.17
Charlson score 3+	961	1.34	−1.34–4.03	1.11	0.99–1.24
1 hospitalization 365–91 days before death	1225	0.00		1.00
2+ hospitalizations 365–91 days before death	862	2.93	1.08–4.78	1.11	1.04–1.19
No hospitalization 90–31 days before death	1188	0.00		1.00
1+ hospitalizations 90–31 days before death	899	8.97	7.18–10.75	1.27	1.19–1.35
No homecare visit	1016	0.00		1.00
<2 visits/week	860	−1.69	−3.58–0.20	0.96	0.90–1.03
2–4 visits/week	152	−5.93	−9.47–−2.40	0.82	0.70–0.95
4–7 visits/week	59	−6.14	−11.49–−0.79	0.82	0.65–1.02
Cardiovascular diseases and diabetes	1235	0.00		1.00
Neurological diseases and dementia	333	−0.83	−3.37–1.71	1.04	0.94–1.15
Respiratory diseases	216	0.14	−2.80–3.08	1.14	1.03–1.26
Liver and kidney diseases	244	4.21	1.38–7.04	1.26	1.16–1.37
Frailty markers, cutaneous, and rheumatological diseases	59	0.21	−5.11–5.53	1.06	0.86–1.30

Statistically significant results are reported in bold.

Quartile regression (75th percentile).

Robust multivariate analysis.

CI, confidence interval; EOL, end of life; RR, relative risk.

The relative risk (RR) of hospital death and prolonged EOL hospital stay decreased with a dose–response relationship, according to the number of homecare visits/week performed 90–31 days before death. Prolonged EOL hospital stay significantly decreased when 2+ visits/week were provided, when compared with patients with no homecare visits. The value of the 75^th percentile of the adjusted number of ordinary hospital days in the last 30 days of life dropped from 15.63 (95% confidence interval [CI] 14.37–16.90) to 9.50 days (95% CI 4.32–14.67) with a relative reduction of 39.3% (95% CI 5.5%–68.0%). A decreasing trend in hospital death was also observed, although it was significant only for the category 2–4 visits/week (−18.4%, 95% CI −5.4% to −29.7%).

Discussion

Our findings showed that almost half of the patients dying from nononcological chronic diseases received minimal or no homecare access between 90 and 31 days before death, with a wide range across local health units. About one-third of patients received <2 homecare visits/week between 90 and 31 days before death and only about one-tenth received 2+ visits/week. Hospital death and prolonged EOL hospital stay for medical reasons were reduced at increasing intensity of homecare visits, with a dose–response relationship.

Population-based data on the number of people receiving palliative care in Europe are scarce. A recent study—based on a standardized questionnaire retrospectively filled by a sentinel network of GPs in the Netherlands, Belgium, Italy, and Spain—similarly shows that 50%–65% nonsudden decedents received primary palliative care in the last three months of life, while 29%–47% received specialist palliative care, with a median duration of 15–30 days before death.¹⁷ Unlike cancer patients, who generally have an EOL trajectory characterized by a short period of evident decline, chronic patients are better described by trajectories of (1) long-term limitation with intermittent serious episodes, sometimes leading to hospital admissions (e.g., heart failure or chronic obstructive pulmonary disease), and (2) prolonged dwindling due to dementia or generalized failure of body systems.^18,19 As a result, the EOL trajectory of chronic patients in most cases is not easily predictable and patients with organ failure and dementia receive a smaller amount of primary and specialist EOL palliative care compared with cancer patients.^17,20

The population of nononcological chronic patients eligible for palliative care is quite elusive; the WPCA-WHO Global Atlas of Palliative Care at the End of Life recently proposed the following list of diseases: cardiovascular diseases, cirrhosis of the liver, endocrine, blood, and immune disorders, kidney disease, neurological disorders, protein energy malnutrition plus HIV/AIDS, tuberculosis, and other pediatric conditions.³ The presence of pain as well as of other stressful symptoms (e.g., dyspnea, anorexia, severe energy loss, mental confusion with behavioral problems, insomnia, and low mood) or of care delivery problems conducing to health needs is usually high among these patients during the last year of life.^2,21 In the present study, only decedents aged 65–84 years with a hospitalization that occurred 365–91 days before death and with drug prescriptions were chosen to better identify EOL trajectories requiring palliative care. Other approaches have been recently proposed to obtain a more precise estimate of people needing nononcological palliative care, integrating information from both underlying and contributory causes of death and previous hospitalizations; if the focus is on home palliative care, disability is of great importance as well.^19,22,23

A direct measurement of the population-based prevalence of advanced chronically ill patients with limited life prognosis in need of palliative care has also been recently implemented in Spain;²⁴ this important study has been carried out by healthcare professionals using a standardized tool across all settings of care. More prospective research is surely needed for a timely and reliable recognition of patients eligible for integrated HPCCD in the Veneto Region.

Most reviews and experimental studies about the impact of home-based palliative care on EOL hospitalization refer to cancer patients,^4,25 but some trials report on a miscellanea of cancer and noncancer patients^11,26–28 or patients affected by chronic heart failure and chronic obstructive pulmonary disease.^29–31 Experimental studies do have their limits: target populations, team compositions, models of interventions, and usual care definitions are very heterogeneous; balance, blinding, attrition, power calculation, and intention-to-treat analysis are often troublesome.²⁵ For these reasons, observational studies can contribute as well.

A relationship between the intensity of homecare visits and hospital EOL utilization has already been described in a few population-based observational studies of terminally ill patients, with shares of nononcological patients widely ranging from 20% to 75%^8,9,32–35; the timeliness of initiation of homecare visits can be influent as well.³² Unlike other observational studies, this one focuses only on nononcological decedents; patients dead from liver and kidney diseases and COPD and respiratory diseases show poor outcomes and attention is also drawn to a broad spectrum of rheumatologic diseases that can lead to severe frailty and bedsores.⁶ As a statistical issue, robust Poisson regression provides a more precise estimate of the effect size given that the event is not rare and odds ratios (ORs) estimated by logistic regression would overestimate the effect size.³⁶ Were ORs instead of RRs (through a robust Poisson regression) calculated in this study, an intensity of 2–4 homecare visits/week would be associated with an OR of death in hospital equal to 0.57 (95% CI 0.39–0.82), which is not far from the results of experimental or quasi-experimental studies.^2,25,33

Our findings suggest that some diseases (identified on the basis of the main cause of death) may be more difficult to treat at home, requiring a prolonged hospitalization at the end of life or necessitating a hospital death. COPD is frequently associated with stressful symptoms such as shortness of breath, exhaustion, and pain, being identified as a critical area also in a previous population study.³⁷ Hepatic and renal failure involves important signs and symptoms (e.g., icterus, bleeding diathesis, neurological symptoms, asthenia, and vomiting). These complex diseases request a high intensity of care, thus specific clinical skills are needed, as well as good training and better integration of care between specialists and of palliative care physicians.

This study has some strengths: it has a population basis and takes into account home visits provided by both primary and specialized healthcare professionals, within an integrated holistic view of the continuum of care.⁶ It is also, to our knowledge, one of the first population-based studies comparing EOL hospital outcomes in a terminally ill nononcological patient receiving palliative care.

The study has limitations as well: the approach is retrospective³⁸ and the definition of health problems of the target population is based on the underlying cause of death, which poorly captures both palliative care needs and trajectories of EOL disability.^18,19,23 Other relevant outcomes such as quality of life, symptoms, and caregiver outcomes have not been investigated due to the intrinsic limits of claims data.^4,12,39 The study size is not large, especially for decedents receiving 4+ visits per week. Last, we included in the analyses only 13 of 21 local health units since they provided reliable data for homecare visits performed by GPs. Our regional model of palliative care for chronic diseases cannot be generalized to other countries and is itself still in development.^15,16 The integrated palliative care coverage is not yet uniform throughout the region. The palliative care specialist visits may be to some degree underreported in some local health units. Yet, most of the time spent by palliative care specialists to supervise these chronic nononcological patients is devoted to contacts with hospital specialists, to multidimensional evaluation, and to discussion/revision of the customized palliative homecare plans with GPs and health district nurses; staff education, advice, and guidance are of great importance as well, thus leaving little time to home visits of selected patients.

The vast majority of EOL home visits is made by nurses, so their education and coordination (daily and weekly coverage, shift scheduling) are of foremost relevance. During the visit, nurses control symptoms and general conditions of patients in a standardized way; when necessary, they perform both simple and complex healthcare services (e.g., medication, infusions, catheterizations, transfusions). The GPs can further improve their recognition of nononcological patients needing palliative care, thus improving a timely integration with the palliative care units.^40,41 Of interest, the Dutch model puts a strong emphasis on GP palliative formal education and GPs represent the main providers of formal care for terminally ill cancer and noncancer patients.¹⁷ The integrated primary and specialist palliative care approach has also been recently tested in Italy by means of a pilot study involving about 100 GPs enrolled within 10 palliative care units: a range of clinical, psychosocial, and communication actions was implemented, ensuring continuity of care and proportionate treatment and avoiding an unnecessary quality-of-life deterioration in cancer and chronic patients with low life expectancy and high need of palliative care.⁴²

Our preliminary findings need to be confirmed by further research, which could take into account other important variables that may impact the quality of EOL care, such as goal of care and level of social support at home. Furthermore, more research is needed to clarify the reasons why study outcomes varied by cause of death.

Conclusions

Integrated primary and specialist palliative care can be provided to terminally ill patients with nononcological diseases in a population-based way to reduce the risk of inappropriate hospital stay and death. A multidisciplinary approach involving GPs, nurses, out-of-hours clinicians, and palliative specialists may be of great help, overcoming communication problems between settings and providers of care and actively exploring any problem that may reduce patients' quality of end of life.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

World Health Organization: WHO Definition of Palliative Care. www.who.int/cancer/palliative/definition/en (last accessed March 15, 2016).

World Health Organization: Palliative Care for older people: Better practices. www.euro.who.int/__data/assets/pdf_file/0017/143153/e95052.pdf (last accessed March 15, 2016).

Connor

, Sepulveda Bermedo

: Global Atlas of Palliative Care at the End of Life. WPCA-WHO, 2014. www.who.int/nmh/Global_Atlas_of_Palliative_Care.pdf

Gomes

, Calanzani

, Curiale

, et al.: Effectiveness and cost-effectiveness of home palliative care services for adults with advanced illness and their caregivers. Cochrane Database Syst Rev, 2013; 6:CD007760.

Addington-Hall

, Altmann

, McCarthy

: Variations by age in symptoms and dependency levels experienced by people in the last year of life, as reported by surviving family, friends and officials. Age Ageing, 1998; 27:129–136.

Teno

, Gozalo

, Bynum

, et al.: Change in end-of-life care for Medicare beneficiaries: Site of death, place of care, and health care transitions in 2000, 2005, and 2009. JAMA, 2013; 309:470–477.

Quill

, Abernethy

: Generalist plus specialist palliative care—creating a more sustainable model. N Engl J Med, 2013; 368:1173–1175.

Van den Block

, Deschepper

, Drieskens

, et al.: Hospitalisations at the end of life: Using a sentinel surveillance network to study hospital use and associated patient, disease and healthcare factors. BMC Health Serv Res, 2007; 7:69.

Pivodic

, Pardon

, Miccinesi

, et al.: Hospitalisations at the end of life in four European countries: A population-based study via epidemiological surveillance networks. J Epidemiol Community Health, 2015. pii: jech-2015-206073.

10.

Mitchell

: How well do general practitioners deliver palliative care? A systematic review. Palliat Med, 2002; 16:457–464.

11.

Brumley

, Enguidanos

, Jamison

, et al.: Increased satisfaction with care and lower costs: Results of a randomized trial of in-home palliative care. J Am Geriatr Soc, 2007; 55:993–1000.

12.

Pivodic

, Harding

, Calanzani

, et al.: Home care by general practitioners for cancer patients in the last 3 months of life: An epidemiological study of quality and associated factors. Palliat Med, 2016; 30:64–74.

13.

Legge n. 38 del 15 marzo 2010: Disposizioni per garantire l'accesso alle cure palliative e alla terapia del dolore.

14.

Intesa Stato-Regioni 25 luglio 2012 sulla definizione dei requisiti minimi e delle modalità organizzative necessari per l'accreditamento delle strutture di assistenza ai malati in fase terminale e delle unità di cure palliative e della terapia del dolore.

15.

Legge Regionale 19 marzo 2009, n. 7: Disposizioni per garantire cure palliative ai malati in stato di inguaribilità avanzata o fine vita e per sostenere la lotta al dolore.

16.

Legge Regionale 29 giugno 2012, n. 23: Norme in materia di programmazione socio-sanitaria e approvazione del Piano socio-sanitario regionale 2012–2016.

17.

Pivodic

, Pardon

, Van den Block

, et al.: Palliative care service use in four European Countries: A cross-national retrospective study via representative networks of general practitioners. PLoS One, 2013; 8:e84440.

18.

Murray

, Kendall

, Boyd K et al.: Illness trajectories and palliative care. BMJ, 2005; 330:1007–1011.

19.

Gill

, Gahbauer

, Han

, et al.: Trajectories of disability in the last year of life. N Engl J Med, 2010; 362:1173–1180.

20.

Evans

, Pasman

, Donker

, et al.: on behalf of EUROIMPACT. End-of-life care in general practice: A cross-sectional, retrospective survey of ‘cancer’, ‘organ failure’ and ‘old-age/dementia’ patients. Palliat Med, 2014; 28:965–975.

21.

Higginson

: Palliative and terminal care In: Stevens

, Raffery

, eds. Health care Needs Assessment: The epidemiologically based needs assessment reviews. New York, NY: Radcliffe Medical Press; 1997:183–260.

22.

Rosenwax

, McNamara

, Blackmore

, et al.: Estimating the size of a potential palliative care population. Palliat Med, 2005; 19:556–562.

23.

Murtagh

, Bausewein

, Verne

, Groeneveld

, et al.: How many people need palliative care? A study developing and comparing methods for population-based estimates. Palliat Med, 2014; 28:49–58.

24.

Gómez-Batiste

, Martínez-Muñoz

, Blay

, et al.: Prevalence and characteristics of patients with advanced chronic conditions in need of palliative care in the general population: A cross-sectional study. Palliat Med, 2014; 28:302–311.

25.

Davis

, Temel

, Balboni

, et al.: A review of the trials which examine early integration of outpatient and home palliative care for patients with serious illnesses. Ann Palliat Med, 2015; 4:99–121.

26.

Zimmer

, Groth-Juncker

, McCusker

: A randomized controlled study of a home health care team. Am J Public Health, 1985; 75:134–141.

27.

Grande

, Todd

, Barclay

, et al.: Does hospital at home for palliative care facilitate death at home? Randomised controlled trial. BMJ, 1999; 319:1472–1475.

28.

Grande

, Todd

, Barclay

, et al.: A randomized controlled trial of a hospital at home service for the terminally ill. Palliat Med, 2000; 14:375–385.

29.

Cummings

, Hughes

, Weaver

, et al.: Cost effectiveness of Veterans Administration hospital-based home care. A randomized clinical trial. Arch Intern Med, 1990; 150:1274–1280.

30.

Hughes

, Cummings

, Weaver

, et al.: A randomized trial of the cost effectiveness of VA hospital-based home care for the terminally ill. Health Serv Res, 1992; 26:801–817.

31.

Hughes

, Weaver

, Giobbie-Hurder

, et al.; Department of Veterans Affairs Cooperative Study Group on Home-Based Primary Care: Effectiveness of team-managed home-based primary care: A randomized multicenter trial. JAMA, 2000; 284:2877–2885.

32.

Seow

, Barbera

, Howell

, et al.: Using more end-of-life homecare services is associated with using fewer acute care services: A population-based cohort study. Med Care, 2010; 48:118–124.

33.

Seow

, Brazil

, Sussman

, Pereira

, et al.: Impact of community based, specialist palliative care teams on hospitalisations and emergency department visits late in life and hospital deaths: A pooled analysis. BMJ, 2014; 348:g3496.

34.

Gómez-Batiste

, Caja

, Espinosa

, et al.: The Catalonia WHO demonstration project for palliative care implementation: Quantitative and qualitative results at 20 years. J Pain Symptom Manage, 2012; 43:783–794.

35.

Chitnis

, Georghiou

, Steventon

, et al.: Effect of a home-based end-of-life nursing service on hospital use at the end of life and place of death: A study using administrative data and matched controls. BMJ Support Palliat Care, 2013; 3:422–430.

36.

Barros

, Hirakata

: Alternatives for logistic regression in cross-sectional studies: An empirical comparison of models that directly estimate the prevalence ratio. BMC Med Res Methodol, 2003; 3:21.

37.

Teno

, Gozalo

, Bynum

, et al.: Change in end-of-life care for Medicare beneficiaries: Site of death, place of care, and health care transitions in 2000, 2005, and 2009. JAMA, 2013; 309:470–477.

38.

Bach

, Schrag

, Begg

: Resurrecting treatment histories of dead patients: A study design that should be laid to rest. JAMA, 2004; 292:2765–2770.

39.

Steinhauser

, Christakis

, Clipp

, et al.: Factors considered important at the end of life by patients, family, physicians, and other care providers. JAMA, 2000; 284:2476–2482.

40.

Keri

: Gold Standards Framework: The GSF Prognostic Indicator Guidance. End Life C, 2010; 4:62–64.

41.

Oishi

, Murtagh

: The challenges of uncertainty and interprofessional collaboration in palliative care for non-cancer patients in the community: A systematic review of views from patients, carers and health-care professionals. Palliat Med, 2014; 28:1081–1098.

42.

AGENAS. Il progetto Teseo-Arianna. Studio Osservazionale sull'applicazione di un modello integrato di Cure Palliative Domiciliari nell'ambito della rete locale di assistenza a favore di persone con patologie evolutive con limitata aspettativa di vita. Roma, 2012. www.agenas-buonepratiche-cp.it/survey/teseoArianna.faces (last accessed May 15, 2016).