Recours aux soins des hommes présentant un cancer de la prostate dans l’année précédant leur décès: une étude en population en 2015

25 décembre 2019

Auteurs : A. Tanguy-Melac, T. Lesuffleur, A. Fagot-Campagna, C. Gastaldi-Ménager, X. Rébillard, P. Tuppin
Référence : Prog Urol, 2019, 16, 29, 995-1006




 




Introduction


The end-of-life care pathway of individuals with certain cancers has been described as a period of progressive decline over several years, followed by a marked and sudden deterioration during the last weeks or months of life [1]. Early access to palliative care (PC) provides patients with a better quality of life prior to their death [2, 3]. Indicators designed to measure the intensity of end-of-life care indicate a high rate of hospitalisations, emergency department visits and intensive care admissions and chemotherapy during the last month of life, which may appear to be inappropriate [4, 5, 6, 7].


Prostate cancer (PCa) is one of the most common cancers in men with an estimated incidence in France of almost 50,000 cases and 8,100 deaths in 2018 [8]. The annual cost for national health insurance of patients managed for PCa in France in 2014 was €1,009 million, i.e. 8% of all cancer-related expenditure and 0.6% of total reimbursed expenditure [9]. The Système National des Données de Santé (SNDS) [National Health Data System], which is being progressively deployed in France, includes, for each individual, information about reimbursed hospital and outpatient health care consumption, including hospital palliative care (HPC), as well as vital status and cause of death [10, 11].


The objective of this study, based on SNDS data, was to describe the characteristics and comorbidities of men managed for PCa during the year before their death in 2015, their hospital pathway during the last year of life and the intensity of care during the last month of life, including HPC management, their cause of death and the expenditure reimbursed by French national health insurance.


Patients and methods


Data


The SNDS comprehensively collects individual outpatient data, as well as healthcare prescriptions and procedures reimbursed, but it does not provide any clinical data. Nevertheless, it includes information on the presence of long-term chronic diseases (LTD) eligible for 100% reimbursement of healthcare expenditure. All this information is linked to data concerning public and private hospital stays: short stay hospitals (SSH), rehabilitation (Rehab) and hospital at home (HaH) and a specific database indicating whether or not the person is a resident of a skilled nursing home (SNH). Hospital discharge diagnoses and LTD diagnoses are coded according to the International Classification of Diseases 10th revision (ICD 10). Causes of death (ICD 10) are linked in the SNDS using an indirect matching procedure. The overall matching rate was 90% in 2015. The Caisse Nationale d'Assurance Maladie (CNAM), as a health research institute, has permanent access to the SNDS database approved by decree and the French data protection authority.


Cases


The general health scheme fund has developed a tool based on SNDS data with algorithms designed to identify beneficiaries reimbursed for chronic or expensive diseases each year [12]. Algorithms identify 56 non-exclusive groups of diseases, based on principal diagnoses, related or significant associated diagnoses in short-stay hospitals and psychiatric hospitals, LTD, dispensing of specific drugs, and specific procedures. In this tool, algorithms designed to identify people on cardiovascular prevention drugs (antihypertensives or lipid-lowering drugs) or psychotropic drugs are considered on the basis of three annual reimbursements. Cancers, including PCa, are defined by short-stay hospitalisations over a 5-year period and/or LTD status based on specific cancer diagnoses. Cases of actively treated cancer are defined as those requiring, over a 2-year period, either hospitalisation for treatment, with the exception of hospitalisations for assessment only, or hospitalisation for metastasis, or initiation of management for an LTD, or treatment with certain specific therapies indicated (androgen deprivation therapy, chemotherapy). The remaining cases were defined as PCa under surveillance (follow-up, watchful waiting, palliative care, etc.). Actively treated cancers were included in preference to cancers under surveillance.


Population


Of the 66 million inhabitants in France at the end of 2015, the general scheme covers about 77% of the population living in France [10]. This retrospective observational study concerned beneficiaries with PCa treated in 2014 or 2015 and who died in 2015.


Variables


In France, SSH have acute wards and palliative care facilities and specific beds; rehab hospitals are devoted to rehabilitation as well as palliative care depending on their rehabilitation specialisation; and HaH care delivered at home by hospital teams. Palliative care is also provided at home and in SNH by ambulatory teams not attached to hospital units, but this information is not available in the SNDS. In this study, HPC included palliative care delivered during the year, either before death or at the time of death.


Antineoplastic agents consisted of drugs dispensed by a retail pharmacy and reimbursed by national health insurance, using ATC codes L01 to L04: antineoplastic, immunomodulating agents and anti-androgens (LHRH analogues or antagonists, first- or second-generation anti-androgens, oestrogen therapy).


Information on the place of death is available for deaths occurring during hospital stays or in SNH. Death at home cannot be distinguished from death outside home: in a retirement home, in public places, etc.


Statistical analysis


For all health care benefits reimbursed by French national health insurance, SNDS indicates the sums corresponding to the expenditure billed to the beneficiary and the sum reimbursed to the beneficiary. Regarding healthcare expenditures, direct healthcare costs are included. They encompass outpatient, inpatient expenditures but also sick-leaves and disability benefits. However, drugs dispensed during a hospital stay are not individually reimbursed and were consequently not included in this study except for those billed by the hospital in addition to diagnosis-related groups funding. All these expenditures are calculated from the CNAM's perspective.


Data are expressed as the mean±standard deviation (SD). The rates of at least one healthcare reimbursement during the study period were reported. Means were calculated only for those people with at least one reimbursement during the period considered. The total and mean lengths of stay during the year per patient in the same type of hospital were determined. Chi2, ANOVA, Student's t test, Wilcoxon's test and Kruskal-Wallis test were used for comparison. A Sankey diagram was used to illustrate patient flow according to the presence and the types of hospitalisation during the year or during the last 28 days before death. SAS software (version 7.11, SAS Institute Inc., Cary, NC, USA) was used for statistical analysis and R software (3.4.3.) was used for Sankey diagrams.


Results


A total of 11,193 men, with a mean age of 81 years (SD: 9.6 years) including 61% of men 80 years or older, who died in 2015 after being managed for PCa during the previous year (2014-2015), were included (Table 1). About 58% of men died in SSH, 4% died in HaH, 9% died in Rehab, 8% died in SNH and 21% died at home.


At least one other cancer, either actively treated or under surveillance during the previous five years, was identified for one-third of men (lung: 3%, colorectal: 4%, other: 26%). A lower proportion of other cancers was observed for men who died in SNH (17%). Almost two-thirds of men had been managed for cardiovascular disease, 26% for chronic respiratory disease, and 20% for neurological or degenerative disease. Higher frequencies of comorbidities were observed among the men who died in SNH, who were also older. Men managed by HPC during the year, i.e. 47%, were younger (mean age: 79 years vs 83 years) and more often presented another cancer (43%).


During the year before death, 94% of men had been hospitalised at least once, all causes combined: 94% in SSH, 28% in Rehab, 9% in HaH and 15% in SNH (Table 1). The annual mean length of stay was 80 days in the presence of HPC vs 42 days in the absence of HPC. The Figure 1 illustrates the more intense flows during the last two months of life, predominantly from home to SSH and, to a lesser extent, from SSH to Rehab. When focusing on the last 28 days, flows from home to SSH intensified before death: 56% of men were at home 28 days before death and 21% were at home at the time of death, while 24% of men were in SSH 28 days before death versus 58% at the time of death.


Figure 1
Figure 1. 

Hospital and SNH pathway during the last year and the last four weeks of life.




During the month before death (Table 2), 76% of individuals had been admitted at least once to SSH. At least one emergency department visit was identified for 43% of men. During these hospital stays, 14% of men were admitted to the intensive care unit at least once, more often between the ages of 60 and 69 years (20%). At least one chemotherapy session was administered for 7% of men during the last 30 days of life and for 4% of men during the last 14 days of life. An antineoplastic agent was dispensed by a retail pharmacy to 24% of men (18% with HPC vs 29% without HPC) during the last month of life.


The main cause of death was cancer for 63% of men (Table 3). This proportion decreased with increasing age and varied according to the place of death and the presence or absence of HPC (83% versus 46%). The second most common cause of death was cardiovascular disease for 13% of men, especially for those who died at home (18%) or in SNH (17%). The third most common cause of death (5%) was ill- defined conditions. PCa was indicated as the cause of death for 40% of men managed for PCa, with higher rates for the youngest men (52% for men<60 years). Among all men for whom PCa was indicated as the cause of death, 59% died in SSH, 8% died in SNH, 17% died at home and 61% had received HPC.


The overall cost of the last year of life was €434 million (€179M with HPC, €255M without HPC), comprising €260M for hospital expenditure, €170M for office medicine expenditure and €4M for other benefits. The mean annual expenditure per man was €38,750 (€48,601 with HPC, €30,048 without HPC) and the growth of the mean monthly expenditure accelerated during the last month of life (Figure 2).


Figure 2
Figure 2. 

Mean monthly expenditure reimbursed by national health insurance. HPC: Hospital palliative care.





Discussion


In France, compared to other countries, end of life and death are more hospital-centered, especially in short-stay hospitals, with less frequent management at home or in SNH [13, 14, 15, 16]. A similar study conducted in France on all people who died in 2013 showed that 61% of deaths occurred in hospital (SSH 51%), 13% in SNH and 26% at home or outside of hospital for deaths from all causes, and 77% (64%), 7% and 16%, respectively, for cancer deaths [11]. In the present study, a smaller proportion of men with PCa died in hospital with a higher proportion who died at home, i.e. 71% (58%), 8% and 21%, respectively. An Australian study of men with metastatic PCa reported 77% of deaths in hospital and 23% in the community [7]. The proportion of each place of death and the mode of management may vary between countries according to the inclusion criteria (patient managed for cancer or cancer as the actual cause of death), the stage of cancer, other concomitant cancers, comorbidities, organization of end-of-life care [17, 18, 19, 20]. In the United States, an increasing number of deaths in hospices was reported and men diagnosed with prostate cancer are less likely to die from prostate cancer than from another cause [21]. Nevertheless, in France, an intense flow, corresponding to transfer of men at the end of life from their home or nursing home to SSH hospitalisation, was observed. This flow was accentuated during the last month and could be related to clinical progression of PCa (responsible for 40% of deaths), all cancers (PCa and other tumours) (67%), but also other reasons, including complications. More moderate flows to Rehab units were also observed for older men, who more often presented psychiatric or neurological comorbid conditions, but in whom the proportion of cancer deaths was higher (72%) than in SSH, and which may require HPC management.


Few published studies have described intensity of care indicators for prostate cancer during the last month of life. These indicators can also be influenced by the variations listed above. Although the Australian study [7] concerned patients with metastatic PCa, some intensity of care indicators were higher in this study: admission to an SSH during the last month of life for 48% of men (76% in our study), length of stay longer than 14 days for 55% of men (mean length of stay of 16 days in our study), emergency department visits for 7% (vs 43% in our study, but almost all hospitalisations were preceded by an emergency department visit), intensive care unit stay for 2% (vs 14% in our study), an antineoplastic agent during the 14 days before death for 12% (vs hospital chemotherapy sessions for 4% and chemotherapy dispensed by a retail pharmacy for 10% in our study). A recent study did not reveal any effect on survival and quality of life of more intensive health care consumption in men with metastatic PCa [22].


Utilization of HPC during the last year of life was identified for 47% of men treated for PCa. The indication for HPC was not reported, but cancer was indicated as the main cause of death for 83% of men managed by HPC. By exclusively considering causes of death, 61% of men with PCa were managed at least once by HPC, similar to the proportion of men managed by palliative care in the Australian study: 60% of men with metastatic PCa were managed in a hospital-based palliative care referral and 39% in a palliative care hospital bed [7]. As in other studies, patients managed by HPC were younger and presented fewer comorbidities. These studies also reported higher rates of metastatic disease and aggressive cancer among men managed by HPC [23]. Nevertheless, in our study, only 8% of men were managed by HPC prior to their end-of-life stay. This proportion was higher for those men who died at home or in SNH (16%), for whom continuation of HPC after their transfer or return home could not be detected in the SNDS. Although hospital chemotherapy sessions were slightly more frequent among men managed by HPC, retail pharmacy dispensing of antineoplastic agents was slightly less frequent. Nevertheless, analysis of the impact of HPC on intensity of care is limited by the fact that the great majority of HPC was administered during the end-of-life stay.


The active PCa inclusion algorithm in this study identified 83% of all men with PCa as the main cause of death, while the remaining men were classified as PCa under surveillance for 1% of cases, other active cancer in 5% of cases, other cancers under surveillance in 7% of cases and no PCa or other active cancer or cancer under surveillance in 4% of men. The survival and cause of death of men with PCa are markedly influenced by a combination of factors such as age, competitive comorbidities including cardiovascular disease, presence of another cancer or the characteristics of PCa (initial stage, response to treatment) [24]. In our study, 37% of deaths were due to non-neoplastic causes (cardiovascular disease in 13% of cases), 23% were due to other cancers and 40% were due to PCa. An American study analysed causes of death of patients derived from a cancer registry (1973-2012) and revealed a reduction of the proportion of patients for whom the main cause of death cause was PCa (50% in the 1970s and 20% in 2012), about 20% for other cancers and 60% for other causes (cardiovascular disease: 20%) [25]. A number of hypotheses, especially epidemiological, can be proposed to explain these differences of comorbidities between these populations, as well as the nature and the frequency of the treatments administered, such as anti-androgens [26]. The availability of effective chemotherapy, starting with docetaxel and now second- generation anti-androgens, may have contributed to this reduction. The benefits of these treatments have been demonstrated in terms of the quantity and quality of life of patients with castration-resistant metastatic PCa. Another American study, published in 2000, evaluating a cohort of deceased patients with PCa, reported PCa as the cause of death in 39% of cases, as our study, [27]. For patients with PCa, included in the Thames cancer registry between 1997-2006, 50% of deaths were due to PCa, 12% of deaths were due to other cancers, and 38% of deaths were due to other causes (cardiovascular disease: 18%) [28]. Our study reports the presence of another treated cancer or cancer under surveillance in 30% of men and another cancer was reported as the cause of death for 23% of men. An American cohort (1973-2006) of patients with PCa and another cancer reported 41% of cancer-related deaths among the men who died at the end of 2011: PCa in 19%, other cancer in 23% [29].


Few studies have specifically reported the mean cost per man treated for PCa during the last year of life, which represented €38,750 for national health insurance in the present study. In the United States, the mean cost of health care during the last year of life for men who died from PCa was lower than that for people dying from another cause ($43,572 vs. $45,830), and very similar to the cost of €38,200 identified in our study [30]. A Canadian study on the direct costs during the last 6 months of life, calculated from administrative data, revealed a cost of $16,020 per 100-day period [31]. A recent review of the literature reported lower costs for patients managed by palliative care [32], but this was not the case in our study (€48,601 with HPC, €30,048 without HPC), in which HPC was identified during the end-of-life stay and the cost difference was observed right from the first months of the last year of life. Mean expenditure disparities can thus be the result of differences in population structure. The aim of our paper is report descriptive statistics regarding healthcare expenditures and not to directly compare healthcare expenditures for individuals with and without of HPC (such a study would require the constitution of a "control" population comparable to that with HPC even if the SNDS is not necessarily the most appropriate). Variation in the definition of palliative care can also lead to different result. Nevertheless, these findings suggest more intensive management of PCa or other diseases prior to death in hospital for men managed by HPC. The prescription of second-generation endocrine therapy, recently marketed for advanced and usually metastatic castration-resistant PCa, should increase the cost of management of many patients during their last year of life.


Conclusions


This study provides a wealth of new data concerning the characteristics and management of men with PCa, regardless of the cause of death, during their last year of life in a country with predominant hospital-based management. These results must be refined for PCa with the help of healthcare professionals and patients, in order to guide and evaluate end-of-life health policies and improve monitoring and assessment of palliative care use.


Disclosure of interest


The authors declare that they have no competing interest.



Acknowledgements


The authors would like to thank the members of the Centre national de la fin de vie et des soins palliatifs , especially Dr V. Fournier and S. Bretonnière, for their participation, as well as Prof. R. Aubry and Dr Russo.




Table 1 - Description of the study population one year before death in 2015.
  Total  Place of death 
Hospital palliative care 
    Home or other  SNH  HaH  Rehab  SSH    No  Yes   
N   11,193  2,333  953  449  975  6,483    5,943  5,250   
100.0  20.9  8.5  4.0  8.7  57.9    53.1  46.9   
  P   P  
Age at inclusion (years)                      
Mean (±SD)  80.8±9.6  81.8±9.4  87.5±6.6  79.0±9.8  81.9±9.0  79.4±9.5  ***  82.5±9.2  78.8±9.6  *** 
<60  2.4  2.5  0.0  3.8  1.9  2.7  ***  1.6  3.4  *** 
60-69  12.0  9.0  1.7  14.7  9.3  14.8    9.1  15.2   
70-79  24.7  22.2  8.8  28.7  22.9  27.9    20.7  29.1   
80-89  42.7  46.0  47.2  39.0  45.7  40.7    45.5  39.7   
>90  18.2  20.3  42.3  13.8  20.2  13.9    23.1  12.6   
Comorbidities                      
Lung cancer  2.6  1.5  0.6  3.6  2.4  3.3  ***  1.5  3.9  *** 
Colorectal cancer  4.3  3.3  2.4  5.3  2.5  5.2  ***  3.3  5.5  *** 
Other cancers  26.4  18.6  14.1  29.4  25.9  30.9  ***  19.7  34.0  *** 
Cardiovascular/neurovascular  65.1  58.7  69.8  59.5  67.3  66.8  ***  68.9  60.9  *** 
Diabetes  23.1  23.7  17.4  24.3  21.9  23.9  **  23.9  22.2 
Psychiatric  6.9  6.3  13.2  5.3  8.5  6.0  ***  7.9  5.7  *** 
Neurological or degenerative  20.2  19.2  48.7  17.8  20.6  16.4  ***  23.7  16.2  *** 
Respiratory chronic  25.8  23.2  19.0  24.3  23.1  28.2  ***  26.6  24.9  *** 
Inflammatory chronic  3.0  2.5  2.2  2.7  3.3  3.4  3.1  3.0 
Rare diseases  0.6  1.0  0.4  0.7  0.4  0.5  0.6  0.5 
HIV or AIDS  0.2  0.3  0.0  0.0  0.0  0.2  0.2  0.2 
Chronic dialysis  1.2  0.8  0.5  0.7  1.0  1.5  **  1.3  1.1 
Liver or pancreas  8.1  4.0  4.3  7.3  7.4  10.3  ***  7.0  9.4  *** 
Treatments                      
Psychotropic drugs  47.0  46.8  64.6  49.2  46.5  44.4  ***  47.6  46.3  *** 
Hypnotics  18.9  18.3  23.0  18.7  18.1  18.6  **  18.9  18.8  ** 
Anxiolytics  25.8  26.0  35.6  29.2  23.8  24.4  ***  25.9  25.8 
Neuroleptics  5.4  5.4  17.5  5.6  3.5  3.9  ***  6.9  3.8  *** 
Antidepressants  21.8  20.3  38.0  22.0  20.0  20.3  ***  22.3  21.4  ** 
Antihypertensives  71.4  71.7  65.9  69.5  73.1  72.0  ***  73.7  68.8  *** 
Lipid-lowering drugs  38.3  37.3  25.3  39.2  40.1  40.2  ***  38.8  37.7  ** 
At least one stay                      
SSH  93.5  77.1  81.8  97.1  99.4  100.0  ***  88.2  99.6  *** 
Mean length of stay during the year (SD)  44.4±42.5  32.3±33.3  31.4±31.2  43.8±33.1  46.7±38.8  49.0±45.86  ***  33.6±38.8  55.2±43.3  *** 
HaH  8.8  2.1  1.9  100.0  3.6  6.6  ***  1.7  16.7  *** 
Rehab  28.2  17.3  31.9  18.0  100.0  21.5  ***  23.1  34.1  *** 
All typesa  93.9  77.8  82.9  100.0  100.0  100.0  ***  88.5  100.0  *** 
Mean length of stay during the year (SD)  60.0±64.5  34.7±47.8  46.4±58.9  99.3±79.3  99.1±71.3  62.5±63.6  ***  41.9±53.6  80.0±67.1  *** 
SNH  14.7  3.0  100.0  8.9  2.9  8.6  ***  20.1  8.6  *** 
Palliative care 46.9  16.9  15.8  92.9  69.2  55.7  ***       
Other than the end-of-life stay  7.8  16.9  15.8  3.8  4.8  4.1  ***       



Légende :
In SNH: skilled nursing home+Rehab: rehabilitation hospital+HaH: hospital at home, palliative care at home and in SNH cannot be identified. ***P <0.001, **P <0.01, *P <0.05.

[a] 
Sum of stays in SSH: short-stay hospital+Rehab+HaH.


Table 2 - Intensity of care indicators during the last 30 days before death in 2015.
  Total  Place of death 
Hospital palliative care 
Age-group 
    Home  SNH  HaH  Rehab  SSH  No  Yes  <60 years  60-69 years  70-79 years  80-89 years  ≥90 years 
N   11,193  2,333  953  449  975  6,483  5,943  5,250  271  1,341  2,762  4,785  2,034 
100.0  20.8  8.5  4.0  8.7  57.9  53.1  46.9  2.4  12.0  24.7  42.7  18.2 
 
SSH stay                            
At least one stay  76.2  35.3  31.8  59.9  66.3  100.0  65.5  88.2  83.4  86.4  82.4  74.3  64.4 
Number of staysb  1.7±1.4  1.6±1.6  1.4±1.1  1.7±1.9  1.5±1.6  1.8±1.4  1.7±1.5  1.8±1.4  2.0±1.4  2.1±1.7  1.9±1.6  1.6±1.4  1.4±0.8 
Length of stay (days±SD)b  15.8±10  11.1±9.1  13.3±9.4  14.3±9.4  16.5±9.2  16.5±10.1  12.8±9.6  18.3±9.7  16.8±10.7  17.1±10.1  17.2±10.2  15.4±9.9  13.2±9.4 
Rehab stay                            
At least one stay  15.5  4.6  6.3  5.6  100.0  8.8  10.5  21.2  12.5  11.4  15.2  16.8  16.0 
HaH stay                            
At least one stay  7.4  1.2  0.5  100.0  1.8  5.0  1.1  14.4  14.0  10.9  9.1  6.3  4.4 
Emergency department visi                          
At least one visit  43.1  17.1  19.4  25.4  33.3  58.7  42.9  43.4  32.8  40.3  40.9  44.7  45.7 
Followed by hospitalisation  42.4  16.2  18.7  24.7  32.7  58.1  42.2  42.7  32.1  39.2  40.3  44.0  45.1 
Number of visitsb  1.2±0.4  1.1±0.3c  1.1±0.2c  1.1±0.5c  1.1±0.3c  1.2±0.5c  1.2±0.4c  1.2±0.5c  1.2±0.4  1.2±0.5  1.2±0.5  1.1±0.4  1.2±0.4 
Stay for organ failure                            
At least one stay  13.8  2.7  1.8  4.0  6.1  21.5  17.0  10.3  8.1  19.9  16.2  13.8  7.5 
Antineoplastic agents a                           
During last 14 days                           
Chemotherapy session  4.2  2.6  0.5  6.2  3.2  5.3  3.7  4.7  11.4  8.7  6.9  2.5  0.5 
Antineoplastic agents  9.9  14.1  12.7  2.9  3.3  9.4  13.2  6.1  7.7  8.5  8.9  9.9  12.1 
During last 30 days                           
Chemotherapy session  7.3  4.2  0.9  8.9  5.3  9.5  5.5  9.3  21.4  17.5  11.7  3.8  0.8 
Antineoplastic agents  23.6  28.3  27.1  12.2  9.9  24.2  28.9  17.6  19.9  21.8  20.9  24.7  26.4 
Radiotherapy (session)                            
During last 14 days  2.0  0.6  0.6  2.0  1.5  2.8  1.1  3.0  6.6  3.5  3.0  1.4  0.4 
During last 30 days  3.0  1.0  1.0  3.8  3.0  3.9  1.5  4.7  7.4  5.9  4.2  2.3  0.5 



Légende :
All P -values are <0.05.

[a] 
Antineoplastic agents consisted of drugs dispensed by a retail pharmacy and reimbursed by national health insurance, using ATC codes L01 to L04: antineoplastic, immunomodulating agents and androgen deprivation (LHRH analogues or antagonists, first- or second-generation anti-androgens, oestrogen therapy).
[b] 
Among those men with at least one visit, with or without an overnight stay,
[c] 
All P -values that are not <0.05.


Table 3 - Main causes of death for men who died in 2015 and who were managed for PCa during the previous year.
  Total  Place of death 
  Hospital palliative care 
Age 
 
    Home  SNH  HaH  Rehab  SSH    No  Yes  <60  60-69  70-79  80-89  ≥90   
All causes                                
N causes (%)  10,328 (100.0)  2,030 (19.7)  885 (8.6)  388 (3.8)  923 (8.9)  6,102 (59.1)    4,891 (47.4)  5,437 (52.6)  250 (2.4)  1,242 (12.0)  2,560 (24.8)  4,421 (42.8)  1,855 (18.0)   
  P   P  
Proportion of PCa   40,1  35.6  35.6  55.1  47.5  40.2    32.7  46.8  52.0  43.3  42.8  39.2  35.0   
100,0  (17.4)  (7.6)  (5.2)  (10.6)  (59.2)    (38.6)  (61.4)  (3.1)  (13.0)  (26.4)  (41.8)  (15.7)   
Tumours  63.1  51.3  46.4  83.5  72.2  66.8  <0.001  45.6  82.5  82.8  78.5  73.0  58.9  46.3  <0.001 
Cardiovascular  13.2  18.0  17.1  5.2  9.5  12.1    19.6  6.1  1.2  6.7  9.7  15.0  19.8   
Ill-defined conditions  5.4  11.6  11.2  4.4  2.9  2.9    7.5  3.0  5.2  3.5  4.5  5.1  8.5   
Gastrointestinal system  2.2  0.9  1.0  0.3  1.5  3.0    3.3  0.9  2.0  2.3  1.7  2.4  2.3   
Respiratory system  4.5  3.5  4.4  1.0  3.6  5.1    6.8  1.8  2.0  1.9  3.0  5.2  6.7   
External causes  3.3  6.7  2.1  0.5  2.5  2.6    5.2  1.1  3.6  3.1  2.7  3.5  3.6   
Neurodegenerative  2.1  2.1  6.9  2.1  1.7  1.4    3.0  1.0  0.8  1.0  1.3  2.6  2.7   
Endocrine  1.6  2.6  2.8  1.5  1.1  1.2    2.2  1.0  0.4  0.8  1.1  2.0  2.2   
Infectious  1.0  0.3  0.7  0.5  0.5  1.3    1.5  0.4  0.0  0.7  0.7  1.1  1.2   
Genitourinary  1.3  0.6  1.1  1.0  1.0  1.7    1.7  0.9  0.0  0.6  0.8  1.6  2.2   
Other  2.3  2.4  6.3  0.0  3.5  1.9    3.6  1.3  2.0  0.9  1.5  2.6  4.5   
Distribution of tumours                                
Prostate  63.6  69.3  76.6  66  65.8  60.2  <0.001  64.4  63.0  62.8  55.1  58.6  66.5  75.6  <0.001 
Larynx, trachea, bronchus, lung  8.3  6.1  1.9  6.5  7.5  9.8    7.9  8.6  8.7  13.9  10.2  6.9  2.1   
Bladder  4.2  3.9  4.1  4.2  4.3    4.4  4.1  4.8  4.3  4.5  4.1  3.7   
Colon  3.1  3.5  2.9  3.7  2.4  3.1    3.1  3.1  1.4  2.6  3.1  3.4  3.1   
Lymphatic tissue  2.3  1.8  1.5  1.2  1.5  2.7    2.5  2.1  0.5  1.5  2.7  2.2  2.6   
Liver  2.1  1.2  2.2  1.5  2.6    1.8  2.3  1.4  2.8  2.7  2.1  0.5   
Pancreas  2.1  1.6  0.2  1.5  1.5  2.5    1.6  2.4  3.4  2.8  2.8  1.4  1.4   
rectal  1.2  0.9  0.2  0.6  1.1  1.5    1.0  1.3  0.5  1.4  1.6  0.9  1.4   
Kidney  0.6  1.5  1.2  1.5  0.9    0.9  1.0  1.0  1.0  1.1  1.0  0.6   
Oesophagus  0.8  0.2  0.6  0.8  1.1    1.0  0.9  1.0  1.4  1.2  0.8  0.3   
Stomach  0.9  0.8  0.5  0.9  1.1  1.0    0.6  1.1  1.4  1.3  0.6  1.0  0.6   
Lip, oral cavity, pharynx  0.7  0.9  0.2  0.3  0.5  0.7    0.9  0.5  2.4  1.0  1.0  0.4  0.1   
Other, not specified  12.1  11.3  9.9  13.1  13.0  12.4    12.4  12.1  15.5  14.6  12.7  11.5  9.0   



Légende :
***P <0.001, **P <0.01, *P <0.05.


References



Murray S.A., Sheikh A. Care for all at the end of life BMJ 2008 ;  336 : [958.1-959].
Bakitas M.A., Tosteson T.D., Li Z.Lyons K.D., Hull J.G., Li Z., et al. Early versus delayed initiation of concurrent palliative oncology care: patient outcomes in the ENABLE III randomized controlled trial J Clin Oncol 2015 ;  33 : 1438-1445 [cross-ref]
Triplett D.P., LeBrett W.G., Bryant A.K., Bruggeman A.R., Matsuno R.K., Hwang L., et al. Effect of palliative care on aggressiveness of end-of-life care among patients with advanced cancer J Oncol Pract 2017 ;  13 : e760-e769
Earle C.C., Neville B.A., Landrum M.B., Souza J.M., Weeks J.C., Block S.D., et al. Evaluating claims-based indicators of the intensity of end-of-life cancer care Int J Qual Health Care 2005 ;  17 : 505-509 [cross-ref]
De Roo M.L., Leemans K., Claessen S.J., Cohen J., Pasman H.R., Deliens L., et al. Quality indicators for palliative care: update of a systematic review J Pain Symptom Manage 2013 ;  46 : 556-572 [cross-ref]
Barbera L., Seow H., Sutradhar R., Chu A., Burge F., Fassbender K., et al. quality indicators of end-of-life care in patients with cancer: what rate is right? J Oncol Pract 2015 ;  11 : e279-e287
Collins A., Sundararajan V., Millar J., Burchell J., Le B., Krishnasamy M., et al. The trajectory of patients who die from metastatic prostate cancer: a population-based study BJU Int 2018 ;
Defossez G., Le Guyader-Peyrou S., Uhry Z., Grosclaude P., Remontet L., Colonna M., et al. Estimations nationales de l'incidence et de la mortalité par cancer en France métropolitaine entre 1990 et 2018  Étude à partir des registres des cancers du réseau Francim. Résultats préliminaires. Synthèse Saint-Maurice: Santé publique France (2019). 20
Tuppin P., Pestel L., Samson S., Cuerq A., Rivière S., Tala S., et al. The human and economic burden of cancer in France in 2014, based on the Sniiram national database Bull Cancer (Paris) 2017 ;  104 : 524-537 [inter-ref]
Tuppin P., Rudant J., Constantinou P., Gastaldi-Ménager C., Rachas A., de Roquefeuil L., et al. Value of a national administrative database to guide public decisions: from the Système National d'Information Interrégimes de l'Assurance Maladie (SNIIRAM) to the Système National des Données de Santé (SNDS) in France Rev Epidemiol Sante Publique 2017 ;  65 : S149-S167 [inter-ref]
Poulalhon C., Rotelli-Bihet L., Moine S., Fagot-Campagna A., Aubry R., Tuppin P. Use of hospital palliative care according to the place of death and disease one year before death in 2013: a French national observational study BMC Palliat Care 2018 ;  17 : 75
CnamTS Améliorer la qualité du système de santé et maîtriser les dépenses. Propositions de l'Assurance Maladie pour 2016. Rapport au ministre chargé de la Sécurité sociale et au Parlement sur l'évolution des charges et des produits de l'Assurance Maladie au titre de 2016 (loi du 13 août 2004)  Paris: Caisse nationale de l'Assurance maladie des travailleurs salariés (2018). 231-237[www.ladocumentationfrancaise.fr/].
Cohen J., Pivodic L., Miccinesi G., Onwuteaka-Philipsen B.D., Naylor W.A., Wilson D.M., et al. International study of the place of death of people with cancer: a population-level comparison of 14 countries across 4 continents using death certificate data Br J Cancer 2015 ;  113 : 1397-1404 [cross-ref]
Chino F., Kamal A.H., Leblanc T.W., Zafar S.Y., Suneja G., Chino J.P. Place of death for patients with cancer in the United States, 1999 through 2015: racial, age, and geographic disparities Cancer 2018 ;  124 : 4408-4419 [cross-ref]
Bekelman J.E., Halpern S.D., Blankart C.R., Bynum J.P., Cohen J., Fowler R., et al. Comparison of site of death, health care utilization, and hospital expenditures for patients dying with cancer in 7 developed countries JAMA 2016 ;  315 : 272-283 [cross-ref]
Cabañero-Martínez M.J., Nolasco A., Melchor I., Fernández-Alcántara M., Cabrero-García J. Place of death and associated factors: a population-based study using death certificate data Eur J Public Health 2019 ;
Legler A., Bradley E.H., Carlson M.D. The effect of comorbidity burden on health care utilization for patients with cancer using hospice J Palliat Med 2011 ;  14 : 751-756 [cross-ref]
Öhlén J., Cohen J., Håkanson C. Determinants in the place of death for people with different cancer types: a national population-based study Acta Oncol 2017 ;  56 : 455-461
Falchook A.D., Dusetzina S.B., Tian F., Basak R., Selvam N., Chen R.C. Aggressive end-of-life care for metastatic cancer patients younger than age 65 years J Natl Cancer Inst 2017 ;  109 (9) :
Teno J.M., Gozalo P.L., Bynum J.P., Leland N.E., Miller S.C., Morden N.E., 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 [cross-ref]
Epstein M.M., Edgren G., Rider J.R., Mucci L.A., Adami H.O. Temporal trends in cause of death among Swedish and US men with prostate cancer J Natl Cancer Inst 2012 ;  104 : 1335-1342 [cross-ref]
Golan R., Bernstein A.N., Gu X., Dinerman B.F., Sedrakyan A., Hu J.C. Increased resource use in men with metastatic prostate cancer does not result in improved survival or quality of care at the end of life Cancer 2018 ;  124 : 2212-2219 [cross-ref]
Osagiede O., Colibaseanu D.T., Spaulding A.C., Frank R.D., Merchea A., Kelley S.R., et al. Palliative care use among patients with solid cancer tumors: a national cancer database study J Palliat Care 2018 ;  33 : 149-158 [cross-ref]
Daskivich T.J., Fan K.H., Koyama T., Albertsen P.C., Goodman M., Hamilton A.S., et al. Effect of age, tumor risk, and comorbidity on competing risks for survival in a U.S. population-based cohort of men with prostate cancer Ann Intern Med 2013 ;  158 : 709-717 [cross-ref]
Zaorsky N.G., Churilla T.M., Egleston B.L., Fisher S.G., Ridge J.A., Horwitz E.M., et al. Causes of death among cancer patients Ann Oncol 2017 ;  28 : 400-407
Ziehr D.R., Chen M.H., Zhang D., Braccioforte M.H., Moran B.J., Mahal B.A., et al. Association of androgen-deprivation therapy with excess cardiac-specific mortality in men with prostate cancer BJU Int 2015 ;  116 : 358-365 [cross-ref]
Newschaffer C.J., Otani K., McDonald M.K., Penberthy L.T. Causes of death in elderly prostate cancer patients and in a comparison nonprostate cancer cohort J Natl Cancer Inst 2000 ;  92 : 613-621 [cross-ref]
Chowdhury S., Robinson D., Cahill D., Rodriguez-Vida A., Holmberg L., Møller H. Causes of death in men with prostate cancer: an analysis of 50,000 men from the Thames Cancer Registry BJU Int 2013 ;  112 : 182-189 [cross-ref]
Dinh K.T., Mahal B.A., Ziehr D.R., Muralidhar V., Chen Y.W., Viswanathan V.B., et al. Risk of prostate cancer mortality in men with a history of prior cancer BJU Int 2016 ;  117 : E20-E28
Dinan M.A., Li Y., Zhang Y., Stewart S.B., Curtis L.H., George D.J., et al. Resource use in the last year of life among patients who died with versus of prostate cancer Clin Genitourin Cancer 2016 ;  14 : 28-37
Krahn M.D., Zagorski B., Laporte A., Alibhai S.M., Bremner K.E., Tomlinson G., et al. Healthcare costs associated with prostate cancer: estimates from a population-based study BJU Int 2010 ;  105 : 338-346 [cross-ref]
Dalal S., Bruera E. End-of-life care matters: palliative cancer care results in better care and lower costs Oncologist 2017 ;  22 : 361-368 [cross-ref]






© 2019 
Elsevier Masson SAS. Tous droits réservés.