Revista Clínica Española (English Edition) Revista Clínica Española (English Edition)
Original article
Adaptation of the Palliative Prognostic Index in patients with advanced medical conditions
Recalibración del Palliative Prognostic Index en pacientes con enfermedades médicas avanzadas
M.D. Nieto Martína,??, , , M. Bernabeu Wittela, L. de la Higuera Vilaa, A. Mora Rufeteb, B. Barón Francoc, M. Ollero Baturonea, en representación de los investigadores del proyecto PALIAR
a Servicio de Medicina Interna, Hospital Universitario Virgen del Rocío, Sevilla, Spain
b Servicio de Medicina Interna, Hospital General Universitario de Elche, Elche, Alicante, Spain
c Servicio de Medicina Interna, Hospital Juan Ramón Jiménez, Huelva, Spain
Received 14 January 2013, Accepted 07 April 2013

To analyze the accuracy of the Palliative Prognostic Index (PPI) in patients with advanced medical diseases and to recalibrate it in order to adapt it to the profile of these patients.


Multicenter, prospective, observational study that included patients with one or more advanced medical diseases. Calibration (Hosmer–Lemeshow goodness of fit) and discriminative power (ROC and area under the curve [AUC]) of PPI were analyzed in the prediction of mortality at 180 days. Recalibration was carried out by analyzing the scores on the PPI of each quartile upward of dying probability. Accuracy of PPI was compared with that obtained for the Charlson index.


Overall mortality of the 1788 patients was 37.5%. Calibration in the prediction of mortality was good (goodness of fit with p=.21), the prognostic probabilities ranging from 0 to 0.25 in the first quartile of risk and from 0.48 to 0.8 in the last quartile. Discriminative power was acceptable (AUC=69; p=.0001). In recalibrated groups, mortality of patients with 0/1–2/2.5–9.5/≥10 points was 13, 23, 39 and 68%, respectively. Sensitivity (S) and negative predicative value (NPF) of the cutoff point above 0 points were 96 and 87%, respectively, while specificity (sp) and positive predictive value (PPV) of the cutoff point above 9.5 points were 95 and 68%. Calibration of the Charlson index was good (p=.2), and its discriminative power (AUC=.52; p=.06) was suboptimal.


PPI can be a useful tool in predicting 6-month survival of patients with advanced medical conditions.


Analizar el rendimiento del Palliative Prognostic Index (PPI) en los pacientes con enfermedades médicas en estadio avanzado, y recalibrarlo para adaptarlo al perfil de estos pacientes.


Estudio prospectivo observacional multicéntrico. Se incluyeron pacientes con una o más enfermedades médicas avanzadas. Se analizó la calibración (bondad de ajuste de Hosmer-Lemeshow) y el poder discriminativo (curva ROC y área bajo la curva [AUC]) del PPI en la predicción de la mortalidad a los 180 días. La recalibración se llevó a cabo analizando las puntuaciones en el PPI de cada cuartil ascendente de probabilidad de fallecer. Se comparó la precisión del PPI con la obtenida con el índice de Charlson.


La mortalidad global de los 1.788 pacientes fue del 37,5%. La calibración en la predicción de mortalidad fue buena (bondad de ajuste con p=0,21), oscilando la probabilidad pronosticada entre 0-0,25 en el primer cuartil de riesgo, y 0,48-0,8 en el último cuartil. El poder discriminativo fue aceptable (AUC=0,69; p<0,0001). En los grupos recalibrados, la mortalidad de los pacientes con 0/1-2/2,5-9,510 puntos fue del 13, 23, 39 y 68%, respectivamente. La sensibilidad y el valor predictivo negativo del punto de corte de la escala por encima de 0 fueron 96 y 87%, respectivamente; la especificidad y el valor predictivo positivo del punto de corte de la escala por encima de 9,5 fueron del 95 y 68%. La calibración del índice de Charslon fue buena (p=0,2), y el poder discriminativo subóptimo (AUC=0,52; p=0,06).


El PPI en los pacientes con enfermedades médicas en estadio avanzado puede ser de utilidad para el pronóstico de supervivencia a 6 meses.

Palliative Prognostic Index, Advanced chronic diseases, Multi-condition patients, Prognostic score
Palabras clave
Palliative Prognostic Index, Enfermedades crónicas avanzadas, Paciente pluripatológico, Escalas pronósticas

What we know

In everyday clinical practice, it is difficult to determine the moment when patients with advanced chronic medical diseases enter the final phase of their lives. The aim of this study was to analyze the performance of the Palliative Prognostic Index, in patients with advanced chronic medical diseases, in predicting mortality in the next 6 months and to recalibrate the index in order to adapt it to these patients.

What this article provides

After its recalibration, the Palliative Prognostic Index had good sensitivity (96%), specificity (95%) and negative predictive values (87%) for predicting death at 6 months and somewhat lesser quality in terms of the positive predictive value (68%). The use of the index could be helpful when making clinical decisions in patients with advanced chronic medical diseases.

The Editors


The progressive aging of the population and the growing number of individuals with chronic diseases represent an emerging health problem.1 According to the World Health Organization, it is expected that chronic diseases will be responsible for 73% of deaths worldwide and for 60% of the disease burden by 2020.2 Populations with chronic progressive diseases constitute a major emerging healthcare and social paradigm.1–3 However, it is difficult to determine the moment when patients with advanced chronic medical diseases enter the final phase of their lives.4 In fact, models for estimating survival at or before 6 months in patients with advanced chronic medical diseases have low positive predictive value because these diseases have unpredictable evolutions, with exacerbations that are followed by periods of clinical stability, during which the patient may recover their baseline state.4,5 This contributes to uncertainty for the patients, family and practitioners and hinders decision making and presents obstacles when planning therapy and care for these patients.

The prognostic scales for determining the proximity of the end of life for patients with cancer usually include 1 or more items specifically concerning cancer (such as the number of metastases), and these scales have not been validated for patients with advanced chronic medical diseases. In these patients, the use of the Palliative Prognostic Index (PPI), a validated quantitative scale for predicting mortality in patients with terminal cancer, may prove interesting.6–8 This prognostic index is designed to predict survival at 3 and 6 weeks by cataloging 5 clinical noncancer-specific dimensions (Table 1).9,10 Its usefulness and most appropriate cutoff points for patients with advanced chronic medical diseases are not known.

Table 1.

Palliative Prognostic Index. Dimensions and weight of each item.

Dimension  Points 
Palliative Performance Status9
10–20 points 
30–50 points  2.5 
≥60 points 
Oral intake
Severely reduced  2.5 
Moderately reduced 
Dyspnea at rest
Present  3.5 
Total score  0–15 

The aim of this study was to analyze the calibration and discriminative power of the PPI in a multicenter cohort of patients with advanced chronic medical diseases in predicting mortality in the next 6 months and to recalibrate the index score and risk strata in order to adapt it to these patients.

Patients and methods

This was a prospective, observational study in which 41 Spanish hospitals participated. The eligible population was selected from hospital areas, home hospitalization, day units and outpatient clinics. The patient enrollment period lasted from February 2009 to September 2010.

According to the previously published methodology of the PALIAR project,4 patients were enrolled with 1 or more of the following conditions:

  • (a)

    Heart failure with baseline dyspneaclass III of the New York Heart Association (NYHA).

  • (b)

    Respiratory failure with baseline dyspneaclass III of the Medical Research Council (MRC), baseline oxygen saturation below 90% or home oxygen therapy.

  • (c)

    Chronic renal failure, stage 4–5 of the National Kidney Foundation (NKF) (glomerular filtration rate below 30ml/min) or baseline serum creatinine3mg/dl).

  • (d)

    Chronic liver disease with a score on the Child–Pugh scale>7.

  • (e)

    Neurological disease with established cognitive impairment (Pfeiffer scale score with 7 or more errors, Lobo mini cognitive exam score18 points or established functional deterioration for baseline daily life activities [Barthel index score<60 points]).

The variables recorded included demographic, epidemiological, clinical and laboratory data present at the time of enrollment.4

In all cases, we determined the PPI scale score (overall score and by area) and the Charlson index. All patients received 180-day follow-up after enrollment, with mortality considered the primary dependent variable. We analyzed the sensitivity (S), specificity (sp), positive predictive value (PPV) and negative predictive value (NPV) of the PPI, using a score >4 and 6 points as the threshold and using mortality at 30, 60, 90, 120, 150 and 180 days of follow-up as the “criterion of truth”.

The accuracy of PPI in predicting mortality at 180 days was analyzed by determining its calibration using the Hosmer–Lemeshow (H–L) goodness of fit test on the predicted likelihood of death and the actual observed mortality, performed in a logistic regression model. The discriminative power was determined using receiver operating characteristic (ROC), curves and the calculation of the area under the curve (AUC).

The recalibration of the index was performed by analyzing the PPI scores for the probability of mortality with ascending quartiles. Subsequently, we constructed ascending and exclusive strata with ranges of PPI scores corresponding to that probability quartile. We did not apply other methods described for the detection of optimal cutoff points for the diagnostic tests, such as the Youden method, due to their current lack of use, given that they tend to overestimate the capacity of the models. We then constructed Kaplan–Meier curves for each risk stratum to test the differences in the survival trajectories of each stratum, analyzing this trajectory using the logrank test and the Tarone–Ware test.

Finally, we compared the accuracy (calibration, using the H–L test; discriminative power, using the ROC curves; and the calculation of the AUC) of the recalibrated PPI with that obtained with the Charlson index. All calculations were performed using the SPSS® statistical package, v.18.0.


We included a total of 1847 patients, 51% of them males, with a mean age of 78.74 years (standard deviation [SD], 10 years). The clinical data of the patients included in the study are listed in Table 2. The mean number of inclusion conditions was 1.35 per patient (SD, 0.6), with the most common being neurological diseases (814 patients; 44.1%), heart failure (718; 38.9%), respiratory failure (615; 33.3%), chronic kidney disease (225; 12.2%) and liver disease (115; 6.2%). Some 29.5% of patients (526) had 2 or more inclusion criteria. The burden of comorbidities was 4.85 diseases per patient (SD, 2.6) and included high blood pressure (1273 patients; 68.9%), atrial fibrillation (663; 35.9%), dyslipidemia (601; 32.5%), chronic obstructive pulmonary disease (560; 30.3%) and diabetes mellitus (525; 28.4%). Seventy percent of patients (1285) met the criteria for patients with multiple diseases.11 The median Charlson index score was 3. The most common symptoms were severe dyspnea (grades 3–4 of the NYHA or MRC) in 952 patients (51.5%), asthenia in 423 (23%), delirium and/or encephalopathy in 403 (22%), anorexia in 363 (20%), chronic pain in 356 (19%) (median score on the visual analog scale, 6/10), insomnia in 332 (18%) and nausea-vomiting in 80 (4.3%) and diarrhea in 41 (2.2%). The median score on the PPI scale was 4.5.

Table 2.

Clinical characteristics of the patients with advanced medical diseases included in the PALIAR study.

Characteristic  Percentage±SD/median (IQR) 
Age  78.7±10 
Male gender  51% 
Time of inclusion
Hospitalization  90% 
Office visit  7% 
Home hospitalization  3% 
Mean chronic organ failure, inclusion criterion  1.35±0.6 
Chronic neurological disease  44.1% 
Heart failure  38.9% 
Respiratory failure  33.3% 
Renal failure  12.2% 
Chronic liver disease  6.2% 
Charlson index  3 (3) 
Prevalence of multiple diseasesa  70% 
Mean number of categories for each patient with multiple diseases  2.9±
Most common comorbidities
Mean number of comorbidities per patient  4.85±2.6 
High blood pressure  68.9% 
Chronic atrial fibrillation  35.9% 
Dyslipidemia  32.5% 
COPDb  30.3% 
Diabetes  28.4% 
Most common symptoms
Severe dyspnea (3–4 of the NYHA and/or mMRC)  51.5% 
Asthenia  23% 
Delirium and/or encephalopathy  22% 
Anorexia  20% 
Chronic pain  19% 
Insomnia  18% 
Nausea-vomiting  4.3% 
Diarrhea  2.2% 
Baseline Barthel Index  40±34 
Dependence of main caregiver  78% 
No. of long-term prescription drugs  8.5±3.5 
No. of hospitalizations last year  2.1 (2) 
Asymptomatic  2% 
Symptomatic-outpatient  23% 
Symptomatic less than 50% bedside vigil  31% 
Symptomatic greater than 50% bedside vigil  22% 
Bedridden  22% 
Palliative Performance Score  45.7±20 
Palliative Performance Index  4.5 (4.5) 
Percentage of patients who meet the NHO criteria  52% 

Abbreviations: SD, standard deviation; ECOG-PS, Eastern Cooperative Oncology Group Performance Status; COPD, chronic obstructive pulmonary disease; mMRC, modified Medical Research Council; NHO, National Hospice Organization; NYHA, New York Heart Association; IQR, interquartile range.


Source: Ollero Baturone et al.11


No inclusion criteria.

Mortality in the 1788 patients who completed the follow-up (96.8% of those recruited) was 15.8%, 22.8%, 28.1%, 31.7% and 37.5% of patients at 30, 60, 90, 120, 150 and 180 days, respectively. The mortality, taking into account the diseases at inclusion, ranged from 33% (patients included with respiratory disease) to 44% (patients included with neurological disease). Some 50.3% of patients met the end-stage condition criteria of the National Hospice Organization12 for the cutoff points greater than 4 (44.4% for more than 6 points). A total of 48.2% of patients with scores greater than 4 (53.3% for more than 6 points) died. The values for S, sp, PPV and NPV for PPI, in the cutoff points at 30 and 180 days predefined by the authors of the index, are listed in Table 3.

Table 3.

Description of the sensitivity, specificity and positive and negative predictive values of the PPI, in the determination of the end of life in the following 180 days, in a cohort of patients with advanced medical diseases.

PPI cutoff points  Sensitivity  Specificity  Positive predictive value  Negative predictive value 
PPI (more than 0 points)  96(180d)–98%(30d)  12(30d)–15%(180d)  17(30d)–40%(180d)  87(180d)–97%(30d) 
PPI (more than 2 points)  90(180d)–94%(30d)  22(30d)–25%(180d)  18(30d)–42%(180d)  78(180d)–94%(30d) 
PPI (more than 4 points)  70(180d)–78%(30d)  49(30d)–54%(180d)  22(30d)–47%(180d)  75(180d)–92%(30d) 
PPI (more than 6 points)  54(180d)–66%(30d)  67(30d)–71%(180d)  27(30d)–53%(180d)  72(180d)–91%(30d) 
PPI (more than 9.5 points)  20(180d)–29%(30d)  93(30d)–95%(180d)  42(30d)–68%(180d)  66(180d)–87%(30d) 

The cutoff points are those established previously, and those obtained after recalibrating the index adapting it to the recruited sample. The range of values obtained in the 6 time points analyzed (30, 60, 90, 120, 150 and 180 days) is given for each dimension.

d: days; PPI: Palliative Prognostic Index.

We analyzed the calibration for the prediction of mortality and found that it was good (goodness of fit of H–L with p=.037), with the predicted probability ranging from 0 to 0.25 for the first quartile of risk and 0.48 to 0.8 for the last quartile. The mortality in the patient group with scores of 0 was 13% (mean survival, 167 days; SD, 2.7); for those with scores 1–2, the mortality was 23% (mean survival, 157 days; SD, 4); for those with scores 2.5–9.5, the mortality was 39% (mean survival, 135 days; SD, 1.9); and for those with scores above 9.5 points, the mortality was 68% (mean survival, 83; SD, 5; p<.0001 for the 2 tests performed). The values for S, sp, PPV and NPV for the cutoff points of the scale above 0 and for 9.5 points are listed in Table 3. The discriminative power of the PPI was acceptable in the overall cohort (AUC, 0.69; confidence interval at 95% [95% CI] 0.665–0.717); p<.0001). The calibration of the Charlson index was good (goodness of fit of the model with p=.2), and the discriminative power was poor (AUC, 0.52; 95% CI 0.49–0.55; p=.06).


The PPI showed good calibration, as well as acceptable discriminative power when applied to patients with advanced chronic medical diseases recruited in the present study. With the recalibration of the index, we obtained 4 clearly differentiated risk strata for mortality at 6 months (ranging from 13% in patients with scores of 0–68% in those with scores of 10 or more points). The sensitivity (96%), specificity (95%) and negative predictive (87% for the cutoff for 0 points) and positive predictive (68% for the cutoff for ≥10 points) values that the PPI achieved in predicting mortality at 180 days were also acceptable.

Currently, the most widely used end-stage patient criteria for patients with advanced chronic medical diseases are the criteria created in 1996 by the US National Hospice Organization.12 These are based on 3 general criteria combining clinical and healthcare aspects, patient preferences and specific criteria for each of the terminal organ failures. However, despite the various changes made to better apply the criteria to healthcare management scenarios other than those in the United States, their validity and reproducibility index have not been optimal. The main limitations of the criteria are their moderate positive predictive value and the subjectivity of the doctor who applies them.13–17 The organ-specific prognostic scales are useful for predicting the prognosis of specific diseases, but their power in patient populations with advanced chronic medical diseases is likely reduced as a result of the marked burden of concurrent diseases and of other comorbidities presented by these patients. In our study, 70% of the patients had multiple diseases, and 30% met 2 or more of the criteria for advanced-stage chronic medical disease. Therefore, it seems reasonable to optimize the instruments used for assessing the prognosis in the final stages of life for these patients by generating new, specifically designed indices or by adapting some of the existing instruments. The adaptation and validation of available indices and scales may contribute to improving the treatment of patients with advanced chronic medical diseases, allowing for palliative care to be started early and gradually, according to the subjective and objective needs of the patients and their environment, thereby improving their quality of life.15,18

This approach is even more applicable in the epidemiological framework of recent years, in which the global burden of patients with advanced chronic medical diseases is surpassing that of cancer-related diseases. In 2004, noncancer-related chronic medical diseases were the cause of death for 130,150 citizens in Spain (56% of the total deaths from chronic diseases), while cancer-related diseases were responsible for 100,244 deaths (44%).12 Also, the adaptation of indices developed for patient populations with advanced-stage cancer to patients with advanced chronic medical diseases may have a pathophysiological and clinical justification, given that the collection and intensity of symptoms and the biological, psychoemotional, functional and family repercussions in both populations may be similar.19–21 We consider that, given its high predictive power in populations with advanced cancer and its set of measurements that are not specifically oncologic and that are highly adaptable to various populations, the PPI may be an instrument to consider for validation in patients with advanced chronic medical diseases.

Most of the 5 dimensions in the PPI individually constitute powerful factors for poor prognosis. Functional impairment in performing basic activities (measured in the PPI with the Palliative Performance Status), delirium, severe dyspnea, and low oral intake have been previously related with deleterious health results.22–25 The presence of edema is the only dimension with a low prediction potential in previous studies, although it is also the least weighted by the PPI (1 of 15 possible points).26,27

When compared with the Charlson index, the PPI achieved a discriminative power that was clearly superior; therefore, for this patient profile, the PPI is probably of more use in clinical decision-making. Recently, an approximation to the Charlson index has been developed for primary care patients in our community, using easily available data from our information systems; however, its reproducibility in other healthcare environments is unknown.28 We believe that, based on its high NPV and good PPV in its extreme strata (patients with 0 points and those with 10 or more points), this approximation could be useful in clinical decision-making. For patients with intermediate scores (those between 1 and 9.5 points), it could also be useful together with the comprehensive clinical assessment. Future studies should explore the accuracy of the PPI in patients affected predominantly by a single advanced disease, although a high percentage (in our case 29.5%) of these patients usually have advanced failure of 2 or more organs, and many of them meet the criteria for multiple diseases (70% in our experience).

The essential limitation of the present study lies in the very nature of the PPI, which, because of its initial aim of predicting survival in patients with oncologic diseases, does not originally incorporate other important measures in patients with advanced chronic medical diseases, such as age, the number and burden of comorbidities, the presence and relationship of individual caregivers or the history of recent hospitalizations.22,29,30 We therefore do not know whether the addition of any of these parameters could have increased the discriminative ability and PPV of the PPI.

In conclusion, the PPI showed good calibration and discriminative power when applied to patients with advanced chronic medical diseases. Once recalibrated, we obtained 4 clearly differentiated strata for risk of mortality at 6 months, which may help in clinical decision making in terms of the healthcare approach for these vulnerable populations.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Annex 1
Project PALIAR researchers

M. Bernabeu-Wittel (1); J. Murcia-Zaragoza (2); M. Ollero-Baturone (1); B. Escolano Fernández (3); G. Jarava Rol (3); C. Hernández Quiles (1); M. Oliver (4); J. Díez-Manglano (5); S. Sanz Baena (6); B. Barón-Franco (7); L.M. Pérez Belmonte (8); L. Moreno-Gaviño (1); J. Galindo-Ocaña (1); P. Macías Mir (9); D. Camacho González (9); D. Nieto Martín (1); J. Praena Segovia (1); B. Massa (10); M. Maiz-Jiménez (3); N. Ramírez-Duque (1); A. Mora-Rufete (11); A. Fernández-Moyano (12); A. Alemán (13); P. Sánchez López (14); J.B. López-Sáez (15); M. Bayón Sayago (16); F. Díez (14); F. Masanés (17); C. Ramos-Cantos (8); D. Mendoza Giraldo (1); G. Ternavasio (18); P. González-Ruano (19); J.A. García García (20); R. Castillo Rubio (21); M. Loring (8); S. Gómez Lesmes (18); S. Serrano Villar (22); I. Novo Valeiro (18); L. de la Higuera (1); M.A. Soria-López (23); B. González Gisbert (24); L. Joya (25); A. Urrutia de Diego (26); H. Llorente Cancho (18); M. Rincón Gómez (1); M. Polvorosa (18); E. González Escoda (27); M. García Gutiérrez (1); M. Cassani Garza (12); L. Alvela (18); M.P. Pérez Gutiérrez (28); J.M. Machín-Lázaro (29); D. Navarro Hidalgo (30); L. Díez (31); M. Muniesa (32); M. Zubiaga (33); A. Fernández (1); L. Feliu-Mazaria (34); G. Tolchinski (35); R. Riera Hortelano (36); P. Retamar (16); P. Giner (37); M.F. Fernández-Miera (38); N. Castro Iglesias (18); A. Fuertes-Martín (18); A. Ruiz-Cantero (3); M.A. García Ordóñez (9); C. Sanromán y Terán (8); A. Martín Pérez (39); F. Formiga (40); A. López-Soto (17); C. Martínez Velasco (32) and M.A. Cuervo (41).

Hospitals participating in the study:

(1) Hospital Virgen del Rocío, Sevilla, (2) Hospital de La Vega Baja, Alicante; (3) Hospital de la Serranía, Málaga, (4) Hospital de Sanlúcar de Barrameda, Cádiz, (5) Hospital Royo Villanova, Zaragoza, (6) Hospital de la Cruz Roja San José y Santa Adela, Madrid, (7) Hospital Juan Ramón Jiménez, Huelva, (8) Hospital de la Axarquía, Málaga, (9) Hospital de Antequera, Málaga, (10) Hospital de Villajoyosa, Alicante, (11) Hospital General Universitario de Elche, Alicante, (12) Hospital San Juan de Dios del Aljarafe, Sevilla, (13) Hospital Morales Meseguer, Murcia, (14) Hospital de Torrecárdenas, Almería, (15) Hospital Universitario de Puerto Real, Cádiz, (16) Hospital Virgen Macarena, Sevilla, (17) Hospital Clínic, Barcelona, (18) Hospital Universitario de Salamanca, (19) Hospital Cantoblanco-La Paz, Madrid, (20) Hospital Nuestra Señora de Valme, Sevilla, (21) Hospital de la Malva-Rosa, Valencia, (22) Hospital Clínico San Carlos, Madrid, (23) Clínica Virgen de la Vega, Murcia, (24) Hospital Pare Jofré, Valencia, (25) Hospital de Leganés, Madrid, (26) Hospital Germans Trias i Pujol, Barcelona, (27) Hospital San Vicente Raspbeig, Alicante, (28) Hospital Clínico Universitario de Valladolid, (29) Hospital Universitario de Guadalajara, (30) Hospital Infanta Margarita, Córdoba, (31) Hospital de la Paz, Madrid, (32) Hospital San Juan de Dios de Pamplona, (33) Complejo Asistencial de Burgos, (34) Hospital General de Palma de Mallorca, (35) Hospital Municipal de Badalona, (36) Hospital San Agustín de Avilés, (37) Hospital San Cecilio, Granada, (38) Hospital Marqués de Valdecilla, Santander, (39) Hospital Sant Joan Despí, Barcelona, (40) Hospital de Bellvitge, Barcelona, and (41) Equipo de Cuidados Paliativos de Badajoz.

Unidad de Pacientes Pluripatológicos
Estándares y Recomendaciones. Ministerio de Sanidad y Política Social
Ministerio de Sanidad y Política Social, (2009)
World Health Organization (WHO)
The World Health Report 2002 – reducing risks, promoting healthy life
World Health Organization (WHO), (2002)
A.F. Jeran,R.S. Azari,T.S. Nesbitt,F.J. Meyers
The TLC model of palliative care in the elderly: preliminary application in the assisted living setting
Ann Fam Med, 2 (2004), pp. 54-60
M. Bernabeu-Wittel,A. Ruiz-Cantero,J. Murcia-Zaragoza,C. Hernandez-Quiles,B. Baron-Franco,C. Ramos-Cantos
Precisión de los criterios definitorios de pacientes con enfermedades médicas no neoplásicas en fase terminal. Proyecto PALIAR
Rev Esp Geriatr Gerontol, 45 (2010), pp. 203-212
Guía Práctica Clínica sobre cuidados Paliativos
Ministerio de sanidad y consumo. Servicio Central de Publicaciones del Gobierno Vasco, (2008)
T. Morita,J. Tsunoda,S. Inoue,S. Chihara
The Palliative Prognostic Index: a scoring system for survival prediction of terminally ill cancer patients
Support Care Cancer, 7 (1999), pp. 128-133
T. Morita,J. Tsunoda,S. Inoue,S. Chihara
Improved accuracy of physicians survival predicition for terminally ill cancer patients using Palliative prognostic Index
Palliat Med, 15 (2001), pp. 419-424
P.C. Stone,S. Lund
Predicting prognosis in patients with advanced cancer
Ann Oncol, 18 (2007), pp. 971-976
F. Anderson,G.M. Downing,J. Hill,L. Casorso,N. Lerch
Palliative Performance Scale: a new tool
J Palliat Care, 12 (1996), pp. 5-11
P. Glare,C. Sinclair,M. Downing,P. Stone,M. Maltoni,A. Vigano
Predicting survival in patients with advanced disease
Eur J Cancer, 44 (2008), pp. 1146-1156
Ollero Baturone M (coordinador), Álvarez Tello M, Barón Franco B, Bernabéu Wittel M, Codina Lanaspa A, Fernández Moyano A, et al. Atención al paciente pluripatológico: Proceso Asistencial Integrado. 2nd ed. Sevilla: Consejería de Salud; 2007. Available from: [accessed 19.05.13].
Medical guidelines for determining prognosis in selected non-cancer diseases. The National Hospice Organization
Hosp J, 11 (1996), pp. 47-63
C. Grbich,I. Maddocks,D. Parker,M. Brown,E. Willis,N. Piller
Identification of patients with noncancer diseases for palliative care services
Palliat Support Care, 3 (2005), pp. 5-14
J. Lynn
Perspectives on care at the close of life. Serving patients who may die soon and their families: the role of hospice and other services
JAMA, 285 (2001), pp. 925-932
P.A. Coventry,G.E. Grande,D.A. Richards,C.J. Todd
Prediction of appropriate timing of palliative care for older adults with non-malignant life-threatening disease: a systematic review
Age Ageing, 34 (2005), pp. 218-227
N.A. Christakis,J.J. Escarce
Survival of medicare patients after enrollment in hospice programs
N Engl J Med, 335 (1996), pp. 172-178
E. Fox,K. Landrum-McNiff,Z. Zhong,N.V. Dawson,A.W. Wu,J. Lynn
Evaluation of prognostic criteria for determining hospice eligibility in patients with advanced lung, heart, or liver disease. SUPPORT Investigators. Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments
JAMA, 282 (1999), pp. 638-645
Estrategia en Cuidados Paliativos del Sistema Nacional de Salud
Ministerio de Sanidad y Consumo; Gobierno de España, (2007)
A.F. Jerant,R.S. Azari,T.S. Nesbitt,F.J. Meyers
The TLC model of palliative care in the elderly: preliminary application in the assisted living setting
Ann Fam Med, 2 (2004), pp. 54-60
L. Kristjanson,S. Nikoletti,D. Porock,M. Smith,M. Lobchuk,P. Pedler
Congruence between patient and family caregiver perceptions of symptom distress in patients with terminal cancer
J Palliat Care, 14 (1998), pp. 24-32
D. Parker,C. Grbich,M. Brown,I. Maddocks,E. Willis,P. Roe
A palliative approach or specialist palliative care? What happens in aged care facilities for residents with a noncancer diagnosis?
J Palliat Care, 21 (2005), pp. 80-87
M. Bernabeu-Wittel,M. Ollero-Baturone,L. Moreno-Gaviño,B. Barón-Franco,A. Fuertes,J. Murcia-Zaragoza
Development of a new predictive model for polypathological patients. The PROFUND index
Eur J Intern Med, 22 (2011), pp. 311-317
L.C. Weng,H.L. Huang,D.J. Wilkie,N.A. Hoenig,M.L. Suarez,M. Marschke
Predicting survival with the Palliative Performance Scale in a minority-serving hospice and palliative care program
J Pain Symptom Manage, 37 (2009), pp. 642-648
M. Drame,N. Jovenin,J.L. Novella,P.O. Lang,D. Somme,I. Laniece
Predicting early mortality among elderly patients hospitalised in medical wards via emergency department: the SAFES cohort study
J Nutr Health Aging, 12 (2008), pp. 599-604
S.M. Figarska,H.M. Boezen,J.M. Vonk
Dyspnea severity, changes in dyspnea status and mortality in the general population: the Vlagtwedde/Vlaardingen study
Eur J Epidemiol, 27 (2012), pp. 867-876
M. Komajda,J.P. Jais,F. Reeves,B. Goldfarb,J.B. Bouhour,Y. Juillieres
Factors predicting mortality in idiopathic dilated cardiomyopathy
Eur Heart J, 11 (1990), pp. 824-831
D. Devroey,V. van Casteren
Symptoms and clinical signs associated with hospital admission and mortality for heart failure
Cent Eur J Public Health, 18 (2010), pp. 209-214
A. Gutiérrez-Misis,M. Sánchez-Santos,A. Otero
Utilización de un proxy al índice de Charlson para estudiar la asociación entre comorbilidad y mortalidad a corto y largo plazo en mayores
Aten Primaria, 44 (2012), pp. 153-161
L.C. Walter,R.J. Brand,S.R. Counsell,R.M. Palmer,S. Landefeld,R.H. Fortinsky
Development and validation of a prognostic index for 1-year mortality in older adults after hospitalization
JAMA, 285 (2001), pp. 2987-2993
S.J. Lee,K. Lindquist,M. Segal,K.E. Covinsky
Development and validation of a prognostic index for 4-year mortality in older adults

Please cite this article as: Nieto Martín MD, et al. Recalibración del Palliative Prognostic Index en pacientes con enfermedades médicas avanzadas. Rev Clin Esp. 2013;213:323–329.

List of investigators are available in Annex 1.

Corresponding author. (M.D. Nieto Martín
Copyright © 2013. Elsevier España, S.L.