TagedEn

TagedFiur TagedEn TagedFiur TagedEn

TagedEnUrologic Oncology: Seminars and Original Investigations 41 (2023) 255.e1−255.e6

Clinical-Bladder cancer
TagedH1Preliminary results regarding automated identification of patients with a
limited six-month survival prognosis using nursing assessment in
uro-oncology patientsTagedEn
TagedPCarmen Roch, MDa,*, Julian Alexander Kielkopf, MDa, Ulrich Stefenelli, MDa,
ubler, MDb, Birgitt van Oorschot, MDa, Anna Katharina Seitz, MDbTagedEn
Hubert K€
a
Interdisciplinary Center for Palliative Medicine, University Hospital W€urzburg, W€urzburg, Germany
b
Department of Urology and Pediatric Urology, University Hospital W€ urzburg, W€urzburg, Germany
Received 16 May 2022; received in revised form 14 September 2022; accepted 9 January 2023

TagedPAbstract
Introduction: Contrary to current recommendations, palliative co-management of tumor patients often occurs late in daily clinical
practice. Palliative care specialist (PCS) co-management should be considered at the latest after a 6-month prognosis has been presumed.
Therefore, identifying patients with a limited prognosis is a reasonable measure.
Methods: Patients were identified using a screening tool for limited prognosis, which combined their tumor stage and data from the nurs-
ing anamnesis. In this retrospective study, a monocentric cohort of patients with urological malignancies−UICC (Union for International
Cancer Control) stages III and IV − were enrolled from March to December 2019, with a 6-month follow-up period ending in May 2020.
Results: Most patients were male and suffered from prostate cancer. Patients with uro-oncological tumors dying within 6 months
correlated significantly with the presence of repeated hospitalizations within three months, pain on admission, malnutrition, impaired
breathing and reduced mobility (P < 0.001). The test was fair in quality (AUC 0.727) at a cut-point of five; a sensitivity of 97% and a
specificity of 25% were obtained. The PPV was 0.64 and NPV was 0.82.
Discussion/Conclusion: We specifically identified the predictors of limited prognosis in urological cancer patients across several entities
using an automated scoring system based on tumor stage and data from the nursing anamnesis. Therefore, we recognized hospitalization as
an important transition point and determined nurses to be valuable partners in identifying unmet palliative care needs without additional
technical, personnel or financial effort. Ó 2023 Elsevier Inc. All rights reserved.

TagedEnTagedPKeywords: Automated screening; Prognosis; Palliative care needs; Specialized palliative careTagedEn

TagedH11. IntroductionTagedEn relevant facilitators that help to improve timely palliative

care [5].TagedEn
TagedPNeeds-based palliative care for cancer patients is recom- TagedPRegardless of the tumor stage, cancer patients often
mended by national and international guidelines regardless describe a high symptom burden, even when undergoing
of the tumor entity or therapeutic intention [1,2]. However, tumor-specific therapy with palliative intentions [6]. Refer-
its implementation in clinical practice is still hampered by a ral to palliative care specialists (PCS) can help address
lack of awareness of surgical disciplines [3,4] and a delayed these symptoms early and capture additional unmet needs
realization of the course of disease [4]. Consensus-based [7].TagedEn
referral criteria and systematic distress screenings are TagedPAccording to current recommendations, screening for
palliative care needs should be considered no later than
when the prognosis for survival is 6 months or less [2,8].
When conducting a PCS assessment, it is important to iden-
TagedEn*Corresponding author. Tel.: +49 931-201-28849.
E-mail address: [email protected] (C. Roch).
tify these patients using previously existing prognosis tools

https://doi.org/10.1016/j.urolonc.2023.01.002
1078-1439/Ó 2023 Elsevier Inc. All rights reserved.
TagedEn255.e2 C. Roch et al. / Urologic Oncology: Seminars and Original Investigations 41 (2023) 255.e1−255.e6

[9−13]. However, using these tools for prognosis screening TagedEnTable 1

usually involves additional bureaucracy for both the patient Items of interest.
and the treatment team, as well as the question of who Items Answer Points
should perform the screening. An automated prediction of
limited prognosis using precollected patient data could at hospital stays in the last no hospital stays 0 point
three months 2 hospital stays 1 point
least allow for a preselection of patients who may benefit
>2 hospital stays 2 points
from a PCS assessment and co-management. This could be breathing impaired breathing 1 point
particularly useful when hospital admissions are viewed as mobility Independent 0 point
a key transition point, in which nurses are important part- mobile with support 1 point
ners in identifying patients with potential palliative needs immobile / mainly 2 points
bedridden
by finding out symptoms and distress.TagedEn
Pain on admission yes 1 point
TagedPUsing our uro-oncological patient collective patients Nutritional risk screening/
were required to be in UICC stage III or IV, we investigated NRS2002
whether patients with a limited prognosis could be identi- unexpected loss of weight 1 point
fied using a screening tool that looks at any relevant clinical in previous months
decreased nutrition intake 1 point
parameters collected from the routine nursing anamnesis on
in the past week
admission.TagedEn sum score 0-8 points

TagedH12. Material and methodsTagedEn TagedH22.3. Missing dataTagedEn

TagedH22.1. Study designTagedEn TagedPUpon analyzing the completeness of the screening data,
missing values regarding the number of hospitalizations in
TagedPThis was a retrospective, monocentric cohort study. All the past three months were found in 32 patients (13.2%).
adult patients who were inpatients at the Department of Regarding the data concerning NRS 2002 (decreased food
Urology at University Hospital W€ urzburg from March to intake in the past week and weight loss in the past 3
December 2019 and had one of the following UICC stage months), missing values were found in 7 patients (2.9%)
III or IV diagnoses on admission were included: prostate each. Data points for breathing, mobility, and pain at the
carcinoma (C61), renal cell carcinoma (C64), renal pelvic time of admission were all fully completed. Patients with
carcinoma (C65), and urinary bladder carcinoma (C67). A missing data were excluded from statistical evaluation
follow-up of 6-month survival ended in May 2020.TagedEn (n = 39; 16.1%). No substitution was made because there
were no significant differences in age and diagnosis, for
example, and therefore no selection bias should be assumed.TagedEn
TagedH22.2. Data recruitmentTagedEn
TagedH13. StatisticsTagedEn
T total of 726 hospitalized patients with urologic malig-
agedPA
nancies were identified via the clinical patient information T ll data collected was analyzed using IBMÒ SPSS
agedPA
system, including 292 patients with UICC stages III and IV (Statistical Package for the Social Sciences) 26.TagedEn
tumors. Of these, a complete data set was obtained for 243 TagedPFirst, a descriptive examination of the socio-demo-
patients.TagedEn graphics was carried out. We examined the difference
TagedPParameter collection from nursing assessment:TagedEn between the score results and the 6-month survival of
TagedPNursing assessment performed at University Hospital patients using a Mann-Whitney U test. To investigate the
W€ urzburg routinely collect data on hospitalizations and diagnostic quality of the score, we performed a receiver
physical capabilities on admission. These include respira- operating characteristic (ROC) analysis. Statistical signifi-
tory impairment, mobility, pain at the time of admission, cance was assumed from P ≤ 0.05.TagedEn
and nutritional status. Depending on the presence or
absence of limitations, points from 0 to 2 (0 = not applica-
TagedH14. ResultsTagedEn
ble, 1 = moderate impairment or applicable, 2 = severe
impairment) are assigned to the individual data points, TagedH24.1. Socio-demographicsTagedEn
which are added up to calculate an overall score (see
Table 1). The development of this scoring system was based TagedPMost patients were male (222/243−91.4%) and suffered
on statistically validated prognostic tools, such as the pallia- from prostate cancer (149/243−61.3%). 21/243 patients
tive performance scale [14] and, as a result, has been (8.6%) were female. The mean age was 70.1 years (SD
adapted to our routine nursing assessment. Our instrument 9.5); the range was 39 to 91 years. The most common rea-
is therefore similar but not identical and for this reason not sons for hospital admission were tumor-related symptoms
yet validated.TagedEn (70/243−28.8%) or tumor-specific therapy (117/243
 TagedEnC. Roch et al. / Urologic Oncology: Seminars and Original Investigations 41 (2023) 255.e1−255.e6 255.e3

TagedEnTable 2
Socio-demographics − n = counts, M = mean, SD = standard deviation, range in years.

Prostate carcinoma (C 61) Renal cell carcinoma (C 64) Carcinoma of the renal pelvis (C65)
n=149 n=52 or carcinoma of the urinary bladder (C67)
n=42

Age (M; SD; (Range)) 71.0; 9.2; (50-91) 69.0; 8.8; (51-86) (68.1; 11.1; 39−84)
Gender
male (%) 100 73.1 83.3
Duration of hospital stay in days 9.5; 6.8 (2−37) 10.7; 6.4 (1-27) 8.8; 8.1 (2−37)
(M; SD (Range))
Reason for admission n=149 (100 %) n=52 (100 %) n=42 (100%)
tumor-related symptoms 42 (28.2) 14 (26.9) 14 (33.3)
tumor-related diagnostic measures 2 (1.3) 7 (13.5) 2 (4.8)
tumor-specific therapy 83 (55.7) 20 (38.5) 14 (33.3)
tumor-related complications 6 (4.0) 4 (7.7) 3 (7.1)
Other 13 (8.7) 4 (7.7) 4 (9.5)
Missing data 3 (2.0) 3 (5.8) 5 (11.9)
Tumor stage UICC
III 37 (24.8) 9 (17.3) 7 (16.7)
IV 112 (75.2) 43 (82.7) 35 (83.3)
SPC co-management
Yes 9 (6.0) 5 (9.6) 6 (14.3)

TagedEn TagedFiur
−48.2%) like radiation therapy, chemotherapy or hormone
therapy. Most patients included in the study had UICC stage
III tumors (78.2%). 20/243 patients (8.2%) were admitted to
SPC, based on clinical judgment of the primary care team.
For further information, please refer to Table 2.TagedEn

TagedH24.2. Diagnostic qualityTagedEn

TagedPTo evaluate the diagnostic quality of the sum score, we

performed a ROC analysis. The score was found to be of
fair quality and had an AUC (area under the curve) value of
0.727 (95% confidence interval 0.64−0.82) (see Fig. 1).
Furthermore, a cutoff value was calculated. A sensitivity of
97% and a specificity of 25% were obtained above a thresh-
old value of 5. The positive predictive value (PPV) was
0.64 and the negative predictive value (NPV) was 0.86.TagedEn

TagedH24.3. Overall survivalTagedEn

TagedPTwo hundred thirty-four patients were still alive on the

closing date of the follow-up period. Nine patients could not
be determined as having survived or died. Fifty patients died Fig. 1. ROC analysis.TagedEn
during the observation period (range 3−173 days, Mean
70.2, Median 56, Standard deviation 50.6). One hundred TagedEnTable 3
eighty-four patients survived the 6-month observation period. Overall survival.
Table 3.TagedEn <6 months >6 months
survival n (%) survival n (%)
TagedH24.4. Correlation between 6-month survival and the sum
Prostate carcinoma (C61) 22 (9.4) 122 (52.1)
scoreTagedEn Renal cell carcinoma (C64) 14 (6.0) 34 (14.5)
Carcinoma of the renal 14 (6.0) 28 (71.8)
TagedPPatients with uro-oncological tumor disease dying within pelvis (C65) or
6 months correlates significantly with the sum score (Mann- carcinoma of the
urinary bladder (C67)
Whitney U test, P < 0.001, R = 0.311). A higher sum score
All 50 (21.4) 184 (78.6)
significantly indicates shorter survival.TagedEn
TagedEn255.e4 C. Roch et al. / Urologic Oncology: Seminars and Original Investigations 41 (2023) 255.e1−255.e6

TagedH15. DiscussionTagedEn therapy, which is associated with significantly better tolera-

bility and quality-of-life preservation [21]. In contrast, tra-
TagedPIn this retrospective study, we identified urological can- ditional chemotherapy much more frequently leads to an
cer patients with a limited prognosis across several entities acute deterioration of the condition and therapy-related hos-
using an automated scoring system based on tumor stage pitalization due to acute toxicity. In this case, patients do
and data from the nursing assessment. This approach over- not necessarily require PCS, but instead require care for
comes barriers among patients, physicians, and caregivers therapy-associated side effects. Therefore, identifying these
[15] and helps to implement palliative care in a timely man- patients is not a priority of palliative care screening, even if
ner [1,16]. According to current guideline recommenda- it is undoubtedly an important task concerning close coop-
tions, a symptom assessment can be performed by a eration between palliative care and urology.TagedEn
palliative care specialist, who can then go on to evaluate a TagedPEven though automated screening detects patients with a
patient’s needs based on the complexity of their symptoms limited prognosis, the proportion of patients without a lim-
and burdens [16]. Not every patient with a limited prognosis ited prognosis who are falsely detected is, in our opinion,
automatically has PCS co-management needs, even though acceptable in the current approach. Here, we have particu-
palliative care is at least recommended when a limited 6- larly focused on sensitivity, as patients who will die within
month prognosis is presumed [2]. Gensheimer 2019 et al. 6 months should be reliably detected. In contrast to conven-
combined over 4,000 digitally captured predictors taken tional screening methods, our screening approach does not
from a clinical information system and constructed a well- examine healthy individuals who are then referred to
functioning prediction model [17]. In terms of an automated undergo further invasive diagnostics if the screening tests
detection of a limited prognosis, this approach is very positive [22]. Morbidity and mortality can no longer be
promising; however, powerful information technology is reduced by prevention in this patient group. Instead, with
required for its implementation, which also means a power- this screening approach, a reminder could be provided so
ful financial background may be necessary. In contrast, our that palliative care is implemented in a timely manner and
approach uses existing processes without any additional patients can participate in the well-known positive effects
examinations, is data efficient, and can be implemented in of timely palliative co-management [23]. Falsely detected
everyday clinical practice without additional personnel and patients would not be harmed as they could also benefit
financial efforts. Furthermore, automated screening may from palliative co-management [18], and there is still an
prevent overly favorable physician assessments and offload ongoing discussion about the right time to integrate pallia-
the responsibility of implementing palliative care in a tive care [5,24]. Using this sum score, the number of
timely manner, which may relieve the burden of clinical patients presented to PCS may significantly increase, espe-
workload. The nursing service is valued and upgraded as cially if a positive screening result automatically prompts a
well. The patient thus benefits from a timely PCS co-man- PROM screening. This transparent patient allocation offers
agement with distressing symptoms being recorded in a clear areas of responsibility that refer to different hospital
structured manner [18].TagedEn facilities (e.g., social services) and enables patient care in a
TagedPIn our study cohort, automated capture of the 6-month timely manner [25]. Ideally, this could result in an earlier
prognosis identified a large proportion of patients as having discharge from hospital with optimized medical care [26].TagedEn
the need for a patient-reported outcome measurement
screening (PROM) and benefitting from a further PCS
assessment. Without this capture, only 26% of the patients TagedH16. LimitationsTagedEn
who died within 6 months would have been presented to
the PCS. As in other screenings, tumor stage and/or disease TagedPOur evaluation was conducted purely retrospectively on
activity were also taken into account [14,19]. The complex- a small cohort. No patients with testicular or penile carcino-
ity of treatment strategies has increased in recent years, and mas were included in the study. Therefore, our results can
well-tolerated cancer therapies are now being used, mean- only be transferred to uro-oncologic patients in general to a
ing that patients with a poor clinical status can be treated limited extent. To achieve more valid results, a prospective
more effectively, in turn making their prognosis difficult to evaluation and validation of the score using a larger number
predict [20]. However, there is evidence that certain thera- of patients should be performed.TagedEn
pies are associated with a limited prognosis because they TagedPThe sole use of an automated prognosis screening relies
are only used as the disease progresses [17]. With this in on the completeness of the nursing assessment. If elements
mind, it is also important to consider that different entities are missing, the prognosis screening cannot be calculated
are associated with different forms of therapy and different adequately and may give too low a score for these patients.
prognoses even at primary diagnosis [7]. For example, In our cohort, approximately 16% of patients could not be
patients with urothelial carcinoma often receive combina- identified using automated screening as the nursing anam-
tion chemotherapy in first-line palliative therapy, whereas nesis was incomplete. Here, education should be provided
for patients with metastatic prostate carcinoma, extended for nursing staff on the importance of obtaining a complete
hormonal therapy is predominantly used in primary nursing assessment.TagedEn
 TagedEnC. Roch et al. / Urologic Oncology: Seminars and Original Investigations 41 (2023) 255.e1−255.e6 255.e5

TagedH17. ConclusionTagedEn TagedP [3] Karlekar M, Collier B, Parish A, Olson L, Elasy T.

Utilization and determinants of palliative care in the trauma
TagedPUro-oncology patients with a limited 6-month survival intensive care unit: results of a national survey. Palliat Med
2014;28:1062–8.TagedEn
prognosis can be identified using automated screening, TagedP [4] Bergman J, Lorenz KA, Acquah-Asare S, Scales CD, Ryan G, Saigal
which may allow for the utilization of synergistic resources CS, et al. Urologist attitudes toward end-of-life care. Urology
and a timely implementation of PCS. Ultimately, not every 2013;82:48–52.TagedEn
patient with a limited survival prognosis also has SPC TagedP [5] Hui D, Heung Y, Bruera E. Timely palliative care: personalizing the
needs. Referral to SPC should only continue to be made for process of referral. Cancers (Basel) 2022;14:1047.TagedEn
TagedP [6] Bagçivan G, Seven M, Paşalak Ş, Bilmiç E, Aydın Y, Oz € G, et al.
identified patients with a combination of limited prognosis, Palliative care needs of the cancer patients receiving active ther-
patient self-assessment, and physician third-party assess- apy. Support Care Cancer 2022;30(1):749–56. https://doi.org/
ment in the form of a holistic needs assessment.TagedEn 10.1007/s00520-021-06485-2. Epub 2021 Aug 9. access 31.01.
2022.TagedEn
TagedP [7] Hugar LA, Wulff-Burchfield EM, Winzelberg GS, Jacobs BL, Davies
TagedH1Statement of ethicsTagedEn BJ. Incorporating palliative care principles to improve patient care
and quality of life in urologic oncology. Nat Rev Urol 2021;18:623–
TagedPThe investigation conforms to the principles outlined in 35.TagedEn
the Declaration of Helsinki. The study protocol was TagedP [8] Szekendi MK, Vaughn J, Lal A, Ouchi K, Williams MV. The preva-
lence of inpatients at 33 U.S. hospitals appropriate for and receiving
reviewed and the need for approval was waived by the
referral to palliative care. J Palliat Med 2016;19:360–72.TagedEn
Research Ethics Committee at University Hospital TagedP [9] Yoon SJ, Suh SY, Hui D, Choi SE, Tatara R, Watanabe H, et al.
W€urzburg (20190708 01).TagedEn Accuracy of the palliative prognostic score with or without clinicians’
prediction of survival in patients with far advanced cancer. J Pain
Symptom Manage 2021;61:1180–7.TagedEn
TagedH1Conflicts of interestTagedEn TagedP[10] Hui D, Paiva CE, Del Fabbro EG, Steer C, Naberhuis J, van de Weter-
ing M, et al. Prognostication in advanced cancer: update and directions
TagedPThe authors have nothing to disclose.TagedEn for future research. Support Care Cancer 2019;27:1973–84.TagedEn
TagedP[11] Highet G, Crawford D, Murray SA, Boyd K. Development and evalu-
ation of the Supportive and Palliative Care Indicators Tool (SPICT): a
TagedH1FundingTagedEn mixed-methods study. BMJ Support Palliat Care 2014;4:285–90.TagedEn
TagedP[12] Glare PA, Chow K. Validation of a simple screening tool for identify-
TagedPThis research did not receive any specific grant from ing unmet palliative care needs in patients with cancer. J Oncol Pract
funding agencies in the public, commercial, or not-for- 2015;11:e81–6.TagedEn
TagedP[13] Morita T, Tsunoda J, Inoue S, Chihara S. The palliative prognostic
profit sectors.TagedEn
index: a scoring system for survival prediction of terminally ill cancer
patients. Support Care Cancer 1999;7:128–33.TagedEn
TagedH1Author contributionsTagedEn TagedP[14] Mosich V, Andersag M, Watzke H. [A German version of the Palliative
Performance Scale (PPS) as a supportive structure to assess survival in
palliative patients]. Wien Med Wochenschr 2019;169:387–93.TagedEn
TagedPCR, AKS and BvO made substantial contributions to the TagedP[15] Bennardi M, Diviani N, Gamondi C, Stussi G, Saletti P, Cinesi I,
conception of the study and drafted the work. JAK, AKS et al. Palliative care utilization in oncology and hemato-oncology: a
and CR contributed to the data acquisition. CR and US for- systematic review of cognitive barriers and facilitators from the per-
mulated the statistics. All authors critically revised the spective of healthcare professionals, adult patients, and their families.
work, approved the final version for publishing, and are BMC Palliat Care 2020;19:47.TagedEn
TagedP[16] German Guideline Programme in Oncology(German Cancer Society,
accountable for all aspects of work.TagedEn German Cancer Aid, AWMF): Palliative care for patients with incur-
able cancer, long version 2.2, 2020, AWMF-registration number 128/
TagedH1Data availabilityTagedEn 001OL, https://www.leitlinienprogramm-onkologie.de/leitlinien/pal-
liativmedizin/. 2020.TagedEn
TagedP[17] Gensheimer MF, Henry AS, Wood DJ, Hastie TJ, Aggarwal S, Dud-
TagedPAll data generated or analyzed during this study is ley SA, et al. Automated survival prediction in metastatic cancer
included in this article. Further enquiries can be directed to patients using high-dimensional electronic medical record data. J
the corresponding author.TagedEn Natl Cancer Inst 2019;111:568–74.TagedEn
TagedP[18] Bergman J, Ballon-Landa E, Lorenz KA, Saucedo J, Saigal CS, Ben-
nett CJ, et al. Community-partnered collaboration to build an inte-
TagedH1ReferencesTagedEn grated palliative care clinic: the view from urology. Am J Hosp
Palliat Care 2016;33:164–70.TagedEn
TagedP [1] Osman H, Shrestha S, Temin S, Ali ZV, Corvera RA, Ddungu HD, TagedP[19] Glare PA, Eychmueller S, McMahon P. Diagnostic accuracy of the
et al. Palliative care in the global setting: ASCO resource-stratified palliative prognostic score in hospitalized patients with advanced
practice guideline. J Glob Oncol 2018;4:1–24.TagedEn cancer. J Clin Oncol 2004;22:4823–8.TagedEn
TagedP [2] Ferris FD, Bruera E, Cherny N, Cummings C, Currow D, Dudgeon D, TagedP[20] Wolska-Washer A, Robak T. Safety and tolerability of antibody-drug
et al. Palliative cancer care a decade later: accomplishments, the conjugates in cancer. Drug Saf 2019;42:295–314.TagedEn
need, next steps − from the American Society of Clinical Oncology. TagedP[21] Rush HL, Murphy L, Morgans AK, Clarke NW, Cook AD, Attard G,
J Clin Oncol 2009;27:3052–8.TagedEn et al. Quality of life in men with prostate cancer randomly allocated
TagedEn255.e6 C. Roch et al. / Urologic Oncology: Seminars and Original Investigations 41 (2023) 255.e1−255.e6

to receive docetaxel or abiraterone in the STAMPEDE trial. J Clin TagedP[24] Hui D, Hannon BL, Zimmermann C, Bruera E. Improving patient and
Oncol 2021:Jco2100728.TagedEn caregiver outcomes in oncology: team-based, timely, and targeted
TagedP[22] Grossman DC, Curry SJ, Owens DK, Bibbins-Domingo K, Caughey palliative care. CA Cancer J Clin 2018;68:356–76.TagedEn
AB, Davidson KW, et al. Screening for prostate cancer: US preven- TagedP[25] Meffert C, R€ucker G, Hatami I, Becker G. Identification of hospital
tive services task force recommendation statement. JAMA patients in need of palliative care−a predictive score. BMC Palliat
2018;319:1901–13.TagedEn Care 2016;15:21.TagedEn
agedP[T 23] Temel JS, Greer JA, Muzikansky A, Gallagher ER, Admane S, Jack- TagedP[26] Earle CC, Landrum MB, Souza JM, Neville BA, Weeks JC, Ayanian
son VA, et al. Early palliative care for patients with metastatic non- JZ. Aggressiveness of cancer care near the end of life: is it a quality-
small-cell lung cancer. N Engl J Med 2010;363:733–42.TagedEn of-care issue? J Clin Oncol 2008;26:3860–6.TagedEn