Abstract
Background
To evaluate the claim that oncologists overestimate expected survival time (EST) in advanced cancer.
Methods
We pooled 7 prospective studies in which observed survival time (OST) was compared with EST (median survival in a group of similar patients estimated at baseline by the treating oncologist). We hypothesized that EST would be well calibrated (approximately 50% of EST longer than OST) and imprecise (<30% of EST within 0.67 to 1.33 of OST), and that multiples of EST would provide well-calibrated scenarios for survival time: worst-case (approximately 10% of OST <1/4 of EST), typical (approximately 50% of OST within half to double EST), and best-case (approximately 10% of OST >3 times EST). Associations between baseline characteristics and calibration of EST were assessed.
Results
Characteristics of 1,211 patients: median age 66 years, male 61%, primary site lung (40%) and upper gastrointestinal (16%). The median OST was 8 months, and EST was 9 months. Oncologists’ estimates of EST were well calibrated (50% longer than OST) and imprecise (28% within 0.67 to 1.33 of OST). Scenarios for survival time based on simple multiples of EST were well calibrated: 8% of patients had an OST less than 1/4 their EST (worst-case), 56% had an OST within half to double their EST (typical), and 11% had an OST greater than 3 times their EST (best-case). Calibration was independent of age, sex, and cancer type.
Conclusions
Oncologists were no more likely to overestimate survival time than to underestimate it. Simple multiples of EST provide well-calibrated estimates of worst-case, typical, and best-case scenarios for survival.
It is a commonly held belief that doctors overestimate survival times for people with advanced cancer, but this is largely based on research in palliative care settings including people with very short survival times (measured in days to weeks) (1-3). However, several studies including people with longer survival times (measured in months to years) reported that oncologists’ estimates were well calibrated, that is, they were just as likely to overestimate survival as they were to underestimate it (4-6). To provide stronger evidence about the calibration of oncologists’ estimates of survival times, larger numbers of people with a range of advanced cancer types and wider ranges of survival times are needed, where the oncologists estimated patients’ expected survival times at baseline and these patients were subsequently followed up to determine their observed survival times.
The accuracy of oncologists’ estimates of survival time has been reported to range from 10% to 40%, depending on the type of estimate and the definition used for accuracy (7-9). For example, Higginson et al. asked health care professionals to estimate the minimum and maximum expected survival time (EST) for individual patients and defined an estimate as accurate if the observed survival time (OST) was within the minimum to maximum range (8). Fairchild et al. defined accuracy as an estimate within 30 days of the observed survival (7), and a number of other studies have used the definition of an estimate within 33% of the observed survival (9-13).
Studies in advanced cancer have reported that most patients wanted information about their EST, including specific scenarios, such as a longest survival with treatment, an average survival, and a shortest survival without treatment (14-16). In previous work, we showed that certain percentiles of an overall survival (OS) curve can be used to define ranges representing best-case, worst-case, and typical scenarios for survival time (4,17). For example in Figure 1, the 90th percentile, when 90% were still alive and 10% had died, could be considered the upper limit of a range for a worst-case scenario (shortest 10% of survival times), and the 10th percentile, when 10% were still alive and 90% had died, could be considered the lower limit of a range for a best-case scenario (longest 10% of survival times). The interval between the 75th and 25th percentiles (middle 50% of survival times) could be considered a range for a typical scenario. In previous work, we have also shown that simple multiples (0.25, 0.5, 2, and 3) of an OS curve’s median can be used to estimate these percentiles (4,17-20). These same simple multiples can be applied to an oncologist’s estimate of an individual patient’s EST, defined as the median survival in a group of similar patients, to provide individualized scenarios for survival time. In a recent study in which oncologists explained prognosis to patients with advanced cancer expressed as worst-case, typical, and best-case scenarios, we found that 91% of patients and 91% of family members found it helpful to receive prognostic information in this format (17). This prognostic information will be most useful to patients if it is accurate, and of limited use if oncologists consistently overestimate survival.
Figure 1.
Survival curve percentiles and their corresponding scenarios. This curve comes from an unrelated study of patients with advanced breast cancer (18).
The aim of this study was to evaluate the usefulness of oncologist estimates of survival by exploring the widely reported claim that oncologists consistently overestimate survival time in advanced cancer. We did this by determining the calibration of EST, and of scenarios based on these estimates, in 7 prospective studies including EST and OST.
Methods
We pooled data from 7 studies that included patients with advanced cancer attending oncology clinics mostly in Australia (4,13,17,21-24). These studies were selected because in each the patient’s oncologist estimated that patient’s EST at baseline (defined as the estimated median survival time in a group of similar patients), and patients were subsequently followed to determine their OST. The EST did not necessarily have to be communicated to the patient. We compared each patient’s EST with their OST. Each patient’s oncologist recorded the patient’s age, sex, Eastern Cooperative Oncology Group (ECOG) performance status, and cancer type. Study procedures followed were in accordance with the ethical standards of the Helsinki Declaration, and informed consent and approval by the health research ethics committee at all participating sites were previously obtained.
Accuracy is a widely understood but variably defined term. For the purpose of this study, we conceptualized accuracy in terms of calibration and precision. Our primary objective was to determine the calibration of oncologists’ estimates of EST, defined as the proportion of patients with an EST longer than their OST (proportion where oncologists overestimated survival). We expected oncologists’ EST to be well calibrated (ie, approximately equal proportions [50%] of estimates being longer than the OST and shorter than the OST). We also evaluated the proportion of patients with a “precise EST,” which was defined as an EST within 0.67 to 1.33 times the OST for comparability with previous studies (4). We used prior research to hypothesize that less than 30% of estimates would meet this definition (5,6,25). We explored whether the calibration and precision of oncologists’ estimates of EST varied according to the length of EST or baseline characteristics of patients (age, sex, cancer type, length of EST, and receipt of a trial intervention or standard of care). We hypothesized that there would not be important associations between these variables.
Additionally, we evaluated the calibration of scenarios for survival time on the basis of simple multiples of the EST. Using our previous findings (17,18), we hypothesized the following:
Approximately 10% of patients would live shorter than one-quarter of their oncologist’s estimate (ie, OST/EST <0.25), corresponding to a worst-case scenario.
Approximately 50% of patients would have a survival time within half to double their oncologist’s estimate (0.5 ≤ OST/EST ≤ 2), corresponding to a typical scenario.
Approximately 10% of patients would live longer than three times their oncologist’s estimate (OST/EST >3), corresponding to a best-case scenario.
Statistical analysis
Associations between oncologists’ estimates of EST and baseline patient characteristics were assessed with univariable Cox regression. For each patient, we calculated the ratio of the OST to their oncologist’s EST and used the Kaplan–Meier distribution of the ratio (OST/EST) to account for censored observations (patients still alive at their last follow-up). A 2-sided P value of less than .05 was considered statistically significant. Statistical analyses were done using R version 4.0.4.
Results
In total, there were 1,211 patients from 7 studies published between 2006 and 2022 (Table 1). The characteristics of the patients are summarized in Table 2. The median age was 66 years, and the majority (61%) were male. The median EST was 9 months (interquartile range = 6–12, absolute range = 2–96). The median OST was 8 months (interquartile range = 4–15, absolute range = 0.03–62).
Table 1.
Study characteristicsa
| Design | Cancer types | Minimum life expectancy eligibility criteria (months) | Other relevant eligibility criteria | Treatments/Interventions given during the study period | Number of patients | Median age (years) | ECOG performance status ≤1 (%) | Country | Ref |
|---|---|---|---|---|---|---|---|---|---|
| Cohort study | Mixed advanced cancers | Nil | Newly referred for noncurative intent treatment | Usual care (Chemotherapy 57%, Radiation 22%, Hormone 45%, Observation 10%, Surgery 1%) | 102 | 64 | Not available | Australia | (4) |
| Placebo-controlled phase 2 trial | Metastatic/locally recurrent gastric cancer | ≥3 | Refractory to ≤2 lines of chemotherapy | Regorafenib vs placebo | 152 | 63 | 100 | Australia, Canada, New Zealand, South Korea | (13) |
| Multi-site, phase 2 trial | Mixed advanced cancers | Nil | Patients wanting quantitative information about their prognosis | Web-based tool estimating and explaining life expectancy | 215 | 67 | 80 | Australia | (17) |
| Randomized control trial | Stage III/IV Non-small-cell lung cancer | ≥3 | Starting first line chemotherapy | Chemotherapy +/- Nitroglycerin | 363 | 64 | 93 | Australia, New Zealand | (21) |
| Randomized control trial | Mixed advanced cancers | 3–12 | Progression on ≥1 line of systemic therapy for advanced cancer | Usual care +/- advanced care planning intervention | 163 | 66 | Not available | Australia | (22) |
| Cohort study | Mixed advanced cancers | Nil | Age ≥65 years and starting first or subsequent line chemotherapy | Single or combination chemotherapy | 102 | 74 | 82 | Australia | (23) |
| Randomized control trial | Mixed advanced cancers | >3 | Baseline score ≥4/10 on scales for depression, anxiety, fatigue, or low energy from the Pt DATA form | Sertraline vs placebo | 114 | 60–69 | 85 | Australia | (24) |
ECOG = Eastern Cooperative Oncology Group performance status; Pt DATA form = Patient Disease And Treatment Assessment Form.
Table 2.
Patient characteristics (N = 1,211)
| Characteristic | No. (%) |
|---|---|
| Median age, years (range) | 66 (16-92) |
| Sex, male | 739 (61) |
| ECOG performance statusa | |
| 0 | 254 (27) |
| 1 | 576 (61) |
| 2 | 111 (12) |
| 3 | 5 (0.5) |
| Cancer type | |
| Lung | 485 (40) |
| Upper gastrointestinal | 189 (16) |
| Colorectal | 110 (9) |
| Breast | 67 (6) |
| Other | 360 (30) |
| Treatment group | |
| Experimental | 404 (33) |
| Control | 469 (39) |
| Routine clinical practice | 338 (28) |
| Estimated survival time | |
| <4 months | 79 (7) |
| 4–8 months | 334 (28) |
| >8–12 months | 280 (23) |
| >12–16 months | 327 (27) |
| >16–20 months | 76 (6) |
| >20 months | 115 (9) |
| Observed survival time | |
| <4 months | 302 (25) |
| 4–8 months | 283 (23) |
| >8–12 months | 178 (15) |
| >12–16 months | 167 (14) |
| >16–20 months | 117 (10) |
| >20 months | 164 (14) |
Not available for 265 patients. ECOG = Eastern Cooperative Oncology Group.
As hypothesized, oncologists’ estimates of EST were perfectly calibrated: 50% of patients had an EST longer than their OST (95% CI = 47% to 53%), and 28% of patients had a “precise EST” (95% CI = 25% to 31%). The calibration of oncologists’ estimates of EST did not vary greatly according to length of EST (Figure 2), except for those with EST of more than 20 months, in whom 60% of EST were longer than the OST. The precision of oncologists’ estimates of EST also did not vary greatly according to length of EST (Figure 3). There were no significant associations between the calibration of oncologists’ estimates and the baseline characteristics of patients in univariable analyses (Table 3).
Figure 2.
Percentage of patients with expected survival time longer than observed survival time (calibration) according to length of expected survival time. EST = estimated survival time; mths = months; N = number of patients; OST = observed survival time. Box represents the percentage and whiskers represent the 95% confidence intervals.
Figure 3.
Percentage of patients with expected survival time within 0.67 to 1.33 times the observed survival time (precision) according to length of expected survival time. EST = estimated survival time; mths = months; N = number of patients; OST = observed survival time. Box represents the percentage and whiskers represent the 95% confidence intervals.
Table 3.
Calibration of oncologists’ estimates according to patient baseline characteristics in univariable analyses
| Variable | Hazard ratio (95% CI) | P-value |
|---|---|---|
| Age, years | 1.00 (0.99 to 1.01) | .45 |
| Sex (ref: female) | 1.02 (0.89 to 1.17) | .78 |
| Cancer type (ref: breast) | ||
| Colorectal | 0.77 (0.52 to 1.13) | .20 |
| Lung | 0.83 (0.60 to 1.16) | |
| Pancreas | 1.13 (0.73 to 1.74) | |
| Prostate | 0.73 (0.45 to 1.18) | |
| Upper GI | 0.91 (0.64 to 1.29) | |
| Other | 0.77 (0.54 to 1.08) | |
| EST category (ref: <4 months) | ||
| 4–8 months | 1.10 (0.85 to 1.45) | .14 |
| >8–12 months | 1.24 (0.94 to 1.64) | |
| >12–16 months | 1.15 (0.87 to 1.52) | |
| >16–20 months | 0.93 (0.62 to 1.40) | |
| >20 months | 1.52 (1.05 to 2.20) | |
| Treatment arm (ref: intervention)a | 0.99 (0.85 to 1.15) | .59 |
Excludes patients in single-arm trials (N = 338).
The proportion of patients with OSTs falling within prespecified ranges for the 3 scenarios corresponded closely with our a priori hypotheses: 8% (95% CI = 7 to 10) of patients (hypothesis = 10%) lived shorter than one-quarter of their oncologist’s estimate (OST/EST <0.25) corresponding to a worst-case scenario; 56% (95% CI = 53 to 59) of patients (hypothesis = 50%) lived within half to double their oncologist’s estimate (0.5 ≤ OST/EST ≤ 2) corresponding to a typical scenario; and 11% (95% CI = 8 to 13) of patients (hypothesis = 10%) lived longer than 3 times their oncologist’s estimate (OST/EST >3) corresponding to a best-case scenario.
Discussion
This study brings together prospectively collected, individualized predictions of EST in more than 1,000 patients with advanced cancers who were rigorously followed, allowing formal comparisons with OST. We present the largest pooled analyses comparing oncologists’ estimates of EST with OST of patients with advanced cancer attending oncology outpatient clinics. Oncologists’ estimates of EST were remarkably well calibrated, with exactly 50% of estimates being longer (or shorter) than the observed survival times, that is, oncologists were equally likely to overestimate survival time as they were to underestimate it. We also found that simple multiples of the oncologists’ estimates of EST provided well-calibrated predictions of worst-case, typical, and best-case scenarios for survival, a format preferred by patients seeking prognostic information (17).
Our findings are contrary to those reported by Glare et al. in a systematic review of physicians’ estimates of survival time in 1,563 patients deemed “terminal” or referred for hospice admission. Doctors’ clinical predictions of survival in Glare’s review were generally overoptimistic, that is, ESTs were longer than OSTs (3); however, the median OST was 29 days. Many other studies supporting the dogma that oncologists’ estimates are “almost always optimistic” (26) are similarly based on studies of patients with very short survival times of usually less than 1 month (1,27,28). Our data did not show evidence of systematic overestimation or underestimation, even in those with an expected survival time of less than 4 months. However, our study included no patients with an expected survival time of less than 2 months.
As hypothesized, oncologists’ estimates were imprecise, with less than 30% of EST within 33% of the OST, similar to other studies (10-12,25). Given the inherent variability of survival time, it is unrealistic to expect point estimates of survival time to be any more precise (29). However, simple multiples of the oncologists’ estimates provided remarkably well-calibrated ranges corresponding to worst-case (approximately 10% of patients), typical (50% of patients), and best-case (approximately 10% of patients) scenarios for survival.
The oncologists in this study did not consistently overestimate survival time. The clinical implication of this is that oncologist estimates of expected survival time are useful for patients with advanced cancer. Patients require accurate prognostic information to make important decisions and plans for the future. Overly optimistic survival estimates may generate false hope and unrealistic expectations leaving patients unprepared for death.
In previous work, we surveyed more than 700 patients and found that the majority preferred to receive prognostic information formatted as worst-case, typical, and best-case scenarios for survival time rather than point estimates of the median survival time (17,30). These findings support our recommendation that oncologists formulate and explain 3 scenarios for survival when thinking and talking about prognosis in advanced cancer, which can be done using our freely available web-based tool at https://ctc.usyd.edu.au/3scenarios/.
The calibration of oncologists’ estimates was independent of the length of the EST and of the baseline characteristics of each patient. Discrepancies between estimated and observed survival time were greatest for extremes of EST (ie, the longest and shortest categories of EST), as expected.
The main strengths of this study are its large sample size and inclusion of a broad range of patients with variable survival times. The heterogeneity of cancer types included shows that oncologists’ estimates are well calibrated across a range of tumor types. Our study also assessed oncologists’ estimates of EST for patients with longer OSTs (median = 8 months) than previous studies and is therefore more representative of the prevalent overall population of people with advanced cancer, who may live for many months or even years.
The participating oncologists were mostly Australian, and all were investigators in trials or prospective studies in advanced cancer, so they may have had greater interest and expertise in estimating prognosis, limiting the general applicability of the results to all oncologists. Similarly, the patients were participants in trials or prospective studies. Four of the 7 studies had a minimum life expectancy of 3 months as an eligibility criterion, so conclusions about the accuracy of expected survival times shorter than this may be limited. We were unable to attribute individual predictions to particular oncologists, and so are unable to assess or comment on the calibration of individual oncologists. We did not have information about what participating patients wanted to know about their prognosis, nor about what they were told or what they understood about their prognosis. These questions warrant further research.
The patients in the 7 included studies were treated largely with chemotherapy or targeted therapies, and future research is needed to determine how prognosis should be formulated and explained for people with advanced cancer treated with immune checkpoint inhibitors, many of whom may have longer survival times. It is not known whether providing feedback on the accuracy of survival estimates to individual oncologists influences the accuracy of future predictions. More research is also needed to encourage and prompt oncologists to offer discussions about prognosis with their patients because we know many patients do not receive the information they desire, or receive this information too late (31).
Oncologists’ estimates of EST for people with advanced cancer were well calibrated but imprecise. Despite long-held dogma, oncologists were no more likely to overestimate survival than to underestimate it. Oncologist estimates of EST also provided well-calibrated estimates of worst-case, typical, and best-case scenarios for survival time, a format of information preferred by patients. When people with advanced cancer request quantitative information about their prognosis, we recommend that oncologists estimate the EST (defined as the median survival time in a group of similar patients) and explain 3 scenarios for survival based on simple multiples of the EST using our freely available web-based tool (https://ctc.usyd.edu.au/3scenarios/). More research is needed to help oncologists and patients initiate earlier conversations about prognosis.
Acknowledgments
We wish to thank participating patients, family members, and health care professionals in this study.
The funder did not play a role in the design of the study; the collection, analysis, and interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication.
Contributor Information
Sharon H Nahm, The NHMRC Clinical Trials Centre, The University of Sydney, Sydney, Australia; Sydney Medical School, The University of Sydney, Sydney, Australia.
Andrew J Martin, The NHMRC Clinical Trials Centre, The University of Sydney, Sydney, Australia.
Josephine M Clayton, Sydney Medical School, The University of Sydney, Sydney, Australia; The Palliative Centre, Greenwich Hospital, HammondCare, Sydney, Australia.
Peter Grimison, The University of Sydney, Sydney, Australia; Chris O’Brien Lifehouse, Sydney, Australia.
Erin B Moth, The University of Sydney, Sydney, Australia; Faculty of Medicine, Health and Human Sciences, Macquarie University, Sydney, Australia.
Nick Pavlakis, The University of Sydney, Sydney, Australia; Royal North Shore Hospital, Sydney, Australia.
Katrin Sjoquist, The NHMRC Clinical Trials Centre, The University of Sydney, Sydney, Australia; Cancer Care Centre, St George Hospital, Kogarah, Sydney, Australia.
Megan E S Smith-Uffen, Department of Medicine, McMaster University, Hamilton, ON, Canada.
Annette Tognela, Macarthur Cancer Therapy Centre, Sydney, Australia.
Anuradha Vasista, Sydney Medical School, The University of Sydney, Sydney, Australia; Nepean Cancer Care Centre, Sydney, Australia.
Martin R Stockler, The NHMRC Clinical Trials Centre, The University of Sydney, Sydney, Australia; Sydney Medical School, The University of Sydney, Sydney, Australia; Concord Cancer Centre, Sydney, Australia.
Belinda E Kiely, The NHMRC Clinical Trials Centre, The University of Sydney, Sydney, Australia; Sydney Medical School, The University of Sydney, Sydney, Australia; Macarthur Cancer Therapy Centre, Sydney, Australia; Concord Cancer Centre, Sydney, Australia.
Data availability
The data underlying this article cannot be shared due to the privacy of individuals who participated in the study. Individual deidentified data can be made available from the repository to accredited researchers who submit a proposal that is approved by the NHMRC Clinical Trials Centre.
Author contributions
Sharon Nahm (Conceptualization; Formal analysis; Investigation; Methodology; Writing—original draft; Writing—review & editing), Andrew J. Martin (Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Resources; Supervision; Writing—review & editing), Josephine M. Clayton (Data curation; Writing—review & editing), Peter Grimison (Data curation; Writing—review & editing), Erin B. Moth (Data curation; Writing—review & editing), Nick Pavlaki (Data curation; Writing—review & editing), Katrin Sjoquist (Data curation; Writing—review & editing), Megan E.S. Smith-Uffen (Data curation; Writing—review & editing), Annette Tognela (Data curation; Writing—review & editing), Anuradha Vasista (Data curation; Writing—review & editing), Martin R. Stockler (Conceptualization; Data curation; Investigation; Methodology; Supervision; Writing—review & editing), Belinda E. Kiely (Conceptualization; Data curation; Investigation; Methodology; Supervision; Writing—review & editing).
Funding
This work was supported by a National Health and Medical Research Council Clinical Trials Centre Postgraduate Research Scholarship, 2021, to author S.N.
Conflicts of interest
MS, who is a JNCI Cancer Spectrum Deputy Editor and co-author on this paper, was not involved in the editorial review or decision to publish this contribution. The authors have no relevant financial or nonfinancial interests to disclose.
References
- 1. Forster LE, Lynn J.. Predicting life span for applicants to inpatient hospice. Arch Intern Med. 1988;148(12):2540-2543. [PubMed] [Google Scholar]
- 2. Christakis NA, Lamont EB.. Extent and determinants of error in doctors' prognoses in terminally ill patients: prospective cohort study. BMJ. 2000;320(7233):469-472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Glare P, Virik K, Jones M, et al. A systematic review of physicians’ survival predictions in terminally ill cancer patients. BMJ. 2003;327(7408):195-198. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Stockler MR, Tattersall MH, Boyer MJ, et al. Disarming the guarded prognosis: predicting survival in newly referred patients with incurable cancer. Br J Cancer. 2006;94(2):208-212. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Kiely BE, Martin AJ, Tattersall MH, et al. The median informs the message: accuracy of individualized scenarios for survival time based on oncologists’ estimates. J Clin Oncol. 2013;31(28):3565-3571. [DOI] [PubMed] [Google Scholar]
- 6. Smith-Uffen MES, Johnson SB, Martin AJ, et al. Estimating survival in advanced cancer: a comparison of estimates made by oncologists and patients. Support Care Cancer. 2020;28(7):3399-3407. [DOI] [PubMed] [Google Scholar]
- 7. Fairchild A, Debenham B, Danielson B, et al. Comparative multidisciplinary prediction of survival in patients with advanced cancer. Support Care Cancer. 2014;22(3):611-617. [DOI] [PubMed] [Google Scholar]
- 8. Higginson IJ, Costantini M.. Accuracy of prognosis estimates by four palliative care teams: a prospective cohort study. BMC Palliat Care. 2002;1(1):1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Perez-Cruz PE, Dos Santos R, Silva TB, et al. Longitudinal temporal and probabilistic prediction of survival in a cohort of patients with advanced cancer. J Pain Symptom Manage. 2014;48(5):875-882. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Hui D, Kilgore K, Nguyen L, et al. The accuracy of probabilistic versus temporal clinician prediction of survival for patients with advanced cancer: a preliminary report. Oncologist. 2011;16(11):1642-1648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Lam PT. Accuracy of clinical prediction of survival in a palliative care unit. Progress in Palliative Care. 2008;16(3):113-117. [Google Scholar]
- 12. Faris M. Clinical estimation of survival and impact of other prognostic factors on terminally ill cancer patients in Oman. Support Care Cancer. 2003;11(1):30-34. [DOI] [PubMed] [Google Scholar]
- 13. Pavlakis N, Sjoquist KM, Martin AJ, et al. Regorafenib for the treatment of advanced gastric cancer (INTEGRATE): a multinational placebo-controlled phase II trial. J Clin Oncol. 2016;34(23):2728-2735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Kaplowitz SA, Campo S, Chiu WT.. Cancer patients’ desires for communication of prognosis information. Health Commun. 2002;14(2):221-241. [DOI] [PubMed] [Google Scholar]
- 15. Greisinger AJ, Lorimor RJ, Aday LA, et al. Terminally ill cancer patients: their most important concerns. Cancer Pract. 1997;5(3):147-154. [PubMed] [Google Scholar]
- 16. Hagerty RG, Butow PN, Ellis PA, et al. Cancer patient preferences for communication of prognosis in the metastatic setting. J Clin Oncol. 2004;22(9):1721-1730. [DOI] [PubMed] [Google Scholar]
- 17. Nahm SH, Stockler MR, Martin AJ, et al. Using three scenarios to explain life expectancy in advanced cancer: attitudes of patients, family members, and other healthcare professionals. Support Care Cancer. 2022;30(9):7763-7772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Kiely BE, Soon YY, Tattersall MH, et al. How long have I got? Estimating typical, best-case, and worst-case scenarios for patients starting first-line chemotherapy for metastatic breast cancer: a systematic review of recent randomized trials. J Clin Oncol. 2011;29(4):456-463. [DOI] [PubMed] [Google Scholar]
- 19. West TA, Kiely BE, Stockler MR.. Estimating scenarios for survival time in men starting systemic therapies for castration-resistant prostate cancer: a systematic review of randomised trials. Eur J Cancer. 2014;50(11):1916-1924. [DOI] [PubMed] [Google Scholar]
- 20. Kiely BE, Alam M, Blinman P, et al. Estimating typical, best-case and worst-case life expectancy scenarios for patients starting chemotherapy for advanced non-small-cell lung cancer: a systematic review of contemporary randomized trials. Lung Cancer. 2012;77(3):537-544. [DOI] [PubMed] [Google Scholar]
- 21. Davidson A, Veillard AS, Tognela A, et al. ; Australasian Lung Cancer Trials Group. A phase III randomized trial of adding topical nitroglycerin to first-line chemotherapy for advanced nonsmall-cell lung cancer: the Australasian lung cancer trials group NITRO trial. Ann Oncol. 2015;26(11):2280-2286. [DOI] [PubMed] [Google Scholar]
- 22. Johnson SB, Butow PN, Bell ML, et al. A randomised controlled trial of an advance care planning intervention for patients with incurable cancer. Br J Cancer. 2018;119(10):1182-1190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Moth EB, Blinman P, Stefanic N, et al. Estimating survival time in older adults receiving chemotherapy for advanced cancer. J Geriatr Oncol. 2020;11(4):617-625. [DOI] [PubMed] [Google Scholar]
- 24. Stockler MR, O’Connell R, Nowak AK, et al. ; Zoloft’s Effects on Symptoms and Survival Time Trial Group. Effect of sertraline on symptoms and survival in patients with advanced cancer, but without major depression: a placebo-controlled double-blind randomised trial. Lancet Oncol. 2007;8(7):603-612. [DOI] [PubMed] [Google Scholar]
- 25. Vasista A, Stockler M, Martin A, et al. Accuracy and prognostic significance of oncologists’ estimates and scenarios for survival time in advanced gastric cancer. Oncologist. 2019;24(11):e1102-e1107. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Lee SR, Smith A.. Survival estimates in advanced terminal cancer. In: Morrison RS, ed. UpToDate. Wolters Kluwer. Accessed February 20, 2022. [Google Scholar]
- 27. Oxenham D, Cornbleet MA.. Accuracy of prediction of survival by different professional groups in a hospice. Palliat Med. 1998;12(2):117-118. [DOI] [PubMed] [Google Scholar]
- 28. Maltoni M, Pirovano M, Scarpi E, et al. Prediction of survival of patients terminally ill with cancer. results of an Italian prospective multicentric study. Cancer. 1995;75(10):2613-2622. [DOI] [PubMed] [Google Scholar]
- 29. Henderson R, Jones M, Stare J.. Accuracy of point predictions in survival analysis. Stat Med. 2001;20(20):3083-3096. [DOI] [PubMed] [Google Scholar]
- 30. Kiely BE, McCaughan G, Christodoulou S, et al. Using scenarios to explain life expectancy in advanced cancer: attitudes of people with a cancer experience. Support Care Cancer. 2013;21(2):369-376. [DOI] [PubMed] [Google Scholar]
- 31. Hagerty RG, Butow PN, Ellis PM, et al. Communicating prognosis in cancer care: a systematic review of the literature. Ann Oncol. 2005;16(7):1005-1053. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data underlying this article cannot be shared due to the privacy of individuals who participated in the study. Individual deidentified data can be made available from the repository to accredited researchers who submit a proposal that is approved by the NHMRC Clinical Trials Centre.