Supportive Care in Cancer https://doi.org/10.1007/s00520-017-3986-z
ORIGINAL ARTICLE
Thinking about one’s own death after prostate-cancer diagnosis Thordis K. Thorsteinsdottir 1,2 Gunnar Steineck 6,8
&
Heiddis Valdimarsdottir 3,4 & Johan Stranne 5 & Ulrica Wilderäng 6 & Eva Haglind 7 &
Received: 25 August 2017 / Accepted: 16 November 2017 # Springer-Verlag GmbH Germany, part of Springer Nature 2017
Abstract Background Prostate-cancer diagnosis increases the risk for psychiatric morbidity and suicide. Thoughts about one’s own death could indicate need for psychiatric care among men with localized prostate cancer. We studied the prevalence and predictors of thoughts about own death among men with prostate cancer. Methods Of the 3930 men in the prospective, multi-centre LAPPRO-trial, having radical prostatectomy, 3154 (80%) answered two study-specific questionnaires, before and three months after surgery. Multivariable prognostic models were built with stepwise regression and Bayesian Model Averaging. Results After surgery 46% had thoughts about their own death. Extra-prostatic tumor-growth [Adjusted Odds-Ratio 2.06, 95% Confidence Interval 1.66–2.56], university education [OR 1.66, CI 1.35–2.05], uncertainty [OR 2.20, CI 1.73–2.82], low control [OR 2.21, CI 1.68–2.91], loneliness [OR 1.75, CI 1.30–2.35], being a burden [OR 1.59, CI 1.23–2.07], and crying [OR 1.55, CI 1.23–1.96] before surgery predicted thoughts about one’s own death after surgery. Conclusions We identified predictors for thoughts about one’s own death after prostate cancer diagnosis and surgery. These factors may facilitate the identification of psychiatric morbidity and those who might benefit from psychosocial support already during primary treatment. Keywords Prostate cancer . Psychological adaptation . Clinical trial . Thoughts about death
Eva Haglind Principal investigator of the LAPPRO trial; Gunnar Steineck Deputy principal investigator This work was supported by: The LAPPRO trial was supported by the Swedish Cancer Society (CAN2008/922, CAN2009/1099, CAN2010/ 593), Västra Götaland Region (ALFGBG-11573) and the Sahlgrenska University Hospital (HTAVGR-6011). Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00520-017-3986-z) contains supplementary material, which is available to authorized users. * Thordis K. Thorsteinsdottir
[email protected] 1
Research Institute in Emergency Care, Landspitali National University Hospital of Iceland, Reykjavik, Iceland
2
Faculty of Nursing, University of Iceland, Reykjavik, Iceland
3
Department of Oncological Sciences, Mount Sinai School of Medicine, New York, NY, USA
4
School of Business, Reykjavik University, Reykjavik, Iceland
5
Department of Urology, Sahlgrenska University Hospital, Gothenburg, Sweden
6
Division of Clinical Cancer Epidemiology, Department of Oncology, Institute of Clinical Sciences, Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden
7
Scandinavian Surgical Outcomes Research Group (SSORG/ Göteborg), Department of Surgery, Institute of Clinical Sciences, Sahlgrenska Academy at University of Gothenburg, Gothenburg, Sweden
8
Division of Clinical Cancer Epidemiology, Department of Oncology–Pathology, Karolinska Institutet, Stockholm, Sweden
Support Care Cancer
Introduction Many associate the word cancer with death, but generally a prostate-cancer diagnosis does not imply an immediate threat of death. Still, during the year after a diagnosis men commit suicide and seek psychiatric care more often than men without cancer diagnosis [1–5]. Qualitative studies have provided narratives in which men with a diagnosis of prostate cancer described the existential challenge [6–8]. Men also worried about future urinary incontinence, bowel problems and decreased sexual health as a consequence of treatment [7]. Intrusive thoughts occur [9], and men experiencing such thoughts about their prostate-cancer diagnosis reported disrupted sleep, depressed mood and impaired quality of life three months after surgery [10]. Additionally those who were uncertain about surgery providing a cure had a higher occurrence of negative intrusive thoughts about prostate cancer [11]. We need to learn more about how men think of their disease and their future after a prostate-cancer diagnosis to establish basis for interventions to prevent this morbidity and mortality. Understanding to what extent men think about their own death may be a possible way forward. With this in mind, we collected data before surgery and after the follow-up visit three months after surgery. We focused on thoughts of own death, and include an analysis of how factors before surgery predicted such thoughts before and after surgery.
from 30 to 89 years, retrieved from the Swedish Tax Agency. The data-collection for the control group was identical to the study group and was approved separately by The Regional Ethics Review board in Gothenburg (Dnr. 608–13).
Questionnaires The study questionnaires had the same clinometric approach as a previous randomized controlled trial and more than twenty large data collections of cancer survivors [13–16]. The basic concept was to atomize symptoms by asking questions about them one by one [17]. The validation process before the start of the study has been described [12]. Clinical data was derived from the study’s medical record forms completed by the doctors. The outcome variables for this study regarding thoughts about one’s own death before and after surgery were measured by the question: BHave you thought about your own death during the past month?^ The response categories were: BNo^, BYes, but more seldom than once a week^, BYes, at least once a week^, BYes, at least three times a week^, BYes, at least once a day^, BYes, at least three times a day^, BYes, at least seven times a day^. Questions about socio-demographic and health-related background as well as possible predictors for thoughts about one’s own death were asked before surgery (See Supplement 1).
Patients and methods
Statistical analysis
From 1 September 2008 to 7 November 2011 men diagnosed with prostate cancer planned for surgery at fourteen urological departments in Sweden were prospectively included in the multicentre non-randomized trial LAPPRO (ISRCTN06393679) aiming primarily at comparing outcomes after open and robotassisted laparoscopic radical prostatectomy. (Current Controlled Trials). All participants gave informed consent before inclusion and The Regional Ethics Review board in Gothenburg approved the trial (Dnr. 277–07). Each urological department reported patients to the study secretariat, which monitored and recorded data at pre-determined clinically relevant time-points. We applied a structured contact process to achieve a high rate of questionnaire completion. After an initial letter of contact a secretary called to ask permission to send out the second questionnaire. The studydesign has previously been described in detail [12]. The analysis includes all who participated in LAPPRO, who were operated on and who completed the two questionnaires, one before surgery and one three months later after the routine clinical follow-up. Prior to answering the second questionnaire participants had been informed about tumor stage based on pathology of the surgical specimen. Reference values for the outcome variable were collected at one time-point from a representative sample of the Swedish population, a total of 3000 individuals in age groups
The prevalence of thoughts of one’s own death between the two time-points and between the two surgical techniques, as well as the prevalence within the control group, were calculated and statistical significant differences analyzed with Chisquare test. The relative risk for thoughts about own death among patients at both time points compared to men in the control group was calculated with 95% confidence interval (CI). Results from our research division revealed that presence of a symptom BAt least once a week^ was significantly associated with lower level quality of life [10, 18]. This category was set as a cut-off and associations calculated using the dichotomized variable. The possible predictor variables were similarly dichotomized as existing or not based on the same prevalence. The statistical modeling was performed step-by-step. First, we calculated the percentages of participants in each category of the possible predictor variables for the occurrence of thoughts about one’s own death. The prevalence ratios for each association (univariate analysis) were calculated with 95% CI. Since the variables were closely associated, two statistical models were applied to determine the relative importance of the predictor variables in explaining the occurrence of thoughts about one’s own death. Firstly, both a forward-
Support Care Cancer
backward selection model was used, and as confirmation all the statistically significant variables (P < 0.05) from the univariate analysis were entered into a purposeful regression model [19], with and without adjusting for all the entered variables. Secondly, the possible predictors were entered into Bayesian Model Selection, performed on 100 imputed datasets that were created using multiple imputations by chained equations to compensate for randomly missing answers [20]. The analyses were repeated without thoughts about own death before surgery as a possible predictor in the models. An association with a posterior probability (PP) over 50% was seen as probable [21]. The statistical analysis was performed in SAS 9.4 (SAS Institute Inc., Cary, NC) and R [22].
Results A total of 3930 men had a radical prostatectomy in the LAPPRO trial. Of those 3154 (80.3%) returned both questionnaires required for this analysis. Table 1 describes the patients’ socio-demographics and health background. The age of participants ranged from 37 to 79 years. The mean age of those who answered only one questionnaire out of two was not statistically different from the mean age of those answering both questionnaires and neither was clinical tumor stage nor results on the outcome (Fisher’s exact test). Questionnaires were returned by 1073 (36%) of the control group, whereof 509 were men (mean age 64 years). Twenty-four per cent of the patients had thoughts about their own death at least once a week before surgery and 17% three months after surgery. In the control group 14% of the men had such thoughts at the time of answering. Table 2 shows the results from each response category on thoughts about one’s own death. Before surgery, the proportion of men reporting thoughts about their own death once a week or more often was statistically significantly different from those in the control group (24% vs. 14%, p < 0.001) and difference was also found between the types of surgery (robot 23% vs. open 27%, p = 0.02) (Chi-square test). The calculated relative risk of thoughts about own death among the patients before surgery as compared to the control group was 1.78 (CI 1.42 to 2.24). The prevalence after surgery was not statistically significantly different from the control group. The univariate analysis of all the hypothesized associations of the possible predictors with thoughts about one’s own death, before and after surgery respectively (Supplement 1), revealed that thoughts about own death before surgery were associated with such thoughts after surgery (Prevalence ratio (PR) 5.56, CI 4.74 to 6.51). Factors that were statistically significantly associated with the thoughts before surgery, but not after surgery, were: not working (PR 0.83, CI 0.73 to 0.95), co-existing illnesses (PR 1.18, CI 1.05 to 1.34) and
robot surgery (PR 0.85, CI 0.74 to 0.97). Clinical tumor stage, known before surgery, was not significantly associated with thoughts before surgery, only after surgery (PR 1.23, CI 1.05 to 1.44). The Bayesian Model Selection including thoughts about own death before surgery in the analysis revealed a posterior probability (PP) over 50% for extra-prostatic tumor-growth (pT3/pT4) (PP 100%), thoughts about own death (PP 100%), low sense of control (PP 100%), loneliness (PP 92.8%), university education (PP 86.2%) and crying (PP 58.7%). Table 3 shows the effect measures of the final models, not including thoughts about own death before surgery, relative risks and odds ratios, selected by the model-selection procedures. A posterior probability over 50% was found for eight associations before surgery and seven after. In the final multivariate model the statistically significant predictors that increased thoughts about own death before surgery were: Uncertainty about the future [Adjusted Odds Ratio (OR) 5.64, CI 4.45 to 7.16; PP 100%], negative intrusive thoughts [OR 2.33, CI 1.85 to 2.93; PP 100%], low sense of control [OR 1.94, CI 1.47 to 2.55; PP 99.87%], being a burden on others [OR 1.87, CI 1.45 to 2.43; PP 100%], uncertainty of cure [OR 1.86, CI 1.31 to 2.64; PP 93.79%], university education [OR 1.73, CI 1.40 to 2.13; PP 100%], crying [OR 1.52, CI 1.20 to 1.92; PP 89.21%], but spirituality [OR 0.74, CI 0.60 to 0.91; PP 69.70%] decreased these thoughts. After surgery the seven predictors that statistically significantly increased the occurrence were: Uncertainty about the future [OR 2.20, CI 1.73 to 2.82; PP 100%], low sense of control [OR 2.21, CI 1.68 to 2.91; 100%], extra-prostatic tumor-growth [OR 2.06, CI 1.66 to 2.56; PP 100%], loneliness [OR 1.75, CI 1.30 to 2.35; PP 97.74%], university education [OR 1.66, CI 1.35 to 2.05; PP 100%], being a burden on others [OR 1.59, CI 1.23 to 2.07; PP 77.20%] and crying [OR 1.55, CI 1.23 to 1.96; PP 94.75%].
Discussion In this cohort of patients with prostate cancer selected for surgery with a curative intention we confirmed the observation from qualitative studies that a prostate-cancer diagnosis may imply an existential challenge for some. Before surgery approximately a quarter of the men thought at least weekly about their own death. After surgery, however, three quarters of the men reported that such thoughts occurred more seldom or never. It can therefore not be assumed that all patients, who have recently been diagnosed with prostate cancer, have thoughts about their own death. Predictors for patients’ thoughts about their own death three months after surgery were: uncertainty about the future, low sense of control, loneliness, being a burden on others and crying before surgery, as
Support Care Cancer Table 1 Characteristics of the study participants undergoing surgery for prostate cancer between 1 September 2008 and 7 November 2011
No. Total non-replying (out of the study population; N = 3930) Not returning questionnaire before surgery Not returning questionnaire 3 months after surgery Participants (returning two questionnaires) Age categories (mean age 62.6; SD = 6.1) 37–55 years 56–79 years Median number of days after confirmed diagnosis, first questionnaire (lower quartile 68; upper quartile 169; mean 184; SD = 296) Median number of days before surgery, first questionnaire (lower quartile 0; upper quartile 16; mean 12; SD = 36) Median number of days after surgery, second questionnaire (lower quartile 84; upper quartile 105, mean 101; SD = 34.) Level of education Primary school Secondary school, 3 years Upper secondary school
%
561 449
14.3 11.6
149
3.8
3154
80.3
427
13.5
2727
86.5
108
6 90
588 924
18.6 29.3
383
12.1
College/University Other Not indicated Marital status Married or living with partner
1189 53 17
37.7 1.7 0.5
2650
84.0
Living alone, but has partner Living alone, no partner Not indicated Widower
215 271 18 117
6.8 8.6 0.6 3.7
1631 25 1322
51.7 0.8 41.9
176
5.6
453 1301 1371 11 18 235
14.4 41.3 43.5 0.4 0.6 7.5
1851 1217 86 1613 21 49 39 202 1056 20
58.7 38.6 2.7 51.1 0.7 1.6 1.2 6.4 33.5 0.6
Employment status before surgery Primarily employed Primarily on sick leave Primarily retired Other Residence Rural Village or town City (population > 500,000) Abroad (not in Sweden) Not indicated Birth country, not in Sweden Clinical stage, rectal palpation before surgery Non-palpable tumor (T1) Palpable tumor (T2/T3) Not indicated Co-existing illnesses1 one or more, total Stroke Embolism elsewhere Neurological illness, other Diabetes Hypertension Heart failure
Support Care Cancer Table 1 (continued) No.
%
Ischemic disease
60
1.9
Heart attack Pulmonary disease
30 79
1.0 2.5
Gastric or duodenal ulcer Kidney disease
41 43
1.3 1.4
Diagnosis of depression Inguinal hernia Prostatitis
74
2.4
234 319
7.4 10.1
2568 70 18
81.4 2.2 0.6
First sign of prostate cancer PSA elevation Enlarged prostate by examination Blood in urine
261
8.3
Primary pain
Primary urinary problems
26
0.8
Primary impotence Other signs
13 92
0.4 2.9
106
3.4
2364 790
75.0 25.0
868 2222
28.1 71.9
Not indicated Type of radical prostatectomy Robot-assisted laparoscopic Open Confirmed Pathological tumor stage (TNM-stage1 from the surgery) Extra-prostatic/Invasive (pT3/pT4) Organ confined (pT2)
Numbers do not always add to total participation rate because of internal missing data. 1 According to patients’ answers (yes/no) to BHave you had any of the following diseases during the past year?^ 2 According to the TNM classification of malignant tumors [23]
well as being diagnosed with extra-prostatic tumor-growth and having a university education. A PubMed search performed on 26 March 2017 combining Bthoughts about death^ and the MeSH term Bprostate cancer^ produced no studies with information on the prevalence of such thoughts before surgery or three months after surgery. However, studies have indicated that the prostate-cancer diagnosis can trigger distressful thought processes months or years after the diagnosis [9, 24]. In a survey at outpatient cancerclinics, where 15% of patients had prostate cancer, 7.8% of the 2365 participants had thoughts of Bbeing better off dead^ or of hurting themselves according to a question in a widely used depression screening tool [25]. Further follow-up of those reporting such thoughts revealed that one third acknowledged thoughts on being Bbetter off dead^ but not of suicide [26]. However, 23% of those reporting such thoughts at the lowest frequency of the screening tool, i.e. several days within the last two weeks, were found to be suicidal when further evaluated [27]. Our own study with data from before and three months after surgery revealed an association between having negative intrusive thoughts about the prostate cancer and a lower level quality of life [10]. In addition, we found a higher prevalence of
negative intrusive thoughts about the prostate cancer than of thoughts about one’s own death (Supplement 1) at both time points, showing that fewer participants have distressful thoughts of an existential character than about the cancer itself. Both the time point for measurement and the phrasing of the question must be considered carefully when comparing our numbers with other studies. We found a clear association between having negative intrusive thoughts and having thoughts about one’s own death. The prevalence of thoughts about own death three months after surgery was significantly lower than before surgery, and not different from the general population. The men who were to undergo robot surgery had lower prevalence compared to those operated by open surgery, but this difference disappeared after surgery. Expectations regarding the chosen surgical technique may play a role, but a reasonable explanation is that surgery, irrespective of the technique, has a major impact on some men’s ability to cope with the existential challenge of the prostate-cancer diagnosis. This may reveal the magnitude of the influence of the existential challenge of the prostate-cancer diagnosis on thoughts of death. We also found that men receiving a post-surgery diagnosis of more invasive disease were more likely to have thoughts
1693 (53.8) 422 (53.5) 1271 (53.9) *Not stated n = 19. **Not stated n = 8
314 (40.0) 1039 (44.0) 283 (55.6)
1353 (43.2)
923 (29.3)
327 (10.4) 88 (11.2)
225 (28.5) 698 (29.6)
239 (10.1) 420 (13.4)
1025 (32.7) 258 (32.8)
103 (13.1) 317 (13.5)
767 (32.7)
44 (8.6)
157 (30.8)
Yes, but less than once a week No, never
107 (3.4)
74 (2.4) 23 (2.9)
21 (2.7) 86 (3.7)
51 (2.2) 128 (4.1)
161 (5.1) 53 (6.7)
41 (5.2) 87 (3.7) 5 (1.0)
108 (4.6)
17 (0.5) 9 (1.1) 8 (0.3) 12 (1.5) 20 (0.9) 1 (0.2)
32 (1.0)
5 (0.2) 1 (0.1) 4 (0.2) 16 (0.5) 5 (0.6) 11 (0.5) 4(0.8)
15 (3.0)
Yes, at least three times a week Yes, at least once a week
Answering categories
Yes, at least seven times a day Yes, at least three times a day Yes, at least once a day
3 months after robotassisted prostatectomy Frequency (column %) Total prevalence before surgery* Frequency (total %) Before open prostatectomy Frequency (column %) Before robot-assisted prostatectomy Frequency (column %) Normal population sample men (n = 509) Frequency (column %)
Have you had thoughts about your own death during the past month?
Table 2
Results from 3154 men on thoughts about their own death before and 3 months after surgery for prostate cancer and the control group
3 months after open prostatectomy Frequency (column %)
Total prevalence 3 months after surgery** Frequency (total %)
Support Care Cancer
about their own death three months after surgery. On the other hand, the pre-surgical clinical tumor-stage did not statistically significantly predict thoughts about death before surgery. Before surgery 1059 men reported having thoughts weekly or more often about an uncertain future and 471 men reported feelings of not having control over their life (Supplement 1); an association was found with thoughts about one’s own death at both time points. The questions did not specifically address existential challenges, but our results reveal an association with the disease prognosis. Also, thoughts about own death may relate to concerns about urinary incontinence, sexual health and the effect on marriage quality, which are all factors that have been shown to affect men’s quality of life after prostate-cancer treatment [16, 28]. Men with a higher educational level had a higher prevalence of thoughts about their own death. This could possibly indicate that more awareness, associated with higher education, predicts increased vulnerability. More than one fifth of the men reported having cried weekly or more often before surgery. Studies on men’s crying are not common. A study of depressive symptomatology among men receiving brachytherapy for prostate cancer found that crying among men with localized prostate cancer was an indicator of depression [29]. Loneliness was also related to thoughts about one’s own death and may be a distinct symptom, but could also be related to depression. Research on middle-aged and elderly people indicates that loneliness and depressive symptomatology may act in synergy and result in diminished well-being [30]. A Swedish study revealed that loneliness reflected an emotional isolation that was experienced by men diagnosed with prostate cancer although they were living with a partner. The study showed that 30% of the 431 participating men were not able to confide in their partner [31]. Crying, loneliness and the feeling of being a burden associated with thoughts about own death in our study population may represent an emotional isolation and could predispose depression and the need for professional support. We applied clinical epidemiological methods and systematically recognized the possible causes of bias in each step of the research process [32]. The strengths of our study were also the prospective design, a rather homogenous population and the high response rate, all of which minimize selectioninduced problems. We systematically tried to minimize measurement errors during the preparatory phase of the study and by applying self-reported outcome measures [33]. The questions were based on concepts and wording validated by men having the same diagnosis and treatment as the study population. Our questionnaires followed a Bone question for one phenomenon^ principle and we strived for a clear definition of the concepts studied. The questionnaires were collected by a third party to prevent interviewer-related problems [34]. One limitation of the study was that we asked about thoughts about one’s own death, but not about thoughts about taking one’s
Support Care Cancer Table 3 Predictors measured before surgery for thoughts about one’s own death before surgery as well as three months after surgery. Variables in the final model selected by successive model-formation*. Values on Unadjusted odds ratio (95% confidence interval) Predictors for thoughts about one’s own death before surgery Younger age University education Living alone Not working Co-existing illness Low to moderate physical health Alcohol consumption Not certain to be cured by prostatectomy No significance of religion or spirituality Hopelessness Low self-esteem Feeling indifferent Low sense of control Burden on others Crying Feelings of loneliness Uncertainty about the future Self-reported depression Anti-depressant medication Active contact with healthcare Negative thoughts Robot assisted surgery Predictors for thoughts about one’s own death after surgery Younger age University education Living alone Pathological tumor stage Low self-assessed physical health Not certain of cure Alcohol consumption Low significance of religion or spirituality Hopelessness Low self-esteem Indifferent Low sense of control Burden on others Crying Loneliness Uncertainty about the future Self-reported depression Anti-depressants Active contact with healthcare Negative thoughts
1.58(1.26 to 1.97) 1.75 (1.48 to 2.07) 1.26(1.01 to 1.56) 0.78(0.66 to 0.93) 1.25(1.06 to 1.47) 2.27(1.92 to 2.68) 1.83(1.33 to 2.52) 4.45(3.40 to 5.82) 0.64(0.54 to 0.75) 8.15(6.50 to 10.21) 2.90(2.44 to 3.46) 6.14(5.11 to 7.36) 9.38(7.57 to 11.64) 5.04(4.13 to 6.16) 3.45(2.88 to 4.13) 6.15(4.84 to 7.81) 13.78(11.31 to 16.80) 6.21(4.45 to 8.68) 1.68(1.16 to 2.44) 0.83(0.69 to 0.99) 7.63(6.35 to 9.17) 0.80 (0.67 to 0.97)
1.32(1.02 to 1.71) 1.72(1.43 to 2.08) 1.36(1.07 to 1.73) 1.88(1.55 to 1.29) 1.95(1.62 to 2.36) 2.78(2.09 to 3.70) 1.86(1.31 to 2.64) 0.68 (0.57 to 0.83) 4.55(3.63 to 5.71) 1.88(1.55 to 2.29) 3.66(3.00 to 4.46) 5.21(4.20 to 6.47) 3.29 (2.65 to 4.08) 2.70 (2.21 to 3.30) 4.50 (3.53 to 5.74) 4.43(3.64 to 5.39) 3.48 (2.49 to 4.85) 1.69 (1.12 to 2.53) 0.82 (0.67 to 1.00)
posterior probabilities from Bayesian Model Averaging including all possible predictors enumerated in Supplement 1
Adjusted odds ratio**(95% confidence interval)
1.73(1.40 to 2.13)
1.86(1.31 to 2.64) 0.74(0.60 to 0.91)
1.94(1.47 to 2.55) 1.87(1.45 to 2.43) 1.52(1.20 to 1.92) 5.64 (4.45 to 7.16)
2.33(1.85 to 2.93)
1.66 (1.35 to 2.05) 2.06(1.66 to 2.56)
2.21(1.68 to 2.91) 1.59(1.23 to 2.07) 1.55(1.23 to 1.96) 1.75(1.30 to 2.35) 2.20(1.73 to 2.82)
Posterior probability (%)***
0.00 100.00 0.00 0.098 2.171 7.589 0.00 93.785 69.703 8.407 0.294 4.790 99.867 100.00 89.207 4.065 100.00 49.251 0.00 0.557 100.00 0.000
0.00 100.00 0.00 100.00 29.778 2.724 0.779 3.276 3.012 0.00 17.878 100.00 77.203 94.748 97.741 100.00 0.00 0.00 0.459 1.329
*An identical final model was obtained by forward variable-selection and backward variable-elimination both for thoughts about one’s own death before surgery and another identical final model was obtained by forward variable-selection and backward variable-elimination both for thoughts about one’s own death three months after surgery **Odds ratios from a model including all selected variables predicting thoughts about one’s own death before surgery and another model including all selected variables predicting thoughts about one’s own death three months after surgery ***Posterior probability; according to Kass and Raftery and co-workers a value <50% indicate an associations as improbable, 50 to 75% indicate an association as slightly probable, 75 to 95% an association as moderately probable and 95 to 100% an association as strongly probable [21]
Support Care Cancer
own life and have no information that would enable a comparison with the available reports on suicides [2–4]. Here we can only conclude that thoughts about own death relate to certain socio-demographic background factors and psychological distress and may signal a need for professional attention. The incidence of thoughts about one’s own death in the general population sample aids comparisons to Bnormality^, although the respondents constituted only 36% of the control group. A Chinese saying states BYou cannot prevent the birds of sorrow from flying over your head, but you can prevent them from building nests in your hair^. Studies on survivors of testicular and breast cancer indicate that patients want healthcare professionals to be proactive at the time of diagnosis [35, 36]. We designed this data collection as part of the LAPPRO trial, considering a broad intervention at the time of surgery to help men meet the existential challenge of receiving a prostate-cancer diagnosis. Having found that the prevalence of thinking about one’s own death at three months was close to the background level, we plan to move in another direction. We did find that a small group of men thought about their own death several times daily; many of these men were depressed. One way forward may be increased efforts to identify and treat depression at the time of diagnosis or at the first clinical follow-up after surgery. The question BAre you depressed?^ has a high sensitivity, but a low positive predictive value [37], and could be used as a clinical screening-tool.
4.
5.
6.
7. 8.
9.
10.
11.
12. Acknowledgements The authors gratefully acknowledge the participants in the LAPPRO trial, the members of the steering committee, the investigators at the participating hospitals, and the personnel at the trial secretariat for their provision of study material and administrative support.
Compliance with ethical standards
13.
Disclaimer No conflicts of interest declared. The authors alone are responsible for the content and writing of the manuscript. 14.
References 1.
Bill-Axelson A, Garmo H, Lambe M, Bratt O, Adolfsson J, Nyberg U, Steineck G, Stattin P (2010) Suicide risk in men with prostatespecific antigen-detected early prostate cancer: a nationwide population-based cohort study from PCBaSe. Sweden European Urology 57(3):390–395. https://doi.org/10.1016/j.eururo.2009.10. 035 2. Fall K, Fang F, Mucci LA, Ye W, Andrén O, Johansson J-E, Andersson S-O, Sparén P, Klein G, Stampfer M, Adami H-O, Valdimarsdottir U (2009) Immediate risk for cardiovascular events and suicide following a prostate cancer diagnosis: prospective cohort study. PLoS Med 6(12):e1000197. https://doi.org/10.1371/ journal.pmed.1000197 3. Fang F, Fall K, Mittleman MA, Sparen P, Ye W, Adami H-O, Valdimarsdottir U (2012) Suicide and cardiovascular death after a cancer diagnosis. N Engl J Med 366(14):1310–1318. https://doi. org/10.1056/NEJMoa1110307
15.
16.
17.
18.
Fang F, Keating NL, Mucci LA, Adami H-O, Stampfer MJ, Valdimarsdottir U, Fall K (2010) Immediate risk of suicide and cardiovascular death after a prostate cancer diagnosis: cohort study in the United States. J Natl Cancer Inst 102(5):307–314. https://doi. org/10.1093/jnci/djp537 Llorente MD, Burke M, Gregory GR, Bosworth HB, Grambow SC, Horner RD, Golden A, Olsen EJ (2005) Prostate cancer: a significant risk factor for late-life suicide American. Journal of Geriatric Psychiatry 13(3):195–201. https://doi.org/10.1097/00019442200503000-00004 Jonsson A, Aus G, Bertero C (2010) Living with a prostate cancer diagnosis: a qualitative 2-year follow-up. The Aging Male 13(1): 25–31. https://doi.org/10.3109/13685530903424170 Kelly D (2009) Changed men: the embodied impact of prostate cancer qualitative. Health research 19:151–163 Westman B, Bergenmar M, Andersson L (2006) Life, illness and death - existential reflections of a Swedish sample of patients who have undergone curative treatment for breast or prostatic cancer. Eur J Oncol Nurs 10(3):169–176. https://doi.org/10.1016/j.ejon. 2005.06.002 Bisson JI, Chubb HL, Bennett S, Mason M, Jones D, Kynaston H (2002) The prevalence and predictors of psychological distress in patients with early localized prostate cancer. BJU Int 90(1):56–61. https://doi.org/10.1046/j.1464-410X.2002.02806.x Thorsteinsdottir T, Hedelin M, Stranne J, Valdimarsdottir H, Wilderang U, Haglind E, Steineck G (2013) Intrusive thoughts and quality of life among men with prostate cancer before and three months after surgery. Health Qual Life Outcomes 11(1):154. https://doi.org/10.1186/1477-7525-11-154 Thorsteinsdottir T, Valdimarsdottir H, Hauksdottir A, Stranne J, Wilderäng U, Haglind E, Steineck G (2017) Care-related predictors for negative intrusive thoughts after prostate cancer diagnosis— data from the prospective LAPPRO trial. Psycho-Oncology 26(11):1749–1757. https://doi.org/10.1002/pon.4359 Thorsteinsdottir T, Stranne J, Carlsson S, Anderberg B, Bjorholt I, Damber J-E, Hugosson J, Wilderang U, Wiklund P, Steineck G, Haglind E (2011) LAPPRO: a prospective multicentre comparative study of robot-assisted laparoscopic and retropubic radical prostatectomy for prostate cancer Scandinavian. Journal of Urology and Nephrology 45:102–112 Bill-Axelson A, Holmberg L, Garmo H, Rider JR, Taari K, Busch C, Nordling S, Haggman M, Andersson SO, Spangberg A, Andren O, Palmgren J, Steineck G, Adami HO, Johansson JE (2014) Radical prostatectomy or watchful waiting in early prostate cancer. N Engl J Med 370(10):932–942. https://doi.org/10.1056/ NEJMoa1311593 Enblom A, Johnsson A, Hammar M, Onelov E, Steineck G, Borjeson S (2012) Acupuncture compared with placebo acupuncture in radiotherapy-induced nausea–a randomized controlled study. Ann Oncol 23(5):1353–1361. https://doi.org/10.1093/ annonc/mdr402 Kreicbergs U, Valdimarsdottir U, Onelov E, Henter JI, Steineck G (2004) Talking about death with children who have severe malignant disease. N Engl J Med 351(12):1175–1186. https://doi.org/10. 1056/NEJMoa040366 Steineck G, Helgesen F, Adolfsson J, Dickman PW, Johansson JE, Norlen BJ, Holmberg L (2002) Quality of life after radical prostatectomy or watchful waiting. N Engl J Med 347(11):790–796. https://doi.org/10.1056/NEJMoa021483 Steineck G, Bergmark K, Henningsohn L, Al-Abany M, Dickman PW, Helgason A (2002) Symptom documentation in cancer survivors as a basis for therapy modifications. Acta Oncol 41(3):244– 252. https://doi.org/10.1080/02841860260088782 Lind H, Waldenstrom AC, Dunberger G, Al-Abany M, Alevronta E, Johansson KA, Olsson C, Nyberg T, Wilderang U, Steineck G, Avall-Lundqvist E (2011) Late symptoms in long-term
Support Care Cancer gynaecological cancer survivors after radiation therapy: a population-based cohort study. Brit J Cancer 105(6):737–745. https://doi.org/10.1038/bjc.2011.315 19. Bursac Z, Gauss CH, Williams DK, Hosmer DW (2008) Purposeful selection of variables in logistic regression. Source Code Biol Med 3(1):17. https://doi.org/10.1186/1751-0473-3-17 20. Genell A, Nemes S, Steineck G, Dickman PW (2010) Model selection in medical research: a simulation study comparing Bayesian model averaging and stepwise regression. BMC Med Res Methodol 10(1):108. https://doi.org/10.1186/1471-2288-10-108 21. Kass RE, Raftery AE (1995) Bayes factors. J Am Stat Assoc 90(430):773–795. https://doi.org/10.1080/01621459.1995. 10476572 22. Development Core Team R (2005) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna 23. Sobin L, Gospodarowicz MK, Wittekind C (2010) TNM classification of malignant Tumours. In: Editor (ed) Book TNM classification of malignant Tumours. Wiley, Blackwell City, p 256 24. Lepore SJ, Helgeson VS (1998) Social constraints, intrusive thoughts, and mental health after prostate cancer. J Soc Clin Psychol 17(1):89–106. https://doi.org/10.1521/jscp.1998.17.1.89 25. Walker J, Waters RA, Murray G, Swanson H, Hibberd CJ, Rush RW, Storey DJ, Strong VA, Fallon MT, Wall LR, Sharpe M (2008) Better off dead: suicidal thoughts in cancer patients. J Clin Oncol 26(29):4725–4730. https://doi.org/10.1200/JCO.2007.11.8844 26. Walker J, Hansen CH, Butcher I, Sharma N, Wall L, Murray G, Sharpe M (2011) Thoughts of death and suicide reported by cancer patients who endorsed the Bsuicidal thoughts^ item of the PHQ-9 during routine screening for depression. Psychosomatics 52(5): 424–427. https://doi.org/10.1016/j.psym.2011.02.003 27. Walker J, Hansen CH, Hodges L, Thekkumpurath P, O'Connor M, Sharma N, Kleiboer A, Murray G, Kroenke K, Sharpe M (2010) Screening for suicidality in cancer patients using item 9 of the nineitem patient health questionnaire; does the item score predict who requires further assessment? Gen Hosp Psychiatry 32(2):218–220. https://doi.org/10.1016/j.genhosppsych.2009.11.011 28. Sunny L, Hopfgarten T, Adolfsson J, Steineck G (2007) Economic conditions and marriage quality of men with prostate cancer. British Journal of Urology International 99(6):1391–1397. https://doi.org/ 10.1111/j.1464-410X.2007.06807.x
29.
30.
31.
32.
33.
34.
35.
36.
37.
Sharpley CF, Christie DH, Bitsika V, Oar AJ (2013) The effects of low- and high-dose-rate brachytherapy on depressive symptoms in prostate cancer patients. Int J Clin Oncol 19(6):1080–1084. https:// doi.org/10.1007/s10147-013-0647-1 Cacioppo JT, Hawkley LC, Thisted RA (2010) Perceived social isolation makes me sad: 5-year cross-lagged analyses of loneliness and depressive symptomatology in the Chicago health, aging, and social relations. Study Psychol Aging 25(2):453–463. https://doi. org/10.1037/a0017216 Helgason AR, Dickman PW, Adolfsson J, Steineck G (2001) Emotional isolation: prevalence and the effect on well-being among 50-80-year-old prostate cancer patients. Scand J Urol Nephrol 35(2):97–101 Steineck G, Hunt H, Adolfsson J (2006) A hierarchical step-model for causation of bias-evaluating cancer treatment with epidemiological methods. Acta oncologica (Stockholm, Sweden) 45(4):421– 429. https://doi.org/10.1080/02841860600649293 Litwin MS, Lubeck DP, Henning JM, Carroll PR (1998) Differences in urologist and patient assessments of health related quality of life in men with prostate cancer: results of the CaPSURE database. J Urol 159(6):1988–1992. https://doi.org/10.1016/ S0022-5347(01)63222-1 Månsson A, Henningsohn L, Steineck G, Månsson W (2004) Neutral third party versus treating institution for evaluating quality of life after radical cystectomy. Eurpean Urology 46(2):195–199. https://doi.org/10.1016/j.eururo.2004.04.010 Skoogh J, Steineck G, Johansson B, Wilderang U, Stierner U (2013) Psychological needs when diagnosed with testicular cancer: findings from a population-based study with long-term follow-up. BJU Int 111(8):1287–1293. https://doi.org/10.1111/j.1464-410X. 2012.11696.x Stinesen-Kollberg K, Thorsteinsdottir T, Wilderang U, Steineck G (2013) Worry about one's own children, psychological well-being, and interest in psychosocial intervention. Psycho-Oncology 22(9): 2117–2123. https://doi.org/10.1002/pon.3266 Skoogh J, Ylitalo N, Larsson Omerov P, Hauksdottir A, Nyberg U, Wilderang U, Johansson B, Gatz M, Steineck G (2010) ‘A no means no’–measuring depression using a single-item question versus hospital anxiety and depression scale (HADS-D). Ann Oncol 21(9):1905–1909. https://doi.org/10.1093/annonc/ mdq058