If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Preferred and perceived participation roles of oncological patients in medical decision-making: Results of a survey among users of the German Cancer Information Service
Cancer patients are facing a variety of treatment and other disease-related decisions. This study aims to provide insights into preferred and perceived participation roles in decision-making among patients with diverse tumors using the German Cancer Information Service (CIS).
Methods
Patients’ decision-making preferences and experiences were assessed as a part of a cross-sectional survey among CIS users. An adapted version of the Control Preferences Scale (CPS) was used to assess preferred and perceived participation roles in eight different areas of medical decision-making (e. g., choice of medication, termination of treatment). Logistic regression analyses were applied to explore preference matching and to analyze associations between participation roles and sociodemographic variables. Moreover, we examined preferences and perceptions of participation roles across different decision situations.
Results
In the final sample (N = 1566, 64.9% female, mean age = 61.6), almost half of the patients (47.1%) preferred to take a collaborative role in decisions on treatment methods, whereas 36.3% preferred an active role and 15.9% a passive role. Collaborative role preferences frequently (40.7%) coincided with experiencing a passive role and predicted a reduced chance of a match between preferences and experiences (OR = 0.57, p = .001). A higher level of education was associated with a lower chance of preferring and perceiving a passive role (OR = 0.85, p < .01). Compared with men, women had increased odds of preferring (OR = 1.45, p < .05) and of actually taking (OR = 2.04, p < .001) a passive role in medical encounters. Preferred participation roles regarding treatment methods were highly correlated with preferences in all other decision areas (r > .50, p < .001) except decisions about family involvement.
Conclusions
The study reveals well-known deficits in the fulfilment of patients’ collaborative role preferences across different areas of medical decision-making in a sample of CIS users characterized by high information-seeking behavior. Participation roles were not only influenced by the patients' level of education but also by their gender. The gender effect may be more pronounced than previous studies suggest. These effects should be considered in the development of interventions to promote shared decision-making. Additionally, study results indicate that preferences for participation in decisions about treatment methods, as assessed by the CPS, can be generalized to other areas of medical decision-making.
Zusammenfassung
Hintergrund
Krebspatient*innen sind mit einer Vielzahl von Behandlungsentscheidungen und weiteren krankheitsbezogenen Entscheidungen konfrontiert. Ziel dieser Studie ist es, Einblicke in den präferierten und erlebten Grad der Beteiligung an Entscheidungen bei Patient*innen mit unterschiedlichen Tumorerkrankungen zu geben, die den Krebsinformationsdienst (KID) nutzen.
Methoden
Präferenzen und Erfahrungen von Krebspatient*innen mit Entscheidungsfindungsprozessen wurden im Rahmen einer querschnittlichen Befragung von KID-Nutzer*innen erfasst. Eine adaptierte Version der Control Preferences Scale (CPS) erfasste den präferierten und erlebten Grad der Partizipation in acht verschiedenen Bereichen medizinischer Entscheidungsfindung (u. a. Medikamentenwahl, Therapiebeendigung). Mittels logistischer Regressionsanalysen wurden die Passung von Präferenz und Erfahrung sowie Zusammenhänge zwischen dem Grad der Partizipation und soziodemografischen Variablen untersucht. Zudem wurden Präferenzen und Erfahrungen über verschiedene Entscheidungsbereiche hinweg analysiert.
Ergebnisse
In der untersuchten Stichprobe (N = 1566, 64,9% weiblich, mittleres Alter = 61,6), präferierte fast die Hälfte der Patient*innen (47,1%) eine kollaborative Behandlungsentscheidung, während 36,3% eine aktive Rolle und 15,9% eine passive Rolle bevorzugten. Die Präferenz einer kollaborativen Entscheidungsfindung ging häufig (40,7%) mit der Erfahrung einer passiven Rolle einher und war ein Prädiktor für eine geringe Passung zwischen Präferenz und Erfahrung (OR = 0,57, p = .001). Ein höheres Bildungsniveau verringerte die Wahrscheinlichkeit, eine passive Rolle zu bevorzugen und zu erleben (OR = 0,85, p < .01). Im Vergleich zu Männern hatten Frauen eine erhöhte Wahrscheinlichkeit, eine passive Rolle zu bevorzugen (OR = 1,45, p < .05) und tatsächlich zu erleben (OR = 2,04, p < .001). Präferenzen und Erfahrungen der Partizipation waren über alle Entscheidungsbereiche hinweg hoch korreliert (r > .50, p < .001), mit Ausnahme von Entscheidungen über den Einbezug der Familie.
Fazit
Die Studie zeigt bekannte Defizite bei der Befriedigung des Bedürfnisses von Patient*innen nach kollaborativer Entscheidungsfindung in unterschiedlichen Bereichen der medizinischen Entscheidungsfindung unter Nutzer*innen des KID, die sich durch aktives Informationssuchverhalten auszeichnen. Nicht nur das Bildungsniveau, sondern auch das Geschlecht der Befragten beeinflusste ihren Grad an Partizipation. Das Geschlecht könnte somit einen bedeutsameren Einfluss haben, als frühere Studien nahelegen. Diese Ergebnisse sollten folglich bei der Entwicklung von Interventionen zur Förderung partizipativer Entscheidungsfindung berücksichtigt werden. Darüber hinaus zeigen die Studienergebnisse, dass sich der präferierte Grad der Beteilung an Behandlungsentscheidungen, wie er von der CPS erfasst wird, auf andere Bereiche medizinischer Entscheidungsfindung übertragen lässt.
The possibilities of curative and palliative treatment of various cancers as well as their diagnostic detection have dramatically increased in recent years. For patients and their treating physicians this results in various complex decisional situations across all phases of illness. For a collaborative communication between physician and patient, in the sense of shared decision-making (SDM) [
], the physicians should acknowledge and respect patients’ individual preferences in all disease-related decisions. Besides patients’ preferences regarding diagnostic methods, treatment options, and life circumstances this involves taking into account patients’ preferences for the involvement in the decision itself, as some patients prefer to take a passive role and SDM should never be forced upon patients [
Yet, a large number of studies shows that the majority of oncological patients of various tumor entities want to be actively involved in medical decisions [
Cuypers M, Lamers RE, de Vries M, Husson O, Kil PJ, van de Poll-Franse LV. Prostate cancer survivors with a passive role preference in treatment decision-making are less satisfied with information received: Results from the PROFILES registry, Urologic Oncology: Seminars and Original Investigations; 2016: Elsevier.
]. According to a systematic review of 31 studies published between 2009 and 2020, almost half of cancer patients prefer a collaborative role, while about a quarter each prefer an active or passive role in the decision-making process with their physician [
]. Patients with active participation preferences may take medical advice but eventually want to decide on their own. The review also showed a well-known gap between patients’ preferences and their actual experiences with SDM, especially for patients with collaborative decision-making preferences [
]. The phase of illness may influence participation preferences. Patients with advanced cancer have a variable need for participation and there are findings suggesting that with progression of disease patients want to give up control [
The present study aims to focus on a specific patient subgroup characterized by high information-seeking behavior which constantly increases due to the possibilities of the Internet [
]. We surveyed preferred and perceived participation amongst users of the German Cancer Information Service (CIS). The CIS is a division of the German Cancer Research Center and provides evidence-based, up-to-date, individually tailored information on all aspects of cancer to patients, relatives and general public via telephone or email. Almost all users of the CIS actively search for disease-related information even before contacting the CIS, most frequently using the Internet (an overview about information needs and sources of CIS users is published elsewhere [
] our first study goal is to analyze preferred and perceived participation roles and their sociodemographic and illness-related determinants in a convenience sample of CIS users.
The second goal of the study is of a methodological nature, although it has important implications for the improvement of clinical care. The preferred level of participation is widely measured with the Control Preferences Scale (CPS), which asks patients to indicate their need for involvement in “treatment decisions” [
]. To better meet patients’ participation preferences in medical encounters, considerations to include the CPS into routine screenings for cancer patients seem to be a feasible and effective procedure [
Cancer patients' control preferences in decision making and associations with patient-reported outcomes: A prospective study in an outpatient cancer center.
]. However, it is unclear whether patient responses in the CPS can be transferred to the many different decision-making situations that cancer patients are confronted with, because the one-item CPS focuses on “treatment decisions” in general. According to Ernst and colleagues [
], patients with solid cancer are less interested in involvement in decisions about medication and end of treatment than in other situations of medical decision-making (e.g. diagnostic tests). Hence, the second goal of our study is to explore in a larger sample, if patients’ needs for participation in treatment decisions as measured by the CPS can be generalized to other situations of medical decision-making, which would make the CPS a suitable screening instrument.
In summary, using a sample of CIS-users with diverse tumors and phases of illness who are generally characterized by their high information-seeking behavior we want to address the following research questions:
1
A.) How are preferred and perceived participation roles distributed in this sample, and how high is the concordance between them?
B.) Are sociodemographic variables or advanced cancer associated with a perceived or preferred passive role in oncological patients?
2
Are preferred and perceived participation roles maintained across different decision-making situations?
Methods
Data source
We used data from a survey of users of the CIS hotline, which ran from June 2016 through April 2017. Upon consent to participate, users received a link to an online questionnaire (LimeSurvey) or a paper-based version at their own request. The CIS conducts such surveys every 5 or 6 years as a tool of quality assurance. A “SDM module” was included in the patient questionnaire to retrospectively assess patients’ SDM needs and experiences. The present analysis is based on the “SDM module” solely.
Sample
Initially, N = 4.217 callers were invited to participate in the survey and N = 2.990 (71%) gave their consent. In total, N = 2010 (67% response rate) actually participated in the CIS survey and N = 1890 patients responded to the “SDM module” (63% response rate). We excluded patients who exhibited more than 20% missing values within the variables examined in this study, resulting in a final sample size of N = 1566. Among them, 929 patients (59.3%) answered the survey online, while 637 patients (40.7%) completed the paper-based version. The participants in the online survey were younger (t(1413) = −11.91, p < .001) and more educated (t(1279) = 9.39, p < .001) than the participants in the paper-based survey. There was no difference in the gender distribution between the groups (χ2 (2) = 4.49, p > .05).
Measures
For this manuscript, sociodemographic variables, disease-related variables and preferred and perceived participation roles in the medical decision-making process were examined. Sociodemographic variables included age, gender, level of education, and native language. Disease-related variables included tumor site and phase of illness.
Preferred participation roles
An adapted version of the CPS by Degner and Sloan [
] was used to assess preferred participation roles, i.e., patients’ preferred extent of involvement in decision-making with their physician. Patients had to choose one of the following statements describing their preferred style of decision-making with regard to a certain medical decision: (1) I would like to decide on it myself, (2) I would like to decide on it myself in the end, after I have seriously considered my physician's opinion, (3) I would like my physician and me to decide on it in joint responsibility, (4) I would like my physician to make the final decision on it but take my opinion into account, or (5) I would like to leave the decision to my physician. For the analysis, the response options were grouped into three categories, (1) and (2) reflecting an active approach, (3) a reflecting a collaborative approach, and (4) and (5) reflecting a passive approach, as proposed by Degner and Sloan [
] that has not been psychometrically tested. This adapted version captures preferred participation roles in eight distinct decision areas: (1) diagnostic tests, (2) treatment methods, (3) choice of medicine, (4) consultation with another physicians, (5) place of treatment, (6) time of treatment, (7) ending treatment, and (8) family involvement.
Perceived participation roles
Perceived participation roles were assessed analogously to preferred participation roles. For this purpose, patients had to indicate how they experienced the decision-making process in the eight decision areas so far. Analog response options (e.g., “I have decided on it myself”) and the same categorization scheme as in the assessment of preferred participation roles were used.
Statistical analysis
The data were analyzed using Statistical Package for Social Sciences (SPSS), version 27 (IBM Corp, Chicago, Illinois).
For all statistical analyses on research questions 1A and B only participation roles in decision area 2 (treatment methods) were analyzed because this decision area corresponds to participation roles as defined by the original CPS [
]. The choice of this outcome variable allows us to compare the preferred and perceived participation roles in our sample of CIS users with studies surveying other populations (e.g. the general population of German cancer patients) that predominantly applied the CPS.
Cross-tabulation was applied to describe the frequency distribution of preferred and perceived participation roles and concordance between them (research question 1A). Based on those results, we exploratively applied logistic regression to investigate whether role concordance (‘0 = discordance’ and ‘1 = concordance’) depended on preferred participation roles. Active and collaborative participation preferences were compared against passive participation preferences. The regression model controlled for age, gender, and level of education.
To investigate the effect of age, gender, and level of education on preferred and perceived participation roles binary logistic regression was used (research question 1B). The passive participation role was chosen as the dependent variable as the goal was to investigate factors associated with preferences and experiences of the paternalistic model of decision making. Active and collaborative roles were combined to form a reference category, as they both represent patient participation, albeit to varying degrees. In the first model we defined the preferred passive participation role as the dependent variable, and in the second model we defined the perceived passive participation role as the dependent variable (1 = passive role; 0 = active or collaborative role, respectively). Age, gender, and level of education were entered as independent variables. Based on the theoretical assumption that gender and age could have a conditional effect on participation roles we also included a respective interaction term (gender x age). As recommended for the analysis of conditional effects [
] we mean-centered age as a metric variable before including it into the model. To investigate the influence of advanced cancer on participation roles (research question 1B) we have run two additional logistic regression models, one to predict preferred participation and one to predict perceived participation, including only patients with advanced cancer (i.e. patients with recurrent disease and patients in a palliative situation) and patients receiving treatment for the first time. ‘Advanced cancer’ was compared against ‘first treatment’ (reference category) because it can be assumed that patients receiving their first treatment have a better prognosis, or at least a higher level of optimism regarding their prognosis. The regression models controlled for age, gender, age x gender (interaction), and level of education.
We calculated Spearman’s rank correlations to explore how strongly preferences and perceptions of participation roles are related across different decision situations (research question 2). As the reliability of a correlation analysis increases with the number of levels of the tested variables, the five response options of the variables were not grouped into three categories for this analysis.
Results
Sample characteristics
The sample characteristics are presented in Table 1. The sample included 1017 women (64.9%) and 547 men (34.9%). Participants were aged between 25 and 99 years (M = 61.6, SD = 11.4).
737 patients (47.1%) wanted to share decisions with their physicians (collaborative role) in decisions on treatment methods (i.e., decision area 2), 568 patients (36.3%) preferred to take an active role, and 249 patients (15.9%) preferred a passive role. When asked about perceived participation, 576 patients (36.8%) stated that they had taken a passive part, 521 patients (33.2%) stated that they had resumed an active role, and 380 patients (24.3%) felt that they had decided collaboratively with their physician. The frequencies of preferred and perceived participation roles in patients in the first treatment block vs. patients with advanced cancer of disease are shown in Table 2.
Table 2Preferred and Perceived Participation in different Phases of Illness.
Figure 1 displays a high concordance between preferred and perceived participation regarding active roles (66.0%) and passive roles (80.2%), whereas most of the patients (40.7%) preferring a collaborative role actually perceived that they played a passive role. The weighed kappa was .40 (p < .001), indicating a moderate role concordance.
Figure 1Concordance between patients preferred and perceived participation roles.
Based on this finding, we exploratively applied logistic regression to investigate the influence of preferred participation roles on the experience of a concordant role (see Table 3). Preferred participation roles distinguished between concordant and discordant roles (χ2(2) = 144.23, p < .01). Patients with active or collaborative role preferences had significantly decreased odds of experiencing a concordant role (p < .001) independent of their gender, age, and level of education. The effect was more pronounced for collaborative role preferences (OR = 0.20) than for active role preferences (OR = 0.53).
Table 3Associations between concordant roles and preferred participation roles (controlled for gender, age, and education).
Associations with Sociodemographic Variables and Advanced Cancer
The logistic regression model investigating role preferences significantly distinguished between passive and active/collaborative role preferences (χ2(4) = 12.08, p < .05). Compared with men, women had significantly increased odds of preferring a passive participation role (OR = 1.45, p < .05). Higher education was associated with a lower chance of preferring a passive role (OR = 0.85, p ≤ .01), while no significant associations with age or a conditional effect of age and gender were found (see Table 4).
Table 4Associations between passive role preferences and sociodemographic factors.
With respect to actually perceiving a passive participation role (see Table 5), female gender was associated with a higher chance (OR = 2.04, p < .001), while higher education was associated with a lower chance (OR = 0.85, p ≤ .001).
Table 5Associations between passive role perceptions and sociodemographic factors.
The additional binary logistic regression analyses showed that patients with advanced cancer had decreased odds to prefer (OR = 0.62, p < .05) or experience (OR = 0.70, p < .05) a passive role than patients receiving treatment for the first time (see Appendix A, Table A and B).
Cross-situational consistency of participation roles
Figure 2 presents the distribution of preferred and perceived participation for distinct decision areas. The preference for an active role was greatest regarding the decision to involve family members (65.7%) and lowest regarding medication (27.6%). The same proportions were observed with respect to perceived participation.
Figure 2Preferred and perceived participation roles in distinct decision areas (note that percentages do not include missing values).
The preferred participation roles in decisions about treatment methods were strongly correlated with the preferred participation roles in all other decision areas (r > .5, p < .001), except for family involvement, for which the effect size of the correlation was medium (r = .29, p < .001). Perceived participation in decisions about treatment methods was also significantly correlated with the perception of participation in all other decision areas. However, correlations were less strong (r > .2, p < .001) than in the correlation analysis with preferred participation roles. A complete overview of the correlations between the different decision areas regarding preferred and perceived participation can be found in Table C and Table D in the Appendix A.
Discussion
The primary aim of the present study was to generate findings about preferred and perceived participation roles in medical decisions in a large, heterogeneous sample of oncological patients using the German Cancer Information Service (CIS). The assessment of participation roles in different medical decisions also provided an opportunity to test if participation preferences in treatment decisions, as assessed with the CPS, can be generalized to other areas of medical decision making.
Distribution and Concordance between Preferred and Perceived Participation roles
Almost half of the patients using the CIS (47%) wanted to share decisions with their physician. This confirms once more that SDM is the preferred communication style of most oncological patients [
]. Compared with previous reports, we found a higher proportion of patients preferring an active participation role in decisions about treatment methods (36%). Users of the German CIS, who are actively seeking information about their disease, might genuinely be more inclined to take an active role in the medical decision-making process. Conversely, the percentage of patients with passive decision preference was much lower amongst CIS users (15%) than in an epidemiological study conducted in Germany between 2008 and 2010 (N = 4020) in which about one third of cancer patients each preferred a an active, collaborative, and passive role in the decision-making process [
]. Participants of the epidemiological study were selected to be representative of German oncological patients in terms of tumor entities and cancer care facilities. In comparison, our convenience sample of CIS users included a higher percentage of women (65% vs. 51%), breast cancer patients (43% vs. 23%) and patients with more than 9 years of education (81% vs. 31%). Since education was negatively associated with passive role preferences (OR = 0.85, p ≤ .01), we assume that the higher proportion of patients with active participation preferences in our study is mainly explained by the educational level. The higher proportion of breast cancer patients in our sample probably does not play a relevant role, as previous findings suggest that the proportion of breast cancer patients with active participation preferences equals the proportion of patients with other tumor entities [
In addition to preferred participation roles, we analyzed role concordance, i.e., the match between the preferred and perceived role. If the preferred and experienced roles do not match, i.e., are discordant, then patients’ needs have presumably not been met. On average, almost 38.1% of patients reported such role discordance and role discordance was highest for patients with collaborative decision preferences. This also replicates findings from the systematic review published in 2021 [
]. Less than 40% of patients with collaborative decision preferences reported that they had felt involved to their desired extent. It is sobering to say that just as many of them reported finding themselves in an undesired passive role in medical encounters. A logistic regression analysis predicting role concordance with preferred participation roles confirmed that when preferring a collaborative role, patients’ chances of experiencing a concordant, i.e., collaborative role, were reduced by 80% (p < .001) compared with patients with passive role preferences (OR = 0.20, p < .001). A similar but smaller effect was found for preferences of active roles compared with passive role preferences (OR = 0.57, p < .001). According to these results, patients who know that they want to make the decision autonomously likely manage to adopt a demanding attitude in the physician's consultation. On the other hand, those who want the physician to decide easily fit into the rather paternalistic communication offer made by many physicians. Patients who want to share decisions, however, do not have their need properly acknowledged and met. To remedy this deficiency, routinely screening patient preferences by means of CPS, as suggested by Schuler et al. [
Cancer patients' control preferences in decision making and associations with patient-reported outcomes: A prospective study in an outpatient cancer center.
] would certainly be an important first step. This way, the physician could catch up about the patient’s individual decision-making preference before the consultation, which could also serve as a reminder of the goal to apply SDM principles in the encounter. However, the main problem remains that many physicians continue to lack a clear understanding of what SDM really means. In our clinical experience, many physicians confuse SDM with active, patient-driven decision-making. When attempting to implement an SDM-based communication style, physicians often ask their patients, what treatment option they want to choose. Most patients feel overwhelmed by such a question and justifiably respond that they do not have sufficient medical expertise to answer this question, which the inexperienced physician then in turn misinterprets as a request for a paternalistic conversation style. To address such misconceptions, SDM trainings for physicians are of great importance (see Conclusions).
Associations with Sociodemographic Variables and Advanced Cancer
We further examined sociodemographic determinants of oncological patients’ preferences and perceptions of participation. As in other populations [
], a lower level of education was associated with a passive role preference (OR = 0.85, p ≤ .01) and with perceptions of a passive role (OR = 0.85, p ≤ .001) among users of the German CIS. Evidently, a lack of education, which is linked to low health literacy [
D. Schaeffer, E.-M. Berens, S. Gille, L. Griese, J. Klinger, et al, Gesundheitskompetenz der Bevölkerung in Deutschland vor und während der Corona Pandemie: Ergebnisse des HLS-GER 2. Universität Bielefeld, Interdisziplinäres Zentrum für Gesundheitskompetenzforschung (2021) https://doi.org/10.4119/unibi/2950305
], is a meaningful obstacle to SDM on the patient side. This clearly supports approaches of developing easy-to-understand patient decision aids (PtDAs) to empower patients and facilitate SDM by providing, at a minimum, information on treatment options and their benefits and risks [
Interestingly, in our sample of CIS users, female gender was associated with passive role preferences and even stronger with passive role perceptions, while in previous studies evidence for an effect of gender was very mixed (for a review see [
]. The causes of the found gender differences are probably complex, and as there is a lack of empirical evidence and research under controlled conditions, general conclusions cannot be drawn from this result, especially since our sample of CIS users is not representative for the population of cancer patients. Sociological theories about gender socialization claim that gender is a learned construct that influences how we perceive ourselves and others [
]. Following this argument, one could generally assume that women are more often considered to have a passive role preference and feel less empowered to actively demand participation, as assertiveness is considered a male characteristic. More specifically, gender is considered a very salient status characteristic that is activated in most encounters and invokes cultural stereotypes that men are more competent than women [
]. According to expectation states theory, people predict the behavior of their interactional partners by locating their status in the social structure. High-status persons are given more opportunities to participate in a conversation [
]. Regression analyses revealed no conditional effect of gender and age, so that it cannot be assumed that the gender effect only applies to older or younger patients.
Another interesting finding is that patients with advanced cancer preferred (OR = 0.62, p < .05) and perceived (OR = 0.70, p < .05) to be less passive than patients receiving treatment for the first time. This finding may also be specific to patients with high information-seeking behavior as a qualitative study suggests, that patients with advanced cancer must actively bring up their treatment preferences as they are rarely invited by their physicians to share them [
Cancer patients' control preferences in decision making and associations with patient-reported outcomes: A prospective study in an outpatient cancer center.
] reported that younger patients take and prefer more active roles in decision-making, age had no impact on participation roles in our sample. This might again reflect a selection bias, as users of the CIS presumably are more active on their behalf than average patients of the same age.
Cross-situational consistency of participation roles
A special characteristic of this study was that it analyzed preferred and perceived participation roles in eight distinct decision areas. In summary, high correlations of participation preferences and perceptions across different decision situations were found, providing evidence that preferred and perceived participation roles are quite stable here. The preference for an active role was greatest regarding the decision to involve family members (65.7%) and lowest regarding medication (27.6%). A similar pattern was found in a study by Ernst and colleagues [
]. The decision to involve family members has a smaller impact on the development of the disease and could be considered by patients as less important than decisions with medical consequences, such as questions of medication. Alternatively, involving family members, compared with the other decisional situations, depends more on the patients’ personal values and less on the physicians’ professional knowledge. Therefore, patients may wish to play a very active role in this decision, and the link to participation preferences in other decisional situations is less pronounced. In line with that, participation preferences concerning treatment methods (i.e., those that can be viewed analogously to the CPS by Degner and Sloan [
]) were very highly correlated (r > .5, p < .001) with all decision areas except for family involvement. We therefore conclude that a one-item assessment of preferred participation roles as used in the CPS [
] may therefore be generalized across decisions that involve weighing facts that require medical expertise. The CPS can therefore be well incorporated into existing routine screening questionnaires applied in German comprehensive cancer centers to provide physicians with a quick overview of the patient's need for participation in all areas of medical decision making, as suggested by Schuler et al. [
Cancer patients' control preferences in decision making and associations with patient-reported outcomes: A prospective study in an outpatient cancer center.
Comparing preferred and perceived participation roles in distinct decision situations (Figure 2), the proportion of perceived collaborative roles is clearly smaller than the proportion of preferred collaborative roles in all decision situations. Hence, a relevant secondary finding is that discordance was found across different decision areas.
Strengths and limitations
The chosen survey setting – the CIS – guaranteed the analyses of preferred and perceived participation roles in a large and diverse sample of German cancer patients, including a variety of tumor sites and phases of illness. Moreover, due to the use of distinct channels to implement the survey, i.e. online or paper-based questionnaire, different groups of patients were reached, with older and less educated patients preferring the paper-based version over the online version of the survey. This led to a wide range of sociodemographic characteristics in the sample, although not necessarily representative to the general population of German cancer patients. For example, women and well-educated patients are overrepresented amongst CIS users. Previous studies have shown that women contact the CIS more often than men [
Trends in cancer information services over 25 years: An analysis of inquiries from patients and relatives made to the Cancer Information Service of the German Cancer Research Center from 1992 to 2016.
A. Heimer, M. Henkel, Analysis of information seeking of people with counselling requests: representative survey on behalf of the German cancer information service of the German cancer research center, Akademische Verlagsgesellschaft AKA GmbH. 2012: Heidelberg.
]. All study findings must be interpreted against the background that the surveyed CIS users are active information seekers and presumably have higher self-management strategies and health literacy than the general population of cancer patients.
Conclusions
This study, like many others before, shows that SDM is not an abstract ideal developed in academic circles; rather, the majority of oncological patients (47.1%) actually want to share decisions with their physicians. The attention of practitioners should be drawn to the fact that patients’ desire for participation is also high in decision areas that are not regarded as typical SDM-suitable decisions. This applies, for example, to decisions about the use of diagnostic tests and about the place of treatment. Patients with advanced cancer also showed a high need for SDM within this sample, so that practitioners should be careful not to underestimate their need for involvement.
The fact that 40.7% of patients with a desire for SDM (i.e., preference of a collaborative role) experience being in a passive role in decision-making illustrates that patients feel insufficiently involved in medical decisions. This role discordance was found in all decision areas. In addition to limiting factors of the health care system that should not be forgotten in this debate (e.g. time constraints), improving physicians’ SDM skills seems to be an adequate way to promote SDM in clinical practice. There are already many such training programs, and SDM also increasingly finds its way into the curricula of medical psychology at German universities [
] as well as for a consensus on criteria for certifying SDM training programs in continuing education as they vary greatly in their quality and effectiveness [
]. Therefore, it is important to also develop interventions that empower patients to claim their right to SDM. Promising programs exist, but the implementation of SDM into the routine care in Germany is realized at a rather slow pace [
]. An exemplary model project that takes an equally physician- and patient-centered approach to improving SDM is currently being implemented University Hospital Campus Kiel (Germany), involving 27 clinical departments with more than 650 physicians [
Making shared decision-making (SDM) a reality: protocol of a large-scale long-term SDM implementation programme at a Northern German University Hospital.
]. The project comprises trainings for physicians and other healthcare professionals, interventions to activate and empower patients based on the idea of prompt-sheets, and the development and implementation of decision aids. If the gender effects observed in this study can be replicated in other studies with comparable sample sizes, we propose to develop more interventions that specifically address female patients. Gender may be an important factor in the empowerment of oncological patients but is not sufficiently researched [
Ethics approval was obtained from the Ethic Committee of the Heidelberg University Hospital (approval number: S-193/2016). Written informed consent to participate was obtained from all participants.
Conflict of interest
The authors declare that there is no conflict of interest.
CRediT author statement
Pia Grabbe: Formal analysis, Visualization, Writing – Original draft
Cuypers M, Lamers RE, de Vries M, Husson O, Kil PJ, van de Poll-Franse LV. Prostate cancer survivors with a passive role preference in treatment decision-making are less satisfied with information received: Results from the PROFILES registry, Urologic Oncology: Seminars and Original Investigations; 2016: Elsevier.
Cancer patients' control preferences in decision making and associations with patient-reported outcomes: A prospective study in an outpatient cancer center.
D. Schaeffer, E.-M. Berens, S. Gille, L. Griese, J. Klinger, et al, Gesundheitskompetenz der Bevölkerung in Deutschland vor und während der Corona Pandemie: Ergebnisse des HLS-GER 2. Universität Bielefeld, Interdisziplinäres Zentrum für Gesundheitskompetenzforschung (2021) https://doi.org/10.4119/unibi/2950305
Trends in cancer information services over 25 years: An analysis of inquiries from patients and relatives made to the Cancer Information Service of the German Cancer Research Center from 1992 to 2016.
A. Heimer, M. Henkel, Analysis of information seeking of people with counselling requests: representative survey on behalf of the German cancer information service of the German cancer research center, Akademische Verlagsgesellschaft AKA GmbH. 2012: Heidelberg.
Making shared decision-making (SDM) a reality: protocol of a large-scale long-term SDM implementation programme at a Northern German University Hospital.
00075-7&id=gr1.jpg){kind=link}
00075-7&id=gr2.jpg){kind=link}