4.1 How cancer patients perceive the role of the GP in cancer care in general

Similar to Lang et al., 2017, almost 90% of the respondents stated that it was important to them that their GP was regularly informed on cancer therapy. However, only one third of the patients actually consulted the GP when making decisions about diagnostic methods and treatment of cancer. This could be due to the fact that many patients consider the GP to be not competent enough in the treatment of cancer and feel more comfortable in the care of a specialist in this respect (Ben-Ami et al., 2014; Brandenbarg et al., 2017; Coindard et al., 2016). This contrasts with the fact that the majority of patients would like to have a more active involvement of the GP in cancer care. Actually, GPs want to take over a more active role as well (Meiklejohn et al., 2016; Noteboom et al., 2021). In fact, involving the GP in the treatment shows positive effects on the patients' well-being and is essential for the care of the steadily increasing number of cancer patients (Rubin et al., 2015). In our survey, patients who received their diagnosis from their GP consulted him significantly more often, as Lang et al., 2017 has already shown. In order to combine the specialist's competence and the all-embracing care of the GP, shared-care models (Rubin et al., 2015) should be considered more often. In shared care models, cancer patients are cared for in partnership by oncology specialists and GPs, who are in constant exchange of information and make joint decisions regarding the course of procedure, so that a smooth transition of the patient between the oncology and GP settings is possible (American Society of Clinical Oncology [ASCO], 2021). Almost 50% of the patients reported having consulted their GP about side effects and discomforts of cancer therapy, which is in line with some previous studies that this is one of the most common reasons for consulting a GP during cancer therapy (Duineveld et al., 2019; Lang et al., 2017). Other studies, however, show that psychosocial needs and organisational problems are also very important reasons for consultation during cancer therapy (Hoekstra, 2014; Roorda et al., 2012).

4.2 What patients expect of their GP regarding information on and offers of CAM procedures

Only one fifth of the GPs of the patients in our study knowingly offered CAM, which is considerably lower than the 60% of German GPs using CAM reported by Joos et al. (2011). Moreover, the prevalence of CAM usage in our study was much lower (14% after cancer diagnosis) than the 40%–60% otherwise reported in the literature for Germany (Horneber et al., 2012; Huebner et al., 2014; Linde et al., 2014; Paul et al., 2013). This can be explained by the fact that only methods prescribed or recommended by the GP were asked for. The use of CAM will continue to increase in the future and accordingly play an ever greater role in everyday clinical practice (Bozza et al., 2015; Canizares et al., 2017). Consistent with this, a large proportion of patients in our study stated that they would like to have more extensive offers of CM (71%) and AM (54%). In contrast to many previous publications (Bauer et al., 2018; Dubois et al., 2019; Keene et al., 2019; Krug et al., 2016; Molassiotis et al., 2005) which have shown a positive association between CAM usage and female gender or higher educational level, we did not find such an association in our study. The results of our survey showed that female patients and patients with a higher level of education neither received CAM offers from their GP more often nor did they have a significantly higher interest in CAM. In our survey, only 7% of patients before and 14% of patients after their cancer diagnosis actually received a CAM offer from their GP. Despite this doubling of the number, which points to the increasing importance of CAM after the cancer diagnosis (Buckner et al., 2018; Velicer & Ulrich, 2008), the rate of patients getting offers remains small.

Only a small proportion of our patients received a CAM offer by the GP's initiative, while three times as many patients received such an offer only upon request. In fact, many cancer patients are not asked by their physicians on interest in CAM (Davis et al., 2012; King et al., 2015; Sanford et al., 2019) and the conversation regarding CAM is in most cases initiated by patients (Tilburt et al., 2019). Yet many patients would like their doctor to initiate such a discussion (Shelley et al., 2009). This is also confirmed by the results of our study, as patients with a strong interest in CAM significantly more often wanted to be actively informed by their GP. The majority of cancer patients in Germany are dissatisfied with the information they receive from doctors on CAM, as the work of Bauer et al. (2018) shows. Presumably, this contributes to the fact that the internet is used as one of the most important source of information by patients (Bauer et al., 2018; Bozza et al., 2015; Labidi et al., 2020). In summary, these results suggest a discrepancy between doctor and patient regarding the communication on CAM. This discrepancy might be due to the fact that doctors assume a very low level of CAM usage by their patients and accordingly do not initiate a conversation (Shelley et al., 2009). Moreover, many doctors, not only in Germany, consider their knowledge of CAM to be insufficient (Dahlhaus et al., 2015; Pirotta et al., 2010; Shelley et al., 2009), although they see it as their duty to provide their patients with scientifically founded information on this topic (Dahlhaus et al., 2015). In particular, many patients believe that GPs have the best knowledge of CAM, which makes them a key player with regard to CAM in cancer care (Wortmann, 2016). This is also in line with the results of our survey, where a large majority agreed that their GP should have a comprehensive knowledge of CAM, not only on the benefits, side effects and interactions, but also on the costs and evidence of the procedures. Nevertheless, the apparent mismatch in doctor–patient communication regarding CAM may increase the effect that some patients conceal their CAM usage from their physicians (Wortmann, 2016). Training on communication about CAM in oncology may help improve doctor–patient exchange and to prevent safety-related problems in the use of CAM (Witt et al., 2020).

Consistent with some previous studies, the most common CAM procedures used by patients in our survey were vitamins and supplements taken orally or as injections/infusions (Huebner et al., 2014; Paul et al., 2013; Wortmann, 2016), followed by homeopathy. In almost 90% of the cases, the aim was to strengthen the body in general. These aims are in line with those reported in other studies (Bozza et al., 2015; Hierl et al., 2017; Huebner et al., 2014; Paul et al., 2013; Wortmann, 2016). In contrast to Gras et al. (2019), combating side effects played a subordinate role in our survey (only 20% of the patients). CAM procedures with low evidence like supplements (Rock et al., 2012; Vernieri et al., 2018) were used much more frequently than physical activity (25%) with strong scientific evidence, although according to many scientific studies the goal of strengthening the body overall could be achieved much better and more safely with sports (Bouillet et al., 2015; Buffart et al., 2017; Fuller et al., 2018; Stout et al., 2017). Likewise, GPs should also point towards a healthy nutrition and the option of nutritional education as a scientifically based intervention that can have a proven impact on the outcome of cancer (Aprile et al., 2021; Arends et al., 2017; Greenlee et al., 2019).

4.3 Patients' attitudes towards homeopathy, acupuncture and vitamin B17

Regarding homeopathy and acupuncture, a large majority of the patients in our study felt that the GP should recommend these procedures, even though a scientifically proven effect is lacking. In line with this, other studies also show that scientific evidence plays a rather subordinate role for many CAM users and that the experiences and advice of doctors as well as family and friends are more decisive (Oh et al., 2010; Verhoef et al., 2007, 2009). Furthermore, this result might also be attributed to the fact that most patients expect little to no side effects from these procedures (Stoneman et al., 2013; Zia et al., 2017). It should also be taken into account that one of the main reasons for using CAM is the more extensive time the doctor spends with the patient (Huebner et al., 2014).

With regard to vitamin B17, the statement that GPs should advise against vitamin B17 due to a lack of scientifically proven effect was most frequently agreed with. Along with this, the work of Bauer et al. (2018) shows that this method is most frequently advised against by oncologists due to the toxic side effects by hydrogen cyanide (Bromley et al., 2005; Sauer et al., 2015). Moreover, the majority of patients in our study thought that vitamin B17 should be recommended by the GP if the patient has no other therapy options left. Another German study shows that the majority of doctors in palliative care would also use AM for patients with cancer if there are no other treatment possibilities left (Conrad et al., 2014).

4.4 Limitations

There are several limitations to our study. Most importantly, our dataset is not representative in some regards. Out of a total of 180 respondents, two thirds were women and one third men. This may be due to the fact, that women are more interested in CAM (Bauer et al., 2018; Dubois et al., 2019; Keene et al., 2019; Krug et al., 2016; Molassiotis et al., 2005). In addition, some cancer entities such as colorectal and lung cancer were underrepresented. Other cancers such as leukaemia and lymphoma, on the other hand, were very strongly represented in this study, which is due to the distribution of a large part of the questionnaires in the haematological day clinic of a university hospital. As our survey was first intended to give a general overview of what cancer patients expect from their GP with regard to CAM, the different cancer stages of the respondents were not differentiated. However, the different cancer stages might have an impact on patients' expectations of their GP in general as well as with regard to CAM, so this should be further investigated in future studies. Furthermore, the questions in Parts 3, 4 and 5 of the questionnaire were only answered incompletely by many patients, so that the analytical evaluation of these questions was partly difficult due to small subsets.