The study by Westman et al. [1] clearly shows that (early) accessibility to palliative care (PC) is far from optimal. There are no data to assume that this is better in other (European) countries than Sweden. So, indeed improvement in this respect is mandatory. A number of studies confirm the importance of early access to PC [2–4]. It adds to improvement in both patient care and quality of life. (Early) cooperation between oncologists and palliative care workers may hence improve clinical decision making (CDM) by patients regarding for instance starting, continuing or stopping cancer-directed treatment. Avorn [5] recently suggested that in CDM, choices made by patients are often less rational than we believe. I recently treated a 28-year-old woman with metastatic melanoma with a considerable and symptomatic tumor burden and elevated LDH, BRAF unmutated with the combination of ipilimumab and nivolumab followed by nivolumab. She was referred to a palliative care worker, mainly because of pain, before beginning treatment. After acceptable relief of the pain treatment options were thoroughly discussed between her and her family, the palliative care worker and her general physician. While thus being fully informed and aware of the low chance and often slow rate of response and fair chance of adverse events, she claimed that she was young and raises two young children and refraining from treatment was no option. Eventually, she survived 8 months of which 7 in acceptable quality of life.

In this case, the decision was not only made after sharing all aspects with the patient, but to some extent by another factor I have to admit: compassion and identification with the patient by her doctor. Avorn [5] might designate this situation in a provocative way (in the US) as “the questionable gray zone in which choices are driven by revenues of the oncologist and institutions, depending on such treatments for their economic success.

From the paper by Westman et al. [1], we may conclude that early access to PC occurs infrequently. I believe that early participation of PC workers in CDM contributes to shared decision making (SDM). These days, we ought to inform patients and their companions optimally and give them the possible options. The no (-cancer-directed-) treatment option requires more attention, as oncologists seem to be in a default treatment plus position. After a time of considerations (time out procedure), patients ultimately decide. No need mentioning that patients as well decide on both rational and less rational arguments. If SDM has taken place correctly, most patients feel better informed and have less regrets regarding their treatment decisions in cancer [6, 7]. It is likely that while applying SDM appropriately, all factors involved in CDM become more visible, both in doctors and in patients. Such a transparency might reduce the chance of disproportionate influence by factors such as recent experiences (last-case bias) and financial drives [5].

Another issue is (non-) compliance to our recommendations [5]. It seems less likely to be a problem if we more carefully considered all patients’ objections and pay more attention to their desires during the SDM process, before recommending any treatment. In addition, the early contribution of a PC worker in the team may be helpful in assessing the patient’s desires. In this respect, special attention can be given to patients with a lower socioeconomic status, who may have lower levels of compliance and hence sometimes lower survival, as proposed by Cavalli–Bjorkman [8]. In the patient described above, the process of SDM gave her extension of life with a fair quality. Non-compliance with my treatment advice, however, would have been highly acceptable for me. As the treatment given is allowed through guidelines, but extremely costly, withholding it would increase my popularity among health care authorities and insurance companies, but impair my own quality of life.

In summary, early access to PC, application of SDM procedures, offering all cancer-directed treatment versus supportive care options, and a time out procedure, especially in patients who passed the curative phase, are likely to improve CDM. PC workers joining the multidisciplinary consultation may provide additional perspectives in the eventual advice to the patient, regarding palliative cancer treatment or supportive care only, but this has to be proven first. If so, arguments exist to improve the low accessibility to PC, as described by Westman.