In this study the level of shared decision-making among paediatric patients with SCD was assessed after groupwise and individual training of the paediatricians, and introducing an SDM tool for the children. This led to a fairly high observed level of SDM, while perceived levels of SDM were high.

When comparing these results with our earlier baseline measurement in a very similar patient population with the same paediatricians [16], the observed level of SDM increased substantially from a median of 25 to 50% on the OPTION-scale. This translates into an improvement from a ‘low’ to a ‘moderate’ effort to engage patients/parents in the decision-making process. These effects are similar to other areas in medicine, like breast cancer [30], and vascular surgery [31, 32]. While the SDM-promoting training and tools have likely contributed to this improvement, a growing awareness amongst the team of paediatric haematologists about the added value of SDM and their willingness to apply SDM in clinical encounters may also have stimulated their effort to involve patients/caregivers in decision-making processes.

Our study design did not allow us to identify which of the interventions led to the improvements in level of SDM, Still, we suggest that active involvement of participants through training and decision support tools is essential to increase the level of SDM during clinical encounters, as studies in other medical realms pointed in the same direction [30, 32, 33].

The ‘3 good questions’ cards address several aspects of SDM, like the paediatricians’ effort to make the patient aware that a health-related decision needs to be made for which there are various treatment options available, deliberation about various treatment options and the risk and benefits of each of these options and can serve as a reminder for the paediatrician to elicit the child and parents’ preferences, worries, and expectations.

Obviously, the level of child involvement is a gliding scale, depending on its age. The higher the child’s age, the less parents may be involved in the decision-making process. Based on the audio-recordings we observed effort from paediatricians to involve children younger than 12. Even though caregivers of these children are legally entitled to be the designated substitute (shared) decision-makers on behalf of the child. Within this context and since our study focus on the level of SDM between paediatric haematologists on the one hand and children and their caregiver on the other, we included all variations of shared decision-making between patient/caregiver on the one hand and paediatricians on the other. Other studies in the Netherlands seem to indicate that involving young children in a decision-making process is not standard procedure [34].

An important observation was that the teach back technique, that has proven to be useful to verify the patients and parents’ understanding of the information provided [35], was rarely observed, even though it was thoroughly discussed during both the individual- and team SDM training and patients and caregivers felt well-informed by the paediatrician as the SDM-Q-Doc results showed. We did not, however, assess the patients’ knowledge they had retained about the information they received from the clinician. To understand the goal and added value of SDM, introducing and explaining the ‘3 good questions’ cards may suit the pivotal role the nursing staff play in a paediatric haematology clinic. Their contribution could help further improve the level of SDM [36,37,38,39,40].

Patients and their parents perceived more SDM than the paediatricians did, especially in the higher range of OPTION scores. Apparently, when OPTION scores increased patients and parents perceived more participation in the decision-making process, whereas paediatricians seemed to have become more aware of their limited SDM skills.

Regarding consultation duration, a longer duration might allow for a better shared decision-making process. However we did not find a difference in consultation duration before and after the intervention. Moreover, literature shows that applying SDM does not require longer consultation duration [41].

The OPTION-5 instrument has not yet been validated for a triadic decision-making process and has been limitedly used in paediatric settings [23, 31, 42, 43]. However, given its strong psychometric properties and the general applicability of SDM as a collaborative decision-making technique, we deem it a valuable method of measuring SDM in a paediatric setting.

Another limitation of our study is the small number of participating paediatric haematologists as well as the single centre setting, which limits the generalizability of the results. However, there are no reasons to suspect that our results are not applicable to other paediatric centers or diseases.

Furthermore, despite the SDM-Q-9 scales’ strong internal consistency [44], it cannot be ruled out that the SDM-Q-9 is sensitive to child and parents’ vulnerability towards authority and/or expert halo bias [42], so that patients score satisfaction with their physician and care rather than the level of perceived SDM. Also, the possibility of paediatricians’ unawareness of their own bias regarding the child’s and parents’ levels of health literacy could have impacted their SDM-Q-DOC scores, since language- and cultural barriers are known factors to impact the physicians’ effort in engaging the child and/or caregiver in the decision-making process [45, 46]. Finally, social desirability bias cannot be ruled out because of the researcher’s presence in the same room with patient and or caregiver when the SDM-Q-9 questionnaire was completed.

The results of our post-intervention study are promising. However, due to the small sample size further research is needed, preferably in a larger population, to explore the impact interventions like groupwise and individual training have on the level of SDM. Also, studies are needed to explore the impact of implementing the 3 Good Questions decision support tool in clinical practice. These effects should be studied both separately as well as in combination in order to determine if, and to what extent, these interventions may strengthen each other.