Secondary care providers currently involved in cancer follow-up care identified perceived barriers and facilitators to formal involvement of GP practices. Among the barriers were the perceived inability to perform structured follow-up in GP practices and risk of re-referral delays. Additionally, concerns about the lack of specialized oncological knowledge among GPs were voiced. At the same time, the accessibility, familiarity, and contextual knowledge of GPs were reported as clear benefits to formal GP involvement.

When asked about requirements for future formal involvement of GPs, many interviewees suggested improvements to hospital logistics to better accommodate shared care between hospital and primary care. Additionally, formal training for GPs, sufficient case-load, and proper staffing, remuneration, and time for GPs were identified as essential prerequisites for follow-up in primary care. Finally, they suggested that formal GP involvement should first be implemented for frail older patients and for prevalent cancer types, whereas patients with a high risk of recurrences or a complicated course were suggested to remain in hospital-based follow-up. Individualized follow-up plans were mentioned as a necessity.

Our finding that specialists voiced concerns about GPs’ ability to provide structured follow-up was also reported by GPs in another study [17]. This is interesting because GP practices in the Netherlands are currently responsible for managing structured care for patients with chronic diseases like diabetes, respiratory problems, and cardiovascular risk management, and they perform this particularly well. This care is typically provided by trained nurse practitioners under GP supervision. One could argue that the fact that many cancer patients have chronic comorbidities and frequently visit their GPs [8, 26,27,28], integrating cancer follow-up into this type of care is a viable approach. However, it is plausible that specialists and GPs view cancer as different from the aforementioned chronic diseases, because of its complex pathophysiology and rapidly developing knowledge base. This could also be reflected in our finding that specialists felt that GPs lack specialized oncological knowledge. Earlier studies reported similar concerns voiced by GPs; they felt that their knowledge of follow-up routines, test interpretation, side effects, and long-term treatment effects for cancer was not up-to-date and they might lack the required experience [17, 29]. However, a qualitative study involving both primary and secondary care providers found that additional specific training for GPs was not deemed necessary in a shared care setting, with clear protocols and effective communication being considered more important [30].

Several studies have shown that cancer follow-up in primary care is feasible and safe, showing no differences in recurrence detection and survival [4, 6]. Yet, secondary care providers in our study expressed concerns about re-referral delays in case of a suspected recurrence, a challenge also highlighted in other studies [15]. Earlier studies showed that a significant proportion of recurrences are detected symptomatically by GPs in between follow-up visits [31]. In addition, meta-analyses comparing intensive and less-intensive follow-up for breast or colorectal cancer showed no differences in overall survival [32,33,34]. One might even hypothesize that regularly scheduled follow-up visits could also cause delays, either from the patient or the doctor, as both may wait for the next planned appointment.

A clear benefit of a GP involvement in follow-up care reported by specialists was the accessibility and familiarity of GPs, potentially reducing anxiety and stress around follow-up examinations and generally improving patient well-being. This finding is also supported by studies among GPs and patients, which highlight improved continuity of care, increased psychosocial attention, better accessibility, and greater familiarity [15, 17, 18, 35,36,37]. GPs generally feel well-equipped to provide psychosocial care in cancer follow-up, and with their long-term patient relationships and deep understanding of individual goals and contexts, they are well-equipped to further personalize follow-up care. This may explain why there are also studies showing that the majority of general practitioners felt proficient in performing cancer follow-up care tasks [38].

In addition, the COVID-19 pandemic has led to significant changes in clinical practice, with telehealth playing a crucial role in improving communication between primary and secondary care [39,40,41]. Communication issues remain a challenge, as does the current shortage of general practitioners, which hinders the formal involvement of GPs.

A shared-care model, leveraging the expertise of medical specialists and the accessibility of GPs, may prove to be an effective way to increase the GPs’ role in cancer follow-up. It can enhance the personalization, effectiveness, and efficiency of oncological follow-up. Rather than a standardized, disease-focused follow-up protocol, it would allow for care tailored to patient-specific goals and priorities. With the growing evidence supporting less intensive follow-up, this approach appears feasible. According to medical specialists as well as GPs and patients [17, 18], key components include the individualized care with input from specialists, GPs, and patients, as well as clear re-referral pathways and communication between care providers. Overcoming communication barriers and addressing GPs’ lack of confidence through training and rapid specialist consultations are also crucial.

In practice, using colorectal cancer as an example, shared care could potentially begin immediately after treatment to ease the transition, with the frequency and intensity of follow-up tailored to the patient's needs, preferences, and recurrence risk. An individualized follow-up plan should be developed, ideally through a three-way discussion, although this may not always be feasible. A visit to the GP early in the follow-up period should be scheduled to re-establish contact and discuss patient preferences. For instance, the first year might include alternating appointments, such as GP visits at 3 and 9 months and specialist follow-ups at 6 and 12 months, depending on the agreed-upon plan. The recent SCORE RCT showed that this follow-up schedule with shared care was non-inferior to current hospital care and acceptable to patients, with no differences reported in quality of life or recurrence detection [11, 42]. GPs could manage CEA monitoring, perform physical examinations as necessary, and assess the patient’s well-being. After the colonoscopy or CT scan at 1 year, GP care could potentially expand, depending on the patient’s risk profile and preferences, as suggested in this study. In such a shared-care model, patients, GPs, and specialists each have their roles. An individualized follow-up care plan can help patients take a more active role in their follow-up, for instance, by enabling them to schedule appointments themselves and take greater ownership of their care. To ensure continuity of care, however, it may be beneficial to have a hospital case manager or dedicated point of contact in primary care to coordinate and organize the alternating appointments between the GP and the hospital. Introducing a new care model requires time for adaptation; therefore, it is crucial to prepare patients early and clearly communicate the transition, so they eventually perceive it as the new standard of care.

This is the first study to explicitly examine the opinions of secondary care providers, on the formal involvement of the GP and, in particular on possible implementation strategies. We therefore consider this study a relevant addition to the discussion about possible future oncological follow-up strategies of care.

A possible limitation is that the interviews in this study concerned the views of secondary care providers based on their experiences with care in hospitals. There may be a lack of knowledge among them about the current working methods in general practice.

Moreover, adopting a deductive approach with predefined categories could have restricted the flexibility of our analysis, leading us to potentially overlook relevant information that did not conform to these categories. Nevertheless, we believe that while deductive coding may have somewhat limited our analysis, its integration with inductive coding has allowed us to delve deeper into these predefined topics. We think this approach effectively addresses our specific research question and ensures that our findings extend beyond simply reiterating existing information.

Lastly, in this study, we focused exclusively on breast cancer and colorectal cancer, although other types of cancer would also be of interest. We examined two of the five most common cancer types; however, while prostate cancer was a strong candidate, we also considered types with somewhat more complex follow-ups.