Ietje Perfors

158 Chapter 6 Other studies evaluating primary care interventions after cancer diagnosis in the curative patient population indicate either positive effects on patient satisfaction 25,27,28 or no effects 31 and showed less ED visits in the older population. 32 These studies examined various interventions, which involved information provision using patient health records 25,27,28 or intensified primary care with the focus on GP 28 or on a primary care team. 32 The variety of interventions, different healthcare systems, or the use of self-developed questionnaires to measure patient satisfaction 25,27–29 might explain the more positive outcome as compared to our study. The higher number of ED visits among intervention patients was in contrast with our expectations. Reasons for ED consultations were mostly oncology- related and seemed unavoidable, for example ED visits because of fever during chemotherapy. Although the ED records did not provide clues for the increased ED use, it may be related to the fact that GPs referred cancer patients at a lower threshold because of study participation. Another explanation might be that patients in the intervention group have more co- morbidities. However, adjustment only slightly affected the estimates. This study has both strengths and limitations. A strength is the pragmatic approach and the implementation of the study in daily practice. Consequently, this pragmatic RCT adds to the scarce evidence on the real-life effects of involving a primary care team during and after curative cancer treatment, in a daily practice setting. Also, outcome measurements were aligned to individual patient’s cancer treatments, thereby enabling different cancer types to be included. Present study results might not be generalizable to all cancer patients who are to be treated with curative intent, since our study population might be a selection of patients who were positive towards primary care. Also, our intervention may have been prematurely implemented, as supported by the relatively high uptake but incorrect scheduling (TOC) and relatively high number of discontinued HON contacts. Future studies might benefit from following strategies to develop and evaluate a complex intervention as presented in the framework of the Medical Research Council 33 more strictly. This approach would require more elaborate pilot evaluations to optimize