|Abstract / Summary
Sarcoma patients often experience delay before a diagnosis is made, which may affect patient experience, prognosis and treatment outcome. Delay before diagnosis has also been reported for many other cancer forms, and the problem has received increased attention in recent years. Denmark implemented Cancer Patient Pathways (CPPs) in 2008/09 to abate the problem, and these pathways describe the ideal diagnostic pathway for suspected cancer patients, hereunder criteria for referral, suggested diagnostic work-up and time limits for each phase. The effects of this implementation for patients suspected of a sarcoma have not been investigated, and the diagnostic journey leading up to referral to a specialised sarcoma centre is sparsely described. To ensure timely diagnosis and optimal management of sarcoma patients it is important to identify where waiting time happens and possible areas of improvement. Overall, this thesis sought to investigate the immediate effects of the Danish CPP for sarcomas and describe the diagnostic journey of suspected sarcoma patients in the new steady state after CPP implementation.
In study I, we examined the effect of the CPP on time intervals at the Aarhus Sarcoma Centre (ASC) two years before and two years after implementation, among all patients referred because of a suspicion of malignancy from hospitals outside the catchment area of Aarhus University hospital (AUH). Further, we described reasons for delay among patients exceeding time limits. In study II, we examined the presence and predictive values of alarm symptoms/signs defined as CPP referral criteria in the same study population. In study III, we looked closer at 64 sarcoma patients found in study II that had been referred to the ASC after an unplanned excision or biopsy performed elsewhere. In study IV we described the time intervals of the entire patient journey from first symptom to diagnosis and examined whether waiting times were affected by presenting signs and symptoms. In study V, we compared time intervals and proportion of malignancies between patients referred to the CPP at the ASC after initial investigations (MRI/CT/histology) performed at local hospitals other than AUH and patients referred from the Aarhus area without these investigations. The study population for studies IV and V consisted of a prospectively collected population of all consecutive patients referred to the CPP for sarcoma during a one-year period (1st of September 2014 to 31st of August 2015).
In study I, we found that time intervals at the ASC had been reduced over the four-year period, most likely due to the CPP implementation. Most delays in 2010 were caused by a need for supplementary diagnostics, indicating that the adherence to time limits are vulnerable to deviations from the standard diagnostic programme. In study II, we showed that defined alarm symptoms were frequent among referred patients and predictive of a sarcoma; however, one third of sarcoma patients were not referred due to alarm symptoms but diagnosed incidentally after imaging or
unplanned surgery. Study III showed that 50% of the patients referred after unplanned surgery in 52
study II had small superficial tumours, and 61% had not presented with alarm symptoms and thus fell outside of CPP referral criteria. Study IV showed that the main part of the total interval from first symptom to diagnosis for suspected sarcoma patients (median 155 days) could be attributed to the patient interval (median 54 days), followed by the local hospital interval (median 26.5 days). The presenting signs and symptoms modified some of the time intervals, with presence of alarm symptoms and GP suspicion shortening health system intervals. Study V showed that patients investigated with MRI and/or CT and/or histology at local hospitals had significantly longer median time intervals compared to patients referred without these investigations, with an estimated age- adjusted difference in median diagnostic interval of 41 days. The proportion of malignancies was significantly higher in the group referred after MRI/CT compared to the group referred directly (37.5% vs 14.3%).
In conclusion, we showed that the Danish CPP for sarcomas has accelerated the diagnostic process at the ASC and that the defined alarm symptoms are predictive of a sarcoma and prevalent among referred patients. However, a large proportion of patients fall outside of the alarm symptom referral criteria, and patients presenting with nonspecific symptoms have longer waiting times. A CPP can only benefit the patients referred to it, and initial GP suspicion of malignancy is thus crucial for an earlier diagnosis. Finally, the CPP time limits do not apply to the diagnostic process happening before referral to a CPP and performing imaging investigations at local hospitals significantly lengthens time intervals. This problem needs to be addressed, possibly by providing easier access to imaging for GPs or inclusion of the diagnostic process at local hospitals as a part of the sarcoma CPP with appurtenant time limits.