Measurement of clinical delay intervals among younger adults with colorectal cancer using health administrative data: a population-based analysis

Abstract

BackgroundClinical delays may be important contributors to outcomes among younger adults (<50 years) with colorectal cancer (CRC). We aimed to describe delay intervals for younger adults with CRC using health administrative data to understand drivers of delay in this population.MethodsThis was a population-based study of adults <50 diagnosed with CRC in Ontario, Canada from 2003 to 2018. Using administrative code-based algorithms (including billing codes), we identified four time points along the pathway to treatment—first presentation with a CRC-related symptom, first investigation, diagnosis date and treatment start. Intervals between these time points were calculated. Multivariable quantile regression was performed to explore associations between patient and disease factors with the median length of each interval.Results6853 patients aged 15–49 were diagnosed with CRC and met the inclusion criteria. Males comprised 52% of the cohort, the median age was 45 years (IQR 40–47), and 25% had stage IV disease. The median time from presentation to treatment start (overall interval) was 109 days (IQR 55–218). Time between presentation and first investigation was short (median 5 days), as was time between diagnosis and treatment start (median 23 days). The greatest component of delay occurred between first investigation and diagnosis (median 78 days). Women, patients with distal tumours, and patients with earlier stage disease had significantly longer overall intervals.ConclusionsSome younger CRC patients experience prolonged times from presentation to treatment, and time between first investigation to diagnosis was an important contributor. Access to endoscopy may be a target for intervention.

Analysis of Time to Treatment and Survival Among Adults Younger Than 50 Years of Age With Colorectal Cancer in Canada

“…20 While it has been used in CRC, breast cancer, oral cancer, pancreatic cancer, and we have previously described its application to young patients with CRC, it has only been directly compared against patient charts for oral cancer. 9,20,21,23 Our study does not capture the interval between symptom onset and presentation, 37 and it is possible young adults have delays to presentation that contribute to worse outcomes. It has been recognized that delays in cancer care can also be related to health behavior and psychological factors such as patient knowledge, reluctance to seek help, fear or denial, and financial concerns.…”

“…We have described our application of these methods to measure delay intervals among 6853 Ontario adults younger than age 50 years with CRC, and we were able to assign a date of presentation to more than 97% of the cohort. 9 Briefly, we searched for the earliest clinical encounter for CRC-related signs or symptoms, up to a maximum of 18 months prior to diagnosis.…”

“…7 Several authors have suggested lack of screening, low suspicion of CRC from physicians, and poor access to care leading to excessively delayed treatment could be contributory. 5,7,8 To evaluate delay in this population, we previously examined the time from first presentation to treatment initiation 9 ; adults younger than 50 years of age in Ontario, Canada, waited a median of 109 days. One-quarter of patients waited longer than 218 days.…”

“…One-quarter of patients waited longer than 218 days. 9 Given that delay is proposed as a potential mechanism for worse outcomes in young people with CRC, there is insufficient understanding of the relationship between time to treatment (or other intervals) and survival in these patients. The literature specifically examining adults younger than age 50 years is sparse.…”

“…To evaluate delay in this population, we previously examined the time from first presentation to treatment initiation; adults younger than 50 years of age in Ontario, Canada, waited a median of 109 days. One-quarter of patients waited longer than 218 days .…”