Exploring the impact of comorbidities on cancer stage at diagnosis and 30 day mortality; a retrospective cohort study

Problem

NHS England has prioritised increasing the proportion of patients diagnosed early with cancer as part of the NHS Long Term Plan. However, this may be challenging, as the rising prevalence of chronic conditions may complicate the cancer diagnostic process. Here we investigate whether patients with pre-existing conditions are more likely to be diagnosed with late-stage cancer or die within 30 days of cancer diagnoses.

Approach

We used linked primary care (Clinical Practice Research Datalink), secondary care (Hospital Episode Statistics) and cancer registration (NCRAS) data. Patients diagnosed with any of 21 stage-able common cancers during 2012-2016 were included. A count of comorbidities was based on medical records for conditions included in the quality and outcome framework. We used logistic regression to investigate which patient groups (comorbidities, age, gender, smoking history and deprivation level) were more likely to be diagnosed at late-stage or die within 30 days of diagnosis.

Findings

119,667 patients were included. Preliminary results suggest that 8.7% (n=10,408) of patients died within 30 days of their diagnosis and 50.5% (48,060/95,149) of patients were diagnosed with late-stage cancer. There was evidence that stage and 30-day mortality were independently associated with age, gender, deprivation, and the presence of comorbidities (p<0.001), with male, older, and more deprived patients more likely to be diagnosed at late-stage or die within 30 days of diagnosis. Patients with several comorbidities were less likely to be diagnosed at late-stage or die within 30 days. For individual cancers there was no evidence that co-morbidities were associated with late-stage and only for X and Y cancers is there any evidence of detriment in 30-day mortality.

Consequences

Although some patient groups were more likely to be diagnosed with late-stage cancer and die within 30 days, patients with multi-morbidity were more likely to be diagnosed early. Potentially, regular monitoring of a chronic condition may provide opportunities to detect cancer earlier. As multi-morbidity has previously been linked to poorer survival chances, it may be that specific comorbidity types or morbidity burden better explain the relationship between multi-morbidity and cancer outcomes.