Abstract
Background
Rural dwellers with colorectal cancer have poorer outcomes than their urban counterparts. The reasons why are not known but are likely to be complex and be determined by an interplay between geography and health service organization. By comparing the associations related to travel-time to primary and secondary healthcare facilities in two neighbouring countries, Denmark and Scotland, we aimed to shed light on potential mechanisms.
Methods
Analysis was based on two comprehensive cohorts of patients diagnosed with colorectal cancer in Denmark (2010-16) and Scotland (2007-14). Associations between travel-time and cancer pathway intervals, tumour stage at diagnosis and one-year mortality were analysed using generalised linear models. Travel-time was modelled using restricted cubic splines for each country and combined. Adjustments were made for key confounders.
Results
Travel-time to key healthcare facilities influenced the diagnostic experience and outcomes of CRC patients from Scotland and Denmark to some extent differently. The longest travel-times to a specialised hospital appeared to afford the most rapid secondary care interval, whereas moderate travel-times to hospital (about 20-60 minutes) appeared to impact on later stage and greater one-year mortality in Scotland, but not in Denmark. A U-shaped association was seen between travel-time to the GP and one year-mortality.
Conclusions
This is the first international data-linkage study to explore how different national geographies and health service structures may determine cancer outcomes. Future research should compare more countries and more cancer sites and evaluate the impact and implications of differences in national health service organisation.
Rural dwellers with colorectal cancer have poorer outcomes than their urban counterparts. The reasons why are not known but are likely to be complex and be determined by an interplay between geography and health service organization. By comparing the associations related to travel-time to primary and secondary healthcare facilities in two neighbouring countries, Denmark and Scotland, we aimed to shed light on potential mechanisms.
Methods
Analysis was based on two comprehensive cohorts of patients diagnosed with colorectal cancer in Denmark (2010-16) and Scotland (2007-14). Associations between travel-time and cancer pathway intervals, tumour stage at diagnosis and one-year mortality were analysed using generalised linear models. Travel-time was modelled using restricted cubic splines for each country and combined. Adjustments were made for key confounders.
Results
Travel-time to key healthcare facilities influenced the diagnostic experience and outcomes of CRC patients from Scotland and Denmark to some extent differently. The longest travel-times to a specialised hospital appeared to afford the most rapid secondary care interval, whereas moderate travel-times to hospital (about 20-60 minutes) appeared to impact on later stage and greater one-year mortality in Scotland, but not in Denmark. A U-shaped association was seen between travel-time to the GP and one year-mortality.
Conclusions
This is the first international data-linkage study to explore how different national geographies and health service structures may determine cancer outcomes. Future research should compare more countries and more cancer sites and evaluate the impact and implications of differences in national health service organisation.
| Original language | English |
|---|---|
| Article number | 102004 |
| Number of pages | 9 |
| Journal | Cancer Epidemiology |
| Volume | 74 |
| Early online date | 19 Aug 2021 |
| DOIs | |
| Publication status | Published - 31 Oct 2021 |
Bibliographical note
FUNDINGThis project was conducted without external funding and with the support of the Data Safe Haven (DaSH) of the University of Aberdeen and Statistics Denmark.
Acknowledgements
We wish to acknowledge Dr Vicky Munro and Dr Joanne Lumsden from the Aberdeen Centre of Data Health Science who worked with the authors in preparing the Scottish Data for Transfer to Statistic Denmark. We also acknowledge Statistics Denmark for hosting the combined dataset for analysis and Kaare Rud Flarup for
help with data management. We also acknowledge Professor Lesley Anderson and Dr Lisa Iversen for providing comments on the manuscript.
Keywords
- Colorectal Neoplasms
- Travel; Diagnosis
- Primary Health Care
- Time-to-Treatment
- Neoplasm Staging
- Mortality
Access to Document
- Murchie_etal_CE_Geographic_Variation_In_AAM
© 2021. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/