Use of primary care during the year before childhood cancer diagnosis: a nationwide population-based matched comparative study

PLoS One. 2013;8(3):e59098. doi: 10.1371/journal.pone.0059098. Epub 2013 Mar 12.


Objective: Childhood cancer is rare and symptoms tend to be unspecific and vague. Using the utilization of health care services as a proxy for symptoms, the present study seeks to determine when early symptoms of childhood cancer are seen in general practice.

Methods: A population-based matched comparative study was conducted using nationwide registry data. As cases, all children in Denmark below 16 years of age (N = 1,278) diagnosed with cancer (Jan 2002-Dec 2008) were included. As controls, 10 children per case matched on gender and date of birth (N = 12,780) were randomly selected. The utilization of primary health care services (daytime contacts, out-of-hours contacts and diagnostic procedures) during the year preceding diagnosis/index date was measured for cases and controls.

Results: During the six months before diagnosis, children with cancer used primary care more than the control cohort. This excess use grew consistently and steadily towards the time of diagnosis with an IRR = 3.19 (95%CI: 2.99-3.39) (p<0.0001) during the last three months before diagnosis. Children with Central Nervous System (CNS) tumours had more contacts than other children during the entire study period. The use of practice-based diagnostic tests and the number of out-of-hours contacts began to increase four to five months before cancer diagnosis.

Conclusions: The study shows that excess health care use, a proxy for symptoms of childhood cancer, occurs months before the diagnosis is established. Children with lymphoma, bone tumour or other solid tumours had higher consultation rates than the controls in the last five months before diagnosis, whereas children with CNS tumour had higher consultation rates in all twelve months before diagnosis. More knowledge about early symptoms and the diagnostic pathway for childhood cancer would be clinically relevant.

Publication types

  • Comparative Study
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Case-Control Studies
  • Child
  • Child, Preschool
  • Denmark
  • Female
  • General Practice / statistics & numerical data*
  • Humans
  • Infant
  • Infant, Newborn
  • Male
  • Neoplasms / diagnosis*
  • Population Surveillance
  • Primary Health Care*
  • Referral and Consultation
  • Registries
  • Time Factors

Grant support

The study was carried out at The Research Unit for General Practice in Aarhus, Denmark and funded by the Child Cancer Foundation, the Novo Nordisk foundation, the Danish Cancer society and the PLU Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.