Pre-diagnostic prescribing patterns in dyspnoea patients with as-yet-undiagnosed lung cancer: A longitudinal study of linked primary care and cancer registry data

Cancer Epidemiol. 2023 Oct:86:102429. doi: 10.1016/j.canep.2023.102429. Epub 2023 Jul 19.

Abstract

Introduction: Patients with as-yet undiagnosed lung cancer (LC) can present to primary care with non-specific symptoms such as dyspnoea, often in the context of pre-existing chronic obstructive pulmonary disease (COPD). Related medication prescriptions pre-diagnosis might represent opportunities for earlier diagnosis, but UK evidence is limited. Consequently, we explored prescribing patterns of relevant medications in patients who presented with dyspnoea in primary care and were subsequently diagnosed with LC.

Method: Linked primary care (Clinical Practice Research Datalink) and National Cancer Registry data were used to identify 5434 patients with incident LC within a year of a dyspnoea presentation in primary care between 2006 and 2016. Primary care prescriptions relevant to dyspnoea management were examined: antibiotics, inhaled medications, oral steroids, and opioid analgesics. Poisson regression models estimated monthly prescribing rates during the year pre-diagnosis. Variation by COPD status (52 % pre-existing, 36 % COPD-free, 12 % new-onset) was examined. Inflection points were identified indicating when prescribing rates changed from the background rate.

Results: 63 % of patients received 1 or more relevant prescriptions 1-12 months pre-diagnosis. Pre-existing COPD patients were most prescribed inhaled medications. COPD-free and new-onset COPD patients were most prescribed antibiotics. Most patients received 2 or more relevant prescriptions. Monthly prescribing rates of all medications increased towards time of diagnosis in all patient groups and were highest in pre-existing COPD patients. Increases in prescribing activity were observed earliest in pre-existing COPD patients 5 months pre-diagnosis for inhaled medications, antibiotics, and steroids, CONCLUSION: Results indicate that a diagnostic window of appreciable length exists for potential earlier LC diagnosis in some patients. Lung cancer diagnosis may be delayed if early symptoms are misattributed to COPD or other benign conditions.

Keywords: COPD; Diagnostic window; Lung cancer; Pre-diagnostic prescribing; Primary care.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Anti-Bacterial Agents / therapeutic use
  • Dyspnea* / diagnosis
  • Dyspnea* / drug therapy
  • Dyspnea* / etiology
  • Humans
  • Longitudinal Studies
  • Lung Neoplasms* / complications
  • Lung Neoplasms* / diagnosis
  • Lung Neoplasms* / drug therapy
  • Practice Patterns, Physicians'*
  • Primary Health Care
  • Pulmonary Disease, Chronic Obstructive / complications
  • Pulmonary Disease, Chronic Obstructive / diagnosis
  • Pulmonary Disease, Chronic Obstructive / drug therapy
  • Routinely Collected Health Data
  • Steroids / therapeutic use

Substances

  • Anti-Bacterial Agents
  • Steroids