Evidence of increasing mortality with longer diagnostic intervals for five common cancers: A cohort study in primary care

https://doi.org/10.1016/j.ejca.2013.01.025Get rights and content

Abstract

Background

Early diagnosis is considered a key factor in improving the outcomes in cancer therapy; it remains unclear, however, whether long pre-diagnostic patient pathways influence clinical outcomes negatively. The aim of this study was to assess the association between the length of the diagnostic interval and the five-year mortality for the five most common cancers in Denmark while addressing known biases.

Methods

A total of 1128 patients with colorectal, lung, melanoma skin, breast or prostate cancer were included in a prospective, population-based study in a Danish county. The diagnostic interval was defined as the time from the first presentation of symptoms in primary care till the date of diagnosis. Each type of cancer was analysed separately and combined, and all analyses were stratified according to the general practitioner’s (GP’s) interpretation of the presenting symptoms. We used conditional logistic regression to estimate five-year mortality odds ratios as a function of the diagnostic interval using restricted cubic splines and adjusting for comorbidity, age, sex and type of cancer.

Results

We found increasing mortality with longer diagnostic intervals among the approximately 40% of the patients who presented in primary care with symptoms suggestive of cancer or any other serious illness. In the same group, very short diagnostic intervals were also associated with increased mortality. Patients presenting with vague symptoms not directly related to cancer or any other serious illness had longer diagnostic intervals and the same survival probability as those who presented with cancer suspicious/serious symptoms. For the former, we found no statistically significant association between the length of the diagnostic interval and mortality.

Conclusion

In full coherence with clinical logic, the healthcare system instigates prompt investigation of seriously ill patients. This likely explains the counter-intuitive findings of high mortality with short diagnostic intervals; but it does not explain the increasing mortality with longer diagnostic intervals. Thus, the study provides further evidence for the hypothesis that the length of the diagnostic interval affects mortality negatively.

Introduction

Over the past few decades, a steady stream of comparative studies of Nordic and European cancer registries has documented survival deficits and an unfavourable stage distribution among Danish cancer patients.1, 2, 3, 4 The Danish government has responded by striving to reduce delays in diagnosis and treatment. Yet, until recently, there was no evidence of any benefit of expediting diagnosis and treatment in symptomatic cancer patients.

Given the complex nature of diagnosing cancer, it is a challenging task to design a study that validly compares cancer patients with short and long delays. Observational studies frequently show the opposite of what we expect: patients with short delays in diagnosis and treatment have higher mortality than the rest.5, 6, 7, 8 Many studies illustrating this paradox take the results to show that there is no association between delay and mortality and find the results reassuring.9, 10, 11 However, by premising their findings on non-random observations, they may have reached wrong conclusions.

We have recently called attention to the fundamental analytical problem of confounding by indication in observational studies caused by differentiated clinical triage.12, 13 This bias arises when general practitioners (GPs) and subsequently diagnostic centres give priority to seriously ill patients who may have higher inherent mortality (the ‘sick-quick’ group) and are comparatively more reluctant to expedite patients with less obvious symptoms of cancer (the ‘low risk–slow diagnosis’ group).14, 15, 16, 17 Most studies lack information on what triggered the GP either to refer immediately or to adopt a watchful waiting approach, and they may hence be reporting biased results.

The aim of this study was to assess the association between the length of the diagnostic interval and the five-year mortality for the five most common cancers in Denmark while addressing the above methodological and analytical issues.

Section snippets

Materials and methods

We conducted a cohort study with overall five-year mortality as the primary outcome and the length of the diagnostic interval (defined as the time from first presentation of symptoms in primary care to the date of diagnosis) as the exposure variable.

Results

We identified 1543 incident colorectal, lung, melanoma skin, breast and prostate cancer patients above the age of 17. In 248 (16%) of the cases, the GPs did not participate or complete the questionnaire (Fig. 1). We compared patients with participating GPs to patients with non-participating GPs and found no differences with respect to age, gender, comorbidity, emergency admission, previous cancer, tumour stage or five-year mortality after discharge. However, compared with patients with

Discussion

In patients with alarm or any serious symptoms, mortality grew the longer the diagnostic intervals in those 40% from this group who experienced the longest delays. In patients with alarm or any serious symptoms, having very short diagnostic intervals was also associated with a high mortality. Inversely, in patients presenting with vague symptoms, we saw much longer diagnostic intervals, the same survival probability and an opposite, concave trend between the length of the diagnostic interval

Conclusion

This study challenges the conclusions of many previous studies and provides evidence for the hypothesis that the length of the diagnostic interval affects mortality negatively. The study thereby supports efforts to shorten the clinical pathway such as standardised cancer packages that guarantee a fast track from referral to diagnosis and treatment in patients with suspected cancer.

Contributors

MLT was responsible for data acquisition from public and medical databases, performed the statistical analyses and wrote the paper. She is the guarantor. MF supervised the statistical analyses and designed programmes in Stata (version 11) for generating cubic splines with specific reference values (centercsplines.ado) and estimates with standard errors for linear combinations (calcest.ado) – both applied in Fig. 2. RPH designed the GP and patient questionnaires and was responsible for this part

Funding

The study was supported by grants from the Danish Cancer Society, the Danish Agency for Science, Technology and Innovation, and Aarhus University. The sponsors were not involved in any part of the study.

Conflict of interest statement

None of the authors have received any support from or have any relationship with companies that might have an interest in the submitted work, but FO is chairman of the Danish Cancer Society, which is a charity-based and independent national patient organisation.

Ethical approval

The study was approved by the Danish Data Protection Agency and the Danish National Board of Health. According to the Committees on Biomedical Research Ethics in the Central Denmark Region, the Act on the Biomedical Research Ethics Committee System and the Processing of Biomedical Research Projects does not apply to this project. All authors had full access to data and take full responsibility for the accuracy of the data analysis.

Acknowledgements

The authors thank Professor William Hamilton and MD, PhD Marianne Djernes Lautrup for fruitful discussions about clinical practice in primary and secondary care, which helped shape the arguments of this paper. We thank also the Department of Clinical Epidemiology for contributing data on comorbidity to the study.

References (45)

  • H. Moller et al.

    A visual summary of the EUROCARE-4 results: a UK perspective

    Br J Cancer

    (2009)
  • R.D. Neal

    Do diagnostic delays in cancer matter?

    Br J Cancer

    (2009)
  • J.K. Olsson et al.

    Timeliness of care in patients with lung cancer: a systematic review

    Thorax

    (2009)
  • K.S. Rupassara et al.

    A paradox explained? Patients with delayed diagnosis of symptomatic colorectal cancer have good prognosis

    Colorectal Dis

    (2006)
  • R. Sainsbury et al.

    Effect on survival of delays in referral of patients with breast-cancer symptoms: a retrospective analysis

    Lancet

    (1999)
  • L.H. Iversen et al.

    Therapeutic delay reduces survival of rectal cancer but not of colonic cancer

    Br J Surg

    (2009)
  • M.L. Tørring et al.

    Time to diagnosis and mortality in colorectal cancer: a cohort study in primary care

    Br J Cancer

    (2011)
  • M. Nylenna

    Diagnosing cancer in general practice. When is cancer suspected?

    Br Med J (Clin Res Ed)

    (1986)
  • M. Nylenna

    Diagnosing cancer in general practice. From suspicion to certainty

    Br Med J (Clin Res Ed)

    (1986)
  • W. Hamilton

    Cancer diagnosis in primary care

    Br J Gen Pract

    (2010)
  • K.J. Rothman

    Epidemiology: an introduction

    (2002)
  • NORDCAN [homepage on the Internet]. Association of the Nordic Cancer Registries. Available from:...
  • Cited by (192)

    View all citing articles on Scopus
    View full text