Elsevier

The Lancet

Volume 353, Issue 9159, 3 April 1999, Pages 1119-1126
The Lancet

Fast track — Articles
Influence of delay on survival in patients with breast cancer: a systematic review

https://doi.org/10.1016/S0140-6736(99)02143-1Get rights and content

Summary

Background

Most patients with breast cancer are detected after symptoms occur rather than through screening. The impact on survival of delays between the onset of symptoms and the start of treatment is controversial and cannot be studied in randomised controlled trials. We did a systematic review of observational studies (worldwide) of duration of symptoms and survival.

Methods

We identified 87 studies (101 954 patients) with direct data linking delay (including delay by patients) and survival. We classified studies for analysis by type of data in the original reports: category I studies had actual 5-year survival data (38 studies, 53912 patients); category II used actuarial or multivariate analyses (21 studies, 25 102 patients); and category III was all other types of data (28 studies, 22940 patients). We tested the main hypothesis that longer delays would be associated with lower survival, and a secondary hypothesis that longer delays were associated with more advanced stage, which would account for lower survival.

Findings

In category I studies, patients with delays of 3 months or more had 12% lower 5-year survival than those with shorter delays (odds ratio for death 1·47 [95% CI 1·42–1·53]) and those with delays of 3–6 months had 7% lower survival than those with shorter delays (1·24 [1·17–1·30]). In category II, 13 of 14 studies with unrestricted samples showed a significant adverse relation between longer delays and survival, whereas four of five studies of only patients with operable disease showed no significant relation. In category III, all three studies with unrestricted samples supported the primary hypothesis. The 13 informative studies showed that longer delays were associated with more advanced stage. In studies that controlled for stage, longer delay was not associated with shorter survival when the effect of stage on survival was taken into account.

Interpretation

Delays of 3–6 months are associated with lower survival. These effects cannot be accounted for by lead-time bias. Efforts should be made to keep delays by patients and providers to a minimum.

Introduction

Detection and treatment of cancer at an early stage improves the prospects for long-term survival. In many types of cancer, the outlook for patients with small, localised tumours is much better than that of patients with advanced or metastatic disease. If cancers are diagnosed early, microscopic spread of disease may not have occurred. Excision of the primary tumour may therefore be curative. For some cancers, evidence shows that diagnosis and treatment before symptoms appear is associated with improved survival rates. The influence on survival of delays in diagnosis and treatment after onset of symptoms has, however, been much less clear.

Whether longer delays in patients who have symptoms are associated with worse outcomes cannot be studied in randomised controlled trials. Detailed analysis of observational data is therefore necessary, taking into account, wherever possible, potential confounding factors. Among patients with breast cancer, uncertainty persists about the possible impact of delays on survival, despite many different reports.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86 The question, however, remains important. Breast cancer is a common disease with more than 30 000 new cases per year in the UK.87, 88 Most (>75%) patients present after symptoms occur, rather than through the National Health Service breast-screening programme. In about a third of these patients, the time between the first symptom and start of treatment is more than 3 months89 and, in about a quarter of patients, more than 6 months.25, 43, 48, 51 Although age-standardised mortality ratios have fallen in the past decade90, 91, 92 breast cancer remains a major cause of death for women in the UK.

Our primary hypothesis was that longer delays between onset of symptoms and diagnosis or treatment are associated with worse survival rates in patients with breast cancer and that this association is not simply attributable to lead-time bias. We tested also a secondary hypothesis that patients with longer duration of symptoms would generally present with more advanced disease and that this relation between delay and stage would account for the poorer survival anticipated in patients with longer delays. These hypotheses were based on the findings from a study of patients attending the Guy's Hospital Breast Unit, London, UK,89 which we consequently excluded from this review.

This study forms part of a larger report commissioned by the National Health Service Cancer Research and Development Programme.93 A review of factors that influence delays by patients or providers is published in this issue of The Lancet (see page 1127).94

Section snippets

Data search

We did the review according to principles laid down to protect against bias.95, 96 Details of the methods are reported elsewhere, including the full search strategy and lists of all duplicate and excluded reports.93 We started the review in June, 1996, and completed it in March, 1998. We attempted to identify all observational studies (worldwide) related to women presenting with symptoms of breast cancer, with direct data linking duration of symptoms and survival. We did not limit the year of

Results

We identified 87 original studies that included information directly linking duration of symptoms (delay) and survival. The studies involved a total of 101 954 patients and were published between 1907 and 1996 (table 1). One report related to two separate studies of different cohorts.

Discussion

The combined results from each of the three categories support our primary hypothesis that longer delays are associated with worse survival rates. In category I the association between longer delay and worse survival was significant among studies with unrestricted samples and among those confined to patients with operable disease. In category II, 13 of 14 studies with unrestricted samples showed a significant adverse relation between longer delays and survival. Four of five studies confined to

References (103)

  • C Haagensen et al.

    Carcinoma of the breast I: results of treatment

    Ann Surg

    (1942)
  • B Nohrman

    Cancer of the breast: clinical study of 1042 cases treated at Radiumhemmet between 1936 and 1941

    Acta Radiol Suppl

    (1949)
  • C Eberbach

    Carcinoma of the breast

    Wisconsin Med J

    (1949)
  • M Shimkin et al.

    Cancer of the breast: analysis of frequency, distribution and mortality at the University of California Hospital. 1918–1947 inclusive

    Surg Gynecol Obstet

    (1952)
  • W Harnett

    The relation between delay in treatment of cancer and survival rate

    Br J Cancer

    (1953)
  • W Maclean

    Carcinoma of the breast: 10 year survey at Winnipeg General Hospital

    J Intl Coll Surg

    (1953)
  • M Miller et al.

    Some observations concerned with carcinoma of the breast

    Am J Roentgenol

    (1954)
  • G Robbins et al.

    The significance of delay in relation to prognosis of patients with primary operable breast cancer

    Cancer

    (1957)
  • General Registrar

    Statistical review of England and Wales for the year 1952 (supplement on Cancer)

    (1957)
  • N Treves et al.

    A report of 549 cases of breast cancer in women 35 years of age or younger

    Surg Gynecol Obstet

    (1958)
  • T Anglem et al.

    Cancer of the breast and biological predeterminism: a re-evaluation

    Cancer

    (1961)
  • C Lalanne

    Taux d'accroissement et prognostic des tumeurs malignes du sein

    Acta Union Intl Contra Cancrum

    (1962)
  • P Rigby-Jones

    Prognosis of malignant tumours of the breast in relation to rate of growth and axillary lymph node involvement as observed clinically

    Acta Union Intl Contra Cancrum

    (1962)
  • HJG Bloom

    The influence of delay on the natural history and prognosis of breast cancer: a study of cases followed for 5–20 years

    Surgery

    (1965)
  • J DeVitt

    The clinical stages of breast cancer: what do they mean?

    Can Med Assoc J

    (1967)
  • The Registrar General's statistical review of England and Wales for the year 1961: supplement on cancer

    (1967)
  • Donegan W. Diagnosis in cancer of the breast. London,...
  • P Peltokallio et al.

    Results of breast cancer treatment

    Acta Chir Scand

    (1969)
  • B Sheridan et al.

    The effects of delay in treatment on survival rates in carcinoma of the breast

    Med J Aust

    (1971)
  • R Handley

    Observations and thoughts on cancer of the breast

    Proc R Soc Med

    (1972)
  • R Hermann et al.

    Results of conservative operations for breast cancer

    Arch Surg

    (1985)
  • K Cummings et al.

    The effects of behavioral and biological factors on survival from breast cancer

    Cancer Detect Prev

    (1983)
  • C Simmons et al.

    Cancer of the breast: end results Massachusetts General Hospital, 1924, 1925 and 1926

    N Engl J Med

    (1934)
  • B Shore

    Carcinoma of the breast: report of four hundred eighteen cases treated at St Luke's Hospital from 1922–1933

    Surg Gynecol Obstet

    (1940)
  • B Hoopes et al.

    The Halsted radical mastectomy: 5 year results in 246 consecutive operations at same clinic

    Surgery

    (1942)
  • J Hawkins

    Evaluation of breast-cancer therapy as a guide to control programs

    J Natl Cancer Inst

    (1944)
  • W Nicolson et al.

    Carcinoma of the breast

    Ann Surg

    (1948)
  • C Haagensen et al.

    Carcinoma of the breast: III results of treatment 1935–1942

    Ann Surg

    (1951)
  • D Smithers et al.

    Cancer of the breast: a review

    Br J Radiol

    (1952)
  • I Macdonald et al.

    Prognosis of mammary carcinoma in young women

    Cancer

    (1956)
  • C Moore et al.

    Carcinoma of the breast: extent of original disease related to end results

    Arch Surg

    (1957)
  • T White

    Breast cancer in women thirty years old or less

    N W Med Seattle

    (1960)
  • J Berkson

    Prognosis of malignant tumours of the breast: a review of recent experiences of the Mayo clinic

    Acta Union Intl Contra Cancrum

    (1962)
  • D Brinkley et al.

    A 15 year follow-up study of patients treated for carcinoma of the breast

    Br J Radiol

    (1968)
  • M Elwood et al.

    Delay in diagnosis and long term survival in breast cancer

    BMJ

    (1980)
  • P Dohrman et al.

    Symptom duration, tumor staging and survival in patients with carcinoma of the breast

    Surg Gynecol Obstet

    (1982)
  • J Feldman

    The effects of patient delay and symptoms other than a lump on survival in breast cancer

    Cancer

    (1983)
  • M Charlson

    Delay in the treatment of carcinoma of the breast

    Surg Gynecol Obstet

    (1985)
  • S Vernon et al.

    Ethnicity, survival and delay in seeking treatment for symptoms of breast cancer

    Cancer

    (1985)
  • C Huguley et al.

    Breast self-examination and survival from breast cancer

    Cancer

    (1988)
  • Cited by (0)

    View full text