Alarm symptoms in early diagnosis of cancer in primary care: cohort study using General Practice Research Database

doi:10.1136/bmj.39171.637106.AE

BMJ 2007;334:1040 (19 May), doi:10.1136/bmj.39171.637106.AE (published 10 May 2007)

Research

Alarm symptoms in early diagnosis of cancer in primary care: cohort study using General Practice Research Database

Roger Jones, Wolfson professor of general practice¹, Radoslav Latinovic, database manager², Judith Charlton, research assistant², Martin C Gulliford, senior lecturer in public health²

¹ Department of General Practice and Primary Care, Division of Health and Social Care Research, King's College London School of Medicine at Guy's, King's College and St Thomas' Hospitals, London SE1 6SP, ² Department of Public Health Sciences, Division of Health and Social Care Research, King's College London School of Medicine at Guy's, King's College and St Thomas' Hospitals, London SE1 3QD

Correspondence to: R Jones roger.jones{at}kcl.ac.uk

Objective To evaluate the association between alarm symptomsand the subsequent diagnosis of cancer in a large populationbased study in primary care.

Design Cohort study.

Setting UK General Practice Research Database.

Patients 762 325 patients aged 15 years and older, registeredwith 128 general practices between 1994 and 2000. First occurrencesof haematuria, haemoptysis, dysphagia, and rectal bleeding wereidentified in patients with no previous cancer diagnosis.

Main outcome measure Positive predictive value of first occurrenceof haematuria, haemoptysis, dysphagia, or rectal bleeding fordiagnoses of neoplasms of the urinary tract, respiratory tract,oesophagus, or colon and rectum during three years after symptomonset. Likelihood ratio and sensitivity were also estimated.

Results 11 108 first occurrences of haematuria were associatedwith 472 new diagnoses of urinary tract cancers in men and 162in women, giving overall three year positive predictive valuesof 7.4% (95% confidence interval 6.8% to 8.1%) in men and 3.4%(2.9% to 4.0%) in women. After 4812 new episodes of haemoptysis,220 diagnoses of respiratory tract cancer were made in men (positivepredictive value 7.5%, 6.6% to 8.5%) and 81 in women (4.3%,3.4% to 5.3%). After 5999 new diagnoses of dysphagia, 150 diagnosesof oesophageal cancer were made in men (positive predictivevalue 5.7%, 4.9% to 6.7%) and 81 in women (2.4%, 1.9 to 3.0%).After 15 289 episodes of rectal bleeding, 184 diagnoses of colorectalcancer were made in men (positive predictive value 2.4%, 2.1%to 2.8%) and 154 in women (2.0%, 1.7% to 2.3%). Predictive valuesincreased with age and were strikingly high, for example, inmen with haemoptysis aged 75-84 (17.1%, 13.5% to 21.1%) andin men with dysphagia aged 65-74 (9.0%, 6.8% to 11.7%).

Conclusion New onset of alarm symptoms is associated with anincreased likelihood of a diagnosis of cancer, especially inmen and in people aged over 65. These data provide support forthe early evaluation of alarm symptoms in an attempt to identifyunderlying cancers at an earlier and more amenable stage.

Add to CiteULike CiteULike Add to Complore Complore Add to Connotea Connotea Add to Del.icio.us Del.icio.us Add to Digg Digg Add to Reddit Reddit Add to StumbleUpon StumbleUpon Add to Technorati Technorati What's this?

Relevant Articles

Commentary: Diagnosing ovarian cancer—more problems than answers: Robin Fox
BMJ 2009 339: b3233. [Full Text]

Alarm symptoms and identification of non-cancer diagnoses in primary care: cohort study: Roger Jones, Judith Charlton, Radoslav Latinovic, and Martin C Gulliford
BMJ 2009 339: b3094. [Abstract] [Full Text] [PDF]

US cancer groups highlight symptoms of early ovarian cancer: Janice Hopkins Tanne
BMJ 2007 334: 1290-1291. [Extract] [Full Text] [PDF]

Alarming results: Douglas Kamerow
BMJ 2007 334: 0. [Extract] [Full Text]

Alarm symptoms in primary care: Robin Fox and John Fletcher
BMJ 2007 334: 1013-1014. [Extract] [Full Text] [PDF]

Risk in primary care of colorectal cancer from new onset rectal bleeding: 10 year prospective study: Jennifer du Toit, William Hamilton, and Kevin Barraclough
BMJ 2006 333: 69-70. [Abstract] [Full Text] [PDF]

Is the two week rule for cancer referrals working?: Roger Jones, Greg Rubin, and Pali Hungin
BMJ 2001 322: 1555-1556. [Extract] [Full Text] [PDF]

Rectal bleeding and colorectal cancer in general practice: diagnostic study: Hans Wauters, Viviane Van Casteren, and Frank Buntinx
BMJ 2000 321: 998-999. [Full Text] [PDF]

Rectal bleeding: prevalence and consultation behaviour: Ann Crosland and Roger Jones
BMJ 1995 311: 486-488. [Abstract] [Full Text]

This article has been cited by other articles:

Thirumaran, M, Sundar, R, Sutcliffe, I M, Currie, D C (2009). Is investigation of patients with haemoptysis and normal chest radiograph justified?. Thorax 64: 854-856 [Abstract] [Full text]
Fox, R. (2009). Commentary: Diagnosing ovarian cancer--more problems than answers. BMJ 339: b3233-b3233 [Full text]
Jones, R., Charlton, J., Latinovic, R., Gulliford, M. C (2009). Alarm symptoms and identification of non-cancer diagnoses in primary care: cohort study. BMJ 339: b3094-b3094 [Abstract] [Full text]
Jones, R (2008). Primary care research and clinical practice: gastroenterology. Postgrad. Med. J. 84: 454-458 [Abstract] [Full text]
Soler, J.-K., Okkes, I., Wood, M., Lamberts, H. (2008). The coming of age of ICPC: celebrating the 21st birthday of the International Classification of Primary Care. Fam Pract 25: 312-317 [Abstract] [Full text]
(2007). Symptoms That Predict an Increased Likelihood of Cancer. JWatch General 2007: 3-3 [Full text]
(2007). When Do Patients Need Referral for Cancer Work-Up?. JWatch Emergency Med. 2007: 3-3 [Full text]
Fox, R., Fletcher, J. (2007). Alarm symptoms in primary care. BMJ 334: 1013-1014 [Full text]

Rapid Responses:

Read all Rapid Responses

Misleading conclusions and very little utility: Dr. Harsha Kumar H N
bmj.com, 12 May 2007 [Full text]
Oral self examination in the early detection of oropharyngeal cancer: Eduardo M. Curbeira-Hernandez. MDD, et al.
bmj.com, 15 May 2007 [Full text]
Likelihood of cancer is insufficient to define referral policy for alarm symptoms.: David Carr
bmj.com, 18 May 2007 [Full text]
Alarm symptoms must be assessed in the context of age-specific cancer incidence: Mark N Upton
bmj.com, 19 May 2007 [Full text]
Re: Alarm symptoms must be assessed in the context of age-specific cancer incidence: LS Lewis
bmj.com, 20 May 2007 [Full text]
Re: Re: Alarm symptoms must be assessed in the context of age-specific cancer incidence: Mark N Upton
bmj.com, 21 May 2007 [Full text]
Re: Re: Re: Alarm symptoms must be assessed in the context of age-specific cancer incidence: LS Lewis
bmj.com, 22 May 2007 [Full text]
Re: Re: Re: Re: Alarm symptoms must be assessed in the context of age-specific cancer incidence: Mark N. Upton
bmj.com, 22 May 2007 [Full text]
Re: Re: Re: Re: Re: Alarm symptoms must be assessed in the context of age-specific cancer incidence: LS Lewis
bmj.com, 22 May 2007 [Full text]
Re: Re: Re: Re: Re: Re: Alarm symptoms must be assessed in the context of age-specific cancer incidence: Mark N. Upton
bmj.com, 23 May 2007 [Full text]
It�s all in the recording: quality versus quantity: Carl Heneghan
bmj.com, 25 May 2007 [Full text]
Re: Re: Re: Re: Re: Re: Re: Alarm symptoms must be assessed in the context of age-specific cancer incidence: LS Lewis
bmj.com, 25 May 2007 [Full text]
When is an alarm symptom not an alarm symptom?: David Kernick
bmj.com, 29 May 2007 [Full text]
Dysphonia as an alarming symptom in patients aged 40 or more years old.: Alejandro D�az-Gonz�lez MD, et al.
bmj.com, 2 Jun 2007 [Full text]
Early Diagnosis of Colorectal Cancer: Brian Gareth Ellis, et al.
bmj.com, 7 Jun 2007 [Full text]
Alarm symptoms in primary care: Further evidence: Knut A Holtedahl
bmj.com, 1 Oct 2007 [Full text]