Spectrum bias: Difference between revisions
Jump to navigation
Jump to search
imported>Robert Badgett (This is content I wrote at WikiPedia) |
imported>James A. Flippin (Clarify) |
||
| Line 1: | Line 1: | ||
Initially identified in 1978<ref name="pmid692598">{{cite journal |author=Ransohoff DF, Feinstein AR |title=Problems of spectrum and bias in evaluating the efficacy of diagnostic tests |journal=N. Engl. J. Med. |volume=299 |issue=17 |pages=926-30 |year=1978 |pmid=692598|id=PMID 692598 |doi=}}</ref>, spectrum bias | Initially identified in 1978<ref name="pmid692598">{{cite journal |author=Ransohoff DF, Feinstein AR |title=Problems of spectrum and bias in evaluating the efficacy of diagnostic tests |journal=N. Engl. J. Med. |volume=299 |issue=17 |pages=926-30 |year=1978 |pmid=692598|id=PMID 692598 |doi=}}</ref>, '''spectrum bias''' is a common flaw in studies of diagnostic tests in which the control group is comprised of people who are certainly disease-free and the test group is comprised of people known to have the disease. This leads to an overestimation of the [[sensitivity (tests)|sensitivity]] and [[specificity (tests)|specificity]] of the test.<ref name="pmid692598"/><ref name="pmid14695644">{{cite journal |author=Goehring C, Perrier A, Morabia A |title=Spectrum bias: a quantitative and graphical analysis of the variability of medical diagnostic test performance |journal=Statistics in medicine |volume=23 |issue=1 |pages=125-35 |year=2004 |pmid=14695644|id=PMID 14695644|doi=10.1002/sim.1591}}</ref> Examples are initial overestimates of the [[sensitivity (tests)|sensitivity]] and [[specificity (tests)|specificity]] of the carcinoembryonic antigen<ref name="pmid3510056">{{cite journal |author=Fletcher RH |title=Carcinoembryonic antigen |journal=Ann. Intern. Med. |volume=104 |issue=1 |pages=66-73 |year=1986 |pmid=3510056|id=PMID 3510056 |doi=}}</ref> and urinary dipstick tests.<ref name="pmid1605428">{{cite journal |author=Lachs MS, Nachamkin I, Edelstein PH, Goldman J, Feinstein AR, Schwartz JS |title=Spectrum bias in the evaluation of diagnostic tests: lessons from the rapid dipstick test for urinary tract infection |journal=Ann. Intern. Med. |volume=117 |issue=2 |pages=135-40 |year=1992 |pmid=1605428 | id = PMID 1605428 |doi=}}</ref> | ||
In such instances, the sample is biased to particular areas of the spectrum of disease (disease free, slightly diseased, moderately diseased, extremely diseased), namely the disease free and extremely diseased. While diagnostic tests among such populations may appear to be highly specific and sensitive, when used on the population at large, they could in fact miss patients who are only mildly ill or early in the stages of the disease's progression. | |||
If properly analyzed, recognition of heterogeneity of subgroups can lead to insights about the test's performance in varying populations.<ref name="pmid12353947">{{cite journal |author=Mulherin SA, Miller WC |title=Spectrum bias or spectrum effect? Subgroup variation in diagnostic test evaluation |journal=Ann. Intern. Med. |volume=137 |issue=7 |pages=598-602 |year=2002 |pmid=12353947|id=PMID 12353947 |doi=}}</ref> | If properly analyzed, recognition of heterogeneity of subgroups can lead to insights about the test's performance in varying populations.<ref name="pmid12353947">{{cite journal |author=Mulherin SA, Miller WC |title=Spectrum bias or spectrum effect? Subgroup variation in diagnostic test evaluation |journal=Ann. Intern. Med. |volume=137 |issue=7 |pages=598-602 |year=2002 |pmid=12353947|id=PMID 12353947 |doi=}}</ref> | ||
Revision as of 16:48, 19 June 2007
Initially identified in 1978[1], spectrum bias is a common flaw in studies of diagnostic tests in which the control group is comprised of people who are certainly disease-free and the test group is comprised of people known to have the disease. This leads to an overestimation of the sensitivity and specificity of the test.[1][2] Examples are initial overestimates of the sensitivity and specificity of the carcinoembryonic antigen[3] and urinary dipstick tests.[4]
In such instances, the sample is biased to particular areas of the spectrum of disease (disease free, slightly diseased, moderately diseased, extremely diseased), namely the disease free and extremely diseased. While diagnostic tests among such populations may appear to be highly specific and sensitive, when used on the population at large, they could in fact miss patients who are only mildly ill or early in the stages of the disease's progression.
If properly analyzed, recognition of heterogeneity of subgroups can lead to insights about the test's performance in varying populations.[5]
References
- ↑ 1.0 1.1 Ransohoff DF, Feinstein AR (1978). "Problems of spectrum and bias in evaluating the efficacy of diagnostic tests". N. Engl. J. Med. 299 (17): 926-30. PMID 692598. PMID 692598. [e]
- ↑ Goehring C, Perrier A, Morabia A (2004). "Spectrum bias: a quantitative and graphical analysis of the variability of medical diagnostic test performance". Statistics in medicine 23 (1): 125-35. DOI:10.1002/sim.1591. PMID 14695644. PMID 14695644. Research Blogging.
- ↑ Fletcher RH (1986). "Carcinoembryonic antigen". Ann. Intern. Med. 104 (1): 66-73. PMID 3510056. PMID 3510056. [e]
- ↑ Lachs MS, Nachamkin I, Edelstein PH, Goldman J, Feinstein AR, Schwartz JS (1992). "Spectrum bias in the evaluation of diagnostic tests: lessons from the rapid dipstick test for urinary tract infection". Ann. Intern. Med. 117 (2): 135-40. PMID 1605428. PMID 1605428. [e]
- ↑ Mulherin SA, Miller WC (2002). "Spectrum bias or spectrum effect? Subgroup variation in diagnostic test evaluation". Ann. Intern. Med. 137 (7): 598-602. PMID 12353947. PMID 12353947. [e]