Odds ratio: Difference between revisions
Jump to navigation
Jump to search
imported>Richard D. Gill m (removed "essentially") |
mNo edit summary |
||
| (9 intermediate revisions by 3 users not shown) | |||
| Line 1: | Line 1: | ||
{{subpages}} | {{subpages}} | ||
The '''odds ratio''' is a technical term often used in statistics, and especially in medical statistics and epidemiology. The odds ratio is the ratio of the relative incidence (odds) of a target disorder in the experimental group relative to the relative incidence in a control group. It reflects how the risk of having a particular disorder is influenced by the treatment. An odds ratio of 1 means that there is no benefit of treatment compared to the control group.<ref name="urlOdds and odds ratio">{{cite web |url=http://www.jr2.ox.ac.uk/bandolier/booth/glossary/Odds.html |title=Odds and odds ratio |author=Anonymous |authorlink= |coauthors= |date= |format= |work= |publisher=Bandolier |pages= |language= |archiveurl= |archivedate= |quote= |accessdate=}}</ref> | The '''odds ratio''' is a technical term often used in [[statistics]], and especially in medical statistics and [[epidemiology]]. The odds ratio is the ratio of the relative incidence (odds) of a target disorder in the experimental group relative to the relative incidence in a control group. It reflects how the risk of having a particular disorder is influenced by the treatment. An odds ratio of 1 means that there is no benefit of treatment compared to the control group.<ref name="urlOdds and odds ratio">{{cite web |url=http://www.jr2.ox.ac.uk/bandolier/booth/glossary/Odds.html |title=Odds and odds ratio |author=Anonymous |authorlink= |coauthors= |date= |format= |work= |publisher=Bandolier |pages= |language= |archiveurl= |archivedate= |quote= |accessdate=}}</ref> | ||
The odds ratio is a difficult concept for non mathematicians and recommendations for how to teach its use are available.<ref name="pmid18181004">{{cite journal |author=Prasad K | The odds ratio is a difficult concept for non mathematicians, and recommendations for how to teach its use are available.<ref name="pmid18181004">{{cite journal |author=Prasad K ''et al.''|title=Tips for teachers of evidence-based medicine: understanding odds ratios and their relationship to risk ratios |journal=J Gen Intern Med |volume=23 |pages=635–40 |year=2008 |pmid=18181004 |doi=10.1007/s11606-007-0453-4 |url=http://dx.doi.org/10.1007/s11606-007-0453-4 |issn=}}</ref> | ||
For a mathematician, the relative incidence of a disorder would be called the odds on having the disorder. An odds ratio is the ratio between the [[odds]] on some event under two circumstances. | For a mathematician, the relative incidence of a disorder would be called the 'odds on having the disorder'. An ''odds ratio'' is the ratio between the [[odds]] on some event under two circumstances. | ||
{| class="wikitable" align="right" | |||
|+ Effect of t'ment on disorder | |||
! rowspan="2"| !! colspan="2"|Disorder | |||
|- | |||
! Yes!!No | |||
|- | |||
| Experimental treatment|| <i>A</i>||<i>B</i> | |||
|- | |||
| Control treatment|| <i>C</i>||<i>D</i> | |||
|} | |||
If we have <i>A</i> individuals with the disorder and <i>B</i> without in the experimental group, and <i>C</i> individuals with the disorder and <i>D</i> without in the control group, the empirical odds ratio is | |||
:<math>Odds\text{ } ratio = \frac{A * D}{B * C}</math> | :<math>Odds\text{ } ratio = \frac{A * D}{B * C}</math> | ||
==Example== | ==Example== | ||
This example is from the Titanic (example from Power<ref name="pmid">{{cite journal |author=Power M |title=Resource reviews |journal=Evidence-based Medicine |volume=13 | This example is from the Titanic (example from Power<ref name="pmid">{{cite journal |author=Power M |title=Resource reviews |journal=Evidence-based Medicine |volume=13 |pages=92 |year=2008 |pmid=18515638 |doi= |url=http://ebm.bmj.com/cgi/content/full/13/3/92 |issn=}}</ref>): | ||
{| class="wikitable" align="right" | {| class="wikitable" align="right" | ||
| Line 38: | Line 50: | ||
The odds on a male passenger surviving the Titanic disaster was 1 to 5 against surviving. The odds on a female passenger surviving the Titanic disaster was 2 to 1 in favour of surviving. The odds ratio for survival between men and women is (1/5) / (2/1) = 1/10 | The odds on a male passenger surviving the Titanic disaster was 1 to 5 against surviving. The odds on a female passenger surviving the Titanic disaster was 2 to 1 in favour of surviving. The odds ratio for survival between men and women is (1/5) / (2/1) = 1/10 | ||
Rather than classifying the passengers on the Titanic as male or female, and comparing their odds on survival, one could classify passengers as survivors or not survivors, and compare the odds on being male or female. The odds ratio for sex (male versus female) between survivors and non-survivors is also 0.10. This follows from the algebraic identity <math>(AD/BC)=(A/B)/(C/D)=(A/C)/(B/D)</math>. | Rather than classifying the passengers on the Titanic as male or female, and comparing their odds on survival, one could classify passengers as survivors or not survivors, and compare the odds on being male or female. The odds ratio for sex (male versus female) between survivors and non-survivors is also 0.10. This follows from the algebraic identity <math>(AD/BC)=(A/B)/(C/D)=(A/C)/(B/D)</math>. | ||
The identity between the odds ratios is much used in epidemiology, especially in retrospective case-control studies. Consider for example the relation between lung cancer and exposure to asbestos. Instead of sampling from the two populations of people who are exposed to asbestos and people who have not been exposed to asbestos, and studying the odds ratio for lung cancer between the two populations defined by exposure status, one could just as well sample from the two populations of people who got lung cancer and people who did not get lung cancer, and study the odds ratio for asbestos-exposure between the two populations defined by health status. This is extremely valuable when studying | The identity between the odds ratios is much used in epidemiology, especially in retrospective case-control studies. Consider for example the relation between lung cancer and exposure to asbestos. Instead of sampling from the two populations of people who are exposed to asbestos and people who have not been exposed to asbestos, and studying the odds ratio for lung cancer between the two populations defined by exposure status, one could just as well sample from the two populations of people who got lung cancer and people who did not get lung cancer, and study the odds ratio for asbestos-exposure between the two populations defined by health status. This is extremely valuable when studying illnesses which are rather rare, even in the population exposed to the risk factor of interest. | ||
==Interpretation== | ==Interpretation== | ||
The odds ratio is generally used to measure the association between a risk factor and disease. However, using the odds ratio to measure the ability of a risk factor to diagnose disease is problematic.<ref name="pmid2213074">{{cite journal |author=Boyko EJ, Alderman BW |title=The use of risk factors in medical diagnosis: opportunities and cautions |journal=J Clin Epidemiol |volume=43 | The odds ratio is generally used to measure the association between a risk factor and disease. However, using the odds ratio to measure the ability of a risk factor to diagnose disease is problematic.<ref name="pmid2213074">{{cite journal |author=Boyko EJ, Alderman BW |title=The use of risk factors in medical diagnosis: opportunities and cautions |journal=J Clin Epidemiol |volume=43 |pages=851–8 |year=1990 |pmid=2213074 |doi= |url=http://linkinghub.elsevier.com/retrieve/pii/0895-4356(90)90068-Z |issn=}}</ref> The odds ratio should be at least 16 to have reasonable diagnostic ability.<ref name="pmid15105181">{{cite journal |author=Pepe MS ''et al.'' |title=Limitations of the odds ratio in gauging the performance of a diagnostic, prognostic, or screening marker |journal=Am J Epidemiol|volume=159 |pages=882–90 |year=2004 |pmid=15105181 |url=http://aje.oxfordjournals.org/cgi/pmidlookup?view=long&pmid=15105181 }}</ref> | ||
The odds ratio is similar to the [[relative risk ratio]]. The two ratios will be numerically similar if the rates in the two groups being compared are both similar and both less than 20% to 30%.<ref name="pmid7730891">{{cite journal |author=Sinclair JC, Bracken MB |title=Clinically useful measures of effect in binary analyses of randomized trials |journal=J Clin Epidemiol |volume=47 | The odds ratio is similar to the [[relative risk ratio]]. The two ratios will be numerically similar if the rates in the two groups being compared are both similar and both less than 20% to 30%.<ref name="pmid7730891">{{cite journal |author=Sinclair JC, Bracken MB |title=Clinically useful measures of effect in binary analyses of randomized trials |journal=J Clin Epidemiol |volume=47 |pages=881–9 |year=1994 |pmid=7730891 |url=http://linkinghub.elsevier.com/retrieve/pii/0895-4356(94)90191-0 |issn=}}</ref><ref name="pmid9804732">{{cite journal |author=Altman DG, Deeks JJ, Sackett DL |title=Odds ratios should be avoided when events are common |journal=BMJ |volume=317 |issue=7168 |pages=1318 |year=1998 |month=November |pmid=9804732 |pmc=1114216 |doi= |url=http://bmj.com/cgi/pmidlookup?view=long&pmid=9804732 |issn=}}</ref><ref name="pmid12954046">{{cite journal |author=Page J, Attia J |title=Using Bayes' nomogram to help interpret odds ratios |journal=ACP J Club |volume=139 |pages=A11–2 |year=2003 |pmid=12954046 |url=http://www.acpjc.org/Content/139/2/Issue/ACPJC-2003-139-2-A11.htm |issn=}} [http://www.acpjc.org/Content/139/2/Issue/ACPJC-2003-139-2-A11.htm Helpful chart]</ref> | ||
The odds ratio may be the most stable ratio across different prevalences.<ref name="pmid11914297">{{cite journal |author=Furukawa TA | The odds ratio may be the most stable ratio across different prevalences.<ref name="pmid11914297">{{cite journal |author=Furukawa TA ''et al.'' |title=Can we individualize the 'number needed to treat'? An empirical study of summary effect measures in meta-analyses |journal=Int J Epidemiol |volume=31 |pages=72–6 |year=2002 |pmid=11914297 |doi= |url=http://ije.oxfordjournals.org/cgi/pmidlookup?view=long&pmid=11914297 |issn=}}</ref> | ||
The odds ratio may be used to derive the [[number needed to treat]]:<ref name="pmid11914297" | The odds ratio may be used to derive the [[number needed to treat]]:<ref name="pmid11914297" /><ref name="pmid9139558">{{cite journal |author=McQuay HJ, Moore RA |title=Using numerical results from systematic reviews in clinical practice |journal=Ann Intern Med |volume=126|pages=712–20 |year=1997 |pmid=9139558 |url=http://www.annals.org/cgi/content/full/126/9/712 |issn=}}</ref> | ||
For odds ratios less than 1:<ref name="pmid9139558"/> | For odds ratios less than 1:<ref name="pmid9139558"/> | ||
| Line 63: | Line 74: | ||
[[Category:Mathematics Workgroup]] | [[Category:Mathematics Workgroup]] | ||
[[Category:Mathematics Content]] | [[Category:Mathematics Content]] | ||
[[Category:Mathematics tag]] | [[Category:Mathematics tag]][[Category:Suggestion Bot Tag]] | ||
Latest revision as of 06:00, 28 September 2024
The odds ratio is a technical term often used in statistics, and especially in medical statistics and epidemiology. The odds ratio is the ratio of the relative incidence (odds) of a target disorder in the experimental group relative to the relative incidence in a control group. It reflects how the risk of having a particular disorder is influenced by the treatment. An odds ratio of 1 means that there is no benefit of treatment compared to the control group.[1]
The odds ratio is a difficult concept for non mathematicians, and recommendations for how to teach its use are available.[2]
For a mathematician, the relative incidence of a disorder would be called the 'odds on having the disorder'. An odds ratio is the ratio between the odds on some event under two circumstances.
| Disorder | ||
|---|---|---|
| Yes | No | |
| Experimental treatment | A | B |
| Control treatment | C | D |
If we have A individuals with the disorder and B without in the experimental group, and C individuals with the disorder and D without in the control group, the empirical odds ratio is
Example
This example is from the Titanic (example from Power[3]):
| Survival | ||
|---|---|---|
| Yes | No | |
| Males | 142 | 709 |
| Females | 308 | 154 |
Male passengers:
142 survived, 709 died
- Odds of survival = 142/709 = 0.20
- Probability (risk or chance) of survival = 142/(142+709) = 17%
Female passengers:
308 survived, 154 died
- Odds of survival = 308/154 = 2.00
- Probability (risk or chance) of survival = 308/(308+154) = 67%
Comparison:
- Odds ratio (OR) for survival = 0.20/2.00 = 0.10
- Relative risk (RR) for survival = 17%/67% = 0.25
The odds on a male passenger surviving the Titanic disaster was 1 to 5 against surviving. The odds on a female passenger surviving the Titanic disaster was 2 to 1 in favour of surviving. The odds ratio for survival between men and women is (1/5) / (2/1) = 1/10
Rather than classifying the passengers on the Titanic as male or female, and comparing their odds on survival, one could classify passengers as survivors or not survivors, and compare the odds on being male or female. The odds ratio for sex (male versus female) between survivors and non-survivors is also 0.10. This follows from the algebraic identity .
The identity between the odds ratios is much used in epidemiology, especially in retrospective case-control studies. Consider for example the relation between lung cancer and exposure to asbestos. Instead of sampling from the two populations of people who are exposed to asbestos and people who have not been exposed to asbestos, and studying the odds ratio for lung cancer between the two populations defined by exposure status, one could just as well sample from the two populations of people who got lung cancer and people who did not get lung cancer, and study the odds ratio for asbestos-exposure between the two populations defined by health status. This is extremely valuable when studying illnesses which are rather rare, even in the population exposed to the risk factor of interest.
Interpretation
The odds ratio is generally used to measure the association between a risk factor and disease. However, using the odds ratio to measure the ability of a risk factor to diagnose disease is problematic.[4] The odds ratio should be at least 16 to have reasonable diagnostic ability.[5]
The odds ratio is similar to the relative risk ratio. The two ratios will be numerically similar if the rates in the two groups being compared are both similar and both less than 20% to 30%.[6][7][8]
The odds ratio may be the most stable ratio across different prevalences.[9]
The odds ratio may be used to derive the number needed to treat:[9][10]
For odds ratios less than 1:[10]
For odds ratios greater than 1:[10]
References
- ↑ Anonymous. Odds and odds ratio. Bandolier.
- ↑ Prasad K et al. (2008). "Tips for teachers of evidence-based medicine: understanding odds ratios and their relationship to risk ratios". J Gen Intern Med 23: 635–40. DOI:10.1007/s11606-007-0453-4. PMID 18181004. Research Blogging.
- ↑ Power M (2008). "Resource reviews". Evidence-based Medicine 13: 92. PMID 18515638. [e]
- ↑ Boyko EJ, Alderman BW (1990). "The use of risk factors in medical diagnosis: opportunities and cautions". J Clin Epidemiol 43: 851–8. PMID 2213074. [e]
- ↑ Pepe MS et al. (2004). "Limitations of the odds ratio in gauging the performance of a diagnostic, prognostic, or screening marker". Am J Epidemiol 159: 882–90. PMID 15105181.
- ↑ Sinclair JC, Bracken MB (1994). "Clinically useful measures of effect in binary analyses of randomized trials". J Clin Epidemiol 47: 881–9. PMID 7730891.
- ↑ Altman DG, Deeks JJ, Sackett DL (November 1998). "Odds ratios should be avoided when events are common". BMJ 317 (7168): 1318. PMID 9804732. PMC 1114216. [e]
- ↑ Page J, Attia J (2003). "Using Bayes' nomogram to help interpret odds ratios". ACP J Club 139: A11–2. PMID 12954046. Helpful chart
- ↑ 9.0 9.1 Furukawa TA et al. (2002). "Can we individualize the 'number needed to treat'? An empirical study of summary effect measures in meta-analyses". Int J Epidemiol 31: 72–6. PMID 11914297. [e]
- ↑ 10.0 10.1 10.2 McQuay HJ, Moore RA (1997). "Using numerical results from systematic reviews in clinical practice". Ann Intern Med 126: 712–20. PMID 9139558.