Content deleted Content added
→By predictive values: someone got confused by double negatives in explaining what negative predictive value means. |
m fix span tags |
||
Line 20:
|-
! By [[predictive value]]s
| Direct quotients from reference group ||<span style="color:DarkGreen;">Most straightforward: Predictive value equals probability</span> ||<span style="color:DarkRed;"> Usually low: Separate reference group required for every subsequent pre-test state</span> ||<span style="color:DarkGreen;"> Available both for [[binary classification|binary]] and [[Continuity (mathematics)|continuous]] values</span>
|-
! By [[Likelihood ratios in diagnostic testing|likelihood ratio]]
| Derived from [[sensitivity and specificity]] || Post-test odds given by multiplying pretest odds with the ratio ||<span style="color:DarkGreen;"> Theoretically limitless</span> ||<span style="color:DarkGreen;"> Pre-test state (and thus the pre-test probability) does not have to be same as in reference group
|-
! By [[relative risk]]
| Quotient of risk among exposed and risk among unexposed || Pre-test probability multiplied by the relative risk ||<span style="color:DarkRed;"> Low, unless subsequent relative risks are derived from same [[multivariate regression analysis]]</span> ||<span style="color:DarkGreen;"> Relatively intuitive to use
|-
! By [[diagnostic criteria]] and [[clinical prediction rules]]
| Variable, <span style="color:DarkRed;">but usually most tedious</span> || Variable ||<span style="color:DarkGreen;"> Usually excellent for all test included in criteria</span> ||<span style="color:DarkGreen;"> Usually most preferable if available</span>
|}
Line 47:
| rowspan="2" style="background: #ffdddd;"| '''Test<br>outcome'''
| style="background: #ffdddd;"|'''''Positive'''''
| style="background: #eeeeff;"| <span style="color:#007700;"> '''True Positive'''</span>
| <span style="color:#770000;"> '''False Positive'''<BR>([[Type I and type II errors#False positive rate|Type I error]])</span>
| <span style="color:#770077;"> → [[Positive predictive value]]</span>
|-
| style="background: #ffdddd;"|'''''Negative'''''
| style="background: #eeeeff;"|<span style="color:#770000;"> '''False Negative'''<BR>([[Type I and type II errors#False negative rate|Type II error]])</span>
| <span style="color:#007700;"> '''True Negative'''</span>
| <span style="color:#770077;"> → [[Negative predictive value]]</span>
|-
| colspan="2" |
| style="background: #eeeeff;" | <span style="color:#770077;"> ↓<br>[[Sensitivity and specificity|Sensitivity]]</span>
| <span style="color:#770077;"> ↓<br>[[Sensitivity and specificity|Specificity]]</span>
|
|}
Line 122:
| rowspan="2" style="background: #ffdddd;"| '''Fecal<br>occult<br>blood<br>screen<br>test<br>outcome'''
| style="background: #ffdddd;"|'''''Positive'''''
| style="background: #eeeeff;"| <span style="color:#007700;"> '''TP = 2'''</span>
| <span style="color:#770000;"> '''FP = 18'''</span>
| → Positive predictive value<BR>= TP / (TP + FP)<BR>= 2 / (2 + 18)<BR>= 2 / 20<BR>'''= 10%'''
|-
| style="background: #ffdddd;"|'''''Negative'''''
| style="background: #eeeeff;"|<span style="color:#770000;"> '''FN = 1'''</span>
| <span style="color:#007700;"> '''TN = 182'''</span>
| → Negative predictive value<BR>= TN / (FN + TN)<BR>= 182 / (1 + 182)<BR>= 182 / 183<BR>'''≈ 99.5%'''
|-
| |||