Easy Probability Estimation for Diagnosing Early Axial SpA
The scores of relevant findings derived from the LR values given in Rudwaleit et al. are listed in Table 1. Since the validity of incorporating negative LRs has never been validated, we recommend to ignore negative test results in an early state of possible axial SpA and therefore have included in Table 1 only positive LRs. This is also reasonable because some of the SpA features may not be present at disease onset but may develop later, and their absence in early disease does not mean anything. How the post-test probability of axial SpA depends on the sum of the scores if the pre-test probability is 5% (chronic back pain, i.e. back pain of >3 months duration with age at onset <45 years) is shown in Fig. 1.
(Enlarge Image)
Figure 1.
Dependence of the probability of axial SpA on the sum of scores according to Table 1 for a pre-test probability of 5% (chronic back pain, i.e. back pain of >3 months duration with age at onset <45 years).
A diagnosis of definitive axial SpA can be made if the score sum is >51 (probability >90% as proposed in Rudwaleit et al.). A diagnosis of probable axial SpA can be made if the score sum is >43 (probability >80%). If the score sum is <13, axial SpA is improbable (probability < 15%). For score sums between 13 and 43, additional tests are necessary to come to a decision.
As in Rudwaleit et al., we regard a diagnosis of axial SpA as definitive if the probability is at least 90%. This is the case if the sum of the scores is >51. Correspondingly, a diagnosis of probable axial SpA can be made if the sum of the scores is larger than 43 (probability > 80%). If the sum of scores is <13 (probability < 15%), axial SpA is improbable. For score sums between 13 and 43, additional tests are necessary to come to a decision.
Results
The scores of relevant findings derived from the LR values given in Rudwaleit et al. are listed in Table 1. Since the validity of incorporating negative LRs has never been validated, we recommend to ignore negative test results in an early state of possible axial SpA and therefore have included in Table 1 only positive LRs. This is also reasonable because some of the SpA features may not be present at disease onset but may develop later, and their absence in early disease does not mean anything. How the post-test probability of axial SpA depends on the sum of the scores if the pre-test probability is 5% (chronic back pain, i.e. back pain of >3 months duration with age at onset <45 years) is shown in Fig. 1.
(Enlarge Image)
Figure 1.
Dependence of the probability of axial SpA on the sum of scores according to Table 1 for a pre-test probability of 5% (chronic back pain, i.e. back pain of >3 months duration with age at onset <45 years).
A diagnosis of definitive axial SpA can be made if the score sum is >51 (probability >90% as proposed in Rudwaleit et al.). A diagnosis of probable axial SpA can be made if the score sum is >43 (probability >80%). If the score sum is <13, axial SpA is improbable (probability < 15%). For score sums between 13 and 43, additional tests are necessary to come to a decision.
As in Rudwaleit et al., we regard a diagnosis of axial SpA as definitive if the probability is at least 90%. This is the case if the sum of the scores is >51. Correspondingly, a diagnosis of probable axial SpA can be made if the sum of the scores is larger than 43 (probability > 80%). If the sum of scores is <13 (probability < 15%), axial SpA is improbable. For score sums between 13 and 43, additional tests are necessary to come to a decision.
SHARE
