Skip to main content
Sports Medicine - Open

Table 3 Likelihood ratio estimates in support of the null hypothesis (no effect of prior concussion on clinical outcomes)

From: Clinical Outcome Following Concussion Among College Athletes with a History of Prior Concussion: A Systematic Review

 

Total studies

Studies reporting both “positive” results and null results

Number of “Positive” studies

Likelihood ratio

80% Power, 5% alpha

45% Power, 25% alpha

All studies

16

7

218.07+

3.76+

Symptom duration

13

4

18.78−

1.55−

Return to play

7

2

9.45−

1.46−

Lowest risk for biasa

7

4

611.5+

4.14+

  1. “Positive” refers to a statistically significant result (i.e., < 0.05). Studies that reported both statistically significant results and null results were classified as reporting “positive” results suggestive of an association between prior concussion history and clinical outcome (with the exception of Lempke et al. [29] as noted in Table 1). + (positive) Higher likelihood ratios (LR) indicate increased likelihood in favor of the alternative hypothesis (i.e., prior concussion history is associated with clinical outcomes following subsequent concussion); − (negative) Higher likelihood ratios (LR) indicate increased likelihood in favor of the null hypothesis (i.e., prior concussion history is not associated with clinical outcomes following subsequent concussion). Magnitude of LRs was characterized as follows: LR = 1 = no effect, LR between 1 and 8 = weak, LR greater than 8 and less than 32 = moderate, LR greater than 32 = strong. aGreater than 6 credits on the Newcastle–Ottawa Scale
Back to article page