Assessment & Screening — all lessons
Screening
Preparticipation health screeningResting and Body Composition Measures
Resting assessments and anthropometric measurementsFitness Testing
Cardiorespiratory assessmentsMuscular strength, endurance and power assessmentsAthlete assessment and evaluationMovement and Postural Assessment
Postural and functional assessment (Prehab assessment tool)Limits of movement screens and postural assessmentLimits of movement screens and postural assessment
Why screening tests rarely predict individual injuries — sensitivity, specificity, base rates and predictive value explained with worked examples — and how to use assessments responsibly.
Learning objectives
- Define sensitivity, specificity and positive/negative predictive value
- Work through a base-rate example showing why accurate prediction is hard
- Distinguish risk factors at group level from prediction at individual level
- Communicate assessment purpose honestly to clients and coaches
Before this lesson
Why this matters
Screening tools are widely sold as injury predictors. Understanding why they rarely deliver protects clients from unnecessary labels and helps you invest time where evidence is strongest.
Fundamental concepts
| Term | Meaning |
|---|---|
| Sensitivity | Of those who will be injured, the % who test positive |
| Specificity | Of those who won’t be injured, the % who test negative |
| Positive predictive value (PPV) | Of those who test positive, the % who are actually injured |
| Negative predictive value (NPV) | Of those who test negative, the % who remain uninjured |
The base-rate problem: a worked example
A squad of 100 athletes; 10 will sustain the injury this season. A test has 70% sensitivity and 70% specificity.
| Will be injured (10) | Won’t be injured (90) | |
|---|---|---|
| Test positive | 7 (true +) | 27 (false +) |
| Test negative | 3 (false −) | 63 (true −) |
PPV = 7 ÷ 34 ≈ 21%. Most “high-risk” athletes will not be injured, and 3 of the 10 who will be injured were missed. This is why even genuine risk factors make poor individual predictors [bahr2016].
What the evidence shows
- Knee valgus: one prospective study associated landing valgus with ACL injury [hewett2005], but a larger later study found the drop vertical jump could not usefully identify future ACL injuries [krosshaug2016].
- FMS: a meta-analysis found composite scores were not useful predictors of injury [moran2017].
Bahr summarises that a useful screen requires (1) a strong, well-established risk factor, (2) a test that identifies individuals with high accuracy, and (3) an effective intervention for those identified — a combination rarely met [bahr2016].
Practical application
Common mistakes and misconceptions
Case study
A club asks to screen players to “identify those at risk” before the season. The S&C coach proposes instead a squad-wide injury-prevention warm-up, load monitoring, and baseline assessments used for individual programming — explaining the base-rate example above to the coaching staff.
Further learning: video lessons
Video links are only marked “verified” after a human has watched them. Unverified entries open a YouTube search for that channel and topic rather than a guessed link.
Why screening tests don’t predict injury — Roald Bahr
BJSM / conference lectures (search)
You will understand: the statistical logic behind screening limits
Why: Primary author explaining his review.
Sensitivity, specificity and predictive values
Khan Academy
You will understand: the core diagnostic-test statistics
Why: Clear statistical foundation.
Knowledge check
Knowledge check · 0/5
1. Sensitivity is the proportion of…
2. If 10% of a squad will be injured and a test has 70% sensitivity and 70% specificity, the probability that a positive-testing athlete is injured is roughly…
3. Bahr (2016) argued that to be useful a screening test must…
4. The best-supported approach to preventing common sports injuries is…
5. Which is a valid use of movement assessment?
Short-answer & application questions
- Recalculate the base-rate example for an injury rate of 30% and explain the change.
- Explain the difference between a risk factor and a predictor.
- Draft a short note to a coach explaining why you will not use a screen to select “high-risk” players.
Key takeaways
- A real risk factor at group level is usually not a useful predictor for an individual.
- With low injury rates, even reasonably accurate tests produce many false positives.
- Large prospective studies and meta-analyses show common screens predict injury poorly.
- Use assessments to guide training and track change; apply injury-prevention programmes to everyone.
Connected topics
Biomechanics and movement analysis
anatomyThe knee complex
prehab rehabInjury prevention programmes
corrective exerciseRationale for corrective exercise (and its limits)
References
- Bahr R (2016). Why screening tests to predict injury do not work—and probably never will…: a critical review. British Journal of Sports Medicine, 50(13):776–780. doi:10.1136/bjsports-2016-096256 Narrative review check pending
- Moran RW, Schneiders AG, Mason J, Sullivan SJ (2017). Do Functional Movement Screen (FMS) composite scores predict subsequent injury? A systematic review with meta-analysis. British Journal of Sports Medicine, 51(23):1661–1669. doi:10.1136/bjsports-2016-096938 Meta-analysis check pending
- Krosshaug T, Steffen K, Kristianslund E, et al. (2016). The vertical drop jump is a poor screening test for ACL injuries in female elite soccer and handball players. American Journal of Sports Medicine, 44(4):874–883. doi:10.1177/0363546515625048 Cohort check pending
- Hewett TE, Myer GD, Ford KR, et al. (2005). Biomechanical measures of neuromuscular control and valgus loading of the knee predict anterior cruciate ligament injury risk in female athletes. American Journal of Sports Medicine, 33(4):492–501. doi:10.1177/0363546504269591 Cohort check pending
Educational content only — not medical advice. Consult a qualified clinician for diagnosis or treatment.