Selection Bias in Sacroiliac Joint Pain Studies
Selection bias in sacroiliac joint pain research refers to systematic errors in patient recruitment that artificially inflate or distort the reported 15–30% prevalence of SIJ pain among chronic low back pain patients, primarily by pre-selecting populations already enriched for SIJ pathology rather than consecutive, unselected patients with axial pain.
Definition and Impact on Prevalence Estimates
Selection bias occurs when study populations are not representative of the broader patient population with chronic low back or buttock pain. The risk of patient selection bias was high in 52–55% of diagnostic accuracy studies evaluating sacroiliac joint pathology 1. This fundamentally undermines the validity of the commonly cited 15–30% prevalence figure 2, 3.
Mechanisms of Selection Bias in SIJ Studies
The bias manifests through several pathways:
Referral filter bias: Studies conducted in specialized pain clinics or tertiary centers receive patients already suspected of having SIJ pathology, creating enriched populations that overestimate true prevalence 1
Diagnostic criteria variability: The lack of standardized inclusion criteria means different studies select fundamentally different patient populations, making prevalence estimates non-comparable 4, 5
Exclusion of competing diagnoses: Many studies exclude patients with disc herniation, radiculitis, or discogenic pain, artificially concentrating SIJ pathology in the remaining cohort 4
Real-World Prevalence vs. Biased Estimates
When unselected consecutive patients presenting with ≥50% sacroiliac region pain were systematically evaluated, only 3–6% had confirmed primary SIJ pathology, while 88–90% had lumbar spine pathology as the actual pain generator 6. This starkly contrasts with the 15–30% prevalence cited in most literature 2, 3.
The False-Positive Problem
Selection bias interacts dangerously with diagnostic test limitations:
- Single uncontrolled SIJ injections have approximately 20% false-positive rates 5, 7
- When applied to pre-selected populations already suspected of SIJ pain, this compounds the overestimation
- Applicability concerns regarding patient selection were high in 36% of included diagnostic studies 1
Clinical Implications
The sacroiliac joint is a common site of referred pain from the lumbar spine (88–90% of cases), not a primary pain generator in most patients complaining of SIJ region pain 6. Selection bias in the literature obscures this critical clinical reality.
Avoiding Selection Bias in Clinical Practice
To avoid being misled by biased prevalence estimates:
- Recognize that most patients with SIJ region pain have lumbar spine pathology, not primary SIJ dysfunction 6
- Use controlled comparative local anesthetic blocks rather than single injections to minimize false-positives 5, 7
- Maintain high clinical suspicion for lumbar spine sources even when pain localizes to the SIJ region 6
- Apply batteries of provocative tests (≥3 positive tests improve specificity), though recent evidence questions their predictive value 3
Quality Assessment Findings
Risk of bias assessment revealed that patient selection bias and applicability concerns were the predominant methodological weaknesses across SIJ diagnostic literature 1. This means the 15–30% prevalence figure should be interpreted with extreme caution, as it likely represents highly selected populations rather than true community or primary care prevalence.
The evidence suggests the actual prevalence of primary SIJ pathology in unselected patients with chronic low back pain is substantially lower than commonly reported, likely in the 3–10% range rather than 15–30% 6, 4, 5.