Statistics quick reference — spinal stenosis¶
Last curated: 2026-08-30
Use rule. Every row preserves population, horizon and method. Estimates that answer different case definitions or use different comparators are displayed side by side and are never averaged. All PMIDs were retrieved live from PubMed E-utilities on 2026-08-30.
Prevalence by definition and setting¶
| Figure | Population / year | Method | Source |
|---|---|---|---|
| 11% (95% CI 4–18%) | General adult populations; studies through review search | Pooled clinically diagnosed LSS | Jensen 2020, PMID 32095908 |
| 25% (19–32%) | Primary-care samples | Pooled clinically diagnosed LSS | Jensen 2020, PMID 32095908 |
| 29% (22–36%) | Secondary-care samples | Pooled clinically diagnosed LSS | Jensen 2020, PMID 32095908 |
| 39% (39–39%) | Mixed primary/secondary-care sample stratum | Pooled clinically diagnosed LSS | Jensen 2020, PMID 32095908 |
| 11% (5–18%) | Asymptomatic samples | Pooled radiological LSS | Jensen 2020, PMID 32095908 |
| 38% (−10% to 85%) | General-population samples | Pooled radiological LSS; very sparse/heterogeneous | Jensen 2020, PMID 32095908 |
| 15% (13–18%) | Primary-care samples | Pooled radiological LSS | Jensen 2020, PMID 32095908 |
| 32% (22–41%) | Secondary-care samples | Pooled radiological LSS | Jensen 2020, PMID 32095908 |
| 21% (16–26%) | Mixed-care samples | Pooled radiological LSS | Jensen 2020, PMID 32095908 |
| 41 papers / 55 samples | Adult LSS prevalence literature | Systematic-review evidence base | Jensen 2020, PMID 32095908 |
| Two-thirds high risk of bias | Same evidence base | Review risk-of-bias assessment | Jensen 2020, PMID 32095908 |
Imaging definitions and discordance¶
| Figure | Population / year | Method | Source |
|---|---|---|---|
| <10 mm | Published lumbar imaging literature | Common anteroposterior-diameter threshold for central stenosis | Steurer 2011, PMID 21798008 |
| <70 mm² | Published lumbar imaging literature | Common canal cross-sectional-area threshold | Steurer 2011, PMID 21798008 |
| 10 quantitative parameters | 25 imaging studies | Systematic review of radiological criteria | Steurer 2011, PMID 21798008 |
| 4 of 63 treatment studies | Primary studies in treatment reviews | Reported quantitative imaging entry criteria | Steurer 2011, PMID 21798008 |
| 437 patients | NORDSTEN surgical candidates | Baseline cross-sectional MRI cohort | Aaen 2022, PMID 34797405 |
| 71.3% | Same cohort | Severe Schizas C/D morphology | Aaen 2022, PMID 34797405 |
| 86.8% | Same cohort | Dural-sac area <75 mm² | Aaen 2022, PMID 34797405 |
| 58.1% | Same cohort | Pfirrmann grade 4/5 disc degeneration | Aaen 2022, PMID 34797405 |
| 11.9% | Same cohort | Facet tropism ≥15° | Aaen 2022, PMID 34797405 |
| 83.7% | Same cohort | Multifidus fatty-degeneration grades 2–4 | Aaen 2022, PMID 34797405 |
| No clinically relevant association | Same cohort | Multivariable relation of MRI severity to ODI, ZCQ and pain | Aaen 2022, PMID 34797405 |
Lumbar natural history¶
| Figure | Population / horizon | Method | Source |
|---|---|---|---|
| 146 enrolled; 89% followed | Moderate, nonsurgical central LSS; 3.3 years | Consecutive observational cohort | Wessberg 2017, PMID 28417234 |
| Mean age 68; 58% women | Same cohort | Baseline description | Wessberg 2017, PMID 28417234 |
| 32% improved / 55% unchanged / 13% worsened | Same cohort | Leg-pain MCID categories | Wessberg 2017, PMID 28417234 |
| 36% improved / 54% unchanged / 10% worsened | Same cohort | Back-pain MCID categories | Wessberg 2017, PMID 28417234 |
| 7% crossed to surgery | Same cohort | Revision of initial nonsurgical decision | Wessberg 2017, PMID 28417234 |
| No group walking improvement | Same cohort | Longitudinal walking outcome | Wessberg 2017, PMID 28417234 |
| No spontaneous improvement below 0.5 cm² | Same cohort | Subgroup by minimum canal area | Wessberg 2017, PMID 28417234 |
| 202 patients; 121±40 months | Symptomatic LSS with ≥5-year follow-up | Retrospective cohort | Ng 2025, PMID 40939213 |
| 19.3% deteriorated | Same cohort | Weakness, sphincter change or walking ≤10 min composite | Ng 2025, PMID 40939213 |
| 92.3% of deteriorators | Deteriorating subgroup | Reduced walking tolerance | Ng 2025, PMID 40939213 |
| 20.5% of deteriorators | Deteriorating subgroup | Myotomal weakness | Ng 2025, PMID 40939213 |
| 5.1% of deteriorators | Deteriorating subgroup | Sphincter disturbance | Ng 2025, PMID 40939213 |
| Adjusted hazard ratio 10.11 | Same cohort | Developmental stenosis predicting deterioration | Ng 2025, PMID 40939213 |
| AUC 0.781 at <55 mm² | Same cohort | Dural-sac area predicting 5-year deterioration | Ng 2025, PMID 40939213 |
DCM natural history and severity¶
| Figure | Population / horizon | Method | Source |
|---|---|---|---|
| 6 studies | Conservatively managed DCM | Meta-analysis of prospective cohorts/RCTs | Sarraj 2024, PMID 37549831 |
| 91% stable (95% CrI 83–97%) | Mild DCM; 1 year | Modeled neurological-stability survival | Sarraj 2024, PMID 37549831 |
| 85% (72–94%) | Mild DCM; 2 years | Same | Sarraj 2024, PMID 37549831 |
| 84% (70–94%) | Mild DCM; 3 years | Same | Sarraj 2024, PMID 37549831 |
| 75% (54–90%) | Mild DCM; 5 years | Same | Sarraj 2024, PMID 37549831 |
| 66% (40–86%) | Mild DCM; 15 years | Same | Sarraj 2024, PMID 37549831 |
| 65% (39–86%) | Mild DCM; 20 years | Same | Sarraj 2024, PMID 37549831 |
| 83% stable (76–89%) | Moderate/severe DCM; 1 year | Modeled neurological-stability survival | Sarraj 2024, PMID 37549831 |
| 72% (62–81%) | Moderate/severe DCM; 2 years | Same | Sarraj 2024, PMID 37549831 |
| 71% (60–80%) | Moderate/severe DCM; 3 years | Same | Sarraj 2024, PMID 37549831 |
| 55% (41–68%) | Moderate/severe DCM; 5 years | Same | Sarraj 2024, PMID 37549831 |
| 44% (27–59%) | Moderate/severe DCM; 15 years | Same | Sarraj 2024, PMID 37549831 |
| 43% (25–58%) | Moderate/severe DCM; 20 years | Same | Sarraj 2024, PMID 37549831 |
| 757 patients / 26 sites | Prospective AO Spine cohorts | mJOA severity-threshold derivation | Tetreault 2017, PMID 27342612 |
| mJOA 15–17 | Clinically diagnosed DCM | Mild category anchored to Nurick | Tetreault 2017, PMID 27342612 |
| mJOA 12–14 | Same | Moderate category | Tetreault 2017, PMID 27342612 |
| mJOA 0–11 | Same | Severe category | Tetreault 2017, PMID 27342612 |
Rehabilitation trials¶
| Figure | Population / horizon | Method | Source |
|---|---|---|---|
| 259 participants | LSS, age >60; 6-week interventions | Three-arm randomized trial | Schneider 2019, PMID 30646197 |
| Mean age 72.4±7.8; 52.9% women | Same trial | Baseline characteristics | Schneider 2019, PMID 30646197 |
| −2.0 SSS points (95% CI −3.6 to −0.4) | Manual/individualized vs medical care; 2 months | Adjusted between-group difference | Schneider 2019, PMID 30646197 |
| −2.4 (−4.1 to −0.8) | Manual/individualized vs group exercise; 2 months | Adjusted between-group difference | Schneider 2019, PMID 30646197 |
| 20% vs 7.6% vs 3.0% | Manual/individualized vs medical vs group exercise | ≥30% symptom/function responders | Schneider 2019, PMID 30646197 |
| 65.3% vs 48.7% vs 46.2% | Same arms | Walking-capacity responders at 2 months | Schneider 2019, PMID 30646197 |
| No between-group difference | Same arms; 6 months | Mean outcomes and responder rates | Schneider 2019, PMID 30646197 |
| 43 vs 43 participants | MRI-confirmed LSS | Supervised PT vs home-exercise RCT | Minetama 2019, PMID 30986577 |
| +30.2 percentage points (95% CI 9.1–48.6) | 6 weeks | ZCQ symptom-severity responder difference | Minetama 2019, PMID 30986577 |
| +32.6 percentage points (11.6–50.6) | 6 weeks | ZCQ physical-function responder difference | Minetama 2019, PMID 30986577 |
| +39.5 percentage points (18.8–56.7) | 6 weeks | Walking-distance responder difference | Minetama 2019, PMID 30986577 |
| +34.9 percentage points (13.9–52.7) | 6 weeks | Leg-pain responder difference | Minetama 2019, PMID 30986577 |
| +25.6 percentage points (4.9–43.9) | 6 weeks | Daily-step responder difference | Minetama 2019, PMID 30986577 |
| 13 trials / 1,440 participants | Neurogenic-claudication exercise RCTs | Systematic review/component analysis | Comer 2024, PMID 37715644 |
| 23 interventions / 60 components | Same review | Component inventory | Comer 2024, PMID 37715644 |
Epidural injection¶
| Figure | Population / horizon | Method | Source |
|---|---|---|---|
| 400 randomized | Central LSS with moderate/severe leg pain and disability | Double-blind glucocorticoid+lidocaine vs lidocaine | Friedly 2014, PMID 24988555 |
| −1.0 RMDQ points (95% CI −2.1 to 0.1; P=0.07) | 6 weeks | Adjusted incremental steroid effect | Friedly 2014, PMID 24988555 |
| −0.2/10 leg pain (−0.8 to 0.4; P=0.48) | 6 weeks | Adjusted incremental steroid effect | Friedly 2014, PMID 24988555 |
| No significant route interaction | Interlaminar vs transforaminal subgroup | Prespecified analysis | Friedly 2014, PMID 24988555 |
Decompression versus nonoperative care¶
| Figure | Population / horizon | Method | Source |
|---|---|---|---|
| 289 randomized + 365 observational | Surgical candidates, ≥12 weeks symptoms | SPORT parallel cohorts | Weinstein 2010, PMID 20453723 |
| +12.6 bodily-pain points (95% CI 8.5–16.7) | 4 years | Adjusted as-treated surgical effect | Weinstein 2010, PMID 20453723 |
| +8.6 physical-function points (4.6–12.6) | 4 years | Adjusted as-treated surgical effect | Weinstein 2010, PMID 20453723 |
| −9.4 ODI points (−12.6 to −6.2) | 4 years | Adjusted as-treated surgical effect | Weinstein 2010, PMID 20453723 |
| 70% vs 52% underwent surgery | Randomized-to-surgery vs randomized-nonoperative; 8 years | Treatment crossover | Lurie 2015, PMID 25569524 |
| 55% vs 52% retained | Randomized vs observational cohorts; 8 years | Follow-up completeness | Lurie 2015, PMID 25569524 |
| No significant years 6–8 effect | Randomized cohort, as-treated | Late treatment-effect convergence | Lurie 2015, PMID 25569524 |
Decompression versus fusion¶
| Figure | Population / horizon | Method | Source |
|---|---|---|---|
| 247 randomized | One/two-level LSS; 135 with spondylolisthesis | Swedish trial | Försth 2016, PMID 27074066 |
| ODI 27 vs 24 (P=0.24) | Fusion vs decompression; 2 years | Primary outcome | Försth 2016, PMID 27074066 |
| 397 vs 405 m (P=0.72) | Fusion vs decompression; 6-minute walk | Objective function | Försth 2016, PMID 27074066 |
| 7.4 vs 4.1 days (P<0.001) | Fusion vs decompression | Hospital stay | Försth 2016, PMID 27074066 |
| 22% vs 21% | Fusion vs decompression; mean 6.5 years | Additional lumbar surgery | Försth 2016, PMID 27074066 |
| ODI change −20.6 vs −21.3 | Decompression vs fusion; 2 years | NORDSTEN-DS mean change | Austevoll 2021, PMID 34347953 |
| Difference 0.7 (95% CI −2.8 to 4.3) | Same | Between-group ODI change | Austevoll 2021, PMID 34347953 |
| 71.4% vs 72.9% | Same | ≥30% ODI responder, modified ITT | Austevoll 2021, PMID 34347953 |
| Difference −1.4 points (−12.2 to 9.4) | Same | Responder contrast; noninferiority margin −15 | Austevoll 2021, PMID 34347953 |
| 12.5% vs 9.1% | Decompression vs fusion; 2 years | Reoperation | Austevoll 2021, PMID 34347953 |
| 63% vs 63% | Decompression vs fusion; 5 years | ≥30% ODI response, imputed modified ITT | Kgomotso 2024, PMID 39111800 |
| Difference 0.4 points (−11.2 to 11.9) | Same | Five-year responder contrast | Kgomotso 2024, PMID 39111800 |
| ODI change −17.8 vs −17.8 | Same | Five-year mean change | Kgomotso 2024, PMID 39111800 |
| 16% vs 18% | Decompression vs fusion; baseline to 5 years | Subsequent lumbar operation | Kgomotso 2024, PMID 39111800 |
| ODI 25 vs 28 (P=0.226) | Decompression vs fusion; Swedish trial 5 years | Primary outcome | Karlsson 2024, PMID 38945544 |
| EQ-5D 0.69 vs 0.59 (P=0.027) | Same | Health-related quality of life | Karlsson 2024, PMID 38945544 |
| 22% vs 24% (RR 1.1, 95% CI 0.69–1.8) | Decompression vs fusion | Subsequent spinal surgery | Karlsson 2024, PMID 38945544 |
Interspinous spacer outcomes¶
| Figure | Population / horizon | Method | Source |
|---|---|---|---|
| 391 implanted (190 vs 201) | Moderate LSS after ≥6 months nonsurgical care | Superion vs X-STOP randomized noninferiority trial | Patel 2015, PMID 25494323 |
| 99.5% vs 99.0% | Superion vs X-STOP | Implantation success | Patel 2015, PMID 25494323 |
| 70% leg-pain reduction | Both groups; 2 years | Within-group change | Patel 2015, PMID 25494323 |
| 77% leg-pain success | Both groups; 2 years | ≥20-mm improvement | Patel 2015, PMID 25494323 |
| 68% back-pain success | Both groups; 2 years | ≥20-mm improvement | Patel 2015, PMID 25494323 |
| 65% ODI success | Both groups; 2 years | ≥15-point improvement | Patel 2015, PMID 25494323 |
| 74 of 88 (84%) | Superion recipients available at 5 years | Success on ≥2 of 3 ZCQ domains | Nunley 2017, PMID 28919727 |
| 75% / 81% / 90% | Same | ZCQ symptom/function/satisfaction success | Nunley 2017, PMID 28919727 |
| 75% free from index-level repeat procedure | Same; 5 years | Reoperation/revision/fixation-free proportion | Nunley 2017, PMID 28919727 |
Falls, patient-derived outcomes and perioperative vulnerability¶
| Figure | Population / horizon | Method | Source |
|---|---|---|---|
| 64 of 135 fell | Surgical DCM cohort; enrollment to 1 year | Prospective multicenter study | Inose 2021, PMID 33496544 |
| 65 of 132 (49%) fell | DCM fall-diary respondents | Prospective 8-center cohort | Kimura 2020, PMID 31809466 |
| 497.4 → 90.3 falls/100 person-years | Same cohort; pre- vs postoperative | Diary-derived incidence | Kimura 2020, PMID 31809466 |
| 34% → 8% motor deterioration per fall | Same cohort | Pre- vs postoperative event proportion | Kimura 2020, PMID 31809466 |
| 8-person focus group + 224-person survey | People living with DCM/supporters | Mixed-methods outcome generation | Davies 2022, PMID 33203262 |
| 52 initial + 16 added outcomes | Same | Patient-derived consequence list | Davies 2022, PMID 33203262 |
| All outcomes recognized by ≥5% | Same survey | Endorsement threshold | Davies 2022, PMID 33203262 |
| 36 of 89 (40.5%) delirium | Spine-surgery patients aged ≥70 | Prospective validated assessment | Brown 2016, PMID 27696373 |
| 17 of 36 (47.2%) hypoactive only | Delirium cases | Phenotype assessment | Brown 2016, PMID 27696373 |
| OR 3.66 (95% CI 1.48–9.04) | Same cohort | Delirium associated with higher length-of-stay quintile | Brown 2016, PMID 27696373 |
| OR 3.49 (1.35–9.00) | Same cohort | Delirium associated with higher charge quintile | Brown 2016, PMID 27696373 |
| OR 0.22 (0.07–0.69) | Same cohort | Delirium associated with discharge home | Brown 2016, PMID 27696373 |
Known conflicts and caveats¶
- Clinical versus radiological prevalence: pooled values answer different case definitions; imaging-only estimates must not be used as symptomatic-disease prevalence (Jensen 2020, PMID 32095908).
- Short versus long lumbar natural history: the 3.3-year cohort selected tolerable moderate symptoms; the 10-year cohort used a deterioration composite and retrospective selection (Wessberg 2017, PMID 28417234; Ng 2025, PMID 40939213).
- SPORT causal contrast: four-year as-treated effects favored surgery, while eight-year randomized-cohort effects converged amid 70%/52% crossover and roughly half retention (Weinstein 2010, PMID 20453723; Lurie 2015, PMID 25569524).
- Fusion conflict: SLIP reported a more favorable signal for fusion, whereas the larger Swedish and NORDSTEN programs did not show better disability outcomes from routine fusion (Ghogawala 2016, PMID 27074067; Försth 2016, PMID 27074066; Kgomotso 2024, PMID 39111800).
- Spacer durability: five-year percentages are among 85–88 available participants from an original 391-person device trial, and the randomized comparator was another spacer rather than decompression (Patel 2015, PMID 25494323; Nunley 2017, PMID 28919727).
- DCM stability curves: credible intervals widen markedly at 15–20 years, reflecting few studies and attrition; they are not individual prognostic calculators (Sarraj 2024, PMID 37549831).
- Delirium transportability: the 40.5% estimate comes from a single academic cohort with active validated surveillance, not routine chart coding (Brown 2016, PMID 27696373).
- Falls before/after surgery: large pre/post reductions are observational and can reflect activity exposure and reporting changes as well as surgery (Kimura 2020, PMID 31809466).