Skip to content

Statistics — degenerative disc disease and related low-back pain

Last curated: 2026-08-30

Use notes

  • DDD-specific population burden is not identifiable from current case definitions; global rows use low-back pain and are labeled accordingly.
  • Imaging prevalence is not symptomatic prevalence.
  • Treatment rows preserve the study population and comparator; they are not cross-trial rankings.
  • Every PMID was returned by live PubMed E-utilities in the 2026-08-29 build or 2026-08-30 deepening session and re-fetched in the independent 2026-08-30 audit.

Global burden of low-back pain

Figure Estimate (95% uncertainty interval) Year Population Method/source
Prevalent cases 619 million (554–694) 2020 204 countries/territories GBD Bayesian meta-regression, PMID 37273833
Projected prevalent cases 843 million (759–933) 2050 Global SDI regression × projected population, PMID 37273833
Age-standardized YLD rate 832/100,000 (578–1,070) 2020 Global GBD 2021, PMID 37273833
Age-standardized prevalence-rate change −10.4% (−10.9 to −10.0) 1990–2020 Global GBD 2021, PMID 37273833
Age-standardized YLD-rate change −10.5% (−11.1 to −10.0) 1990–2020 Global GBD 2021, PMID 37273833
YLDs attributed to occupational factors, smoking and high BMI 38.8% (28.7–47.0) 2020 Global low-back-pain YLDs Comparative risk assessment, PMID 37273833

Asymptomatic imaging prevalence

Source for all rows: systematic review of 33 studies and 3,110 asymptomatic people; generalized linear mixed-effects age model (Brinjikji 2015, PMID 25430861).

Feature Age 20 Age 80 Population/method
Disc degeneration 37% 96% Asymptomatic CT/MRI cohorts; modeled decade prevalence
Disc bulge 30% 84% Same
Disc protrusion 29% 43% Same
Annular fissure 19% 29% Same
Absolute age gradient: degeneration — +59 percentage points Age 80 minus age 20
Absolute age gradient: bulge — +54 points Age 80 minus age 20
Absolute age gradient: protrusion — +14 points Age 80 minus age 20
Absolute age gradient: annular fissure — +10 points Age 80 minus age 20

Childhood-to-adult trajectory

Figure Estimate Year/age Population Method/source
MRI signal change 18% age 8 Healthy schoolchild cohort Longitudinal MRI, PMID 36194584
MRI signal change 10% age 11–12 Same Longitudinal MRI, PMID 36194584
MRI signal change 38% age 18–19 Same Longitudinal MRI, PMID 36194584
Lifetime low-back pain 54% age 19 Same Interview; no significant association with signal change, PMID 36194584
≥1 Pfirrmann grade ≥3 disc 5% age 8 40 people with complete four-wave data 26-year MRI cohort, PMID 41786224
≥1 Pfirrmann grade ≥3 disc 12% age 11 Same PMID 41786224
≥1 Pfirrmann grade ≥3 disc 48% age 19 Same PMID 41786224
≥1 Pfirrmann grade ≥3 disc 72% age 34 Same PMID 41786224
Pfirrmann summary progression +0.55/year (0.48–0.63) ages 11–19 Same GEE model, PMID 41786224
Pfirrmann summary progression +0.08/year (0.05–0.11) ages 19–34 Same GEE model, PMID 41786224

Adult longitudinal MRI

Figure Estimate Follow-up Population Method/source
Progression by feature 7–13% of discs 5 years 75 male MZ twin pairs Repeated MRI, PMID 16540872
Participants with progression 7–46% 5 years Same Feature dependent, PMID 16540872
New axial annular tears 1.5% of discs 5 years Same MRI, PMID 16540872
Axial tears no longer visible 2.0% 5 years Same MRI, PMID 16540872
New sagittal HIZ 0.5% 5 years Same MRI, PMID 16540872
HIZ no longer visible 1.6% 5 years Same MRI, PMID 16540872
Endplate irregularity increase 2.1% 5 years Same MRI, PMID 16540872
Endplate irregularity decrease 1.8% 5 years Same MRI, PMID 16540872
Familial share of progression variance 47–66% 5 years Same Longitudinal models, PMID 16540872
Loading + resistance-training variance 2–10% 5 years Same Longitudinal models, PMID 16540872
Mean disc-height reduction 2.2–3.6% 5 years 134 male MZ twins Quantitative MRI, PMID 18475246
Mean bulge increase L1–L4 7–10% 5 years Same Quantitative MRI, PMID 18475246
Mean bulge increase L4–S1 4% 5 years Same Quantitative MRI, PMID 18475246
Occupational lifting variance in height loss adjusted R² 4.9% 5 years Same Regression, PMID 18475246
Smoking variance in height loss adjusted R² 3.5% 5 years Same Regression, PMID 18475246
Upper-disc height change −8.7% (−1.0 mm) 15 years 105 men Quantitative MRI, PMID 24262855
Lower-disc height change −11.3% (−1.3 mm) 15 years Same Quantitative MRI, PMID 24262855
Lumbar L1–S1 shortening 0.13 mm/year 15 years Same Morphometric MRI, PMID 24262855

Modic-change natural history

Source: 72 community adults at baseline, 56 at approximately two years (Teichtahl 2017, PMID 27324605).

Figure Estimate Population/method
Baseline type 1 participants 3/72 (4.2%) Community MRI cohort
Baseline type 1 lesions 6 Same
Persistent type 1 lesions 4/6 Follow-up MRI
Type 1 → type 2 2/6 Follow-up MRI
Baseline type 2 participants 20/72 (27.8%) Community MRI cohort
Baseline type 2 lesions 47 Same
Incident type 2 lesions 18 in 7/56 (12.5%) Follow-up MRI
Type 2 resolution 1 lesion Follow-up MRI
Reduced disc height predicting incident type 2 OR 1.9 (1.1–3.3) Longitudinal association
Baseline type 3 participants 1/72 (1.4%) Community MRI cohort

Risk factors

Figure Estimate Population Method/source
Smoking-associated degeneration 18% greater mean score Identical twins highly discordant for smoking Within-pair MRI, PMID 1948392
Heavy vs sedentary work, severe L5–S1 degeneration OR 1.86 (1.19–2.92) 1,022 postmenopausal women Adjusted cohort analysis, PMID 35084078
Injury-related height difference injured twin +0.3 mm; p=0.302 37 injury-discordant MZ pairs Within-pair MRI, PMID 20838276
Injury-related maximum-level height difference +0.05%; p=0.302 Same Within-pair MRI, PMID 20838276

Pfirrmann measurement reliability

Source: 300 discs in 60 patients, three observers (Pfirrmann 2001, PMID 11568697).

Figure Estimate Method
Mean patient age 40 years (range 10–83) Reliability sample
Grade I discs 14 T2 MRI classification
Grade II discs 82 Same
Grade III discs 72 Same
Grade IV discs 68 Same
Grade V discs 64 Same
Intraobserver κ 0.84–0.90 Repeated rating
Interobserver κ 0.69–0.81 Three readers
Exact agreement 83.8% All discs
One-grade disagreement 15.9% All discs
≥2-grade disagreement 1.3% All discs

Exercise and psychological care

Figure Estimate (95% CI) Population Method/source
Exercise trials 249 Chronic nonspecific low-back pain Cochrane review, PMID 34580864
Mean age 43.7 years Same Review sample
Women 59% average Same Review sample
Pain vs no/usual/placebo MD −15.2/100 (−18.3 to −12.2) Same Moderate certainty, PMID 34580864
Function vs no/usual/placebo MD −6.8/100 (−8.3 to −5.3) Same Below review’s 10-point importance threshold
Pain vs other conservative treatment MD −9.1/100 (−12.6 to −5.6) Same Low certainty
Function vs other conservative treatment MD −4.1/100 (−6.0 to −2.2) Same Moderate certainty
Any adverse effect reported 33% exercise vs 29% comparison groups 86 reporting studies Mostly minor, PMID 34580864
Psychological-intervention trials 97 RCTs; 13,136 participants Chronic nonspecific low-back pain Network meta-analysis, PMID 35354560
CBT + physiotherapy, function post-treatment SMD 1.01 (0.58–1.44) Same Network meta-analysis
Pain education + physiotherapy, function post-treatment SMD 0.62 (0.08–1.17) Same Network meta-analysis
Behavioral therapy + physiotherapy, pain post-treatment SMD 1.08 (0.22–1.94) Same Network meta-analysis

Pharmacological evidence

Source: overview of seven Cochrane reviews, 103 studies and 22,238 participants (Cashin 2023, PMID 37014979).

Figure Estimate (95% CI) Population/comparator
Acute LBP paracetamol pain MD +0.49/100 (−1.99 to 2.97) Placebo; high certainty
Acute LBP paracetamol disability MD +0.05/24 (−0.50 to 0.60) Placebo
Acute LBP NSAID pain MD −7.29/100 (−10.98 to −3.61) Placebo; moderate certainty
Acute LBP NSAID disability MD −2.02/24 (−2.89 to −1.15) Placebo; high certainty
Muscle relaxant adverse events RR 1.50 (1.14–1.98) Acute LBP; placebo
Chronic LBP NSAID pain MD −6.97/100 (−10.74 to −3.19) Placebo; low certainty
Chronic LBP NSAID disability MD −0.85/24 (−1.30 to −0.40) Placebo
Strong-opioid pain SMD −0.43 (−0.52 to −0.33) Chronic LBP; placebo
Strong-opioid disability SMD −0.26 (−0.37 to −0.15) Chronic LBP; placebo
Opioid nausea risk difference +0.10 (0.07–0.14) Chronic LBP; placebo
Opioid constipation risk difference +0.07 (0.04–0.11) Chronic LBP; placebo
Opioid dizziness risk difference +0.08 (0.05–0.11) Chronic LBP; placebo
Antidepressant pain SMD −0.04 (−0.25 to 0.17) Chronic LBP; placebo
Antidepressant disability SMD −0.06 (−0.40 to 0.29) Chronic LBP; placebo

STarT Back trial

Figure Estimate (95% CI) Population/method
Randomized 851 (568 stratified; 283 control) 10 English general practices, PMID 21963002
Four-month RMDQ between-group difference 1.81 (1.06–2.57) Intention-to-treat
Twelve-month RMDQ difference 1.06 (0.25–1.86) Intention-to-treat
Four-month effect size 0.32 (0.19–0.45) Standardized
Twelve-month effect size 0.19 (0.04–0.33) Standardized
Incremental QALY +0.039 12 months
Mean back-pain healthcare cost £240.01 vs £274.40 Stratified vs control

Basivertebral-nerve ablation

Figure Estimate Population/method
SMART randomized 225: 147 ablation, 78 sham Modic 1/2 chronic axial pain, PMID 29423885
SMART baseline ODI 42 Mean age 47
SMART three-month ODI change −20.5 vs −15.2 Ablation vs sham; p=0.019 per protocol
SMART ≥10 ODI responders 75.6% vs 55.3% Ablation vs sham
INTRACEPT randomized 140 planned; 104 in interim ITT Ablation vs standard care, PMID 31229663
INTRACEPT baseline ODI 46.1 Mean age 50
Symptoms ≥5 years 67.3% Interim ITT population
Three-month ODI change −25.3 vs −4.4 Adjusted difference 20.9; p<0.001
Three-month VAS change −3.46 vs −1.02 Adjusted difference 2.44 cm; p<0.001
≥10 ODI responders 74.5% vs 32.7% p<0.001

Fusion

Source: systematic review of 26 articles and 3,060 fusion participants with ≥12-month validated outcomes (Phillips 2013, PMID 23334400).

Figure Weighted average (SD) Method
Back-pain VAS improvement 36.8/100 (14.8) Across fusion cohorts
ODI improvement 22.2 (14.1) Across fusion cohorts
SF-36 physical-component improvement 12.5 (4.3) Across fusion cohorts
Satisfaction 71.1% (5.2%) Across studies
Radiographic fusion rate 89.1% (13.5%) Across techniques
Overall reoperation 12.5% (12.4%) Across studies
Index-level reoperation 9.2% (7.5%) Across studies

Cell and biologic trials: conflicting estimates shown side by side

Trial/product Figure Population/method Source
Disc progenitor, high dose VAS −62.8% at 52 weeks 60-person four-arm randomized phase I/II PMID 38925869
Disc progenitor, high dose Absolute VAS −42.8 points Same PMID 38925869
Disc progenitor, high dose Disc volume +249.0 mm³ at 52 weeks Blinded radiological assessment PMID 38925869
Disc progenitor, high dose Disc volume +402.1 mm³ at 104 weeks Same PMID 38925869
Disc progenitor trial Severe AEs 18.3%; serious AEs 6.7% Serious events all vehicle/placebo, judged unrelated PMID 38925869
RESPINE allogeneic BM-MSC Primary responders 74% vs 69%; p=0.77 114 randomized; MSC vs sham PMID 39393844
RESPINE No between-group secondary-outcome difference 12 months PMID 39393844
MPC ± HA trial 404 randomized at 49 sites MPC, MPC+HA or saline; 36 months PMID 40174800
MPC ± HA trial Primary composite not significant All randomized participants PMID 40174800
MPC+HA subgroup opioid cessation 27.8% vs 7.8% at 36 months Baseline opioid users; duration <68-month subgroup context PMID 40174800
Earlier MPC trial 100 randomized at 13 sites 6m/18m MPC+HA vs HA/saline PMID 33045417

Cauda equina and red-flag accuracy

Figure Estimate (95% CI) Population/method
CES community incidence 0.3–0.5/100,000/year Two community populations, systematic review, PMID 32059184
CES adult community estimate 0.6/100,000/year One population, PMID 32059184
CES working-age estimate 7/100,000/year One population, PMID 32059184
CES among primary-care LBP 0.08% One study, PMID 32059184
CES among secondary-care LBP 0.27% pooled Four studies, PMID 32059184
Confirmed CES among suspected 19% 18 studies, mostly UK single-center, PMID 32059184
CES red-flag studies 7 studies; 569 participants MRI reference standard, PMID 31132655
Pooled red-flag sensitivity range 0.19–0.43 Across signs/symptoms, PMID 31132655
Pooled red-flag specificity range 0.62–0.88 Across signs/symptoms, PMID 31132655

Deepening sweep: additional discriminating estimates

Population, exposure and measurement

Figure Estimate (95% CI where available) Population/method Source
Composite substantial lumbar degeneration/abnormality 69.9% 335 population-based KORA participants; mean age 56.2 years; cross-sectional 3-T MRI PMID 36241920
KORA sample composition 43.3% women; mean BMI 27.7±4.5 kg/m² Same PMID 36241920
Daytime lumbar-disc height change −6.1% (−7.6 to −4.7) to −8.0% (−10.6 to −5.4) 12 asymptomatic adults aged 18–30; morning-to-evening MRI PMID 35072794
Daytime lumbar-disc volume change −5.4% (−7.6 to −3.3) to −8.5% (−11.0 to −6.0) Same PMID 35072794
Maximum posterior L5–S1 deformation −13.1% (−16.1 to −10.2) Same PMID 35072794
Higher occupational loading and imaging degeneration OR range 1.6–3.3 Four-study meta-analysis; level-specific estimates PMID 31656182
Patient-handling health personnel and lumbar disc disease OR 2.45 (1.41–4.26) Systematic review/meta-analysis; heterogeneous occupational controls PMID 32635557
High-intensity zone and abnormal disc morphology OR 47.79 (17.07–133.77) Meta-analysis against discographic morphology PMID 28178999
High-intensity zone and pain reproduction OR 8.65 (6.01–15.23) Meta-analysis against provocation discography PMID 28178999
Radiomics development cohort 1,397 adults; 737 extracted features Northern Finland Birth Cohort; automated segmentation PMID 40539398

Guideline implementation and treatment interpretation

Figure Estimate (95% CI where available) Population/method Source
Structured exercise: pain SMD −0.33 (−0.58 to −0.08) Eight RCTs; WHO evidence review; exercise vs no intervention PMID 37991647
Structured exercise: functional limitation SMD −0.31 (−0.57 to −0.05) Eight RCTs; adults and older adults PMID 37991647
TENS: immediate pain MD −0.90 (−1.54 to −0.26) Nine sham-controlled RCTs; approximately 2 weeks PMID 37991646
Inappropriate lumbar-imaging referrals, no red flags 34.8% (27.1–43.3) Systematic review/meta-analysis PMID 29730460
Fusion-arm repeat surgery at four years 23% Two merged RCTs; 124 total randomized PMID 19635718
Rehabilitation-arm crossover to surgery at four years 24% Same PMID 19635718
Adjacent-segment radiographic degeneration 5.9%/year (4.8–7.2) 31 studies; 4,206 fusion patients PMID 26836484
Adjacent-segment clinical disease 1.8%/year (1.3–2.4) Same PMID 26836484
Workers’ compensation return to work after fusion 26–36% Published work-related chronic-low-back-pain series summarized in cohort report PMID 24231782
Workers’ compensation reoperation after fusion 22–27% Same PMID 24231782
RMDQ patients classified improved 14–51%, depending on method 447 primary-care consulters; six-month change PMID 16360560

Red-flag discrimination

Finding Estimate Population/method Source
Major trauma for vertebral fracture LR+ 12.8; LR− 0.37 Systematic review PMID 18177783
Age >50 years for vertebral fracture LR+ 2.2; LR− 0.34 Same PMID 18177783
Bilateral sciatica for MRI-confirmed cauda-equina compression sensitivity 32.4%; PPV 17.2%; NPV 88.3% 260 suspected-CES referrals PMID 32475252
Spinal malignancy prevalence among low-back-pain cohorts 0.1–1.6% 70-study scoping review PMID 41156041

These rows are intentionally not merged with earlier estimates. They use different phenotypes, thresholds, settings and comparators.

Known conflicts and caveats

  1. Low-back-pain burden cannot be assigned to DDD because the GBD case definition is symptom-based.
  2. Asymptomatic prevalence and symptomatic association are compatible; neither establishes individual causality.
  3. Cross-sectional age prevalence is not a longitudinal progression rate.
  4. Twin heritability/familial estimates are population- and phenotype-specific.
  5. Exercise and psychological estimates come from nonspecific chronic low-back pain, not imaging-defined DDD.
  6. Pharmacologic trials are generally short relative to chronic exposure and may understate cumulative harms.
  7. Basivertebral-ablation trials used narrow Modic 1/2 criteria; sponsor involvement and crossover affect interpretation.
  8. Fusion weighted averages mix techniques, comparators and study designs and should not be treated as randomized causal estimates.
  9. Cell products are not a class: dose, source, manufacture, carrier and phenotype differ.
  10. The positive disc-progenitor result and negative RESPINE primary result should not be averaged; they test different products.
  11. CES estimates vary by denominator and definition; suspected-CES cohorts are heavily UK and single-center.
  12. MCID thresholds vary with baseline, anchor, intervention and time; threshold crossing is not a universal success definition.
  13. High-intensity-zone odds ratios use discography-linked reference constructs and are not population diagnostic likelihood ratios.
  14. Diurnal MRI changes are short-term deformation in young asymptomatic volunteers, not degeneration progression.
  15. Adjacent-segment radiographic degeneration occurs more often than clinical disease; combining them exaggerates symptomatic burden.