Oswestry Disability Index

Low back pain:

(Copay et al, 2008; n = 457, mean age = 54.4 years, average BMI = 29.9 kg/m2, 59.1% female; patients were selected from a Lumbar Spine Study Group with scores at baseline and at 1 year; most frequent diagnoses were spinal stenosis (28.5%), lumbar disc pathology/degeneration (15.8%), spondylolisthesis (14.7%); patients either went under decompression surgery (11.5%) or fusion surgery (88.5%); study is to determine the clinically and statistically appropriate MCID value for patients who underwent lumbar spinal surgery when utilizing anchor-based and distribution-based methods)

• When using the Health Transition Item (HTI) as an anchor, SEM = 4.62
• When using the Satisfaction with Results Scale as an anchor, SEM = 4.21

(Dawson et al, 2010; Cohort 1: n = 214, age = 25.9 (9.0), weight (kg) = 68.1 (14.6), 88.3% female, 49.5% with annual upper back pain, 92.1% with annual lower pain; Cohort 2: n = 33, age = 23.2 (6.5), weight = 68.0 (12.2), 90.9% female, 52.3% with annual upper back pain, 94.0% with annual lower back pain; patients were enrolled as Bachelor of Nursing students in Australia; Cohort 2 is a subsection of Cohort 1, having completed the ODI Version 2.0 on two separate occasions instead of one)

• For students reporting back pain (BP) in the previous year (n = 214), SEM = 2.56% (90% CI)
• For students reporting BP on the day of the study (n = 82): SEM = 2.77% (90% CI)

(Grotle et al, 2012; n = 87 for control group and n = 42 for test retest group, mean age = 35 (5.0); patients recruited from primary care clinics in Norway using the following inclusion criteria: pelvic girdle pain located distal, lateral, or both in relation to the L5 - S1 area, in the buttocks, symphsis or both, with pain onset during pregnancy or within 3 weeks after delivery; diagnostic criteria determined by the following tests: Posterior Pelvic Pain Provocation Test, Active Straight Leg Raising Test, pain provocation of the long dorsal sacroiliac ligament, and pain provocation of the symphysis by palpation and by a modified Tredelenburg test)

• SEM = 4.02

(Johnsen et al, 2013; n = 113, age = 25 - 55, mean age = 41, 52.6% female, mean duration of low back pain = 5.74 years; inclusion criteria: age between 25 - 55, lower back pain for over a year, degenerative changes in the intervertebral disc in one of the two lowest levels of the lumbar spine and an Oswestry Disability Index of equal to or greater than 30%; exclusion criteria: generalized chronic pain syndrome and degeneration established in more than two levels; patients randomized to surgery with insertion of artificial disc or non-surgical, multidisciplinary back rehabilitation program; only those outcomes of patients who completed the SF6D, EG5D and ODI at baseline and at 2 year follow up were included)

• SEM = 4.24 for patients stable for a period of 3 months

(Miekisiak et al, 2013; n = 164 with 70 men and 94 women, age = 49.7 (11.8); n = 84 (52%) provided retest questionnaires within 2 - 14 days of baseline data; patients were recruited as surgical candidates and inclusion crtiteria included chronic LBP with or without radiation to the leg, 18 years or older, and good comprehension of the Polish language; a culturally, translated version of the ODI was used, after review by a committee including one neurosurgeon, one psychologist, one physical therapist, and a translator)

• SEM = 3.54 (2.62 - 4.79)

Low back pain:

(Copay et al, 2008)

• Using the Health Transition Item (HTI) as an anchor, MDC = 12.81 (CI = 95%)
• Using the Satisfaction with Results Scale as an anchor, MDC = 11.67 (CI = 95%) 

(Dawson et al, 2010) 

• For students reporting back pain (BP) in the previous year (n = 214), MDC90 = 5.94%
• For students reporting back pain (BP) in the previous year (n = 82), MDC90 = 6.43% 

(Grotle et al, 2012) 

• Individual MDC = 11.1
• Group MDC = 1.7

(Johnsen et al, 2013) 

• MDC = 11.74, CI = 95%

(Miekisiak et al, 2013)

• For the total sample population (MDC95 = 10)

(Monticone et al, 2012; n = 179, 62.6% female; patients recruited from rehabilitation centers between 2009 - 2010; inclusion criteria included diagnosis of sub-acute or chronic common lower back pain, age 18 - 70 years, and the ability to read and speak fluent Italian; exclusion criteria were acute common lower back pain and specific causes of lower back pain including disc herniation, canal stenosis, and with recent myocardial infarctions, cerebrovascular events)

• For the total sample population (MDC = 13.67)
• For the population with subacute lower back pain (MDC = 15.35)
• For the population with chronic lower back pain (MDC = 12.72)

Low back pain:

(Copay et al, 2008)

• MCID = 12.8 (2.92 - 15.36), based on different calculation methods yielding widely different threshold values

(Johnsen et al, 2013)

• MIC = 12.88 with 88% sensitivity and 85% specificity

(Glassman et al, 2006; n = 497, with 270 females and 227 males, age = 47 years (17 - 86); inclusion criteria included one (65%) or two (35%) level lumbar spinal fusions with preoperative baseline data and at least 1 year post-operative data with many contributing also 2 year post-operative data; 131 patients also had a prior decompressive surgical procedure)

• FDA standards for good to excellent operative outcomes include a 15 point improvement in ODI, plus maintenance or improvement in SF - 36 score. Older studies indicate that a 4 or 10 point decrease might identify a significant improvement

(Monticone et al, 2012)

• With the MCID taken as the point on the ROC curve closest to the upper left corner of the figure, the MCID is 9.5 (AUC = 0.71, sensitivity = 76%, specificity = 63%)

Lower back pain:

(Cho et al, 2012; n = 166; age = 53.8 years with follow up 3.5 years (2 - 7 years); patients underwent multilevel revision surgery for spinal deformity with a minimum 2 year follow up; primary diagnosis included de novo scoliosis, degenerative, trauma, neuromuscular scoliosis, congenital deformity, ankylosing spondylitis, tumor, Scheuerman kyphosis, and rheumatoid arthritis)

• Comparison of ODI scores at ultimate follow-up: follow-up complication, mean ODI = 37.2; no complication, mean ODI = 26.6

• Assessing the difference between final and preoperative ODI scores: no or perioperative major complication groups, mean ODI = 17.9; follow up complication group, mean ODI = 9.5

(Copay et al, 2008)

• ODI at baseline = 52.50 (14.06)

• ODI after 1 year = 37.60 (21.50)

(Dawson et al, 2010) 

Additionally, Cohort 1: 

• For students with non-serious back pain (n = 80), as defined by not requiring sick leave and/or treatment, ODI = 7.3% (6.2), range (0 - 24)% 

• For students with serious back pain (n = 134), as defined by requiring sick leave and/or reatment, ODI = 9.7% (7.9), range (0 - 38)% 

(Frost et al, 2008; n = 201, age = 42.5; patients were invited to take part if they were 18 years old and over with at least a 6 week history of low back pain or without leg pain or neurologic signs; patients were excluded if they present with serious pathologies (ie. systemic rheumatological, gynecological problems, ankylosing), were unable or unwilling to complete the questionnaires independently, had received physical therapies, or were referred for intensive functional restoration programs, data assessed at baseline and 12 months)

(Glassman et al, 2006; n = 497; mean age = 47 (17 - 86)

• 1-year post-op (n = 413): mean 22.2 pt improvement

• 2-year post-op (n = 141): mean 22.1 pt improvement 

• Prior decompression surgery (n = 131): baseline, mean 54.4; 1-year post-op, mean 17.5 pt improvement; 2-year post-op: mean 16.6 pt improvement

• No prior decompression surgery (n = 366): baseline, mean 49.8; 1-year post-op, mean 21.7 pt improvement; 2-year post-op, mean 24.4 pt improvement 

(Grotle et al, 2012)

(Monticone et al, 2012; n = 179, age = 47.7 (12.3)

• Baseline ODI mean score: 26.8 (16.6)

• ODI mean score (after 8 week exercise program): 17.9 (16.0) 

(Wittink et al, 2004)

• Mean ODI score = 45.16 (19.56)

(Miekisiak et al, 2013)

• Mean ODI score = 48.45 (18.94), range = 9 - 24

Low back pain:

(Dawson et al, 2010)

• Excellent test retest reliability (n = 33) for ODI subsections of walking (ICC = 0.78 (0.60 - 0.88)), sleep (ICC = 0.82 (0.67 - 0.91)), and total ODI score (ICC = 0.88 (0.77 - 0.94)), all with CI 95% 
• Adequate test retest reliability (n = 33) for ODI subsections of pain intensity (ICC = 0.65 (0.40 - 0.81)), lifting (ICC = 0.74 (0.53 - 0.87)), sitting (ICC = 0.71 (0.48 - 0.85)), standing (ICC = 0.59 (0.31 - 0.78)), social life (ICC = 0.52 (0.22 - 0.73), and travel (ICC = 0.51 (0.21 - 0.73))
• Poor test retest reliability (n = 33) for ODI subsection of sex (ICC = 0.25 (-0.11 - 0.55))
• Test retest reliability not established for personal care

(Gronbald et al, 1993; n = 94, age = 42.7 (10.3) years, range 18 - 79; inclusion criteria including low back pain with or without radiation to the legs for at least 2 months, indicating a chronic back pain state; exclusion criteria including major disease, back pain > 3 months) 

• Excellent test retest reliability when test-retest interval is 1 week (ICC = 0.83)

(Grotle et al, 2012)

• Excellent test retest reliability (ICC= 0.94 (0.89 - 0.97); 95% CI)

(Miekisiak et al, 2013) 

• Excellent test retest reliability (ICC = 0.97 (0.94 - 0.98); 95% CI)

Low Back Pain:

(Grotle et al, 2012)

• Excellent internal consistency for overall ODI score (Cronbach’s alpha = 0.83)
• Adequate internal consistency for ODI traveling section-total score correlation (Cronbach’s alpha = 0.72)
• Poor internal consistency for ODI subsections-total score correlation: pain (Cronbach’s alpha = 0.50), personal hygiene (0.61), lifting (0.58), walking (0.61), sitting (0.34), standing (0.51), sleeping (0.42), sexual activity (0.53), social activity (0.44)

(Miekisiak et al, 2013) 

• Excellent internal consistency (Cronbach’s alpha = 0.90) 

(Wittink et al, 2004; n = 87 and age = 46.9 years; included patients classified according to six most common ICD9 diagnostic groups by anatomic site (head and neck, upper or lower abdominal, low back) and/or pathophysiology (complex regional pain syndrome of myofascial pain) 

• Excellent internal reliability (Cronbach’s alpha = 0.86); this figure is better than previously reported in the literature

Low back pain: 

(Copay et al, 2008)

• Poor correlation between baseline and change scores on ODI (r = 0.11)
• Adequate correlation between HTI anchor and ODI change (rho = 0.35) and between Satisfaction with results anchor and ODI change (rho = 0.46)

(Frost et al, 2008)

• Excellent correlation between improved vs nonimproved patients (ROC = 0.75 (0.68-0.82), 95% CI)

(Johnsen et al, 2013)

• Adequate correlation between ODI and EQ5D baseline scores (r = 0.58) and between ODI and SF6D baseline scores (r = 0.38)

(Wittink et al, 2004) 

Overlap of Multidimensional Personality Inventory (MPI) with the ODI Instruments in R2 values

Overlap of SF36 Domains with the ODI Instruments in R2 values

Low Back Pain:

(Frost et al, 2008)

• Adequate correlation between transition rating index and the ODI (version 2.1) (r = 0.47)

(Grotle et al, 2012)

• Excellent correlation with: Pelvic Girdle Questionnaire (PGQ) total (r= 0.75), PGQ Activity subscale (r = 0.72), PGQ Symptom subscale (r= 0.71), Disability Rating Index (r = 0.71), SF2 - Physical Functioning (r = 0.66)
• Adequate correlation with: Fear-avoidance beliefs questionnaire (r = 0.33), SF1 - General Health (r = 0.51), SF3 - Role-physical (r = 0.54), SF4 - Bodily pain (r = 0.56), SF6 - Social functioning (r = 0.52) 
• Poor correlation with: Pain catastrophizing scale (r = 0.26), SF5 - Vitality (r = 0.07), SF7 - Mental Health (r = 0.12), SF8 - Role-emotional (r = 0.17)
• Adequate discriminant validity between ODI and pain localization (ROC = 0.726, 95% CI, 0.621 - 0.830, p = 0.015)
• Poor discriminant validity between ODI and pregnancy (ROC = 0.63, 95% CI, 0.535 - 0.769, p = 0.001) 

(Haro et al, 2008; n = 42, age 66.8 (10.9); patients were surgical patients followed for more than 2 years; surgical indications included no response to conservative treatment and neurological deterioration; neurological symptoms were classified as nerve root type, cauda equine type or combined type; surgical interventions included decompression at the location of the dural or root indentation by myelography)

• Excellent correlation between the visual analogue scale (VAS) of low back pain with ODI V2, preoperation and postoperation respectively (r= -0.71 and r = -0.75)

(Little and MacDonald et al, 1994; n = 144; patients had all received lumbar spinal surgery to operatively manage various diagnosis such as spinal canal stenosis, prolapsed intervertebral discs, low back pain)

• Excellent correlation between patients’ subjective assessment of improvement postoperatively and change in ODI (r = 0.61)

(Miekisiak et al, 2013)

• Excellent correlation between the Polish ODI and RMDQ (r = 0.607, p < 0.001)
• Adequate correlation between VAS score for leg pain and ODI (r = 0.56, p < 0.001)

Low back pain: 

(Copay et al, 2008)

• MDC was selected as the most appropriate MCID threshold value by comparing potential MCID value calculations and verifying with two different anchors (global health assessment, the Health Transition Item, and rating of satisfaction with the results of the surgery, the Satisfaction Results Scale)

Low back pain: 

(Dawson et al, 2010)

• Adequate floor effects (13% scored ODI = 0)
• 36% of student nurses scored less than MDC90, which is outside of the recommended 15% maximum

(Frost et al, 2008)

• Adequate floor effect with baseline data, with 16 (6%) subjects scoring less than 8%, including 2 subjects who scored 0

(Grotle et al, 2012)

• Excellent, no floor or ceiling effects reported

(Little & MacDonald et al,1994)

• Possible floor effect as researchers assumed that participants that were worse did not feel comfortable to admit it

(Miekisiak et al, 2013) 

• Excellent, no floor or ceiling effects reported

Low back pain:

(Copay et al, 2008)

• Small effect size (r = 0.11)

(Frost et al, 2008)

• Large effect size for improved patients using the ODI scale, -0.88 to -1.00, and worse patients, 0.77

• Small effect size for unchanged patients using the ODI scale, 0.007 

• Low Back Pain: SF-36 effect size to the ODI comparing each domain, analyzing 81 patients during a 5-week period:

(Johnsen et al, 2013) 

• Excellent correlation between change scores of ODI and EQ5D (r = 0.64) and between changes scores of ODI and SF6D (r = 0.77)

• Excellent correlation between change scores of ODI and global score categories (rho = 0.84) and between change scores of SF6D and global score categories (rho = 0.76)

• Adequate correlation between change scores of EQ5D and global score categories (rho = 0.55)

• Excellent discrimination between improved and non-improved patients for ODI (ROC = 94% (87.5 - 97.6), CI = 95%) and for SF6D (ROC = 90% (82.1 - 94.6), CI = 95%)

• Adequate discrimination between improved and non-improved patients for EQ5D (ROC = 83% (75 - 90), CI = 95%)