Supplementary File 2 (SF2): Characteristics of Included Studies, Presented for Each Measurement

Supplementary File 2 (SF2): Characteristics of included studies, presented for each measurement property analysed

Internal Consistency
Author/Year/Country / Type of study* / Age / Gender / Diagnosis, other disease characteristics / N= / COSMIN score and some reason for score / Findings of study / Score
Crawford 2007, USA / Type A / 45.4 y, range
18-81 / 78 f, 186 m / Meniscus injuries/pathology / 264 / GOOD. Not described, but it can be deduced how missing items were handled / Cronbach´s α : 0.773 / G/+
Fu 2010, China / Type B / 2 groups, 16-65y, and
65-85 y / 30f, 54 m / Knee injuries or OA pain, 28 with ligament injuries / 84 / GOOD. Good sample size.
Not described, but it can be deduced how missing items were handled / Cronbach´s α : 0.97 / G/-
Haverkamp 2006, The Netherlands / Type B / 54.6 y, range
21-84 / 42% m / Variety of knee problems. 62% OA, 29% meniscal lesions, 9% ligament problems / 145 / GOOD. Not described, but it can be deduced how missing items were handled / Cronbach´s α : 0.92 / G/+
Higgins 2007, USA / Type A / 37.5 y
(SD 14.9) / 73% m / Any knee dysfunction, pre/ post-surgery / 1517 / GOOD. Percentage of missing items NOT described. / Two domains of the IKDC: cronbach´s α 0.87 resp 0.88 / G/+
Irrgang 2001, USA / Type A / 37.5 y
(SD 16.2) range
6.2-86.6 / 53% (252 of 479) m, 54 cases with gender unknown / Ligamentous injuries (150), meniscal injuries (108), PFPS (93), Patellar dislocation (15), OA(92), Other conditions (22), not recorded (83) / 533 / GOOD. Not described, but it can be deduced how missing items were handled / Cronbachs α : 0.92 / G/+
Kocher 2011, USA / Type Pedi / 14.6 y
(SD 2.4)range 10.2-18.5 / 29 m(40%) 43 f (60%). / Stable knee condition, determined by physician. / 72 / GOOD. Good sample size
Not described, but it can be deduced how missing items were handled / Cronbach´s α : 0.91 / G/+
Lertwanich 2008, Thailand / Type B / 27.6 y (SD=7.3), range
18-50 / 98% m / Mainly ACL injury, meniscus injury or patellofemoral pain / 55 / GOOD. Good sample size
Not described, but it can be deduced how missing items were handled / Cronbach´s α: 0.92 / G/+
Metsavaht 2010, Brazil / Type B / 46.7 y, range
15-84 / 50% f / OA, Meniscal lesion, ACL injury, PFPS, and other knee symptoms / 117 / GOOD. Not described, but it can be deduced how missing items were handled / Cronbach´s α: 0.928 resp0.935 (for both test and retest administrations separately) / G/+
Metsavaht 2011, Brazil / Type D / 61.7 y range
35-84 / 41 f (72%),
16m(28%) / diagnosed primarily OA / 57 / GOOD. Good sample size
Not described, but it can be deduced how missing items were handled / Cronbach´s α: 0.811. / G/+
Padua 2004, Italy / Type B / 34 y, range
18-42 / 9f, 41 m / Undergoing ACL reconstruction / 50 / GOOD. Good sample size
Not described, but it can be deduced how missing items were handled / Cronbach´s α: 0.91 / G/+
Schmitt 2010, USA / Type A / 14.2 y,
range
6-18 / 361 f, 312 m / Referred to PT for evaluation and treatment of knee, mainly anterior knee pain, pain in joint, ligament sprain / 673 / GOOD. Not described, but it can be deduced how missing items were handled / Cronbach´s alpha: 0 .911. / G/+

*Type of study: Type A= investigated the measurement properties of the IKDC, Type B= investigated the measurement properties of the IKDC while translating it to another language, Type C= investigated the measurement properties of another instrument, and used the IKDC to compare it against, Type D= compared The IKDC to other measurement instruments without specifically investigating measurement properties, Type Pedi= investigated the measurement properties of the Pedi-IKDC, The IKDC developed for use in children with knee disorders.

Score: Methodological quality assessed with COSMIN score: E=excellent, G= good, F= fair, P=poor. Studies scoring Poor has not been included in the scoring of results of each measurement property, since their result does not contribute to evidence. The results of for each measurement property in each study got the scoring: += positive scoring, ?= indeterminate scoring -= negative scoring.

y= years, f= female, m= male, PT= physical therapist, ACL= anterior cruciate ligament, PFPS= patellofemoral pain syndrome, OA= osteoarthritis.

Test re-test reliability
Author/Year/Country / Type of study* / Age / Gender / Diagnosis, other disease characteristics / N= / Time interval / COSMIN score and some of the reasons for score / Findings of study / Score
Crawford 2007, USA / Type A / 50.6 y, range 19-73 / 10 f, 21 m / Meniscus injuries/pathology / 31 / 4 weeks / FAIR. Moderate sample size / ICC : 0.95 / F/+
Fu 2010, China / Type B / 65-85 y / not reported / Knee injuries or OA pain / 30 / 7-10 days / FAIR. Moderate sample size / ICC : 0.87 / F/+
Greco 2010, USA / Type A / 43.8 y (SD 10.4) range 21-60 / 62% m / Stable cohort, had undergone articular cartilage surgery several years ago. / 17 / 6 and 12 months / POOR. Small sample size / ICC: 0.91 (at 6 months) and 0.93 (at 12 months) / P
Haverkamp 2006, The Netherlands / Type B / 54.6 y, range 21-84 / 42% m / Variety of knee problems. 62% OA, 29% meniscal lesions, 9% ligament problems / 71 / 7 days / GOOD. Assumable that patients were stable.
Not described, but it can be deduced how missing items were handled / ICC: 0.96
Pearson test r=0.92 / G/+
Irrgang 2001, USA / Type A / 37.2 y (SD 10.5) range 17.4-57.4 / 39% f / Subjects participating in other long term studies (probably with some sort of knee condition) / 33 / 49.7 days
(SD 24.2, range 4-92) / FAIR. Moderate sample size / ICC: 0.94
(95% CI for a true change of 0.88-0.97) / F/+
Kocher 2011, USA / Type Pedi / 14.6 y (SD 2.4)
range 10.2-18.5 / 29 m (40%)
43 f(60%) / Stable knee condition, determined by physician. / 72 / 17 days (range 3-67) / GOOD: Good sample size.Not described, but it can be deduced how missing items were handled / ICC: 0.91
(95% CI for true change of 0.86-0.95) / G/+
Lertwanich 2008, Thailand / Type B / 27.6 y(SD=7.34) median: 26, range 18-50
(data for the entire cohort n=55) / 98.2% m
(n = 54) / Mainly ACL injury, meniscus injury or patellofemoral pain / 32 / 7days / POOR. Test conditions were NOT similar for both measurements / ICC: 0.92 / P
van Meer 2012, The Netherlands / Type A / 27y, range18-48 / 40%f / ACL injuries and reconstructions / 50 / 5 days (range 3-12) / GOOD:Good sample size. / ICC2,1,: 0.93 (95% CI: 0.89-0.96) / G/+
Metsavaht 2010, Brazil / Type B / 46.7 y,
range 15-84
(data for the entire cohort) / 50% f / OA, Meniscal lesion, ACL injury, PFPS, and other knee symptoms / 85 / 7 days / GOOD. Good sample size. / ICC: 0.988 (95% CI: 0.982-0.992) / G/+
Padua 2004, Italy / Type B / 34 y, range 18-42. (data for the entire cohort) / 9f, 41 m / Patients undergoing ACL reconstruction / 20 / 5 days / POOR. Poor sample size / ICC: 0.90 (p<0.001) / P
Piontek 2012, Poland / Type B / Not reported / 26 f / ACL reconstruction >2 years ago / 30 / 2 days-2 weeks / POOR. No ICC, Spearman or Pearson test calculated. / P
Siqueira 2012, / Type D / 18-39 y, mean 24.29, SD 4.09 / 27 f, 4 m / Patients with PFPS / 31 / 2 days / FAIR.Moderate sample size.
Pearson or Spearman correlation coefficient calculated WITHOUT evidence provided that no systematic change has occurred. / Spearman: r=0.96 / F/+

y=years, f=female, m=male,ACL= anterior cruciate ligament, PFPS= patellofemoral pain syndrome, OA= osteoarthritis, CI= Confidence Intervall, Q= question, no= number.

Measurement Error
Author/Year/Country / Type of study* / Age / Gender / Diagnosis, other disease characteristics / N= / Time interval / COSMIN score and some of the reasons for score / Findings of study / Score
Greco 2010, USA / Type A / 43.8 y
(SD 10.4) range 21-60 / 62% m / Stable cohort, had undergone articular cartilage surgery several years ago. / 17 / 6 and 12 months / POOR. Small sample size
. / SEM: 5.6 (at 6 m), 4.9 (at 12 m)
MDC: 15.6 (at 6m), 13.7 (at 12m) / P
Irrgang 2001, USA / Type A / 37.2 y
(SD 10.5) range 17.4-57.4. / 39.4% f / Subjects participating in other long term studies (probably with some sort of knee condition) / 33 / 49.7 days (SD 24.2, range 4-92) / FAIR. Moderate sample size / SEM: 4.6 (95% CI for a true change of +-9.0) / F/?
van Meer 2012,The Netherlands / Type A / 27y, range18-48 / 40% f / ACL injuries and reconstructions / 50 / 5 days (range 3-12) / GOOD.Good sample size / SEM: 4.4
SDCind: 12.2 / G/?
Metsavaht 2010, Brazil / Type B / 46.7 y, range 15-84 / 50.4% f
(data for the entire cohort) / OA, Meniscal lesion, ACL injury, PFPS, and other knee symptoms / 85 / one week / GOOD. Good sample size / Mean error: 0.50 (SD= 3.37, 95%, LoA: -6.1 to 7.1). / G/?

m=male, f=female, SEM= standard error of measurement, SDCind= smallest detectable change individuals, OA= Osteoarthritis, ACL= anterior cruciate ligament, PFPS= Patellofemoral Pain Syndrome.

Content Validity
Author/Year/Country / Type of study* / Age / Gender / Diagnosis, other disease characteristics / N= / COSMIN score and some reasons for score / Findings of study / Score
Irrgang 2001, USA / Type A / 37.5 y
(SD 16.2) range 6.2-86.6 / 52.5% (252 of 479) m, 54 cases with gender unknown / Ligamentous injuries (150), meniscal injuries (108), PFPS (93), Patellar dislocation (15), OA (92), Other conditions (22), not recorded (83) / 533 / GOOD. Assessment of whether all items are relevant for the purpose of the measurement instrument (discriminative, evaluative and/or predictive) was not described but assumed / All items were considered to be relevant for the construct to be measured and for the target population / G/+
van Meer 2012, The Netherlands / Type A / 26
range 18-57 / 25% f / ACL injuries and reconstructions / 19 expert panel, 26 patients / EXCELLENT. / 89%relevant items (16 of 18) / E/+

OA= Osteoarthritis, PFPS= PatellofemoralPain Syndrome.

Structural Validity
Author/Year/Country / Type of study* / Age / Gender / Diagnosis, other disease characteristics / N= / COSMIN score and some reasons for the score / Findings of study / Score
Higgins 2007, USA / Type A / 37.5 y (SD= 14.9) / 73% m / Any knee dysfunction, pre- or post surgery / 1517 / GOOD. Percentage of missing items NOT described / Two factors; symptom and knee articulation (11 items) and activity level (4 items) / G/?
Irrgang 2001, USA / Type A / 37.5 y
(SD 16.2) range 6.2-86.6 / 52.5% (252 of 479) m, 54 cases with gender unknown / Ligamentous injuries (150), meniscal injuries (108), PFPS (93), Patellar dislocation (15), OA (92), Other conditions (22), not recorded (83) / 533 / GOOD. Not described, but it can be deduced how missing items were handled / One factor that accounted for 50.2% of the total variance. / G/+
Schmitt 2010, USA / Type A / 14.2 y,
range 6-18 / 361 f, 312 m / Referred to PT. for evaluation and treatment of knee, mainly anterior knee pain, pain in joint, ligament sprain. / 673 / GOOD. Not described, but it can be deduced how missing items were handled / One factor that accounted for 48% of the total variance / G/+

y=years, m=male, f=female, PT= physical therapist, PFPS= Patellofemoral Pain Syndrome, OA= Osteoarthritis, FA= Factor Analysis

Hypothesis testing
Author/Year/Country / Type of study* / Age / Gender / Diagnosis, other disease characteristics / N= / COSMIN score and some of the reason for the score / Findings of study / Score
Agel 2009, USA / Type D / 31 y,
range 14-62 / not reported / Proximal tibial opening wedgeosteotomies, meniscal transplants, refrigerated osteoarticularallografts, and multiple ligament knee reconstructions / n=130
(total 490 administrations at different times) / POOR. No information on the measurement properties of the comparator instrument(s) / Correlation to the Modified Cincinnati.
100% of hypothesis confirmed / P
Björklund 2009, Sweden / Type C / 27 y,
range 18-50 / 13 f, 22 m / ACL-reconstructed, postop 4 and 8 months / 35 / FAIR. Moderate sample size / Correlation to TAK test.
100% of hypothesis confirmed / F/+
Briggs 2009, USA / Type C / 37 y,
range 18-77 / 749 f, 1034 m / ACL injury - preoperatively / 1783 / GOOD. Not described, but it can be deduced how missing items were handled / Correlation toLysholm(confirmed) and to Tegner (not confirmed).
50% of hypothesis confirmed / G/-
Cook 2010, USA / Type D / 49.3 y (SD 13.5) / 17f, 25 m / Anterior knee pain more than 3 months / 52 / GOOD. Expected direction of the correlations or differences NOT stated / Correlation to clinical tests for PFPS.
0% of hypothesis confirmed. / G/-
Crawford 2007, USA / Type A / 45.4 y
range 18-81 / 78f , 186 m / Meniscus injuries/pathology / 264 / FAIR. Poor description of the constructs measured by the comparator instrument(s) / Correlation to 7 hypothesis, i.e. Tegner activity scale, workers compensation claim, difficulties with ADL, work and sport, assessment of overall knee function, meniscus tears.
100% of hypothesis confirmed / F/+
Haverkamp 2006, The Netherlands / Type B / 54.6 years, range 21-84 / 42% male / Variety of knee problems. 62% OA 29% meniscal lesions, 9% ligament problems / 145 / POOR. No description of the constructs measured by the comparator instrument(s) / Correlation to VAS pain, Oxford 12 Questionnaire, WOMAC.
100% of hypothesis confirmed / P
Kocher 2011, USA / Type Pedi / 14.6 y (SD 2.5) range 10.0-18.9 / 288 m (48.9%), 301 f (51.1%) / Children with knee complaints. 16% previous knee disorders, 84% no previous knee disorders. / 589 / GOOD. Expected magnitude of the correlations or differences NOT stated. Adequate measurement properties of the comparator instrument(s) but not sure if these apply to the study population / Correlations to eleven hypothesis developed from the CHQ.
100% of hypothesis confirmed / G/+
Kong 2012, Korea / Type C / 23.43 y (SD 3.17) and 24.73y ( SD 2.16) for each group / 100% m / ACL reconstructed and healthy controls / 60 / POOR. Unclear what was expected, regarding Hypothesis a priori / Correlation to three functional performance tests (co-contraction test, shuttle run, carioca test). Unclear hypotheses / P
Laboute 2009, France / Type C / 27.4 y / 150 f,
613 m / Short and long time follow-up after ACL reconstruction. / 613 / POOR. Unclear what was expected in hypothesis, regarding correlations or mean differences / Correlation to PPLP scoring scale.Unclear hypotheses / P
van Meer, 2012
The Netherlands / Type A / 25y,
range 18-57 / 25% f / Patients with an acute ACL injury or pre op ACL reconstruction. / 100 / GOOD. No reference about MP for VAS and patient-rated improvement of stability. / Correletion to VAS- pain (confirmed), Lysholm (confirmed) and “patient rated improvement for instability” (not confirmed).
66% of hypothesis confirmed. / G/-
Metsavaht 2010, Brazil / Type B / 46.7 y,
range 15-84 / 50% f / OA, Meniscal lesion, ACL injury, PFPS, and other knee symptoms / 117 / FAIR. Some information on MP (or a reference to a study on MP) of the comparator instrument(s) in any study population.
Hypothesis vague or not formulated, but possible to deduce what was expected / Correlation toLysholm score and WOMAC.
100% of hypothesis confirmed, but vague hypothesis. / F/+
Metsavaht 2011, Brazil / Type D / 61.7 y,
range 35-84 / 41 f (72%), 16 m (28%) / Diagnosed primary OA / 57 / FAIR. Hypothesis vague or not formulated, but possible to deduce what was expected / Correlation to WOMAC and Lysholm, but a proportional bias between IKDC and WOMAC, and IKDC and Lysholm.
Hypothesis graded as not confirmed. / F/-
Metsavaht 2012, / Type C / 52.1 y, SD 15.1, range 15-78. / 29% m / Patients undergoing treatment for a variety of knee conditions (OA, ACL, meniscus, PFSS etc.) / 87 / GOOD.Not described but it can be deduced how missing items were handled.
Expected magnitude of the correlations or differences NOT stated. / Correlation to the LEFS.
100% of hypothesis confirmed / G/+
Reinke 2011, USA / Type D / 16-35y / 41f, 28 m / Between 24-39 months after ACL reconstruction / 69 / FAIR. Hypothesis vague or not formulated, but possible to deduce what was expected / Correlation to four hop-tests.
75% of hypothesis confirmed / F/+
Shelbourne 2013, USA / Type D / ACL rec 35 y, SD 12, range 13-73. Arthroscopy 47y, SD 15, range 11-88. / ACL-rec.: 57.4% m, Arthroscopy 69.4% m / ACL-reconstructed patients and patients undergoing knee arthroscopy / ACL rec: 3209
Arthroscopy: 1813 / POOR. No measurement properties of comparator instrument reported / Correlation to SANE confirmed.
100% of hypothesis confirmed / P
Siqueira 2012, Brazil / Type D / 18-39 y, mean 4.29, SD 4.09 / 27 f, 4 m / Patients with PFPS / 31 / POOR.Unclear if hypotheses were formulated a priori and what was expected. / Correlation to KOS.
Unclear hypothesis / P