Skip to main content

Table 1 Risk of bias and confounding

From: Preoperative predictors for outcomes after total hip replacement in patients with osteoarthritis: a systematic review

First author, year

Follow-up (years)

Follow-up (% missing)

Looking at independent effects

Bethge, 2010 [14]

1

28.9

Age, gender and self-efficacy expectations

Clement, 2011 [43]

1

ND

Age

Clement, 2011 [47]

1

ND

Age, SF-12 scores and length of stay

Cushnaghan, 2007 [34]

Mean 8.8

48 cases

Prediction model

53 controls

Davis, 2011 [35]

5

HHS: 28

Age, gender, operating consultant, and a diagnosis of cancer, atherosclerotic disease, cardiac disease, diabetes mellitus, osteoporosis and phlebitis

SF-36: 32

Duivenvoorden, 2013 [15]

1

31

Age, gender, time spent on waiting list and unbalanced characteristics between study population and patients lost to follow-up

Gandhi, 2010 [36]

Mean 3.3

14 at 1 year follow-up

Age, gender, BMI, SF-36 Mental Health (MH) scores, method of fixation (cemented vs uncemented), and comorbidity

Gordon, 2014 [37]

1

8

Age, gender, Charnley classification, previous contralateral THR, and preoperative pain VAS

Gordon, 2014 [38]

1

8

Gender, previous contralateral hip surgery, pain, and Charnley classification

Greene, 2014 [16]

1

66.7

Bayesian model averaging with age, gender, Charnley classification, presence of comorbidities, whether the included hip was the first or second in the time interval, marital status, and education level

Fortin, 2002 [39]

2

25.7a

Age, gender, education, and comorbidity

Haverkamp, 2013 [28]

Mean 2.3

18.6a

No

Heiberg, 2013 [17]

1

27.3

Prediction model

Ieiri, 2013 [49]

1 and 3

ND

Canonical correlation analysis

Johansson, 2010 [29]

2

ND

No

Judge, 2014 [19]

1

30.8

Age, sex, SF-36 mental health, comorbidities, fixed flexion, analgesic use, college education, OA in other joints, expectation of less pain, radiological K&L grade, ASA grade, years of hip pain

Judge, 2013 [40]

each year up to 5

20 at 1 year, 30 at 5 year

Multivariable model

Judge, 2012 [48]

Mean 8

61.3b

Prediction model

Judge, 2011 [18]

1

31.6

Age, sex, school education, ASA grade, K&L grade, BMI, medication use

Katz, 2012 [27]

12

ND

Patient age, sex, race, Medicaid eligibility, comorbidity and hospital and surgeon annual THA volume

Kennedy, 2011 [44]

Up to 1.3

ND

Age, gender, body mass index

Keurentjes, 2013 [20]

1.5–6

54.1a

Age, sex, Charnley Comorbidity Classification and BMI

McHugh, 2013 [41]

1

11.7

Multivariable model

Meding, 2000 [22]

Mean 2.7

11.4 at 1 year, 37.2 at 3 years, 64.8 at 5 years and 84.3 at 7 years

Age and gender

Nikolajsen, 2006 [30]

1–1.5

6.4

No

Nilsdotter, 2003 [42]

Mean 3.6

9.6

Multivariable model

Nilsdotter, 2002 [32]

1

16.2

No

Nilsdotter, 2001 [31]

1

11.9

No

Röder, 2007 [24]

Mean 4.3

ND (registry study)

Gender, age, and follow-up year

Rolfson, 2009 [23]

1

ND (registry study)

Age, gender and comorbidity

Sadr Azodi, 2008 [25]

3

ND (registry study)

Age at the time of surgery, calendar period, and fixation principle

Sarasqueta, 2012 [21]

1

29

Prediction model

Stickles, 2001 [26]

1

ND (registry study)

No

Street, 2005 [33]

1 and 2

ND

No

Tanaka, 2010 [45]

1

ND

Age, sex, changes in the LLD, vertical ATD, femoral offset, and the horizontal and vertical position of the center of the femoral head, stage of hip OA (advanced or terminal); HHS, and the duration of OA

  1. SF-12 12-item Short Form Health Survey, HHS Harris Hip Score, SF-36 36-item Short Form Health Survey, BMI body mass index, MH mental health, THR total hip replacement, VAS visual analogue scale, OA osteoarthritis, K&L grade Kellgren-Lawrence, ASA American Society of Anesthesiologists, LLD leg length discrepancy, ATD articulotrochanteric distance
  2. ND not described or partly described
  3. aFor patients with THA and TKA, not separately described
  4. bAt 6 months