Table 4

Regression models clustered by surgeon: adjusted beta values for the linear regression model for WT2 and adjusted OR for the logistic regression model for proportion of patients receiving surgery within benchmark waiting time (26 weeks)

Linear regression model for WT2
Explanatory variableHipKnee
Beta value (SE)P valueBeta value (SE)P value
Age−0.00151 (0.000908)0.09680.000236 (0.000827)0.776
Sex−0.0106 (0.0205)0.6070.000489 (0.0155)0.975
Period (pre-WCIS vs post-WCIS)−0.0185 (0.0211)0.381−0.0754 (0.0156) < 0.001
BMI0.00259 (0.00155)0.0941 0.00116 (0.00102)0.253
Pre-operative Oxford-12 score−0.00779 (0.00138) < 0.001 −0.00314 (0.00103) 0.0024
Medical comorbidity−0.0858 (0.0398) 0.0312 −0.00099 (0.0344)0.977
Pre-operative SF-12 mental component summary−0.00027 (0.000851)0.753 0.000038 (0.000673)0.955
Logistic regression model for proportion receiving surgery within benchmark waiting time
Explanatory variableHipKnee
OR (95% CI)P valueOR (95% CI)P value
Age1.01 (1.00 to 1.03)0.06511.00 (0.99 to 1.01)0.841
Sex1.04 (0.78 to 1.40)0.7760.88 (0.69 to 1.12)0.308
Period (pre-WCIS vs post-WCIS)1.36 (1.00 to 1.84)0.05031.39 (1.09 to 1.78) 0.0084
BMI0.97 (0.95 to 0.996) 0.0227 1.00 (0.98 to 1.01)0.635
Pre-operative Oxford-12 score1.05 (1.03 to 1.07) < 0.001 1.02 (1.01 to 1.04) 0.0046
Medical comorbidity1.56 (0.86 to 2.84)0.1431.48 (0.84 to 2.59)0.175
Pre-operative SF-12 mental component summary1.00 (0.99 to 1.01)0.9811.00 (0.99 to 1.01)0.456
  • L inear regression: n=882 (hip; 54.9% missing); 1398 (knee; 48.7%); missing data were treated as missing.

  • Continuous data were log-transformed to reduce the skewness of the distribution (dependent variable=logWT2); regressed using a mixed linear model.

  • The dependent variable was waiting time 2 (time elapsed between decision to undergo surgery and surgery taking place; reported in number of weeks).

  • Age (years), BMI, Oxford-12, SF-12 mental component summary are all continuous variables.

  • Medical comorbidity: no=0; yes=1; reflective of 13 conditions.

  • Medical comorbidities: heart disease, high blood pressure, lung disease, diabetes, ulcer or stomach disease, kidney disease, liver disease, anaemia or other blood disease, cancer, depression, osteoarthritis or degenerative arthritis other than hip or knee, back pain, rheumatoid arthritis, other medical problem.

  • MCS-12—short form-12 mental component summary: range 0 (worst) to 100 (best).

  • WT2=β0+β1Age+β2Sex+β3Period+β4BMI+β5Oxford-12+β6MedicalComorbidity+β7MCS (clustered by Surgeon).

  • The variation of the random effect for hip is 0.01380 and for knee is 0.01413.

  • −2 Res log likelihood: 433.6 (hip); 460.3 (knee).

  • L ogistic regression: n=1002 (hip; 56.7% missing); 1571 (knee; 53.1% missing); missing data reported where missing.

  • Continuous data were log-transformed to reduce the skewness of the distribution; regressed using a generalised linear mixed model.

  • Regression was modelled based on the likelihood of meeting the 26-week clinically recommended benchmark (benchmark=1).

  • The dependent variable was benchmark (patient did not receive surgery within benchmark=0; patient did receive surgery within benchmark=1).

  • Age (years), BMI, Oxford-12, SF-12 mental component summary are all continuous variables.

  • Medical comorbidity: no=0; yes=1; reflective of 13 conditions;.

  • Medical comorbidities: heart disease, high blood pressure, lung disease, diabetes, ulcer or stomach disease, kidney disease, liver disease, anaemia or other blood disease, cancer, depression, osteoarthritis or degenerative arthritis other than hip or knee, back pain, rheumatoid arthritis, other medical problem.

  • MCS-12—short form-12 mental component summary: range 0 (worst) to 100 (best).

  • Benchmark=β0+β1Age+β2Sex+β3Period+β4BMI+β5Oxford-12+β6MedicalComorbidity+β7MCS (clustered by Surgeon).

  • The variation of the random effect for hip is 2.16 and for knee is 1.69.

  • −2 Res log pseudo-likelihood: 4653 (hip); 7249.4 (knee).

  • MCS, mental component summary; SF-12, short form-12; WCIS, Winnipeg Central Intake Service; WT2, waiting time 2.