Skip to main content

Table 1 Demographic characteristics of matched and unmatched cohorts

From: Thymectomy for non-thymomatous myasthenia gravis: a propensity score matched study

Characteristics Unmatched (n = 395) Matched (n = 98)
  Thymectomy (n = 183) Controls (n = 212) Absolute standardized difference Thymectomy (n = 49) Controls (n = 49) Absolute standardized difference
Age (mean ± SD) 34.8 ± 14.1 63.7 ± 12.7* 2.33 49.8 ± 14.3 50.1 ± 13.2 0.02
Females [n (%)] 124(68) 89(42)* 2.03 25(51) 27(50) 0.08
Time to Dx (mean ± SD) 14.8 ± 26.1 16.9 ± 31.8 0.55 17.8 ± 25.6 18.3 ± 27.3 0.02
MGFA class at diagnosis n [(%)]
 MGFA I 16(9) 31(15)* 0.21 5(10) 4(8) 0.07
 MGFA II 57(31) 81(38) 0.15 21(43) 23(47) 0.08
 MGFA III 70(38) 73(34) 0.08 17(35) 15(31) 0.08
 MGFA IV/V 40(22) 27(13)* 0.22 6(12) 7(14) 0.05
Medications [n(%)]
Prednisone 130(71) 166(78) 0.16 35(71) 33(67) 0.09
Azathioprine 109(60) 118(56) 0.08 30(61) 31(63) 0.04
Mycophenolate 32(18) 41(19) 0.05 6(12) 7(14) 0.05
Other immunosupressants 28(15) 21(10) 0.15 4(8) 5(10) 0.07
Follow-up (mean ± SD) 118.9 ± 115.9 67.4 ± 52.8* 0.44 95.9 ± 103.1 76.4 ± 54.6 0.16
  1. Continuous data are expressed as median and range.
  2. Nominal data are expressed as number and proportion of patients.
  3. MGFA: Myasthenia Gravis Foundation of America. The proportion of patients in each MGFA class at diagnosis, before treatment, is presented for both groups.
  4. Time to Diagnosis (Dx) and Follow-up are in months.
  5. Other immunosupressants include: methotrexate, rituximab, cyclosporine and cyclophosphamide.
  6. Absolute Standardized Difference > 0.1, indicating poor balance of the variable between groups.
  7. *p < 0.05. p-values are less sensitive than the Absolute Standardized Difference to detect imbalance.
  8. The matched dataset has excellent balance of all the variables, except mild residual imbalance of follow-up times. This was accounted with time-to-event analyses.