Skip to main content
Journal of Patient-Reported Outcomes

Table 6 Impact of aggregate DIF within the EE, DP, and PA subscales on individual physicians’ subscale scores and burnout symptom prevalence estimates a

From: Examining the measurement equivalence of the Maslach Burnout Inventory across age, gender, and specialty groups in US physicians

Scale- DIF Grouping variable

Reference (R) group; Focal (F) group (Sample n)

Mean absolute difference (SD) between individual physicians’ latent burnout symptom score from DIF unadjusted and adjusted multi-group IRT models

Correlation between individual physicians’ latent burnout symptom score from DIF unadjusted and adjusted multi-group IRT models

Prevalence of EE/DP/low PA from DIF-unadjusted Multi-group IRT Model, % (n)

Prevalence of EE/DP/low PA from DIF-adjusted Multi-group IRT Model, % (n)

Absolute difference in EE/DP/low PA prevalence estimates between DIF- unadjusted and adjusted multi-group IRT model

EE – Sex

Male (R); Female (F) (n = 6083)

0.00 (0.01)

0.9999

46.7% (2842)

46.6% (2835)

0.1%

EE – Age Group

≥65 years (R); 55–64 years (F) (n = 3271)

0.00 (0.01)

0.9999

40.8% (1336)

40.7% (1330)

0.1%

EE – Specialty

GIM (R); Emergency medicine (F) (n = 744)

0.01 (0.02)

0.9997

49.2% (366)

49.1% (365)

0.1%

GIM (R); General pediatrics (F) (n = 762)

0.01 (0.02)

0.9998

47.6% (363)

47.9% (365)

0.3%

GIM (R); Neurology (F)

0.01 (0.03)

0.9995

49.3% (318)

49.1% (317)

0.2%

GIM (R); Pediatric subspecialty (F) (n = 717)

0.01 (0.01)

0.9999

43.5% (312)

43.2% (310)

0.3%

DP – Specialty

GIM (R); Family medicine (F) (n = 946)

0.03 (0.04)

0.9991

39.5% (374)

40.0% (378)

0.5%

GIM (R); General pediatrics (F) (n = 784)

0.05 (0.07)

0.9977

33.7% (264)

33.7% (264)

0.0%

GIM (R); Neurology (F) (n = 672)

0.02 (0.05)

0.9986

37.8% (254)

38.1% (256)

0.7%

GIM (R); Obstetrics and gynecology (F) (n = 708)

0.04 (0.05)

0.9984

37.4% (265)

37.7% (267)

0.3%

PA – Age Group

≥65 years (R); 45–54 years (F) (n = 2540)

0.02 (0.03)

0.9996

13.6% (346)

13.6% (346)

0.0%

PA – Specialty

GIM (R); Anesthesiology (F) n = 642)

0.03 (0.05)

0.9987

18.5% (119)

18.2% (117)

0.3%

GIM (R); Emergency Medicine (n = 761)

0.01 (0.03)

0.9996

19.2% (146)

18.9% (144)

0.3%

GIM (R); General surgery subspecialty (F) (n = 766)

0.03 (0.03)

0.9993

14.9% (114)

14.9% (114)

0.0%

GIM (R); Neurology (F) (n = 653)

0.02 (0.06)

0.9981

14.1% (92)

14.4% (94)

0.3%

GIM (R); Psychiatry (F) (n = 923)

0.02 (0.02)

0.9998

13.2% (123)

13.0% (121)

0.2%

  1. a Calculated for all analyses resulting in significant subscale-level signed DRF; DIF-unadjusted multi-group IRT models were those where item parameters for studied items were constrained to equality; DIF-adjusted multi-group IRT models were those where item parameters for studied items were freely estimated. Prevalence of EE, DP, and low PA was obtained by finding the theta score corresponding with the commonly used respective cut offs of ≥ 27, ≥ 10, and ≤ 33 on each scale based on EAP sum scoring produced from the DIF-unadjusted and DIF-adjusted multi-group IRT models
Back to article page