Skip to main content

Table 3 Multivariable lasso regression model of characteristics distinguishing between presence versus absence of HATMD

From: Clinical, psychological, and sensory characteristics associated with headache attributed to temporomandibular disorder in people with chronic myogenous temporomandibular disorder and primary headaches

Predictor

Penalized

coefficienta

Inferential

OR (95% CL)b

Monthly headache frequency in the last 3 months

0.311

1.49 (1.02, 2.17)

Familiar masticatory muscle pain responses on examination

0.242

2.98 (1.58, 5.62)

Systolic blood pressure

−0.141

n/a

HIT-6 score

0.130

n/a

CSQ-R coping self-statements

−0.119

n/a

Diastolic blood pressure

−0.097

n/a

Time since headache onset

−0.067

n/a

Familiar TMJ pain responses on examination

0.063

n/a

Maximum unassisted jaw opening

−0.059

n/a

Migraine (headache type)

0.057

n/a

Pain-free jaw opening

−0.012

n/a

CSQ-R ignoring pain

−0.004

n/a

  1. Abbreviations: CSQ-R Coping Strategies Questionnaire-Revised, HIT-6 Headache Impact Test-6, TMD Temporomandibular disorder, TMJ Temporomandibular joint
  2. a Penalized coefficients are non-zero values from the Stata “lasso logit” command that used 42 potential predictors variables. (Variables not listed have coefficients of zero.)
  3. b Odds ratios are from the Stata “dslogit” command that specified two a priori predictor variables (monthly headache frequency in the last 3 months and familiar muscle pain responses on examination) with the remaining 40 potential predictors as control variables. The command uses a double-selection process for inference yielding odds ratios (OR) and 95% confidence limits (CLs) for the two a priori variables