Skip to main content

Table 2 Odds ratios from logistic regression models predicting survey response in univariate (Unadjusted) and multivariate models (Adjusted)a

From: Evidence of non-response bias in the Press-Ganey patient satisfaction survey

Variable

Comparison

Unadjusted odds ratiob

Adjusted odds ratioa

Adjusted P value

Sex

Female (Reference)

--

--

--

Male

0.757 (0.697–0.823)c

0.782 (0.718–0.852)

<.001

Age

18–29 (Reference)

--

--

--

30–49

1.693 (1.457–1.967)c

1.708 (1.468–1.988)

<.001

50–64

2.697 (2.334–3.117)c

2.715 (2.344–3.146)

<.001

>= 65

2.247 (1.930–2.617)c

3.396 (2.815–4.096)

<.001

Insurance

Private (Reference)

   
 

Medicaid/Self pay/other

0.291 (0.238–0.354)c

0.311 (0.255–0.379)

<.001

Medicare

0.775 (0.700–0.857)c

0.478 (0.413–0.553)

<.001

Workers comp

0.578 (0.466–0.718)c

0.588 (0.472–0.733)

<.001

Subspecialty

Adult reconstruction (Reference)

--

 

--

Foot and ankle

0.970 (0.814–1.155)

1.048 (0.876–1.253)

0.6086

Sports

0.836 (0.706–0.989)c

0.981 (0.824–1.168)

0.8282

Hand

0.707 (0.593–0.844)c

0.795 (0.663–0.953)

0.013

Spine

0.686 (0.554–0.851)c

0.755 (0.607–0.940)

0.012

Non-operative

0.682 (0.585–0.796)c

0.757 (0.645–0.888)

<.001

Trauma

0.504 (0.400–0.635)c

0.623 (0.492–0.789)

<.001

Distance

> = 50 miles (Reference)

   

<50 miles vs

1.106 (0.988–1.239)

1.103 (0.980–1.241)

0.104

  1. aDue to missing data on at least one of the predictor variables, 251 subjects were excluded, leaving 2737 in the responder group and 13791 in the non-responder group
  2. bThese values are given as the odds ratio (OR), with the 95 % confidence-interval (CI) in parentheses, predicting survey response in univariate (unadjusted) and multivariate (adjusted) models, adjusting for gender, age, insurance type, sub-specialty and distance from the clinic
  3. cSignificant at the 0.05 level in the unadjusted analysis