Skip to main content

Table 3 Concordance between self-reported and registered-based health care utilization (using the latter as reference), Belgian Health Interview Survey 2008 (population aged 15 years and over)

From: Regional differences in the validity of self-reported use of health care in Belgium: selection versus reporting bias

 

Accurate reporting

Overreporting a

Underreporting b

Agreement

PPV c

NPV d

 

%

(95 % CI)

%

(95 % CI)

%

(95 % CI)

kappa

(95 % CI)

%

(95 % CI)

%

(95 % CI)

Contact GP < 12 m

84.8

(83.7–85.9)

5.9

(5.3–6.7)

9.2

(8.4–10.2)

0.47

(0.45–0.50)

91.5

(90.8–92.2)

53.6

(51.1–56.1)

Contact specialist < 12 m

75.6

(74.3–76.9)

4.9

(4.3–5.5)

19.5

(18.3–20.8)

0.49

(0.47–0.51)

89.2

(88.2–90.1)

58.7

(57.2–60.3)

Contact dentist < 12 m

80.3

(79.0–81.5)

14.6

(13.5–15.8)

5.1

(4.5–5.8)

0.59

(0.57–0.60)

71.6

(70.2–72.9)

87.7

(86.7–88.7)

Contact dentist < 6 m

85.2

(84.0–86.3)

7.7

(6.9–8.6)

7.1

(6.3–8.0)

0.65

(0.63–0.67)

74.3

(72.5–76.0)

90.2

(89.4–91.0)

Contact physiotherapist < 12 m

88.3

(87.3–89.3)

5.6

(5.0–6.3)

6

(5.3–6.9)

0.60

(0.58–0.62)

67.9

(65.5–70.2)

92.2

(91.5–92.8)

Inpatient hospitalisation < 12 m

95.2

(94.5–95.7)

2.9

(2.5–3.4)

1.9

(1.6–2.3)

0.75

(0.72–0.77)

75.7

(73.1–78.1)

97.2

(96.8–97.5)

Day patient hospitalisation < 12 m

89.7

(88.8–90.5)

2.1

(1.7–2.5)

8.2

(7.5–9.1)

0.43

(0.40–0.46)

69.7

(66.1–73.2)

90.5

(89.8–91.1)

  1. < 12 m: in the 12 months prior to the date of the interview
  2. < 6 m: in the 6 months prior to the date of the interview
  3. a Overreporting: self-reported contact, but no register-based contact
  4. b Underreporting: no self-reported contact, but register-based contact
  5. c PPV: positive predictive value
  6. d NPV: negative predictive value