Table 3. Odds ratios (and 95%CI) of SAQ nonresponse based on the multilevel analysis model, Belgian Health Interview Survey, 2013 (rescaled estimates).
| * | Model 0 (Empty model) | Model 1 | Model 2 | Model 3a |
|---|---|---|---|---|
| OR (95% IC) | OR (95% IC) | OR (95% IC) | ||
| Fixed effects at individual-level | ||||
| Gender | ||||
| Male | 1.11 (0.96–1.28) | 1.11 (0.96–1.28) | 1.10 (0.96–1.27) | |
| Female | 1.00 | 1.00 | 1.00 | |
| Age group (years) | ||||
| 15–34 | 1.08 (0.90–1.30) | 1.06 (0.88–1.28) | 1.06(0.88–1.28) | |
| 35–54 | 1.00 | 1.00 | 1.00 | |
| ≥ 55 | 0.77 (0.65–0.93)* | 0.77 (0.64–0.92)* | 0.78 (0.65–0.93)* | |
| Nationality | ||||
| Belgian | 1.00 | 1.00 | 1.00 | |
| European | 1.39 (1.10–1.77)* | 1.40 (1.11–1.78)* | 1.40 (1.11–1.77)* | |
| Non-European | 1.58 (1.18–2.13)* | 1.60 (1.19–2.15)* | 1.59 (1.19–2.12)* | |
| Education level | ||||
| Primary/No degree | 1.95 (1.50–2.54)* | 1.91 (1.47–2.48)* | 1.89 (1.46–2.44)* | |
| Secondary inferior | 1.10 (0.86–1.40) | 1.06 (0.83–1.35) | 1.05 (0.83–1.34) | |
| Secondary superior | 1.17 (0.97–1.41) | 1.15 (0.95–1.37) | 1.14 (0.95–1.37) | |
| Superior education | 1.00 | 1.00 | 1.00 | |
| Region | ||||
| Flanders | 1.00 | 1.00 | 1.00 | |
| Brussels | 2.38 (1.45–3.93)* | 2.27 (1.37–3.78)* | 2.10 (1.29–3.42)* | |
| Wallonia | 1.75 (1.05–2.91)* | 1.72 (1.02–2.89)* | 1.73 (1.06–2.83)* | |
| Chronic conditions | ||||
| Yes | 1.17 (0.99–1.38) | 1.20 (1.02–1.42)* | 1.20 (1.02–1.41)* | |
| No | 1.00 | 1.00 | 1.00 | |
| Household income | ||||
| Quintile 1 | 2.06 (1.61–2.64)* | 2.05 (1.60–2.63)* | 2.02 (1.59–2.58)* | |
| Quintile 2 | 1.76 (1.36–2.27)* | 1.78 (1.38–2.30)* | 1.76 (1.36–2.26)* | |
| Quintile 3 | 1.09 (0.84–1.41) | 1.10 (0.85–1.43) | 1.10 (0.85–1.42) | |
| Quintile 4 | 1.00 | 1.00 | 1.00 | |
| Quintile 5 | 0.92 (0.70–1.21) | 0.91 (0.69–1.19) | 0.91 (0.70–1.19) | |
| Fixed effects at interview-level | ||||
| Quarter of interview | ||||
| Quarter 1 | 1.00 | 1.00 | ||
| Quarter 2 | 1.66 (1.26–2.20)* | 1.65 (1.26–2.17)* | ||
| Quarter 3 | 2.06 (1.56–2.71)* | 2.06 (1.55–2.66)* | ||
| Quarter 4 | 2.43 (1.87–3.15)* | 2.39 (1.85–3.08)* | ||
| Length of face-to-face interview | ||||
| Short (facetime ≤ 13’) | 1.19 (0.97–1.46) | 1.19 (0.97–1.45) | ||
| Average (13’ < facetime ≤ 37’) | 1.00 | 1.00 | ||
| Long (facetime > 37’) | 1.00 (0.83–1.20) | 1.00 (0.83–1.19) | ||
| Fixed effects at interviewer level b | ||||
| Number of interviews performed by interviewer | ||||
| Low (count < 47) | 3.30 (1.60–6.79)* | |||
| Average (47 ≤ count < 60) | 1.00 | |||
| High (60 ≤ count < 98) | 3.08 (1.37–6.92)* | |||
| Very high (≥ 98) | (1.43–10.22)* |
Observations: 7,089 respondents, 183 interviewers, ICC = intraclass coefficient of correlation; PCV = proportional change in variance. The PCV expresses the change in the interviewer’s level variance between the initial model and the model with more terms.
a Best fitting model;
b Only variables that were significantly related to nonresponse after adjusting for all other variables.
*p value <0.05