Table 4.
Latent class conditional logit model estimates and membership predictors (three class)
| Attributes and levels | Latent class conditional logit | |||||
|---|---|---|---|---|---|---|
| Class 1 | Class 2 | Class 3 | ||||
| Coeff | SE | Coeff | SE | Coeff | SE | |
| Contribution to antibiotic resistance | ||||||
| Decrease by a quarter | 2.895 | [0.207]** | 0.085 | [0.154] | 0.161 | [0.033]** |
| Stay at the current level (Ref.) | 1.302 | [0.120]** | 0.318 | [0.110]** | 0.031 | [0.029] |
| Double | −1.123 | [0.118]** | −0.243 | [0.107]* | −0.021 | [0.028] |
| Triple | −3.074 | [0.217]** | −0.160 | [0.131] | −0.170 | [0.033]** |
| Treatment duration | ||||||
| 3 Days | 0.130 | [0.077] | 0.038 | [0.112] | 0.120 | [0.022]** |
| 7 Days | −0.094 | [0.077] | 0.301 | [0.082]** | 0.025 | [0.022] |
| 14 Days (Ref.) | −0.035 | [0.082] | −0.338 | [0.125]** | −0.144 | [0.022]** |
| Side effects | ||||||
| 1% | 0.217 | [0.061]** | 0.418 | [0.106]** | 0.275 | [0.023]** |
| 10% | 0.212 | [0.077]** | −0.127 | [0.091] | −0.004 | [0.021] |
| 25% (Ref.) | −0.429 | [0.079]** | −0.291 | [0.085]** | −0.271 | [0.023]** |
| Days needed to recover | ||||||
| 3–7 Days | 0.241 | [0.068]** | 0.314 | [0.075]** | 0.200 | [0.021]** |
| 8–14 Days | 0.017 | [0.074] | −0.052 | [0.098] | 0.022 | [0.022] |
| 15–28 Days (Ref.) | −0.258 | [0.075]** | −0.262 | [0.098]** | −0.222 | [0.022]** |
| Days before you start your medication | ||||||
| Straight after seeing a GP | 0.155 | [0.051]** | 0.189 | [0.054]** | 0.048 | [0.013]** |
| Wait 3 days (Ref.) | −0.155 | [0.051]** | −0.189 | [0.054]** | −0.048 | [0.013]** |
| Treatment failure | ||||||
| 5% | 0.030 | [0.083] | 0.131 | [0.095] | 0.144 | [0.022]** |
| 10% | 0.097 | [0.076] | −0.005 | [0.106] | 0.030 | [0.022] |
| 25% (Ref.) | −0.128 | [0.062]** | −0.126 | [0.075] | −0.175 | [0.022]** |
| Out-of-pocket costs | ||||||
| $10 | 0.239 | [0.100]* | 2.401 | [0.197]** | 0.156 | [0.027]** |
| $30 | 0.192 | [0.070]** | 0.321 | [0.072]** | 0.091 | [0.021]** |
| $70 (Ref.) | −0.431 | [0.088]** | −2.723 | [0.225]** | −0.246 | [0.029]** |
| Latent class share (%) | 0.245 | 0.252 | 0.504 | |||
| Latent class membership predictors (Ref. Class 3) | ||||||
| Young adults (aged 18–29 years) | 0.070 | [0.274] | −0.247 | [0.267] | ||
| Old adults (aged 60–69 years) | 0.332 | [0.164]* | 0.013 | [0.169] | ||
| Self-reported health: Fair/poor | 0.397 | [0.175]* | 0.595 | [0.162]** | ||
| Highly educated (Bachelor's degree or above) | 0.351 | [0.147]* | 0.001 | [0.151] | ||
| No. of prescribed antibiotics in the last 2 years: None | 0.426 | [0.149]** | 0.130 | [0.153] | ||
| Attitude towards receiving antibiotics for a cold or flu: Definitely want | −1.526 | [0.510]** | −0.719 | [0.304]* | ||
| Attitude towards receiving antibiotics for a cold or flu: Definitely do not want | 0.130 | [0.155] | 0.113 | [0.161] | ||
| Whether heard of the need to reduce unnecessary antibiotics in healthcare: Yes | 0.544 | [0.177]** | −0.277 | [0.152] | ||
| Observations | 29,844 | |||||
| N | 1658 | |||||
| LL | −7757 | |||||
| BIC | 16133 | |||||
Notes: **p < 0.01, *p < 0.05. Robust standard errors (SEs) are reported in the table. Effect coding was used for all attributes. LL, log likelihood; BIC, Bayesian information criterion