Sample Demographic Characteristics

Characteristic Overall Public
N = 91		 Supplementary
N = 89		 p-value
Age (years) 39 (16) 45 (15) 32 (15) <0.001
Sex


0.8
    Male 67 (37%) 33 (36%) 34 (38%)
    Female 113 (63%) 58 (64%) 55 (62%)
Previous anesthesia 167 (93%) 85 (93%) 82 (92%) 0.7
Previous PONV 41 (23%) 20 (22%) 21 (24%) 0.8
Previous postoperative pain 113 (63%) 55 (60%) 58 (65%) 0.5
Previous postoperative hypothermia 35 (19%) 15 (16%) 20 (22%) 0.3
Previous intraoperative awareness 9 (5.0%) 5 (5.5%) 4 (4.5%) >0.9

Willingness to Pay by Healthcare Coverage (Table 2)

Characteristic Public
N = 911		 Supplementary
N = 891		 p-value2
WTP - PONV (50% efficacy) 32 (1, 128) 64 (32, 256) 0.001
WTP - PONV (100% efficacy) 128 (32, 256) 256 (128, 1,024) <0.001
WTP - Awareness (50% efficacy) 32 (1, 256) 256 (64, 1,024) <0.001
WTP - Awareness (100% efficacy) 256 (64, 512) 2,048 (512, 4,096) <0.001
WTP - Pain (50% efficacy) 32 (1, 128) 256 (64, 512) <0.001
WTP - Pain (100% efficacy) 256 (32, 512) 512 (256, 2,048) <0.001
WTP - Hypothermia (50% efficacy) 16 (1, 128) 128 (32, 512) <0.001
WTP - Hypothermia (100% efficacy) 128 (32, 512) 512 (128, 1,024) <0.001
1 Median (Q1, Q3)
2 Wilcoxon rank sum test

Comparative WTP Values by Efficacy Level

Comparative WTP Values by Efficacy Level (Figure 2)

Consolidated Impact of Previous Complication History on WTP (Radial Chart)

Esse gráfico radial consolida o impacto do histórico prévio de cada complicação no WTP, mostrando claramente como ter experimentado previamente a complicação influencia o valor mediano disposto a pagar.

Preparação dos dados em formato longo para todas as complicações

bd_long_all <- bd %>%
  pivot_longer(
    cols = c(ponv50, ponv0, recall50, recall0, pain50, pain0, hypothermia50, hypothermia0),
    names_to = "Complication_Efficacy",
    values_to = "WTP"
  ) %>%
  mutate(
    Efficacy = ifelse(grepl("50$", Complication_Efficacy), "50%", "100%"),
    Complication = case_when(
      grepl("ponv", Complication_Efficacy) ~ "PONV",
      grepl("recall", Complication_Efficacy) ~ "Awareness",
      grepl("pain", Complication_Efficacy) ~ "Pain",
      grepl("hypothermia", Complication_Efficacy) ~ "Hypothermia"
    )
  )

Modelo de regressão linear misto

model <- lmer(WTP ~ contas + Efficacy + sexo + idade + (1|id), data = bd_long_all)

# Tabela consolidada dos resultados
model_summary <- tbl_regression(model, 
                                exponentiate = FALSE,
                                label = list(
                                  contas ~ "Healthcare Coverage (Supplementary vs Public)",
                                  Efficacy ~ "Efficacy (100% vs 50%)",
                                  sexo ~ "Sex (Female vs Male)",
                                  idade ~ "Age (years)"
                                )) %>% 
  add_global_p() %>%
  modify_header(label = "**Predictors**")

model_summary
Predictors Beta 95% CI1 p-value
Healthcare Coverage (Supplementary vs Public)

0.004
    Public — —
    Supplementary 247 79, 416
Efficacy (100% vs 50%)

<0.001
    100% — —
    50% -508 -581, -434
Sex (Female vs Male)

0.2
    Male — —
    Female -81 -202, 40
Age (years) -4.6 -8.5, -0.71 0.020
1 CI = Confidence Interval