Data Analysis

Paper analyzed

Abou-Khalil R, Andary J, El-Hayek E. Apple cider vinegar for weight management in Lebanese adolescents and young adults with overweight and obesity: a randomised, double-blind, placebo-controlled study. BMJ Nutrition, Prevention & Health 2024;e000823. doi: 10.1136/bmjnph-2023-000823

https://nutrition.bmj.com/content/early/2024/01/18/bmjnph-2023-000823.info

Manual copy of the data table 2

pdata <- tribble(
  ~group, ~week, ~weightm, ~weightsd,
  "1) 5 ml", 0, 79,   27.1,
  "2) 10 ml", 0, 79.3, 28.5,
  "3) 15 ml", 0, 77.2, 27.2,
  "placebo", 0, 79.1, 28.6,
  
  "1) 5 ml", 4, 78.1, 26.5,
  "2) 10 ml", 4, 78.1, 27.4,
  "3) 15 ml", 4, 74,   26.6,
  "placebo", 4, 79.3, 28.4,
  
  "1) 5 ml", 8, 77.3, 26.0,
  "2) 10 ml", 8, 77.2, 27.4,
  "3) 15 ml", 8, 72.6, 26.4,
  "placebo", 8, 78.4, 28.7,
  
  "1) 5 ml", 12, 73.9, 23.7,
  "2) 10 ml", 12, 71.9, 26.1,
  "3) 15 ml", 12, 70.2, 25.0,
  "placebo", 12, 78.8, 28.6,
  
) 

Simple parameter estimation

# sample sizes given in paper and 95% critical value
sample_size_per_group <- 30
t_value <- qt(0.95, df = sample_size_per_group-1)

Comparison of between group differences using CIs

pdata |> 
  # calculate 95% ci for each measurement
  mutate(weight_ci = weightsd/sqrt(sample_size_per_group) * t_value) |> 
  
  # generate plot
  ggplot() +
  aes(x = week, y = weightm, color = group, group = group) +
  geom_point(position=position_dodge(1.5)) +
  geom_errorbar(aes(ymin = weightm - weight_ci, ymax = weightm + weight_ci), 
                width = 1,
                position=position_dodge(1.5)) +
  geom_line(position=position_dodge(1.5)) +
  labs(title = "Between-group comparisons using theoretical 95% confidence intervall",
       subtitle = "Assuming normal distribution of weight in each group",
       x = "week", y = "weight (kg)",
       caption = "Data copied from Table 2 of the paper.",
       color = "Experimental condition") +
  NULL

Comments

No within-group comparison as no individual data was reported.