Longitudinal analyses: LME vs Pairwise comparisons

Hi all,

I have two specific questions regarding my datasets to understand how to analyze them correctly.

  1. I am evaluating a dietary intervention in two groups (control vs specific diet). The first time point all the animals are under a control diet. In the following time points, they have a control vs a specific diet. However, there is a change in the diet at each time point (an increase in the dose of prebiotics). Is it correct to perform LME? Or maybe is better to use pairwise comparisons with time 0?

Group Time Diet
Group_1 0 C
Group_2 0 C
Group_1 1 C
Group_2 1 Dose1
Group_1 2 C
Group_2 2 Dose2

  1. Another dataset I have is different skin biopsies and we want to compare the skin layers. Since they are “dependent” among them within each individual, would it be appropriate to use a longitudinal approach? For example, an LME considering each skin layer a number (for example, how they are physically closer: swab, epidermis, dermis)? Or pairwise comparisons would be better?

I would really appreciate any thoughts on that. Many thanks for your attention and patience!

Best regards,


Hi @AnnaC,

You could use both. The regular change in diet certainly complicates matters, so on the one hand the pairwise comparisons would be a better approach for isolating the effects of specific treatments.

On the other hand, LME may be appropriate, depending on your treatments and hypothesis. Are the changing diets a gradual introduction of some ingredient and you expect a gradual increase in, say, alpha diversity, or is this a cross-over study and you expect the changing diets to reverse the effect? If the former (or in any case where you expect or see a linear change), LME would be useful, and you can “eye-ball” how well LME may fit your data by using the volatility action to look at the shape of the data. If the latter (cross-over), then your data probably aren’t linear, LME would make a poor fit, and paired comparisons would be a better choice.

Sure! LME is a really useful method for longitudinal experiments but it is not in itself an inherently longitudinal method. This is why the parameter name state is used throughout q2-longitudinal instead of calling it “time”… because many of these methods can be applied to both longitudinal as well as other experimental designs that involve dependent samples.

I suppose here’s one caveat: assigning numbers arbitrarily may make too many assumptions about the relationship between these sites… I’d feel more comfortable using a quantitative measure (e.g., distance from surface), but I suppose you could use volatility to see what the data look like after assigning those labels. Given that you only have 3 dependent samples, and especially because it sounds like there could be methodological differences in how they were collected (surface swab vs. skin biopsy?) pairwise comparisons may be more appropriate in this case.

Good luck!


Many thanks, again @Nicholas_Bokulich! Now I have a clearer idea.

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.