Prenatal and Postnatal Hair Steroid Levels Predict Post-Partum Depression 12 Weeks after Delivery

Reviewer 2

Although the authors have revised the manuscript and provided

additional details for the methods section, several issues remain; the

writing is still unclear in many parts, it is difficult to interpret many

of the primary findings, and the revisions brought to light additional

concerns.

We thank the Reviewer 2 for their scrutiny and valuable comments. We

believe that her/his comments helped us again to make significant

improvement to the quality of the manuscript.

Specifically:

The introduction is still not cohesive; the authors introduce the

purpose of their study regarding hair cortisol sample, but then the

next headings are about cortisol levels through blood, saliva and

urine without any introduction to the methods used for cortisol

sampling. Throughout each section (particularly the cortisol and other

steroid sections), there is no real summary of results; just a listing of

what prior studies had found (many of which are contradictory). It

does not flow well or succinctly summarize findings for the reader.

Per request, we introduced a short opening paragraph:

In the following sections, we provide a brief overview of studies

on steroids during and after pregnancy. Such steroids, and above all,

cortisol, were mainly assessed in blood, saliva and urine, and to a much

lesser extent in hair.

Subheading: Plasma cortisol in healthy pregnant women

Summary: Thus, increases in plasma cortisol from the second to

the third trimester reflect physiological adaptations of

pregnancy among healthy women.

Subheading: Salivary, blood and urine cortisol in pregnant women with

major depressive disorders

Summary: Collectively, the pattern of results from cortisol

concentrations in saliva, blood and urine in pregnant women with PPD

does not seem to be fully consistent. Nevertheless, a hypercortisolemia

was mainly associated with transient symptoms of depression, while a

hypocortisolemia was associated with chronic symptoms of depression.

Subheading: Hair cortisol in women with post-partum depression

Summary: To summarize, studies on hair cortisol during and after

pregnancy are scarce and the pattern of results appears inconsistent.

Such inconsistent findings appear to reflect above all methodological

issues such as the timing of psychological assessments, and the lack of

a thorough diagnosis of PPD.

Subheading: Other steroids during pregnancy and after delivery

Summary: To conclude, research on further steroids such a

cortisone, progesterone, testosterone or DHEA is scarce, and results are

inconsistent and based on blood and saliva samples, while data from

hair samples are missing so far.

I disagree with the justification and use of stepwise multiple

regression. Please refer to the citations below regarding the numerous

limitations and shortcomings of using this technique. Your models

should be based on theory; not guided by statistics themselves.

Further, the limitations of using this analysis have not been

acknowledged. Smith, G. (2018). Step away from stepwise. Journal

of Big Data, 5(1), 32. Mundry, R., & Nunn, C. L. (2008). Stepwise

model fitting and statistical inference: turning noise into signal

pollution. The American Naturalist, 173(1), 119-123. Whittingham,

M. J., Stephens, P. A., Bradbury, R. B., & Freckleton, R. P. (2006).

Why do we still use stepwise modelling in ecology and

behaviour? Journal of animal ecology, 75(5), 1182-1189.

We thank Reviewer 2 for drawing out attention to the statistical issues

related to stepwise multiple regression analyses. We admit that we have

been unaware of the literature raising concerns as regards the method of

a stepwise multiple regression analysis. We have learned from Harrell

(2001) that stepwise multiple regression analyses bear the risk that

standard errors are biased toward 0, that p-values are biased toward 0,

and that parameters estimates are biased away from 0.

We have performed a ‘simple’ multiple regression analysis. We have

modified and adapted Table 4: we have reported the full result model,

including also those variables, which did not reach statistical

significance.

The overall pattern of results did not change.

The present results are limited to the final trimester; given that the

introduction focuses on timing of assessment (between first and

second trimesters, etc.), the implications of steroid assessment only in

the final trimester of this study should also be discussed

We thank Reviewer 2 for this candid comment. In the Limitation section,

the text reads as follows:

Sixth, we assessed hair steroids of the last trimester and the

first 12 weeks after delivery; accordingly, the present results

do not reflect steroid concentrations of the all trimesters, and

the results do not tell us more about hair steroid

concentrations since the very beginning of pregnancy.

However, this would have been particularly interesting: as

reported in the present data, and as mentioned in xxx, women

with post-partum depression have a higher risk to report

symptoms of depression already before pregnancy.

Accordingly, assessing hair steroids throughout pregnancy or

even before pregnancy would have allowed to get a more

comprehensive understanding of the underlying

neuroendocrinological processes preceding a PPD.

Please cite the cutoff score of 12 on the EPDS. The validated score

appears to be either 10 or 13. Matthey, S., Henshaw, C., Elliott, S., &

Barnett, B. (2006). Variability in use of cut-off scores and formats on

the Edinburgh Postnatal Depression Scale–implications for clinical

and research practice. Archives of women's mental health, 9(6), 309-

315.

We mentioned that the cut-off score of 12 points was used. The text

reads now:

However, we employed the more restrictive cut-off score of 12

for the self-rating of post-partum depression; others

recommend to use cut-off score of 13 {Matthey, 2006 #4358}.

Expand the limitations section regarding the sampling technique

(e.g., the group without PPD does not represent the full sample);

further, the sample size is relatively small, and these results

(particularly the exploratory ones) require replication with larger

sample

In the Limitation section, the text reads as follows:

Seventh, we assessed a smaller sample of women with and

without PPD; accordingly, the present results, particularly the

exploratory ones, require replication with larger samples.

The inclusion criteria for PPD require that subjects report 12 or

higher on the EPDS; and the inclusion criteria for no PPD require an

EPDS score of 5 or lower. Similar gaps in criteria are present for the

BDI; yet no one seemed to fall into the gap categories given that 65

women were categorized as PPD and 389 were categorized without

PPD. It seems strange that no one would have categorized by EPDS

of 5-12. Is that correct? Please provide a justification for and

additional information about the random selection of 50 women

without PPD- why was the full sample not used?

Thank you!

We specified this point, and the text reads now:

Additionally, of the 495 women approached, 41 (8.3%) did

neither fulfill the criteria for participants with or without PPD,

and 389…”

The following information were added:

Randomization occurred based on the following criteria: matching

the group of participants with PPD as regards sample size, age,

gestational age, and Apgar-score 10’ after delivery. A total of 167

participants fulfilled these criteria; their codes were put in sealed

and separate envelopes; sealed envelopes were put in an opaque

ballot box and stirred. A staff member of the hospital not otherwise

involved in the study picked out 50 envelopes.

To avoid possible and further confounders, we tested the sample of 48

participants with PPD against 50 participants without PPD.

Was there a program that was used to assist with the random

selection? The large sample size estimate stems from an article that

used EPDS scores within 96 hours of delivery and again at an

outpatient appointment. This effect size does not apply to hair steroid

levels. Please justify.

Thank you for drawing our attention to this point. To avoid that the

sample was statistically underpowered, we have chosen a study with

higher effect sizes. Further, neighter Caparros-Gonzalez et al, nor Braig

et al, nor Orta et al., which assessed hair steroids assessed women with a

firmly performed diagnose of post-partum depression.

Minor revisions include:

As currently written, the abstract does not sufficiently explain the

procedure; this was a cross-sectional study conducted 12-weeks after

delivery. At this time, subjects gave hair sample (reflecting 12 weeks

pre-and post-delivery) and completed surveys.

We have adapted the abstract, which reads as follows:

delivery and with or without PPD. Method. The present study

was a cross-sectional study conducted twelve weeks after

delivery. At that time, 48 women (mean age: 25.9 years) with

PPD and 50 healthy controls (mean age: 25.2 years) completed

questionnaires on depressive symptoms. Further, at the same

time point, 6cm lengths of hair strands were taken, providing

samples of hair steroids 12 weeks before and 12 weeks after

delivery in order to analyze hair steroids (cortisol, cortisone,

progesterone, testosterone, DHEA).

Sleep is still referred to in section 2.8 & table 3

Thank you! We have carefully checked the text once again, and we have

deleted all information as regards sleep.

Some information from section 2.6 should be moved to the

Introduction section (e.g., the justification for using hair sampling).

As per request, we have moved the first paragraph of section 2.6. Hair

strands sampling to the Introduction section. This was an excellent

suggestion of Reviewer 2, as it further helped us to justify the

methodological approach used in the present study.