Is what I am listening to/reading evidence based? | Dr. Ellen Chetwynd

What does it actually mean to practice evidence-based lactation care – and how do you tell trustworthy research from noise? In this episode Dr. Ellen Chetwynd walks through how clinicians can evaluate research quality and the evidence behind a topic. She shares how she personally introduces new concepts to her practice with patients, how she reads a research article, and how she looks at the evidence that supports a concept.

Guest Bio Ellen Chetwynd, PhD, MPH, BSN, RN, IBCLC, Clinician-Researcher, Adjunct Assistant Professor at University of North Carolina. She received the United States Lactation Consulting Association Award for Excellence in 2019 and as of the release of this, November 2026, she is the immediate past Editor in Chief of the Journal of Human Lactation.

Listen to the full episode to learn along with me. Read my personal notes below to see what I found noteworthy and what I’ll revisit in the future.

My Personal Notes from this Episode

How Dr. Chetwynd Reads a Research Article (50:00 min)

First she reads the last 1-2 sentences of the Background Section to learn the foundation of the article 

  • The most impotent sentence: the very end of the background section.
  • It should tell you the gap in the literature and then what they’re going to do to fill that gap. This tells you the foundation of the article.
  • “Despite all of this information that we have, we still don’t know X. So I’ve designed a study in which I approach X by taking this population and doing Y”

Next she looks at the tables and the figures: all of the information you need to understand the table/figure should be in the actual table/figure.

  • If there is something about the table/figure she doesn’t understand then she goes to the methods/results. 
    • Methods will tell you what the researchers did to create that table. 
    • Results will lay out what happened when they did that thing.

Next she looks at the first sentence of the discussion section. This should say something like ‘we did this study, and we were able to support and extend the current literature on this topic area by our results which say …(here they state the primary result of the study)’ many articles miss this important sentence

  • Support = we didn’t start on a wild hair, we’re following previous research and taking a microstep from the previous study
  • Extend = not just repeating a previous study, but incrementally moving it forward a bit
  • Primary result = there may be multiple results but a good paper will tell you the primary – exactly the most important thing you’re suppose to get from that paper

She does not pay much attention to the discussion and even less attention to the conclusions

The meat of the study is in the methods and results which you can shortcut by taking a look at the tables.

How Dr. Chetwynd has been looking into the evidence that supports a common practice in our field: asymmetric latch (27:00 min)

  • She started with a textbook and looked at those references 
  • Then she started looking at different adjacent fields (for example how speech and language pathologists want young children to hold their head when eating)
  • She finds what works in her practice with her patients (taking into account her own expert opinion)
  • She spoke with the ‘mothers of our field’, the professionals who helped to shape the field of lactation.

She found that asymmetric latch is expert opinion based – not evidence based 

One primary technique in body led breastfeeding is to start by using an asymmetric latch then bringing baby into a neutral position at the breast. She has found this works in her own practice. See BodyLedBreastfeeding.com and Body Led Breastfeeding podcast for more info.

How Dr. Chetwynd personally introduced a new concept to her practice (a change in how she fits for flanges) (1:06:00 min)

She got the nipple measuring tool and told her clients “I’m just learning this, we’re going to try some stuff out, this is what I recommend, and I ask them for their feedback” 

she SLOWLY began to adapt this to her own practice. 

She asks for feedback from her patients.

She is very clear with her patients: I have some information, but not All the information. “I don’t have all the information and I try to be really clear about that” (in this example, practice is changing because of this one study, there is still lots of information that is unknown)

She talks to other LC’s to learn their thoughts and what they are finding if they are implementing these changes.

If considering applying research to your practice… Put it to your research group: chew it, talk about it, get geeky about it – have they tried it in their own clinical practice?

EVIDENCE BASED PRACTICE

Steps to take to determine if something is evidence based: (26:00 min)

  • Choose something predigested like ABM Protocols or In the Field articles from the Journal of Human Lactation.
  • References are really important to pay attention to. 
  • If you’re looking for research info: look to hear from a researcher or translational clinician. 
  • Don’t just listen to a single source – move around between various podcasts, conferences and presentations. 
  • Speaker should have experience in the field.

Evidence based medicine means taking that research, using our own clinical expertise and then interacting with your patient in a way that takes into account all of their cultural expectations and desires for their own heath care. It is not the research alone. (5:00 min)

It takes a long time, years and years, to amount evidence for a topic. Typically this includes many many studies that prove the same thing. If you were to base your practice on one single study your practice would swing wildly over and over again. These individual studies and wild swings will get more and more centered and more and more centered as our years of evidence in the field get longer. (13:00 min)

Science is not an individual sport. It takes a team to chew on research. (3:30 min) 

We should all have a group of people to geek out with. We need a team. (1:05:00 min)

A really good source of evidence: Academy of Breastfeeding Medicine protocols are experts in the field that look around at everything they can find, they write on it and chew on it and come up with guidelines.

  • to dive even deeper into a specifics aspect of an ABM protocol – go directly to the articles that are referenced and read them – she calls these the ‘formational or foundational articles’, the articles the authors are using to support what they are saying. The reference should support what is said in the sentence. (20:00 min)

Individualized care – like body work/massage/lymph drainage- is hard to protocolize and do studies on because it is so individualized (21:30 min)

History of evidence in the field of lactation (5:00 min)

  • Before evidence based practice the field of lactation relied heavily on expert opinion
  • Many things in our field are expert opinion, not necessarily evidence based. (We have not been doing research in our field for very long, started in the 90’s)
  • It is hard to get strong evidence for lactation because it is a manual technique – so we are individualizing it to the individual person – and also our primary outcome, length of breastfeeding, is confounded by so many different things (42:00 min)
    • We’re not going to have many RCTs that are supporting the evidence for what we do in a clinical study.
    • There are so many different variables that impact length of BF, it’s hard to narrow it down to the specific element we are studying

HOW TO READ RESEARCH: THE BASICS

Hierarchy of Evidence Pyramid: (35:00 min)

  • Draw a line underneath the case control studies
    • Everything above this line is designed to control for things that might influence the study. Leading to less bias in the result.
  • She would break systematic reviews and meta analysis apart and put meta analysis at the very very top.

Cannot RCT (randomize control trial) for breastfeeding vs formula feeding: unethical due to the beneficial well being of the dyad.

Epidemiologic Research

Since we cannot do many RCTs we need to take into account epidemiologic research: techniques needed if we cannot randomize a population. 

Often instead (of randomizing for BF vs formula) we’ll look at duration of BF (for example babies who got BF for 1-2 weeks vs babies who got BF for 3 months or longer) BUT the problem with this as your two arms is there are SO many reasons why people might end up formula feeding, its hard to pin down the reason to what you are studying –  which could lead to an incredibly biased result (if we were looking at this as an RCT)

Confounder: something that was different from one group to the other

  • Examples: insurance company, WIC participant, access to lactation support, institutional racism
  • Did they breastfeed longer because of what was being studied or because of a different factor?

We control to try to create similarity between the groups. We cannot randomize. But we can equalize the arms of the study using various statistical techniques so that they ACT as if they are equal. 

Systematic Review vs Meta Analysis (1:15:00 min)

Meta analysis : combines the participants from a bunch of different studies to act as one larger study that has more participants and greater power. The studies have to have enough similarities for this to be done. 

Systematic review: usually happens if the author wonders if a meta analysis can be performed, but turns out there is not enough similarities to perform the meta analysis. Cannot combine the actual participant of the study to give you a combined result. Instead you analyze each study in various ways – pull out different threads of information – and come to a conclusion based on the results of those studies. This is not quite as powerful as a meta analysis. 

When Reading the Research Article (56:00 min)

When looking for a good result – in general look for:

  • small p value with a moderate population
  • really, really small p value if it goes along with a huge population
  • confidence interval to be narrow and NOT to cross the 0 or 1 significance
  • Does it apply to your clinical practice? Is it relevant? Do you have access to it (the piece of equipment, etc)? 

If there is a significant P value then it is statistically significant (emphasis on statistically)

  • P value is usually .05 but not always – the researcher is who decides the P value

Always look at a P value and the population size at the same time. Never just the p value by itself.

  • A smaller P value goes with a big population
  • A larger P value goes with a small population
  • N: whole population | n: is a subsection of the population (like one arm of the population)

Confidence intervals or confidence limits:

  • a very wide confidence interval generally means that the study is less accurate.
  • A super narrow confidence interval means there is a higher accuracy.
  • Look at whether it crosses 1 or 0 – depending on your test statistic

KEEP READING the difficult to understand parts of the studies. Practice by making yourself read the sentences and see if you can figure out at least a part of what they’re talking about. As you make yourself do the work, you will learn more about how to interpret the studies.

Don’t just read the abstract and be done: sometimes abstracts are not a good representation of what they put In the paper. 

Applying research to clinical practice: (1:02:00 min)

Put it to your research group: chew it, talk about it, get geeky about it, have they tried it in their own clinical practice?

talk to other people

see what’s happening in the podcast world, what are people saying about the research article?

introduce it to your practice slowly and carefully

use your own common sense: you are a clinical expert in the field and what your patient says matters.

It is my job to keep myself up with the state of the science in my field of practice

Part of our job as an evidence-based clinician is to combine what we learn from evidence with what we know as an expert in our field and integrating ourselves with others inside and outside of our field and to sustain a positive relationship with our patients.

Best ways to spend my time learning about lactation: listening to podcasts and reading original research (don’t spend too much time reading blogs, unless you know and trust the author)

Resources

Good sources of ‘pre-chewed’ evidence based info (26:50 min)

Academy of Breastfeeding Medicine Protocols “experts in the field look around at everything they can find at the time that they write it, they chew on it and they come up with guidelines”

From the Field articles in the Journal of Human Lactation “somebody is giving your all of their framing on a topic area”

Podcasts

Journal of Human Lactation Podcast

References

Mitchell KB, Johnson HM, Rodriguez JM, et al. The Mastitis Spectrum : The Academy of Breastfeeding Medicine protocol #36 , revision 2022. Breastfeeding Med. 2022 May; 17(5):360-376.

  • ABM Clinical protocol The Mastitis Spectrum ( 14:00 min)

Anderson L, Kynoch K, Kildea S, et al. Effectiveness of breast massage for the treatment of women with breastfeeding problems: A systematic review. JBI Database System Rev Implement Rep 2019; 17:1668-1694.

  • Breast massage study referenced in the ABM Protocol (14:30 min)

Witt AM, Bolman M, Kredit S, et al. Therapeutic breast massage in lactation for the management of engorgement, plugged ducts, and mastitis. J Hum Lact 2016;32:123-131.

  • Therapeutic breast massage study referenced in the ABM Protocol (14:30 min)

Ezzo J, Manheimer E, McNeely ML, et al. Manual Lymphatic drainage for lymphedema following breast cancer treatment. Cochrane Database Syst Rev 2015;CD003475.

  • Lymphatic drainage study references in the ABM Protocol (14:30 min)

Theurich MA, Kar E. Applying the Evidence Pyramid to Breastfeeding and Lactation Research. Journal of Human Lactation. 2026;42(1):135-140. doi:10.1177/08903344251406587

  • Evidence Based Pyramid (35:00 min) specifically as applied to clinical lactation care

Nolet PS, Emary PC, Murray J, Harris GH, Gleberzon B, Chopra A, De Ciantis M, Overton R. Conceptualizing the evidence pyramid for use in clinical practice: a narrative literature review. J Can Chiropr Assoc. 2025 Nov;69(3):238-254. Epub 2025 Nov 30. PMID: 41425288; PMCID: PMC12716892.

  • Evidence Based Pyramid (35:00 min)

Anders LA, Mesite Frem J, McCoy TP. Flange Size Matters: A Comparative Pilot Study of the Flange FITSTM Guide Versus Traditional Sizing Methods. J Hum Lact. 2025 Feb;41(1):54-64. doi: 10.1177/08903344241296036. Epub 2024 Nov 30. PMID: 39614713; PMCID: PMC11829504.

  • Pilot study on flange sizing (1:01:45 min)

Nakamura M, Luo Y, Ebina Y. Systematic Review on the Efficacy of Moisturizing Therapy in Treating Nipple Trauma and Nipple Pain. Journal of Human Lactation. 2025;41(1):39-53. doi:10.1177/08903344241301401

  • Systematic Review on moisture levels for treatment of nipples (1:10:20 min)

Nakamura M, Asaka Y, Ogawara T, Yorozu Y. Nipple Skin Trauma in Breastfeeding Women During Postpartum Week One. Breastfeed Med. 2018 Sep;13(7):479-484. doi: 10.1089/bfm.2017.0217. Epub 2018 Aug 3. PMID: 30074830.

  • Another Maya Nakamura study (1:12:17 Min)
  • She took 776 pictures of nipple trauma


Nakamura M, Asaka Y. An Evaluation of the Signs of Nipple Trauma Associated With Breastfeeding: A Delphi Study. J Hum Lact. 2022 Aug;38(3):548-558. doi: 10.1177/08903344221076527. Epub 2022 Feb 10. PMID: 35139667.

  • Another Maya Nakamura Study (1:12:17 Min)
  • A Delphi study establishing definitions for different kinds of nipple lesions


Nakamura M, Sugimori H, Ebina Y. Development of Nipple Trauma Evaluation System With Deep Learning. J Hum Lact. 2025 Feb;41(1):105-114. doi: 10.1177/08903344241303867. Epub 2024 Dec 24. PMID: 39718190.

  • Another Maya Nakamura Study (1:12:17 Min)
  • Trained AI to recognize the severity of the lesions using deep learning and based on these definitions/images

Nakamura M, Sugimori H, Ebina Y. Feasibility and Acceptability of a Deep-Learning-Based Nipple Trauma Assessment System for Postpartum Breastfeeding Support. Healthcare (Basel). 2025 Aug 22;13(17):2091. doi: 10.3390/healthcare13172091. PMID: 40941442; PMCID: PMC12428512.

  • Another Maya Nakamura Study (1:12:17 Min)
  • Tested the clinical acceptability of the software

Where to learn more from Dr. Ellen Chetwynd

Hosted by Suzanne Sanders, IBCLC

This podcast is intended for healthcare professionals only and is for informational purposes. Content is not medical advice and should not replace professional judgment or individualized patient care. Exercise your own clinical judgment before applying any information discussed. Product or service mentions do not constitute endorsements.

Leave a Reply