Critically analyzing a research paper is an essential skill for lactation professionals, enabling you to discern the quality and applicability of the findings to your practice. This process involves evaluating the study’s design, methodology, data analysis, and interpretation of results to ensure the research is robust and reliable. By understanding how to scrutinize each section of a research paper, from the introduction to the conclusion, you can identify potential biases, limitations, and the overall validity of the study. This critical approach not only enhances your ability to apply evidence-based practices but also ensures that you are providing the most accurate and effective support to breastfeeding families.

In this blog post, I will guide you through the essential steps to effectively read and understand research papers, tailored specifically for lactation professionals. You’ll learn how to approach each section of a research paper—from the abstract to the conclusion—and gain insights into interpreting study results, understanding study methods, and recognizing the importance of sample selection and potential biases. Additionally, we’ll cover key biostatistics terms to enhance your ability to critically evaluate research findings. By the end of this post, you’ll be equipped with the skills to confidently apply evidence-based knowledge to your practice, ultimately improving the support you provide to breastfeeding families. 

How to Read the Paper

Research papers are not meant to be read start to finish. Reading a research paper effectively involves a strategic approach to grasp its core content and assess its relevance and reliability. Start with the Abstract, read the first paragraph and last sentence of the Introduction, examine the Methods, evaluate the Conclusion, and finally the limitations. For a step-by-step guide to reading research papers, read my post: Unlocking Research: A Step-by-Step Guide for Lactation Professionals on Reading Research Papers.

Read the Introduction

The introduction of a research paper provides several essential elements that set the stage for the entire study. For lactation professionals, understanding the purpose and content of the introduction is crucial for grasping the context and significance of the research.

The first element of the introduction is background information. This provides context for the research by explaining the broader topic or issue being studied. The next element of the introduction is the statement of the problem, which identifies the specific problem or gap in knowledge that the research aims to address. Finally, the introduction outlines the main goals or questions the study aims to answer.

You can gain all of this by simply reading the first paragraph and the last sentence of the introduction. The next step in critically analyzing a research paper is to look at the Methods to determine study design and what population was studied.

Read the Methods

1-Study Design

Study design refers to the overall strategy and structure of a research study that determines how data will be collected, analyzed, and interpreted. For lactation professionals, understanding study design is crucial as it affects the validity and reliability of the research findings, and consequently, their applicability to clinical practice.

In a 2020 study by Coentro et al., milk transfer with nipple shields was assessed using breast pumps to measure milk transfer, not babies. This is a perfect example of why you need to understand the study design. You cannot directly compare the findings of this study to a baby nursing a the breast because we know that pumps work differently than a baby.

Here’s an overview of key study designs relevant to lactation research:

2-Population

When reading the methods section of a research paper, you need to pay close attention to details about the population being studied. You want to know the demographics of the population being studied, the inclusion and exclusion criteria, the sample size- how many people were studied, and how they were selected.

Sample Selection:

How participants are chosen for the study. A representative sample ensures that findings are generalizable to the broader population. It’s important to consider how participants were selected for a study when determining if you can apply the findings to the population you serve.

For example, let’s say a researcher wanted to understand the impact of return to work on breastfeeding so she surveyed a group of moms from a local breastfeeding support group that she can easily access. This is a convenience sample because the researcher only chose to survey this group because the participants were easily accessible. The findings of this study cannot be generalized to every mom in the US or really to any other area that is not identical to the population of the mothers surveyed.

In my study, I recruited mothers from social media and via lactation consultant networks. While my reach was nationwide, it is still important to consider the implications of recruiting in this way. My sample was limited to only moms that use social media or  were recruited by a lactation consultant. This means I don’t have representation of mothers who don’t use social media or don’t have access. This population is more diverse than the population from the first example, but how the mothers were recruited should be considered when considering applicability of the findings to your population.

Randomization:

In RCTs, randomization helps eliminate selection bias by giving each participant an equal chance of being assigned to any group. This doesn’t occur much with studies looking at breastmilk intake because it would not be ethical to say one group can have breastmilk but the other cannot. Randomization in lactation often presents in studies about support or education. For example, a mothers may be randomized to receive a prenatal breastfeeding class or no class to assess the impact of prenatal breastfeeding on breastfeeding outcomes. 

Blinding:

Single-blind (participants unaware of group assignment) or double-blind (both participants and researchers unaware) methods reduce bias in data collection and interpretation. An example of blinding in lactation studies is when mothers randomized to use lanolin or a placebo to evaluate effect on nipple healing. Single blind means only the researchers know which group the mother is in. Double blind means neither the mother nor the researcher know which group the mother is in. 

Control Groups:

Comparing an intervention group to a control group helps isolate the effect of the intervention. The control group in research is the group that receives no intervention. For example, in the example of the mothers receiving prenatal breastfeeding or not, the group receiving no prenatal breastfeeding education is the control group. The purpose of the control group is to compare the intervention to doing nothing. 

Read the Discussion

The discussion section of a research paper is where the authors interpret and contextualize their findings. It bridges the gap between the results and the broader implications of the study, providing insights that go beyond the data. Understanding the discussion section is crucial as it offers a comprehensive analysis of the study’s significance, limitations, and potential applications.

Read the Limitations

The limitations section of a research paper is crucial as it provides insight into the constraints and potential weaknesses of the study. Understanding these limitations is essential for interpreting the results accurately and assessing their applicability to clinical practice.

Consider Bias

Bias in research refers to systematic errors or influences that can affect the validity and reliability of study results, leading to conclusions that may not accurately reflect the true situation. For lactation professionals, understanding bias is crucial because it helps in critically evaluating the quality of research and its applicability to clinical practice. Bias can occur at various stages of research, including study design, data collection, analysis, and interpretation.

Selection Bias occurs when the participants selected for the study are not representative of the larger population. Measurement bias happens when there are errors in how data is collected. Recall bias occurs when participants do not accurately remember or report past behaviors or events, such as breastfeeding practices or dietary habits. Observer bias arises when researchers’ expectations or knowledge about the study influence their observations or interpretations. Publication bias refers to the tendency for studies with positive or significant results to be published more frequently than those with negative or non-significant findings. Attrition bias occurs when there is a loss of participants over time in a longitudinal study, which can affect the validity of the results if the dropout rate is related to the study outcome.

Another source of bias that can impact lactation studies is the funding of the study. For products, research funded by the manufacturer may reflect that the product performs better than it acutally does. No matter the product being studied, if the research is funded by the manufacturer, the reader should assess the findings through the lens of potential bias.

What About the Results? 

The results section provides the empirical data and findings that form the basis of the study’s conclusions and implications. It presents the raw data collected during the study, often through tables, figures, and statistical analyses. It details the statistical methods used to analyze the data and whether the results are statistically significant. Without a solid understanding of biostatistics, this section will likely be meaningless to the reader. However, if you decide to skim it, here are a few key biostatistic terms you should know: 

Biostatistics Basics

Inferential Statistics:

• P-Value: Indicates the probability that the observed results occurred by chance. A p-value less than 0.05 is typically considered statistically significant, suggesting that the results are unlikely to be due to chance.
• Confidence Interval (CI): A range of values that is likely to contain the true value of an unknown population parameter. For example, a 95% CI means we can be 95% confident that the interval contains the true mean.
• Effect Size: A measure of the strength of the relationship between two variables or the magnitude of an intervention effect. Common measures include Cohen’s d and Pearson’s r.
• Statistical Power: The probability that a study will detect an effect if there is one. Higher power reduces the risk of Type II errors (failing to detect an effect that is present).

Statistical Tests:

• T-Test: Compares the means of two groups to determine if they are statistically different from each other. It can be paired or unpaired.
• ANOVA (Analysis of Variance): Compares the means of three or more groups to see if at least one group mean is different from the others.
• Chi-Square Test: Assesses the association between categorical variables. It tests whether the observed frequencies in each category differ from the expected frequencies.
• Regression Analysis: Examines the relationship between a dependent variable and one (simple regression) or more (multiple regression) independent variables. It helps in predicting outcomes and identifying trends.

Common Terms and Concepts:

• Sample Size (n): The number of participants or observations in a study. Larger sample sizes generally provide more reliable results.
• Bias: Systematic error that can affect the validity of the study results. Types include selection bias, measurement bias, and publication bias.
• Randomization: The process of randomly assigning participants to different groups to reduce bias.
• Blinding: Keeping study participants, caregivers, and sometimes researchers unaware of group assignments to prevent bias.
• Confounding Variables: Extraneous variables that can influence the outcome of the study, leading to incorrect conclusions if not controlled.

Association

Association is an identifiable relationship between an exposure and a disease or outcome. Association implies that exposure might cause the measured outcome. A study might find an association between breastfeeding education programs and increased breastfeeding rates. This means there is a relationship, but the study might not specify whether the relationship is linear or measure its strength.

Correlation

Correlation is a precise measure of the strength and direction of a linear relationship. A study might calculate the correlation between the number of breastfeeding support sessions attended and the duration of breastfeeding. If the Pearson’s r is +0.8, this indicates a strong positive linear relationship, suggesting that more support sessions are associated with longer breastfeeding durations.

Causation

Causation implies that there is a true mechanism that leads from exposure to outcome. The cause must precede the effect in time. For instance, in lactation research, if a new breastfeeding technique is said to increase milk production, the implementation of the technique must come before the observed increase in milk production.

Wrap Up

Mastering the skill of reading and understanding research papers is invaluable for lactation professionals. By systematically approaching each section—starting with the abstract for an overview, examining the introduction for context, scrutinizing the methods for study design, analyzing the results for data interpretation, and critically evaluating the discussion—you can thoroughly assess the validity and applicability of research findings. Understanding sample selection, recognizing potential biases, and being familiar with key biostatistics terms further enhance your ability to interpret and apply research effectively. With these tools, you can confidently integrate evidence-based practices into your work, thereby providing the highest level of support to breastfeeding families. Stay engaged with the latest research, and continue to deepen your expertise for the benefit of the families you serve.

References

Dawson, P. (2002, May 1). Revised growth charts for children. American Family Physician. https://www.aafp.org/pubs/afp/issues/2002/0501/p1941.html