Too often, we hear words being thrown around like “studies show” and “research supports” without fully understanding what they mean.
You’re a chiropractor, not a statistician.
It would be ludicrous to assume that every chiropractor knows how-to, let alone want to, break down research articles into their clinically significant parts.
So, here’s your chance to let us do the hard work for you and give you a crash-course on how to quickly and accurately read a chiropractic research paper while only taking away the good stuff.
Chiropractic Research: The Basics
I have to give a few disclaimers before we jump right into this article.
By no means am I an expert, I do not know everything about research, and this post is not comprehensive.
I’m just a chiropractor who thinks it’s essential to have the ability to read the study as a tool in your chiropractic tool belt.
Additionally, the following paragraph examples are for demonstrative purposes and are not real statistical values from any research.
First things first – Chiropractic Research Terminology
Before we jump into any research article, we need to understand any important terminology you may encounter—the first being prevalence and incidence.
Prevalence is the number of individuals in a population with a condition at a given point in time.
For example, a study of 1,000 individuals found that 725 of them had low back pain.
Incidence is the number of individuals in a population that develop a condition over a period of time.
For example, a study of 1,000 individuals found 125 new cases of low back pain in one year.
The clinical takeaway here is that the prevalence in chronic conditions (like low back pain) is usually higher than its incidence.
Additionally, when measuring incidence, the first occurrence is typically the only one counted, even though some individuals may have had more than one episode of low back pain throughout the year in our example.
Next, we’ll go straight to some statistics terms so that your eyes don’t gloss over when you get to the article’s results section.
The most straightforward to understand is descriptive statistics.
These include data points like the average (or mean), standard deviation, and percentages used primarily compared to a baseline measurement.
For example, when you read that one in four individuals over the age of 50 involved in a study reported having low back pain.
On the other hand, we have analytical (or inferential) statistics.
This is where the data become a bit more complicated but is also more critical.
It allows researchers to interpret and draw conclusions about target populations based on analyses like confidence intervals and p-values.
Descriptive statistics and analytical statistics are equally important because descriptive data is necessary to complete analytical data.
Types of Evidence + The Chiropractic Research Pyramid
Now that we’ve scratched the surface of the basic terminology you may read in scientific literature, let’s talk about the different types of studies you may encounter.
We’re going to tackle these in the order of the evidence pyramid, from the bottom up.
I like to keep in mind Dr. Souza’s description: “as you move up the pyramid, the amount of available literature decreases, but increases in relevance to the clinical setting.”
At the bottom of the pyramid are things like ideas, editorials, and opinion writing pieces.
To be quite frank, similar to what you’d read in a blog post.
These may be well-referenced and supported by evidence but have not gone through any peer review processes and are not actual research studies where formal conclusions were drawn.
Evidence you find at the bottom of the pyramid is generic and typically isn’t recommended to inform any serious clinical decision-making.
The next tier up in the pyramid includes case reports and case series.
These are usually studies of individual treatment or collections of studies on the treatment of a particular cohort.
These provide more value than the level below as they can inform a clinician about how others may have managed a particular condition in the past.
However, case reports and case series lack a control group, making it difficult to draw further conclusions about clinical treatment methods.
Next, we have the case-control study.
Typically retrospective, these studies most often compare individuals who already have a specific condition with individuals who do not.
Already we can see that any information gathered from this study design is valuable, as there is now a control group.
The largest con of the case-control study is the likelihood of confounding variables, making any causal relationship challenging to determine.
For example, saying that neck pain is related to drinking coffee because individuals who work at a computer more commonly drink coffee.
We’ve only got a few study designs left, so stay with me because these are the most important! The next two, cohort studies and randomized control trials (also called RCT’s), go hand in hand.
A cohort study lacks randomization of the participants and can be either interventional or observational.
Interventional designs follow participants who currently have a specific condition and receive treatment and compare it to another group not affected by the condition or not receiving the treatment.
Observational designs follow participants who don’t have a condition to determine the risk of them obtaining it.
You can think of a randomized control trial as a better version of a cohort study.
The concept of following participants with a condition and then utilizing a comparison group is essentially the same.
What makes it better is in the name—participants are randomly allocated to these groups, which reduces the risk of bias.
Lowering the bias lowers the opportunity for confounding variables throughout the study, making any significant results that much more attributable to the treatment being tested.
While the results from a randomized control trial are better than others, there are still some pitfalls.
Methods used in randomized control trials are often difficult to apply directly to practice, and the participants don’t typically represent an average patient population.
Lastly, we’ll talk about the two at the top of the pyramid which contribute the most useful information: systematic reviews and meta-analyses.
In general, these are viewed as the most comprehensive.
A meta-analysis combines results from various similar studies and then collectively analyzes all of the results as if they were just one study.
These are great because by just reading one research article, you can better understand many different studies’ interrelation.
On the other hand, it’s rare to glean useful conclusions from the results of these analyses because the included and excluded studies are at the authors’ discretion.
Systematic reviews gather high-quality published research, evaluate the findings, and then summarize the identified evidence.
The most crucial detail about the systematic review is that the appraisal and summarization are reproducible.
The more reproducible a process in research is, the more confidence it gives readers to interpret that process.
Piecing Together a Chiropractic Research Study
Becoming more familiar with study designs is a huge piece of the puzzle, so the next natural step is to polish up the different pieces of a scientific article.
First, the abstract. I like to call this the highlight reel of chiropractic research.
It’s the miniature version of a research article, including the study’s goals, methods, results, and conclusions. It’s an excellent place for an at-a-glance understanding.
To understand the publication, though, you should read further.
The next thing you’ll come across is the introduction. This explains what motivated the authors to complete the research.
It provides background to the study, reviews previous relevant work, and shows where current knowledge of specific topics is lacking.
It also should address the clinical importance of the topic at hand and should immediately give you—as a provider—a reason to keep reading.
This importance will often be expressed in epidemiologic or public health terms like morbidity, mortality, or any of the above mentioned terms.
To think critically about this section, analyze the evidence the author presents. Are sources cited to support or oppose the topic at hand?
Following the introduction is the methods section.
As mentioned before, methods are most valid and reliable when they are reproducible.
When reading a methods section, you should answer any question that you may have about the population demographics, the data from measurements taken, or the statistical methods used for analysis.
For example, you may want to know the age of the studied patients or how any measurements were standardized.
After the methods, you’ll find the result and discussion.
The results section is precisely what you think it would be. I typically look here for the hard facts and numerical data.
If you are a statistics wizard, this is where you would look for the breakdown of any statistical analysis that was performed.
Here you can also find any graphs, tables, or other figures the author may have used to illustrate the data.
The discussion, like the results section, is aptly named. You’ll find the most useful clinical information in this section.
Most authors use the discussion as an opportunity to take data from the results section and translate it to any clinical implications.
The discussion should also review any flaws and limitations of the study. For example, if the sample size was smaller than anticipated or there was difficulty in standardizing the collected measurements, it would be mentioned here.
If the author doesn’t state the study’s weaknesses, you should question the objectiveness of the research and take a closer look with a critical eye.
And to conclude—the conclusion.
The conclusion will be the last part of the study wherein authors restate their summarized findings and recommend any future study designs.
The conclusion is a great place to look if you are interested in performing research of your own as it can advise the next steps to take.
All studies should be referenced appropriately and it’s essential to at least glance through this section to determine if there is any bias.
The biggest takeaway here? Your best information may not come from the abstract; dig a bit deeper into the article.
Make sure you’re asking questions throughout the article: Were the researchers objective? Do they state if the results of the study have any clinical significance? The point of research is to learn and never stop asking questions continuously!
Why is Chiropractic Research Important?
By now you may be wondering why all of this is important.
As a chiropractor, you have so many skills in your tool belt, such as: thorough history taking, taping, soft tissue work, and spinal manipulation, to name a few.
Searching and reading relevant chiropractic research are essential parts of your tool belt because evidence-based practice drives decisions we make every day as clinicians.
The practical applications for having this skill are endless; let’s say you just attended a seminar and want to know if there is any evidence for the products or methods. Even more so if you’re going to explain or defend that treatment modality to a patient.
What if you have a question about a particular patient’s complaint or want to know if spinal manipulation is supported in treating that complaint?
The list goes on! You can see how beneficial it is for you and your patients when you can stay informed and educated.
The Next Steps on Your Chiropractic Research Journey
So, what’s next? You can’t spend all day searching the internet for the latest chiropractic research articles; even if you did, it would be impossible to pick out the most clinically relevant, then read and understand them.
But you can let the experts do the work for you.
Here’s a helpful link where you can subscribe to alerts offered by most major publishing sources.
While you may receive more articles than you can consume, at least you’ll have the tools in your chiropractic tool belt to review these articles quickly and with a critical eye!