Improve your oral health with personalized insights.
Get your kit

How we research our recommendations

Brian Maurer

May 9, 2022
5

minute read

Reviewed by:

Dr. David Lin, Ph.D.

How we research our recommendations

The oral health and care space today, like much of the health and wellness space, is unfortunately filled with too much noise. This noise includes junk science, anecdotal evidence ("it worked for me!"), and bad actors pushing products that haven't actually been shown effective. This makes our blood boil.

One of our goals at Bristle is to combat this by bringing a rigorous scientific lens to oral health to enable patients to better understand and own their health. We make hygiene and lifestyle recommendations based on a user's results that have been shown to be clinically effective. Below we will explain how we research the products and interventions we recommend. If you ever have any questions, please email us at info@bristlehealth.com.

Step 1: Checking for scientific literature

Our first step in vetting a new product or treatment is to look at the scientific literature, which usually starts at PubMed Central (PMC). This is an archive of biomedical and life science journal literature created by the National Institutes of Health and National Library of Medicine and serves as a "Google for scientific literature". For this example, let's search "nano-hydroxyapatite toothpaste" which is a toothpaste ingredient that has recently grown in popularity. We get these results:

Step 2: Vetting the journal and authors

The next step is to dig in to each article. We start by looking at which journal the paper was featured in:

  • Does the journal have any conflicting interests or bias?
  • Is it well respected in the research community?
  • Has the journal been found to be predatory?
  • Was the paper peer reviewed? (reviewed by other unaffiliated experts in the field)

Then we look at the authors:

  • Are they experts in a related field?
  • Are they from a legitimate research institution?
  • Do they have any conflicts of interest that may bias their outcomes?

(this doesn't automatically invalidate the study, but should be kept in mind)

Step 3: Evaluating study design

With these considerations in mind, we then will assess the study design. Some studies are listed in clinicaltrials.gov, and the data are published. Sometimes claims in the abstracts or summaries are not supported by the data, so we do a deep dive and read each paper thoroughly. There are a number of factors in properly designing a study that can affect the results and conclusions. Some of these include:

  • What was the sample size? (how many people/animals/time points were used?) Typically the larger the better.
  • Did they use proper controls? (a control is used as a baseline. In our example this would be someone using an alternative toothpaste that does not contain hydroxyapatite).
  • What type of study did they run? Randomized control trial (RCTs), cohort study, case-control? RCTs usually provide the most reliable answers as they provide the most rigorous process for removing bias and establishing cause-effect relationships. An explanation of study types can be found here.
  • Did they use the appropriate statistical analysis?
  • Did their data support their conclusions?
  • Are their findings consistent with other papers?
  • Are there any refutations of their findings? What did those studies find?

Step 4: Consulting our expert network

Once we have found a sufficient amount of evidence, our next step is to consult with our network of oral health experts to get their assessment of the treatment and how they have seen it perform first hand. Our expert network includes leading dental school professors and researchers, practicing physicians, and oral health policy experts from institutions like the University of Pacific School of Dentistry and CareQuest Institute for Oral Health.

In some cases we have found that although the journal evidence finds a positive conclusion, in the "real world" (not in a study) the treatment may not be effective for other reasons. Examples are a mouthwash that tastes bad so patients don't use it, or it leads to other undesirable effects like teeth staining.

Step 5: Evaluating commercial products

Once we have verified the treatment to be effective, we then look into suppliers of those products. In this research we again consult with our expert network and look into:

  • The product's active ingredients list
  • Customer reviews (on their site and third party sites)
  • Company background: Are they reputable? Does their mission align with ours?
  • Do we like the product? What do we or don't we like about it? (we test them on ourselves)

Step 6: Assigning confidence scores

Once we have performed the evaluation steps above we will assign the treatment each recommendation a confidence score. This confidence score is shown in the top right corner of each treatment recommendation card on your report:

Below is a breakdown of our confidence labels.

High confidence

A “High confidence” label indicates that, following our rigorous review of internal research, cited clinical evidence, and consultation with oral health experts, we confidently recommend this ingredient as an effective treatment for the health risks listed below.

Moderate confidence

A “Moderate confidence” label indicates that we have collected a significant amount of information about the efficacy of this ingredient, however, we require more and/or higher quality clinical evidence before assigning a “High confidence”. We believe the findings of the cited studies are useful and likely correct, but more research is needed to confidently recommend this ingredient as an effective treatment for its related health risks.

Being investigated

A “Being investigated” label indicates that we consider early research on this ingredient to be promising, but we require more and/or higher quality clinical research to confidently recommend this ingredient as an effective treatment for the health risks below.

Conclusion

By this point we have established the scientific evidence behind the product, the performance of the product in the real world (experts + our testing), and the credibility of the company behind it, so we can feel confident listing the recommendation on your report.

Improve your oral health with personalized insights
Get your kit

Join our community!

Want helpful tips and research digests delivered directly to your inbox? Sign up here.
You can unsubscribe from these communications at any time.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

How we research our recommendations

Brian Maurer
May 10, 2022
Reviewed by:
Dr. David Lin, Ph.D.
5
  minute read

The oral health and care space today, like much of the health and wellness space, is unfortunately filled with too much noise. This noise includes junk science, anecdotal evidence ("it worked for me!"), and bad actors pushing products that haven't actually been shown effective. This makes our blood boil.

One of our goals at Bristle is to combat this by bringing a rigorous scientific lens to oral health to enable patients to better understand and own their health. We make hygiene and lifestyle recommendations based on a user's results that have been shown to be clinically effective. Below we will explain how we research the products and interventions we recommend. If you ever have any questions, please email us at info@bristlehealth.com.

Step 1: Checking for scientific literature

Our first step in vetting a new product or treatment is to look at the scientific literature, which usually starts at PubMed Central (PMC). This is an archive of biomedical and life science journal literature created by the National Institutes of Health and National Library of Medicine and serves as a "Google for scientific literature". For this example, let's search "nano-hydroxyapatite toothpaste" which is a toothpaste ingredient that has recently grown in popularity. We get these results:

Step 2: Vetting the journal and authors

The next step is to dig in to each article. We start by looking at which journal the paper was featured in:

  • Does the journal have any conflicting interests or bias?
  • Is it well respected in the research community?
  • Has the journal been found to be predatory?
  • Was the paper peer reviewed? (reviewed by other unaffiliated experts in the field)

Then we look at the authors:

  • Are they experts in a related field?
  • Are they from a legitimate research institution?
  • Do they have any conflicts of interest that may bias their outcomes?

(this doesn't automatically invalidate the study, but should be kept in mind)

Step 3: Evaluating study design

With these considerations in mind, we then will assess the study design. Some studies are listed in clinicaltrials.gov, and the data are published. Sometimes claims in the abstracts or summaries are not supported by the data, so we do a deep dive and read each paper thoroughly. There are a number of factors in properly designing a study that can affect the results and conclusions. Some of these include:

  • What was the sample size? (how many people/animals/time points were used?) Typically the larger the better.
  • Did they use proper controls? (a control is used as a baseline. In our example this would be someone using an alternative toothpaste that does not contain hydroxyapatite).
  • What type of study did they run? Randomized control trial (RCTs), cohort study, case-control? RCTs usually provide the most reliable answers as they provide the most rigorous process for removing bias and establishing cause-effect relationships. An explanation of study types can be found here.
  • Did they use the appropriate statistical analysis?
  • Did their data support their conclusions?
  • Are their findings consistent with other papers?
  • Are there any refutations of their findings? What did those studies find?

Step 4: Consulting our expert network

Once we have found a sufficient amount of evidence, our next step is to consult with our network of oral health experts to get their assessment of the treatment and how they have seen it perform first hand. Our expert network includes leading dental school professors and researchers, practicing physicians, and oral health policy experts from institutions like the University of Pacific School of Dentistry and CareQuest Institute for Oral Health.

In some cases we have found that although the journal evidence finds a positive conclusion, in the "real world" (not in a study) the treatment may not be effective for other reasons. Examples are a mouthwash that tastes bad so patients don't use it, or it leads to other undesirable effects like teeth staining.

Step 5: Evaluating commercial products

Once we have verified the treatment to be effective, we then look into suppliers of those products. In this research we again consult with our expert network and look into:

  • The product's active ingredients list
  • Customer reviews (on their site and third party sites)
  • Company background: Are they reputable? Does their mission align with ours?
  • Do we like the product? What do we or don't we like about it? (we test them on ourselves)

Step 6: Assigning confidence scores

Once we have performed the evaluation steps above we will assign the treatment each recommendation a confidence score. This confidence score is shown in the top right corner of each treatment recommendation card on your report:

Below is a breakdown of our confidence labels.

High confidence

A “High confidence” label indicates that, following our rigorous review of internal research, cited clinical evidence, and consultation with oral health experts, we confidently recommend this ingredient as an effective treatment for the health risks listed below.

Moderate confidence

A “Moderate confidence” label indicates that we have collected a significant amount of information about the efficacy of this ingredient, however, we require more and/or higher quality clinical evidence before assigning a “High confidence”. We believe the findings of the cited studies are useful and likely correct, but more research is needed to confidently recommend this ingredient as an effective treatment for its related health risks.

Being investigated

A “Being investigated” label indicates that we consider early research on this ingredient to be promising, but we require more and/or higher quality clinical research to confidently recommend this ingredient as an effective treatment for the health risks below.

Conclusion

By this point we have established the scientific evidence behind the product, the performance of the product in the real world (experts + our testing), and the credibility of the company behind it, so we can feel confident listing the recommendation on your report.

Join our community

Get helpful tips and research digests delivered directly to your inbox
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.