1 of 27

Social emergency Medicine Research

SEMINAL STUDY

2.1

1.4 // Projects in Social Emergency Medicine

3.1 // Policy, Advocacy & Activism

2 RESEARCH & METHODS

2 of 27

Research in Social Emergency Medicine

Social Emergency Medicine Series

3 of 27

Disclosures

None

4 of 27

Objectives

Discuss 3 common quantitative research approaches used in Social Emergency Medicine

Describe the pros and cons of qualitative vs. quantitative research

Discuss the interaction between research, policy, and politics in regard to Social Emergency Medicine

5 of 27

Outline

Introduction
Quantitative vs. Qualitative Research
Overview of Specific Research Methods
Conducting Research (IRB, funding, publishing)
Measuring SDH in Research
Politics and Research in SEM
Using Research to Affect Policy Change

6 of 27

Introduction

7 of 27

This Photo by Unknown Author is licensed under CC BY

8 of 27

Quantitative vs. Qualitative Research

9 of 27

This Photo by Unknown Author is licensed under CC BY-NC

10 of 27

This Photo by Unknown Author is licensed under CC BY-SA-NC

11 of 27

Measurable

Discrete

Gives point estimate

Comparable

Pros

Lacks Nuance

Lacks Individuality

Cons

Quantitative Research

12 of 27

Specific Research Methods

13 of 27

This Photo by Unknown Author is licensed under CC BY

Hierarchy of Evidence

14 of 27

This Photo by Unknown Author is licensed under CC BY-NC-ND

Randomized Controlled Trial

15 of 27

This Photo by Unknown Author is licensed under CC BY

Observational Studies

Since not everything can be randomized, we use some of the retrospective, observational, and epidemiological study types to gather the best evidence we can.

Sometimes, advanced statistical methods can be used to simulate randomization in these types of studies. For example, propensity score matching.

In this chart you will see the most common type of observational and epidemiologic studies. Many of you may (reluctantly) remember this from medical school.

An example of a cross sectional study would be to all patients in the Emergency Department if they have ever been food insecure

An example of a case control study would be to identify all patients with an opioid prescription in a community and compare those who have had an overdose and those who have not and then see if they had a prescription for naloxone within the year prior. You would then analyze if patients with an overdose were less likely to have a naloxone prescription. Remember, this is just an association and not causation.

An example of a prospective cohort study would be to identify people who are suffering homelessness who are currently in the Emergency Department and follow them for 6 months to see who ends up with stable housing. Then, you can analyze what factors differ those who obtained stable housing and those who didn’t.

16 of 27

Confounders

One of the major limitations of the observational studies is the presence of confounders.

This is why you cannot establish causation with observational studies. If you know about a potential confounder, you can use statistical methods to partially control for it. But, oftentimes the confounder is not known or suspected.

However, in instances where randomization cannot be performed, multiple observational studies are sometimes the best evidence that will ever be available and causation is assumed. For example, tobacco and cancer. While there is no randomized study of smoking to not smoking, multiple observational studies in different populations, with different cancers have shown an association. When controlling for many possible confounders, the association is still quite strong. There is a pathophysiologic mechanism supporting the role of smoking and cancer. This body of evidence is accepted by most as evidence of causation.

17 of 27

This Photo by Unknown Author is licensed under CC BY-SA

Evidence Based Medicine vs. Eminence Based Medicine

Evidence based medicine involves using the best evidence we have to make a decision regarding the treatment for a patient.

There will never be an RCT for every intervention. Some cannot be studied. The classic example of this is an RCT for parachutes for jumping out of a plane. No one would ever do this.

But, you also must consider that not every population is included in an RCT. For example, the initial studies for COVID-19 antivirals used only patients who were not vaccinated. Nonetheless, many physicians prescribed these antivirals to high risk vaccinated individuals based on their understanding and extrapolation of the data and risks/benefit.

Further, patient values and expectations must also play an equal role in treatment decisions.

Not necessarily shown in this chart, it is also important to consider social determinants of health like access to care, homelessness, and finances when making treatment decisions. Prescribing a drug that is 90% effective but the patient cannot afford is 0% effective. It might, then, be better to prescribe a drug that is 75% effective but is affordable to the patient.

18 of 27

Conducting Research

19 of 27

This Photo by Unknown Author is licensed under CC BY-SA-NC

This Photo by Unknown Author is licensed under CC BY-SA

20 of 27

Research starts with designing the concept of a study.

Then, you must consider if what resources you need to conduct your study. Do you need research assistants to conduct a survey? Do you need pay for laboratory testing? Do you need a statistician? You may need funding or grants to do this. Your time is not infinite and is likely one of the major resources needed for any study.

If your study involves human subjects, you will need to obtain IRB approval. Based on what type of study you have and your institution, you may also need to have they study reviewed by a Scientific Review Committee to evaluate the methods. Recently, the US Department of Health and Human services updated their regulations so that multi-site studies only need a single IRB review at one of the sites.

Conducting the actual study will vary widely based on what type of study you have designed. It may involve you giving out a survey to medical students after teaching them about reproductive rights, or it may be a multi-institutional study on dispensing naloxone in the ED. The latter may have a research coordinator and several research assistants and require frequent submissions to various oversight groups.

Once you study is done, you will need to write it for publication. This typically follows the IMRAD (Introduction, Methods, Results, and Discussion) approach to scientific publication. You will need to select a journal to submit it to. Then, if not rejected, it will undergo editorial and peer review and likely be sent back to you for edits. You will resubmit your paper after edits and it will be reviewed again. Eventually, it will be accepted for publication.

Conducting research may be a required component of your position, particularly if you teach/practice at an academic institution

21 of 27

Social Determinants of Health in Research

Historically, social determinants of health have been largely ignored in clinical research. There are many components to this.

Minorities of all types, including pregnant patients, have been underrepresented in research studies. Despite efforts to try to improve this, significant disparities still exist.

Measuring social determinants of health can also be difficult. Inclusive language can help, but t is not the only intervention that will help. For example, race is often recorded in the medical record based on observation by registration staff and not self report. Some medical records do not have different fields for sex and gender.

Many tests and clinical guidelines are also based on misunderstanding of the “biology” of race. One well known example is the estimated glomerular filtration rate calculation. This typically included a race component, which was added with the understanding that Black people (males) had more muscle mass than White people. There is and was no biological evidence for this.

22 of 27

This Photo by Unknown Author is licensed under CC BY-SA-NC

Black patients have been particularly abused historically in research – which understandably leads to a lot of distrust in medicine and research. In the most notable examples, Black prisoners in Tuskegee with Syphilis were coerced into participating into an observational study where they were not given treatment for syphilis. During the course of the study, it became accepted and clear that penicillin could treat Syphilis. However, none of the study participants were told this or offered treatment. Many of them died of this. This experiment is regarded by many as the impetus for the creation of the modern-day IRBs.

Another example is that of Henrietta Lacks, who had her cancer cells removed during surgery without her permission. These cells were then developed andsold and are one of the most widely used cancer cell lines in the US today. Ms. Lacks and her family received no compensation or recognition for any of this. It was largely ignored until the early 2000s, when a book on the topic brought it to light.

These are well known examples in Black patients, and it is understandable why there would remain distrust in the medical and research.

However, this can further exacerbate disparities as Black patients are less likely to be willing to participate in research and then they continue to be excluded from research evidence. This needs to be actively combatted.

Note, these are some well known examples, but other minorities have had similar experiences and may have similar feelings.

23 of 27

This Photo by Unknown Author is licensed under CC BY-SA-NC

This Photo by Unknown Author is licensed under CC BY-SA

Research and Policy

Most social issues in Emergency Medicine are grounded in long-standing local, state, and national policies.

High quality research is needed to effect policy change. Understanding the research is key, but in order to enter the political realm, presentation skills are also important. Despite that Washington D.C. is often heralded as an “evidence free zone”, evidence is, in fact, needed to convince political leaders and the public of the need for change. While you may not have an interest in conducting research, understanding how to present research findings to non-medical political leaders and the public is an equally important skill.

However, the interaction between research and policy/politics is two directional. One of the most glaring recent examples is the political interaction with COVID-19 policies regarding masks, vaccinations, and treatments. Policies and opinions were often developed based on political views instead of expert understanding of the evidence.

24 of 27

Summary

Research in social emergency medicine can be difficult for many reasons

Just like in other aspects of medicine, the approach to social emergency medicine topics should be driven by evidence

Having quality research is key to affecting change

25 of 27

Objectives

Discuss 3 common quantitative research approaches used in Social Emergency Medicine

Describe the pros and cons of qualitative vs. quantitative research

Discuss the interaction between research, policy, and politics in regard to Social Emergency Medicine

26 of 27

This Photo by Unknown Author is licensed under CC BY-SA

27 of 27

SEMINAL STUDY

Policy, ADvocacy & Activism

3.1

4.1 // Race & Racism: Case and Introduction

4.2 // race & racism: Effects of Racism

Up next: