Effectiveness of Vaccines

January 22, 2021

Information

Marietta Vázquez, MD, professor of pediatrics (general pediatrics); vice-chair, diversity, equity, and inclusion, Department of Pediatrics; director, Yale Children's Hispanic Clinic (Y-CHiC), discussed the progression from efficacy in trials to effectiveness under “real-world conditions” for COVID-19 vaccines.

ID6104

To CiteDCA Citation Guide

00:00Thank you for for the for the invitation.
00:04I'll be. Brief,
00:07with the time that I have today,
00:10I'm going to be talking about effectiveness,
00:13so going from efficacy to effectiveness
00:15I I was a principle investigator in a
00:19randomized clinical trial for fieser
00:21an I've participated as a speaker in
00:24educational events for both Merck and Sanofi.
00:27So we're talking about going
00:29from efficacy to effectiveness.
00:32Strictly speaking,
00:32efficacy and effectiveness have
00:34very different meanings in the
00:37context of vaccine development.
00:39But there oftentimes they're
00:41used interchangeably, so,
00:42so in terms of similarities,
00:45they both present or represent the
00:49percentage reduction of disease amongst.
00:53People who are who are vaccinated but
00:55but there are obviously differences.
00:57Vaccine efficacy.
00:58As you've heard,
00:59the previous speakers refer to
01:01how well the vaccines perform
01:03under the most ideal conditions,
01:05and these are mostly randomized
01:07clinical trials.
01:07These are experimental studies
01:09that are considered the gold
01:10standard vaccine effectiveness,
01:12which I'll talk about in the few
01:14minutes that I have remaining refers
01:16to how well a vaccine performs under
01:19real world conditions under less than
01:22perfectly controlled circumstances.
01:23So what happens when a vaccine?
01:26Is given to individuals who have
01:28comorbidities, who are different ages.
01:30When the very strict criteria of
01:32all of us who've gotten are some of
01:35us have gotten our covid vaccine,
01:37I counting the interval.
01:39did I get this one? You know how many days?
01:42But as we know,
01:44in real in in real practice this
01:46is not often adhere to,
01:48so this slide shows you the.
01:51Scheme of randomized clinical trials.
01:53When you have a large group of
01:55susceptible individuals and those
01:57are randomized to receive or not
01:59receive the intervention in question.
02:00In this case it will be a vaccine and
02:03the direction of the study is looking
02:06forward in time to try to figure out.
02:09Which the proportion of.
02:11In each one of these groups of individuals
02:14who develop disease and is the vaccine works.
02:18Obviously, those who receive the vaccine
02:20are going to have lower rates of infection,
02:23but even though these studies
02:25are the gold standard,
02:26they have certain disadvantages.
02:28Their advantages, well, we've.
02:29I've mentioned before.
02:31They are the gold standard.
02:32Their experimental studies where the
02:34allocation or the intervention is
02:36given randomly, which minimizes bias,
02:38and for the most part,
02:40these studies are blinded, so it.
02:42Really ask helps you an unbiased
02:45ascertainment of the outcome,
02:47but they have disadvantages.
02:49They have poor statistical power for things,
02:52diseases that are buried rare.
02:54You need very large samples.
02:57They are very costly an oftentimes to
03:00selected population is studied again,
03:02sort of.
03:03Drawing parallels with Covid vaccine.
03:06Young individuals were not studied.
03:08Pregnant women were not studied.
03:11An individuals with significant
03:13comorbidities are oftentimes excluded.
03:14The follow-up is limited.
03:16Usually these are conducted for
03:18one of three years.
03:20Again, if you think of covid vaccine,
03:23everything was done much quicker,
03:25but for the most part there
03:27follow-up is limited,
03:28and this leads us to think
03:31about generalizability.
03:32How how did those results from the
03:35randomized clinical trial translate
03:37to the patient that we see and an?
03:40What is the the efficacy in in
03:43the general clinical setting?
03:45So, so the questions that remain
03:47on answers would be the efficacy.
03:49In actual practice, its efficacy,
03:51overtime, what happens five years later?
03:53What happens 10 years later and then when
03:56you start thinking of subgroup analysis,
03:58so populations if all of a sudden
04:01cardiac patients seemed to
04:03have a higher rates of disease,
04:05will it work as well in patients
04:07will have that type of comorbidity?
04:09As I mentioned,
04:11different age groups and ethnic groups
04:13which is very important especially now.
04:16In the face of COVID-19,
04:18where we see an infection that
04:21seems to to affect both in morbidity
04:24and mortality in in higher rates,
04:28certain a certain ethnic group.
04:31So effectiveness studies and now
04:32use the case control methodology,
04:34'cause that's the one that my
04:36collaborators and I have used for many,
04:38many years, is an observation.
04:40ULL study an in this case.
04:42This is the schematic of what
04:44case control study would be,
04:45and this is the study that we
04:47used to assess effectiveness.
04:49You select cases you select
04:50participants based on the outcomes.
04:52So the direction of the observation
04:54is completely the opposite.
04:55The cases are going to be the
04:57individuals who develop the
04:59disease in question controls other
05:00persons without the disease.
05:02And then you look backwards in time.
05:05To try to ascertain which proportion
05:07in each one of these groups developed,
05:10sorry were that were vaccinated,
05:12and if the vaccine is effective,
05:14then you would expect fewer of
05:16the cases to have been previously
05:18vaccinated as compared to the controls.
05:21And these studies have many advantages
05:23there statistically powerful
05:25with much smaller sample sizes,
05:27and I'll show you in some of the
05:30examples in the next few slides,
05:32you don't need longitudinal follow-up
05:34because there is no intervention.
05:36They are ethically acceptable
05:38because let's face it,
05:39after a vaccine is licensed
05:41and recommended for use,
05:43nobody is going to want to do a randomized
05:46trial where some people get something
05:48that is recommended and others are
05:51are going to are going to be withheld,
05:53something that that the
05:55scientific groups recommends,
05:56but they have disadvantages.
05:57They're not experimental,
05:58and there are susceptible to bias so.
06:02Is it really?
06:04When we now put this into into context
06:07with what's happening with Kovid vaccine,
06:09even though CDC and many others
06:11plan to track the effectiveness,
06:14we know that randomized clinical trials
06:16for covid vaccine have shown that this
06:20vaccine is highly highly effective.
06:22But it remains to be seen if indeed
06:25this efficacy of about 90% or higher
06:28in the ideal setting will will
06:31remained in everyday practice an we
06:33know that ongoing studies are needed
06:36to be able to address the questions
06:39that from the get go we have,
06:42but also unexpected relevant clinical
06:44clinical questions that that remain so.
06:46Our research team at Yale with
06:49collaborators over many years
06:51in other medical centers such as
06:53CDC and others have conducted
06:55studies of effectiveness of.
06:57Vaccines for multiple vaccines over the year,
07:00and I will share for this talk just
07:03one or two studies due to time,
07:06but show you how results of a study on
07:09this slide of effectiveness of varicella
07:12vaccine and as you can see here under 1000,
07:16subjects were included in the study,
07:18and even though the number of
07:20participants was small by enrolling cases,
07:23children with varicella and
07:25controls children without,
07:26we were able to assess the
07:28effectiveness of this vaccine.
07:30With great certainty the the method of
07:33measure that's used is the Max odds ratio,
07:36and that's what's used on
07:39case control studies,
07:40and we were able to assess the
07:44effectiveness of the vaccine.
07:46Furthermore,
07:46not only are we able to assess
07:49the overall effectiveness,
07:51but we are able to target
07:53important questions in subgroups,
07:55so again using the varicella
07:57vaccine as the example,
07:59we were able to evaluate the
08:01effectiveness by time elapsed from
08:03the time of vaccination in this study,
08:06led in part to changes in actual vaccine
08:10recommendation in the United States.
08:12Originally this vaccine was
08:14recommended just to receive one dose.
08:16And then with data from studies
08:19of effectiveness,
08:20the recommendation changed to include two
08:24doses routinely to Everett to everybody.
08:28Another one of the of the analysis
08:31that can be done an and we're done,
08:34is questions that arise that
08:36we didn't even think about.
08:38So in the varicella vaccine
08:40in after licensure,
08:41there were questions of whether what was
08:44the ideal age to give it to a child.
08:47Then again,
08:48at that point it would have been impossible
08:51to then go back to the randomized trial,
08:54but using the same methodology with
08:56data that were already accrued.
08:58Our team was able to answer this
09:01question an address whether the
09:04effectiveness in different groups
09:06by age were different,
09:08which ended up not not being different.
09:12Once a vaccine is licensed and
09:15recommended for the population in general,
09:17as I mentioned,
09:18it's unethical to conduct a
09:20randomized clinical trial,
09:21so figuring out whether using two
09:24doses was better than one would have
09:27been would have been impossible.
09:29But given the fact that when the
09:31recommendation for two doses came about,
09:34not everybody at once got to doses
09:36using the case control methodology,
09:39we were able to compare the effectiveness
09:42of individuals with Windows.
09:43Versus Versus 2 doses. So you know.
09:47Again,
09:47studies of Affectedness are very
09:50useful in being able to to acquire
09:53data like this one,
09:54and then another example from
09:57a different study. Is that?
10:00We can even conduct an be able to
10:02assess effectiveness for individuals who
10:05never received the vaccine themselves,
10:07and this study was done about
10:0910 years ago when the topic of
10:12vaccination during pregnancy was not
10:14well and graced by patients and not
10:17by the entire medical community,
10:19and the rates of influenza
10:21vaccination were low,
10:22and for this study we set
10:25out to evaluate whether
10:26one vaccine given to a pregnant
10:29mother would be able to protect both.
10:32The mother and the infant,
10:34which is not only very important
10:36in terms of protection,
10:37but also in terms of in
10:39terms of cost effectiveness,
10:41where you have one vaccine,
10:42being able to to protect two
10:44individuals for briefly for this study.
10:47We did very similar structure.
10:48A case control study where the
10:50cases were infants who were
10:52hospitalised due to influenza and
10:54controls were infants who were very
10:56similar who did not have influence.
10:58And again we look backwards in time.
11:01But rather than assessing.
11:03Influenza vaccine in the infants.
11:05It was different.
11:06We were assessing influenza vaccination in
11:09the in their mothers in the pregnant women,
11:12and again the data from this
11:15study greatly greatly.
11:16It showed that the vaccine given to
11:19a mother during pregnancy was highly
11:22effective in preventing her infant
11:24in the first six months of life from
11:28being admitted with lab documented influence,
11:30and an these data.
11:33Lead to increases in the rate of
11:37vaccination during pregnancy and
11:39I think have helped move forward
11:42the vaccination.
11:43Program so really just sum up.
11:48After randomized clinical trials are
11:50conducted in Vaccine Start license,
11:52an many other vaccines are going to
11:55continue to come through the market.
11:57There are questions and our
11:59certainties that we have.
12:01At this point.
12:02We can all have them in our heads whatever
12:05questions concerning Covid vaccine.
12:07There are certainly plenty of controversies,
12:10but we know that many other questions will
12:13come up questions that we didn't think about.
12:16An observation ull studies that
12:18assess effectiveness of the vaccines,
12:20such as case control studies,
12:22can be used successfully to address
12:25both immediate and long-term
12:26questions regarding these vaccines.
12:28Thank you very much.