In my previous article in this series, I looked at whether we can trust the data on covid and vaccination. That is a very important question, and if you aren’t sure about it, I would encourage you to go read that piece first. The short version: Challenges with testing and reporting mean case counts are more like estimates than exact tallies, but the more important figures for deaths and hospitalizations seem generally trustworthy. Also, the rumor that the CDC is counting vaccinated and unvaccinated cases differently is based on a misunderstanding of a completely different data collection program. Looking at the reporting systems in place, and cross-referencing the official picture with other real-world data, there doesn’t seem to be a good reason to doubt the basic picture of the pandemic which is presented by the official numbers.
In this post, I’ll look at what the numbers tell us about vaccine efficacy. I’ll look first at whether covid is still a threat, then at how protective the vaccines actually are, then at whether natural immunity is better, and finally at the problems of waning immunity and the delta variant.
Is covid even a threat anymore?
Just a few months ago, it seemed like covid was on its way out. Now, not so much. This survey from Jim Geraghty (a conservative writer who has been a good source of covid info throughout the pandemic) includes many links to local news stories in states across the country describing hospitals strained to capacity.
I was genuinely astounded to see a recent article from Dr. Mercola dismiss stories like these, writing, “as for viral social media posts by doctors and nurses claiming hospitals are overflowing with unvaccinated COVID patients, don’t believe them. Most are bots.” A few days ago, an 11-county area in Texas had 1 ICU bed available, while 575 patients waited for a hospital bed to open up in Houston. In Florida, funeral homes report having to stack bodies waiting for cremation. In Alabama, there were “negative 29” ICU beds available. The overwhelmed nurses in this video report from Mississippi don’t look like bots. Again, these are local news reports, based on interviews with real medical workers dealing with the actual situation on the ground where they are.
Does this mean we need to panic? Does it mean we’re all going to die? Does it mean we need to lock down and wear quadruple-layer masks for the rest of our lives? Does it even mean we all need to be vaccinated? No, no, no, and not necessarily.
As Christians, we know we are secure in the care of our God and need not fear “the pestilence that stalks in darkness, or the destruction that lays waste at noon” (Ps. 91:6). Whatever comes to us personally, whether sickness or health, will be from our Father and for our good. We can also be thankful that covid is a relatively mild pandemic by historic standards, with low mortality rates compared to many other viruses. Furthermore, despite covid’s real dangers, we need not conclude that every proposed solution is wise, worthwhile, or effective. But none of these truths changes the fact that covid remains a genuine threat which is certainly affecting our neighbors and may very well touch us personally. That doesn’t tell us whether the vaccines are an effective solution—but it does make this an important question to consider.
What are the vaccines supposed to accomplish?
For the sake of speed, the vaccine trials only measured vaccine efficacy against symptomatic covid. The priority was getting vaccines which would keep people out of hospitals, and if the vaccines could accomplish that, the question of whether they prevented people from getting sick at all was a secondary priority. Since the trials were only designed to measure efficacy against symptomatic covid, that was all which could ethically be promised at the end of 2020 when they first came out.
Unfortunately, “Right now, we can’t promise anything other than reduction of symptoms” was misunderstood or misrepresented by many to mean, “The vaccines cannot do anything other than reduce symptoms.” An unanswered question was assumed to have a negative answer, and that negative answer has stuck, even after a plethora of studies have demonstrated that in fact the vaccines are quite effective at preventing infection.
As early as December 2020, a trial funded by Pfizer found the vaccines 95% effective at preventing covid. By February, studies from the Mayo Clinic and the UK measured 86-89% protection from infection, and a large study in March found the vaccines were 90% effective at preventing infection in healthcare workers. Too many other studies to link here have repeatedly reinforced these findings. At this point, there is simply no empirical basis for claiming that the vaccines don’t prevent infection.
If the vaccines provide significant protection against infection, that means they help to protect our neighbors by making us less likely to pass along the virus. However, a number of studies from the spring also demonstrated a significant reduction in viral load (and therefore presumably in infectiousness) in breakthrough infections of those who were vaccinated. This implies that those who are vaccinated are not only much less likely to become seriously sick, but are also less contagious if they do.
In the last couple months, waning immunity and the delta variant have changed some of these numbers, but the same basic picture remains. I’ll look at those recent developments in a moment, after taking a brief detour below…
But isn’t natural immunity better?
But, you might argue, natural immunity offers more effective protection against covid than vaccination does. The data has been debatable on this question, but evidence seems to be building that natural immunity from prior covid infection is in fact more protective than immunity from vaccination. This large preprint study out of Israel published just a couple days ago seems to strongly support that conclusion, and experts I’ve found credible throughout the pandemic are taking it seriously even though it has not yet been peer reviewed.
Observing a large study population from June 1 through August 14 (a period when delta was dominant in Israel), the study found that the likelihood of a previously infected person being reinfected was roughly 13 times lower than the likelihood of a vaccinated person getting a breakthrough infection. However, the different was between a tiny risk and a truly miniscule one. During the 2.5-month study period, only 1.05% of vaccinated people tested positive for covid, 21 out of 46,035 were hospitalized, and zero died. For comparison, 4 of 46,035 previously infected people were hospitalized during that period and, again, zero died.
It’s always a mistake to grab a single study and assume it is conclusive, but this one seems consistent with an emerging body of evidence that tentatively supports a double conclusion: vaccine immunity is remarkably protective, and natural immunity is even better.
So why not just rely on natural immunity? Well, the thing about relying on natural immunity from getting covid to protect you from getting covid is that means you have to get covid before you’re protected from getting covid (again).
Even though getting covid is very far from a death sentence, even for an elderly person, there are good reasons to avoid it. There’s the chance of death, obviously, even for younger people. There is also a relatively larger chance of long-term harms. Derek Thompson, who I’ve found careful and accurate even though I often disagree with him politically, calculates the rate of long-term illness at roughly 1-in-100 for a young man in his 30s. There is also the danger of harming others, either by passing covid along to someone more vulnerable or by filling up a hospital bed needed by, say, a cancer patient.
If you think covid is a serious enough risk to consider vaccination, then the protectiveness of natural immunity seems irrelevant since you have to catch covid to acquire it. (Of course, if vaccines are unsafe, it might be wiser to just accept the risks of covid and hope for the best. I’ll look at the safety question in my next post.)
Isn’t vaccine immunity waning? And what about delta?
As I explained above, a large body of studies in the spring demonstrated that the covid vaccines were surprisingly protective against covid infections and also seemed to reduce the contagiousness of breakthrough infections. However, recently both of those conclusions have been questioned, as the much more transmissible delta variant has emerged and as immunity in those vaccinated months ago seems to be waning. Together, these two trends mean that more vaccinated people are getting sick.
The antibodies from vaccination (or infection) cannot safely remain in our blood forever, so they gradually dissipate, leaving behind T cells and B cells which remember the virus and are prepared to fight it and produce new antibodies when it is detected. However, delta’s remarkably high viral load makes it more likely to get a foothold before our immune system can get up to full speed and beat it back. This means delta is producing relatively more infections among the vaccinated. But it’s important to understand what “more” looks like here, especially in comparison to those who do not have protection from vaccination.
I’m going to start with two graphs from the New York City Department of Health based on data from hospitals in the city. It is the best visual representation I have seen of trends which have been fairly consistent since delta came on the scene. The first graph shows the number of covid cases in NYC between June 26 and August 7, divided by vaccination status. Two things stand out. First, delta is more contagious for everyone. Second, vaccination makes a big difference.
The second graph shows hospitalization rates (the more important metric), where the difference is even more dramatic.
San Diego recently released data for July 26 through August 24. As of today, their vaccination dashboard shows 74.4% of their eligible population has been vaccinated. In the past month, the 3/4 of the population who are vaccinated had 3,913 cases, 12 hospitalizations, and 2 deaths. The 1/4 who are unvaccinated accounted for 28,509 cases, 603 hospitalizations, and 27 deaths.
Just three days ago, a new study in Los Angeles reported, “On July 25, infection and hospitalization rates among unvaccinated persons were 4.9 and 29.2 times, respectively, those in fully vaccinated persons.” (The study also notes that, “In July, when the delta variant was predominant, cycle threshold values were similar for unvaccinated, partially vaccinated, and vaccinated persons,” addressing concerns that samples of vaccinated and unvaccinated people are being assessed differently.)
National data from the last several weeks is not yet available because CDC data collection moves with bureaucratic slowness, but other local and state-level data shows a similar pattern, and I haven’t seen any which counters the basic picture from these cities. Delta is more transmissible, including for the vaccinated, but those who have been vaccinated are much less likely to get sick and even more unlikely to be hospitalized or die.
On a related note, there is some evidence that delta may have altered the contagiousness of those who get a breakthrough infection. The CDC’s call last month for renewed mask mandates was based upon a few studies which suggested that viral load in breakthrough infections was the same as that in the unvaccinated. The two main studies were an analysis of hospital data from India and a just-published study of a recent outbreak in Provincetown, MA. However, the director of the team which conducted the Indian study has said that the claim about high viral loads was removed from the latest version of the study after data from a third hospital was examined, and the methodology of the Provincetown study has been questioned. There are also a number of recent studies which have reached the opposite conclusion: 1, 2, 3, 4. At minimum, we can say the jury is still out on the question of the infectiousness of breakthrough cases. And even if they turn out to be equally infectious, the most recent data still shows the vaccinated are significantly less likely to be infected in the first place.
But what about other countries?
Some vaccine skeptics point to trends in other, more-vaccinated countries as evidence of a downturn in vaccine effectiveness that awaits us as well. The most darkly amusing example I’ve seen is an article breathlessly announcing that Gibraltar, with a 99% vaccination rate, had recently seen a 2,500% increase in cases per day. The accompanying graph shows an alarmingly sharp increase in cases beginning in late June as delta arrived. However, if you squint at the numbers on the graph, you’ll observe the increase is from 0-2 new cases per day (consistently since February) to around 30—among a population of 34,000. This delta “wave” also produced their first covid deaths since mid-March. Two of them. If that is vaccine failure, I’ll take it! (This mini-wave did prompt renewed lockdowns on the island. But that illustrates political insanity, not vaccine failure.)
The most common example of alleged vaccine failure is Israel, however. Israel had an early and broad vaccination campaign, so they offer some sense of what other nations can expect as months pass after vaccination. Delta became their dominant strain over the summer, and they have seen cases, hospitalizations, and deaths dramatically increase, including among the vaccinated. However, the significant difference between vaccinated and unvaccinated numbers persists in Israel, despite delta and some degree of waning immunity.
The latest figures from Israel’s Health Ministry show that unvaccinated Israelis, only 20% of the population, account for 50% of current serious hospitalizations. This is especially striking since the unvaccinated population is generally younger, and therefore should naturally be at less risk. On August 8, the Health Ministry reported that severe illness was 2.8 times more likely for the unvaccinated under 60, and 5 times more likely for those over 60.
This piece offers what seems to be a balanced take on what Israel’s experience suggests and the important questions which remain unanswered. It is not impossible that the bottom is about to drop out of vaccine immunity so the vaccines become essentially useless. However, for the moment Israel remains an example of the effectiveness of the covid vaccines to greatly diminish the risk of serious illness, even in the face of delta.
But does maintaining the protection of vaccination mean we’ll be forever dependent on booster shots? Israel last month began offering a third booster shot to the vaccinated amid their delta wave, and the Biden administration recently authorized the same policy in the US. However, many experts who are pro-vaccination have pushed back against “boosters for all” because they believe it is not yet warranted and lacks sufficient safety data.
This note from a recent report on Israel’s booster program helps to explain why many medical experts question the need for universal boosters (emphasis added):
Despite the decreasing effectiveness of the vaccine in preventing infection, it still provides significant protection against severe illness and death even without a booster shot.
As of August 21, the Health Ministry recorded 215.9 severe COVID-19 cases per 100,000 people among the unvaccinated over the age of 60, compared to 21 per 100,000 people among those who had received two doses of the Pfizer vaccine. This makes unvaccinated older people more than 10 times as likely to experience a severe case than their immunized counterparts.
With new vaccines combating a novel virus, much remains still unknown. But for the moment, all the data tells us that vaccination makes it significantly less likely that you will get covid, significantly less likely that you will get a serious case of covid, and significantly less likely that you will pass covid on to someone else. It would be an odd decision to avoid that protection now just because it might diminish by some unknown amount in months and years to come.
But, of course, that assessment assumes the vaccine is not itself a risk to your health. And we all know that safety fears are the main reason many are choosing not to be vaccinated. I will look at the safety question in my next post in this series.