Since the epidemic began, we’ve been doing virtual brown-bag seminars, with speakers from Berkeley and around the world. Jessica Metcalf, and epidemiologist and demographer at Princeton and David Reher, an historian and demographer at the University Complutense of Madrid, both gave wonderful talks. Afterwards they responded to some of the chatroom questions by email. (Thanks, to Julia Mason for preparing this post!)
12:17:02 From Joshua Goldstein: Question for Jessica: what is the IFR CFR (infection fatality rate and case fatality rate) gap for other outbreaks, eg. SARS?
Jessica Metcalf: Very few asymptomatic individuals for SARS (as far as we know, extensive serology wasn’t done, so hard to tell); that is thought to be why SARS was easy to contain. As mentioned on the call, many infections have asymptomatic phases.
12:18:22 From Joshua Goldstein: What is your opinion about what kind of infection rates we’ll see in wave 1, wave 3, and further on?
Jessica Metcalf: As long as individuals remain susceptible, I anticipate people will keep getting infected…. Although numbers decline once the threshold for herd immunity reached (i.e., 1/R0 so between ½ and 1/3) chains of transmission continue to occur, and the final size can be up to 80% of the population even for these values of R0. The exact distribution across the waves will depend on social distancing, etc.
12:19:14 From David Steinsaltz: There’s an interesting new paper by Modi et al. trying to lower-bound the IFR based on an estimate of excess mortality in Italian regions https://www.medrxiv.org/content/10.1101/2020.04.15.20067074v2.full.pdf
Jessica Metcalf: Indeed, many such efforts underway.
12:20:14 From Donehower: Question for Jessica: Has there been another infectious disease with such a long asymptomatic period? Does this combine with the lack of thorough testing to upend some of the math? Are we modeling with unprecedented levels of uncertainty?
Jessica Metcalf: The asymptomatic period is not unusually long. The latent period (time from infection to infectiousness) is estimated at around 5 days, with ~ one day of transmission occurring without symptoms, and then ~5 days of symptoms, for people who show symptoms.
12:24:19 From Ann: is it possible that children have cross-immunity from Polio vaccinations?
Jessica Metcalf: The enteroviruses (which include polio) are a completely different group of viruses. That doesn’t mean that cross-reactive immunity is impossible, but the features that the immune system is looking for are less likely to resemble each other.
12:26:21 From Magali Barbieri: Could there be a relationship between the low rates of infection/disease for women and those for children (the latter inheriting their mother’s immunity)?
Jessica Metcalf: Maternal immunity (from maternal antibodies) rarely lasts beyond a year or at most a few years, but it is possible we’re seeing low infection rates all the way up to the late teens (although, again, the data is shaky and uncertain…).
12:30:34 From Magali Barbieri: What do we now know about the relationship between contagiosity and severity of the disease?
Jessica Metcalf: No clear evidence for a relationship between symptoms and transmission yet – while it seems that coughing might be a good way of moving virus around, in fact some people have suggested that very effective transmission achieved by talking.
12:30:56 From rlee@demog.berkeley.edu: Might high transmissibility be heritable from one round of infections to the next, either because of a different strain of virus or because of the kind of person transmitted to, e.g. at a given workplace where all would have more contacts?
Jessica Metcalf: No evidence on strains of virus associated with differences in virus performance yet; certainly context and behavior are likely to matter for the magnitude of transmission.
12:32:41 From Nathan Seltzer: Are there differences in IFR/CFR for children across racial/ethnic groups? At what point in the age distribution do we see a divergence in outcomes across race? Do we have enough data for these comparisons yet?
Jessica Metcalf: Not enough data to address this.
12:34:42 From David Steinsaltz: Some evidence against the idea that children are just less likely to be exposed: The H1N1 pandemic in 2009 is believed to have infected 30-40% of school-aged children, 12-15% of young adults, but 3% of the elderly.
Jessica Metcalf: Well, yes. That and all the age contact pattern data, evidence for half a dozen directly transmitted infections, etc.
12:36:06 From jrcarey@ucdavis.edu: For Jessica: Case Fatality Ratio for 185 countries posted on Wash Post daily ranges from 3% for e.g. Germany to 15% in Belgium. What do you consider to be a good estimate of CFR?
Jessica Metcalf: As a demographer, I would never discount age ☺
12:36:40 From Amy Tsui: Given that 1/4 of US COVID-19 deaths happen to nursing home residents, is there a way to factor in such clustering in your model? Also many children <5 are in child care and have likely built up some immunities through continuous exposure to flu viruses. The interaction of age and such clustered living, seems to be a factor to be considered.
Jessica Metcalf: Clustering is certainly important – schools are thought to be critical for transmission as a result of clustering that occurs in classrooms, etc. How you appropriately include such phenomena in models will depend on the question you’re trying to answer. Flu virus immunity is unlikely to have any relevance to coronavirus immunity – they are totally different viruses.
12:45:44 From Joshua Goldstein: Question for Jessica on David’s talk. Is viral exposure in recent years protective or scarring for elderly living in institutions?
Jessica Metcalf: It depends on what kind… Viral exposure can mean all kinds of things.
12:47:54 From luisimac21: In developing countries only about 10% of elderly live alone or in an empty nest. Does this make them more vulnerable (higher probability of getting infected)?
Jessica Metcalf: Contact certainly increases the risk of infection. I think we still don’t know how this will play out.
David Reher: The basic issue is that we do not know as much about contagion as we should. The much lower incidence of elderly living alone in developing countries is well known, of course. In my view, these people should have somewhat lower rates of contagion than people living in larger households. That does not mean that the rate of infection will be near 0. Not at all. We should also remember that living alone makes a number of other factors affecting contagion and illness more difficult to manage. The key issue, in my point of view, is to identify the differences in contagion and in the management of contagion/disease for people living alone in comparison to people living in larger households. We do not know this at present either in the developing or in the developed world. Once contagion takes place, the management of illness can be more effectively carried out both at a societal level and for the person who is ill if he/she is living with someone else, especially a spouse. The weight of the population living alone in any given society multiplied by these (as yet unknown rates) will help us understand how they affect societies as a whole. Anything related to single living, ceteris paribus, will be less important at a societal level in the developing world.
[A recent paper by Mossong et al in Plos (2008) showed that persons living alone have fewer contacts with others than those in larger households, but still they had a mean of 8.8 contacts per day. This was estimated outside the period of epidemics, but clearly has much to say about what we are discussing.] -PLoS Med 5(3) e74. doi:10.1371/journal.pmed.0050074
12:50:17 From Joshua Goldstein: Question for David: are we seeing only the super spreader equivalent in nursing homes, where 80% of the nursing homes are actually quite safe?
David Reher: We might indeed be seeing that, at least in part. I believe, however, that the key here is the ability of the nursing homes to prevent the entry of contagion. They have been only moderately (at best) successful with this. Once in the door, however, limiting the spread of contagion in already-crowded facilities is extremely difficult. From a social standpoint, the question is what needs to be done in order to prevent the entry of contagion: Testing every visitor (not really feasible)? Forbidding entry from anyone outside the home? How does contagion enter nursing homes? Delivery persons, food supplies, staff, kin? Should nursing-home staffs be confined to the nursing homes where they work during an epidemic? You can see the difficulties involved. Once contagion is in the doors, it might indeed be a question of super spreaders o, perhaps, because contagion and age are linked, independent of co-residence, with the very old much more likely to become infected and, in those cases, to develop serious symptoms. A perfect storm?
12:50:33 Question From Donehower: Is there potential for this virus to FINALLY make us take care jobs seriously – make them well paid, secure jobs that people make into careers? Is anyone talking about that?
David Reher: I have no answer to what is a clearly important question. Its importance, however, is not directly related to the Covid epidemic. On a larger scale, we might also ask whether or not the epidemic will make people value health workers generally (who are highly vulnerable during the epidemic) more highly. I hope so, but I do not know.
12:50:41 From Joshua Goldstein: Another question for David: what about 5 years down the line, will the elderly be locked in for a really long time?
David Reher: I do not know Josh. They might, but where? In group quarters? On desert islands? That sort of solution would do no good for them at all –especially in group quarters, would it? The feasibility of that sort of reaction in any given society, I suspect, would be conditioned by the importance given to the elderly in that society. Culture? Maybe.
12:51:09 From rlee@demog.berkeley.edu: As we move toward opening economies, would it be a good policy to separate elderly from co-residence situations with working age and children, and isolating them? Brutal, but would it save the lives of the elderly?
David Reher: Ron. My response here is similar to the one for Josh. Where would you put the elderly? If they live in group quarters, death rates would be correspondingly higher. Forcing them to live alone with no contact at all with the outside world? That might save lives (or perhaps not), but in some contexts it could have a very debilitating effect (mentally, socially, emotionally) on the elderly persons themselves. It would cut them off –de facto- from societies and from their families. I suspect that many might become very depressed. In a familistic society like in Spain, at least in theory, most elderly maintain close and on-going contacts with their kin (but not everyone). In any case, I really don’t know, we haven’t been there yet.
13:00:12 From William Dow: Rather than REQUIRING elderly to be isolated, I’d like to see ideas for providing comfortable OPTIONS for worried elderly who want to temporarily get out of risky living situations.
David Reher: Will. A question for you: have you got any suggestions? The issues are potentially very important. If we had a go-round conversation about this, I imagine a number of good suggestions would surface. Maybe we should –at least for the future. When doing this, however, it is important to remember that everyone is a protagonist (a subject) – and it is not just a question of younger people deciding what to do with the elderly. I imagine that – once again in a familistic society — many elderly would not take kindly to this sort of suggestion. It would make an interesting script for a Sci-Fi movie. At another level, why would society want to implement something like this? Would it be to protect the elderly from disease and from society, or to protect society from the elderly?
13:01:39 From rlee@demog.berkeley.edu: Lots of empty hotel rooms could be made available for elderly, but I think very few would voluntarily leave family.
David Reher: I concur!
13:08:37 From William Dow: David: once we have reliable rapids, could we test every nursing home worker for active infection when they arrive at work everyday?
David Reher: Ah Will, reliable records. I agree!!! That is the entire point of my talk, isn’t it? Testing care workers daily is not really possible, mostly because there are few immediate results to tests and the workers might complain about the invasion of their civil rights. Cloistering care workers in the rest homes? Another potential solution that is just as unpopular. I wonder.
Jessica Metcalf: Issues with this – you could be infected, but not test positive for the first few days since viral titres are low; and even later, false negatives are possible. So you are always running a risk. The risk could be carefully mapped to decide if it was acceptable – more and more data on these false negative/positive rates are becoming available.
13:08:50 From rlee@demog.berkeley.edu: Nursing homes are caught between the need for isolation but also need for workers to care for the elderly, and they sometimes compel infected workers to continue to work. They are also criticized for abandoning the elderly due to no workers.
David Reher: I agree completely. As we can see, there are not too many easy solutions to this.