Some Interesting Research on Transmission

First, I am going to make a quick comment about masks.  It should be apparent from the resurgence of cases in Europe and the US how little masks do, if anything, to slow transmission.  The Danish study looms, so we are seeing a frenzy of activity by masking advocates and use of bogus studies like the IMHE one recently published in Nature Medicine and highlighted in a front-page story in my home town paper.  That study is the classic example of “if you tell it they work, it will tell you they work”.  Worthless.  You can take it from all this activity that the Danish study, a large, randomized trial, is indeed negative.  Maybe there are design or other aspects of that study that impact its conclusions.  Be nice if the powers that be would just let it be published.  What we are seeing is the exact opposite of the scientific attitude–“I want to believe that it works, so I am going to ignore any evidence that it doesn’t and promote even the sketchiest evidence that it does.”

This is a very useful and well-done study on the geographic patterns of infection spread.  (NBER Paper)   The authors attempted to disentangle factors influencing transmission levels, including the effect of mitigation efforts.  They also found that voluntary behavior changes had a bigger impact than mandatory ones.  Using cell phone data, significant drops in mobility were noted even before stay-at-home and other orders were issued, and some of these drops persisted even after lifting of these orders.  The lockdowns reduced employment by 12% and consumer spending by 7%.  Variation in lockdown stringency or timing did not provide explanatory power to variations in case rates across geographies.   They identified population density and overall population as two of the most significant factors, explaining over 50% of the variation.

Superspreaders is a frequently discussed explanation for transmission.  In this study, five examples of superspreading events were examined.  (Medrxiv Paper)   The researchers were also trying to estimate viral loads.  The authors describe the typical size of most droplets being emitted as quite small, but large enough to accommodate a substantial number of virus particles.  These small particles are aerosolized, or capable of floating for an extended time.  They also focus on variation in viral load and what that may mean in various settings.  Their premise is that superspreading events could occur both from a patient with a very high viral load or from a patient with an average viral load but emitting for some time in certain environments, primarily indoor areas with poor ventilation.  The five superspreader events were used to estimate probable viral loads and other transmission characteristics.  They believe an infectious dose is on the order of 300 to 2000 viral particles.   They believe that what they refer to as “enhanced” transmission by an infected person with an average viral load, but emitting for some time in an enclosed space, is more likely to explain superspreading events than emission by a person with a very high viral load, partly because those peak loads occur for only a relatively brief time.  If you are interested in how to think about transmission this is good paper to read, lot of explanation of variables to be considered.

Science magazine has an article reviewing knowledge on transmission issues.  (Science Paper)    The authors first note that household transmission is the primary location for spread, accounting for a very large fraction of infections.  Household includes any primary residence, including congregate care settings like nursing homes.  Within households, spouses are twice as likely to be infected as other household members, and symptomatic index cases appear more likely to transmit.  Older persons also appear to be at higher risk of getting infected and transmitting.  Community transmission obviously also occurs, but where and how is still not fully delineated.  Viral loads and other factors, like having coughing or sneezing symptoms may impact infectiousness.  Settings where one seriously infectious person has close contact with a number of other persons obviously can faciliate significant transmission.  It appears that the epidemic is characterized by over-dispersion in spread–that is, a few infected people are responsible for much of the transmission and many people are responsible for no transmission.  Broad and untargeted interventions (i.e., tell everyone to stay at home and make everyone wear masks) are unlikely to have a major impact on spread, but more targeted ones aimed at identifying and isolating those who may be highly infectious could help.

This paper from India examined the contributions of asymptomatic and symptomatic persons to spread.  (SSRN Paper)   This was basically a contact tracing study of over 3400 cases.  91% of these were asymptomatic.  Asymptomatic cases tended to be significantly younger.  While there was some spread by asymptomatic persons, the great majority of transmission occurred from symptomatic persons.