Title : Using vaccine status, number of infections, and pre-omicron versus omicron variant status to more precisely predict the likelihood of both long covid as well as the severity of long covid in the U.S.
Abstract:
The risk of contracting long Covid varies widely by study. One meta-analysis reviewing over 40,000 long Covid patients across multiple studies from around the world found that the level of reported risk across such studies varied tremendously, ranging from 3% to 75%. This is due, at least in part, to the fact that many factors exert a causal impact on long Covid rates, as well as severity of long Covid, and that the incorporation of multiple key factors such as number of vaccinations, type of variant, number of infections, etc…, is necessary for a more precise population level picture. In 2020, The National Center for Health Statistics in the U.S. partnered with the U.S. Census Bureau to include health related questions within the Household Pulse Survey. This is a 20 minute survey intended to provide rapid and ongoing information about the impact of Covid within the United States. Data collection began in April 2020 and follows a two weeks on, two weeks off collection approach, and remains ongoing currently. Each module has its own separate weight and may be merged with other modules in order to increase sample size. The survey provides representative estimates for the adult population living in households at three different levels of geography: the 15 largest Metropolitan Statistical Areas, the 50 states plus the District of Columbia, and the U.S. nationally as a whole. This research merged multiple modules together for a sample size of over 500,000 in order to analyze both the risk of long Covid as well as the severity of long Covid based upon multiple key factors. These factors include (1) the number of Covid infections, (2) the time period of each Covid infection, (3) the number of Covid vaccines, (4) the time period of each Covid vaccine, (5) whether infections fell within the pre-Omicron time period, the post-Omicron time period, or across both time periods. Variables were constructed to capture multiple combinations, e.g., the number of infections pre-Omicron only versus post-Omicron only versus both pre-and post-Omicron by the number of vaccines received, etc… Results indicated that among those infected, the number of infections substantially increased the likelihood of both long Covid as well as severity of long Covid. This held even among those who experienced Omicron era infections only. The number of infections had a greater impact than the type of variant or the number of vaccines received, though the latter did provide some reduction against both the likelihood as well as the severity of long Covid. The results indicate a wide range of risk for both the likelihood of long Covid, as well as the severity of long Covid, and provide greater specificity in prediction based upon the various combinations of the aforementioned factors.
