Summary of Thesis

Background: Excess mortality due to seasonal influenza is substantial, and pandemics like COVID-19 call for timely mortality estimates. Methods used to estimate influenza-associated mortality typically use all-cause deaths, which is readily available in many countries, or cause-specific mortality data, which may be more specific to influenza but have substantial delays.

Method: For Denmark, Spain, and the United States, we estimated age-stratified excess mortality for i) all-cause, ii) pneumonia and influenza, iii) respiratory and circulatory, iv) respiratory, and v) circulatory causes of death for the 2015/16 and 2016/17 seasons. We quantified differences between the different categories with respect to their weekly and seasonal excess mortality estimates. The estimates were obtained using the EuroMOMO model on mortality data from 2010 through 2017.

Results: The respective periods of weekly excess mortality for all-cause and cause-specific deaths were similar in their chronological patterns. Seasonal all-cause excess mortality estimates for the 2015/16 and 2016/17 seasons were 15,068 deaths (10,582-19,558) and 46,292 deaths (42,047-50,540), for the United States. For Denmark they were 20.3 (15.8-25.0) and 24.0 (19.3-28.7) per 100,000 population. For Spain they were 22.9 (18.9-26.9) and 52.9 (49.1-56.8) per 100,000 population. Seasonal respiratory and circulatory excess mortality estimates were two to three times lower than the all-cause estimates.

Discussion: There are benefits to using a simple model based on all-cause mortality as it is timely and may approximate cause-specific estimates and the influenza-associated mortality. These findings have important implications for the development of future timely mortality monitoring systems during pandemics such as COVID-19.

Why is this important

It adds another perspective on the costs and benefits of the different widely used ways in which mortality can be estimated and monitored - especially wrt. seasonal pandemics like influenza. For instance, this can give more clarity on when to use one method over another depending on the intended purpose of the method within the field of public health and biosecurity/pandemic preparedness. One of the models I analyzed turned out to be one of the most used models during covid.

Strengths and weaknesses

Strengths

• I decided to do research on an existing model and collaborative network (EuroMOMO) because I thought it’d be important during a pandemic. This turned out to be true as the model and network I did research on became one of the most widely used models for mortality estimation during covid. 
• I had involved multiple excellent institutions which meant that people who normally don’t talk that much together and disagree quite substantially came together. This was probably super good for keeping high epistemic norms as opposed to confirming the view of a single research group.
• I included large datasets from three countries.

Weaknesses

• I didn’t spend enough time understanding the subfield and the opinion of the experts before I got excited about a particular idea/path forward.
• I was too ambitious and lacked good judgment about what can actually be accomplished within a time frame - planning fallacy is real! One example was that getting access to the relevant data took much much longer than my collaborators said. A piece of advice that helped me a lot was “it’s better to have one bird in your hand than ten on the roof”. 
• I had picked supervisors who are highly applied researchers (public health) so when covid-19 emerged I completely lost access to my supervisors which made the research quite difficult. Hardly a weakness but seems worth mentioning.
• Too many people were involved in my project so I had multiple supervisors which meant that no one had a sense of responsibility for my progress. 
  
  

Recommendations based on my experience

General tips:

• Planning fallacy is real so make sure that you have a backup/MVP version of your project. Especially if it involves other people and getting access to datasets. It can easily take 2 months longer than expected.
• If you’re relatively excited about research as a career path but aren’t sure whether you want to do a PhD, then taking a bit extra time to turn your thesis into an actual research project (e.g., by postponing your studies by a few months or a full semester) can be a great way to do explore research as a career path. 
• Having good chemistry with your supervisor - feeling as if they’re willing to invest in you - is important, particularly if you’re considering research as a career path.
• Approach your thesis from a learning/experiment lens. This will help you to see the challenges as an essential part of the learning process as opposed to personal shortcomings.
• Attempt to do at least some of it abroad - it’ll expand your world-view.