As we have seen, the 10th revision of the International Classification of Disease (ICD) is used to code events. The series of codes which interest us are the J codes:

ICD Code

• J09  Influenza due to certain identified influenza virus

• J10 Influenza due to other identified influenza virus 

• J11 Influenza, virus not identified 

• J12 Viral pneumonia not elsewhere classified

• J13 Pneumonia due to Streptococcus pneumoniae 

• J14 Pneumonia due to Haemophilus influenzae 

• J15 Bacterial pneumonia, not elsewhere classified

• J16 Pneumonia due to other infectious organisms, not elsewhere classified

• J17 Pneumonia in diseases classified elsewhere

• J18 Pneumonia, organism unspecified

Share

The first thing to note is that only deaths directly attributed to influenza are the ones with the J09-J11 codes. All the rest are Flu/F word-related. 

To be 100% sure that influenza is in the frame, we would have to look at notes examine lab results, and post-mortem results, if they exist. However, let’s assume the J09-J10 codes are accurate, for now. 

In February, we reported carrying out a simple calculation for English data. We took the ONS data, which provides separate reporting of the yearly data, and by subtracting data coded J09-J11 for the period 2015-2018 from the J09-J18 “pot”, we got the following totals:

The yearly influenza deaths are minimal. For example, in 2015, there were about 3 deaths per million, rising to roughly 24 per million in 2018. 

However, the other viruses, bacteria or agents which are not influenza (e.g. measles, S Pneumonia, M Pneumoniae) or are unknown or not identified are the ones adding far more to the tally than influenza. 

How do other European nations fare? A TTE reader, Eugenio Florean, managed to get similar coded data from ISTAT, Italy’s ONS equivalent, in record time (no mean feat). Here are the data for all age groups:

Considering that Italy has just under 59 million inhabitants and England just under 56 million, the figures appear very similar and very low.

However, the US CDC spews out data for tens of thousands of US influenza deaths. 

How is this difference possible? How can it come about? 

Part of the answer lies (as always) in the use of models. CDC uses four types:

• Rate-difference models

• Serfling least squares cyclical regression models,

• Serfling–Poisson regression models,

• Autoregressive integrated moving average models.

Readers will ask us what all this means. We must confess that we find it difficult to understand, let alone explain. 

But the text assured us that the four models give similar results, so all is well in the thousands. The only slight worry is the sentence: “Human subject review was not required for this study as only aggregate national data without personal identifiers were used in analyses”.

I.e. never mind who these folk are or what their circumstances were; we use their anonymised data and add it all up. This is a sign of the presence of a serious syndrome, probably incurable: the big data syndrome.

What about the rest of the EU then? Again, the European CDC counterpart, ECDC, assures readers of the fact sheet for citizens that around 40,000 folk succumb to influenza each year in the EU. Alas, for Germany, we were unable to find the J09-J10 data, but we found a report lumping J00-J99 diseases of the respiratory system with all ages for 2015, including lung cancer with a total of 68,300 - Still mega numbers but at odds with more specific coding. 

The rarity of the outcome “influenza death” is one of the explanations for its absence as an outcome in three long-standing Cochrane reviews of influenza vaccines, regardless of study design (yes, observational data were included). This is especially important as most of the studies were prospective with protocols. 

So what is going on? As Pontius Pilate said, “what is truth?” or should we correct him to say where is truth?

Reference

Thompson WW et al. (2009), Estimates of US influenza-associated deaths made using four different methods. Influenza and Other Respiratory Viruses, 3: 37-49. https://doi.org/10.1111/j.1750-2659.2009.00073.x