“XXL consumption figures are being used to influence alcohol policy in Germany,” says Dr Winfried Maatz

Interview | When it comes to German figures on alcohol consumption, data are cherry-picked, estimates and computed figures are jumbled together, relative figures are published without absolute figures – and lo and behold: Germany becomes a country of heavy drinkers. Which was presumably the plan. However, this is far from true and the physician Dr Winfried Maatz proves it in his recently published book*.

Your book is a real eye-opener. Your examples show how skilfully figures are used in the current debate on alcohol consumption to achieve desired outcomes. Could you please explain what prompted you to delve into this sea of numbers, and what do you mean by the “strategy of high figures”?

Dr Maatz: The actual reason I began to look closely at Germany’s alcohol consumption data was the WHO’s “No Safe Level” thesis. As a doctor and a scientist, I found it incomprehensible that every drop of alcohol should pose a health risk.

Consumption figures are apparently being inflated to achieve certain political goals.

Dr Maatz: That’s the strategy. XXL figures for alcohol consumption shall help push through tax increases, bans on alcohol and advertising, as well as warning labels on alcoholic drinks.

This tactic is particularly striking when it comes to the number of hospital treatments. Allegedly, in 2023, more people in Germany were receiving hospital treatments for alcohol problems than for pneumonia. How can that be?

Dr Maatz: As a doctor, I was involved in coding diagnoses on a daily basis in my clinical practice, so I am very familiar with the classification system (ICD). Here, less common diseases, such as alcohol-related diseases, are put into one group. In contrast, common diseases, such as pneumonia, are divided into several groups. If you then compare the number of alcohol-related diseases with just one of the five groups of pneumonia, you arrive at this misinterpretation. However, this shouldn’t just have been obvious to me as a doctor; common sense should have made anyone suspicious.

This shows how important it is to scrutinise figures and the methodology behind statistics. Anyone making such comparisons should always be aware that the methodology has a decisive influence on the result. Do you have another example?

Dr Maatz: A very recent one, in fact. The 2026 Yearbook on Addiction, published by the German Centre for Addiction Issues (DHS), states that 7 million people were receiving hospital treatments for alcohol consumption in 2024. With a total of 17.5 million hospital treatments, that would correspond to 40 percent. A figure like that should have raised eyebrows. In reality, however, there were only 283,500 treatments for alcohol consumption – a mere 1.6 percent. The fact that treatments for alcohol consumption have fallen by 28.9 percent over the past ten years, isn’t mentioned either.

It is outrageous to see which figures are included in per capita consumption of pure alcohol (Litres Pure Alcohol-LPA), such as unrecorded alcohol consumption or alcohol consumption whilst on holiday abroad. How on earth does one arrive at this figure?

Dr Maatz: This is a particularly good example of how figures are inflated. In the DHS Yearbook on Addiction, an addiction researcher added 0.5 LPA per capita of unrecorded alcohol consumption to the total in 2024.

That is the equivalent of approximately 8.9 litres of beer, 0.6 litres of spirits and 2.5 litres of wine. Per capita!

Dr Maatz: This estimate is highly questionable, as only 0.007 LPA of unrecorded alcohol can be calculated for Germany’s hobby brewers and vintners. Furthermore, it is scientifically inaccurate to mix computed figures with estimated figures and present the overall result as a calculated figure.

That brings us straight to the wildly contradictory figures on per capita LPA consumption in Germany. There are six different figures bandied about. Depending on the source, they vary between 9.0 and 11.6 LPA per capita. That is a considerable difference.

Dr Maatz: Yet per capita alcohol consumption is a very important indicator. It is used to monitor the increase or decrease in a country’s alcohol consumption over time. This figure is also used to compare countries with each other, and a country whose alcohol consumption exceeds the European average is then singled out as a “high-consumption country”.

So, temperance organisations might have a particular interest in high figures?

Dr Maatz: Absolutely. And that brings us back to the strategy.

Could you please explain that in more detail? How can different consumption figures be calculated from the same set of data?

Dr Maatz: I’ve wondered about that myself. During my research, I discovered certain methods, which are used to massage alcohol consumption upwards. That’s why I’ve made per capita alcohol consumption data a central issue in my book.

So how exactly are alcohol consumption figures adjusted upwards?

Dr Maatz: Well, first of all, you need to know that the figures for alcohol consumption in Germany are compiled by the DHS and by the Federal Association of the Spirits Industry (BSI). Up until 2022, the DHS and the BSI used the same database, and the per capita LPA figures reported by the two were almost identical.

Allow me to ask a follow-up question. There is only one per capita LPA figure, but there are various types of alcoholic drinks.

Dr Maatz: Per capita consumption is defined as the quantity of pure alcohol per person in the population aged 15 and over. This figure is derived from the quantities of beer, wine, sparkling wine and spirits. An average alcohol content is applied to each type of drink, which is then used to calculate the quantity of pure alcohol. Using this data, the BSI calculated a per capita consumption of 9.8 LPA for 2020, whilst the DHS arrived at a figure of 10.0 LPA. But in 2023, the DHS changed its team of authors, and they simply altered the previous calculation bases. They increased the average ABV for beer to 5.0%, from 4.8% ABV previously, and the ABV for wine and sparkling wine to 12.0%, from 11.0%. The key change, however, concerned the calculation of alcohol content for spirits. Instead of using retailers’ sales figures, the new authors switched to the volume sales of taxed alcohol issued by the Federal Statistical Office.

Hang on. Tax figures aren’t the same as consumption figures. The millions of bottles of spirits sitting on supermarket shelves may have been taxed, but they’re far from having been drunk.

Dr Maatz: That’s right. But by making these rather unconventional “changes”, the addiction researchers arrived at 10.3 LPA per capita for 2023, compared with 9.4 litres from the BSI. That’s a very significant difference of 0.9 LPA per capita. In total figures, the difference amounts to nearly 64 million litres of pure alcohol for Germany.

That’s an incredible amount. Does that already include the estimated volume of unrecorded alcohol?

Dr Maatz: No, the authors only did that when calculating per capita alcohol consumption for 2024. The figures have just been published in the DHS Yearbook on Addiction 2026.

I’m keen to see the figures.

Dr Maatz: The current DHS Yearbook on Addiction reports an even higher figure of 11.0 LPA per capita. This includes the estimated 0.5 LPA of unrecorded alcohol. As a result, for the first time in 30 years, addiction researchers have been able to show an increase in alcohol consumption in Germany.

And the public health authorities just accept that?

Dr Maatz: Not only that. The Green Party is basing its motion in the Bundestag to increase alcohol duties on this figure. With 11.0 LPA, Germany also immediately becomes a “high-consumption country”, as it exceeds the European average of 9.9 LPA. And the authors claim in the DHS Yearbook on Addiction 2026 that Germany will not achieve the 10 percent reduction target agreed with the WHO by 2025. Yet the figure of 11.0 LPA per capita is not only scandalous because it includes unrecorded alcohol consumption, it is also definitively incorrect due to a calculation error.

The DHS is publishing an incorrect figure in its yearbook? If so, then the whole structure of claims, assertions and estimates built upon those 11.0 LPA would collapse like a house of cards.

Dr Maatz: That’s right. The first error in the calculation of the 11.0 LPA lies with the Federal Statistical Office, which reported a massive increase in sales of taxed alcohol for 2024. The total figure of 194 million litres represented such a substantial increase that I had already cast doubt on it as early as 2025. At the time, the Federal Statistical Office was unable to provide me with an explanation. However, after the statisticians published a massive decline in taxed alcohol for 2025, I approached them again and demanded a plausible explanation. And indeed, in June 2026, the Federal Statistical Office confirmed to me that – due to double accounting – an inflated volume of taxed alcohol had been tabled for 2024.

And what is the correct figure?

Dr Maatz: In fact, only 155.2 million litres of alcohol were subject to duty in 2024. The next error is that the authors of the DHS Yearbook on Addiction adopted the figure of 194 million litres without verification when calculating per capita consumption. This means that the figure of 11.0 litres for 2024 is definitely incorrect.

What is the actual per capita consumption?

Dr Maatz: If one takes the BSI’s figures and calculation methods for the year 2024 as a basis, per capita consumption of pure alcohol stood at 9.3 litres. And this does not even take into account the negative net balance of 0.36 LPA per capita resulting from the cross-border trade for Germany.

That means Germany is not a high-consumption country when it comes to alcohol.

Dr Maatz: Not at all. Instead, Germany has already achieved - ahead of schedule - the target agreed with the WHO to reduce alcohol consumption by 10 percent by 2025.

Sweden is repeatedly held up as a role model for Germany. But in Sweden – supposedly such a teetotalling nation – not everything is as it seems, is it?

Dr Maatz: For many years, Sweden has been reported to have a consistently low per capita alcohol consumption. However, one must bear in mind that Swedes purchase considerable quantities of alcoholic drinks abroad. This, however, is not recorded in the excise tax statistics. Another aspect that has often been overlooked: Alcohol consumption in Sweden has risen slightly over the past 20 years, whilst in Germany it has fallen by 42 percent. Although Sweden has a lower per capita consumption of 7.5 LPA compared with Germany, it has high rates of binge drinking within the EU. There is at least one episode of heavy drinking recorded per month, affecting in some cases over 40 percent of the population.

Your data analyses do not reflect well on the WHO, the addiction researchers at the DHS or the scientists at the German Nutrition Society (DGE). Could it be that so-called “evidence-based” findings are merely speculation?

Dr Maatz: You cannot say it is “mere speculation”. But “evidence-based” does not automatically mean “high quality”. What matters are the “levels of evidence”. None of the studies on alcohol consumption reaches the highest level, which carries the recommendation A [strong recommendation]. Prospective randomised trials – the gold standard of evidence-based research – are not to be found amongst studies on alcohol consumption.

Given such a complex subject and in the face of such skilfully compiled figures, it is hardly surprising that the media has willingly spread the message of the anti-alcohol lobby. Should they have been more conscientious?

Dr Maatz: In my experience, reports from other media outlets are often simply reproduced without verification. For instance, in its 2024 position paper, the German Nutrition Society (DGE) cites a per capita consumption of 12.2 LPA for Germany. Yet this figure dates from 2019.

Your text is packed with charts and tables. There is also an extensive appendix with even more tables and figures. All of this makes it difficult to read. Why did you nevertheless opt for this format?

Dr Maatz: I can understand your criticism to a certain extent. As a doctor who has carried out scientific research, I am bound by a duty of rigour. Figures must be disclosed so that results can be verified. Furthermore, the numerous references in the appendix can save readers a great deal of time when searching for information on specific topics. From my own experience, I can say that it can be quite time-consuming to find very specific data in the WHO and OECD tables.

Your book mentions that a second volume is to follow. What will it be about?

Dr Maatz: During my research, I gathered so much material that I couldn’t fit it all into a single book. That’s why I’ve decided to split it into two. The second volume is subtitled “Die Dosis macht das Gift” (The dose makes the poison). I will focus on studies into the alleged health risks posed by alcohol. The temperance movement’s strategy will also be a central theme. I was actually surprised to discover how extended the temperance organisations’ networks are. I wasn’t aware of that.

Dr Maatz, thank you for speaking to us.

The interview was conducted by Ina Verstl.

*“Alkoholkonsum nüchtern betrachtet – Band 1: Zahlen, Daten und Fakten” by Dr Winfried Maatz is published by BoD and costs EUR 18.50.