Drug-Related Deaths in Germany

In Germany, the number of drug-related deaths rose to 2227 cases in 2023, the highest level since recording started in 1973 (1). In proportion to population, this amounts to 2.64 cases per 100 000 persons—the second-highest figure since 1991 (2.66). Despite this trend, our knowledge of the characteristics of persons who died as a result of drugs is limited. In our study, we analyzed how often the various substances or substance classes were involved in drug-related deaths in 2023, also stratified by the availability or unavailability of a toxicology report (TR).

Methods

Our study is based on the database maintained by the German Federal Criminal Police Office (Bundeskriminalamt, BKA) comprising all deaths that came to the attention of the State Criminal Investigation Offices (Landeskriminalämter, LKAs) in 2023 (2). The database holds information on the age, sex and federal state of the deceased, as well as the substances involved (18 substances [substance classes] and the category “other”). In addition, the following information is noted: long-term harm to health, suicide, accident while under the influence of drugs, and availability of a TR. The BKA data were slightly adjusted for our analysis. Information from open text fields on substances involved were systematically integrated.

In the statistic, a drug-related death is assumed if at least one of the following criteria is met (3):

• Death as the result of unintended overdosing

• Death as the result of damage to health caused by prolonged drug abuse (“long-term harm“)

• Suicide out of despair over living circumstances or under the influence of withdrawal symptoms

• Fatal accident of a person under the influence of drugs.

In this analysis, we classified a death where only one substance was stated as a monovalent drug-related death, while a death where more than one substance was stated (also within one substance class as well as in connection with cannabis, alcohol or one of the “other“ substances) was classified as a polyvalent death. If no substance (class) was mentioned, substance involvement was classified as “unknown“ (10.1 % of cases; in 91.6% of these cases, no TR was available).

We calculated descriptive statistics for demographic characteristics, substance involvement as a percentage for the 19 substances and the three created substance categories overall and stratified by availability of a TR.

Results

In the report year 2023, there were 2227 drug-related deaths (1844 male [82.8%], 383 female [17.2%]). The median age was 41.0 years (range: 0–74 years), with no difference between the two sexes (p = 0.86). In 39.6 % of cases, results from a TR were available; however, large regional disparities were noted in this respect. Table 1 shows the proportions of monovalent and polyvalent cases overall and also stratified by whether or not a TR was available. Noticeably, the number of substances was 50% higher when a TR was available (p<0.0001). Table 2 lists the substances involved overall and by availability of a TR. If a TR was available, both opioids and non-opioids were more frequently detected compared to cases with no TR.

Table 1. The descriptive statistics on drug-related deaths in 2023.

Total (N = 2227)	No toxicological report (n = 1345)	Toxicological report (n = 882)
Age (mean, SD; in years)
41.0 (SD = 12.3)	42.4 (SD = 12.0)	38.9 (SD = 12.3)
Mean number of substances detected
2.4	2.0	3.0
Monovalent cases
522 (23.4%)	340 (25.3%)	182 (20.6%)
Polyvalent cases
1479 (66.4%)	798 (59.3%)	681 (77.2%)
Unknown (no substance stated)
226 (10.1%)	207 (15.4%)	19 (2.2%)
Presence of long-term harm
647 (29.1%)	514 (38.2%)	133 (15.1%)

SD, standard deviation

Table 2. Substances or substances classes detected, stratified by the unavailability or availability of a toxicological report^*1.

Substance/substance class*2	Total (N = 2227)		No toxicological report (n = 345)		Toxicological report (n = 882)
	%	[95% CI]	%	[95% CI]	%	[95% CI]
• At least one opiate/opioid involved	56.1	[54.0; 58.1]	53.3	[50.6; 56.0]	60.3	[57.1; 63.6]
– Heroine/morphine	32.0	[30.0; 33.9]	31.6	[29.1; 34.1]	32.5	[29.4; 35.6]
– At least one substitution agent	29.4	[27.4;31.2]	27.8	[25.4; 30.2]	31.7	[28.7; 34.8]
– including: methadone	26.6	[24.8; 28.5]	25.8	[23.5; 28.1]	27.9	[24.9; 30.9]
– Other opiate-/opioid-based drugs	8.3	[7.1; 9.4]	6.5	[5.2; 7.9]	10.9	[8.8; 12.9]
• At least one non-opioid involved	76.0	[74.2; 77.8]	68.6	[66.1; 71.0]	87.4	[85.2; 89.6]
– Cocaine/crack	27.4	[25.5; 29.2]	21.7	[19.5; 23.9]	36.1	[32.9; 39.2]
– Amphetamine	18.1	[16.5; 19.6]	14.3	[12.5; 16.2]	23.7	[20.9; 26.5]
– Methamphetamine	5.5	[4.5; 6.4]	4.1	[3.0; 5.1]	7.6	[5.8; 9.3]
– Amphetamine derivatives	5.2	[4.2; 6.1]	3.6	[2.6; 4.6]	7.6	[5.8; 9.3]
– At least one NPS*3	4.0	[3.2; 4.9]	2.4	[1.6; 3.2]	6.6	[4.9; 8.2]
– Psychoactive substances	25.6	[23.8; 27.5]	16.8	[14.8; 18.8]	39.1	[35.9; 42.3]
– Alcohol	27.9	[26.0; 29.7]	33.0	[30.5; 35.5]	20.1	[17.4; 22.7]
– Cannabis	21.4	[19.7; 23.1]	22.7	[20.4; 24.9]	19.4	[16.8; 22.0]
– Other substances	10.0	[8.7; 11.2]	4.3	[3.2; 5.4]	18.6	[16.0; 21.2]

^*1 Multiple entries of substances were possible.

^*2 Substances with prevalence rates of less than 4% are not shown for reasons of space. This applies to buprenorphine, other opioid-based replacement treatments, fentanyl, synthetic opioids, synthetic cannabinoids, ketamine, other new psychoactive substances (NPS), as well as GHB/GBL

^*3 New psychoactive substance (NPS); this term covers synthetic opioids, synthetic cannabinoids, ketamine, and other NPS.

GHB,gamma-hydroxybutyric acid;

GBL, gamma-butyrolactone

Discussion

In 2023, more drug-related deaths associated with the consumption of illegal substances were recorded in Germany than ever before. This study is the first to use the BKA’s statistic of drug-related deaths for scientific analyses and in doing so makes an empirical contribution to a topic that is the subject of considerable public debate, but about which little is actually known empirically. An older validation study for this statistic showed a very high degree of congruence with the National Cause of Death Registry (4). Further research utilizing this data source is important for discussions surrounding rising drug-related death rates, as these insights could be used to develop preventive measures and interventions to reduce deaths related to the consumption of illegal substances in the future.

A higher rate of toxicological reports is urgently needed to inform the further debate. Currently, cost coverage is a common problem. TR findings are available in less than 40% of cases. The comparison between cases with and without a TR highlights the immense importance of TRs: If a TR is available, 50% more substances are detected and a death is more frequently classified as polyvalent. In addition, some substances and substance classes show a marked increase in prevalence when a TR is available (in some cases more than twice as high). To some extent, these differences may be due to the fact that TRs are more likely to be conducted for deaths of younger persons; however, the age difference between cases with and without a TR is small. Future studies should control for various factors. However, since this requires a larger number of cases, it is hardly possible to carry out these analyses with data from one reporting year only.

The statistic of drug-related deaths has some limitations:

• It only includes deaths known to the police and thus underestimates the number of drug-related deaths (4). Unlike in the case of tobacco and alcohol (5), there is no systematic estimate for drug-related mortality which would, for example, take secondary diseases into account.

• Given the frequent unavailability of a TR, it is often not possible to make valid statements on the substances involved.

• Generally, the registry does not assess whether death was actually caused by the individual substances, but merely documents their presence. Even if a TR is available, it rarely documents individual substances as cause of death by way of annotations.