There are relatively large regional differences in Sweden in the proportion of newborns receiving antibiotics for suspected sepsis, according to a study at the University of Gothenburg. The researchers want to call attention to overuse as well as highlight good examples.
When newborns receive antibiotics, it is almost always for suspected sepsis. Despite Sweden’s low antibiotic use compared to other countries, last year researchers found unjustifiably high usage levels in newborns. The gradual decline in sepsis prevalence over time has not affected antibiotic use.
This new study provides an overview of antibiotic use in six major regions in Sweden. The dataset also includes the corresponding data from individual hospitals with neonatal units where newborns are treated, a total of 37 hospitals.
The study is based on data from more than one million infants born in Sweden between 2012 and 2020. The infants encompassed were born at 34 weeks’ gestation or later. The results are presented in the journal Archives of Disease in Childhood: Fetal & Neonatal Edition.
Considerable differences
The study presents the proportion of newborns receiving antibiotics during the first week of life, an internationally dominant metric. In the western world, where sepsis rates are low, the goal is to treat no more than one percent of infants while maintaining low sepsis prevalence and mortality rates.
In the study, western Sweden ranked highest in antibiotic use, followed by northern Sweden, south-eastern Sweden, southern Sweden, central Sweden, and eastern Sweden. The figures for western Sweden and eastern Sweden were 3.0 percent and 1.3 percent, respectively. Sepsis prevalence in all Swedish regions was below one per thousand live births.
At hospital level, the lowest antibiotic use of newborns during the first week of life was 0.9 percent, and the highest 4.3 percent.
Greater awareness key
One of the driving forces behind the study is Johan Gyllensvärd, a PhD student in pediatrics at Sahlgrenska Academy at the University of Gothenburg and a practicing pediatrician at Ryhov County Hospital in Jönköping.
“The threshold for antibiotic intervention varies, explained in part by prevalence levels, but it’s also down to local practices. It’s very much a matter of the policies and traditions seen at the different hospitals,” he says.
Sepsis is a potentially life-threatening condition caused by the immune system’s reaction to an infection. Treatment must be started immediately. At the same time, antibiotics disrupt the bacterial flora, which is sensitive in newborns. Overuse also increases the risk of becoming a carrier of antibiotic-resistant bacteria.
“Antibiotics are essential in the treatment of sepsis in newborns, and if you have a sicker population, then it’s reasonable for antibiotic use to be higher. However, we need to increase awareness of the overuse of antibiotics and improve at distinguishing infected and uninfected infants. This study helps highlight good examples and demonstrates that it’s possible to have a treatment rate of one percent or less of all newborns while maintaining low sepsis and mortality rates,” Johan Gyllensvärd explains.

Fact box:

Proportion of newborns receiving antibiotics during the first week of life

Major regions

Southern Sweden: 1.7 percent (Ystad, Halmstad, Helsingborg, Karlskrona, Kristianstad, Malmö, Växjö, Lund)

South-eastern Sweden: 2.1 percent (Västervik, Jönköping, Kalmar, Norrköping, Linköping)

Eastern Sweden: 1.3 percent (Visby, Danderyd, Södersjukhuset, Huddinge, Solna)

Western Sweden: 3.0 percent (Borås, Trollhättan, Skövde, Gothenburg)

Central Sweden: 1.6 percent (Hudiksvall, Eskilstuna, Falun, Gävle, Karlstad, Västerås, Örebro, Uppsala)

Northern Sweden: 2.2 percent (Gällivare, Skellefteå, Örnsköldsvik, Luleå, Sundsvall, Östersund, Umeå)

Hospitals

The five hospitals with the lowest proportions: Västerås 0.9 percent, Örebro 1.0 percent, Södersjukhuset 1.0 percent, Visby 1.1 percent, and Norrköping 1.2 percent.

The five hospitals with the highest proportions: Luleå, 2.6 percent, Umeå 2.6 percent, Skövde 2.9 percent, Borås 3.7 percent, and Trollhättan 4.3 percent.

Sources: University of Gothenburg, Swedish Medical Birth Register, Swedish Neonatal Quality Register