OBJECTIVES: If a patient shows an infection, the physician has to decide which kind of anti-infective substance has to be prescribed. One method to figure this out is to order  laboratory cultures. We want to find out whether there is a correlation between the frequencies of cultures reimbursed and the number of prescriptions of anti-infective agents. METHODS: Claims data (2006) from physicians in free practice and institutes for laboratory medicine data for laboratory cultures were conducted out of different reimbursement catalogs in Austria. RESULTS: The rate of reimbursed cultures per prescription was 11% (for bacterial cultures and antibacterial medication J01, J04), 63% (for mycotic cultures and antimycotic medication J02), 24% for viral cultures and antiviral medication J05) and 25% (for parasitic cultures and antiparasitical medication P01,02,03). If just those infections verified by cultures had caused a prescription, the costs for anti-infective agents would have been 17% of the current number. If all prescriptions for anti-infective agents had been based on a culture, the prescription costs would have been 7,5 times higher. There are price differences for cultures due to various contract partners. If the lowest fee had been paid for each test we would have saved 33% of the current turnover for bacterial, 37% for mycotic, 38% for viral, and 6% for parasitic infections. If the highest price had been paid we would have paid 75% (bacterial), 46% (mycotic), 164% (viral) and 33% (parasitic) more than the current turnover. CONCLUSIONS: Our further research will focus on the different categories of prescription for anti-infective agents, testing and also the guideline conformity of its use.