Materials and Methods Within the WHO program ‘MONICA’ psychosocial factors were examined in a representative sample of men aged 25–64 years old at 3 time points (1984, 1988 and 1994). Total sample was 2149 persons. The period of study was 10–20 years. Incident cases of AH were counted and Spilberger's test for estimation of personal anxiety (PA) was used. Cox–proportional regression model was used for an estimation of RR.

Results During 20 years of investigation the incidence of AH was 16.9%. High levels of personal anxiety (HLA) were present in 57.4% of the group with incident AH. Incident AH with HLA was significantly higher in several subgroups: single, divorced and widowed men; incomplete secondary and primary education; men undertaking hard physical work. The RR of AH development during the first 5 years of observation in the HLA group compared to average level of anxiety (ALA) was 6.8 times higher (95% CI 3.2 to 14.2; p<0.05); in 10 years the risk was 5.0 times higher (95% CI 2.9 to 11.8 p<0.05) higher, and in 20 years it was 1.8 times higher (95% CI 1.1 to 3.2 p<0.05).

Conclusion Personal anxiety should be considered in the development of AH.