Objective: Various guidelines recommend that the risks of diabetes mellitus, dyslipidemia, and lithium intoxication should be appropriately managed by regular monitoring of blood sugar and serum lipid levels in patients treated with atypical antipsychotics, and regular monitoring of serum lithium concentrations in patients treated with lithium carbonate. However, in Japan, these recommendations are not always observed. The present study used a database, constructed from research on medical policies performed in 2006, to investigate the frequencies of blood sugar, serum lipid, and serum lithium monitoring in psychiatric outpatients in Japan.
 Methods: This database contained the health insurance claims of 3,674 psychiatric outpatients extracted from the 47 prefectures throughout Japan. The present study examined two subordinate surveys: 1) the frequency of monitoring blood sugar and serum lipid levels in those on atypical antipsychotics during a period of one month (February) in 2006; 2) the frequency of monitoring the serum lithium concentration in those on lithium carbonate, also obtained during February 2006.
 Results: In Survey 1, the subjects were 228 male and 271 female recipients, with an average age of 45.1 years; 86.8% of these subjects suffered from psychosis. In Survey 2, the subjects were 70 male and 64 female recipients, with an average age of 49.9 years; 57.5% of these subjects suffered from mood disorder and 36.6% suffered from psychosis. In Survey 1, the blood sugar level, HbA1c, and urinary sugar were monitored in 28 (5.6%), 5 (1.0%), and 8 (1.6%) subjects, respectively. At least one of these three tests was performed in 32 subjects (6.4%). The serum lipid level was monitored in 40 subjects (8.0%). In Survey 2, no serum lithium concentrations were measured. The frequency of monitoring the serum lithium concentration per month was estimated to be below 2.2%.
 Conclusions: The Maudsley Prescribing Guidelines recommend that both blood sugar and serum lipid levels should be monitored at least once a year (equivalent to a measurement frequency of 8.3% per month). Survey 1 revealed that, although the actual frequency of monitoring the serum lipid level agreed with the recommendation, that of monitoring the blood sugar level was lower than recommended. However, when compared with the results of a study in the U. S. (equivalent to a measurement frequency of 2-17% per month), the frequency of monitoring the blood sugar level was not that low in Japan. The frequency of monitoring the serum lithium concentration per month was estimated to be below 2.2%. The guidelines and package inserts recommend that the serum lithium concentration should be monitored once every 2-6 months. Therefore, the frequency of monitoring the serum lithium concentration in psychiatric outpatients treated in Japan was considered to be too low from the standpoint of medical safety.
 <Authors' abstract>