(before/after) trial based on the frequency of etiological diagnoses of LBP, the testing of cli-
nical knowledge with multiple choice tests, and on priorities given in confronting problems
of treatment of LBP by the three GP's groups that had voluntarily adhered to the study.
RESULTS: A statistically significant increase of diagnostic capacity (the frequency of exact
responses in the multiple choice test) was achieved. There was also a shifting of priorities in
confronting problems in the treatment of LBP, from the diagnosis as a primary issue before
intervening, to communication with the patient. CONCLUSIONS: Within the Italian context,
the study also confirms the efficiency of a multifaceted intervention based on a training
process aimed at modifying the clinical behaviour of the GP with respect to LBP.
Cricelli C. Exploring patient- and doctor-related variables associated with antibiotic pre-
scribing for respiratory infections in primary care. Eur J Clin Pharmacol. 2003;59:651-7.
prescribe antibiotics or parenteral antibiotics for acute respiratory infections (ARIs) and to de-
scribe the variability as well as the appropriateness of antibiotic use and its predictive factors
in general practice. METHODS: We conducted a cross-sectional study among patients aged
15-85 years with a diagnosis of ARIs, using information from 469 GPs from the Health Se-
arch Database. Diagnoses were linked with antibiotic prescriptions and other patients and
doctor-related variables. Available scientific evidence was used to establish the appropriate-
ness of first-choice antibiotic treatment. Frequency analyses and logistic regressions were used
to identify variables associated with antibiotic use and appropriateness. RESULTS: On 67,761
cases of ARIs, antibiotics were prescribed in 63.2%, varying from 80.9% for acute bronchitis
to 43.9% for croup, influenza and common cold. Significant associations with antibiotic use
were found for geographic location and number of patients under care. The use of diagnostic
tests significantly lowered the risk. Geographic location and living in an urban area were asso-
ciated with parenteral antibiotic use. Amoxicillin (16.7%) and amoxicillin-clavulanate (17.9%)
were the most common antibiotics prescribed. Appropriateness was reported in 39.0% of ca-
ses, with geographic location, physician's gender and diagnostic tests being the factors more
predictive of appropriate antibiotic use. CONCLUSIONS: There is still excessive antibiotic use
for ARIs. Its overuse is influenced by the physicians' characteristics and by the environment in
which they practice, whilst diagnostic tests might reduce inappropriateness. Therefore, effec-
tive strategies for changing diagnostic and therapeutic behaviour are needed.
Prevalence estimates for chronic diseases in Italy: exploring the differences betwe-
en self-report and primary care databases. J Public Health Med. 2003;25:254-7.
care morbidity and to examine how the differences vary across the diseases and are influen-
ced by patients' demographic characteristics. METHODS: A comparison of the prevalence
of four chronic conditions for 432747 patients from the Health Search Database (HSD) and
119799 individuals from a Health Interview Survey was carried out. A linear regression was
performed to study the associations between age and difference in morbidity. RESULTS: Si-
milar prevalence was found for diabetes and hypertension, whereas for chronic obstructive
pulmonary disease (COPD) and gastroduodenal ulcer lower HSD prevalence was reported.
Among females, age was always associated with morbidity difference. Among males, signi-
ficant associations were found only for COPD (R2 = 0.81; p = 0.001) and gastroduodenal
ulcer (R2 = 0.93; p < 0.001). CONCLUSIONS: The difference between population and pri-
mary care morbidity is affected by disease under investigation and patients' demographic
characteristics. Therefore, in choosing the more cost-effective approach to collect data such
evidence should be taken into account, and results should be interpreted with great caution.