To the Editor: Compliance with published clinical practice guidelines (CPG) has been reported to be a problem worldwide. A recent study by Wolfe et al1 concluded that most American family physicians find CPGs to be helpful, and familiarity with CPGs is fairly uniform across most subgroups studied.
We recently performed a study in Estonia on doctors‘ compliance with a CPG for type 2 diabetes mellitus (DM).2, 3 We studied 354 family doctors, a random sample from the list of all family doctors in Estonia: 163 responded (response rate 46%). Respondents reported the guidelines were available in 76% of cases, which is about 3 times higher than reported by Wolfe et al.1 Nevertheless, our results on how doctors change their practices when CPGs are available at their offices are not as encouraging those reported by Wolfe et al.1 In the that study, it was found that 28% to 33% of the respondents reported they had changed their practice as a result of CPGs. The study also showed that of the subgroup of doctors who were aware of CPGs, 81% to 91% changed patient care. In our study, performed 3 years after the guideline was approved and distributed, we found that there was a great variability in doctors’ self-reported performance, depending on the guideline recommendation, and the results are far from what we expected (Table 1).
We found no relationship between the use of a CPG for DM and the location of practice (rural or urban) or whether the practice was solo or group. In our study, the more experienced doctors reported better availability and better use of the DM CPG compared with their younger colleagues, which is the opposite of the results in the study of American doctors.
It has been reported in many studies that the attitudes of family physicians toward CPGs depends on varied circumstances, including cultural and legal factors. Lack of compliance with the guidelines may indicate deficiencies in the physician’s knowledge, implementation problems, lack of belief in the guidelines, or problems in patient compliance.4 Another possible explanation is that doctors involved in surveys might have overestimated changes in their practice. It would be useful to study the patients to detect the actual change in practice.
