Irrespective of the health system, general practitioners/family doctors should care for individuals in the context of their family [1]. Family has a powerful influence on health and illness and family interventions have been shown to improve health outcomes for a variety of health problems [4]. However, there are differences in the physicians' involvement with families. Doherty and Baird have described five levels of physician involvement with families [5]. In Estonia, it has been aimed that a family doctor should work at least at level three, i.e. he or she has to communicate appropriate medical information and advice to family members, to be aware of gross family dysfunctions and to deal with the family members' feelings and concerns related to the condition of the patient [5,6]. It may differ in different countries what is expected and valued in general practice care [7]. What constitutes good medical care is determined culturally within a specific historical and geographic context [8].

First, the FDs were asked whether their patients have registered on the list by families or not and whether they regard it as appropriate that the family should be cared by one doctor, or whether they think that children should have a separate primary care physician. The FDs rated their opinion on the degree of involvement with various problems in the family: they should deal only with the treatment and counselling of a particular patient, or to cooperate also with family members, in addition to treatment of a particular patient, or they should deal also with emotional and relationship problems of the family members. The questions about FDs' awareness of various issues related to their patients' families such as familial diseases and diseases of family members, financial coping, relationships in the family, living conditions (overcrowding), drug addiction, alcohol abuse, smoking and leisure activities were inquired on a three-step scale : yes, in the case of each patient; yes, in certain cases; no, it is not necessary. FDs' were asked to self-estimate their ability to manage problems in families such as substance abuse and relationship problems. Questions about the FDs' ability to manage problems in families were open-ended, for example: how do you assess your possibilities as a FD to influence relationship problems in the family? If you assess that you cannot influence them, please specify why? If you assess you can influence them, please specify how? Also, the FDs were asked to estimate whether their professional training for dealing with the problems of families is adequate or inadequate. Several questions were related to sociodemographic characteristics (sex, age) and professional history (character and size of practices and length of service in primary health care). The questionnaire was piloted for clarity and relevance in a group of five FDs, and minor changes were introduced.

In February 2002, the questionnaire was mailed to 236 FDs. a random sample of family doctors of Estonia. The random sample of FDs was formed by choosing the name of every 3^rd doctor, in alphabetical order, from the list of doctors who had passed residency or retraining courses in family medicine by that time in Estonia (n = 715). A note of reminder and a new questionnaire was sent to the non-responders 4 weeks after the first mailing. Of the 236 mailed questionnaires, 151 were returned after two mailings (64%), five of them were excluded as the respondents did not actually work as FDs. Thus, altogether 146 questionnaires were included in the study, of these 124 were fully completed, while in 22 cases some of the answers (1 to 3 per questionnaire) were missing.

The data were analysed using SPSS for the Windows version 10. The chi-square test was used to test the differences in the proportions, all p-values calculated were two-tailed, the p-values higher than 0.05 were considered non-significant (NS).

Our study revealed that care of patients in the context of the family is an important issue for FDs and that Estonian family doctors have good possibilities to take care of the whole family. Although all patients have the right to choose an individual family doctor, the FDs who responded to the questionnaire were sure that family members mostly have one and the same family doctor. This is concordant with the results of a recent survey among patients according to which 74% of the respondents reported that they had one and the same family physician for the whole family [15]. In Estonia, similar with the other Eastern Europe countries shift from separate pediatric and adult primary care system to family doctors system occurred in the 1990s [16]. Only 17% of the FDs in our study were of the opinion that children and adults should have different primary care doctors.

In our study, altogether 90% of the family doctors thought that they should communicate and cooperate with family members in management of the health problems of patients. It is a good result considering that family medicine is a new and developing speciality in Estonia. At the same time, most family doctors are not yet ready to deal with the family members' emotional and relationship problems. Primary care is an important early intervention site of most serious relationship issues, domestic violence, etc [17]. However, studies conducted in other communities have identified that physicians need more continuing education concerning these topics [18,19]. Currently, a two-day course on domestic violence and child abuse, which are serious problems also in Estonia [20], is included in the postgraduate residence training curriculum.

Concerning the family doctors' attitudes to the importance of awareness of their patients' family-related issues, the results indicated that the awareness of drug addiction, diseases of family members, living conditions of the family and alcohol problems in the family were considered the most essential. The awareness of the patients' relationship problems, economic problems and leisure activities was not so highly valued. A recent survey among patients in Estonia revealed that they were also more disposed to involve the family physician in such problems as harmful habits and diseases in the family, but they were less willing to share the relationship problems [15]. This can reflect the current situation in Estonia where the number of drug users as well as alcohol users has significantly increased during the last five years [21]. Lately, much attention has been paid to the problem by politicians, doctors and the mass media.

The low willingness to be aware of the patients' relationship problems is partly related to insufficient preparedness in this field, as the doctors who considered themselves to be adequately prepared to tackle relationship issues were also more often willing to do this. From another point of view, family issues, especially relationships and economic situation, are always delicate topics and require consideration of the patient's attitude to corresponding activities. It has been shown that patients vary considerably in their preferences for physician inquiries into such problems as social functioning, psychosocial issues and health risks. Also, it may reflect cultural differences: in the Nordic countries biomedical talk is more common, while in the southern regions psychosocial dialogue is prevalent [22].