General practitioners who never perform Pap smear: the medical offer and the socio-economic context around their office could limit their involvement in cervical cancer screening.

BACKGROUND: In France, with the growing scarcity of gynecologists and a globally low and socially differentiated coverage of cervical cancer screening (CCS), general practitioners (GPs) are valuable resources to improve screening services for women. Still all GPs do not perform Pap smears. In order to promote this screening among GPs, the characteristics of physicians who never perform CCS should be more precisely specified. Besides already-known individual characteristics, the contextual aspects of the physicians' office, such as gynecologist density in the area, could shape GPs gynecological activities. METHODS: To analyze county (département) characteristics of GPs' office associated with no performance of CCS, we used a representative sample of 1063 French GPs conducted in 2009 and we constructed mixed models with two levels, GP and county. RESULTS: Almost 35% (n = 369) of the GPs declared never performing CCS. GPs working in counties with a poor GP-density per inhabitants were more likely to perform CCS (odds ratio (OR) = 0.52 for each increase of density by 1 GP per 10,000 inhabitants, 95% confidence interval (CI) = 0.37-0.74). On the contrary, GPs working in counties with an easier access to a gynecologist were more likely not to perform CCS (OR = 1.06 for each increase of density by 1 gynecologist per 100,000 women, 95%CI = 1.03-1.10 and OR = 2.02 if the first gynecologist is reachable in less than 15 min, 95%CI = 1.20-3.41) as well as GPs working in areas with a poverty rate above the national average (OR = 1.66, 95%CI = 1.09-2.54). These contextual characteristics explain most of the differences between counties concerning rates of not performing CCS. CONCLUSIONS: Specific programs should be developed for GPs working in contexts unfavorable to their involvement in CCS.

Contextual determinants of participation in cervical cancer screening in France, 2010.

BACKGROUND: Some contextual factors associated with participation in cervical cancer screening are reported in the literature, but few studies have examined their combined effect. Our objective was to assess the role of contextual characteristics, separately and in combination, in participation in cervical cancer screening in France. METHODS: Marginal Poisson regression models - taking into account the correlation between women in a given commune - were conducted using data from the Baromètre Santé 2010 survey. The characteristics of the commune of residence of the women studied were the potential spatial accessibility to general practitioners (GP) and gynecologists, the agglomeration category, and the socioeconomic level. RESULTS: The analyses were performed in 3380 women, 88.2% of whom were up to date with their cervical cancer screening. Once the individual characteristics were taken into account, the screening participation rate was similar in all the communes, with the exception of those with poor access to a gynecologist and good access to a GP, where the rate was 6% lower (95%CI: 0.5-11%) than in the communes with good access to both GP and gynecologist. The same association with accessibility was observed in small agglomerations. Compared to women living in the more advantaged communes, the screening participation rate was 8% (2-12%) lower in those living in the more disadvantaged ones, except when accessibility to both types of physician was high. DISCUSSION: We observed an association between potential spatial accessibility to care in women's residential communities and their cervical cancer screening practices, in particular in small agglomerations, rural communes, and more disadvantaged communes.

Disengagement of general practitioners in cervical cancer screening.

In the absence of organized cervical cancer screening (CCS) programs, gynecologists remain principal actors in obtaining a Pap smear, followed by general practitioners (GPs). In France, with the growing scarcity of gynecologists and social inequalities in access to opportunistic screening, GPs are valuable resources for women's gynecologic follow-up. We aimed to investigate the characteristics of GPs who do not perform CCS, analyzing the effect of GPs' sex and their evolution over time. On the basis of data from three cross-sectional surveys conducted among representative samples of French GPs in 1998, 2002, and 2009 (n=5199), we constructed univariate and multivariate logistic mixed models (level 2: county, level 1: GP) with random intercept stratified on GPs' sex to investigate the characteristics of the GPs associated with no practice of CCS ever. Almost one-third of all GPs did not perform CCS ever and it increased with time. Male GPs were always more likely not to perform it (odds ratio=0.50, 95% confidence interval=0.42-0.59). The percentage of GPs not performing CCS increased more markedly among male than among female GPs, and increased more among the youngest age group. Increasingly fewer GPs engage in CCS when the growing scarcity of medical gynecologists calls for more participation. Female GPs remain significantly more active in CCS than male GPs. The participation in CCS is determined differently according to the practitioner's sex.

Gender, socio-economic status, migration origin and neighbourhood of residence are barriers to HIV testing in the Paris metropolitan area.

In France, numerous HIV patients still discover their HIV status as a result of AIDS-related symptoms. We investigated factors related to the absence of any HIV testing in men and women separately, using the data from the SIRS cohort, which includes 3023 households representative of the Paris metropolitan area in 2005. The failure to use HIV testing services was studied in relation to individual socio-economic and demographic factors as well as some psychosocial characteristics. The effect of the characteristics of the residential neighbourhood was also analysed using multilevel models. In multivariate analysis, the factors associated with no history of HIV testing in women were an age >44 years, the absence of any pregnancy during the previous 15 years, a low education level, unemployment, to have had no or only one steady relationship in one's lifetime, to have a religious affiliation and to live in a poor neighbourhood. In men, factors were age <30 or >44 years, to have had no or only one steady relationship during one's lifetime, to have a religious affiliation and to perceive oneself as being at low risk of HIV infection. An association according to the "migration origin" was observed among men: foreigners and French men born to (at least) one foreign parent were more likely not to have been tested than French men born to two French parents. We conclude that gender, social and territorial differences exist in HIV testing among people living in the Paris area. More systematic proposals of HIV test in primary care would be an effective policy to overcome these persistent social stratifications.

The combined effects of activity space and neighbourhood of residence on participation in preventive health-care activities: The case of cervical screening in the Paris metropolitan area (France).

Estimates from multilevel regression of 1768 women living in the Paris metropolitan area showed that women who reported concentrating their daily activities in their perceived neighbourhood of residence had a statistically greater likelihood of not having undergone cervical screening during the previous 2 years. Furthermore, the characteristics of the administrative neighbourhood of residence (such as the practitioner density or the proportion of residents with a recent preventive consultation) had a statistically greater impact in terms of delayed cervical screening on women who concentrated the vast majority of their daily activities within their perceived neighbourhood of residence than among those who did not. The residential environment might promote or damage, to a greater extent, the health behaviour of people whose daily activities are concentrated within their perceived neighbourhood, since we can assume that their exposure to their neighbourhood characteristics is stronger. It could thus be useful to study more often the combined effects of activity space and neighbourhood of residence on participation in preventive health-care activities.