Elsevier

Social Science & Medicine

Volume 66, Issue 5, March 2008, Pages 1046-1056
Social Science & Medicine

Socioeconomic context, distance to primary care and detection of hepatitis C: A French population-based study

https://doi.org/10.1016/j.socscimed.2007.11.044Get rights and content

Abstract

Previous research suggests that hepatitis C detection is lower in rural than in urban areas and depends on geographic accessibility to care. To what extent differences in socioeconomic context could explain these disparities remains unknown. This study assessed the respective influence of the socioeconomic context and of the distance to primary care on the hepatitis C detection rates in a well-defined population of 1,005,817 inhabitants covered by a universal health insurance system in the Côte d'Or and the Doubs areas of France. The 1938 cases of hepatitis C newly diagnosed in the survey area between 1994 and 2001 were included. The patients' ‘cantons’ of residence were classified into six socioeconomic clusters according to 19 quantitative indicators. Age and sex-standardized hepatitis C detection rate ratios (DRR) for the six clusters were estimated and a multilevel Poisson model investigated whether detection rates varied across clusters after adjusting for distance to the nearest general practitioner. Standardized DRR (95% CI) were higher in both “lower-income urban” and “upper-income urban” clusters and lower in the following clusters: “outer suburb”, “industrial rural” and economically “dynamic rural”. Adjusting for distance to general practitioner reduced the rate difference between socioeconomic clusters: for a distance to general practitioner equalling 0 km (practice in the municipality) and compared to the “upper-income urban” cluster, only the “lower-income urban” cluster had higher DRR. In the six clusters, detection rates decreased as the distance to general practitioner increased (DRR 0.89 95% CI 0.84–0.96 for a 1-km change) but the decrease was more marked in both urban clusters. In this population with good primary care affordability, geographic proximity to general practitioner, rather than socioeconomic context of neighborhood, appeared to be the main factor limiting hepatitis C detection.

Introduction

Hepatitis C virus (HCV) infection is a worldwide cause of life-threatening liver disease, which poses challenging public health problems both in developed and developing countries (Perz & Alter, 2006). In most infected people, the infection becomes chronic and remains asymptomatic for many years, even though liver damage worsens. Identifying infected persons and offering medical evaluation and treatment is one of the main objectives of health programs that have been implemented in several countries to tackle the expected hepatitis C-related health burden (CDC, 1998, Department of Health, 2004, Viral Hepatitis Prevention Board, 2005). These programs plan to promote HCV infection screening for persons at high risk in primary care settings. Indeed, general practitioners are in a good position to identify asymptomatic people belonging to high risk groups. Collecting data on cases of HCV infection newly detected in a well-defined area enables program stakeholders to monitor the screening activity and its possible temporal and spatial variations.

Since 1994, such a reporting system has been implemented in two French ‘départements’, made up of 74 ‘cantons’ (administrative territorial divisions) and totalling 1,005,817 inhabitants. This database enables us to estimate population-based detection rates of hepatitis C and to search for rate variation within the ‘départements’, according to the place of residence, its environmental and social context and the availability of health care facilities. Indeed, within several countries, health care utilization depends on both individual and contextual characteristics and, in particular, varies between urban and rural areas (Arcury et al., 2005, Casey et al., 2001, Chaix et al., 2005a, Chaix et al., 2005b, Coughlin et al., 2002). In the United States, several studies revealed that people living in rural areas were less likely than urban residents to utilize regular check-up, cancer screening and other recommended preventive services (Arcury et al., 2005, Casey et al., 2001, Coughlin et al., 2002).

As stated by Penchansky and Thomas (1981), access to health services is a general concept which summarizes five specific dimensions describing the “fit” between the patient and the health care system: availability, accessibility, accommodation, affordability and acceptability. In the literature, the first two dimensions are often considered simultaneously as both contribute to “spatial accessibility”, which can be measured in particular by distance to health care provider (Gugliardo, 2004, Ricketts et al., 1994). Since the mid-1990s, ensuring the availability of hepatitis C screening services that offer comprehensive care for infected persons has been a priority of French health authorities (Viral Hepatitis Prevention Board, 2005). Regional networks have been created, made up of general practitioners in charge of screening and of specialists who prescribe antiviral treatments. However, spatial accessibility deserves special attention in the French health system, since large geographic variations in physician-to-population ratios exist in France, resulting in a shortage of primary health care providers and specialists in rural areas (Lucas Gabrielli & Tonnellier, 2001). By contrast, the lack of affordability may not be considered as a notable barrier to being screened for HCV infection by a general practitioner, since in 1998 the French National Health Insurance System covered 99.8% of the population, providing good ability to pay for primary care (Bocognano et al., 1999).

In previous work, we found that detection rates of hepatitis C were lower in rural than in urban areas and decreased as the distance to primary care increased, the urban–rural difference becoming insignificant after adjusting for distance to primary care provider (Monnet et al., 2006). Considering the strong effect of distance to general practitioner on detection rates and the low plausibility that it could influence the prevalence of the disease, the results of this first study raised the important issue of possible inequalities in access to HCV screening related to the place of residence.

However, the role of the difference of socioeconomic and cultural context in the urban–rural detection gap needed to be evaluated. Following the numerous studies showing that the context of neighbourhood strongly influenced health outcomes (Pickett & Pearl, 2001), several recent works have shown that contextual characteristics could also modify health care access (Chaix et al., 2005a, Litaker et al., 2005). Indeed, the socioeconomic and cultural levels of the context may particularly affect the last dimension of access pointed out by Penchansky and Thomas (1981), acceptability, inasmuch as these factors could influence beliefs and expectations in the population about the need and relevance of hepatitis C screening. Recent research conducted in France on representative national samples has shown that health care utilization not only depended on individual characteristics but also on the socioeconomic context of the area of residence as well as on its supply of physicians (Chaix et al., 2005a, Chaix et al., 2005b). In fact, important contrasts exist between urban and rural areas, concerning health care resources, socioeconomic and cultural factors, as well as differences in premature and avoidable mortality (Trugeon, Fontaine, & Lémery, 2006). Therefore, this study examined (1) whether or not hepatitis C detection rates also varied according to the socioeconomic context of the area of residence and (2) if the decrease in hepatitis C detection rate related to reduced spatial accessibility to primary care remained significant after adjusting for the differences in socioeconomic and cultural context of the area of residence.

Section snippets

Setting

The two ‘départements’ included in the study were the Côte d'Or and the Doubs, with respectively 506,755 and 499,062 inhabitants (1999 census). Their geographic characteristics and their resources of health professionals have been described elsewhere (Monnet et al., 2006). In 2000, the Côte d'Or ranked beyond the 90th percentile for the Gross Domestic Product per capita among French ‘départements’, with an economy based on service industries, food-processing and agriculture. The Doubs

Variation of detection rates across socioeconomic clusters

Firstly, we calculated age and sex-standardized detection ratios across clusters with their 95% confidence intervals (CI). The ratio numerator was the number of cases newly detected over the period in each cluster; the denominator corresponded to the expected number of cases in the cluster, calculated from the age–sex specific detection rates in both ‘départements’ (reference rates) and age–sex distribution in the cluster population.

Then we used a multilevel Poisson model (Goldstein, 2003) to

Variation of detection rates across socioeconomic clusters

Between January 1994 and December 2001, 1938 newly detected cases of hepatitis C (1150 men, 788 women) were recorded, giving age and sex-standardized mean yearly rates of 26.6 for 105 inhabitants, 95% confidence interval (95% CI) (25.0–28.2) (men) and 17.3 for 105 inhabitants, 95% CI (16.0–18.6) (women). Compared to the mean value of the whole region, age and sex-standardized detection ratios (95% CI) were higher in both “lower-income urban canton” and “upper-income urban canton” clusters,

Discussion

In this French area of more than one million inhabitants, hepatitis C detection rates varied according to the socioeconomic context of the area of residence. Age and sex-standardized detection ratios were higher in both metropolitan “lower-income” and “upper-income” areas and lower in the outer suburbs and in both non-remote rural areas. This detection gap was markedly reduced after adjusting for distance to GP, confirming the assumption that disparities in spatial accessibility to primary care

Acknowledgements

We are grateful to Catherine Quantin, from the Department of Biostatistics of the University Hospital of Dijon, for her valuable suggestions. We thank the doctors and medical analysis laboratories which took an active part in reporting cases to the registry and Patrick Hillon for his scientific support. We are also indebted to the regional union of medical insurance centers (Bourgogne), to Pamela Albert, Estelle Collin-Naudet and Philippe Evrard for their technical assistance.

References (37)

  • CDC

    Recommendations for prevention and control of hepatitis C virus (HCV) infection and HCV-related chronic disease. Centers for Disease Control and Prevention

    MMWR Recommendations and Reports

    (1998)
  • Centre for Multilevel Modelling

    MLwiN 2.0

    (2005)
  • B. Chaix et al.

    Comparison of a spatial approach with the multilevel approach for investigating place effects on health: the example of healthcare utilisation in France

    Journal of Epidemiology and Community Health

    (2005)
  • B. Chaix et al.

    Access to general practitioner services: the disabled elderly lag behind in underserved areas

    European Journal of Public Health

    (2005)
  • A. Couffinhal et al.

    Part I: Determinants of social inequalities in health and the role of health care system

    Policies for reducing inequalities in health, what role can the health system play? A European perspective

    (2005)
  • S. Coughlin et al.

    Breast and cervical carcinoma screening practices among women in rural and nonrural areas of the United States

    Cancer

    (2002)
  • Department of Health

    Hepatitis C action plan for England

    (2004)
  • A. Diez-Roux

    Multilevel analysis in public health research

    Annual Review of Public Health

    (2000)
  • Cited by (0)

    This project was supported by a research grant (04029) from the French National Agency for AIDS Research.

    View full text