Actual and preferred place of death of cancer patients. Results from the Italian survey of the dying of cancer (ISDOC)

OBJECTIVE: To describe actual and preferred place of death of Italian cancer patients and to analyse the preferences met regarding the place of death. DESIGN: Mortality follow back survey of 2000 cancer deaths, identified with a two stage probability sample representative of the whole country. Information on patients' experience was gathered from the non-professional caregiver with an interview. A section of the interview covered information on the actual and preferred place of death of the patients. SETTING: 30 Italian local health districts randomly selected after stratification in four geographical areas. PARTICIPANTS: 1900 of 2000 (95.0%) caregivers of cancer deaths identified. MAIN OUTCOME MEASURES: Prevalence of actual and preferred places of death. RESULTS: Valid interviews were obtained for 66.9% (n = 1271) of the caregivers. Place of death was home for 57.9% of Italian cancer patients, hospital for 34.6%, hospice for 0.7%, nursing home for 6.5%, and ambulance for 0.4%. Wide and significant differences within Italy were seen (home deaths ranged between 94.0% in the south and 28.2% in the north east). Home was the preferred place of death for 93.5% of patients that expressed a preference, with minimal differences within the country (between 89.5% and 99.0%). Overall 67.1% of the sample died in the place where they preferred to die. CONCLUSIONS: Policymakers should encourage health services to focus on ways of meeting individual preferences on place of death. As home was the preferred place of death for most cancer patients, effective programmes to enable the patients to remain at home should be implemented.     

Where cancer patients die: an epidemiologic study

In a sample of deaths among cancer patients, the relationship of place of death to age, sex, length of time between diagnosis and death, cancer site, and patients' socioeconomic status was investigated. The Rochester (N.Y.) Regional Tumor Registry provided these data for all cancer patients who died in Monroe County, N.Y., during 1976, 1977, and 1978. Patients who had not been residents of the county were excluded from the sample, as were patients under 15 years of age at death and those whose cancers had been diagnosed only at autopsy. Analysis with a logit model was used to estimate odds ratios that compared the probabilities of death in an acute care hospital and in a chronic care facility with the probability of death at home. Patients whose cancers had been diagnosed less than 1 month before their deaths were significantly more likely to die in a hospital than were patients whose cancers had been diagnosed earlier. Cancer sites, too, were significantly related to place of death: persons with leukemia or lymphoma were most likely to die in a hospital, followed by patients with lung, breast, and upper gastrointestinal tract cancers; persons with colorectal, genitourinary, and miscellaneous cancers were most likely to die at home. The patients whose deaths were studied were classified by socioeconomic area (SEA) ranking. Patients who had resided in higher level SEAs were more likely to die at home than those from lower level SEAs; however, this trend was reversed among patients from the lowest level SEAs, who had a relatively high rate of death at home and a low rate of death in chronic care facilities.     

Do social factors affect where patients die: an analysis of 10 years of cancer deaths in England.

BACKGROUND: This study investigated whether indices of social deprivation were related to the proportion of cancer patients who died at home. METHODS: Data were derived from death registrations for all cancer deaths 1985-1994 in England. Two indices of deprivation (Underprivileged Area Score (UPA), or Jarman, and Townsend scores) were calculated for each electoral ward; 1991 Census data were used. The scores use combinations of variables, including the percentage in overcrowded homes, the percentage of elderly people living alone, the percentage of one-parent families, etc. A high score indicates more deprivation. The main outcome measures were the proportion (in five and ten year averages) of cancer deaths which occurred at home, calculated for every electoral ward (with populations usually ranging from 5000 to 11,000). Spearman rho was used to test for correlations between the proportion of cancer deaths at home and deprivation score. Electoral wards were categorized by deprivation score into three groups of equal size and analysed over 10 years. Multivariate analysis was used to determine the relative association of different patient based and electoral ward variables with cancer death at home. p < 0.05 (two-tailed) was taken as significant. RESULTS: There were over 1.3 million death registrations from cancer in the 10 years. The proportion who died at home was 0.27, in hospital 0.47, and other setting 0.26. There were wide variations (0.05-0.75) in the proportion of people who died at home in different electoral wards. Small inverse correlations were found between the percentage who died at home and the UPA (-0.35; p < 0.001) and Townsend (-0.26; p < 0.001) scores. The correlations were greatest in North Thames (-0.63, UPA) and smallest in West Midlands (-0.20, UPA). The proportion of home deaths for the different bands of deprivation were: 0.30 (low deprivation), 0.27 (middle deprivation) and 0.24 (high deprivation). Plotting the trends over 10 years suggests no change in this relationship. Multiple regression analysis predicted several ward and patient characteristics and accounted for 30 per cent of the variation. Increased age (patient variable), Jarman score and ethnic minorities (both ward variables) were associated with fewer patients dying at home. Being male and having cancer of the digestive organs were associated with home death. CONCLUSION: There are wide variations in the percentage of cancer deaths at home in different electoral wards. Social factors are inversely correlated with home cancer death, and may explain part of this variation. Home care in deprived areas may be especially difficult to achieve.     

The place of death of cancer patients: can qualitative data add to known factors?

Research on the distribution of cancer deaths by setting-hospital, hospice, home, other--is longstanding, but has been given fresh impetus in the UK by policy commitments to increase the proportion of deaths occurring in patients' homes. Studies of factors associated with the location of cancer deaths fall into two main categories: geo-epidemiological interrogations of routinely collected death registration data, and prospective and retrospective cohort studies of terminally ill cancer patients. This paper summarises the findings of these studies and considers the place of death factors that are generated in semi-structured interviews with 15 palliative care service providers working in the Morecambe Bay area of north-west England. These qualitative data are found not only to confirm and considerably enrich understanding of known factors, but also to bring new factors into view. New factors can be grouped under the headings: service infrastructure, patient and carer attitudes, and cultures of practice. Such an approach provides useful information for policy makers and practitioners in palliative care.     

Dying at home or in an institution using death certificates to explore the factors associated with place of death

INTRODUCTION: The knowledge of determinants of place of death is important for public health policy aimed at improving the quality of end-of-life care. METHODS: We investigated the influence of clinical, socio-demographic, residential and health care system factors on the place of death, using data from all 55,759 deaths in 2001 in Flanders (Belgium), gathered via official death certificates and data from anonymously linked health care statistics. A multivariate logistic regression was used to examine the associated factors (home versus hospital as dependent categories). RESULTS: Of all deaths in Flanders, 53.7% took place in hospital, 24.3% at home and 19.8% in a care home. The probability of home deaths varied by region, by rural or urban residence and by the hospital bed availability in the region and dying at home was less likely among those suffering from certain non-malignant chronic diseases, the less educated and those living alone. CONCLUSION: Although most people wish to die at home, most deaths in Flanders (Belgium) in 2001 did not take place there. The clinical, socio-demographic and residential factors found to be associated with the place of death could serve as focal points for a policy to facilitate dying in the place of choice, including at home.     

AIDS deaths shift from hospital to home. AIDS Mortality Project Group.

OBJECTIVE. This study monitors trends in place of death among persons with acquired immunodeficiency syndrome (AIDS) as a measure of health care usage patterns and terminal health care among persons infected with human immunodeficiency virus (HIV). METHODS. Sixteen health departments collected death certificates for 55,186 persons with AIDS whose deaths occurred through 1991. Place of death was categorized as hospital, residence, hospice or nursing home, and other. RESULTS. The percentage of AIDS deaths at hospital facilities decreased from 92% in 1983 to 57% in 1991. In 1988, 23% of deaths occurred at home or in hospices and nursing homes. This trend was more evident among men, Whites, and men who had sex with men; less so among persons with other modes of exposure; and not at all among injecting drug users and children with perinatally acquired AIDS. Place of death varied by geographic location, with the greatest percentage of hospital deaths in the Northeast (91%) and the greatest percentage of at-home deaths in the West (27%). CONCLUSIONS. The percentage of AIDS deaths at home or in hospices and nursing homes has increased since 1983. These trends may reflect changes in hospital use for end-stage HIV infection. Decreasing hospitalization and increasing outpatient services and home care will decrease costs and may allow HIV-infected persons improved social support.     

Proportion of cancer deaths occurring in hospital, Canada, 1994-2000

BACKGROUND: Most terminally ill cancer patients would prefer not to die in hospital, but only a minority achieve their wish. Our objective was to examine the proportion of cancer deaths occurring in Canadian hospitals. METHODS: The two sources of data (1994-2000) were: 1) all hospital separations (HS) with a primary diagnosis of cancer and discharge as 'dead'; 2) all death certificates (DC) with cancer as underlying cause of death. Proportions of hospital deaths were estimated with two different numerators: 1) hospital cancer deaths from HS data, and 2) deaths with hospital as location from DC data; the denominator for both were all cancer deaths identified from the DC data. RESULTS: Proportions of hospital deaths from HS data decreased from 55% to 40% over 1994-2000, was slightly lower for females, decreased with age, but varied widely among provinces. Proportions of hospital deaths from DC data started at 80% and showed a small downward trend over the years. While age, sex, and cancer site distributions stayed the same, the proportion of hospital deaths from DC date again varied among provinces. For provinces with the home category completed on the DC data, 1999-2000, Alberta had most home deaths at 15.6% and PEI least at 5.7%. INTERPRETATION: This is the first Canada-wide data on place of death for terminal cancer, which is important for determining and comparing present-day practices, as well as for planning for the future.     

Place of death: preferences among cancer patients and their carers

The place of death of cancer patients has become an important theme in UK cancer and palliative care policy. This paper examines the place of death preferences of 41 terminally ill cancer patients and 18 of their informal carers, living in the Morecambe Bay area of north-west England. We interviewed cancer patients referred to the research team by 13 specialist palliative care professionals; patients had an estimated 3 months of life remaining. The study design involved an in-depth qualitative interview with each patient soon after referral to the study, followed by an interview some 4 weeks later and subsequent tracking interviews by telephone at 2-4 week intervals until death occurred. Interviews were also conducted with main coresident carers soon after patient referral to the study and again in the post-bereavement period. Thirteen factors were identified as shaping the place of death preference of patients and carers. These are organised into four thematic domains: the informal care resource, management of the body, experience of services, and existential perspectives. In documenting these factors, this paper adds significantly to current knowledge on the factors that shape place of death preference, a field of enquiry acknowledged to be underdeveloped (J. Palliative Med. 3 (2000) 287). More importantly, it uncovers some of the reasons that underpin these preferences. Our research revealed a much stronger preference for deaths in a hospice than had been anticipated, leading us to take a qualified stance on the current policy drive in favour of home deaths by those charged with delivering UK cancer and palliative care services.     

A comparison of hospice and conventional care

Interviews with relatives and others who knew a group of people dying of cancer in England are reported. The bulk of the paper compares 45 such patients who received hospice care with 126 who received conventional care. The sampling procedure showed that 2.9% of people aged 15 or over at death died in a hospice, and 6.9% received some form of hospice service. The hospice patients differed in several ways from other cancer patients. They had fewer conditions other than cancer recorded on the death certificate, were believed to be more religious and were more likely to suffer from a variety of symptoms and restrictions, including pain. Hospice patients were reported more likely to know that they were dying and respondents' levels of satisfaction with hospice home nursing and in-patient hospice care were significantly higher than for other forms of care. Hospice home nurses were found to have adopted a more advisory approach to nursing care than other home nurses who focussed more on practical care. When final admissions were considered, in-patient hospice care involved fewer medical interventions and, in the last year of life, those receiving hospice services were less likely to have an operation. There were few differences between the two groups in what happened at the time of death, although for in-patient deaths, respondents judged the staff of hospices to be more understanding. The relatives of hospice patients were more likely to be visited by a nurse at home after the death. Few differences in bereavement reaction were found, but those that were suggested that respondents for the hospice group were adjusting better.(ABSTRACT TRUNCATED AT 250 WORDS)     

Discrepancy in the preferences of place of death between terminally ill cancer patients and their primary family caregivers in Taiwan

There is a worldwide common preference for dying at home. However, death at home does not come without significant challenges and potential consequences for families. Given the interactive nature of decisions regarding the place of death, the family's perspective is important and needs to be investigated. The purposes of this study were to compare (1) Taiwanese terminally ill cancer patients' and their family caregivers' preferences for the patient's place of death; and (2) important factors that are considered in choosing the preferred place of death from both points of view. A total of 617 dyads of terminally ill cancer patients and their family caregivers were surveyed. The majority of both terminally ill cancer patients and their family caregivers preferred to die at home (61.0% and 56.9%, respectively). A higher proportion of the family caregivers indicated a preference for hospital death for the patients. There was a moderate association between the two respondents in the preferences of place of death. Results underscore discrepancies between patients and their families in the importance given to cultural concerns, quality of health care, worries of being a burden to others, lack of availability of families, relationships with health care providers, and being surrounded by the home environment. Effective interventions need to be developed which can lighten the caregiving burden and help families retain dying patients at home, avoid unnecessary re-hospitalizations and unfavorable hospital deaths, and improve accordance with the patient's wishes.     

Geographical variations of place of death among Italian communities suggest an inappropriate hospital use in the terminal phase of cancer disease

This study aimed to assess geographic variations of place of death among Italian communities, to investigate the determinants of home death, and to examine trends in the proportion of home deaths over a period of 8 y. A cross-sectional analysis was performed on 13 provinces from two Italian regions (Liguria and Toscana), and the geographic variations and determinants of home death were studied for the 17,597 residents, who died of cancer in 1991. Trends for both the regions of the proportion of home deaths were examined for the period 1987 and 1995. A remarkable heterogeneity in the observed proportion of home deaths among the 13 provinces was observed, ranging between 31.4 and 40.4% in Liguria and between 37.7 and 73.3% in Toscana. The estimated proportion of home deaths after adjustment for age, gender, marital status, education, place of birth, characteristics of the living area, and cancer site remained substantially the same. The proportion of home deaths significantly increased with increasing age, and years of education. It was higher among females, married and widowed patients, native patients, and for residents in a semi-urban or rural areas. A significant decrease in the percentage of patients who died at home between 1987 and 1995 in both regions was observed. The wide geographical and social differences observed between and within the communities for the frequency of home deaths are not explained by the distribution of known determinants, and possibly suggest patterns of inappropriate hospital admissions in the terminal phase of disease.     

Deaths of children occurring at home in six European countries

Objectives Until now there have been no population‐based European data available regarding place of death of children. This study aimed to compare proportions of home death for all children and for children dying from complex chronic conditions (CCC) in six European countries and to investigate related socio‐demographic and clinical factors. Methods Data were collected from the death certificates of all deceased children aged 1–17 years in Belgium, the Netherlands, Norway, England, Wales (2003) and Italy (2002). Gender, cause and place of death (home vs. outside home) and socio‐demographic factors (socio‐economic status, degree of urbanization and number of hospital beds in the area) were included in the analyses. Data were analysed using frequencies and multivariate logistic regression. Results In total 3328 deaths were included in the analyses; 1037 (31.2%) related to CCC. The proportion of home deaths varied between 19.6% in Italy and 28.6% in the Netherlands and was higher for children dying from CCC in all the countries studied, varying between 21.7% in Italy and 50% in the Netherlands. Among children dying from CCC, home death was more likely for cancer patients and those aged over 10 years. After controlling for potentially related clinical and socio‐demographic factors, differences in the proportion of home deaths between countries remained significant, with higher proportions in Belgium and the Netherlands as compared with Italy. Conclusions Although home deaths comprise a substantial proportion of all deaths of children with CCCs, variation among disease categories and across countries suggest that considerable potential still exists for further improvements in facilitating end‐of‐life care in the home for those children and families who desire to be in this location.     

Differences in breast cancer incidence in Australia and England by age,extent of disease and deprivation status: women diagnosed 1980–2002

Objective: To compare breast cancer incidence in England and Australia by age, extent of disease and deprivation. Methods: We analysed data for women aged 15–99 years diagnosed with breast cancer in England or Australia during the period 1990 to 1994, and in West Midlands or New South Wales during the period 1980 to 2002. We calculated three‐year rolling average incidence rates and incidence rate ratios (IRR) between West Midlands and New South Wales by age, extent of disease and category of deprivation. Results: Breast cancer incidence was higher in England than in Australia, and in West Midlands than in New South Wales but became more similar over time. Socio‐economic differences in incidence were greater in New South Wales than in West Midlands. The most deprived women in West Midlands were diagnosed at a later stage of disease than the most deprived women in New South Wales. Incidence among elderly women was higher in West Midlands than in New South Wales. There were also high proportions of tumours with unknown stage among elderly women in West Midlands. Conclusions: Although the overall incidence of breast cancer is similar, differences by age, extent of disease and deprivation exist. Implications: The underlying reasons for these patterns require further examination.     

The ageing and de-institutionalisation of death—Evidence from England and Wales

Increasingly, age of death is postponed until very old age, and care of those who are dying is challenged by medical co-morbidities and the presence of dementia. Although most people would prefer to die at home, currently in England and Wales only about 20 per cent of those aged 65 years and over die at home, and this proportion falls to about 10 per cent among those aged over 85 years. To explore recent and likely future trends in age and place of death, mortality statistics from 2006 to 2013 were analysed and projected to 2050 using age- and gender-specific rates. Results confirmed recent increasing age at death and indicated a trend for increasing proportions of older people to die at home. Projections indicated large increases in home-based deaths, particularly for men aged 65 and over. Consistent with people’s wishes, there may be a partial return to the view that dying at home is a normal experience. Resource allocations are likely to need to shift to support people dying at home and their formal and informal carers.     

Trends and patterns in place of death for Medicare enrollees

Two changes in the Medicare program in 1983 may have affected where aged persons die--the change from retrospective hospital reimbursement to the prospective payment system and passage of the Medicare hospice benefit. Patterns and trends in where people die--hospitals, other institutions such as nursing homes, decedents' homes, and other places--for persons 65 years of age or over from 1980 through 1986 are examined. The proportion of deaths in hospitals declined somewhat after implementation of prospective payment. The hospice benefit may have caused the shift among cancer patients away from hospital deaths toward deaths at home.     

Where a Cancer Patient Dies: The Effect of Rural Residency

CONTEXT: Surveys indicate 50% to 80% of cancer patients would choose to die at home if possible, although far fewer actually do. In Nova Scotia (NS), cancer deaths occurring out-of-hospital increased from 19.8% in 1992 to 30.2% in 1997. The impact of rural residency on this trend has not been studied. PURPOSE: To determine the association between dying of cancer in a rural locale and the likelihood of it being an out-of-hospital death. METHODS: Secondary analysis of linked population-based administrative health data files. Subjects were all Nova Scotians who died of cancer from 1992 to 1997. Measures included location of death, dichotomized as a hospital death or an out-of-hospital death; and urban-rural residency, using an enumeration area urban-rural indicator created from postal code information adjusted for individual characteristics. RESULTS: Of the 13,652 total cancer deaths, 6171 occurred in rural NS, of whichl 1471 (23.8%) died out-of-hospital. Out-of-hospital deaths in rural NS increased from 16.2% in 1992 to just over 27% in 1997. Compared with urban cancer patients, the adjusted odds of an out-of-hospital death in rural NS was lower (adjusted odds ratio, 0.87; 95% confidence interval, 0.79-0.95). CONCLUSIONS: There was an increasing trend during the 1990s for cancer patients to die out-of-hospital. Compared with their urban counterparts, patients in rural areas were less likely to do so. Those with cancer living in the rural setting who wish to die at home may face unique challenges.     

Mini-symposium — Public Health Observatories: Assessing and communicating the health and judicial impact of alcohol use

BACKGROUND: Alcohol consumption places an increasing burden on health services, criminal justice agencies and private industry throughout the UK. Despite a national strategy to tackle alcohol-related harm, there remains a lack of epidemiology on alcohol use and related harms at local levels. Utilising national data sources and existing research studies, Regional Public Health Observatories are appropriately placed to calculate such measures and examine their relationship with deprivation. METHODS: For the North West of England, borough and health locality data were extracted from national sources. Alcohol consumption utilised lifestyle survey data and estimates of related harm were calculated by applying existing alcohol attributable fractions to deaths, hospital episodes and crime data. Contribution of alcohol to reduction in life expectancy was also calculated and all measures were correlated with deprivation. RESULTS: For the North West, the annual burden of alcohol was estimated at over 3700 deaths, 56,700 hospital episodes and 71,000 crimes. Annual alcohol-related death rates for men varied from 0.43 to 1.17 per 1000 between Local Authorities and there was a six-fold variation in alcohol-related crime rate between areas. For males, alcohol reduced life expectancy by more than five months in the area worst affected. For both sexes, more deprived areas had significantly higher levels of alcohol-related hospital episodes, crime, and contribution to reduced life expectancy. CONCLUSIONS: Alcohol-related harm requires urgent, multi-agency attention at local levels throughout the UK. Applying appropriate research methods to national data sources provides limited but valuable local measures of alcohol use and its impact on health and crime. Regional Public Health Observatories can facilitate partnership working through such provision of national intelligence tailored to support local and regional action. Application of these techniques also helps identify additional requirements for better local intelligence on health-related behaviours and the necessity for consistency across local health and behaviour surveys.     

COVID-19 and deprivation amplification: An ecological study of geographical inequalities in mortality in England

‘Deprivation amplification’ is used to understand the relationship between deprivation, scale and COVID-19 mortality rates. We found that more deprived Middle Super Output Areas (MSOAs) in the more deprived northern regions suffered greater COVID-19 mortality rates. Across England, the most deprived 20% of MSOAs had higher mortality than the least deprived (44.1% more COVID-19 deaths/10,000). However, the most deprived MSOAs in the north fared worse than equally deprived areas in the rest of England (14.5% more deaths/10,000, beta = 0.136, p < 0.01). There was also strong evidence of spatial clustering and spill-overs. We discuss these findings in relation to ‘deprivation amplification’, the ‘syndemic pandemic’, and the health and place literature.     

Cause-specific perinatal death rates, birth weight and deprivation in the West Midlands, 1991-93

OBJECTIVES: To study the relationship between cause-specific perinatal death rates, material deprivation and birthweight among births in 3 consecutive years in the West Midlands Health Region. STUDY DESIGN: Retrospective cohort study. SETTING: West Midlands Health Region (WMHR). STUDY POPULATION: All births (live and stillbirths) to mothers with addresses in the WMHR in 1991, 1992 and 1993. MAIN OUTCOME MEASURES: Cause-specific perinatal death rates--crude and stratified by birthweight. METHODS: Perinatal deaths in the WMHR in 1991-93, collected as part of the national Confidential Enquiry into Stillbirths and Deaths in Infancy, were classified into causes of death by the extended Wigglesworth classification. Crude rates for cause-specific perinatal deaths and rates stratified by birthweight < 2500 g and > or = 2500 g were calculated for each enumeration district (ED) quintile derived by ranking enumeration districts for the whole of the region by Townsend Deprivation Index. Cause-specific rates of death were investigated for a linear trend across ED quintiles. The relative risk of death (most vs least deprived) from specific causes was calculated. Using rates for the least deprived quintile as the reference, deaths from each cause 'attributable' to social inequality were calculated. RESULTS: Positive linear trends in perinatal deaths were noted with increasing deprivation for each specific cause of death except those classified as 'other causes' (Wigglesworth Class E). Relative risk (most vs least deprived) of perinatal death with a congenital anomaly was 1.98 (confidence interval, CI: 1.36,2.89). For deaths related to antepartum events, intrapartum events and immaturity the risks were 1.81 (CI: 1.39,2.38), 1.48 (CI: 1.10,1.98) and 1.92 (CI 1.45,2.56), respectively. Forty-three (35.7%) perinatal deaths per year were due to congenital anomalies, 63 (29.7%) antepartum events, 36 (21.9%) intrapartum events and 61 (32.7%) immaturity and these were statistically 'attributable' to social inequality. Cause-specific perinatal death rates for babies < 2500 g showed no correlation with deprivation; however, for babies > or = 2500 g the association with deprivation persisted. CONCLUSIONS: All cause-specific rates except those due to 'other causes' showed a positive linear trend with increasing deprivation. These trends were found for infants born > or = 2500 g but were not seen for low birthweight infants (< 2500 g). Almost 30% of deaths were statistically 'attributable' to social inequality. The results of this study suggest that material deprivation plays an important role in the causal pathway leading to perinatal death and needs to be addressed in preventive programmes aimed at the reduction of perinatal deaths.     

Socio-economic deprivation is associated with increased proximity to general practices in England: an ecological analysis

BACKGROUND: As health status is consistently negatively correlated with socio-economic deprivation, the need for health services is generally assumed to be greater in more deprived communities. The Inverse Care Law predicts that access to good quality primary care services in more deprived wards will be less than that in affluent wards. However, the relationship between deprivation and geographical proximity to health services has received little attention. METHODS: We investigated the relationship between geographical proximity to general practices and a number of markers of socio-economic deprivation at the electoral ward level in the North East of England using various domains of the Index of Multiple Deprivation 2000 (IMD2000). RESULTS: More deprived wards, as measured by the employment, education and income domains of the IMD2000, had greater proximity to general practices, as measured by the access domain of the IMD2000, than affluent wards. This results held in both urban and rural wards. CONCLUSIONS: Contrary to our expectations and the predictions of the Inverse Care Law, geographical proximity to general practices was greater in more deprived, compared to more affluent wards. However, geographical proximity to services does not necessarily ensure that services will be accessed or that they are of good quality.