SEARCH

SEARCH BY CITATION

Keywords:

  • access to health care;
  • diabetes;
  • ethnicity and inequality;
  • socioeconomic factors;
  • systematic review

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

The major increase in the prevalence of diabetes mellitus (DM) has led to the study of social inequalities in health-care. The aim of this study is to establish the possible existence of social inequalities in the prevention, diagnosis, treatment, control and monitoring of diabetes in Organisation for Economic Co-operation and Development (OECD) countries which have universal healthcare systems. We searched MEDLINE, EMBASE and the Cochrane Database of Systematic Reviews for all relevant articles published up to 15 December 2007. We included observational studies carried out in OECD countries with universal healthcare systems in place that investigate social inequalities in the provision of health-care to diabetes patients. Two independent reviewers carried out the critical assessment using the STROBE tool items considered most adequate for the evaluation of the methodological quality. We selected 41 articles from which we critically assessed 25 (18 cross-sectional, 6 cohorts, 1 case-control). Consistency among the article results was found regarding the existence of ethnic inequalities in treatment, metabolic control and use of healthcare services. Socioeconomic inequalities were also found in the diagnosis and control of the disease, but no evidence of any gender inequalities was found. In general, the methodological quality of the articles was moderate with insufficient information in the majority of cases to rule out bias. This review shows that even in countries with a significant level of economic development and which have universal healthcare systems in place which endeavour to provide medical care to the entire population, socioeconomic and ethnic inequalities can be identified in the provision of health-care to DM sufferers. However, higher quality and follow-up articles are needed to confirm these results.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Inequality in terms of health originates in the political, economic and social inequalities within a society and it refers to the different health-related opportunities and resources that are available to individuals belonging to a particular social class, gender, ethnic group or area, with the result that the most disadvantaged groups suffer worse health than the rest (Mackenbach & Bakker 2003, Marmot et al. 2008). The relevance of inequalities in terms of health-care is especially evident among patients suffering from chronic or long-term illnesses, as these people have a relationship with the healthcare system during a greater length of time (Eakin et al. 2002).

Diabetes is the fourth most common cause of death worldwide, and currently affects 246 million people around the globe, and it is estimated that around 380 million people will have diabetes by the year 2025 (International Diabetes Federation 2006). However, this disease is not distributed equally among all population groups, as higher incidence, appearance of complications and higher mortality rate have been observed among women, ethnic minorities and people with a low socioeconomic level (Robbins et al. 2001, Brown et al. 2004). In addition, although no systematic review has been found in this respect, various primary studies have indicated that these population groups have greater difficulty accessing healthcare services regarding prevention (Meisinger et al. 2005), diagnosis (van Laar et al. 2007), treatment (Klinke et al. 2004) and control of the disease (Peyrot et al. 2006), which is especially relevant given that healthcare provision is fundamental in the case of diabetes mellitus (DM) for maintaining a good quality of life, reducing complications and increasing the chances of survival for those affected (Del Prato et al. 2007). These studies have been carried out both in countries with healthcare systems that guarantee a universal provision of health-care as well as in countries where such provision is basically linked to private insurances; although it has not been established yet what effect the type of healthcare system has on these inequalities. Due to this, in the present systematic review, only countries belonging to the Organisation for Economic Co-operation and Development (OECD) which have a universal healthcare system in place have been analysed, as these countries benefit from a high level of economic development as well as from healthcare systems being therefore in a better position to prevent inequalities.

The aim of this study is to carry out a systematic review on observational research on gender, ethnic and socioeconomic inequalities in the prevention, diagnosis, treatment, control and monitoring of DM in the OECD countries with a universal healthcare system in place.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Data sources and search strategy

A bibliographic search in the databases MEDLINE and EMBASE was carried out. All Cochrane Library’s databases were also searched in order to retrieve the systematic reviews on the subject. For the first two databases a search strategy was designed, whereas for the third the existence of systematic reviews was investigated by entering the keyword: ‘Diabetes’. The search strategy was limited to comparative study, journal article and meta analysis published in English, French, Portuguese, Spanish and Italian between 1967 and December 2007. Three search themes were combined using the Boolean operator ‘and’. For the first one, diabetes, the exploded versions of Medical Subject Headings (MeSH) DM, type 1 or mellitus, type 2 or diabetes, gestational or diabetic ketoacidosis were combined. The second theme, inequalities, combined exploded versions of MeSH terms sex factors or ethnic groups or socioeconomic factors or poverty or social conditions or social class or gender identity or social justice or prejudice or text words educational factors or geographic factors or inequit* or equity* or disparit* or inequalit* or bias. The third theme, country where the studies were carried out, combined exploded versions of MeSH terms Spain or Portugal or Greece or Italy or Great Britain or Ireland or France or Germany or Austria or Belgium or the Netherlands or Denmark or Finland or Norway or Sweden or Canada or Japan or Australia or New Zealand.

A bibliographical database was created using Reference Manager Professional Edition Version 10© (ISI ResearchSoft; Thomson Reuters, New York, NY, USA) and used to store the references found.

Study selection

The articles included had to fulfil the following criteria: (i) those analysing the association between any of the inequality axes (socioeconomic, gender and ethnic) and any type of healthcare provision (prevention, diagnosis, treatment, control and monitoring), (ii) studies carried out in OECD countries with a universal healthcare system (Spain, Portugal, Greece, Italy, Great Britain, Ireland, France, Germany, Austria, Belgium, the Netherlands, Denmark, Finland, Norway, Sweden, Canada, Japan, Australia and New Zealand), (iii) those published in Spanish, French, Italian, Portuguese or English, (iv) original studies with a comparison group, systematic reviews and meta-analysis. Articles not related to DM, non-original studies (narrative reviews, letters, editorials, opinion articles, etc.) and qualitative studies were excluded.

In order to select the studies, the titles and abstracts were read, and those which fulfilled the criteria for inclusion were selected. Following this, the complete texts of the selected articles were read, and those which did not meet the criteria described above were excluded. This was done by two independent reviewers, and any discrepancies were resolved by a third reviewer.

Data extraction, critical assessment and level of evidence

Data were extracted from the articles selected using a data extraction sheet. This sheet was designed and tested so that the specific features of this type of article could be included. Information was extracted from the axis of inequality studied, type of health-care provided, country where the study was carried out, type of DM, sample size, year of publication, design used for the studies and main results. The exposure variables were gender, ethnicity and socioeconomic level (measured as level of income and/or education and/or occupational status). The type of health-care was classified as diagnosis, treatment, control of the disease measured by clinical criteria, and healthcare services related to the control of DM. This information, along with data regarding the methodological quality assessment, was stored in a database created using Microsoft Access.

For the evaluation of the methodological quality of the studies, the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) (von Elm et al. 2007) was used, a tool which has recently been used in various systematic reviews to evaluate the methodological quality of observational studies (Olmos et al. 2008, Scales & Dahm 2008). This tool was initially developed to evaluate the clarity in communication and therefore, in this study, only those items that were considered most adequate for the evaluation of the methodological quality were selected. These items (a total of 14) were those in the methods, results and other analyses sections, and 1 point was assigned to the study for every correct item. The methodological quality was only analysed in the case of finding three or more articles sharing the same research question (the same axis of inequality studied and the same result variables), and the assessment was carried out by two independent reviewers (I.R.C. and A.O.L.). Discrepancies were resolved by a third reviewer (I.R.P.).

The level of scientific evidence was established based on the methodological quality and epidemiological design of the studies. Cohort and case-control studies with a low risk of bias are thought to be the types of observational studies that provide the highest levels of scientific evidence, while cross-sectional studies provide less evidence (Scottish Intercollegiate Guidelines Network 2010).

Summary of results

Because of the high level of heterogeneity of the articles, it was impossible to provide a quantitative summary of the results, so a qualitative summary was produced. To do this, the articles included in this review were categorised and analysed based on the inequality axis and healthcare types assessed.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

Study selection

As a result of searching in MEDLINE and EMBASE, 1430 bibliographic references were retrieved. No studies were found in the Cochrane Collaboration. After eliminating 183 duplicates, 1040 references were excluded after their abstracts were read. The main reason for exclusion was that the studies focused on diabetes, but did not study the relationship between prevention, diagnosis, treatment or control and monitoring of diabetes and inequalities. After the full text of 193 articles was read, a further 152 were excluded. Of these, 105 were excluded because although they focused on diabetes, they did not study the relationship between prevention, diagnosis, treatment or control and monitoring of diabetes and inequalities. Forty seven were excluded because, even though they examined some variables that measure inequalities in diabetes, their primary focus was not the topic studied here. Among the 41 selected articles, seven groups of articles containing 25 different studies could be identified as they shared the same research question and the same result variables, and therefore these were critically assessed (Figure 1). Given that some studies analysed several axes of inequality and/or several types of health-care simultaneously, they were included in more than one group. Those groups are as follows:

image

Figure 1.  Study selection.

Download figure to PowerPoint

Study characteristics

Although the search strategy allowed retrieving articles published between 1967 and 2007, the selected studies were published between 1995 and 2007, and a substantial increase was observed from 2003 onwards (Tables 1–7). Eleven studies were carried out in Great Britain, three in Germany, two in France, New Zealand, Australia and Sweden and one in Spain, Canada and Holland. In 18 studies a cross-sectional design was used, six were cohort studies and one was case-control. In general, the methodological quality of the articles was moderate, with an average grading of 9.29 criteria being fulfilled out of a total of 14 (66.36%). It was observed that 91.67% adequately presented the key information related to the design, 80% of the articles described the contexts, places and dates as well as recruitment, follow-up and data-gathering periods, 87.5% contained information on eligibility criteria, information sources and patient selection methods, 83.3% accurately defined the variables as result, exposure, predictor and effect modifier variables, 95.8% provided the data sources as well as details of the methods of assessment, 87.5% explained the steps taken to control potential sources of bias, 83.3% indicated the number of participants in every phase of the study, 91.7% described the number of result events and provided summary-measures over time, and 83.3% provided non-adjusted and adjusted estimations, as well as precision. However, in 33.3% problems arose in the explanation of how sample size was determined, only 25% explained some of the statistical methods, 70% did not provide information on the participants’ characteristics and the exposures and potential confounding factors, and only 4.2% describe measures adopted for overcoming potential sources of bias.

Table 1.   Gender inequalities in access to healthcare services related to the control of diabetes. Characteristics and results
SourceCountryDesignNType of diabetesQuality assessment*Healthcare servicesResults
  1. DM, diabetes mellitus; N, sample size; OR, odds ratio.

  2. *Result of the critical evaluation carried out using the STROBE tool.

Wredling et al. 1998SwedenCross-sectional348 men and 301 womenType 1 DM and Type 2 DM6Medical check-ups and educational programmesType 1 DM: Women went for more medical check-ups (P < 0.05) and to educational programmes (P < 0.05) and men to the nursing service (P < 0.05). Type 2 DM: Men went more often to the nursing service and received more education on the care of diabetes (P < 0.05).
El Fakiri et al. 2003HollandCross-sectional338Type 1 DM and Type 2 DM13GP, internal medicine specialist, cardiologist, dietician, physiotherapist, podiatrist, hospital stay, complementary medicine and mental healthcareNo significant differences were observed.
Millett & Dodhia 2006Great BritainCross-sectional4089 men and 3972 womenType 2 DM10Diabetic retinopathy screening serviceNo significant differences were observed.
Table 2.   Ethnic inequalities in the treatment of diabetes. Characteristics and results
SourceCountryDesignNType of diabetesEthnicity*Quality assessmentType of treatmentResults
  1. DM, diabetes mellitus; N, sample size; OR, odds ratio.

  2. *Ethnic group or country of origin.

  3. Result of the critical evaluation carried out using the STROBE tool.

Sedgwick et al. 2003Great BritainCross-sectional1899Type 2 DMBlack African, afro-Caribbean, Caucasian10Antihypertensives and insulinNo significant differences were observed.
Lloyd et al. 2004Great BritainCross-sectional303 primary healthcare servicesType 2 DMIndia, Bangladesh, Pakistan, Africa and Caribbean7Cost associated with the prescription of oral antidiabeticsAmong Pakistanis, Bangladeshis and Hindus a moderate correlation was observed between the area of residence and the prescription of oral antidiabetics (r 0.60).
Robinson et al. 2006New ZealandCross-sectional5917Type 2 DMEurope, Maori, Pacific, other Asian, Indian, others10Aspirin, antihypertensives and insulin‘Other Asians’ were treated less with aspirin (OR = 0.74, CI:95% = 0.56–0.99), with two or more antihypertensives (OR = 0.55, CI:95% = 0.40–0.74) and with insulin (OR = 0.44, CI:95% = 0.28–0.70) than the other groups.
Tomlin et al. 2006New ZealandCross-sectional13 281Type 1 DM and Type 2 DMEurope, Maori, Pacific Islands8Diet, statins and antihypertensivesType 1 DM: No significant differences were observed. Type 2 DM: Maori and Pacific Islands residents, compared with Europeans, were treated less with diet and statins (P < 0.05) but received more antihypertensives (P < 0.01).
Table 3.   Ethnic inequalities in the control of diabetes measured by clinical variables. Characteristics and results
SourceCountryDesignNType of diabetesEthnicity*Quality assessmentClinical variablesResults
  1. DM, diabetes mellitus; HbA1c, glycosylated haemoglobin; BMI, Body Mass Index; N, sample size; OR, odds ratio.

  2. *Ethnic group or country of origin.

  3. Result of the critical evaluation carried out using the STROBE tool.

Tomlin et al. 2006New ZealandCross-sectional13 281Type 1 DM and Type 2 DMEurope, Maori and Pacific Islands8HbA1cEuropeans were better at glycaemic control (P < 0.05).
Tubiana-Rufi et al. 1995FranceCross-sectional165Type 1 DMFrench and immigrants10HbA1cChildren of immigrant mothers have higher HbA1c values than children of native mothers (P < 0.05).
Ralph-Campbell et al. 2006CanadaCohort394Type 2 DMAboriginal and non-aboriginal10HbA1c, blood pressure, BMI and cholesterolemiaNo significant differences were observed.
McElduff et al. 2005Great BritainCohort2890Not specifiedSouthern Asian and European10HbA1c, blood pressure and cholesterolemiaInequalities diminished over time in the control of diabetes between both groups.
Davis et al. 2001Great BritainCohort2999Type 2 DMCaucasian, afro-Caribbean and Hindu10Weight, glycaemia, HbA1c and blood pressureNo significant differences were observed.
Kousta et al. 2006Great BritainCase-control850Gestational diabetesEuropean, Asian or Hindu and afro-Caribbean9BMI, abdominal circumference, blood pressure, cholesterolemia, glycaemia and insulinemiaNon-European women were worse at glycaemic control than European women (P < 0.05).
Davis et al. 2007AustraliaCross-sectional837Type 2 DMAboriginal and Anglo-Celtic6HbA1c, glycaemia, BMI, abdominal circumference and cholesterolemiaThe aboriginal population were worse at glycaemic control than the Anglo-Celtic population.
Bruce et al. 2003AustraliaCross-sectional1264Type 2 DMAboriginal and Anglo-Celtic10HbA1cAustralian aboriginals were better at glycaemic control than Anglo-Celtic patients (OR = 0.30, CI:95% = 0.11–0.81).
Table 4.   Ethnic inequalities in access to healthcare services related to the control of diabetes. Characteristics and results
SourceCountryDesignNType of diabetesEthnicity*Quality assessmentHealthcare servicesResults
  1. DM, diabetes mellitus; N, sample size; OR, odds ratio.

  2. *Ethnic group or country of origin.

  3. Result of the critical evaluation carried out using the STROBE tool.

Buch et al. 2005Great BritainCross- sectional11 682Type 2 DMCaucasian and ethnic minorities6Diabetic retinopathy screening serviceThe Caucasian population went more frequently for examinations than ethnic minorities (P < 0.05).
Sedgwick et al. 2003Great BritainCross- sectional1899Type 2 DMBlack African, afro-Caribbean and Caucasian10Nursing, dietetic and ophthalmic serviceThe Caucasian population had worse access, as afro-Caribbean patients more frequently attended the nursing service (OR = 1.34, CI:95% = 1.04–1.74) and the dietetic service (OR = 1.49, CI:95% = 1.19–1.86). Black African patients also went more frequently to dietetic consultations (OR = 2.15, CI:95% = 1.40–3.29) and ophthalmic services (OR = 1.72, CI:95% = 1.10–2.70).
Ralph-Campbell et al. 2006CanadaCohort394Type 2 DMAboriginal and non-aboriginal10Diabetic retinopathy and nephropathy screening service, number of visits to the doctor, to the emergency department and time spent in hospital because of diabetesNon-aboriginals went more often to check-ups for nephropathy (OR = 2, CI:95% = 1.1–3.9) and diabetic retinopathy (OR = 7.1, CI:95%= 1.1–45.6).
Bruce et al. 2003AustraliaCross- sectional1264Type 2 DMAnglo-Celtic, southern European, other European, Asian, indigenous Australian and other10Education about diabetes and advice on dietIndigenous Australians received less diabetes education and advice on diet (P < 0.05) than the other groups.
Robinson et al. 2006New ZealandCross- sectional5917Type 2 DMEuropean, Maori, Pacific, other Asian, Indian and other10Medical consultationsMaoris more frequently went for medical consultations, whereas the group ‘other Asians’ went less frequently (P < 0.05).
Tomlin et al. 2006New ZealandCross- sectional13 281Type 1 DM and Type 2 DMEurope, Maori and Pacific Islands8Diabetic retinopathy and foot screening serviceType 1 DM: no significant differences were observed. Type 2 DM: European patients went more frequently for foot examinations (P < 0.05) and for diabetic retinopathy examinations (P < 0.05).
Table 5.   Socioeconomic inequalities in the diagnosis of diabetes. Characteristics and results
SourceCountryDesignNType of diabetesSocioeconomic level indicators analysedQuality assessment*Indicator of late diagnosisResults
  1. DM, diabetes mellitus; N, sample size; OR, odds ratio.

  2. *Result of the critical evaluation carried out using the STROBE tool.

Blanc et al. 2003FranceCohort72Type 1 DMAnnual family income7Diabetic ketoacidosis (pH<7)Association between a low income level and greater frequency of severe ketoacidosis at the point of diagnosis (P < 0.05) and of wrong diagnosis (P < 0.01).
Lawlor et al. 2007Great BritainCohort408Type 2 DMOccupational status12Late diagnosisNo significant differences were observed.
Sadauskaite-Kuehne et al. 2002SwedenCross- sectional401Type 1 DMOccupational status and educational level of parents8Diabetic ketoacidosisAssociation between diabetic ketoacidosis and low occupational status (OR = 0.21, CI:95% = 0.08–0.57) and educational status of the mother (OR = 0.40, CI:95% = 0.20–0.79).
Rathmann et al. 2005GermanyCross- sectional1354Type 2 DMLevel of education, income and occupational status8Undiagnosed diabetesAssociation (in women only) between undiagnosed diabetes and low occupational status (OR = 0.5, CI:95% = 0.3–0.8) and low level of income (OR = 0.7, CI:95% = 0.5–1.03).
Table 6.   Socioeconomic inequalities in the control of diabetes measured by clinical variables. Characteristics and results
SourceCountryDesignNType of diabetesSocioeconomic level indicators analysedQuality assessment*Clinical variable studiedResults
  1. DM, diabetes mellitus; BMI, body mass index; HbA1c, glycosylated haemoglobin; N, sample size; OR, odds ratio; SAHRU, small area health research unit; UPA, under-privileged area.

  2. *Result of the critical evaluation carried out using the STROBE tool.

Muhlhauser et al. 1998GermanyCross- sectional684Type 1 DMIncome level, occupational status and level of education9BMI, HbA1c, ketoacidosis and hypoglycaemiaAssociation between low socioeconomic level and high values of HbA1c (P < 0.05) and worse self-monitoring of glycaemia (OR = 1.38, CI:95% = 1.18–1.61).
Reisig et al. 2007GermanyCross- sectional373Type 2 DMIncome level, occupational status and level of education10HbA1c, cholesterolemia and blood pressureAssociation between high values of HbA1c and low income level (OR = 2.49, CI:95% = 1.22–5.07), a lower educational level (OR= 1.80, CI:95% = 0.95–3.43) and worse occupational status (OR = 2.40, CI:95% =0.88–6.60).
Weng et al. 2000Great BritainCohort610Not specifiedUPA index (based on post code)9HbA1cAssociation between low values of HbA1c and BMI and high socioeconomic level (P = 0.003).
Edwards et al. 2003Great BritainCross- sectional5435Type 2 DMTownsend index (based on post code)11Blood pressure, HbA1c, cholesterolemia, BMIAssociation between living in richer areas and low BMI (P = 0.009) and HbA1c levels (P = 0.006).
Lawlor et al. 2007Great BritainCohort408Type 2 DMOccupational status12HbA1c, insulinemia, cholesterolemia, blood pressure and BMIAssociation (in women only) between low occupational status and high levels of insulinemia, triglyceridemia, HDL cholesterol and BMI (P < 0.05).
O’Connor et al. 2006Great BritainCross- sectional1030Type 2 DMSAHRU index (based on post code)1HbA1cNo significant differences were observed.
Larranaga et al. 2005SpainCross- sectional2985Type 2 DMOccupational status, level of education and proportion of family members with a low quality of life12Cholesterol, obesity and BMIAssociation between low socioeconomic level and high levels of HbA1c (OR =1.8, CI:95% = 1.3–2.4), of LDL cholesterol (OR = 1.9, CI:95% = 1.2–3.0) and obesity (OR = 1.7, CI:95% =1.3–2.3).
Table 7.   Socioeconomic inequalities in access to healthcare services related to the control of diabetes. Characteristics and results
SourceCountryDesignNType of diabetesSocioeconomic level indicators analysedQuality assessment*Healthcare servicesResults
  1. DM, diabetes mellitus; IMD, index of multiple deprivation; N, sample size; OR, odds ratio; UPA, under-privileged area.

  2. *Result of the critical evaluation carried out using the STROBE tool.

Muhlhauser et al. 1998GermanyCross-sectional684Type 1 DMIncome level, occupational status and level of education9Educational courses, medical visits and hospital admissionsAssociation between low social status and high level of participation in educational courses on diabetes (OR = 1.29, CI:95% = 1.10–1.50), low frequency of medical visits (OR = 1.36, CI:95% = 1.17–1.56) and less time spent in hospital (P < 0.05).
Larranaga et al. 2005SpainCross-sectional2985Type 2 DMOccupational status, level of education and proportion of family members with a low quality of life12Primary healthcare servicesAssociation between low socioeconomic level and a greater use of primary healthcare services (P < 0.05).
Weng et al. 2000Great BritainCohort610Not specifiedUPA9Medical visitsAssociation between high socioeconomic level and high frequency of medical visits (P < 0.05).
Millett & Dodhia 2006Great BritainCross- sectional8066Type 1 DM and Type 2 DMIMD10Diabetic retinopathy screening serviceAssociation between low socioeconomic level and low attendance of retinopathy examination programmes (OR = 0.58, CI:95% = 0.36–0.95).
Buch et al. 2005Great BritainCross- sectional11 682Type 2 DMIMD6Diabetic retinopathy screening serviceNo significant differences were observed.

Outcomes

Gender inequalities in access to healthcare services related to the control of DM

In Sweden, women with type 1 DM (DM1) went more frequently for medical check-ups and to educational programmes than men, whereas men visited the nursing service more frequently. In the case of patients with type 2 DM (DM2), men visited medical centres more frequently to receive education on the care of DM (Wredling et al. 1998). In Holland, gender inequalities were studied regarding the use of medical services, observing no significant differences (El Fakiri et al. 2003). A study carried out in Great Britain revealed that there were no gender inequalities with respect to adherence to diabetic retinopathy screening services (Millett & Dodhia 2006) (Table 1).

Ethnic inequalities in the treatment of DM

This group included two studies analysed in Great Britain. The first one found that black Caribbean patients were treated more frequently than white patients using antihypertensive treatment (Sedgwick et al. 2003). The second study revealed that living in certain parts of the country with a higher proportion of immigrants was correlated with a lower prescription of oral antidiabetics (Lloyd et al. 2004). Two other studies were carried out in New Zealand. In the first, it was observed that the ethnic group ‘other Asians’, in comparison with Europeans, were treated less with aspirin, with two or more antihypertensives and with insulin (Robinson et al. 2006). In the second, it was found that Maoris and dwellers of the Pacific Islands were treated less frequently with diet and statins than Europeans (Tomlin et al. 2006) (Table 2).

Ethnic inequalities in the control of DM measured by clinical variables

A study carried out in New Zealand noted that Europeans had better glycaemic control compared with the other ethnic groups studied (Tomlin et al. 2006). In France, higher values of glycosylated haemoglobin (HbA1c) were observed in the children of immigrant mothers than in those of native mothers (Tubiana-Rufi et al. 1995). A study carried out in Canada investigated inequalities between aboriginal and non-aboriginal patients in the control of DM, observing that, with adjustments for age, gender, civil status, family income, level of education and duration of the DM, no significant differences existed between the two groups in terms of blood pressure, HbA1c, body mass index (BMI) and total cholesterol (Ralph-Campbell et al. 2006). In a cohort study carried out in Great Britain, it was observed that inequalities in the control of DM between Southern Asians and Europeans decreased over time (McElduff et al. 2005). However, in another study in the same country no significant differences were observed between the ethnic groups in question with respect to plasma glucose or HbA1c (Davis et al. 2001). Also in Great Britain, the study by Kousta (Kousta et al. 2006) highlighted the fact that non-European women with gestational DM presented more difficulties in glycaemic control (P < 0.05). Finally, a study conducted in Australia revealed that aboriginals had significantly higher values of HbA1c than Anglo-Celtic patients (Davis et al. 2007). On the other hand, a study in the same country observed that indigenous people carried out better glycaemic control than Anglo-Celtic patients (Bruce et al. 2003) (Table 3).

Ethnic inequalities in access to healthcare services related to the control of DM

In Great Britain, the Caucasian population visited diabetic retinopathy screening service more frequently than ethnic minorities (Buch et al. 2005). Another study in the same country, observed a greater use of the nursing, dietetic and ophthalmic services among the black African and Afro-Caribbean population compared with Caucasians (Sedgwick et al. 2003). In Canada, aboriginals visited diabetic nephropathy and diabetic retinopathy screening services more frequently than non-aboriginals (Ralph-Campbell et al. 2006). The study by Bruce et al. (2003) in Australia revealed that the indigenous population received less education and advice about DM than other patients. Robinson et al. (2006) found that in New Zealand, Maoris went for more medical consultations than patients in the group ‘other Asians’. Another study in the same country observed that DM2 European patients attended foot examinations and diabetic retinopathy programmes more frequently (Tomlin et al. 2006) (Table 4).

Socioeconomic inequalities in the diagnosis of DM

In a study carried out in France, children with DM1 from low-income families more frequently developed severe ketoacidosis and were more frequently wrongly diagnosed before being admitted to hospital (Blanc et al. 2003). In Great Britain, a cohort study did not find any association between the diagnosis of DM in women aged 60–79 and socioeconomic level (Lawlor et al. 2007). In Sweden, children with DM1 were at greater risk from ketoacidosis at the point of diagnosis if the mother did not work (Sadauskaite-Kuehne et al. 2002). According to Rathmann et al. (2005), in Germany and only in women was there found to be an association between undiagnosed DM and low occupational status and income level (Table 5).

Socioeconomic inequalities in the control of DM measured by clinical variables

A study carried out in Germany found that diabetic patients with a lower socioeconomic level had higher HbA1c values and were worse at glycaemic control (Muhlhauser et al. 1998). Another study in Germany revealed that there is a greater risk of not properly controlling levels of HbA1c in lower income patients, those with a lower level of education and those with lower occupational status (Reisig et al. 2007). In Great Britain, patients with a lower socioeconomic level presented higher HbA1c and BMI values than those with a higher socioeconomic level (Weng et al. 2000). In the same country, patients coming from richer areas had lower BMI (P < 0.05) and HbA1c (P < 0.05) levels than those who lived in more deprived areas (Edwards et al. 2003). Additionally, in Great Britain women with a low occupational status had worse levels of insulinemia, triglyceridemia, HDL cholesterol and BMI (Lawlor et al. 2007). On the other hand, O’Connor et al. (2006) noted that there were no socioeconomic inequalities either in the frequency of measuring of HbA1c or in their values. In Spain, an association was found between a low socioeconomic level and high levels of HbA1c, low-density lipoprotein (LDL) cholesterol and of obesity (Larranaga et al. 2005) (Table 6).

Socioeconomic inequalities in access to healthcare services related to the control of DM

A study conducted in Germany revealed that both male and female patients with lower social status participated more regularly in educational courses on DM, whereas having higher social status was associated with more frequent medical consultations and less time spent in hospital (Muhlhauser et al. 1998). In Spain, there was an association between belonging to a low socioeconomic group and making more frequent visits to primary healthcare services (Larranaga et al. 2005). In Great Britain, it was observed that patients with a high socioeconomic level visited their doctor more frequently (Weng et al. 2000), whereas another study found that there was a connection between low socioeconomic level and a low level of adherence to the diabetic retinopathy screening service (Millett & Dodhia 2006), which was not confirmed in the study by Buch et al. (2005), who did not find significant differences (Table 7).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

This systematic review shows some indications of the existence of social inequalities in DM health-care. The results of the articles analysed point towards the existence of ethnic inequalities in the treatment and control of DM measured by clinical criteria as well as socioeconomic inequalities in the diagnosis and control of the disease. Additionally, a high methodological quality cohort study in Canada confirmed the existence of ethnic inequalities in the use of healthcare services.

Specifically, the findings of this review show that there is no scientific evidence of the existence of gender inequalities. Only three articles were found on this subject, and they were of moderate methodological quality and had a cross-sectional design. They therefore provided a low level of scientific evidence, so we cannot conclude that gender inequalities exist in terms of access to DM healthcare services, although it cannot be ruled out. This lack of scientific evidence regarding the possible existence of gender inequalities in diabetes is surprising, as it has been observed that even in countries with a universal healthcare system such as Canada (Janzen 1998, Denton et al. 2004) or Great Britain (Macintyre et al. 1996, Arber 1997), women have lower levels of perceived health, lower quality of life, and do not have such good access to healthcare services.

On the other hand, this review has found evidence about ethnic inequalities in the treatment and health-care, which would be consistent with the findings of the existing literature, as it is recognised that in general, ethnic minority groups receive less medical care for their health problems (Boyd et al. 2005, Vázquez et al. 2006, Dorling et al. 2007, Knight et al. 2009). However, it is possible that the differences observed in this study reflect in part socioeconomic inequalities, for as in the majority of the studies reviewed the necessary adjustments depending on socioeconomic level in order to minimise the magnitude of the differences between ethnic groups, have not been made.

Finally, the results of this study indicate that DM sufferers with a low socioeconomic level suffer from a greater delay in diagnosis and are worse at controlling their condition. If this is the case, it would be consistent with the findings of the existing literature, both in terms of patients with other chronic diseases such as cancer (Woods et al. 2006, Dejardin et al. 2008), HIV (Giuliano & Vella 2007, Tsai et al. 2009) or mental illnesses (Bonizzato & Tello 2003, Muntaner et al. 2003, Amaddeo & Jones 2007), and also in terms of general health problems (Mackenbach & Bakker 2003, Marmot et al. 2008, Knight et al. 2009).

The fact that the existence of inequalities is not entirely consistent may be because studies with different epidemiological designs and outcome measures were included in this review. However, the research focus of this systematic review meant that the results of the articles had to be combined based on the axis of inequality and healthcare type. As such, categorising the results based on epidemiological design as well would have created numerous groups, each with a low number of articles. Furthermore, it is important to take into account that the lack of consistency may also be due to differences between the healthcare systems in the different countries studied, as the fact that they all belong to the OECD and offer universal healthcare services does not mean that these systems are exactly the same. Moreover, there are other limitations that we would like to point out. In a systematic literature review, the results are based on the findings of studies which have already been concluded, and such being the case it is subject to the limitations of quality and quantity of these studies. Concerning the analysis of the results of a systematic review, it is desirable to make a quantitative synthesis of the results of these studies. However, the heterogeneity observed among the selected studies prevented reaching a conclusive synthesis of the results. Thus, although a greater number of studies investigate ethnic and socioeconomic inequalities, these can be problematic in the sense that they have different criteria for defining the concept of ethnicity (understood as race or as country of origin) and socioeconomic level (which is calculated from occupational status, educational level or level of family income), which makes comparison of the results of the studies difficult. Moreover, although it is generally assumed that systematic reviews may be subject to publication bias (papers showing differences are most frequently published), in the field of inequalities we should consider whether the bias is in this sense or in the opposite, given that studies showing inequalities in health might have been rejected due to a lack of reliability on the results. Therefore, in this case it seems difficult to establish whether what has been published underestimates or overestimates the presence of inequalities. Furthermore, the fact that the review included studies with patients with type 1 and patients with type 2 diabetes may be a limitation because of the differences between the two diseases. However, as some of the studies found examined both types of diabetes at the same time (Millett & Dodhia 2006, Tomlin et al. 2006) or did not specify the type of diabetes studied (Weng et al. 2000, McElduff et al. 2005), it was impossible to evaluate them separately. Finally, another possible limitation of the study could be that a critical assessment was only carried out in the event of finding at least three articles sharing the same research question (as it is difficult to draw conclusions from a small number of articles) which, nevertheless, leads to a loss of information. On the other hand, we do not consider the use of some STROBE items as a limitation. There are no validated scales to evaluate observational studies and, as Sanderson et al. 2007 point out, the STROBE is the scale that includes the most adequate items to evaluate the methodological quality of observational research.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References

This systematic review shows that even in countries with a significant level of economic development and which have universal healthcare systems in place and are endeavouring to provide medical care to the entire population, socioeconomic and ethnic inequalities can be identified in the provision of health-care to DM sufferers. Nonetheless, more high quality and follow-up articles are needed to confirm these results.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion
  8. References
  • Amaddeo F. & Jones J. (2007) What is the impact of socio-economic inequalities on the use of mental health services? Epidemiologia e Psichiatria Sociale 16, 1619.
  • Arber S. (1997) Comparing inequalities in women’s and men’s health: Britain in the 1990s. Social Science & Medicine 44, 773787.
  • Blanc N., Lucidarme N. & Tubiana-Rufi N. (2003) Factors associated with childhood diabetes manifesting as ketoacidosis and its severity. Archives of Pediatrics 10, 320325.
  • Bonizzato P. & Tello J.E. (2003) Socio-economic inequalities and mental health. I. Concepts, theories, and interpretations. Epidemiologia e Psichiatria Sociale 12, 205218.
  • Boyd A.E., Murad S., O’shea S., De Ruiter A., Watson C. & Easterbrook P.J. (2005) Ethnic differences in stage of presentation of adults newly diagnosed with HIV-1 infection in south London. HIV Medicine 6, 5965.
  • Brown A.F., Ettner S.L., Piette J. et al. (2004) Socioeconomic position and health among persons with diabetes mellitus: a conceptual framework and review of the literature. Epidemiologic Reviews 26, 6377.
  • Bruce D.G., Davis W.A., Cull C.A. & Davis T.M.E. (2003) Diabetes education and knowledge in patients with type 2 diabetes from the community: the Fremantle Diabetes Study. Journal of Diabetes and its Complications 17, 8289.
  • Buch H.N., Barton D.M., Varughese G.I., Bradbury S., Scarpello J.H. & Walker A.B. (2005) An assessment of the coverage of a district-wide diabetic retinopathy screening service. Diabetic Medicine 22, 840841.
  • Davis T.M., Cull C.A. & Holman R.R. (2001) Relationship between ethnicity and glycemic control, lipid profiles, and blood pressure during the first 9 years of type 2 diabetes: UK prospective diabetes study (UKPDS 55). Diabetes Care 24, 11671174.
  • Davis T.M., McAullay D., Davis W.A. & Bruce D.G. (2007) Characteristics and outcome of type 2 diabetes in urban Aboriginal people: the Fremantle Diabetes Study. Internal Medicine Journal 37, 5963.
  • Dejardin O., Bouvier A.M., Faivre J., Boutreux S., De Pouvourville G. & Launoy G. (2008) Access to care, socioeconomic deprivation and colon cancer survival. Alimentary Pharmacology & Therapeutics 27, 940949.
  • Del Prato S., Horton E. & Nesto R. (2007) We have the evidence, we need to act to improve diabetes care 105. International Journal of Clinical Practice 61, 915.
  • Denton M., Prus S. & Walters V. (2004) Gender differences in health: a Canadian study of the psychosocial, structural and behavioural determinants of health. Social Science & Medicine 58, 25852600.
  • Dorling D., Mitchell R. & Pearce J. (2007) The global impact of income inequality on health by age: an observational study. BMJ 335, 873875.
  • Eakin E.G., Bull S.S., Glasgow R.E. et al. (2002) Reaching those most in need: a review of diabetes self-management interventions in disadvantaged populations. [Review]. Diabetes/Metabolism Research Reviews 18, 2635.
  • Edwards R., Burns J.A., McElduff P., Young R.J. & New J.P. (2003) Variations in process and outcomes of diabetes care by socio-economic status in Salford, UK. Diabetologia 46, 750759.
  • El Fakiri F., Foets M. & Rijken M. (2003) Health care use by diabetic patients in the Netherlands: patterns and predicting factors. Diabetes Research and Clinical Practice 61, 199209.
  • Von Elm E., Altman D.G., Egger M., Pocock S.J., G°tzsche P.C. & Vandenbroucke J.P. (2007) The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. Preventive Medicine 45, 247251.
  • Giuliano M. & Vella S. (2007) Inequalities in health: access to treatment for HIV/AIDS. Annali dell’Istituto Superiore di Sanita 43, 313316.
  • Gan D. (2006) Diabetes Atlas, 3rd edn. International Diabetes Federation, Brussels.
  • Janzen B.L. (1998) Women, gender and health: A review of recent literature. Prairie Women’s Health Centre of Excellence, Winnipeg, Canada.
  • Klinke J.A., Johnson J.A., Guirguis L.M., Toth E.L., Lee T.K., Lewanczuk R.Z. & Majumdar S.R. (2004) Underuse of aspirin in type 2 diabetes mellitus: prevalence and correlates of therapy in rural Canada. Clinical Therapeutics 26, 439446.
  • Knight M., Kurinczuk J.J., Spark P. & Brocklehurst P. (2009) Inequalities in maternal health: national cohort study of ethnic variation in severe maternal morbidities. BMJ 338, b542.
  • Kousta E., Efstathiadou Z., Lawrence N.J. et al. (2006) The impact of ethnicity on glucose regulation and the metabolic syndrome following gestational diabetes. Diabetologia 49, 3640.
  • Van Laar J.J., Grishchenko M., Van Wouwe J.P. & Stronks K. (2007) Ethnic differences in the timely diagnosis of children with Type 1 diabetes mellitus in the Netherlands: clinical presentation at onset. Diabetic Medicine 24, 296302.
  • Larranaga I., Arteagoitia J.M., Rodriguez J.L., Gonzalez F., Esnaola S. & Pinies J.A. (2005) Socio-economic inequalities in the prevalence of Type 2 diabetes, cardiovascular risk factors and chronic diabetic complications in the Basque Country, Spain. Diabetic Medicine 22, 10471053.
  • Lawlor D.A., Patel R., Fraser A., Smith G.D. & Ebrahim S. (2007) The association of life course socio-economic position with diagnosis, treatment, control and survival of women with diabetes: findings from the British Women’s Heart and Health Study. Diabetic Medicine 24, 892900.
  • Lloyd D.C., Scrivener G., Robinson T. et al. (2004) Determinants of prescribing patterns for type II diabetes at Primary Care Trust level. Ethnic differences in Type 2 diabetes care and outcomes in Auckland: a multiethnic community in New Zealand. Journal of Clinical Pharmacy and Therapeutics 29, 389394.
  • Macintyre S., Hunt K. & Sweeting H. (1996) Gender differences in health: are things really as simple as they seem? Social Science & Medicine 42, 617624.
  • Mackenbach J.P. & Bakker M.J. (2003) Tackling socioeconomic inequalities in health: analysis of European experiences. The Lancet 362, 14091414.
  • Marmot M. (2008) Social determinants of health inequalities. The Lancet 365, 10991104.
  • Marmot M., Friel S., Bell R., Houweling T.A. & Taylor S. (2008) Closing the gap in a generation: health equity through action on the social determinants of health. The Lancet 372, 16611669.
  • McElduff P., Edwards R., Burns J.A. et al. (2005) Comparison of processes and intermediate outcomes between South Asian and European patients with diabetes in Blackburn, north-west England. Diabetic Medicine 22, 12261233.
  • Meisinger C., Lowel H., Thorand B. & Doring A. (2005) Leisure time physical activity and the risk of type 2 diabetes in men and women from the general population. The MONICA/KORA Augsburg Cohort Study. Diabetologia 48, 2734.
  • Millett C. & Dodhia H. (2006) Diabetes retinopathy screening: audit of equity in participation and selected outcomes in South East London. Journal of Medical Screening 13, 152155.
  • Muhlhauser I., Overmann H., Bender R. et al. (1998) Social status and the quality of care for adult people with type I (insulin-dependent) diabetes mellitus – a population-based study. Diabetologia 41, 11391150.
  • Muntaner C., Borrell C., Benach J., Pasarin M.I. & Fernandez E. (2003) The associations of social class and social stratification with patterns of general and mental health in a Spanish population. International Journal of Epidemiology 32, 950958.
  • O’Connor R., Houghton F., Saunders J. & Dobbs F. (2006) Diabetes mellitus in Irish general practice: level of care as reflected by HbA1c values. The European Journal of General Practice 12, 5865.
  • Olmos M., Antelo M., Vazquez H., Smecuol E., Maurino E. & Bai J.C. (2008) Systematic review and meta-analysis of observational studies on the prevalence of fractures in coeliac disease. Digestive and Liver Disease 40, 4653.
  • Peyrot M., Rubin R.R., Lauritzen T., Skovlund S.E., Snoek F.J., Matthews D.R. & Landgraf R. (2006) Patient and provider perceptions of care for diabetes: results of the cross-national DAWN Study. Diabetologia 49, 279288.
  • Ralph-Campbell K., Pohar S.L., Guirguis L.M. & Toth E.L. (2006) Aboriginal participation in the DOVE study. Canadian Journal of Public Health 97, 305309.
  • Rathmann W., Haastert B., Icks A., Giani G., Holle R., Meisinger C. & Mielck A. (2005) Sex differences in the associations of socioeconomic status with undiagnosed diabetes mellitus and impaired glucose tolerance in the elderly population: the KORA Survey 2000. European Journal of Public Health 15, 627633.
  • Reisig V., Reitmeir P., Doring A., Rathmann W. & Mielck A. (2007) Social inequalities and outcomes in type 2 diabetes in the German region of Augsburg. A cross-sectional survey. International Journal of Public Health 52, 158165.
  • Robbins J.M., Vaccarino V., Zhang H. & Kasl S.V. (2001) Socioeconomic status and type 2 diabetes in African American and non-Hispanic white women and men: evidence from the Third National Health and Nutrition Examination Survey. American Journal of Public Health 91, 7683.
  • Robinson T., Simmons D., Scott D. et al. (2006) Ethnic differences in Type 2 diabetes care and outcomes in Auckland: a multiethnic community in New Zealand. The New Zealand Medical Journal 119, U1997.
  • Sadauskaite-Kuehne V., Samuelsson U., Jasinskiene E., Padaiga Z., Urbonaite B., Edenvall H. & Ludvigsson J. (2002) Severity at onset of childhood type 1 diabetes in countries with high and low incidence of the condition. Diabetes Research and Clinical Practice 55, 247254.
  • Sanderson S., Tatt I.D. & Higgins J.P. (2007) Tools for assessing quality and susceptibility to bias in observational studies in epidemiology: a systematic review and annotated bibliography. International Journal of Epidemiology 36, 666676.
  • Scales C.D. Jr & Dahm P. (2008) The critical use of population-based medical databases for prostate cancer research. Current Opinion in Urology 18, 320325.
  • Sedgwick J.E., Pearce A.J. & Gulliford M.C. (2003) Evaluation of equity in diabetes health care in relation to African and Caribbean ethnicity. Ethnicity & Health 8, 121133.
  • Tomlin A., Tilyard M., Dawson A. & Dovey S. (2006) Health status of New Zealand European, Maori, and Pacific patients with diabetes at 242 New Zealand general practices. The New Zealand Medical Journal 119, U2004.
  • Tsai A.C., Chopra M., Pronyk P.M. & Martinson N.A. (2009) Socioeconomic disparities in access to HIV/AIDS treatment programs in resource-limited settings. AIDS Care 21, 5963.
  • Tubiana-Rufi N., Moret L., Czernichow P. & Chwalow J. (1995) Risk factors for poor glycemic control in diabetic children in France. Diabetes Care 18, 14791482.
  • Vázquez V., Espejo J. & Faus M.J. (2006) Determinación de las necesidades en atención farmacéutica de los inmigrantes en el campo de Gibraltar. Pharmacy Practice 4, 2433.
  • Weng C., Coppini D.V. & Sonksen P.H. (2000) Geographic and social factors are related to increased morbidity and mortality rates in diabetic patients. Diabetic Medicine 17, 612617.
  • Woods L.M., Rachet B. & Coleman M.P. (2006) Origins of socio-economic inequalities in cancer survival: a review. Annals of Oncology 17, 519.
  • Wredling R., Adamson U., Larsson Y. & Ericsson A. (1998) Are diabetic men and women treated equally? A population-based study. Diabetes, Nutrition & Metabolism – Clinical & Experimental 11, 816.
  • Scottish Intercollegiate Guidelines Network (2010) Key to Evidence Statements and Grades of Recommendations. Available from: http://www.sign.ac.uk/guidelines/fulltext/50/annexb.html [Accessed 6 September 2010]