DIABETICMedicine DOI: 10.1111/dme.12529
Short Report: Epidemiology The association between socio-demographic marginalization and plasma glucose levels at diagnosis of gestational diabetes L. Sampson1, K. Dasgupta2 and N. A. Ross1 1 Department of Geography, McGill University and 2Division of Clinical Epidemiology, Department of Medicine, McGill University Health Centre, Montreal, QC, Canada
Accepted 20 June 2014
Abstract Aims We examined the association between socio-demographic marginalization and plasma glucose levels at diagnosis of gestational diabetes in a multi-ethnic and socio-economically diverse patient group. Methods Medical charts at a Toronto gestational diabetes clinic were reviewed for women with a recorded pregnancy between 1 March 2006 and 26 April 2011. One-hour 50-g glucose challenge test values and postal code data were abstracted. Postal codes were merged with 2006 Canadian census data to compute neighbourhood-level ethnic concentration (% recent immigrants, % visible minorities) and material deprivation (% low education, % low income, single-parent households). We compared women in the highest neighbourhood quintiles for both ethnic concentration and material deprivation with all other women to explore an association between marginalization and diagnostic glucose levels. Multivariate regression models of glucose challenge test values and insulin prescription were adjusted for age, prior gestational diabetes, parity and diabetes family history.
Among 531 patients with complete glucose challenge test data (mean 11.94 mmol/l, SD 1.83), those in the most marginalized neighbourhoods had 0.43 mmol/l higher glucose challenge test values (95% CI 0.08–0.78) compared with the rest of the study population. Other factors associated with higher glucose challenge test values were prior gestational diabetes (0.59 mmol/l increment, 95% CI 0.19–0.99) and diabetes family history (0.32 mmol/l increment, 95% CI –0.01 to 0.66). Each additional 1 mmol/l glucose challenge test result was associated with an increased likelihood of being prescribed insulin (odds ratio 1.33, 95% CI 1.17–1.51).
Results
Conclusions Women living in the most materially deprived and ethnically concentrated neighbourhoods have higher glucose levels at diagnosis of gestational diabetes. They may need close monitoring for timely initiation of insulin.
Diabet. Med. 31, 1563–1567 (2014)
Introduction Gestational diabetes mellitus is defined as glucose intolerance (hyperglycaemia) first recognized during pregnancy [1]. This condition affects approximately 7% of all pregnancies [2] and higher plasma glucose levels at diagnosis are associated with increased risk of complications [3]. Greater understanding of factors that contribute to elevated plasma glucose levels at diagnosis may help to shape practice aimed at lowering the severity of gestational diabetes hyperglycaemia. There is evidence that age, obesity, parity, family history of diabetes and ethnicity are risk factors for gestational diabetes [1,4]. Ethnocultural background and socio-economic status Correspondence to: Nancy A. Ross. E-mail: [email protected]
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
affect incidence of gestational diabetes [2,5]; women with low income have more than twice the risk of developing gestational diabetes than more affluent women [6]. Little is known about the influence of socio-economic status on levels of hyperglycaemia at diagnosis, or the effects of socio-economic status on glycaemic levels once gestational diabetes has been diagnosed. Material resources unequally distributed by race and gender, are stratified by place [7], and place-based characteristics are modifiable amplifiers to health inequalities [8]. Marginalization refers to the patterning of inadequate and unequal access to political, social and economic benefits and exclusion from full societal participation attributable to membership in an identifiably less privileged group [9]. This study assesses the association between marginalization and plasma glucose levels
1563
DIABETICMedicine
What’s new? ● This study demonstrates an association between residential socio-demographic marginalization and elevated levels of plasma glucose at time of gestational diabetes diagnosis in a multicultural urban Canadian population. ● We report a relationship between elevated glucose values at the time of gestational diabetes diagnosis and likelihood of eventual insulin prescription. ● Linking elevated diagnostic test values to patients living on a low income and in ethnic minority neighbourhoods may provide clinicians with an extended risk profile and indicate a need for timely initiation of insulin. at diagnosis of gestational diabetes (1-h 50-g, glucose challenge test) and insulin prescription amongst a multi-ethnic and socio-economically diverse urban population.
Subjects and methods Electronic patient charts from the Toronto East General Hospital Gestational Diabetes Clinic were reviewed for patients referred to the clinic between 1 March 2006 and 26 April 2011. Women were retained if they had a documented 50-g glucose challenge test, demonstrated fulfilment of gestational diabetes criteria [1] and had a valid 6-digit postal code. The index pregnancy was the most recent pregnancy. Medical variables available included age, diagnosis (type of diabetes), reported diagnostic test, parity, gestational diabetes in a previous pregnancy, family history of diabetes, and insulin prescription at any time during follow-up.
Glucose challenge test
Glucose level at 1 h following a 50-g glucose load (glucose challenge test) was recorded (according to 2008 Canadian guidelines). The study sample and cases excluded because of a missing glucose challenge test were similar socio-demographically (data available from authors).
Insulin prescription
Insulin prescription was recorded as a patient being prescribed insulin by their physician during the index pregnancy at any time following diagnosis of gestational diabetes.
Material deprivation and ethnic concentration
Marginalization indicators, material deprivation and ethnic concentration were obtained from the Ontario Marginalization Index (ON-Marg) [10]. Material deprivation is derived
1564
Marginalization and gestational diabetes L. Sampson et al.
from census indicators measuring the proportion of residents with low education, low income, receiving government transfer payments, who are unemployed, single-parent families and with dwellings in need of repair. Ethnic concentration is measured by proportion of new immigrants (< 5 years) and proportion of the population who self-identify as a visible minority (or racialized group). Dissemination area-level factor scores and quintiles for these dimensions of marginalization were assigned to individual subjects via 6-digit postal codes. Postal codes were translated into dissemination areas using the Statistics Canada Postal Code Conversion File [11]. The socio-demographic marginalization data were applied to the patients. The marginalization indicators were collinear (Pearson 0.48, P ≤ 0.001). They were combined as a composite, binary variable: Deprivation/ethnic concentration. Women were coded 1 if they lived in a dissemination area classified as the highest quintile (5 = most marginalized) in both Material deprivation and Ethnic concentration, and 0 if they did not. Multivariate linear regression models were used to examine the association between marginalization and 50-g glucose challenge test diagnosis, while accounting for known risk factors (age, parity, gestational diabetes in previous pregnancy and family history). Logistic regression models using two-tailed significance tests are reported for insulin prescription and conservative reliable estimates were recorded. Breusch–Pagan/Cook–Weisberg, variance inflation factor (VIF) scores and F-statistic tests were carried out (Stata version 12.0; StataCorp., College Station, TX, USA).
Results Five hundred and thirty-one women were included in the final analyses. The mean glucose challenge test was 11.9 mmol/l, with values ranging from 8.3 to 26.2 mmol/l. Subjects were 32 years old on average with a mean of one prior delivery. Approximately half had a family history of diabetes, 75% of subjects had no personal history of gestational diabetes in previous pregnancies and 185 patients presented as primigravida (35% of the study population). One hundred and fifty-six women (30%) lived in the most marginalized neighbourhoods (Q5 for both material deprivation and ethnic concentration) and this group had a mean glucose challenge test value of 12.23 mmol/l (95% CI 11.86– 12.60). The remaining 375 women (70%) were classified as not living in the most marginalized neighbourhoods (Q1– Q4) and had a mean glucose challenge test value of 11.81 mmol/l (95% CI 11.65–11.97). Of the 527 patient charts that reported treatment information, approximately half of the women were prescribed insulin, while the remaining women controlled their blood sugar levels through diet and exercise. More women in the most marginalized neighbourhoods were prescribed insulin (54% vs. 43%).
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
Patients lived in 307 unique dissemination areas. The median household income was $48 327, with an unemployment rate of 9% [12]. For context, the Canadian median household income was $66 650 and the unemployment rate was 6.4% [12]. The average study neighbourhood was comprised of a 54% immigrant population (predominantly South Asian and Chinese) and the majority of patients lived in areas with high levels of ethnic concentration (Q4 and Q5). In multivariate analyses, women living in the most marginalized neighbourhoods had 0.43 mmol/l higher glucose challenge test values (95% CI 0.08–0.78) compared with other patients (Table 1). Having a diagnosis of gestational diabetes in a previous pregnancy was associated with an increase of 0.59 mmol/l (95% CI 0.19–1.00) in glucose challenge test values for the index pregnancy. Patients with a family history of diabetes had a 0.32 mmol/l (95% CI –0.01 to 0.67) increase in glucose challenge test results.
Within the neighbourhoods classified as the most highly ethnically concentrated (Q5), we tested the range of glucose challenge test values across material deprivation quintiles (Fig. 1). Although there is a linear trend, the quintile that is conclusively different from the other quintiles is Q5 (most materially deprived). In Q5, the mean glucose challenge test value was 12.23 mmol/l (95% CI 11.86–12.60), which is statistically different from the mean glucose challenge test of 11.49 mmol/l in Q1 (95% CI 11.21–11.77). Gestational diabetes hyperglycaemia at diagnosis and gestational diabetes in a previous pregnancy were found to be associated with insulin prescription (odds ratio 1.33, 95% CI 1.17–1.51 and odds ratio 1.96, 95% CI 1.23–3.11, respectively). Living in areas of high ethnic deprivation and high ethnic concentration was associated with increased odds of being prescribed insulin of 44% (95% CI 0.95–2.17), but this result was not conclusive.
Table 1 Glucose challenge test (50-g) and insulin prescription
50-g glucose challenge test value at diagnosis (mmol/l) Age (years) Gestational diabetes in previous pregnancy Parity Family history of diabetes Marginalization§
Model 1 (n = 531)
Model 2 (n = 527)
Linear regression: 50-g glucose challenge test
Logistic regression: Insulin prescription
Incremental increase in 50-g glucose challenge test value (mmol/l)
95% CI
Odds ratio
95% CI
—
1.33‡
1.17–1.51
–0.01 to 0.04 0.19–0.99 –0.26 to 0.08 –0.01 to 0.66 0.08–0.78
1.02 1.96† 1.04 1.11 1.44
0.98–1.06 1.23–3.11 0.85–1.27 0.75–1.64 0.95–2.17
— ~ 0.00 0.59† –0.09 0.32* 0.43†
*P ≤ 0.05; †P ≤ 0.010; ‡P ≤ 0.001, two-tailed tests. §Living in dissemination area with co-occurrence of quintile 5 for both material deprivation and ethnic concentration (Ontario Marginalization Index).
FIGURE 1 Glucose challenge test values by material deprivation for patients living in the most ethnically concentrated (quintile 5) dissemination areas.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
1565
DIABETICMedicine
Discussion Women with gestational diabetes living in neighourhoods at extremes of both ethnic concentration and material deprivation had higher glucose levels 1 h following a glucose challenge test compared with those living in less marginalized circumstances, in models adjusting for age, previous gestational diabetes, parity and diabetes family history. Previous gestational diabetes and family diabetes history were also associated with higher 1-h post-load glucose levels. Elevated glucose challenge test values at diagnosis were found to be associated with increased odds of insulin prescription and gestational diabetes in a previous pregnancy, in a model adjusted for age, parity and family history of diabetes. Our findings signal that marginalization is associated with glucose values at diagnosis of gestational diabetes and may flag a need for insulin management. Our findings are consistent with three studies that have demonstrated the influence of place-based social conditions on elevated gestational diabetes risk [13–15]. Feig et al. found that women living in urban rather than rural areas and women from lower-income neighbourhoods were at increased risk for gestational diabetes [15]. A Norwegian-based study similarly reported women from areas of lower socio-economic status (Oslo East) had increased gestational diabetes risk (odds ratio 1.6) [14]. An Australian study used maternal origin as a proxy for ethnicity and found that South Asian women had the highest adjusted odds ratio of any region [13]. The same study reported that women living in the three lowest socio-economic quartiles had higher gestational diabetes incidence relative to women in the higher quartile [13]. Our study builds on these prior findings, demonstrating that marginalization is also associated with the magnitude of glucose elevation. Studies have linked previous diagnosis of gestational diabetes with elevated likelihood of future gestational diabetes [15–17], but, to our knowledge, ours is the first to demonstrate a link with elevated plasma glucose levels during later gestational diabetes pregnancies. Similarly, while prior studies have established an association between family history of Type 2 diabetes and gestational diabetes [1,4], we have determined a family history of diabetes to be associated with a 0.32 mmol/l higher glucose challenge test.
Marginalization and gestational diabetes L. Sampson et al.
[19], obesity could have mediated our findings. Indeed, a study by Cullinan et al. found a strong socio-economic gradient in the prevalence of gestational diabetes in Ireland, some of which they argue is driven by behavioural risk factors such as personal, clinical and lifestyle factors that vary across socio-economic groups [20]. While much of the study population lived in neighbourhoods composed of ethnocultural groups known to have increased diabetes risk, such as South Asian and Chinese, we demonstrate that a socio-economic gradient within those higher risk categories persists. This suggests that material deprivation has an association with gestational diabetes above and beyond ethnocultural predisposition.
Conclusions Neighbourhoods that concentrate minority and racialized populations in situations of material deprivation are not unique to Toronto. Minority populations that cluster together in poverty are emblematic of many urban environments, which should provide clinicians with an extended risk profile for people who reside in neighbourhoods similar to those in this study. Only through identifying these at-risk populations can we aim to not only prevent gestational diabetes, but also recommend increased post-diagnosis follow-up visits to ensure timely initiation of insulin and reduction of long-term complications.
Funding sources
This work was supported by Canadian Institutes of Health Research (CIHR) Catalyst Grant: Population and Public Health to KD and NAR (co-principal investigators) Exploring Hypotheses Linking Type II Diabetes and Neighbourhood Environments (no. 94431) and CIHR Interdisciplinary Capacity Enhancement (ICE) Grant, Health Disparities: Intergenerational Dynamics and Empirical Assessment Strategies (no. HOA-80072).
Competing interests
None declared.
Acknowledgements Limitations and strengths
The main limitation of our study was the inconsistency of physician reporting in the patient hospital charts, which prevented us from capturing some potentially important covariates. These included individual-level ethnicity, socio-economic status, smoking status and pre-pregnancy weight. Obesity has been identified as a major modifiable risk factor associated with gestational diabetes [18]. As poverty and obesity are linked, particularly for Canadian women
1566
The authors would like to thank Lisa Sparrow at the Toronto East General Hospital (TEGH) for her insight into the study population.
References 1 Thompson D, Capes S, Feig DS, Kader T, Keeley E, Kozak S et al. Diabetes and Pregnancy. In: Committee CDACPGE ed. Canadian Diabetes Association 2008 Clinical Practice Guidelines for the Prevention and Management of Diabetes in
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
2
3
4
5 6
7
8 9 10
11
Canada. Toronto: Canadian Journal of Diabetes, 2008: S186– S180. Nomura Y, Marks DJ, Grossman B, Yoon M, Loudon H, Stone et al. Exposure to gestational diabetes mellitus and low socioeconomic status: effects on neurocognitive development and risk of attention-deficit/hyperactivity disorder in offspring. Arch Pediatr Adolesc Med 2012; 166: 337–343. Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, Coustan DR et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008; 358: 1991–2002. Buchanan TA, Xiang AH, Page KA. Gestational diabetes mellitus: risks and management during and after pregnancy. Nat Rev Endocrinol 2012; 8: 639–649. Ferrara A. Increasing prevalence of gestational diabetes mellitus: a public health perspective. Diabetes Care 2007; 30: S141–S146. Joseph KS, Liston RM, Dodds L, Dahlgren L, Allen AC. Socioeconomic status and perinatal outcomes in a setting with universal access to essential health care services. CMAJ 2007; 177: 583–590. Massey DS. The age of extremes: concentrated affluence and poverty in the twenty-first century. Demography 1996; 33: 395– 413. Roux AVD, Mair C. Neighborhoods and health. Ann N Y Acad Sci 2010; 1186: 125–145. Spitzer DL. Engendering health disparities. Can J Public Health 2005; 96: S78–S96. Matheson FI, Dunn J, Smith KLW, Moineddin R, Glazier RH. ON-Marg Ontario Marginalization Index. 1.0 edn. Toronto: Collaboratory for Research on Urban Neighbourhoods, Community Health and Housing, 2012. Holowaty EJ, Norwood TA, Wanigaratne S, Abellan JJ, Beale L. Feasibility and utility of mapping disease risk at the neighbourhood
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
DIABETICMedicine
12 13
14
15
16
17
18
19
20
level within a Canadian public health unit: an ecological study. Int J Health Geogr 2010; 9: 21. Statistics Canada. 2006 Canadian Census. Ottawa: Statistics Canada, 2006. Anna V, van der Ploeg HP, Cheung NW, Huxley RR, Bauman AE. Sociodemographic correlates of the increasing trend in prevalence of gestational diabetes mellitus in a large population of women between 1995 and 2005. Diabetes Care 2008; 31: 2288–2293. Clausen T, Oyen N, Henriksen T. Pregnancy complications by overweight and residential area. A prospective study of an urban Norwegian cohort. Acta Obstet Gynecol Scand 2006; 85: 526–533. Feig DS, Zinman B, Wang X, Hux JE. Risk of development of diabetes mellitus after diagnosis of gestational diabetes. CMAJ 2008; 179: 229–234. Kim C, Newton KM, Knopp RH. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care 2002; 25: 1862–1868. Lauenborg J, Hansen T, Jensen DM, Vestergaard H, Molsted-Prdersen L, Hornnes P et al. Increasing incidence of diabetes after gestational diabetes. Diabetes Care 2004; 27: 1194–1199. Hedderson MM, Williams MA, Holt VL, Weiss NS, Ferrara A. Body mass index and weight gain prior to pregnancy and risk of gestational diabetes mellitus. Am J Obstet Gynecol 2008; 198: e401–e407. Matheson FI, Moineddin R, Glazier RH. The weight of place: a multilevel analysis of gender, neighborhood material deprivation, and body mass index among Canadian adults. Soc Sci Med 2008; 66: 675–690. Cullinan J, Gillespie P, Owens L, Avalos G, Dunne FP. Is there a socioeconomic gradient in the prevalence of gestational diabetes mellitus? Ir Med J 2012; 105: 21–23.
1567
Copyright of Diabetic Medicine is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
Short Report: Epidemiology The association between socio-demographic marginalization and plasma glucose levels at diagnosis of gestational diabetes L. Sampson1, K. Dasgupta2 and N. A. Ross1 1 Department of Geography, McGill University and 2Division of Clinical Epidemiology, Department of Medicine, McGill University Health Centre, Montreal, QC, Canada
Accepted 20 June 2014
Abstract Aims We examined the association between socio-demographic marginalization and plasma glucose levels at diagnosis of gestational diabetes in a multi-ethnic and socio-economically diverse patient group. Methods Medical charts at a Toronto gestational diabetes clinic were reviewed for women with a recorded pregnancy between 1 March 2006 and 26 April 2011. One-hour 50-g glucose challenge test values and postal code data were abstracted. Postal codes were merged with 2006 Canadian census data to compute neighbourhood-level ethnic concentration (% recent immigrants, % visible minorities) and material deprivation (% low education, % low income, single-parent households). We compared women in the highest neighbourhood quintiles for both ethnic concentration and material deprivation with all other women to explore an association between marginalization and diagnostic glucose levels. Multivariate regression models of glucose challenge test values and insulin prescription were adjusted for age, prior gestational diabetes, parity and diabetes family history.
Among 531 patients with complete glucose challenge test data (mean 11.94 mmol/l, SD 1.83), those in the most marginalized neighbourhoods had 0.43 mmol/l higher glucose challenge test values (95% CI 0.08–0.78) compared with the rest of the study population. Other factors associated with higher glucose challenge test values were prior gestational diabetes (0.59 mmol/l increment, 95% CI 0.19–0.99) and diabetes family history (0.32 mmol/l increment, 95% CI –0.01 to 0.66). Each additional 1 mmol/l glucose challenge test result was associated with an increased likelihood of being prescribed insulin (odds ratio 1.33, 95% CI 1.17–1.51).
Results
Conclusions Women living in the most materially deprived and ethnically concentrated neighbourhoods have higher glucose levels at diagnosis of gestational diabetes. They may need close monitoring for timely initiation of insulin.
Diabet. Med. 31, 1563–1567 (2014)
Introduction Gestational diabetes mellitus is defined as glucose intolerance (hyperglycaemia) first recognized during pregnancy [1]. This condition affects approximately 7% of all pregnancies [2] and higher plasma glucose levels at diagnosis are associated with increased risk of complications [3]. Greater understanding of factors that contribute to elevated plasma glucose levels at diagnosis may help to shape practice aimed at lowering the severity of gestational diabetes hyperglycaemia. There is evidence that age, obesity, parity, family history of diabetes and ethnicity are risk factors for gestational diabetes [1,4]. Ethnocultural background and socio-economic status Correspondence to: Nancy A. Ross. E-mail: [email protected]
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
affect incidence of gestational diabetes [2,5]; women with low income have more than twice the risk of developing gestational diabetes than more affluent women [6]. Little is known about the influence of socio-economic status on levels of hyperglycaemia at diagnosis, or the effects of socio-economic status on glycaemic levels once gestational diabetes has been diagnosed. Material resources unequally distributed by race and gender, are stratified by place [7], and place-based characteristics are modifiable amplifiers to health inequalities [8]. Marginalization refers to the patterning of inadequate and unequal access to political, social and economic benefits and exclusion from full societal participation attributable to membership in an identifiably less privileged group [9]. This study assesses the association between marginalization and plasma glucose levels
1563
DIABETICMedicine
What’s new? ● This study demonstrates an association between residential socio-demographic marginalization and elevated levels of plasma glucose at time of gestational diabetes diagnosis in a multicultural urban Canadian population. ● We report a relationship between elevated glucose values at the time of gestational diabetes diagnosis and likelihood of eventual insulin prescription. ● Linking elevated diagnostic test values to patients living on a low income and in ethnic minority neighbourhoods may provide clinicians with an extended risk profile and indicate a need for timely initiation of insulin. at diagnosis of gestational diabetes (1-h 50-g, glucose challenge test) and insulin prescription amongst a multi-ethnic and socio-economically diverse urban population.
Subjects and methods Electronic patient charts from the Toronto East General Hospital Gestational Diabetes Clinic were reviewed for patients referred to the clinic between 1 March 2006 and 26 April 2011. Women were retained if they had a documented 50-g glucose challenge test, demonstrated fulfilment of gestational diabetes criteria [1] and had a valid 6-digit postal code. The index pregnancy was the most recent pregnancy. Medical variables available included age, diagnosis (type of diabetes), reported diagnostic test, parity, gestational diabetes in a previous pregnancy, family history of diabetes, and insulin prescription at any time during follow-up.
Glucose challenge test
Glucose level at 1 h following a 50-g glucose load (glucose challenge test) was recorded (according to 2008 Canadian guidelines). The study sample and cases excluded because of a missing glucose challenge test were similar socio-demographically (data available from authors).
Insulin prescription
Insulin prescription was recorded as a patient being prescribed insulin by their physician during the index pregnancy at any time following diagnosis of gestational diabetes.
Material deprivation and ethnic concentration
Marginalization indicators, material deprivation and ethnic concentration were obtained from the Ontario Marginalization Index (ON-Marg) [10]. Material deprivation is derived
1564
Marginalization and gestational diabetes L. Sampson et al.
from census indicators measuring the proportion of residents with low education, low income, receiving government transfer payments, who are unemployed, single-parent families and with dwellings in need of repair. Ethnic concentration is measured by proportion of new immigrants (< 5 years) and proportion of the population who self-identify as a visible minority (or racialized group). Dissemination area-level factor scores and quintiles for these dimensions of marginalization were assigned to individual subjects via 6-digit postal codes. Postal codes were translated into dissemination areas using the Statistics Canada Postal Code Conversion File [11]. The socio-demographic marginalization data were applied to the patients. The marginalization indicators were collinear (Pearson 0.48, P ≤ 0.001). They were combined as a composite, binary variable: Deprivation/ethnic concentration. Women were coded 1 if they lived in a dissemination area classified as the highest quintile (5 = most marginalized) in both Material deprivation and Ethnic concentration, and 0 if they did not. Multivariate linear regression models were used to examine the association between marginalization and 50-g glucose challenge test diagnosis, while accounting for known risk factors (age, parity, gestational diabetes in previous pregnancy and family history). Logistic regression models using two-tailed significance tests are reported for insulin prescription and conservative reliable estimates were recorded. Breusch–Pagan/Cook–Weisberg, variance inflation factor (VIF) scores and F-statistic tests were carried out (Stata version 12.0; StataCorp., College Station, TX, USA).
Results Five hundred and thirty-one women were included in the final analyses. The mean glucose challenge test was 11.9 mmol/l, with values ranging from 8.3 to 26.2 mmol/l. Subjects were 32 years old on average with a mean of one prior delivery. Approximately half had a family history of diabetes, 75% of subjects had no personal history of gestational diabetes in previous pregnancies and 185 patients presented as primigravida (35% of the study population). One hundred and fifty-six women (30%) lived in the most marginalized neighbourhoods (Q5 for both material deprivation and ethnic concentration) and this group had a mean glucose challenge test value of 12.23 mmol/l (95% CI 11.86– 12.60). The remaining 375 women (70%) were classified as not living in the most marginalized neighbourhoods (Q1– Q4) and had a mean glucose challenge test value of 11.81 mmol/l (95% CI 11.65–11.97). Of the 527 patient charts that reported treatment information, approximately half of the women were prescribed insulin, while the remaining women controlled their blood sugar levels through diet and exercise. More women in the most marginalized neighbourhoods were prescribed insulin (54% vs. 43%).
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
DIABETICMedicine
Patients lived in 307 unique dissemination areas. The median household income was $48 327, with an unemployment rate of 9% [12]. For context, the Canadian median household income was $66 650 and the unemployment rate was 6.4% [12]. The average study neighbourhood was comprised of a 54% immigrant population (predominantly South Asian and Chinese) and the majority of patients lived in areas with high levels of ethnic concentration (Q4 and Q5). In multivariate analyses, women living in the most marginalized neighbourhoods had 0.43 mmol/l higher glucose challenge test values (95% CI 0.08–0.78) compared with other patients (Table 1). Having a diagnosis of gestational diabetes in a previous pregnancy was associated with an increase of 0.59 mmol/l (95% CI 0.19–1.00) in glucose challenge test values for the index pregnancy. Patients with a family history of diabetes had a 0.32 mmol/l (95% CI –0.01 to 0.67) increase in glucose challenge test results.
Within the neighbourhoods classified as the most highly ethnically concentrated (Q5), we tested the range of glucose challenge test values across material deprivation quintiles (Fig. 1). Although there is a linear trend, the quintile that is conclusively different from the other quintiles is Q5 (most materially deprived). In Q5, the mean glucose challenge test value was 12.23 mmol/l (95% CI 11.86–12.60), which is statistically different from the mean glucose challenge test of 11.49 mmol/l in Q1 (95% CI 11.21–11.77). Gestational diabetes hyperglycaemia at diagnosis and gestational diabetes in a previous pregnancy were found to be associated with insulin prescription (odds ratio 1.33, 95% CI 1.17–1.51 and odds ratio 1.96, 95% CI 1.23–3.11, respectively). Living in areas of high ethnic deprivation and high ethnic concentration was associated with increased odds of being prescribed insulin of 44% (95% CI 0.95–2.17), but this result was not conclusive.
Table 1 Glucose challenge test (50-g) and insulin prescription
50-g glucose challenge test value at diagnosis (mmol/l) Age (years) Gestational diabetes in previous pregnancy Parity Family history of diabetes Marginalization§
Model 1 (n = 531)
Model 2 (n = 527)
Linear regression: 50-g glucose challenge test
Logistic regression: Insulin prescription
Incremental increase in 50-g glucose challenge test value (mmol/l)
95% CI
Odds ratio
95% CI
—
1.33‡
1.17–1.51
–0.01 to 0.04 0.19–0.99 –0.26 to 0.08 –0.01 to 0.66 0.08–0.78
1.02 1.96† 1.04 1.11 1.44
0.98–1.06 1.23–3.11 0.85–1.27 0.75–1.64 0.95–2.17
— ~ 0.00 0.59† –0.09 0.32* 0.43†
*P ≤ 0.05; †P ≤ 0.010; ‡P ≤ 0.001, two-tailed tests. §Living in dissemination area with co-occurrence of quintile 5 for both material deprivation and ethnic concentration (Ontario Marginalization Index).
FIGURE 1 Glucose challenge test values by material deprivation for patients living in the most ethnically concentrated (quintile 5) dissemination areas.
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
1565
DIABETICMedicine
Discussion Women with gestational diabetes living in neighourhoods at extremes of both ethnic concentration and material deprivation had higher glucose levels 1 h following a glucose challenge test compared with those living in less marginalized circumstances, in models adjusting for age, previous gestational diabetes, parity and diabetes family history. Previous gestational diabetes and family diabetes history were also associated with higher 1-h post-load glucose levels. Elevated glucose challenge test values at diagnosis were found to be associated with increased odds of insulin prescription and gestational diabetes in a previous pregnancy, in a model adjusted for age, parity and family history of diabetes. Our findings signal that marginalization is associated with glucose values at diagnosis of gestational diabetes and may flag a need for insulin management. Our findings are consistent with three studies that have demonstrated the influence of place-based social conditions on elevated gestational diabetes risk [13–15]. Feig et al. found that women living in urban rather than rural areas and women from lower-income neighbourhoods were at increased risk for gestational diabetes [15]. A Norwegian-based study similarly reported women from areas of lower socio-economic status (Oslo East) had increased gestational diabetes risk (odds ratio 1.6) [14]. An Australian study used maternal origin as a proxy for ethnicity and found that South Asian women had the highest adjusted odds ratio of any region [13]. The same study reported that women living in the three lowest socio-economic quartiles had higher gestational diabetes incidence relative to women in the higher quartile [13]. Our study builds on these prior findings, demonstrating that marginalization is also associated with the magnitude of glucose elevation. Studies have linked previous diagnosis of gestational diabetes with elevated likelihood of future gestational diabetes [15–17], but, to our knowledge, ours is the first to demonstrate a link with elevated plasma glucose levels during later gestational diabetes pregnancies. Similarly, while prior studies have established an association between family history of Type 2 diabetes and gestational diabetes [1,4], we have determined a family history of diabetes to be associated with a 0.32 mmol/l higher glucose challenge test.
Marginalization and gestational diabetes L. Sampson et al.
[19], obesity could have mediated our findings. Indeed, a study by Cullinan et al. found a strong socio-economic gradient in the prevalence of gestational diabetes in Ireland, some of which they argue is driven by behavioural risk factors such as personal, clinical and lifestyle factors that vary across socio-economic groups [20]. While much of the study population lived in neighbourhoods composed of ethnocultural groups known to have increased diabetes risk, such as South Asian and Chinese, we demonstrate that a socio-economic gradient within those higher risk categories persists. This suggests that material deprivation has an association with gestational diabetes above and beyond ethnocultural predisposition.
Conclusions Neighbourhoods that concentrate minority and racialized populations in situations of material deprivation are not unique to Toronto. Minority populations that cluster together in poverty are emblematic of many urban environments, which should provide clinicians with an extended risk profile for people who reside in neighbourhoods similar to those in this study. Only through identifying these at-risk populations can we aim to not only prevent gestational diabetes, but also recommend increased post-diagnosis follow-up visits to ensure timely initiation of insulin and reduction of long-term complications.
Funding sources
This work was supported by Canadian Institutes of Health Research (CIHR) Catalyst Grant: Population and Public Health to KD and NAR (co-principal investigators) Exploring Hypotheses Linking Type II Diabetes and Neighbourhood Environments (no. 94431) and CIHR Interdisciplinary Capacity Enhancement (ICE) Grant, Health Disparities: Intergenerational Dynamics and Empirical Assessment Strategies (no. HOA-80072).
Competing interests
None declared.
Acknowledgements Limitations and strengths
The main limitation of our study was the inconsistency of physician reporting in the patient hospital charts, which prevented us from capturing some potentially important covariates. These included individual-level ethnicity, socio-economic status, smoking status and pre-pregnancy weight. Obesity has been identified as a major modifiable risk factor associated with gestational diabetes [18]. As poverty and obesity are linked, particularly for Canadian women
1566
The authors would like to thank Lisa Sparrow at the Toronto East General Hospital (TEGH) for her insight into the study population.
References 1 Thompson D, Capes S, Feig DS, Kader T, Keeley E, Kozak S et al. Diabetes and Pregnancy. In: Committee CDACPGE ed. Canadian Diabetes Association 2008 Clinical Practice Guidelines for the Prevention and Management of Diabetes in
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
Research article
2
3
4
5 6
7
8 9 10
11
Canada. Toronto: Canadian Journal of Diabetes, 2008: S186– S180. Nomura Y, Marks DJ, Grossman B, Yoon M, Loudon H, Stone et al. Exposure to gestational diabetes mellitus and low socioeconomic status: effects on neurocognitive development and risk of attention-deficit/hyperactivity disorder in offspring. Arch Pediatr Adolesc Med 2012; 166: 337–343. Metzger BE, Lowe LP, Dyer AR, Trimble ER, Chaovarindr U, Coustan DR et al. Hyperglycemia and adverse pregnancy outcomes. N Engl J Med 2008; 358: 1991–2002. Buchanan TA, Xiang AH, Page KA. Gestational diabetes mellitus: risks and management during and after pregnancy. Nat Rev Endocrinol 2012; 8: 639–649. Ferrara A. Increasing prevalence of gestational diabetes mellitus: a public health perspective. Diabetes Care 2007; 30: S141–S146. Joseph KS, Liston RM, Dodds L, Dahlgren L, Allen AC. Socioeconomic status and perinatal outcomes in a setting with universal access to essential health care services. CMAJ 2007; 177: 583–590. Massey DS. The age of extremes: concentrated affluence and poverty in the twenty-first century. Demography 1996; 33: 395– 413. Roux AVD, Mair C. Neighborhoods and health. Ann N Y Acad Sci 2010; 1186: 125–145. Spitzer DL. Engendering health disparities. Can J Public Health 2005; 96: S78–S96. Matheson FI, Dunn J, Smith KLW, Moineddin R, Glazier RH. ON-Marg Ontario Marginalization Index. 1.0 edn. Toronto: Collaboratory for Research on Urban Neighbourhoods, Community Health and Housing, 2012. Holowaty EJ, Norwood TA, Wanigaratne S, Abellan JJ, Beale L. Feasibility and utility of mapping disease risk at the neighbourhood
ª 2014 The Authors. Diabetic Medicine ª 2014 Diabetes UK
DIABETICMedicine
12 13
14
15
16
17
18
19
20
level within a Canadian public health unit: an ecological study. Int J Health Geogr 2010; 9: 21. Statistics Canada. 2006 Canadian Census. Ottawa: Statistics Canada, 2006. Anna V, van der Ploeg HP, Cheung NW, Huxley RR, Bauman AE. Sociodemographic correlates of the increasing trend in prevalence of gestational diabetes mellitus in a large population of women between 1995 and 2005. Diabetes Care 2008; 31: 2288–2293. Clausen T, Oyen N, Henriksen T. Pregnancy complications by overweight and residential area. A prospective study of an urban Norwegian cohort. Acta Obstet Gynecol Scand 2006; 85: 526–533. Feig DS, Zinman B, Wang X, Hux JE. Risk of development of diabetes mellitus after diagnosis of gestational diabetes. CMAJ 2008; 179: 229–234. Kim C, Newton KM, Knopp RH. Gestational diabetes and the incidence of type 2 diabetes: a systematic review. Diabetes Care 2002; 25: 1862–1868. Lauenborg J, Hansen T, Jensen DM, Vestergaard H, Molsted-Prdersen L, Hornnes P et al. Increasing incidence of diabetes after gestational diabetes. Diabetes Care 2004; 27: 1194–1199. Hedderson MM, Williams MA, Holt VL, Weiss NS, Ferrara A. Body mass index and weight gain prior to pregnancy and risk of gestational diabetes mellitus. Am J Obstet Gynecol 2008; 198: e401–e407. Matheson FI, Moineddin R, Glazier RH. The weight of place: a multilevel analysis of gender, neighborhood material deprivation, and body mass index among Canadian adults. Soc Sci Med 2008; 66: 675–690. Cullinan J, Gillespie P, Owens L, Avalos G, Dunne FP. Is there a socioeconomic gradient in the prevalence of gestational diabetes mellitus? Ir Med J 2012; 105: 21–23.
1567
Copyright of Diabetic Medicine is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
