Neighbourhood property value and type 2 diabetes mellitus in the Maastricht study: A multilevel study.

OBJECTIVE: Low individual socioeconomic status (SES) is known to be associated with a higher risk of type 2 diabetes mellitus (T2DM), but the extent to which the local context in which people live may influence T2DM rates remains unclear. This study examines whether living in a low property value neighbourhood is associated with higher rates of T2DM independently of individual SES. RESEARCH DESIGN AND METHODS: Using cross-sectional data from the Maastricht Study (2010-2013) and geographical data from Statistics Netherlands, multilevel logistic regression was used to assess the association between neighbourhood property value and T2DM. Individual SES was based on education, occupation and income. Of the 2,056 participants (aged 40-75 years), 494 (24%) were diagnosed with T2DM.
RESULTS: Individual SES was strongly associated with T2DM, but a significant proportion of the variance in T2DM was found at the neighbourhood level (VPC = 9.2%; 95% CI = 5.0%-16%). Participants living in the poorest neighbourhoods had a 2.38 times higher odds ratio of T2DM compared to those living in the richest areas (95% CI = 1.58-3.58), independently of individual SES.
CONCLUSIONS: Neighbourhood property value showed a significant association with T2DM, suggesting the usefulness of area-based programmes aimed at improving neighbourhood characteristics in order to tackle inequalities in T2DM.

Background

With an increase in the number of people with diabetes from 108 million in 1980 to 422 million in 2014 [1], T2DM—representing about 90% of cases of diabetes—is a growing health problem worldwide. It is known that people with a SES defined by low education, occupation and income, have a higher risk of T2DM compared to people with higher SES. Behavioural risk factors, such as poor diet, low levels of physical activity, and smoking—which are known to be more common in less advantaged individuals—have been traditionally related to diabetes outcomes. Accordingly, the prevalence of T2DM shows large inequalities, disproportionally affecting deprived populations [2]. Moreover, T2DM prevalence is also related to the geographical context where people live, which can shape individuals’ health outcomes through different pathways, resulting in a clear patterning of the disease at various geographical levels. A review and meta-analysis of the existing literature reports that low levels of education, occupation and income—used as SES measures—were associated with an increased risk of T2DM in high-, middle- and low-income countries [3]. Another strand of research has focused on the role played by the environment in which people live in contributing to determine diabetes outcomes. Studies in a review and meta-analysis of the literature focus on the characteristics of the built environment (green spaces, walkability, food environment, air and noise pollution), providing useful insights as regards the structural characterizations of the place of living in relation to diabetes outcomes [4]. However, they tend to neglect the role of individual socioeconomic conditions, which are often not taken into account nor adjusted for as confounders. Similarly, studies focused on the association between individual SES and diabetes outcomes rarely consider the spatial characterization of the phenomenon.

The physical and social environment in which people live influences their lifestyle, behaviours and opportunities, and ultimately also their health [5]. Processes influencing T2DM outcomes take place at different levels in individuals’ lives, and each of these levels should be included in analyses aimed at estimating the effects of social circumstances and the surrounding environment.

In this study, we relied on individual self-reported measures of educational level, occupational status, and household income as indicators of individual SES, while we used the average property value of the neighbourhood as an indicator of neighbourhood SES. Rather than merely reflecting attributes of the individual real estate (e.g. size, quality, and the like), the average property value of a neighbourhood has proven to be mostly the product of area attractiveness [6] and might thus be considered a valid indicator of area SES. Accordingly, the use of property value as a proxy measure for SES has proven efficient in relation to many health outcomes, both at the individual [7,8,9] and contextual levels [10,11]. It has been suggested that measuring neighbourhood SES by using the traditional approach with aggregate measures of education, occupation, and income may not reflect the complex mechanisms interrelated in defining the characteristics of an area [12], while property value—encompassing in its evaluation also housing location and characteristics of the built environment in which homes are located—may capture the environmental and material aspects of SES in broader terms [13].

Thus, by using cross-sectional data from the Maastricht Study and matching them with aggregate-level neighbourhood data from Statistics Netherlands (CBS), we analysed neighbourhood variation in T2DM rates in the area studied and its independence from individual SES in order to evaluate the association between living in a low property value neighbourhood and T2DM.

Research design and methods

Study population

For our analysis, we used data from The Maastricht Study, an observational prospective population-based cohort study. The rationale and methodology have been described previously [14]. In brief, the Maastricht Study focused on the etiology, pathophysiology, complications and comorbidities of T2DM and was characterized by an extensive phenotyping approach. The study uses state-of-the-art imaging techniques and extensive biobanking to determine the health status in a population-based cohort of 10,000 individuals with T2DM individuals oversampled. Eligible for participation were all individuals aged between 40 and 75 years and living in the southern part of the Netherlands. Participants were recruited through mass media campaigns and from the municipal registries and the regional Diabetes Patient Registry via mailings. Recruitment was stratified according to known T2DM status, with an oversampling of individuals with T2DM, for reasons of efficiency.

The present study includes cross-sectional data from the first 3,451 participants in the Maastricht Study who completed the baseline survey between November 2010 and September 2013. The examinations of all participants were performed within a time window of three months. The study has been approved by the Maastricht University and the Academic Hospital’s medical ethical committee (NL31329.068.10) and the Ministry of Health, Welfare and Sports of the Netherlands (Permit 131088-105234-PG). All participants gave written informed consent.

From the overall sample of 3,451 respondents, 1,395 (40.4%) were excluded due to missing data in one or more of the covariates (879 equivalent income, 77 educational level, 627 occupational status, 76 property value), or because of reporting type 1 diabetes (n = 37) or other types of diabetes (n = 4). The final sample size was 2,056 (59.6% of initial sample). We used 2-way t-test to determine whether the included group differed from the excluded one in key characteristics. The percentage of respondents with T2DM in the final sample included in the analysis was 24.0%, compared to 35.5% of the excluded population (P-value<0.001). The mean age of respondents in the included group was 58.9 years, compared to 61.0 years in the excluded one (P-value<0.001). The percentage of female respondents in the final sample was 47.2%, compared to 50.6% in the excluded group (P-value = 0.048). The mean neighbourhood property value was 222,174€ in the final sample included in the analysis, compared to 219,801€ of the excluded population (P-value = 0.666).

Measures

Diabetes status

T2DM status is defined according to the 2006 WHO diagnostic criteria of glucose tolerance status [15]. All participants underwent a standardized 7-point OGTT after overnight fasting. Blood samples were collected at baseline, and 15, 30, 45, 60, 90 and 120 minutes after consumption of the 75g glucose drink. Participants who were insulin-dependent and participants with a fasting glucose level higher than 11.0 mmol/l (as determined by finger prick) did not undergo this test. Prediabetes was defined as IFG (fasting plasma glucose 6.1–6.9 mmol/l and 2-h plasma glucose <7.8 mmol/l), IGT (fasting plasma glucose <7.0 mmol/l and 2-h plasma glucose ≥7.0—< 11.1 mmol/l) or both [16]. T2DM was defined by fasting plasma glucose ≥7.0 mmol/l or 2-h plasma glucose ≥11.1 mmol/l. Participants on diabetes medication and without type 1 diabetes were also considered as having T2DM.

Individual SES

The study sets out self-reported information about participants’ SES, including educational level, occupational status, and income level.

Educational level was reported through seven ordinal categories: 1) no education, primary education not completed, primary education, 2) lower vocational education, 3) lower general education, 4) intermediate vocational education, 5) higher secondary education, 6) higher professional education, 7) university education.

Occupational status was reported by means of a continuous measure, according to the criteria of the International Socioeconomic Index (ISEI-08) of occupational status [17], which ranks occupational positions also by the average level of education and average earnings of job holders.

Household income was adjusted for household size. Household income was divided by the square root of household size, implying, for instance, that a household of four persons has twice the needs of a single-person household, as suggested by equivalent scales of the Organisation for Economic Co-operation and Development (OECD) [18].

The three individual SES measures were normalized for reasons of efficiency and to allow a comparison between the magnitude of the effect of each indicator on T2DM risk.

Neighbourhood property value

Information concerning socioeconomic characteristics of the neighbourhood of residence came from CBS Statistics Netherlands (Centraal Bureau voor de Statistiek). CBS routinely collects a wide range of indicators regarding the territory and its inhabitants, which are made available every year at three geographical levels: neighbourhoods (average area size in the study area: 1.8 km2), districts (9.7 km2), and municipalities (50.8 km2).

We used average neighbourhood property value as proxy for neighbourhood SES. In the Netherlands, property values are assessed every year according to the criteria of the Law on the Valuation of Property (Wet Waardering Onroerende Zaken) [19]. The evaluation is implemented by the single municipalities, and the data are made freely available by CBS (see: www.cbsinuwbuurt.nl), at the three geographical levels. Here, we opted for neighbourhood, since property values are likely to be more homogenous at this lowest level. We divided the original continuous variable, indicating neighbourhood average property value in Euro, into quartiles, in order to have an immediately understandable comparison between people living in disadvantaged rather than advantaged neighbourhoods.

Each participant’s residential address was attributed to the corresponding neighbourhood using conversion files provided by CBS [20]. The neighbourhood code of residence was then used to link the Maastricht Study data with CBS data, so that it was possible to match individual and contextual information for each respondent. For each neighbourhood, property value was assigned by computing the mean of the four years considered in the analysis (from 2010 to 2013). The correlation between the property values of the neighbourhoods in the different years of analysis was always close to one, meaning that for each neighbourhood there was almost no variation over the years and that it was reasonably possible to rely on the average value without biasing the results. The study included 82 neighbourhoods, with an average of 25 cases for each (minimum: 1; maximum: 135).

Confounders

Both sex and age at the year of visit were used as covariates.

Statistical analysis

Given the hierarchical structure of the data (individuals nested in neighbourhoods), multilevel regression models [21] were used to assess simultaneously the effect of individual and contextual characteristics, enabling estimation of the effect of the neighbourhood of residence on the outcome, independently of individual SES.

First, the spatial heterogeneity in T2DM rates was assessed using a multilevel binary logistic empty model (model 1), which enabled measurement of the extent to which the probability of having T2DM varies from one area to another. The variance partition coefficient (VPC) revealed the proportion of variability in the outcome at each level of analysis, providing a first description of the spatial distribution of the disease in the study area and evidencing the existence of a possible contextual dimension for the phenomenon studied. Second, the model was integrated with predictors at level-1 (individual SES) to investigate the extent to which area level differences were explained by the individual composition of the areas (model 2). Third, the level-2 predictor was added to check if neighbourhood property value was associated with T2DM independently of individual SES, i.e. to assess the existence of a contextual effect for T2DM in the population studied (model 3). In all the main analyses, random intercept models were fitted; multi-collinearity was checked before running the models. Additional analyses were performed with random slope models for individual SES measures and cross-level interactions. Finally, as reported more in detail in the discussion, sensitivity analyses were conducted to assess the choice of the predictors, outcomes and area units used, repeating the analysis with different operationalizations.

The multilevel logistic regression models were estimated with the Likelihood Estimation method using STATA 15 software (the analyses were also repeated with MLwiN 3.02).

Results

Table 1 shows the distribution of the variables considered at the individual and neighbourhood levels in the models developed. Most of individuals reporting T2DM were males (73.5%) rather than females (26.5%). T2DM rates increased with age and decreased with educational level, occupational status and household income. Neighbourhoods from higher to lower property values showed proportionally higher rates of T2DM.

Table 1

Percentage distribution of the variables (n = 2,056).

	Totalpercentage(n = 2,056)	T2DMPercentage(n = 494)
Sex
Female	47.2	26.5
Male	52.8	73.5
Age
40–53	26.1	15.6
54–59	23.8	20.5
60–65	26.2	26.5
66–75	24.0	37.5
Educational Level
University education	10.5	5.7
Higher professional education	32.2	25.9
Intermediate vocational, Higher secondary	29.8	30.4
Primary, lower general/vocational	27.6	38.1
Occupational status (ISEI08 score)
88.9–70.6	24.8	19.0
70.5–56.1	25.3	22.7
56.0–39.1	23.7	22.7
39.0–13.2	26.2	35.6
Household Income (€)
6,000–2,437	24.2	18.2
2,386–1,888	24.9	23.5
1,875–1,509	22.8	23.1
1,502–424	28.2	35.2
Property Value (€)
581,000–262,000	24.8	17.2
261,000–226,000	26.0	22.5
225,000–169,000	23.6	23.3
168,000–125,000	25.6	37.0

Table 2 shows the main results of the analyses performed. In model 1, the VPC of 9.2% informed us that, even if most of the variance was found at the individual level, there was a significant variation in T2DM outcome at the neighbourhood level. In model 2, we considered all individual predictors to account for differences in T2DM outcomes, without including any contextual variable. Among the measures of individual SES considered, both educational level (OR = 0.50; CI = 0.28–0.87) and occupational status (OR = 0.51; CI = 0.28–0.91) were inversely associated with T2DM, whilst the effect of household income on the outcome was not statistically significant. In model 3 we included all individual-level predictors, introducing the contextual variable ‘property value of the neighbourhood of residence’ as level-2 predictor. The results showed that, after controlling for all individual socioeconomic characteristics, people living in neighbourhoods with the lowest property values were more than twice as likely to have T2DM, as compared to people living in the better-off neighbourhoods (OR = 2.38; CI = 1.58–3.58). Significance and direction of level-1 predictors remained substantially the same on moving from model 2 to model 3. The decreasing VPC from model 1 (9.2%) to model 3 (2.2%) indicated that the predictors included in each step were able to explain most of the contextual variance, as confirmed by the decreasing AIC, indicating a better fit of the final model with all level-1 and level-2 predictors, as compared with the previous models.

Table 2

Multilevel logistic regression of T2DM (0 = no, 1 = yes).

N = 2,056.

	Model 1			Model 2		Model 3
	AIC: 2231.84VPC: 9.2%			AIC: 2030.71VPC: 4.9%		AIC: 2016.54VPC: 2.2%
	Odds Ratio	95% CI		Odds Ratio	95% CI	Odds Ratio	95% CI
Intercept	0.31	[0.25, 0.37]		0.06	[0.02, 0.16]	0.04	[0.01, 0.11]
Age				1.05	[1.04, 1.07]	1.05	[1.04, 1.07]
Sex
Male				1.00	-	1.00	-
Female				0.31	[0.24, 0.39]	0.31	[0.25, 0.40]
Educational Level				0.50	[0.28, 0.87]	0.53	[0.30, 0.93]
Occupational Status				0.51	[0.28, 0.91]	0.53	[0.29, 0.95]
Household Income				0.52	[0.20, 1.32]	0.69	[0.27, 1.75]
Property Value
Extremely high						1.00	-
Moderately high						1.14	[0.76, 1.71]
Moderately low						1.27	[0.85, 1.91]
Extremely low						2.38	[1.58, 3.58]

We also tested models with random slope for individual SES indicators and with cross-level interactions between individual SES and neighbourhood property value. However, none of these models was statistically significant, meaning that the effect of individual SES was the same in each neighbourhood and that the contextual effect of property value was the same for all individuals, regardless of their own SES (S1–S6 Tables).

Additionally, we assessed the impact of the choice of different indicators for the same models through a sensitivity analysis. First, running the same models with the 4-digit postcode (average area size in the study area: 5.2 km2)–instead of neighbourhood (1.8 km2)–as level-2 unit and with a measure of neighbourhood SES based on education, occupation and income (instead of property value) led to results similar to those reached with neighbourhood as geographical unit and property value as contextual variable (S7 Table). This suggests that the results were not influenced by the choices made in terms of area unit as well as the type of neighbourhood SES indicators. Second, analyses with the same predictors and level-2 unit but with continuous outcomes (fasting and 2-h plasma glucose tolerance status), confirmed the statistically significant association between neighbourhood property value and T2DM after controlling for age, sex and individual SES (S8 and S9 Tables). This provides evidence that our findings were not just due to the dichotomization of the outcome. Finally, running the same models including only neighbourhoods with at least 10 respondents led to analogous results (S11 Table), confirming that the results in the main models were not influenced by the inclusion of second-level units with few respondents.

Discussion

Our results showed a statistically significant association of neighbourhood property value with T2DM outcome, over and above sex, age and individual SES. The independence from SES implies that living in a poor area matters not only for individuals with fewer resources but also for the better off. Thus, people with low SES suffer the cumulative disadvantage of being exposed to the risks deriving from their personal conditions as well as to those deriving from the context in which they live.

Studies investigating the association of neighbourhood characteristics or SES with T2DM have obtained similar findings also in other contexts. A multilevel study of small-area SES in south-eastern France, reported a significantly higher prevalence of diabetes in the more deprived areas independently of individual SES [22]. An Australian study reported a lower risk of T2DM in greener neighbourhoods, even after controlling for demographic and cultural factors [23]. Similarly, a study in Canada reported that neighbourhood walkability was inversely associated with the development of diabetes [24], while a Swedish study suggested that local food environment was associated with T2DM [25]. In the United States, a series of studies reported an association of neighbourhood resources supporting physical activity and healthy diets with a lower incidence of T2DM, metabolic syndrome and insulin resistance [26,27,28,29,30]. Similarly to these studies, the results achieved here are specific of the context studied, and less likely to be applicable to lower- and middle-income counties.

Since property value has proved to predict higher rates of T2DM for people living in least valuable neighbourhoods independently of their education, occupation and income, we speculated on the possible pathways connecting the ecological level to the health outcome, following the debate between a material and a psycho-social explanation for health inequalities [31,32,33]. Least valuable neighbourhoods are possibly less equipped with supermarkets and grocery stores selling healthy food, walkable paths, proximity to green spaces and physical activity amenities, all aspects of the built environment which have been proven to be associated with higher obesity and T2DM rates [34,35,36]. In addition to this material explanation, other scholars have focused on a psychosocial one [37,38], according to which socioeconomic inequalities may be detrimental to health outcomes due to higher levels of chronic stress resulting from the psycho-social impact of the perceived relative social position [39, 40]. In the case of T2DM, this may be related to physiological and metabolic alterations due to stress response, including overstimulation of the neuroendocrine system, which could influence the development of the disease [39]. People living in unfavourable neighbourhoods compare themselves with those living in more affluent areas, feeling inadequate and fostering processes of exclusion and stigmatization. In this sense, the average property value could be more than just a proxy for neighbourhood SES and accessibility to resources through availability from the built environment, making an intrinsic contribution itself to the social and spatial patterning of T2DM outcomes.

We also know that some people living in less advantaged areas may easily go to other neighbourhoods to grant themselves access to resources missing in their resident context. Lower SES people may be limited in their mobility, which restricts their access to health-protecting or health-enhancing resources in other neighbourhoods. Hence, we should likely be cautious in assuming that everyone might have potential access to every service and amenities in the area where he or she is located. People with less purchasing power are likely more prone to rely on services and amenities available in the proximity of their residence; the better off usually have better opportunities to go elsewhere to obtain desired facilities. Accordingly, if the issue of mobility would have to be considered, it would possibly widen the already existing socioeconomic inequalities in access to services and eventually also health.

As far as methodological choices are concerned, three issues should be mentioned. First, dealing with a contextual analysis of the social determinants of health, we are aware that a choice of different area units of analysis might have changed our results significantly. We dealt with the so-called Modifiable Area Unit Problem by running the same analysis with a larger area unit (4-digit postcode) and with a traditional measure of neighbourhood SES based on education, occupation, and income—thus making it possible to check also for the reliability of property value as proxy for neighbourhood SES—and the results were completely in line with those shown in the models reported (S7 Table). Data for this analysis came from the Netherlands Institute of Social Research, which provides SES scores for each 4-digit postcode area in the Netherlands for specific years [41,42].

Second, starting from two continuous measures of level of glucose in blood, in the analysis reported we opted for a model with a dichotomous outcome, without considering prediabetes as a separate category, in line with the American Diabetes Association’s suggestion not to consider prediabetes as a clinical entity itself [43]. However, we also tested models considering prediabetes, separately running two binomial models comparing prediabetes with normal glucose and T2DM with normal glucose. Contrary to our expectations, the first comparison indicated that a low property value did not discriminate between the prediabetes and normal glucose categories (S10 Table). We are not sure about the reasons for the absence of an association with prediabetes, however, the linear models with continuous outcomes confirmed the presence of a territorial gradient in T2DM (S8 and S9 Tables), suggesting that the lack of relationship could be driven by the small sample size in the segmented analysis.

Third, only the comparison between living in extremely high and extremely low property value neighbourhoods was statistically significant. This is because we opted to compare the adverse effects on T2DM of living in a deprived neighbourhood instead of an affluent one. However, on choosing the lowest category as reference in our model, every comparison was statistically significant, resulting in a gradually incremental protective effect of living in a better neighbourhood (S12 Table). However, in both cases the extremely low property value quantile appeared to be substantially different from the other three, suggesting that interventions aimed at reducing spatial inequalities in T2DM onset in Southern Limburg should foremost be directed towards the improvement of the most deprived areas. The improvement to a more feasible intermediate level would already result in a substantial reduction of area-based health inequalities.

We should also highlight possible limitations of this study. First, the cross-sectional nature of the data used prevents us from assessing the direction of causation in our findings, even if it is less likely that having T2DM could affect residential choices and mobility. Reverse causality is possible, however, in the literature selective pathways at the origins of health inequalities are known to play a minor role as compared to the possibility that SES influences health outcomes, and not vice-versa [44]. Second, T2DM is the outcome of life-course processes which encompassed social and contextual exposure long before the moment in which people were involved in the study. For instance, the neighbourhood of residence in earlier life may have had a bigger effect than the current one. We had no data to keep track of this eventual residential mobility, but the area of study is reported to be one with low internal and external mobility [14], so that relocations should not represent an issue of great concern. Third, we had to exclude from the analysis cases for which information on education, occupation or income was missing. This could be a source of bias in the analysis if it altered the real distribution of the variables (S13 Table reports the distribution of the variables before excluding cases with missing data). Finally, we are aware that there are some other factors playing a fundamental role in shaping social and spatial inequalities in T2DM, which have not been included in our analysis. Other components of individual SES (e.g. health behaviours, parental background, disease familiarity, etc.) or of neighbourhood characteristics (e.g. availability of services and amenities, etc.), as well as other factors (e.g. social support, social cohesion, social networks, household composition, etc.) may intervene in the association between SES and T2DM. Nonetheless, our models were adjusted for the most pertinent indicators among those available, avoiding the risks of over-adjusting and underestimating the real effect of neighbourhood SES on T2DM onset.

Nonetheless, this study has also several strengths. The quality and the structure of the data used allowed us to measure each indicator at the proper level of analysis, without incurring the typical fallacies encountered when variables at one level are derived from data collected for units at a different level. In regard to the results, we found a very large odds ratio predicting higher T2DM rates for individuals living in neighbourhoods in the lowest property value quartile compared to those living in the better-off areas. Since we stringently controlled for individual SES, we may have faced over-adjustment [45], given that we cannot exclude a causal pathway between individual SES and neighbourhood property values, in either direction. Thus, the real association between neighbourhood SES and individual SES could be even stronger than the one reported, due to a possible underestimation. Finally, we conducted numerous sensitivity analyses, and they all confirmed our findings: the results from the model with a different area unit and another measure of neighbourhood SES provide stronger evidence for the hypotheses tested, suggesting through different operationalizations of the indicators that the results are not due to chance.

Conclusions

This study has shown that living in neighbourhoods with low property value is associated with higher rates of T2DM, even after controlling for individual SES. Therefore, improving local contexts where people live (e.g. providing greener and more walkable spaces and healthier food environments) and mitigating inequalities between neighbourhoods could reduce T2DM rates and inequalities. Nonetheless, addressing social stratification at the individual level remains the most important strategy to pursue in order to reduce social inequalities in health [46,47]. Focusing too closely on the context may induce neglect of disadvantaged individuals living in affluent areas, whilst concentrating solely on individuals may lead to disregard of some contextual and environmental factors playing an important role in the origin of health inequalities.

N = 2,056. Random slope for educational level (covariance unstructured).

(DOCX)

Click here for additional data file.

N = 2,056. Random slope for occupational status (covariance unstructured).

(DOCX)

Click here for additional data file.

N = 2,056. Random slope for household income (covariance unstructured).

(DOCX)

Click here for additional data file.

N = 2,056. Cross-level interaction between educational level and property value (covariance unstructured).

(DOCX)

Click here for additional data file.

N = 2,056. Cross-level interaction between occupational status and property value (covariance unstructured).

(DOCX)

Click here for additional data file.

N = 2,056. Cross-level interaction between household income and property value (covariance unstructured).

(DOCX)

Click here for additional data file.

Multilevel logistic regression of T2DM (0 = no, 1 = yes) with 4-digits postal code and neighbourhood SES.

(DOCX)

Click here for additional data file.

Multilevel linear regression of fasting plasma glucose tolerance status.

N = 2,077.

(DOCX)

Click here for additional data file.

Multilevel linear regression of 2-h plasma glucose tolerance status.

N = 1,942.

(DOCX)

Click here for additional data file.

Multilevel logistic regression of comparing a) normal glucose levels individual with prediabetes (N = 1,562); b) normal glucose levels individual with T2DM (N = 1,739); c) prediabetes with T2DM (N = 811).

(DOCX)

Click here for additional data file.

N = 1,908. Only neighbourhoods with at least 10 respondents included*.

(DOCX)

Click here for additional data file.

N = 2,056. Property value reverse coded.

(DOCX)

Click here for additional data file.

Percentage distribution of the variables before excluding cases with missing data.

(DOCX)

Click here for additional data file.

27 Feb 2020

PONE-D-19-33108

Neighbourhood Property Value and Type 2 Diabetes Mellitus in the Maastricht Study: a Multilevel Study

PLOS ONE

Dear Mr Consolazio,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Apr 11 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Eyal Oren, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please correct your reference to "p=0.000" to "p<0.001" or as similarly appropriate, as p values cannot equal zero.

3. Thank you for stating the following in the Financial Disclosure section:

"This study has been supported by the European Regional Development Fund as part of OP-ZUID, the province of Limburg, the department of Economic Affairs of the Netherlands (grant 31O.041), Stichting Weijerhorst, the Pearl String Initiative Diabetes, the Cardiovascular Center Maastricht, Cardiovascular Research Institute Maastricht (CARIM), School for Nutrition, Toxicology and Metabolism (NUTRIM), Stichting Annadal, Health Foundation Limburg and by unrestricted grants from Janssen, Novo Nordisk and Sanofi. The regional association of General Practitioners (Zorg in Ontwikkeling (ZIO)) is gratefully acknowledged for its contribution to The Maastricht Study, enabling the invitation of individuals with T2DM by using information from its web-based electronic health record."

We note that you received funding from a commercial source: Janssen, Novo Nordisk and Sanofi

Please provide an amended Competing Interests Statement that explicitly states this commercial funder, along with any other relevant declarations relating to employment, consultancy, patents, products in development, marketed products, etc.

Within this Competing Interests Statement, please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your amended Competing Interests Statement within your cover letter. We will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4. We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

6. We note that Hans Bosma serves as guest Editor for the Health Inequities and Disparities Call for papers, could you please provide an amended Competing Interest Statement to declare this. Please also confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests). If there are restrictions on sharing of data and/or materials, please state these.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This paper essentially builds on the previously published analyses of the cross-sectional relationship between socioeconomic status and diabetes prevalence in the baseline data collection for the Maastrict Cohort Study. Another recent publication (included as reference 2) explores in much more detail the association with a wide range of individual level markers of socioeconomic status. This analysis uses the potential to link the dataset to neighbourhood property values to extend the use of this dataset to explore the association with this area-level variable.

Whilst the approach to analysis and presentation of findings seems appropriate and clear, both the rationale for choosing property value as a proxy of environmental determinants of diabetes risk and the caveats in terms of assuming this is a causal association in the interpretation of the findings could be more adequately addressed as suggested below:

1. As property value reflects both an element related to the size and quality of individual properties and an element related to the perceived attractiveness of the area (eg local amenities, access to public transport, quality of local green spaces etc) it is difficult without any information on, or adjustment for, the former to know the extent to which property value reflect area rather than property specific characteristics. This makes it difficult to be confident of the authors assertion that property value is measuring area SES rather being a proxy for individual variable not adjusted for.

2. It appears from the previously published analysis that there may be other confounding variables and residual confounding by individual factors (for which property value may be a proxy) cannot be excluded as explanation for findings.

3. The lack of a relationship with prediabetes is also a finding that suggests it is less likely that this cross-sectional association is due to areas characteristics associated with property values being on the causal pathway to diabetes. Authors need to at least discuss the potential explanation of reverse causality with health status (including obesity and diabetes) being on a causal pathway that results in an individual living in an area with lower property values.

Reviewer #2: Overall comment

An interesting study with ‘daring’ interpretation that perhaps require further data and explanation.

Background

Further information and examples of the different environmental and material aspects of the different categories of property in particular between that of extreme high and extreme low would aid understanding of this variable on T2DM. Further comparisons to other countries and discussion has been quite well done under the Discussion section but probably requiring a more attention to the differences between the two extreme categories, as shown by the result.

Lines 67-70: the statement is less accurate as the review reported that most of the included studies that adjusted for lifestyle behaviours and psychosocial factors had insignificant effects except one study solely in women.

Methods

1. Slightly more information on “extensive phenotyping approach” may help understanding of the Maastricht Study.

2. Disclose also the differences between the included and excluded respondents from the aspects of NPV based on the CBS Statistics Netherlands. If this information was lost (n=76), please explain how did this happen because isn’t it a readily available data?

3. Were all patients in the regional Diabetes Patient Registry selected and invited? How were the oversampling of people with T2DM executed and completed?

4. How good is the regional Diabetes Patient Registry in terms of coverage of the people with a diagnosis of T2DM in the Southern part of the Netherlands? Could you provide a comparison between those included respondents and those in the registry at large from the aspects of NPV (if this is possible)? Supplemental Table 7 shows that the extreme groups of the occupational status and property value quartiles could be over- and under-represented.

5. I believe the municipal registries hold the total residential information. Could you provide a comparison between those included respondents and those in the regions at large from the aspects of NPV (if this is possible)?

6. Line 158: please provide more detail on the normalization procedure.

Results

Suggest to put the total (n=…) on Table 1 caption itself and for the T2DM column.

Providing key reason/s for the 3 sensitivity analyses would be appreciated here although are quite well discussed in the Discussion. Is the third one for a support of (small) sample size?

Discussion

Could NPV (and its personal SES indicators) a correlate to health behaviours that has an effect on T2DM? Isn’t it established that people of higher SES more available and capable for healthier lifestyles? Greener and more walkable spaces and healthier food environments will not benefit anyone who is not motivated to self-regulation and self-care, or unavailable and not affordable for them. The former (NPV) may act as the external motivator and the later (personal SES) may complement as the internal motivator and both would function through regular/consistent health behaviours on T2DM. The association could be bidirectional and inherited through generations (shared personal and social traits) in the majority. What would be a more plausible explanation for the association between NPV and T2DM?

Lines 311-316: do discuss a bit more on this statistical observation. The extremely low quartile property value seems to be substantially different from all the others (Supplemental Table 6). From another perspective, concluding and suggesting to achieve the external features of the moderately low property value is probably more achievable by many stakeholders to also enjoy its health benefits similar to that in the extremely high property value.

It is probably good to remind the readers that the results are likely not/less applicable to lower/middle income countries.

Are ‘greener and more walkable spaces and healthier food environments’ true observed differences between the different NPV in the southern Netherlands? Are not these relatively accessible to almost all occupants?

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Boon-How Chew

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

8 May 2020

All the issues raised by the editor and the reviewers have been addressed in the rebuttal letter attached.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

26 May 2020

Neighbourhood Property Value and Type 2 Diabetes Mellitus in the Maastricht Study: a Multilevel Study

PONE-D-19-33108R1

Dear Dr. Consolazio,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Eyal Oren, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: N/A

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Thank you for addressing reviewers' comments

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Boon-How Chew

29 May 2020

PONE-D-19-33108R1

Neighbourhood Property Value and Type 2 Diabetes Mellitus in the Maastricht Study: a Multilevel Study

Dear Dr. Consolazio:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Eyal Oren

Academic Editor

PLOS ONE