Changes in prevalence of diabetes over 15 years in a rural Australian population: The Crossroads Studies

https://doi.org/10.1016/j.diabres.2020.108492Get rights and content

Abstract

Aims

Secular trends in diabetes prevalence vary globally. We compared the prevalence of diabetes in two surveys 15 years apart in rural Australia.

Method

Two cross-sectional household surveys and clinics (biomedical assessments including oral glucose tolerance tests (OGTTs)) in 2001–2003 (Crossroads) and 2016–2018 (Crossroads-II).

Setting: Four rural Victorian towns.

Participants: Residents of randomly selected households (Crossroads (n = 5258), Crossroads-II (n = 2649)) with nested clinic assessments for randomly selected participants (n = 1048 and 736 respectively). Response rate 61%.

Main outcome measures: Self-reported diagnosed diabetes, screening history, and diabetes defined by OGTT.

Results

The age standardised prevalence of diagnosed diabetes increased from 5.0(4.4–5.7)% to 7.7(6.7–8.6)%, with crude prevalence increasing overall (5.4 to 10.4% p < 0.001), in the smaller towns (5.4 to 11.1% p = 0.001) and, the regional centre (4.1 to 7.3% p < 0.001). Screening for diabetes over the previous two years increased (rural towns 49.8 to 63.8%; regional centre 44.9 to 63.6%; both p < 0.001). The proportion of undiagnosed diabetes was 23.2% in 2003 and 13.7% in 2018. The age and sex adjusted change in total diabetes was (1.15(0.84–1.59)). Central obesity (adjusted odd ratio 1.28(1.00–1.64)) but not overall obesity (adjusted odd ratio 1.17(0.95–1.46)) increased over time.

Conclusions

Over 15 years, the crude prevalence of diagnosed diabetes increased while the age and sex adjusted total diabetes prevalence did not change significantly. The epidemic may be slowing in some settings.

Introduction

Diabetes and its complications result in increasingly high healthcare costs as well as reduced quality of life [1]. A major driver to these costs is the growing numbers of people living with diabetes worldwide, expected to grow from 463 million in 2019 to 700 million by 2045 [2]. This is mostly due to population growth, ageing and an increase in prevalence driven by new cases (estimated 31.8% of the total growth) [3]. A reduction in mortality among those with diabetes [4], [5], [6] has also contributed to the increasing prevalence. Interestingly, trends in the incidence of clinically diagnosed diabetes appears to have been stabilising since 2006 in high income countries [7].

The number of people with diabetes in Australia is estimated to be 1.2 million [8], many of whom live in rural and remote areas where access to specialist diabetes services can be limited [9]. Those living in areas with the most disadvantage, including rural areas, are significantly more likely to develop abnormal glucose tolerance (AGT), partially mediated by central adiposity and unhealthy behaviours [10]. As with other countries, rural Australians are more likely to be obese and have a higher prevalence of diabetes [11].

The Goulburn Valley in rural Victoria includes areas with significant disadvantage, a mix of Aboriginal, migrant and settled residents, and has experienced increased access to primary care over the last 15 years. A study in this region in 2001–2003 (the Crossroads Study) described the prevalence of diabetes, impaired glucose tolerance (IGT), impaired fasting glucose (IFG) [12] and obesity [13] in the regional centre and surrounding smaller towns. We postulated that while diabetes and obesity prevalence would have continued to increase, rates of diabetes screening and lifestyle risk factors would have improved, associated with improved primary care access and overall secular trends. To demonstrate such changes requires analysis of serial cross-section samples, using the same methodology, in the same population. Therefore, we undertook the Crossroads-II follow-up study [14] in 2016–2018 to investigate changes in the prevalence of diagnosed and total diabetes, diabetes screening, and diabetes risk factors between the baseline and follow-up Crossroads studies.

Section snippets

Materials and methods

The Goulburn Valley region of Victoria, Australia is located 100–300 km from Melbourne and was chosen for the study due to its previously limited access to health services and excess poor health outcomes (2001–2003) [12]. This follow-up study was undertaken across three towns (populations 6000–10,000) situated around the regional centre (population 45,000) [15]. The three shire capitals were selected because of their geographical locations east, north and south of the regional centre, and their

Results

In Crossroads-II, 1895/3022 (62.7%) households (number of occupied residences) responded to the survey. Supplementary Fig. 1 shows the overall response to study participation. Table 1 compares baseline characteristics of responding residents in Crossroads (n = 5,258) and Crossroads-II (n = 2649). Compared to participants in Crossroads, participants in Crossroads-II were older, more likely to be female and be of non-European descent (including more likely to be Aboriginal 50 (1.9%) vs 39(0.7%)

Discussion

This is the first, recent, population-based follow-up survey of dysglycaemia in rural Australia and demonstrates an increase in the prevalence of diagnosed diabetes by 58(29–94)% over 15 years after adjusting for demographic and socioeconomic factors. Minor attenuation occurred with inclusion of lifestyle factors in the model. The prevalence increased to a similar extent in men and women, and age-specific prevalence also increased significantly, in men and women aged ≥ 70 years and among women

Author contribution

DS conceived and designed the study, wrote the first version of the manuscript and researched the data. LB and KG helped design the study, researched the data and reviewed/edited the manuscript. DM contributed to the discussion and reviewed/edited the manuscript. All authors had full access to all of the data.

Data sharing

Data will not be available.

Ethics

This project received ethics approval from Goulburn Valley Health (ID 1648142).

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

We thank the University of Melbourne and the Baker Heart and Diabetes Institute for their support during Crossroads. We thank our project manager Sian Wright and the fieldworkers who assisted with data collection. Many thanks to Carolina Sari for the age standardisation calculations. Finally, we are indebted to all the participants who generously gave their time.

Formatting of funding sources

We gratefully acknowledge the support of the Australian Government Department of Health Rural Health Multidisciplinary Training Programme. Crossroads-II was funded by the NHMRC (APP1113850) and local health and local community partners, including Goulburn Valley Health, Primary Care Connect, Benalla Health, Cobram District Health, Seymour Health, Moira Shire, Goulburn Valley Primary Care Partnerships, Shepparton Access, City of Greater Shepparton, Alfred Health and the Department of Rural

References (24)

  • Australian Bureau of Statistics. Diabetes Mellitus, National Health Survey: First Results 2017-2018. ABS, 2018. (ABS...
  • E.D. Williams et al.

    Area-level socioeconomic status and incidence of abnormal glucose metabolism: the Australian Diabetes, Obesity and Lifestyle (AusDiab) study

    Diabetes Care

    (2012)
  • Cited by (8)

    View all citing articles on Scopus
    View full text