The number of individuals living alone is increasing among older people in developing and developed countries, and this is considered an important demographic and social change (1). In 2017, 33.6% of the households in the European Union and around 40% of those in Nordic countries (except Iceland) consisted of one individual living alone (2). Complex reasons explain this trend, such as, for instance, trends toward longevity, high divorce rates, high rates of widowhood, and low rates of intergenerational co-residence (3). Living alone may cause social isolation and feelings of loneliness and depression, especially when individuals perceive that their social needs are not met. The influence of social isolation on mortality and morbidity has been established among the traditional clinical risk factors (2, 4, 5). Meanwhile, loneliness and depression can negatively impact health and survival. Meta-analytic evidence demonstrates that loneliness is a predictor of all-cause mortality, showing that lonely people have a 22% higher risk of death than do non-lonely people (1). Moreover, living alone also entails a higher cost of living and may increase the economic burden of low-income people (6). Thus, living alone arises as a new concern with aging in patients with chronic non-communicable diseases.

Previous studies have demonstrated a significant association between living alone and mortality among older people (7, 8). The causal pathways connecting living alone with mortality are multifactorial. The social networks of individuals living alone tend to shrink, and these individuals are also likely to be in poorer health. Meanwhile, patients living alone have an increasing trend toward poor health behaviors (5, 9), and are also more likely to experience unmet care needs (10). In addition, several studies found that single living increased worse outcomes post heart attack or myocardial infarction (11, 12). More recently, our previous study demonstrated living alone is an independent risk factor for 1-year all-mortality in acute coronary syndrome patients ≥75 years of age (13). Given type 2 diabetes mellitus (T2DM) has been associated with the onset of atherosclerotic cardiovascular disease among older patients, often presenting as coronary heart disease (CHD), cerebrovascular disease, and cardiovascular death of atherosclerotic origin in patients (14), subsequently promoting premature aging. Therefore, there is an urgent need for cardiovascular events prevention in diabetic individuals. To achieve this goal, it is necessary to identify specific high-risk factors affecting the prognosis of T2DM in primary care.

The percentage of T2DM patients living alone has been reported to be ~7–15% (15). To date, no study has prospectively assessed the association between living alone and incident T2DM, although cross-sectional studies have investigated living alone as a risk factor for T2DM (16, 17). As individuals with T2DM tend to live for a long time with advanced comorbidities, it is significant for public health to determine whether living status is independently associated with poor clinical outcomes. However, previous studies have exclusively focused on the relationship between living alone and the incidence of T2DM. Therefore, the present study examined the association between living alone and clinical outcomes in T2DM. We used the data from the Action to Control Cardiovascular Risk in Diabetes (ACCORD) study (18) and the ACCORD Follow-On Study (ACCORDION) (19) to assess the association between living arrangements and all-cause mortality in patients with T2DM.

In this study, we found an association between living alone and mortality in T2DM. Unadjusted analysis showed that people living alone had a higher incidence of all-cause mortality (the primary endpoint) and non-fatal stroke. However, there were no differences in CV mortality, non-fatal MI, CHF, or CHD between the two groups. Importantly, living alone in patients with T2DM was independently associated with an increased risk of all-cause mortality after adjusting for confounding variables. However, the adjusted analysis revealed that living alone was not an independent predictor of non-fatal stroke. These results highlight the clinical importance of living status in individuals with T2DM. The association between living alone and increased risk of all-cause mortality was observed among the prespecified subgroups. This phenomenon could be significant for public health in consideration of the increasing incidence of living alone and how society as a whole and its healthcare systems adapt to this transformation.

The incidence of living alone continues to grow in the general population, and the present study showed that 20.28% of the ACCORD participants were living alone. Living status has been suggested as a risk factor for T2DM. T2DM patients are diverse in terms of ethnicity, life behaviors, socioeconomic status, and psychosocial factors that may play a role in the prognosis of T2DM. Previous studies revealed that the association between living alone and mortality persisted significantly, even after controlling for confounding variables (21). However, few studies have investigated the relationship between living alone and prognosis in patients with T2DM. Hence, it is necessary to evaluate the relationship between living status and adverse events in T2DM patients.

The present study demonstrated that T2DM patients living alone were characterized by older age, higher prevalence of cardiovascular events, higher prevalence of smoking habits, higher BMI, and higher levels of total cholesterol and LDL-C than those living with others. These findings suggest that multifactorial lifestyle modification interventions are likely to be effective in improving the prognosis of T2DM patients living alone. Moreover, our results showed that patients living alone had lower quality of life, suggesting a critical need for adjusting treatment and management strategies to improve the quality of life of these patients.

Our results expanded previous understanding and confirmed that living alone is an independent risk factor for the prognosis of T2DM patients in a long-term follow-up. However, the potential mechanisms underlying such association are unclear, several factors have been found to be associated with mortality. Several studies have found that individuals living alone have worse self-perceived health and quality of life, depression, and feelings of loneliness (22–24). Consistent with previous reports, our study further confirmed that T2DM patients living alone experienced more feelings of depression (25.71% vs. 21.99%, p < 0.001). Moreover, numerous studies have shown that depression is associated with a higher mortality rate (25). Katon et al. conducted a study on 4,000 patients with T2DM. Over a 3-year follow-up, the mortality in patients with mild or severe depression was 1.7 and 2.3 times higher, respectively, than that of patients without depression (26). Similar findings were reported by Zhang et al. in a survival analysis using the National Health and Nutrition Examination Survey (NHANES)-I data (27).

The majority of studies have also found that living alone is associated with poor diabetes self-care, and especially poor dietary arrangements (28). Dietary patterns are closely related to the optimal management of T2DM in the general population (28–30). The diverse features of people living alone and complex social and demographic changes could influence the dietary patterns of patients with T2DM. All these factors could affect compliance with the guidelines to optimize nutritional status. The relationship between living alone and dietary patterns has also been discussed previously (31, 32). Although a few studies found some healthy behaviors in patients living alone, most studies found that a larger number of patients living alone are less likely to follow healthy dietary habits, including the intake of diverse foods and the consumption of fruits and vegetables (31, 33). Furthermore, there is a possibility that a decline in motivation and pleasure in cooking and/or eating in people living alone, which often manifests in the cooking of simple meals or the consumption of ready-made food. The likely consequences are difficulty in following healthy eating recommendations and in controlling portion size. Aspects of psychological and mental health related to living alone could also affect food intake, resulting in increased or decreased dietary intake.

Previous studies have found that T2DM patients living alone show poor medication adherence, including to prescribed medications and blood measurements. Strict glycemic management is associated with a decreased risk of diabetes-related complications, especially in individuals who have not suffered years of uncontrolled HbA1c levels (34). Projections from the observational United Kingdom Prospective Diabetes Research (UKPDS 35) proposed that a 1 percent decrease in mean HbA1c would lead to a 14% lower risk of all-cause mortality, 21% lower rate of diabetes-related mortality, and 37% decline in the risk of microvascular complications (35). Spencer et al. suggested that education and support from peers allow T2DM patients to achieve better self-management in the long term, leading to good efficacy of HbA1c control (36). A link between living alone and worse HbA1c management has also been observed in a recent study (37). Our results are consistent with those previous findings, as the mean HbA1c level in T2DM patients living alone in our study was higher than that in patients living with one or more adults.

We also found that the risk of CHF and CHD was significantly higher in individuals living alone among propensity score-matched patients. Recent research has described the association between living alone and the incidence of adverse cardiovascular events (35, 38). The Coronary Revascularization Demonstrating Outcome Study in Kyoto of Acute Myocardial Infarction Registry (CREDO—Kyoto AMI) showed that, in a 5-year follow-up, individuals living alone had higher risk of admission for HF (39). The Reduction of Atherothrombosis for Continued Health (REACH) study also showed that living alone was associated with a higher risk of mortality and CV death (40). A possible explanation is that living alone may increase anxiety and depression, causing more psychological distress, poor handling mechanisms and self-care, less access to healthcare services, and less insistence on guideline-recommended therapy and secondary prevention targets.

The first limitation of this study is that it was a post-hoc, exploratory analysis of the ACCORD data; randomization may break, and residual and uncontrolled confounding may still be present. Additionally, the data included in the present study, derived from clinical trials, may not be representative of real-world populations of patients with T2DM. Third, we were unable to account or adjust for unidentified confounders, such as stress and socioeconomic status. Unfortunately, although the statistical modeling included multiple factors, including psychosocial factors and medical history, we acknowledge that there remains a potential for residual confounding. Fourth, living alone was assessed only once at baseline: we did not re-evaluate the living status during follow-up, during which cohabitation status or social circumstances may have changed. Information from prospective clinical trials is needed to clarify the practical effects of living alone in patients with T2DM.

The present study suggests that living status may be a strong marker for predicting the prognosis of T2DM patients, an observation which warrants confirmation in further studies. The main significance of the present study was the identification of specific high-risk factors affecting the prognosis of T2DM. Therefore, these findings have potential implications for public health. Society as a whole needs to be prepared to the negative effects of the increasing rate of individuals living alone.

Publicly available datasets were analyzed in this study. The datasets are available from the ACCORD/ACCORDIN Research Materials obtained from the National Heart, Lung, and Blood Institute (NHLBI) Biologic Specimen and Data Repository Information Coordinating Center.

The studies involving human participants were reviewed and approved by The Second Xiangya Hospital of Central South University. The patients/participants provided their written informed consent to participate in this study.

This study was completed in collaboration with the following authors: YW and ST defined the study theme and methods. LF, YZ, and JS analyzed the data. LF wrote the paper. ZX and ST edited the paper. All authors have read and approved the final manuscript.

This research was supported by the National Natural Science Foundation of China (81801394 to ST) and the Natural Science Foundation of Hunan Province (2019JJ50878 to ST).

The contents of this report do not necessarily reflect the opinions or views of the ACCORD/ACCORDIN study authors or the NHLBI.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.