PDF
Citation
Print
INTRODUCTION
In recent decades, the number of individuals living alone has risen globally, especially in developed countries, and thus emerged as a significant social issue. The proportion of single-person households has reached 29%, equaling 38 million households, in the United States [1]. Similarly, Korea has seen a rapid increase in the percentage of single-person households, from 15.5% in 2000 to 33.4% in 2021 [2]. This trend of living alone is linked to changes in lifestyle habits, inadequate nutritional support, decreased physical activity, and limited social interactions [3,4]. Several studies have suggested that living alone is associated with adverse health outcomes such as mortality, cardiovascular disease, and type 2 diabetes mellitus (T2DM) [5-8]. However, few studies have specifically examined the association between living alone and the development of diabetic complications.
T2DM is associated with the risk of various micro- and macrovascular complications and is one of the most common causes of end-stage kidney disease (ESKD) [9]. The development and progression of diabetic complications during the natural history of the disease are profoundly influenced by glucose control and lifestyle factors. Favorable lifestyle factors could significantly reduce premature mortality in people with diabetes [10]. Given the importance of lifestyle management in people with T2DM, individuals with diabetes living alone (IDLA) might face challenges in maintaining a balanced diet, performing regular physical activity, and accessing medical care [11]. Considering that the prevalence of diabetic vascular complications was higher among patients with more barriers to self-care, IDLA might be particularly vulnerable to ESKD [12].
ESKD is a significant health burden, causing impaired quality of life and premature mortality [13]. Despite the well-documented effects of lifestyle factors on the progression of chronic kidney disease (CKD) [14], research into the specific link between IDLA and the risk of developing ESKD is absent. Understanding this association is critical because it could inform targeted interventions intended to mitigate the risk of ESKD in this vulnerable population. Therefore, we investigated this association using a nationally representative, large-scale database of nearly 2.5 million people with T2DM.
METHODS
Study population and data source
This study used the Korean National Health Information Database (NHID), a public database established by the National Health Insurance System (NHIS). The NHID contains information on healthcare utilization, healthcare service claims and reimbursements, health examinations, sociodemographic variables, diagnosis and treatment variables, and mortality data for the residents of South Korea. The NHIS covers approximately 97% of the Korean population, with the remaining 3% covered by Medical Aid. Policyholders and their dependents who are 20 years of age or older are recommended to undergo health check-ups every 1 or 2 years, depending on their employment status. Further details of the NHID are described elsewhere [15,16]. In our study, the initial cohort comprised 2,616,828 individuals with T2DM who underwent health examinations in 2015–2016. Exclusions were made for individuals younger than 20 years (n=323) and those with missing household information (n=64,258), with missing variables from health examinations (n=85,299), with a previous diagnosis of ESKD (n=12,759), or with a diagnosis of ESKD within a year after the study enrollment date (n=21,576). Ultimately, the study population contained 2,432,613 subjects (Supplementary Fig. 1). This study was approved by the Institutional Review Board of Seoul St. Mary’s Hospital, The Catholic University of Korea (No. KC23ZASI0241). Anonymized and deidentified information was used for the analyses, and thus the requirement for informed consent was waived.
Measurements and definitions
Single-person households were defined based on the number of family members registered in the NHIS. IDLA was defined as continuously living alone for 5 years or more during the index year and the previous years. Information on current smoking, alcohol drinking, and exercise were obtained via self-administered questionnaires. Smoking status was categorized as nonsmoker, ex-smoker, and current smoker. Alcohol consumption was classified into none, mild, and heavy based on a daily intake of 30 g of alcohol [17]. Regular physical activity was defined as >30 minutes of moderate-intensity exercise at least five times per week or >20 minutes of vigorous-intensity exercise at least three times per week [18]. A healthy lifestyle score ranging from 0 to 3 was calculated by assigning one point each for being a non/ex-smoker, having no alcohol consumption, and performing regular exercise [19]. Low household income level was defined as being in the lowest 25% based on the amount of insurance premiums paid or being a recipient of Medical Aid. The presence of T2DM was defined according to the following criteria: (1) at least one claim per year under International Classification of Diseases, 10th Revision–Clinical Modification (ICD-10) codes E11–14 and at least one claim per year for the prescription of antidiabetic medication, or (2) fasting glucose level ≥126 mg/dL [20]. Obesity was defined as a body mass index (BMI) ≥25 kg/m2, in accordance with the Korean Society for the Study of Obesity criteria [21]. Hypertension was defined as the presence of at least one claim per year under ICD-10 codes I10 or I11 and at least one claim per year for the prescription of an antihypertensive agent or as a systolic/diastolic blood pressure ≥140/90 mm Hg. The presence of dyslipidemia was defined according to the presence of at least one claim per year under ICD-10 code E78 and at least one claim per year for the prescription of a lipid-lowering agent or as total cholesterol (TC) ≥240 mg/dL. CKD was defined as an estimated glomerular filtration rate (eGFR) <60 mL/min/1.73 m2. eGFR was calculated using the abbreviated Modification of Diet in Renal Disease formula: 175×serum creatinine (mg/dL)–1.154×age (yr)–0.203×(0.742 if female) [22]. Blood samples for the measurement of serum glucose, TC, high-density lipoprotein-cholesterol, low-density lipoprotein-cholesterol, triglycerides, and serum creatinine were drawn after an overnight fast. Hospitals performing health examinations are certified by the NHIS, and laboratory investigations were performed in accordance with the guidelines of the Korean Association of Laboratory Quality Control.
Study outcome and follow-up
The primary end point was incident ESKD, which was defined as a combination of ICD-10 codes (N18–19, Z49, Z94.0, or Z99.2), the initiation of renal replacement therapy, and/or kidney transplantation during hospitalization. Patients with ESKD are designated as Medical Aid beneficiaries, and all medical care expenses for dialysis are reimbursed by the Korean Health Insurance Review and Assessment Service. Consequently, the diagnostic codes for ESKD patients are strictly managed, ensuring high accuracy. Codes for treatment or medical expense claims were R3280 for kidney transplantation, O7011–O7020 or V001 for hemodialysis, and O7071–O7075 or V003 for peritoneal dialysis. We excluded individuals without previous CKD who had a transplantation or dialysis code on the same date as an acute kidney injury code. Subjects on continuous kidney replacement therapy or acute peritoneal dialysis due to acute kidney injury were also excluded [23]. The study population was followed from baseline to the date of ESKD diagnosis or December 31, 2022.
Statistical analysis
Baseline characteristics are presented as mean±standard deviation, median (interquartile range [IQR]), or number (%). The incidence rates of the outcome were calculated by dividing the number of incident cases by the total follow-up period (1,000 person-years). The incidence probability of the primary outcome according to the number of household members was calculated using Kaplan-Meier curves, and a log-rank test was performed to analyze differences between groups. Hazard ratios (HRs) and 95% confidence interval (CI) values for ESKD by household status were analyzed using a multivariable Cox proportional hazard regression model. Model 1 was unadjusted; model 2 was adjusted for age and sex; model 3 was further adjusted for BMI, smoking, alcohol drinking, regular exercise, income status, hypertension, dyslipidemia, and eGFR; and model 4 was further adjusted for the duration of diabetes, insulin treatment, and oral antidiabetic medications. A sensitivity analysis of the relationship between the duration of living alone (first time registrants in the index year, those living alone for less than 5 years, and those living alone for 5 years or more) and the risk of ESKD was based on yearly sequential household number data. In addition, the association between the risk of ESKD and the maintenance of healthy lifestyle habits (not drinking, not smoking, and being physically active) was evaluated. Potential effect modifications by age group, sex, smoking, alcohol consumption, regular exercise, presence or absence of obesity, eGFR levels, insulin use, number of oral antidiabetic medications (<3 vs. ≥3), duration of diabetes, and income level were evaluated through stratified analyses and interaction testing with a likelihood ratio test. Statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA), and a P value <0.05 was considered to indicate significance.
RESULTS
Baseline characteristics of the study participants
The characteristics of the participants by household size are described in Table 1. Subjects in the IDLA group were younger, and the proportion of individuals with low income was significantly higher (44.7% in the IDLA group, 18.4% in the non-IDLA group). From a lifestyle perspective, those who consumed alcohol or were current smokers were more frequent, and those who exercised regularly were less frequent in the IDLA group. The prevalence of hypertension, dyslipidemia, and CKD was lower in the IDLA group. Fasting glucose and lipid levels were higher in the IDLA group. Blood pressure, eGFR, and BMI were numerically similar between the groups, but statistically significant differences were observed due to the large sample size. The comparison of participant characteristics by the duration of living alone is described in Supplementary Table 1. The proportions of individuals with unhealthy lifestyle factors and low income were higher among those who had lived alone for a longer period.
The risk of ESKD in individuals with diabetes living alone
During a median follow-up of 6.0 years (IQR, 5.3 to 6.3) after the initial index date, 23,707 participants in the non-IDLA group and 2,984 participants in the IDLA group developed ESKD. The cumulative incidence was higher in the IDLA group (Fig. 1). After adjusting for age and sex, a 21% higher risk of ESKD was observed in the IDLA group (Table 2). After further controlling for other adjusting factors such as income status, smoking, alcohol drinking, regular exercise, BMI, hypertension, dyslipidemia, eGFR, duration of diabetes, and medications, living alone was still an independent predictor of ESKD development (HR, 1.10; 95% CI, 1.06 to 1.14). The incidence and risk of ESKD increased with the duration of living alone, particularly for those who lived alone for more than 5 years (HR, 1.11; 95% CI, 1.07 to 1.16) (Supplementary Table 2). The risk of ESKD was closely related to the level of income in both the IDLA and non-IDLA groups. The risk was increased by 44% in those in the IDLA group who received Medical Aid, compared with people in the highest income quartile who were not living alone. The ESKD risk was also significantly higher in people living alone who were in the lower first or second quartile of household income, but no significant differences were noted in those in the third or fourth quartile of household income (Table 3).
Subgroup analyses
Higher adjusted HRs of incident ESKD were observed in younger individuals (HR, 1.16; 95% CI, 1.11 to 1.22) compared with the older group (HR, 1.03; 95% CI, 0.96 to 1.09). The risk of ESKD in IDLA was more significant in those who consumed alcohol than in non-drinkers. In the non-obese subgroup, IDLA demonstrated a higher risk of ESKD compared to those not living alone (HR, 1.17; 95% CI, 1.11 to 1.24) whereas no significant difference was noted in obese subgroup. In the subgroup analysis by eGFR levels, the risk of ESKD was particularly elevated in the IDLA group with preserved kidney function compared to their non-IDLA counterparts (HR, 1.30; 95% CI, 1.18 to 1.43 in stage 1 CKD; and HR, 1.22; 95% CI, 1.13 to 1.32 in stage 2 CKD). However, for individuals with lower eGFR levels (stage 3–5 CKD), the association between living alone and ESKD risk was not statistically significant (Table 4).
Effect of healthy lifestyle on the risk of ESKD in individuals with diabetes living alone
Adherence to at least one favorable lifestyle habit (non-smoking, abstaining from alcohol, or engaging in regular exercise) reduced the risk of ESKD in both the IDLA and non-IDLA groups (Fig. 2). In the non-IDLA group, there was a 16%, 32%, and 43% risk reduction for ESKD when having one, two, and three healthy lifestyle behaviors, respectively. In the IDLA group, there was a 28%, 46%, and 49% risk reduction for ESKD when having one, two, and three healthy lifestyle behaviors, respectively. The relationship between favorable lifestyle habits and the risk of ESKD was thus more profound in the IDLA group (P for interaction=0.002).
DISCUSSION
In this nationwide population-based cohort study involving nearly 2.5 million participants, we demonstrated that the IDLA group was associated with a higher risk of developing ESKD than the non-IDLA group. After controlling for possible adjusting factors, the ESKD risk was 10% higher in single-person households. The risk of ESKD varied with the duration of living alone and was particularly high in those who had lived alone for more than 5 years. Subgroup analyses revealed higher risks for younger individuals, alcohol drinkers, non-obese subjects, those without underlying CKD, and those with low income. Adherence to favorable lifestyle habits (non-smoking, abstaining from alcohol, and regular exercise) significantly reduced the ESKD risk in both the IDLA and non-IDLA groups, but the effect was more profound in those living alone.
Several studies have investigated the health outcomes associated with living alone. A meta-analysis demonstrated a higher risk of mortality in the general population living alone, and this association was stronger in younger people and in men [7]. Additionally, social isolation has been linked to multiple chronic diseases in older Europeans [24]. Living alone was significantly associated with an increased risk of hypertension in older Chinese men, which is a risk factor for kidney injury [25]. The incidence of T2DM has also been shown to be higher among Koreans and Germans living alone [8,26]. Middle-aged individuals living alone were at a higher risk for mortality and cardiovascular events, which are typical macrovascular complications. The increase in cardiovascular risk among a patient group considered to be at low risk implies that other factors, such as social determinants of health, are involved [6]. However, no previous research explored the relationship between living alone and the risk of developing microvascular complications such as nephropathy, retinopathy, and neuropathy in people with T2DM. Understanding the effects of living alone on these microvascular complications is crucial because they are significant contributors to the overall morbidity and mortality associated with diabetes.
Several plausible mechanisms could explain the association between the risk of ESKD and living alone. We demonstrated a significant association between adherence to favorable lifestyle factors and a decreased risk of ESKD. Notably, as the number of favorable lifestyle habits increased, the risk of ESKD decreased more substantially. In previous studies, it has been estimated that approximately 90% of T2DM cases and 65% of hypertension cases, the two major causes of CKD, could be prevented if people adhered to a healthy lifestyle including non-smoking, abstaining drinking and regular exercise [27,28]. Regular physical activity has been shown to attenuate declines in renal function in both CKD and acute kidney injury patients [29,30]. Smoking has been associated with the prevalence, development, and progression of kidney disease. Additionally, smoking was significantly associated with renal impairment and proteinuria in a healthy population [31]. Our study showed that higher proportions of people in the IDLA group had unhealthy lifestyle patterns, which might make them vulnerable to comorbidities and increase their risk of ESKD, despite their younger age and lower prevalence of CKD. These results are consistent with previous studies indicating a higher prevalence of CKD in young-onset diabetes [32].
The social determinants of health, including social networks, healthcare accessibility, and economic status, are significant issues for individuals living alone. Existing studies have revealed that participation in community organizations is consistently related to better health status, suggesting that social support plays a crucial role in managing diabetes [33,34]. People living alone tend to use healthcare services less frequently than those living with others and exhibit higher levels of depressive symptoms [35]. People living alone saw a doctor less often and were less likely to have a basic health examination than those living with others, even when controlling for their differences in general and mental health. Notably, the IDLA group had a lower prevalence of hypertension, dyslipidemia, and CKD in our study, likely due to their younger age. Considering these baseline characteristics of the IDLA group, it is conceivable that not only physical health, but also socioeconomic factors had a significant effect. The finding of our subgroup analysis which showed higher risk of ESKD in IDLA without obesity and CKD also emphasizes the importance of social determinants of health influencing critical health outcomes. Additionally, we explored the correlation between income and ESKD incidence in this study and found that a lower income was associated with greater risk. Specifically, higher-income individuals within the IDLA group did not exhibit a significant increase in ESKD risk, whereas lower-income individuals showed a markedly elevated risk. This suggests that financial resources may play a protective role against ESKD progression by enhancing access to healthcare, adherence to treatment, and lifestyle management. Institutional measures and systematic approaches are needed to address these issues effectively, particularly by supporting low-erincome IDLA individuals to reduce this health disparity.
Diabetic kidney disease is a heterogeneous disease and has numerous etiologic pathways. The poor lifestyles and lack of social support found in the IDLA group can lead to difficulties in maintaining metabolic homeostasis. Optimizing glucose control plays a pivotal role in slowing the progression of diabetic kidney disease, with evidence suggesting that intensive glucose control can slow eGFR [36]. Persistent hyperglycemia stimulates the production of advanced glycation end-products, which induces the production of pro-inflammatory cytokines via the activation of nuclear transcription factors [37]. In a Japanese study, living alone was associated with higher glycosylated hemoglobin (HbA1c) and visit-to-visit HbA1c variability in people with diabetes [38]. Social engagement and loneliness are also associated with neuroimmune markers. In the English Longitudinal Study of Ageing, higher levels of social engagement and living with somebody were related to lower levels of C-reactive protein (CRP), fibrinogen, and white blood cell count [39]. In the Copenhagen Aging and Midlife Biobank cohort, a strong association between the duration of living alone or accumulated number of partnership breakups and low-grade inflammation was suggested in middle-aged men [40]. Conversely, a significant reduction in CRP levels was observed after patients with T2DM participated in resistance exercise. Other pro-inflammatory cytokines, including tumor necrosis factor-α and interleukin 6, were also reduced [41]. Therefore, living alone and unhealthy lifestyle conditions can trigger inflammatory responses in the body, leading to disrupted metabolic homeostasis in IDLA.
Our study has several limitations. First, this was an observational study, and the association found between single-person households and renal endpoints might not be causal. Those diagnosed with ESKD within 1 year from the index date were excluded to minimize the possible effects of reverse causality. Second, the precise causes of ESKD were not identifiable in our study. Third, CKD stages are ideally categorized by both eGFR and albuminuria. However, our database lacked information on albuminuria, which may affect the precision of CKD staging in our analysis. Fourth, the study population consisted of Korean men and women; therefore, it is uncertain whether these findings can be generalized to other ethnic groups with different cultural backgrounds and healthcare systems. Fifth, the dietary patterns of the participants were not available. Sixth, indices that reflect the control status of T2DM, such as HbA1c or glycated albumin levels, were not included in this database. We sought to address this limitation by incorporating variables that reflect diabetes severity, such as diabetes duration, insulin usage, and the number of oral antidiabetic medications. Finally, some social determinants that could be confounders were not available in this database, such as marital status, reason of living alone, loneliness, employment status, and severity of comorbidities. Despite those limitations, our study has significant strengths. We analyzed a large number of participants, which is only possible in a large-scale nationwide study. Our study extended the scope of research to include health outcomes across all age groups. Additionally, our study’s detailed, stratified analyses provide insights into how various factors, such as age, income, and lifestyle habits, interact with household status to influence ESKD incidence.
In conclusion, this study has comprehensively explored the multifaceted associations among lifestyle factors, socioeconomic status, and the progression to ESKD within the context of varying living arrangements. Individuals living alone exhibited a unique pattern of associations between risk factors and the development of renal complications. Adherence to favorable lifestyle habits could reduce this risk, emphasizing the need for targeted interventions to improve lifestyle and social support for those living alone. Our study suggests the importance of considering social determinants of health when managing diabetes and preventing its complications. We also provide valuable insights into the complex interplay of lifestyle choices, social factors, and health outcomes and call for institutional measures and systematic approaches to address these issues effectively. Policies that support targeted lifestyle interventions, healthcare accessibility, and social support integration for single-person households with diabetes could be crucial in reducing the progression of diabetic complications in this vulnerable group.
XML Download