2023 Volume 70 Issue 3 Pages 305-313
We prospectively analyzed the association between mobile phone usage time and the incidence of diabetic retinopathy (DR) in type 2 diabetes (T2D) among participants.We included a total of 4,371 patients with T2D among the participants. Mobile phone usage time was quantified at baseline by summing up the hours spent on mobile phone use. The types of mobile phone usage time in our study include game time, TikTok time, WeChat time, watching movies or reading time, and online shopping time. We categorized patients into four groups according to different mobile phone usage time: ≤1.5 h/day (n = 1,101), 1.6–3.5 h/day (n = 1,098), 3.6–7.5 h/day (n = 1,095), and >7.6 h/day (n = 1,077). Fundus photography was performed every year from January 2012 to January 2020. During a follow-up of 8 years, 1,119 were affected by DR, resulting in an overall incidence of 25.6%. The incidences of mild nonproliferative DR (NPDR), moderate NPDR, severe NPDR, and proliferative DR (PDR) were 10.1%, 5.1%, 5.1%, and 5.2%, respectively. In comparisons with participants in the lowest category (≤1.5 h/day), the hazard ratios (HRs) of DR were 1.19 (95% confidence interval [CI] 1.07, 1.31, p = 0.040) for 1.6–3.5 h/day, 1.60 (95% CI 1.40, 1.81, p < 0.001) for 3.6–7.5 h/day, and 1.85 (95% CI 1.61, 2.09, p < 0.001) for >7.6 h/day, respectively. Our results provide the general population with a feasible and practical alternative for the reduction of mobile phone use behavior time and new measures to prevent the occurrence of DR.
THE PREVALENCE of type 2 diabetes (T2D) has been increasing, and it is estimated that by 2030, 7,079 per 100,000 individuals worldwide will be affected [1]. Such an escalating trend is partly caused by excessive sedentary behaviors in along with the increasingly popular use of, for example, mobile phones, television, and computers [2]. Trends in mobile phone time have been increaseing significantly in the past decade, and mobile phones have become indispensable [3], such as for mobile payment in China. Frequent use of mobile phones will decrease physical activity and increase sedentary behavior. According to the data from the National Health and Nutrition Examination Survey (NHANES), participants spent an average of 7.7 hours per day, or more than >50% of their monitored time, engaging in sedentary behaviors [4]. Approximately 30% of the population spends more than 6 hours engaging in sedentary behaviors during the working day in the United Kingdom. Recently, the World Health Organization issued a new guideline on sedentary behavior and physical activity, which provides new recommendations for reducing sedentary behaviors [5].
Diabetic retinopathy (DR) is one of the main microvascular complications of diabetes and is caused by long-term damage to the retinal microvasculature. Because the incidence of diabetes continues to rise, the number of patients with DR will increase worldwide. DR can cause severe and permanent visual impairment and is the most common cause of blindness in working-age adults [6]. In the course of 20 years of diabetes, more than 60% of patients with T2D and almost all patients with type 1 diabetes (T1D) will develop some degree of retinopathy [7, 8]. In the report of patients with T2D in the United States, approximately 20% had retinopathy at the time of diabetes diagnosis, and most of them developed some degree of retinopathy in the course of the next decades [9]. In the UKPDS study, 39% of male subjects and 35% of female subjects had some degree of DR when they were diagnosed with diabetes [10].
Currently, there is a great concern about the harmful effects of electromagnetic waves, radiofrequency waves and microwaves generated by mobile phones and their telecommunication stations on health [11]. However, there is insufficient evidence for the quantification of a mobile phone usage time threshold. The effect of mobile phones on T2D has not been adequately elucidated. Moreover, few studies have investigated the interaction between mobile phone use and DR in relation to T2D. In this study, we prospectively analyzed the association between mobile phone usage time and the incidence of DR in T2D among participants.
A total of 5,736 patients with T2D were consecutively recruited among inpatients and outpatients at the Department of Endocrinology of the Nanjing Integrated Traditional Chinese and Western Medicine Hospital Affiliated with Nanjing University of Chinese Medicine from January 2012 to February 2012. T2D was diagnosed according to the 1999 World Health Organization criteria [12]. Inclusion criteria were male or female older than 18 years and younger than 60 years, presence of T2D, and a stable glucose-lowering regimen over the previous three months. Exclusion criteria included the following: HbA1c >9%, diagnosed with DR before enrollment, a history of malignancy, use of mobile phones for more than 12 hours a day for professional reasons, mental disorders, and severe kidney dysfunction. Participants with DR at baseline (n = 282) and those with missing information on T2D and mobile phone use (n = 1,083) at baseline were excluded, leaving a total of 4,371 participants for the main analysis. When the interaction between mobile phone usage time and DR in T2D was examined, only participants with complete data were included in the analysis. Participants provided a wide range of health related information through questionnaires, physical measurements, and biological samples. The study protocol was approved by the ethics committees of Nanjing Integrated Traditional Chinese and Western Medicine Hospital Affiliated with the Nanjing University of Chinese Medicine following the principles of the Helsinki Declaration. Each participant provided signed written informed consent.
Assessment of mobile phone usage timeIn the current analysis, mobile phone usage time was quantified at baseline by summing the hours spent on mobile phone use. The types of mobile phone usage time in our study include game time, TikTok time, WeChat time, watching movies or reading time, and online shopping time. At the baseline assessment, participants were asked, “On a typical day, how many hours do you spend watching a mobile phone?”, “On a typical day, how many hours do you spend using a mobile phone? (Do not include using a mobile phone at work),” and “On a typical day, how many hours do you spend playing games on your phone?” Patients were categorized into four groups according to different mobile phone usage time: ≤1.5 h/day (n = 1,101), 1.6–3.5 h/day (n = 1,098), 3.6–7.5 h/day (n = 1,095), and >7.6 h/day (n = 1,077). The aforementioned information was obtained through a questionnaire survey.
Anthropometric and biochemical measurementsEach patient underwent a physical examination at baseline during the initial visit, including measurements of weight, height, and blood pressure. Comorbidity assessments included hypertension, angina pectoris, myocardial infarction, dyslipidemia, nephropathy, and stroke. BMI was calculated as weight divided by the square of height in meters. A standard mercury sphygmomanometer was used to measure blood pressure three times, and the measurements were averaged. Alcohol intake was assessed using the questionnaire and reported as “never,” “special occasions only,” and “daily or almost daily.” Smoking status was obtained with the use of the questionnaire and reported as “never,” “previous,” or “current.” A healthy diet score was adapted from the American Heart Association guidelines [13]. Triglycerides (TG), total cholesterol, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and uric acid were measured by using a biochemical analyzer with standard enzymatic methods (ADVIA® XPT, Siemens, Germany). The glucose oxidase method was used to determine the fasting blood glucose level. HbA1c was measured by using a NycoCard Hemoglobin A1c analyzer with highperformance liquid chromatography (Axis-shield Poc, Norway). The study period was from January 2012 to January 2020.
Assessment of DRFundus photography was performed by two independent ophthalmologists, using a digital nonmydriatic camera (CR-2 PLUS AF, Canon, Japan), who were blinded to subject characteristics, following the standardized protocol of the Department of Ophthalmology, Nanjing Integrated Traditional Chinese and Western Medicine Hospital Affiliated with Nanjing University of Chinese Medicine. Retinopathy was graded according to the International Classification of DR [13, 14]. The severity of DR was classified as (1) non-DR; (2) mild nonproliferative DR (NPDR, microaneurysms only); (3) moderate NPDR (more than microaneurysms only but less than severe NPDR); (4) severe NPDR (any of the following: one or more quadrants have prominent intraretinal microvascular abnormalities, two or more quadrants have definite venous beads, more than 20 intraretinal hemorrhages in each of four quadrants, and no PDR); and (5) proliferative DR (PDR, one or more of the following: preretinal hemorrhage, vitreous hemorrhage, and retinal neovascularization). To decrease the risk of severe vision loss, we recommend the use of scatter (panretinal) laser photocoagulation when severe NPDR is approaching or just reaching high-risk PDR. Patients with ungradable retinal fundus photographs of both eyes were excluded from the study. Fundus photography was performed every year from January 2012 to January 2020.
Physical activityPhysical activity was assessed using the questionnaire at baseline. Participants were asked about participation in different types of activities during the prior three months. The types of activities included walking for pleasure, light physical activity (i.e., doing housework, gardening, yoga, and shopping), heavy physical activity (e.g., running, climbing, digging, lifting heavy objects, square dancing, or using heavy tools), strenuous sports (long-distance running, swimming, playing basketball or football, and skipping rope), and other exercises (e.g., keeping fit, cycling, and bowling). The average time (hours per day) spent on the different types of activities was calculated by multiplying the reported frequency by the average duration. Walking for pleasure, light physical activity, and heavy physical activity were combined into the category of daily-life activities, and strenuous sports and other exercises were combined into an indicator of structured exercise. Total time spent on activities was calculated by summing the average time spent on the five types of activities.
Statistical analysisThe baseline characteristics of the study population were summarized across the categories of mobile phone usage time as n (%) for categorical variables and means (SDs) for continuous variables. Follow-up time was calculated from the recruitment date to the date of the first diagnosis of PDR, death, or end of the follow-up - whichever came first. The trends of continuous variables were assessed with linear polynomial contrasts in ANOVA for normally distributed variables. We used the Cochran-Armitage trend test to examine trends of rates across groups. Status by severity of DR was treated as an ordinal categorical variable (0 = non-DR, 1 = mild NPDR, 2 = moderate NPDR, 3 = severe NPDR, and 4 = PDR). Cox proportional hazards models were used to estimate the hazard ratios (HRs) and 95% CI for the associations between mobile phone usage time and risk of DR. The proportional hazards assumption was tested by the inclusion of an interaction term between mobile phone usage time and the time variable. No evidence of violations of the assumption was found. In model 1, age, HbA1c, use of insulin, and sex were adjusted. In the multivariable-adjusted model 2, age, HbA1c, use of insulin, sex, smoking status, alcohol intake, healthy diet score, metabolic equivalent of task (METs), diabetes duration, therapeutic compliance, hypertension, stroke, AMI, CHD angina, uric acid, dyslipidemia, cholesterol-lowering medication, and antihypertensive medication were additionally controlled for [5]. Because physical activity and nephropathy are strong mediators for the association between mobile phone usage time and DR in our study, adjustment for physical activity and nephropathy in the model constitutes statistical overcorrection and results in underestimation of the true effect of mobile phone use. Therefore, they were not adjusted in the main analyses. Multinomial logistic regression analyses were performed to evaluate the independent association between mobile phone usage time and different stages of DR (i.e., mild NPDR, moderate NPDR, severe NPDR, and PDR) after controlling for clinical risk factors including age, sex, BMI, diabetes duration, HbA1c, blood pressure, and lipid profile, as well as physical activity, nephropathy, and METs, when indicated. All p values were two sided, and p < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS software version 25.0.
The clinical characteristics of the patients according to the categories of mobile phone usage time are shown in Table 1. Of the 4,371 participants during a follow-up of eight years, 1,119 were affected by DR, resulting in an overall incidence of 25.6%. The incidences of mild NPDR, moderate NPDR, severe NPDR, and PDR were 10.1%, 5.1%, 5.1%, and 5.2%, respectively. Patients with excessive mobile phone usage time had shorter diabetes duration; higher BMI, HbA1c, uric acid, blood pressure, and LDL-C; and less physical activity, METs, and therapeutic compliance. They were more likely to use their mobile phones to play games, watch TikTok and movies, use WeChat, and read or shop online. They were also more likely to be male, young, and current drinkers or smokers.
| Mobile phone usage time, h/day | p value | ||||
|---|---|---|---|---|---|
| ≤1.5 | 1.6–3.5 | 3.6–7.5 | >7.6 | ||
| n | 1,101 | 1,098 | 1,095 | 1,077 | — |
| Age, years | 48.38 (13.69) | 44.88 (20.15) | 43.50 (15.33) | 40.25 (14.63) | 0.791 |
| Men | 564 (51.2) | 540 (49.2) | 585 (53.4) | 597 (55.4) | 0.022 |
| Diabetes duration (years) | 6.63 (2.26) | 7.13 (1.64) | 7.00 (2.73) | 5.85 (2.03) | 0.656 |
| BMI, kg/m2 | 26.00 (3.78) | 26.75 (4.17) | 27.87 (2.90) | 28.63 (3.02) | 0.461 |
| HbA1c, % | 7.56 (1.12) | 7.91 (1.10) | 8.01 (0.72) | 8.24 (0.82) | 0.565 |
| Uric acid (umol/L) | 415.75 (87.24) | 421.63 (93.37) | 438.75 (95.47) | 480.13 (61.81) | 0.442 |
| LDL cholesterol (mmol/L) | 4.53 (0.97) | 4.85 (0.81) | 5.09 (1.12) | 5.13 (1.05) | 0.608 |
| SBP, mmHg | 130.63 (18.21) | 132.50 (18.52) | 136.25 (13.02) | 142.25 (15.92) | 0.523 |
| DBP, mmHg | 74.38 (14.00) | 80.00 (14.64) | 83.75 (8.35) | 87.88 (7.95) | 0.146 |
| Therapeutic compliance % (95% CI) | 40.9 (35.0–46.8) | 40.4 (34.5–46.3) | 37.3 (31.4–43.2) | 36.2 (30.3–42.1) | 0.061 |
| METs | 9.13 (4.73) | 5.75 (2.82) | 3.75 (2.49) | 3.06 (1.37) | 0.002 |
| Healthy diet score | 2.36 (0.54) | 2.34 (0.56) | 2.23 (0.70) | 2.19 (0.62) | 0.924 |
| Mobile phone usage time | |||||
| Game time, h/day | 0.15 (0.18) | 0.79 (0.26) | 1.64 (0.48) | 6.53 (0.97) | <0.001 |
| Tiktok time, h/day | 0.16 (0.14) | 0.63 (0.28) | 1.43 (0.63) | 2.51 (1.03) | <0.001 |
| WeChat time, h/day | 0.23 (0.10) | 0.34 (0.16) | 0.64 (0.20) | 1.45 (0.51) | <0.001 |
| Watching movies, h/day | 0.37 (0.20) | 0.39 (0.15) | 0.66 (0.18) | 1.22 (0.41) | <0.001 |
| Reading time, h/day | 0.14 (0.07) | 0.47 (0.24) | 0.53 (0.19) | 0.94 (0.16) | <0.001 |
| Online shopping, h/day | 0.24 (0.15) | 0.31 (0.11) | 0.38 (0.13) | 0.68 (0.22) | <0.001 |
| Physical activity | |||||
| Light physical activity, h/day | 1.21 (0.79) | 1.28 (0.67) | 0.79 (0.40) | 0.39 (0.34) | 0.017 |
| Heavy physical activity, h/day | 1.07 (0.48) | 1.10 (0.53) | 0.30 (0.16) | 0.17 (0.12) | <0.001 |
| Structured exercise, h/day | 0.72 (0.31) | 0.96 (0.55) | 0.27 (0.16) | 0.22 (0.13) | <0.001 |
| Walking for pleasure, h/day | 0.95 (0.33) | 0.82 (0.30) | 0.61 (0.23) | 0.36 (0.20) | 0.001 |
| Total physical activity, h/day | 2.58 (0.92) | 1.96 (0.92) | 0.91 (0.28) | 0.71 (0.49) | <0.001 |
| Smoking | |||||
| Never | 774 (70.3) | 840 (76.5) | 798 (72.9) | 753 (69.9) | 0.002 |
| Previous | 204 (18.5) | 156 (14.2) | 183 (16.7) | 171 (15.9) | 0.051 |
| Current | 123 (11.2) | 102 (9.3) | 114 (10.4) | 153 (14.2) | 0.002 |
| Alcohol intake | |||||
| Daily or almost daily | 237 (21.5) | 261 (23.8) | 222 (20.3) | 285 (26.5) | 0.003 |
| Special occasions only | 345 (31.3) | 330 (30.1) | 321 (29.3) | 381 (35.4) | 0.012 |
| Never | 519 (47.1) | 507 (46.2) | 552 (50.4) | 411 (38.2) | <0.001 |
| Hypertension | 333 (30.2) | 357 (32.5) | 372 (34.0) | 373 (34.6) | 0.132 |
| Stroke | 228 (20.7) | 240 (21.9) | 249 (22.7) | 254 (23.6) | 0.412 |
| AMI | 122 (11.1) | 138 (12.6) | 150 (13.7) | 146 (13.6) | 0.230 |
| CHD Angina | 195 (17.7) | 201 (18.3) | 216 (19.7) | 231 (21.4) | 0.122 |
| Dyslipidemia | 375 (34.1) | 390 (35.5) | 411 (37.5) | 423 (39.3) | 0.062 |
| Nephropathy | 252 (22.9) | 235 (21.4) | 270 (24.7) | 274 (25.4) | 0.114 |
| Antihypertension medications | 333 (30.2) | 358 (32.6) | 360 (32.9) | 329 (30.5) | 0.419 |
| Cholesterol-lowering medications | 357 (32.4) | 348 (31.7) | 396 (36.2) | 340 (31.6) | 0.073 |
Data are mean (SD), or N (%). DBP, diastolic blood pressure; SBP, systolic blood pressure; METs, metabolic equivalent of task; CI, confidence interval; AMI, acute myocardial infarction; CHD, coronary heart disease.
We found a linear dose-responsive relationship between mobile phone usage time and the risk of DR, with a threshold effect. More mobile phone time of participants was consistently associated with a higher risk of DR across the models (Table 2). Each SD (1 SD = 1.8 h/day) increase in mobile phone usage time was associated with a DR HR of 1.24 (95% CI 1.12, 1.36) after adjustment for age, HbA1c, use of insulin, sex, smoking status, alcohol intake, healthy diet score, METs, diabetes duration, therapeutic compliance, hypertension, stroke, AMI, CHD angina, uric acid, dyslipidemia, cholesterol-lowering medication, and antihypertensive medication. In comparisons with participants in the lowest category (≤1.5 h/day), the HRs of DR were 1.19 (95% CI 1.07, 1.31, p = 0.040) for 1.6–3.5 h/day, 1.60 (95% CI 1.40, 1.81, p < 0.001) for 3.6–7.5 h/day, and 1.85 (95% CI 1.61, 2.09, p < 0.001) for >7.6 h/day, respectively (Table 2).
| Mobile phone usage time (h/day) | n cases/n total (%) | Model 1 | Model 2 | ||
|---|---|---|---|---|---|
| HR (95% CI) | p | HR (95% CI) | p | ||
| ≤1.5 | 180/1,101 (16.3) | 1 (reference) | 1 (reference) | ||
| 1.6–3.5 | 243/1,098 (22.1) | 1.21 (1.08, 1.34) | 0.040 | 1.19 (1.07, 1.31) | 0.040 |
| 3.6–7.5 | 312/1,095 (28.5) | 1.58 (1.32, 1.84) | <0.001 | 1.60 (1.40, 1.81) | <0.001 |
| >7.6 | 384/1,077 (35.7) | 1.91 (1.68, 2.14) | <0.001 | 1.85 (1.61, 2.09) | <0.001 |
| Per SD increase | 1.30 (1.11, 1.49) | <0.001 | 1.24 (1.12, 1.36) | <0.001 | |
| p value for trend | <0.001 | <0.001 | |||
1 SD of mobile phone usage time = 1.8 h/day. Model 1: adjust for age, HbA1c, use of insulin, and sex. Model 2: model 1 adjust plus smoking status, alcohol intake, healthy diet score, METs, diabetes duration, therapeutic compliance, hypertension, stroke, AMI, CHD angina, uric acid, dyslipidemia, cholesterol-lowering medication, and antihypertensive medication.
All of the patients were stratified according to mobile phone usage time. For mild NPDR (p for trend = 0.002), moderate NPDR (p for trend <0.001), severe NPDR (p for trend <0.001), and PDR (p for trend <0.001), the incidence of DR by severity increased with increasing times of mobile phone use (Fig. 1). For example, the incidence of PDR was 3.0% in group 1, 3.8% in group 2, 5.8% in group 3, and 8.4% in group 4. The clinical characteristics of the retinal fundus examination are shown in Table 3. Our study demonstrated that there were significant statistical differences in the incidence of microaneurysms, intraretinal hemorrhages, retinal neovascularization, and vitreous hemorrhage among the four groups (p for trend <0.001; Table 3).
Incidence of diabetic retinopathy by severity, according to different mobile phone usage time
| Clinical features (n) | Mobile phone usage time, h/day | p value for trend | |||
|---|---|---|---|---|---|
| ≤1.5 n = 1,101 |
1.6–3.5 n = 1,098 |
3.6–7.5 n = 1,095 |
>7.6 n = 1,077 |
||
| Microaneurysms | 105 | 132 | 153 | 183 | <0.001 |
| Intraretinal hemorrhages | 57 | 75 | 102 | 144 | <0.001 |
| Definite venous beading | 29 | 42 | 39 | 45 | 0.078 |
| Intraretinal microvascular abnormalities | 30 | 31 | 41 | 43 | 0.052 |
| Retinal neovascularization | 27 | 33 | 57 | 81 | <0.001 |
| Vitreous hemorrhage | 18 | 15 | 39 | 63 | <0.001 |
| Preretinal hemorrhage | 21 | 24 | 33 | 31 | 0.076 |
In a multinomial logistic regression model with less mobile phone usage time as the reference group, significant associations existed between mobile phone usage time and the incidence of DR by severity. When mobile phone usage time was included as a categorical variable in the multinomial logistic regression model, after adjustment for age, sex, BMI, diabetes duration, blood pressure, lipid profile, and HbA1c, the highest mobile phone use time was independently associated with moderate NPDR, severe NPDR, and PDR, compared with the lowest mobile phone usage time (moderate NPDR, odds ratio [OR] 1.85, p = 0.002; severe NPDR, OR 2.16, p < 0.001; PDR, OR 2.21, p < 0.001) (Table 4). The link between mild NPDR and mobile phone usage time, as a categorical variable, did not reach statistical significance after controlling for physical exercise, METs and nephropathy. The significant effect of mobile phone usage time (3.6–7.5 h and >7.6 h) on the presence of PDR remained similar after adjustment for physical exercise, nephropathy and METs (Table 4).
| Mild NPDR | Moderate NPDR | Severe NPDR | PDR | |||||
|---|---|---|---|---|---|---|---|---|
| OR (95% CI) | p value | OR (95% CI) | p value | OR (95% CI) | p value | OR (95% CI) | p value | |
| Model 1 | ||||||||
| ≤1.5 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | ||||
| 1.6–3.5 | 1.12 (0.96–1.40) | 0.402 | 1.22 (0.96–1.48) | 0.221 | 1.12 (0.98–1.25) | 0.171 | 1.02 (0.97–1.07) | 0.462 |
| 3.6–7.5 | 1.34 (0.98–1.70) | 0.127 | 1.42 (0.99–1.85) | 0.102 | 1.54 (1.13–1.95) | 0.010 | 1.43 (1.03–1.83) | 0.042 |
| >7.6 | 1.29 (0.93–1.65) | 0.124 | 1.85 (1.32–2.39) | 0.002 | 2.16 (1.73–2.60) | <0.001 | 2.21 (1.82–2.61) | <0.001 |
| Model 2 | ||||||||
| Physical activity | 1.00 (0.97–1.03) | 0.941 | 1.02 (0.98–1.06) | 0.943 | 1.01 (0.97–1.05) | 0.940 | 1.01 (0.97–1.05) | 0.942 |
| ≤1.5 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | ||||
| 1.6–3.5 | 1.11 (0.95–1.27) | 0.400 | 1.25 (0.97–1.53) | 0.225 | 1.13 (0.98–1.28) | 0.170 | 1.01 (0.97–1.05) | 0.460 |
| 3.6–7.5 | 1.32 (0.97–1.67) | 0.124 | 1.40 (0.96–1.84) | 0.101 | 1.52 (1.11–1.93) | 0.009 | 1.41 (1.02–1.80) | 0.040 |
| >7.6 | 1.28 (0.92–1.64) | 0.112 | 1.83 (1.31–2.35) | 0.001 | 2.14 (1.72–2.56) | <0.001 | 2.20 (1.81–2.59) | <0.001 |
| Model 3 | ||||||||
| Nephropathy | 1.01 (0.95–1.07) | 0.954 | 1.02 (0.97–1.07) | 0.969 | 1.04 (0.98–1.06) | 0.972 | 1.00 (0.94–1.06) | 0.942 |
| ≤1.5 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | ||||
| 1.6–3.5 | 1.20 (0.97–1.43) | 0.424 | 1.24 (0.97–1.51) | 0.250 | 1.15 (0.99–1.31) | 0.197 | 1.06 (0.98–1.14) | 0.481 |
| 3.6–7.5 | 1.41 (0.98–1.84) | 0.156 | 1.45 (0.99–1.91) | 0.120 | 1.57 (1.16–1.98) | 0.012 | 1.45 (1.06–1.84) | 0.046 |
| >7.6 | 1.33 (0.96–1.70) | 0.127 | 1.87 (1.34–2.40) | 0.003 | 2.18 (1.78–2.58) | <0.001 | 2.23 (1.84–2.62) | <0.001 |
| Model 4 | ||||||||
| METs | 1.00 (0.95–1.05) | 0.945 | 0.98 (0.95–1.01) | 0.914 | 1.02 (0.97–1.05) | 0.976 | 0.99 (0.94–1.04) | 0.928 |
| ≤1.5 | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | 1.00 (Reference) | ||||
| 1.6–3.5 | 1.12 (0.94–1.30) | 0.411 | 1.21 (0.95–1.47) | 0.200 | 1.17 (0.98–1.36) | 0.120 | 1.02 (0.94–1.10) | 0.414 |
| 3.6–7.5 | 1.34 (0.97–1.71) | 0.125 | 1.39 (0.96–1.82) | 0.100 | 1.57 (1.15–1.99) | 0.016 | 1.43 (1.03–1.84) | 0.043 |
| >7.6 | 1.29 (0.94–1.64) | 0.126 | 1.81 (1.29–2.33) | 0.001 | 2.19 (1.77–2.61) | <0.001 | 2.21 (1.81–2.61) | <0.001 |
Model 1: adjusted for age, sex, BMI, diabetes duration, blood pressure, lipid profile, and HbA1c. Model 2: including all variables in model 1 plus physical activity; Model 3: including all variables in model 1 plus nephropathy; Model 4: including all variables in model 1 plus METs.
DR is a chronic microvascular complication, sight-threatening, and well-characterized, which eventually affects virtually all patients with diabetes mellitus [15]. DR is characterized by gradual and progressive alterations in the retinal microvasculature that lead to nonperfusion areas of the retinal, increased vascular permeability, and pathologic intraocular proliferation of retinal vessels [16, 17]. Duration of diabetes, lack of appropriate glycemic control, renal disease, hypertension, and elevated serum lipid levels are closely associated with the onset and severity of DR [18-20]. Despite decades of study, there are currently no known methods to prevent DR; DR is still the main cause of new-onset blindness among working-aged people in most developed countries. However, with appropriate ophthalmologic care and medical treatment, more than 90% of vision loss caused by PDR can be prevented [21]. Therefore, before finding an effective cure for DR, the main clinical care focus for preventing vision loss should be appropriately directed at early identification, accurate classification, and timely treatment of DR [22].
In this large prospective cohort study, we provide evidence of a mobile phone usage time independent effect on the presence of DR, and we found independent associations of mobile phone use time with mild NPDR, moderate NPDR, severe NPDR, and PDR. Until now, few data are available regarding the relationship between mobile phone use and patients with DR in T2D. Our findings extend the literature showing that mobile phone usage time is associated with an increased risk of DR in a linear fashion and add evidence on reducing mobile phone use behavior. In addition, the hours of mobile phone use were significantly higher in patients with more advanced DR. In our study, mobile phone usage time was significantly associated with the incidence of DR even after adjustment for clinical risk factors, including HbA1c, duration of diabetes and METs. This indicates that the influence of mobile phone usage time on DR is independent of HbA1c, duration of diabetes, and METs. Interestingly, our analyses showed that mobile phone usage time could significantly increase DR risk, with group 2 (1.6–3.5 h) showing the minimal (18%) risk increment and group 4 (>7.6 h) showing the maximal (86%) risk increment. It is demonstrated in our study that excessive use of mobile phones will increase the incidence of microaneurysms, intraretinal hemorrhages, retinal neovascularization, and vitreous hemorrhage.
Multiple mechanisms may account for the benefits of replacing mobile phone use behavior time with different types of physical activities for DR risk. Excessive use of mobile phones leads to a decrease in exercise time, tear secretion, and blinking times; and an increase in sedentary time, causing poor glycemic control and eye fatigue in patients with T2D. At the same time, the flashing screen strongly stimulates the eyes and exacerbates eye fatigue. A large cohort study in the Chinese population showed that there was a significant positive association between sedentary behavior time and T2D [23]. Sedentary behavior is associated with dyslipidemia, obesity, and decreased insulin sensitivity, which may contribute to the development of DR [24, 25]. A recent study conducted a comprehensive meta-analyzed of 37 controlled trials, and the results showed that physical activity could significantly improve insulin metabolism and postprandial glucose [26], which are closely related to DR. Furthermore, several other studies on physical activity may also explain the underlying mechanisms, including reduction of HOMA of insulin resistance, lipoproteins, adipose tissue gene expression, C-peptide, and fasting insulin [27-29]. The link between mild NPDR and mobile phone usage time did not reach statistical significance after physical exercise was controlled for, which suggests that substituting mobile phone use behavior with physical exercise might confer stronger effects in reducing DR risk.
To the best of our knowledge, this is the first study to evaluate the association between mobile phone usage time and the risk of DR. The main advantages of this study include prospective design, a large sample size, and well documented clinical traits, which increase the reliability of our results. We also considered a wide range of potential confounding factors and performed a sensitivity analysis by excluding the DR cases that developed in the first recruitment. We recognize that the current study has several potential limitations. First, mobile phone use behaviors and physical activity types were self-reported and information bias is inevitable. Second, the measurement of mobile phone usage time with four stages may not represent all stages of the participants. Therefore, our research results should be interpreted with caution. Third, the individuals enrolled in this study were inpatients and outpatients with T2D in our hospitals. Therefore, our findings might not apply to all patients with diabetes in other regions.
In conclusion, we provide evidence that mobile phone usage time is associated with the incidence of DR in T2D, and that this association is independent of HbA1c, duration of diabetes, and METs. Our findings suggest that T2D patients should decrease mobile phone usage time, decrease sedentary behavior time, and increase sports time. Our results provide the general population with a feasible and practical alternative for the reduction of mobile phone use behavior time and new measures to prevent the occurrence of DR. Further prospective studies are warranted to obtain a conclusive description of the role of mobile phone usage time in the onset and progression of DR.
The authors have declared no conflicts of interest.
No specific funding was received from any funding agency in the commercial, public or not-for-profit sectors.
YWZ designed the study protocol and drafted the manuscript. HHH, LJ, and LFC analyzed data and checked the manuscript. All authors have read and approved the content of the manuscript.
None.
