Journal of Physical Activity and Health, 2014, 11, 1156 -1161 http://dx.doi.org/10.1123/jpah.2012-0354 © 2014 Human Kinetics, Inc.
Official Journal of ISPAH www.JPAH-Journal.com ORIGINAL RESEARCH
Accelerometer-Assessed Sedentary and Physical Activity Behavior and its Association With Vision Among U.S. Adults With Diabetes Paul D. Loprinzi, Gina Pariser, and Pradeep Y. Ramulu Background: To examine the association between accelerometer-assessed physical activity and visual acuity among a nationally representative sample of adults with evidence of diabetes. Methods: Six hundred seventy adult participants with diabetes (age 20 to 85) from the 2003–2006 NHANES cycles constituted the analyzed sample. Participants wore an accelerometer for 7 days to quantify time spent in sedentary behavior, light-intensity physical activity and moderate-to-vigorous physical activity. Visual acuity was objectively assessed for each eye. Results: In multivariable models, every 1-hour increment in daily sedentary behavior was associated with 23% greater likelihood (OR = 1.23; 95% CI: 1.01–1.52) of having uncorrected refractive error as opposed to normal sight. Performing more than 5 minutes of daily moderate-to-vigorous physical activity was associated with a 82% lower likelihood of having vision impairment as opposed to normal sight (OR = 0.18; 95% CI: 0.06–0.50) while every 1-hour increment in daily light-intensity physical activity was, after adjustments, independently associated with a 38% lower likelihood of vision impairment (OR = 0.62; 95% CI: 0.42–0.92). Conclusion: People with diabetes spending more time in sedentary behavior and less time performing light or moderate-to-vigorous physical activity are more likely to have poorer vision. Keywords: epidemiology, exercise, population Diabetic retinopathy is a leading cause of visual impairment affecting both younger and older adults.1 Poor visual acuity is frequently the result of microvascular retinal changes and is associated with significant decreases in quality of life.2,3 Additionally, microvascular changes in the retinal arterioles reflect systemic vascular disease which may increase an individual’s risk for cardiovascular and cerebrovascular disease, as well as premature mortality.4–7 One of the major risk factors for developing diabetic retinopathy is an individual’s level of glycemic control.8 Consistent evidence indicates that among those with diabetes, regular participation in physical activity is an effective strategy to regulate glycemic control.9,10 There is also emerging evidence showing that regular engagement in physical activity may result in favorable systemic microvascular changes independent of glycemic control,11,12 which ultimately may positively influence retinal microcirculation. This assertion is supported by findings from Tikellis et al13 who showed that higher self-reported physical activity levels were associated with lower prevalence of arteriovenous nicking, wider venular caliber, and retinopathy. In addition, individuals with higher self-reported physical activity were less likely to have diabetic retinopathy.13 However, other studies which also measured physical activity through self-report did not find an association between physical activity and the progression or development of proliferative retinopathy among individuals with diabetes.14,15 Despite these studies, our understanding of the potential protective link between physical activity and visual outcomes in diabetes remains limited. Previous studies have mostly used self-reported Loprinzi ([email protected]) is with the Center for Health Behavior Research; Dept of Health, Exercise Science, and Recreation Management, School of Applied Sciences, The University of Mississippi, Oxford, MS. Pariser is in the Physical Therapy Program, Bellarmine University, Louisville, KY. Ramulu is with the Wilmer Eye Institute, John Hopkins School of Medicine, Baltimore, MD. 1156
physical activity when assessing the association between physical activity and diabetic visual outcomes, though self-report is very poorly correlated with objective measurement of physical activity.16 Additionally, self-reported physical activity is less predictive of relevant outcomes such as body mass index, cholesterol, skinfold thickness, triglycerides, and measures of insulin resistance than accelerometer-determined activity.17,18 Willis and colleagues19 showed that, among the U.S. general population, individuals with visual impairment spent significantly less time in accelerometerassessed moderate-to-vigorous physical activity than those with normal sight, though specific attention was not given to visual outcomes in participants with diabetes. Because of the rise in diabetes prevalence the Center for Disease Control and Prevention Division of Diabetes Translation predicts that the number of Americans with diabetic retinopathy will triple by 2050.20 Addressing the association between physical activity and visual outcomes among adults with diabetes is particularly important because of the potential for physical activity to help prevent visual complications. As a result, we hypothesize that regular participation in physical activity will be associated with more favorable visual outcomes. Most previous studies of physical activity have focused exclusively on moderate-to-vigorous physical activity, though recent evidence indicates that light-intensity physical activity and too much time spent in sedentary behaviors are also associated with health outcomes.21–23 Examining the association between light-intensity physical activity, sedentary behavior, and visual outcomes in adults with diabetes is particularly important as persons with diabetes are more likely to engage in these behaviors, compared with higher intensity levels of physical activity. If shown to be associated with visual acuity, light-intensity physical activity may be an attractive alternative to higher intensity levels among this population given the long-standing sedentary lifestyles and increased likelihood of concurrent comorbidities that may make it difficult to engage in high intensity levels. Here, we use data from the National Health
PA and Vision Among Adults With Diabetes 1157
and Nutrition Examination Survey (NHANES) to examine whether sedentary behavior, light-intensity physical activity, and moderateto-vigorous physical activity are associated with visual outcomes among adults with evidence of diabetes.
Methods
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
Design and Participants The NHANES is a cross-sectional observational study that collects data on a variety of health parameters to assess the health status of Americans. NHANES is conducted by the National Center for Health Statistics (NCHS), and all procedures for data collection were approved by the NCHS ethics review board. All participants provided written informed consent before data collection. Data from the 2003–2004 and 2005–2006 cycles were used in the current study because these were the only cycles where objectively measure physical activity data are currently publically available. Accelerometry data from these 2 cycles were combined to increase the sample size. NHANES uses a complex, multistage probability design to collect a representative sample of noninstitutionalized U.S. civilians, with oversampling of specific subpopulations, such as older adults and non-Hispanic blacks and Mexican Americans. Participants who were under 20 years of age, did not have evidence of diabetes, had insufficient accelerometry data, and had missing data for the covariates were excluded, leaving 670 participants with evidence of diabetes for analysis.
Assessment of Provisional Diabetes Status Participants were asked several questions related to diabetes during home interviews. Participants were asked 1) if they ever had been told by a doctor or health professional that they had or have diabetes, 2) if they are now taking insulin, and 3) if they are now taking diabetic pills to lower blood sugar (subjects in 2003–2004 cycle only). In the current study, participants who answered yes to any of these 3 questions were considered to have evidence of diabetes. In addition, participants with a blood glycohemoglobin (ie, hemoglobin A1C) of 6.5% or greater were considered to have diabetes.24 A subsample of the NHANES participants were examined in a morning fasting session. Fasting glucose was measured from a blood sample and participants with a fasting glucose level of 126 mg/dL or higher were also considered to have diabetes.25
Measurement of Physical Activity At the mobile examination center, participants who were able to walk were asked to wear an ActiGraph 7164 accelerometer on their right hip for 7 days. Accelerometers were affixed to an elastic belt that was worn around the participant’s waist, near the iliac crest. Participants were asked to wear the accelerometer during all activities, except water-based activities and while sleeping. An accelerometer measures the frequency, intensity, and duration of physical activity by generating an activity count that is proportional to the measured acceleration.26 Estimates for sedentary, light-intensity and moderate-to-vigorous physical activity were summarized in 1-minute time intervals. An activity count below 100 counts per minute (ie, 0–99) was used to classify sedentary behavior;27 activity counts between 100 and 2019 counts per minute were used to classify time spent in light-intensity physical activity; and activity counts greater than or equal to 2020 were
used to classify time spent at moderate-to-vigorous physical activity intensity.28 Participants were considered to meet physical activity guidelines if they engaged in at least 150 minutes of moderate-intensity or 75 minutes of vigorous-intensity physical activity per week or some combination of the two. To account for the combination of moderate and vigorous physical activity, vigorous-intensity was multiplied by 2 before being added to moderate-intensity.16 Therefore, participants could meet guidelines if they engaged in at least 150 minutes of moderate plus 2 × vigorous intensity physical activity per week. For the analyses described here, and to represent habitual physical activity patterns, only those participants with at least 4 days with 10 or more hours per day of wear time were included in the analyses.28 To determine the amount of time the monitor was worn, nonwear was defined by a period of a minimum of 60 consecutive minutes of 0 activity counts, with the allowance of 1 to 2 minutes of activity counts between 0 and 100.28
Measurement of Visual Acuity Detailed methodology of the vision assessment is described elsewhere.19 In summary, presenting visual acuity was assessed for each eye. In subjects who had presenting visual acuity worse than 20/30, corrected lenses were removed (if worn) and objective refraction was measured using an ARK-760 autorefractor (Nidek Co Ltd., Tokyo, Japan). As previously described,19 visual acuity of the better-seeing eye was used to classify participants as having normal sight, uncorrected refraction error (URE), or visual impairment (VI). Participants with presenting visual acuity of 20/40 or better in either eye were classified as having normal sight. Participants with presenting visual acuity worse than 20/40, but postrefraction visual acuity in either eye were 20/40 or better, were classified as having URE. Participants with visual acuity worse than 20/40 after autorefraction, or who self-reported not being able to see light with both eyes open, were classified as having VI. Participants with missing data for presenting acuity in both eyes, or with visual acuity worse than 20/40 in both eyes with no autorefraction in either eye, were excluded from the analysis as they were considered to have incomplete visual acuity data.
Measurement of Covariates Several covariates were selected based on previous research demonstrating a link between these covariates and physical activity and/or vision. Information about age, gender, race-ethnicity, marital status, education, comorbidity status, and income poverty ratio, which is a ratio of family income to poverty threshold, were obtained from a questionnaire. Ranging from 0 to 5, an income poverty ratio less than 1 is considered to be below the poverty threshold. Participants were asked if they ever have been told by a doctor or health care professional whether they had certain comorbidities. A dichotomous comorbidity variable was created, which included not having or having at least 1 of the following chronic diseases/events: arthritis, coronary heart disease, heart attack, congestive heart failure, stroke, cancer, emphysema, and chronic bronchitis. Serum cotinine was measured as a marker of active smoking status or environmental exposure to tobacco (ie, passive smoking). Serum cotinine was measured by an isotope dilutionhigh performance liquid chromatography/atmospheric pressure chemical ionization tandem mass spectrometry. Body mass index was calculated from measured weight and height (weight in
1158 Loprinzi, Pariser, and Ramulu
kilograms divided by the square of height in meters). Blood pressure was obtained from the participants using standard protocols, with elevated blood pressure defined as systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg. High-density lipoprotein cholesterol was measured directly in serum. High sensitivity C-reactive protein concentration was quantified using latex-enhanced nephelometry, with elevated C-reactive protein defined as > 0.3 mg/dL.29 Homocysteine, a marker of endothelial function, was measured using the fluorescence polarization immunoassay, with elevated homocysteine defined as > 10.4 μmol/L for women and > 11.4 μmol/L for men.30 Lastly, glycohemoglobin (hemoglobin A1c) was measured using the Primus instrument, which is a fully automated glycohemoglobin analyzer using high performance liquid chromatography.
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
Data Analysis All statistical analyses were performed using procedures from sample survey data using STATA (version 10.0, College Station, TX) to account for the complex survey design used in NHANES. To account for oversampling and nonresponse, and to provide nationally representative estimates, all analyses included the use of appropriate survey sample weights, clustering and primary sampling units. New sample weights were created for the combined NHANES cycles following analytical guidelines for the continuous NHANES. Means and standard errors were calculated for continuous variables and proportions were calculated for categorical variables. Statistical differences between continuous variables and categorical variables were tested using an adjusted Wald test. Statistical differences between categorical variables were tested with design-based likelihood ratio tests. To examine the association between physical activity intensity and visual acuity, a multinomial logistic regression was used, with visual impairment and uncorrected refraction error serving as the outcome variables, and normal sight serving as the reference group. Three multinomial logistic regression models were computed, 1 for sedentary behavior, 1 for light-intensity physical activity, and then 1 for moderate-to-vigorous physical activity. For each multinomial logistic model, the following covariates were included: age, comorbidity status, cotinine, education, marital status, income poverty ratio, race/ethnicity, gender, body mass index, blood pressure, C-reactive protein, homocysteine, high-density lipoprotein cholesterol, and glycohemoglobin. In addition, in the sedentary behavior and light-intensity models, moderate-to-vigorous physical activity was also controlled for. In the moderate-to-vigorous physical activity model, due to skewed data, moderate-to-vigorous physical activity was dichotomized as being above or below the median value of 5.3 minutes of moderate-to-vigorous physical activity per day. Statistical significance was established as an alpha less than 0.05.
Results Demographic and biological characteristics of the analyzed sample are shown in Table 1. The frequency of comorbid illness differed among participants with VI (67.6% [42.3–92.9]), URE (38.4% [19.5–57.3]) and normal sight (60.6% [53.6–67.7]) (P = .04). In addition, income to poverty threshold differed across individuals with normal sight (2.7 [2.5–2.9]), URE (2.3 [2.1–2.6]), and VI (2.0 [1.4–2.6]) (P = .03). There was no difference in sedentary behavior between those with VI (603.1 min/day [511.3–694.8]) and those with
normal sight (517.3 min/day [504.0–530.5]); however, it did approach significance (P = .06). Participants with VI (229.3 min/ day [189.2–269.3]) spent less time in light-intensity physical activity than those with URE (345.0 min/day [307.6–382.3]) or normal sight (313.4 min/day [299.9–327.0]) (P < .01). Similarly, participants with VI (4.5 min/day [0.6–8.4]) spent less time in moderate-to-vigorous physical activity than those with URE (16.6 min/day [10.6–22.7]) or normal sight (12.5 min/day [10.5–14.5]) (P < .01). Although the design-based likelihood ratio test was not significant (P = .19), those with VI (7.1% [0.00–21.3]) had the lowest proportion of meeting physical activity guidelines compared with those with URE (28.1% [12.8–43.4]) and normal sight (19.2% [13.5–24.9]). Table 2 shows the results for the adjusted multinomial logistic regression examining the association between physical activity, sedentary behavior, and acuity outcomes. With regard to the covariates, only A1C was associated with visual acuity. A1C was marginally independently associated with a greater likelihood of uncorrected refraction error instead of normal sight (OR = 1.20, 95% CI: 0.99–1.45; P = .05), but not with a higher risk of visual impairment (OR = 1.16, 95% CI: 0.88–1.54; P = .26). Each additional hour spent in sedentary behavior was associated with a 23% higher (OR = 1.23; 95% CI: 1.01–1.52) likelihood of having uncorrected refractive error, compared with normal sight (P = .04), but was not significantly associated with a higher likelihood of visual impairment (OR = 1.31, 95% CI: 0.96–1.78). Subjects performing more than the median level of daily moderate-to-vigorous physical activity (> 5.3 min/day) had a 82% lower likelihood of visual impairment (OR = 0.18, 95% CI: 0.06–0.50) as compared with those below the median value for daily moderate-to-vigorous physical activity. Each additional hour spent in light-intensity physical activity was associated with a 38% lower likelihood of visual impairment (OR = 0.62; 95% CI: 0.42–0.92) independent of the impact of moderate-to-vigorous physical activity. Neither light nor moderate-to-vigorous physical activity was associated with the likelihood of uncorrected refractive error (P > .40 for both).
Discussion Using a nationally representative sample of U.S. adults with evidence of diabetes, we examined the association between accelerometer-assessed physical activity and objectively-measured visual outcomes. Our cross-sectional findings indicate that uncorrected refractive error is more common among patients with diabetes spending greater amounts of time in sedentary behavior, while visual impairment was more common in individuals engaging in less physical activity. Both light-intensity and moderate-to-vigorous intensity physical activity were independently associated with visual acuity. Several reasons may explain the observed association between physical activity and vision in this sample of individuals with evidence of diabetes. It is possible that decreased physical activity contributes to poor visual outcomes, that poor vision results in decreased physical activity, or that causality is bidirectional. While these various possibilities cannot be distinguished as part of this cross-sectional study, there are compelling reasons to believe that decreased physical activity may play a causal role in the visual outcomes noted. First, visual impairment may be worse with less physical activity as a result of greater retinal vascular disease (ie, macular ischemia, macular edema, and/or neovascularization).
PA and Vision Among Adults With Diabetes 1159
Table 1 Characteristics of Americans With Evidence of Diabetes, Stratified by Vision Status, National Health and Nutrition Examination Survey, 2003–2006 Normal sight (n = 597)
Uncorrected refractive error (n = 49)
Visual impairment (n = 24)
Pa
Age (yr)
59.7 (58.4–61.0)
61.1 (58.5–63.8)
62.6 (57.2–67.9)
0.31
% female
47.3 (41.5–53.1)
46.8 (29.5–64.0)
67.1 (37.6–96.7)
0.49
Variable Demographic
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
Ethnicity
0.08
% Mexican American
8.5 (4.5–11.9)
11.9 (4.4–19.3)
11.8 (1.1–22.4)
% other Hispanic
3.7 (1.4–6.0)
14.2 (0.0–29.6)
N/A
% Non-Hispanic White
68.2 (60.3–76.1)
42.5 (22.9–62.0)
65.7 (41.1–90.4)
% Non-Hispanic Black
13.4 (9.2–17.6)
21.1 (4.7–37.4)
22.3 (4.2–40.4)
% other race
6.2 (3.4–9.0)
10.2 (0.0–23.7)
N/A
Income-to-poverty ratio
2.7 (2.5–2.9)
2.3 (2.1–2.6)
2.0 (1.4–2.6)
Education
0.03 0.23
% some college or above
44.2 (39.0–49.5)
56.3 (37.3–75.2)
32.6 (4.5–60.8)
Marital status
0.23
% married or living with partner
65.9 (60.6–71.3)
63.4 (48.5–78.3)
40.4 (12.8–68.0)
60.6 (53.6–67.7)
38.4 (19.5–57.3)
67.6 (42.3–92.9)
0.04
30.8 (30.0–31.6)
30.9 (29.7–32.2)
31.1 (28.6–33.6)
0.81
Systolic blood pressure (mm/Hg)
133.9 (131.7–136.1)
130.1 (122.0–138.2)
126.3 (110.9–141.8)
0.31
Diastolic blood pressure (mm/Hg)
67.3 (65.6–69.0)
65.6 (62.3–69.0)
64.0 (57.5–70.5)
0.33
7.1 (6.9–7.3)
7.5 (7.0–7.9)
7.9 (7.0–8.7)
0.06
HDL-cholesterol (mg/dL)
50.1 (48.4–51.8)
52.3 (49.9–54.6)
54.4 (49.9–58.9)
0.08
Serum cotinine (ng/mL)
44.2 (31.1–57.2)
57.4 (20.6–94.2)
70.6 (0.0–149.2)
0.53
0.61 (0.51-71)
0.57 (0.36-0.78)
0.53 (0.13-94)
0.70
52.5 (46.9–58.1)
48.7 (27.6–69.9)
37.5 (9.2–65.9)
0.60
33.8 (28.7–38.9)
23.8 (8.0–39.6)
40.2 (10.6–69.9)
0.34
Homocysteine (μmol/L)
10.0 (9.5–10.5)
11.2 (9.8–12.6)
12.4 (9.7–15.1)
0.08
% elevated homocysteined
25.3 (21.9–28.8)
35.5 (11.1–60.5)
49.3 (23.5–75.1)
0.25
Health Comorbidities % with at least 1 comorbidity BMI
(kg/m2)
HgbA1C (%)
C-reactive protein (mg/dL) % elevated
CRPb
% elevated blood
pressurec
Note. Means adjusted for age, gender, body mass index, comorbidity status, cotinine, education, marital status, poverty-to-income ratio, race-ethnicity, C-reactive protein, blood pressure status (elevated or not elevated), homocysteine, high-density lipoprotein cholesterol, and HgbA1C. Results were not adjusted for the evaluated variable (eg, weighted adjusted age did not adjust for age in the model). a For continuous variables, an adjusted Wald test was used to determine if a significant difference occurred across the 3 visual outcomes. A design based likelihood ratio test was used to test for significance for the categorical independent variables. b > 0.3 mg/dL. c Systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg. d > 10.4 μmol/L for women and > 11.4 μmol/L for men.
Retinal microvascular changes may be a result of retinal arteriole inflammation and/or endothelial function,31 and previous studies have demonstrated an inverse association between physical activity and C-reactive protein,12 an important biomarker of inflammation. In addition, hypertension may exacerbate retinal arteriole inflammation among those with diabetes.32 Another
potential mechanism to explain this relationship includes physical activity-induced changes in endothelial function,11 as measured by homocysteine levels. However, in the current study, physical activity was associated with vision even after controlling for these parameters, suggesting that physical activity may influence vision acuity through other mechanisms, or that these biomarkers do
1160 Loprinzi, Pariser, and Ramulu
Table 2 Odds of Visual Impairment or Uncorrected Refractive Error Conferred by Time in Sedentary, Light-Intensity, and Moderate-to-Vigorous Intensity Activity, Adjusted Multinomial Logistic Regression Model From 2003–2006 National Health and Nutrition Examination Survey Participants Odds ratio (95% CI) [P-value] (n = 670)a Accelerometer-derived variables
Uncorrected refractive error
Visual impairment
Sedentary behavior†
1.23 (1.01–1.52)
1.31 (0.96–1.78)
Light-intensity physical activity†
1.02 (0.78–1.33)
0.62 (0.42–0.92)
Referent
Referent
1.44 (0.58–3.56)
0.18 (0.06–0.50)
Moderate-to-vigorous physical activity < 5.3 min/day (median)
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
≥ 5.3 min/day
Note. Bold indicates statistical significance (P < .05). † Odds ratios and 95% confidence interval expressed as a 1-hr change in sedentary behavior and light-intensity physical activity. a Odds ratios reflect odds of uncorrected error or visual impairment as compared with normal sight as determined in multinomial regression models. Three models were run: one with time in sedentary behavior, a second with time spent in light-intensity physical activity, and a third with moderate-vigorous intensity physical activity (MVPA). For the MVPA model, MVPA was dichotomized as less than 5.3 min/day and more than or equal to 5.3 min/day (median). Less than 5.3 min/day served as the referent group. All models controlled for age, gender, body mass index, comorbidity status, cotinine, education, marital status, income to poverty ratio, race-ethnicity, C-reactive protein, blood pressure (elevated or not elevated), homocysteine, high-density lipoprotein cholesterol, and A1c. In addition, for the sedentary and light-intensity physical activity models, MVPA was also controlled for.
not completely capture the impact of diabetes on inflammation and endothelial dysfunction, respectively. Indeed, C-reactive protein, hypertensive status, and homocysteine were all unassociated with visual impairment in the current study. Physical activity was associated with visual acuity even after controlling for glycohemoglobin, an indicator of the control of blood glucose levels, suggesting that physical activity may positively influence visual acuity even among those with poorer control of blood glucose levels. An important finding of the current study was that light-intensity physical activity was independently associated with less frequent visual impairment. This finding is consistent with emerging evidence showing that light-intensity physical activity is independently associated with other health outcomes,33 such as kidney function.23 Additionally, promoting light-intensity physical activity may be more practical in diabetic patients, as coexisting neuropathy, cardiovascular disease, and nephropathy may limit their ability to engage in high-intensity physical activity. A limitation of the current study includes the relatively small sample size of participants with VI. However, the 3.5% (24/670) of participants in the current study with VI is similar to other studies in the general population (2.5% [145/5722]).19 Additionally, it is possible that the most severely affected patients with diabetes were less likely to participate with the study procedures, particularly the accelerometer testing. If true, our findings may underestimate the magnitude of the association between physical activity and VI. Finally, our classification of diabetes partially relied on self-report of disease or medication use, which may have resulted in some patient misclassification. In summary, our analysis showed that adults with diabetes who were more sedentary and less active were more likely to have poorer visual outcomes. To our knowledge, this is the first study to examine the association between accelerometer-assessed lightand moderate-to-vigorous physical activity, sedentary behavior and objectively-measured visual acuity in a nationally representative sample of adults with evidence of diabetes. Further studies
are needed to confirm our findings and to better understand the mechanisms linking physical activity, sedentary behavior, and visual outcomes.
References 1. Congdon NG, Friedman DS, Lietman T. Important causes of visual impairment in the world today. JAMA. 2003;290(15):2057–2060. PubMed doi:10.1001/jama.290.15.2057 2. Fenwick EK, Xie J, Ratcliffe J, et al. The impact of diabetic retinopathy and diabetic macular edema on health-related quality of life in type 1 and type 2 diabetes. Invest Ophthalmol Vis Sci. 2012;53(2):677–684. PubMed doi:10.1167/iovs.11-8992 3. Mazhar K, Varma R, Choudhury F, McKean-Cowdin R, Shtir CJ, Azen SP. Severity of diabetic retinopathy and health-related quality of life: the Los Angeles Latino Eye Study. Ophthalmology. 2011;118(4):649– 655. PubMed doi:10.1016/j.ophtha.2010.08.003 4. Wong TY, Klein R, Couper DJ, et al. Retinal microvascular abnormalities and incident stroke: the Atherosclerosis Risk in Communities Study. Lancet. 2001;358(9288):1134–1140. PubMed doi:10.1016/ S0140-6736(01)06253-5 5. Wong TY, Klein R, Nieto FJ, et al. Retinal microvascular abnormalities and 10-year cardiovascular mortality: a population-based case-control study. Ophthalmology. 2003;110(5):933–940. PubMed doi:10.1016/ S0161-6420(03)00084-8 6. Wong TY, Klein R, Sharrett AR, et al. Retinal arteriolar narrowing and risk of coronary heart disease in men and women. The Atherosclerosis Risk in Communities Study. JAMA. 2002;287(9):1153–1159. PubMed doi:10.1001/jama.287.9.1153 7. Taylor HR, McCarty CA, Nanjan MB. Vision impairment predicts fiveyear mortality. Trans Am Ophthalmol Soc. 2000;98:91–96, discussion 96–99. PubMed 8. Klein R, Klein BE, Moss SE, Cruickshanks KJ. Relationship of hyperglycemia to the long-term incidence and progression of diabetic retinopathy. Arch Intern Med. 1994;154(19):2169–2178. PubMed doi:10.1001/archinte.1994.00420190068008
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
PA and Vision Among Adults With Diabetes 1161 9. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C, White RD. Physical activity/exercise and type 2 diabetes: a consensus statement from the American Diabetes Association. Diabetes Care. 2006;29(6):1433–1438. PubMed doi:10.2337/dc06-9910 10. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C. Physical activity/exercise and type 2 diabetes. Diabetes Care. 2004;27(10):2518–2539. PubMed doi:10.2337/diacare.27.10.2518 11. Loprinzi PD, Cardinal BJ. Interrelationships among physical activity, depression, homocysteine, and metabolic syndrome with special considerations by sex. Prev Med. 2012;54(6):388–392. PubMed doi:10.1016/j.ypmed.2012.03.016 12. Loprinzi P, Cardinal B, Crespo C, et al. Objectively measured physical activity and C-reactive protein: National Health and Nutrition Examination Survey 2003-2004. Scand J Med Sci Sports. 2013;23(2):164–70. PubMed 13. Tikellis G, Anuradha S, Klein R, Wong TY. Association between physical activity and retinal microvascular signs: the Atherosclerosis Risk in Communities (ARIC) Study. Microcirculation. 2010;17(5):381–393. PubMed 14. Cruickshanks KJ, Moss SE, Klein R, Klein BE. Physical activity and the risk of progression of retinopathy or the development of proliferative retinopathy. Ophthalmology. 1995;102(8):1177–1182. PubMed doi:10.1016/S0161-6420(95)30893-7 15. Cruickshanks KJ, Moss SE, Klein R, Klein BE. Physical activity and proliferative retinopathy in people diagnosed with diabetes before age 30 yr. Diabetes Care. 1992;15(10):1267–1272. PubMed doi:10.2337/ diacare.15.10.1267 16. Tucker JM, Welk GJ, Beyler NK. Physical activity in U.S.: adults compliance with the Physical Activity Guidelines for Americans. Am J Prev Med. 2011;40(4):454–461. PubMed doi:10.1016/j. amepre.2010.12.016 17. Atienza AA, King AC. Comparing self-reported versus objectively measured physical activity behavior: a preliminary investigation of older Filipino American women. Res Q Exerc Sport. 2005;76(3):358– 362. PubMed 18. Shephard RJ. Limits to the measurement of habitual physical activity by questionnaires. Br J Sports Med. 2003;37(3):197–206, discussion 206. PubMed doi:10.1136/bjsm.37.3.197 19. Willis JR, Jefferys JL, Vitale S, Ramulu PY. Visual impairment, uncorrected refractive error, and accelerometer-defined physical activity in the United States. Arch Ophthalmol. 2012;130(3):329–335. PubMed doi:10.1001/archopthalmol.2011.1773 20. Fast Facts. Vision Health Initiative. Division of Diabetes Translation. National Center for Chronic Disease Prevention and Health Promotion. http://www.cdc.gov/visionhealth Updated June 1, 2012. Accessed Jan 13, 2013. 21. Owen N, Sparling PB, Healy GN, Dunstan DW, Matthews CE. Sedentary behavior: emerging evidence for a new health risk. Mayo Clin Proc. 2010;85(12):1138–1141. PubMed doi:10.4065/mcp.2010.0444
22. Owen N, Healy GN, Matthews CE, Dunstan DW. Too much sitting: the population health science of sedentary behavior. Exerc Sport Sci Rev. 2010;38(3):105–113. PubMed doi:10.1097/JES.0b013e3181e373a2 23. Hawkins MS, Sevick MA, Richardson CR, Fried LF, Arena VC, Kriska AM. Association between physical activity and kidney function: National Health and Nutrition Examination Survey. Med Sci Sports Exerc. 2011;43(8):1457–1464. PubMed doi:10.1249/ MSS.0b013e31820c0130 24. Summary of revisions for the 2010 Clinical Practice Recommendations. Diabetes Care. 2010;33(Suppl 1):S3. PubMed doi:10.2337/ dc10-S003 25. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33(Suppl 1):S62–S69. PubMed doi:10.2337/dc10-S062 26. Chen KY, Bassett DR, Jr. The technology of accelerometrybased activity monitors: current and future. Med Sci Sports Exerc. 2005;37(11, Suppl):S490–S500. PubMed doi:10.1249/01. mss.0000185571.49104.82 27. Matthews CE, Chen KY, Freedson PS, et al. Amount of time spent in sedentary behaviors in the United States, 2003-2004. Am J Epidemiol. 2008;167(7):875–881. PubMed doi:10.1093/aje/kwm390 28. Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008;40(1):181–188. PubMed doi:10.1249/ mss.0b013e31815a51b3 29. Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003;107(3):499–511. PubMed doi:10.1161/01. CIR.0000052939.59093.45 30. Selhub J, Jacques PF, Rosenberg IH, et al. Serum total homocysteine concentrations in the third National Health and Nutrition Examination Survey (1991-1994): population reference ranges and contribution of vitamin status to high serum concentrations. Ann Intern Med. 1999;131(5):331–339. PubMed doi:10.7326/0003-4819-131-5199909070-00003 31. Wong TY, Klein R, Klein BE, Tielsch JM, Hubbard L, Nieto FJ. Retinal microvascular abnormalities and their relationship with hypertension, cardiovascular disease, and mortality. Surv Ophthalmol. 2001;46(1):59–80. PubMed doi:10.1016/S0039-6257(01)00234-X 32. Mohamed IN, Soliman SA, Alhusban A, et al. Diabetes exacerbates retinal oxidative stress, inflammation, and microvascular degeneration in spontaneously hypertensive rats. Mol Vis. 2012;18:1457–1466. PubMed 33. Camhi SM, Sisson SB, Johnson WD, Katzmarzyk PT, Tudor-Locke C. Accelerometer-determined moderate intensity lifestyle activity and cardiometabolic health. Prev Med. 2011;52(5):358–360. PubMed doi:10.1016/j.ypmed.2011.01.030
Official Journal of ISPAH www.JPAH-Journal.com ORIGINAL RESEARCH
Accelerometer-Assessed Sedentary and Physical Activity Behavior and its Association With Vision Among U.S. Adults With Diabetes Paul D. Loprinzi, Gina Pariser, and Pradeep Y. Ramulu Background: To examine the association between accelerometer-assessed physical activity and visual acuity among a nationally representative sample of adults with evidence of diabetes. Methods: Six hundred seventy adult participants with diabetes (age 20 to 85) from the 2003–2006 NHANES cycles constituted the analyzed sample. Participants wore an accelerometer for 7 days to quantify time spent in sedentary behavior, light-intensity physical activity and moderate-to-vigorous physical activity. Visual acuity was objectively assessed for each eye. Results: In multivariable models, every 1-hour increment in daily sedentary behavior was associated with 23% greater likelihood (OR = 1.23; 95% CI: 1.01–1.52) of having uncorrected refractive error as opposed to normal sight. Performing more than 5 minutes of daily moderate-to-vigorous physical activity was associated with a 82% lower likelihood of having vision impairment as opposed to normal sight (OR = 0.18; 95% CI: 0.06–0.50) while every 1-hour increment in daily light-intensity physical activity was, after adjustments, independently associated with a 38% lower likelihood of vision impairment (OR = 0.62; 95% CI: 0.42–0.92). Conclusion: People with diabetes spending more time in sedentary behavior and less time performing light or moderate-to-vigorous physical activity are more likely to have poorer vision. Keywords: epidemiology, exercise, population Diabetic retinopathy is a leading cause of visual impairment affecting both younger and older adults.1 Poor visual acuity is frequently the result of microvascular retinal changes and is associated with significant decreases in quality of life.2,3 Additionally, microvascular changes in the retinal arterioles reflect systemic vascular disease which may increase an individual’s risk for cardiovascular and cerebrovascular disease, as well as premature mortality.4–7 One of the major risk factors for developing diabetic retinopathy is an individual’s level of glycemic control.8 Consistent evidence indicates that among those with diabetes, regular participation in physical activity is an effective strategy to regulate glycemic control.9,10 There is also emerging evidence showing that regular engagement in physical activity may result in favorable systemic microvascular changes independent of glycemic control,11,12 which ultimately may positively influence retinal microcirculation. This assertion is supported by findings from Tikellis et al13 who showed that higher self-reported physical activity levels were associated with lower prevalence of arteriovenous nicking, wider venular caliber, and retinopathy. In addition, individuals with higher self-reported physical activity were less likely to have diabetic retinopathy.13 However, other studies which also measured physical activity through self-report did not find an association between physical activity and the progression or development of proliferative retinopathy among individuals with diabetes.14,15 Despite these studies, our understanding of the potential protective link between physical activity and visual outcomes in diabetes remains limited. Previous studies have mostly used self-reported Loprinzi ([email protected]) is with the Center for Health Behavior Research; Dept of Health, Exercise Science, and Recreation Management, School of Applied Sciences, The University of Mississippi, Oxford, MS. Pariser is in the Physical Therapy Program, Bellarmine University, Louisville, KY. Ramulu is with the Wilmer Eye Institute, John Hopkins School of Medicine, Baltimore, MD. 1156
physical activity when assessing the association between physical activity and diabetic visual outcomes, though self-report is very poorly correlated with objective measurement of physical activity.16 Additionally, self-reported physical activity is less predictive of relevant outcomes such as body mass index, cholesterol, skinfold thickness, triglycerides, and measures of insulin resistance than accelerometer-determined activity.17,18 Willis and colleagues19 showed that, among the U.S. general population, individuals with visual impairment spent significantly less time in accelerometerassessed moderate-to-vigorous physical activity than those with normal sight, though specific attention was not given to visual outcomes in participants with diabetes. Because of the rise in diabetes prevalence the Center for Disease Control and Prevention Division of Diabetes Translation predicts that the number of Americans with diabetic retinopathy will triple by 2050.20 Addressing the association between physical activity and visual outcomes among adults with diabetes is particularly important because of the potential for physical activity to help prevent visual complications. As a result, we hypothesize that regular participation in physical activity will be associated with more favorable visual outcomes. Most previous studies of physical activity have focused exclusively on moderate-to-vigorous physical activity, though recent evidence indicates that light-intensity physical activity and too much time spent in sedentary behaviors are also associated with health outcomes.21–23 Examining the association between light-intensity physical activity, sedentary behavior, and visual outcomes in adults with diabetes is particularly important as persons with diabetes are more likely to engage in these behaviors, compared with higher intensity levels of physical activity. If shown to be associated with visual acuity, light-intensity physical activity may be an attractive alternative to higher intensity levels among this population given the long-standing sedentary lifestyles and increased likelihood of concurrent comorbidities that may make it difficult to engage in high intensity levels. Here, we use data from the National Health
PA and Vision Among Adults With Diabetes 1157
and Nutrition Examination Survey (NHANES) to examine whether sedentary behavior, light-intensity physical activity, and moderateto-vigorous physical activity are associated with visual outcomes among adults with evidence of diabetes.
Methods
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
Design and Participants The NHANES is a cross-sectional observational study that collects data on a variety of health parameters to assess the health status of Americans. NHANES is conducted by the National Center for Health Statistics (NCHS), and all procedures for data collection were approved by the NCHS ethics review board. All participants provided written informed consent before data collection. Data from the 2003–2004 and 2005–2006 cycles were used in the current study because these were the only cycles where objectively measure physical activity data are currently publically available. Accelerometry data from these 2 cycles were combined to increase the sample size. NHANES uses a complex, multistage probability design to collect a representative sample of noninstitutionalized U.S. civilians, with oversampling of specific subpopulations, such as older adults and non-Hispanic blacks and Mexican Americans. Participants who were under 20 years of age, did not have evidence of diabetes, had insufficient accelerometry data, and had missing data for the covariates were excluded, leaving 670 participants with evidence of diabetes for analysis.
Assessment of Provisional Diabetes Status Participants were asked several questions related to diabetes during home interviews. Participants were asked 1) if they ever had been told by a doctor or health professional that they had or have diabetes, 2) if they are now taking insulin, and 3) if they are now taking diabetic pills to lower blood sugar (subjects in 2003–2004 cycle only). In the current study, participants who answered yes to any of these 3 questions were considered to have evidence of diabetes. In addition, participants with a blood glycohemoglobin (ie, hemoglobin A1C) of 6.5% or greater were considered to have diabetes.24 A subsample of the NHANES participants were examined in a morning fasting session. Fasting glucose was measured from a blood sample and participants with a fasting glucose level of 126 mg/dL or higher were also considered to have diabetes.25
Measurement of Physical Activity At the mobile examination center, participants who were able to walk were asked to wear an ActiGraph 7164 accelerometer on their right hip for 7 days. Accelerometers were affixed to an elastic belt that was worn around the participant’s waist, near the iliac crest. Participants were asked to wear the accelerometer during all activities, except water-based activities and while sleeping. An accelerometer measures the frequency, intensity, and duration of physical activity by generating an activity count that is proportional to the measured acceleration.26 Estimates for sedentary, light-intensity and moderate-to-vigorous physical activity were summarized in 1-minute time intervals. An activity count below 100 counts per minute (ie, 0–99) was used to classify sedentary behavior;27 activity counts between 100 and 2019 counts per minute were used to classify time spent in light-intensity physical activity; and activity counts greater than or equal to 2020 were
used to classify time spent at moderate-to-vigorous physical activity intensity.28 Participants were considered to meet physical activity guidelines if they engaged in at least 150 minutes of moderate-intensity or 75 minutes of vigorous-intensity physical activity per week or some combination of the two. To account for the combination of moderate and vigorous physical activity, vigorous-intensity was multiplied by 2 before being added to moderate-intensity.16 Therefore, participants could meet guidelines if they engaged in at least 150 minutes of moderate plus 2 × vigorous intensity physical activity per week. For the analyses described here, and to represent habitual physical activity patterns, only those participants with at least 4 days with 10 or more hours per day of wear time were included in the analyses.28 To determine the amount of time the monitor was worn, nonwear was defined by a period of a minimum of 60 consecutive minutes of 0 activity counts, with the allowance of 1 to 2 minutes of activity counts between 0 and 100.28
Measurement of Visual Acuity Detailed methodology of the vision assessment is described elsewhere.19 In summary, presenting visual acuity was assessed for each eye. In subjects who had presenting visual acuity worse than 20/30, corrected lenses were removed (if worn) and objective refraction was measured using an ARK-760 autorefractor (Nidek Co Ltd., Tokyo, Japan). As previously described,19 visual acuity of the better-seeing eye was used to classify participants as having normal sight, uncorrected refraction error (URE), or visual impairment (VI). Participants with presenting visual acuity of 20/40 or better in either eye were classified as having normal sight. Participants with presenting visual acuity worse than 20/40, but postrefraction visual acuity in either eye were 20/40 or better, were classified as having URE. Participants with visual acuity worse than 20/40 after autorefraction, or who self-reported not being able to see light with both eyes open, were classified as having VI. Participants with missing data for presenting acuity in both eyes, or with visual acuity worse than 20/40 in both eyes with no autorefraction in either eye, were excluded from the analysis as they were considered to have incomplete visual acuity data.
Measurement of Covariates Several covariates were selected based on previous research demonstrating a link between these covariates and physical activity and/or vision. Information about age, gender, race-ethnicity, marital status, education, comorbidity status, and income poverty ratio, which is a ratio of family income to poverty threshold, were obtained from a questionnaire. Ranging from 0 to 5, an income poverty ratio less than 1 is considered to be below the poverty threshold. Participants were asked if they ever have been told by a doctor or health care professional whether they had certain comorbidities. A dichotomous comorbidity variable was created, which included not having or having at least 1 of the following chronic diseases/events: arthritis, coronary heart disease, heart attack, congestive heart failure, stroke, cancer, emphysema, and chronic bronchitis. Serum cotinine was measured as a marker of active smoking status or environmental exposure to tobacco (ie, passive smoking). Serum cotinine was measured by an isotope dilutionhigh performance liquid chromatography/atmospheric pressure chemical ionization tandem mass spectrometry. Body mass index was calculated from measured weight and height (weight in
1158 Loprinzi, Pariser, and Ramulu
kilograms divided by the square of height in meters). Blood pressure was obtained from the participants using standard protocols, with elevated blood pressure defined as systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg. High-density lipoprotein cholesterol was measured directly in serum. High sensitivity C-reactive protein concentration was quantified using latex-enhanced nephelometry, with elevated C-reactive protein defined as > 0.3 mg/dL.29 Homocysteine, a marker of endothelial function, was measured using the fluorescence polarization immunoassay, with elevated homocysteine defined as > 10.4 μmol/L for women and > 11.4 μmol/L for men.30 Lastly, glycohemoglobin (hemoglobin A1c) was measured using the Primus instrument, which is a fully automated glycohemoglobin analyzer using high performance liquid chromatography.
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
Data Analysis All statistical analyses were performed using procedures from sample survey data using STATA (version 10.0, College Station, TX) to account for the complex survey design used in NHANES. To account for oversampling and nonresponse, and to provide nationally representative estimates, all analyses included the use of appropriate survey sample weights, clustering and primary sampling units. New sample weights were created for the combined NHANES cycles following analytical guidelines for the continuous NHANES. Means and standard errors were calculated for continuous variables and proportions were calculated for categorical variables. Statistical differences between continuous variables and categorical variables were tested using an adjusted Wald test. Statistical differences between categorical variables were tested with design-based likelihood ratio tests. To examine the association between physical activity intensity and visual acuity, a multinomial logistic regression was used, with visual impairment and uncorrected refraction error serving as the outcome variables, and normal sight serving as the reference group. Three multinomial logistic regression models were computed, 1 for sedentary behavior, 1 for light-intensity physical activity, and then 1 for moderate-to-vigorous physical activity. For each multinomial logistic model, the following covariates were included: age, comorbidity status, cotinine, education, marital status, income poverty ratio, race/ethnicity, gender, body mass index, blood pressure, C-reactive protein, homocysteine, high-density lipoprotein cholesterol, and glycohemoglobin. In addition, in the sedentary behavior and light-intensity models, moderate-to-vigorous physical activity was also controlled for. In the moderate-to-vigorous physical activity model, due to skewed data, moderate-to-vigorous physical activity was dichotomized as being above or below the median value of 5.3 minutes of moderate-to-vigorous physical activity per day. Statistical significance was established as an alpha less than 0.05.
Results Demographic and biological characteristics of the analyzed sample are shown in Table 1. The frequency of comorbid illness differed among participants with VI (67.6% [42.3–92.9]), URE (38.4% [19.5–57.3]) and normal sight (60.6% [53.6–67.7]) (P = .04). In addition, income to poverty threshold differed across individuals with normal sight (2.7 [2.5–2.9]), URE (2.3 [2.1–2.6]), and VI (2.0 [1.4–2.6]) (P = .03). There was no difference in sedentary behavior between those with VI (603.1 min/day [511.3–694.8]) and those with
normal sight (517.3 min/day [504.0–530.5]); however, it did approach significance (P = .06). Participants with VI (229.3 min/ day [189.2–269.3]) spent less time in light-intensity physical activity than those with URE (345.0 min/day [307.6–382.3]) or normal sight (313.4 min/day [299.9–327.0]) (P < .01). Similarly, participants with VI (4.5 min/day [0.6–8.4]) spent less time in moderate-to-vigorous physical activity than those with URE (16.6 min/day [10.6–22.7]) or normal sight (12.5 min/day [10.5–14.5]) (P < .01). Although the design-based likelihood ratio test was not significant (P = .19), those with VI (7.1% [0.00–21.3]) had the lowest proportion of meeting physical activity guidelines compared with those with URE (28.1% [12.8–43.4]) and normal sight (19.2% [13.5–24.9]). Table 2 shows the results for the adjusted multinomial logistic regression examining the association between physical activity, sedentary behavior, and acuity outcomes. With regard to the covariates, only A1C was associated with visual acuity. A1C was marginally independently associated with a greater likelihood of uncorrected refraction error instead of normal sight (OR = 1.20, 95% CI: 0.99–1.45; P = .05), but not with a higher risk of visual impairment (OR = 1.16, 95% CI: 0.88–1.54; P = .26). Each additional hour spent in sedentary behavior was associated with a 23% higher (OR = 1.23; 95% CI: 1.01–1.52) likelihood of having uncorrected refractive error, compared with normal sight (P = .04), but was not significantly associated with a higher likelihood of visual impairment (OR = 1.31, 95% CI: 0.96–1.78). Subjects performing more than the median level of daily moderate-to-vigorous physical activity (> 5.3 min/day) had a 82% lower likelihood of visual impairment (OR = 0.18, 95% CI: 0.06–0.50) as compared with those below the median value for daily moderate-to-vigorous physical activity. Each additional hour spent in light-intensity physical activity was associated with a 38% lower likelihood of visual impairment (OR = 0.62; 95% CI: 0.42–0.92) independent of the impact of moderate-to-vigorous physical activity. Neither light nor moderate-to-vigorous physical activity was associated with the likelihood of uncorrected refractive error (P > .40 for both).
Discussion Using a nationally representative sample of U.S. adults with evidence of diabetes, we examined the association between accelerometer-assessed physical activity and objectively-measured visual outcomes. Our cross-sectional findings indicate that uncorrected refractive error is more common among patients with diabetes spending greater amounts of time in sedentary behavior, while visual impairment was more common in individuals engaging in less physical activity. Both light-intensity and moderate-to-vigorous intensity physical activity were independently associated with visual acuity. Several reasons may explain the observed association between physical activity and vision in this sample of individuals with evidence of diabetes. It is possible that decreased physical activity contributes to poor visual outcomes, that poor vision results in decreased physical activity, or that causality is bidirectional. While these various possibilities cannot be distinguished as part of this cross-sectional study, there are compelling reasons to believe that decreased physical activity may play a causal role in the visual outcomes noted. First, visual impairment may be worse with less physical activity as a result of greater retinal vascular disease (ie, macular ischemia, macular edema, and/or neovascularization).
PA and Vision Among Adults With Diabetes 1159
Table 1 Characteristics of Americans With Evidence of Diabetes, Stratified by Vision Status, National Health and Nutrition Examination Survey, 2003–2006 Normal sight (n = 597)
Uncorrected refractive error (n = 49)
Visual impairment (n = 24)
Pa
Age (yr)
59.7 (58.4–61.0)
61.1 (58.5–63.8)
62.6 (57.2–67.9)
0.31
% female
47.3 (41.5–53.1)
46.8 (29.5–64.0)
67.1 (37.6–96.7)
0.49
Variable Demographic
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
Ethnicity
0.08
% Mexican American
8.5 (4.5–11.9)
11.9 (4.4–19.3)
11.8 (1.1–22.4)
% other Hispanic
3.7 (1.4–6.0)
14.2 (0.0–29.6)
N/A
% Non-Hispanic White
68.2 (60.3–76.1)
42.5 (22.9–62.0)
65.7 (41.1–90.4)
% Non-Hispanic Black
13.4 (9.2–17.6)
21.1 (4.7–37.4)
22.3 (4.2–40.4)
% other race
6.2 (3.4–9.0)
10.2 (0.0–23.7)
N/A
Income-to-poverty ratio
2.7 (2.5–2.9)
2.3 (2.1–2.6)
2.0 (1.4–2.6)
Education
0.03 0.23
% some college or above
44.2 (39.0–49.5)
56.3 (37.3–75.2)
32.6 (4.5–60.8)
Marital status
0.23
% married or living with partner
65.9 (60.6–71.3)
63.4 (48.5–78.3)
40.4 (12.8–68.0)
60.6 (53.6–67.7)
38.4 (19.5–57.3)
67.6 (42.3–92.9)
0.04
30.8 (30.0–31.6)
30.9 (29.7–32.2)
31.1 (28.6–33.6)
0.81
Systolic blood pressure (mm/Hg)
133.9 (131.7–136.1)
130.1 (122.0–138.2)
126.3 (110.9–141.8)
0.31
Diastolic blood pressure (mm/Hg)
67.3 (65.6–69.0)
65.6 (62.3–69.0)
64.0 (57.5–70.5)
0.33
7.1 (6.9–7.3)
7.5 (7.0–7.9)
7.9 (7.0–8.7)
0.06
HDL-cholesterol (mg/dL)
50.1 (48.4–51.8)
52.3 (49.9–54.6)
54.4 (49.9–58.9)
0.08
Serum cotinine (ng/mL)
44.2 (31.1–57.2)
57.4 (20.6–94.2)
70.6 (0.0–149.2)
0.53
0.61 (0.51-71)
0.57 (0.36-0.78)
0.53 (0.13-94)
0.70
52.5 (46.9–58.1)
48.7 (27.6–69.9)
37.5 (9.2–65.9)
0.60
33.8 (28.7–38.9)
23.8 (8.0–39.6)
40.2 (10.6–69.9)
0.34
Homocysteine (μmol/L)
10.0 (9.5–10.5)
11.2 (9.8–12.6)
12.4 (9.7–15.1)
0.08
% elevated homocysteined
25.3 (21.9–28.8)
35.5 (11.1–60.5)
49.3 (23.5–75.1)
0.25
Health Comorbidities % with at least 1 comorbidity BMI
(kg/m2)
HgbA1C (%)
C-reactive protein (mg/dL) % elevated
CRPb
% elevated blood
pressurec
Note. Means adjusted for age, gender, body mass index, comorbidity status, cotinine, education, marital status, poverty-to-income ratio, race-ethnicity, C-reactive protein, blood pressure status (elevated or not elevated), homocysteine, high-density lipoprotein cholesterol, and HgbA1C. Results were not adjusted for the evaluated variable (eg, weighted adjusted age did not adjust for age in the model). a For continuous variables, an adjusted Wald test was used to determine if a significant difference occurred across the 3 visual outcomes. A design based likelihood ratio test was used to test for significance for the categorical independent variables. b > 0.3 mg/dL. c Systolic blood pressure ≥ 140 mmHg or diastolic blood pressure ≥ 90 mmHg. d > 10.4 μmol/L for women and > 11.4 μmol/L for men.
Retinal microvascular changes may be a result of retinal arteriole inflammation and/or endothelial function,31 and previous studies have demonstrated an inverse association between physical activity and C-reactive protein,12 an important biomarker of inflammation. In addition, hypertension may exacerbate retinal arteriole inflammation among those with diabetes.32 Another
potential mechanism to explain this relationship includes physical activity-induced changes in endothelial function,11 as measured by homocysteine levels. However, in the current study, physical activity was associated with vision even after controlling for these parameters, suggesting that physical activity may influence vision acuity through other mechanisms, or that these biomarkers do
1160 Loprinzi, Pariser, and Ramulu
Table 2 Odds of Visual Impairment or Uncorrected Refractive Error Conferred by Time in Sedentary, Light-Intensity, and Moderate-to-Vigorous Intensity Activity, Adjusted Multinomial Logistic Regression Model From 2003–2006 National Health and Nutrition Examination Survey Participants Odds ratio (95% CI) [P-value] (n = 670)a Accelerometer-derived variables
Uncorrected refractive error
Visual impairment
Sedentary behavior†
1.23 (1.01–1.52)
1.31 (0.96–1.78)
Light-intensity physical activity†
1.02 (0.78–1.33)
0.62 (0.42–0.92)
Referent
Referent
1.44 (0.58–3.56)
0.18 (0.06–0.50)
Moderate-to-vigorous physical activity < 5.3 min/day (median)
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
≥ 5.3 min/day
Note. Bold indicates statistical significance (P < .05). † Odds ratios and 95% confidence interval expressed as a 1-hr change in sedentary behavior and light-intensity physical activity. a Odds ratios reflect odds of uncorrected error or visual impairment as compared with normal sight as determined in multinomial regression models. Three models were run: one with time in sedentary behavior, a second with time spent in light-intensity physical activity, and a third with moderate-vigorous intensity physical activity (MVPA). For the MVPA model, MVPA was dichotomized as less than 5.3 min/day and more than or equal to 5.3 min/day (median). Less than 5.3 min/day served as the referent group. All models controlled for age, gender, body mass index, comorbidity status, cotinine, education, marital status, income to poverty ratio, race-ethnicity, C-reactive protein, blood pressure (elevated or not elevated), homocysteine, high-density lipoprotein cholesterol, and A1c. In addition, for the sedentary and light-intensity physical activity models, MVPA was also controlled for.
not completely capture the impact of diabetes on inflammation and endothelial dysfunction, respectively. Indeed, C-reactive protein, hypertensive status, and homocysteine were all unassociated with visual impairment in the current study. Physical activity was associated with visual acuity even after controlling for glycohemoglobin, an indicator of the control of blood glucose levels, suggesting that physical activity may positively influence visual acuity even among those with poorer control of blood glucose levels. An important finding of the current study was that light-intensity physical activity was independently associated with less frequent visual impairment. This finding is consistent with emerging evidence showing that light-intensity physical activity is independently associated with other health outcomes,33 such as kidney function.23 Additionally, promoting light-intensity physical activity may be more practical in diabetic patients, as coexisting neuropathy, cardiovascular disease, and nephropathy may limit their ability to engage in high-intensity physical activity. A limitation of the current study includes the relatively small sample size of participants with VI. However, the 3.5% (24/670) of participants in the current study with VI is similar to other studies in the general population (2.5% [145/5722]).19 Additionally, it is possible that the most severely affected patients with diabetes were less likely to participate with the study procedures, particularly the accelerometer testing. If true, our findings may underestimate the magnitude of the association between physical activity and VI. Finally, our classification of diabetes partially relied on self-report of disease or medication use, which may have resulted in some patient misclassification. In summary, our analysis showed that adults with diabetes who were more sedentary and less active were more likely to have poorer visual outcomes. To our knowledge, this is the first study to examine the association between accelerometer-assessed lightand moderate-to-vigorous physical activity, sedentary behavior and objectively-measured visual acuity in a nationally representative sample of adults with evidence of diabetes. Further studies
are needed to confirm our findings and to better understand the mechanisms linking physical activity, sedentary behavior, and visual outcomes.
References 1. Congdon NG, Friedman DS, Lietman T. Important causes of visual impairment in the world today. JAMA. 2003;290(15):2057–2060. PubMed doi:10.1001/jama.290.15.2057 2. Fenwick EK, Xie J, Ratcliffe J, et al. The impact of diabetic retinopathy and diabetic macular edema on health-related quality of life in type 1 and type 2 diabetes. Invest Ophthalmol Vis Sci. 2012;53(2):677–684. PubMed doi:10.1167/iovs.11-8992 3. Mazhar K, Varma R, Choudhury F, McKean-Cowdin R, Shtir CJ, Azen SP. Severity of diabetic retinopathy and health-related quality of life: the Los Angeles Latino Eye Study. Ophthalmology. 2011;118(4):649– 655. PubMed doi:10.1016/j.ophtha.2010.08.003 4. Wong TY, Klein R, Couper DJ, et al. Retinal microvascular abnormalities and incident stroke: the Atherosclerosis Risk in Communities Study. Lancet. 2001;358(9288):1134–1140. PubMed doi:10.1016/ S0140-6736(01)06253-5 5. Wong TY, Klein R, Nieto FJ, et al. Retinal microvascular abnormalities and 10-year cardiovascular mortality: a population-based case-control study. Ophthalmology. 2003;110(5):933–940. PubMed doi:10.1016/ S0161-6420(03)00084-8 6. Wong TY, Klein R, Sharrett AR, et al. Retinal arteriolar narrowing and risk of coronary heart disease in men and women. The Atherosclerosis Risk in Communities Study. JAMA. 2002;287(9):1153–1159. PubMed doi:10.1001/jama.287.9.1153 7. Taylor HR, McCarty CA, Nanjan MB. Vision impairment predicts fiveyear mortality. Trans Am Ophthalmol Soc. 2000;98:91–96, discussion 96–99. PubMed 8. Klein R, Klein BE, Moss SE, Cruickshanks KJ. Relationship of hyperglycemia to the long-term incidence and progression of diabetic retinopathy. Arch Intern Med. 1994;154(19):2169–2178. PubMed doi:10.1001/archinte.1994.00420190068008
Downloaded by Prince William Campus Lbry on 10/02/16, Volume 11, Article Number 6
PA and Vision Among Adults With Diabetes 1161 9. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C, White RD. Physical activity/exercise and type 2 diabetes: a consensus statement from the American Diabetes Association. Diabetes Care. 2006;29(6):1433–1438. PubMed doi:10.2337/dc06-9910 10. Sigal RJ, Kenny GP, Wasserman DH, Castaneda-Sceppa C. Physical activity/exercise and type 2 diabetes. Diabetes Care. 2004;27(10):2518–2539. PubMed doi:10.2337/diacare.27.10.2518 11. Loprinzi PD, Cardinal BJ. Interrelationships among physical activity, depression, homocysteine, and metabolic syndrome with special considerations by sex. Prev Med. 2012;54(6):388–392. PubMed doi:10.1016/j.ypmed.2012.03.016 12. Loprinzi P, Cardinal B, Crespo C, et al. Objectively measured physical activity and C-reactive protein: National Health and Nutrition Examination Survey 2003-2004. Scand J Med Sci Sports. 2013;23(2):164–70. PubMed 13. Tikellis G, Anuradha S, Klein R, Wong TY. Association between physical activity and retinal microvascular signs: the Atherosclerosis Risk in Communities (ARIC) Study. Microcirculation. 2010;17(5):381–393. PubMed 14. Cruickshanks KJ, Moss SE, Klein R, Klein BE. Physical activity and the risk of progression of retinopathy or the development of proliferative retinopathy. Ophthalmology. 1995;102(8):1177–1182. PubMed doi:10.1016/S0161-6420(95)30893-7 15. Cruickshanks KJ, Moss SE, Klein R, Klein BE. Physical activity and proliferative retinopathy in people diagnosed with diabetes before age 30 yr. Diabetes Care. 1992;15(10):1267–1272. PubMed doi:10.2337/ diacare.15.10.1267 16. Tucker JM, Welk GJ, Beyler NK. Physical activity in U.S.: adults compliance with the Physical Activity Guidelines for Americans. Am J Prev Med. 2011;40(4):454–461. PubMed doi:10.1016/j. amepre.2010.12.016 17. Atienza AA, King AC. Comparing self-reported versus objectively measured physical activity behavior: a preliminary investigation of older Filipino American women. Res Q Exerc Sport. 2005;76(3):358– 362. PubMed 18. Shephard RJ. Limits to the measurement of habitual physical activity by questionnaires. Br J Sports Med. 2003;37(3):197–206, discussion 206. PubMed doi:10.1136/bjsm.37.3.197 19. Willis JR, Jefferys JL, Vitale S, Ramulu PY. Visual impairment, uncorrected refractive error, and accelerometer-defined physical activity in the United States. Arch Ophthalmol. 2012;130(3):329–335. PubMed doi:10.1001/archopthalmol.2011.1773 20. Fast Facts. Vision Health Initiative. Division of Diabetes Translation. National Center for Chronic Disease Prevention and Health Promotion. http://www.cdc.gov/visionhealth Updated June 1, 2012. Accessed Jan 13, 2013. 21. Owen N, Sparling PB, Healy GN, Dunstan DW, Matthews CE. Sedentary behavior: emerging evidence for a new health risk. Mayo Clin Proc. 2010;85(12):1138–1141. PubMed doi:10.4065/mcp.2010.0444
22. Owen N, Healy GN, Matthews CE, Dunstan DW. Too much sitting: the population health science of sedentary behavior. Exerc Sport Sci Rev. 2010;38(3):105–113. PubMed doi:10.1097/JES.0b013e3181e373a2 23. Hawkins MS, Sevick MA, Richardson CR, Fried LF, Arena VC, Kriska AM. Association between physical activity and kidney function: National Health and Nutrition Examination Survey. Med Sci Sports Exerc. 2011;43(8):1457–1464. PubMed doi:10.1249/ MSS.0b013e31820c0130 24. Summary of revisions for the 2010 Clinical Practice Recommendations. Diabetes Care. 2010;33(Suppl 1):S3. PubMed doi:10.2337/ dc10-S003 25. Diagnosis and classification of diabetes mellitus. Diabetes Care. 2010;33(Suppl 1):S62–S69. PubMed doi:10.2337/dc10-S062 26. Chen KY, Bassett DR, Jr. The technology of accelerometrybased activity monitors: current and future. Med Sci Sports Exerc. 2005;37(11, Suppl):S490–S500. PubMed doi:10.1249/01. mss.0000185571.49104.82 27. Matthews CE, Chen KY, Freedson PS, et al. Amount of time spent in sedentary behaviors in the United States, 2003-2004. Am J Epidemiol. 2008;167(7):875–881. PubMed doi:10.1093/aje/kwm390 28. Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008;40(1):181–188. PubMed doi:10.1249/ mss.0b013e31815a51b3 29. Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for healthcare professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation. 2003;107(3):499–511. PubMed doi:10.1161/01. CIR.0000052939.59093.45 30. Selhub J, Jacques PF, Rosenberg IH, et al. Serum total homocysteine concentrations in the third National Health and Nutrition Examination Survey (1991-1994): population reference ranges and contribution of vitamin status to high serum concentrations. Ann Intern Med. 1999;131(5):331–339. PubMed doi:10.7326/0003-4819-131-5199909070-00003 31. Wong TY, Klein R, Klein BE, Tielsch JM, Hubbard L, Nieto FJ. Retinal microvascular abnormalities and their relationship with hypertension, cardiovascular disease, and mortality. Surv Ophthalmol. 2001;46(1):59–80. PubMed doi:10.1016/S0039-6257(01)00234-X 32. Mohamed IN, Soliman SA, Alhusban A, et al. Diabetes exacerbates retinal oxidative stress, inflammation, and microvascular degeneration in spontaneously hypertensive rats. Mol Vis. 2012;18:1457–1466. PubMed 33. Camhi SM, Sisson SB, Johnson WD, Katzmarzyk PT, Tudor-Locke C. Accelerometer-determined moderate intensity lifestyle activity and cardiometabolic health. Prev Med. 2011;52(5):358–360. PubMed doi:10.1016/j.ypmed.2011.01.030
