Skip Navigation


American Journal of Epidemiology Advance Access originally published online on May 26, 2006
American Journal of Epidemiology 2006 164(2):143-150; doi:10.1093/aje/kwj166
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
164/2/143    most recent
kwj166v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by LeCaire, T.
Right arrow Articles by D'Alessio, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by LeCaire, T.
Right arrow Articles by D'Alessio, D.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2006 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A.

Original Contribution

Lower-than-Expected Prevalence and Severity of Retinopathy in an Incident Cohort followed during the First 4–14 Years of Type 1 Diabetes

The Wisconsin Diabetes Registry Study

Tamara LeCaire1, Mari Palta1,2, Hongling Zhang3, Catherine Allen1, Ronald Klein4 and Donn D'Alessio1

1 Department of Population Health Sciences, School of Medicine and Public Health, University of Wisconsin–Madison, Madison, WI
2 Department of Biostatistics and Medical Informatics, School of Medicine and Public Health, University of Wisconsin–Madison, Madison, WI
3 Covance, Inc., Madison, WI
4 Department of Ophthalmology and Visual Sciences, School of Medicine and Public Health, University of Wisconsin–Madison, Madison, WI

Correspondence to Dr. Mari Palta, Department of Population Health Sciences, School of Medicine and Public Health, University of Wisconsin–Madison, 610 Walnut Street, Room 689, Madison, WI 53726-2397 (e-mail: mpalta{at}wisc.edu).

Received for publication July 7, 2005. Accepted for publication January 20, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The authors examined the development of diabetic retinopathy in a population-based cohort of persons with incident type 1 diabetes to investigate the possibility of lowered retinopathy prevalence and severity compared with previous US studies. A total of 474 diabetic persons from Wisconsin were followed from diagnosis through 4–14 years' duration during 1990–2002. Retinopathy was determined by fundus photography at 4, 7, 9, and 14 years' duration. Risk of developing retinopathy was modeled on demographic and diabetes-care characteristics by means of a generalized linear model using the complementary log-log link for interval-censored data. Prevalence of retinopathy increased with duration of diabetes, from 6% at 4 years to 73% at 14 years, and was highest among adults (≥20 years of age). Risk of developing retinopathy increased with increasing duration, worse glycemic control, and age up to 20 years. Indicators of diabetes care were related to retinopathy through their effect on glycemic control. Improvements in diabetes care leading to better glycemic control may have contributed to the much lower prevalence and less severe retinopathy observed than expected on the basis of a previous report from the same region of Wisconsin. The observed decreased prevalence has important implications for persons with type 1 diabetes, since retinopathy is a serious microvascular complication.

diabetes mellitus, type 1; diabetic retinopathy; longitudinal studies; risk factors

Abbreviations: DCCT, Diabetes Control and Complications Trial; WESDR, Wisconsin Epidemiologic Study of Diabetic Retinopathy


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Diabetic retinopathy is a leading cause of blindness among US adults (1Go). The Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR), established in 1979–1980, demonstrated an increasing prevalence of retinopathy with increasing duration of diabetes (2Go). That study and others reported retinopathy to be present in nearly all persons with type 1 diabetes by 15 years' duration and in all persons with diabetes by 20 years' duration (2Go–4Go).

Since that time, the Diabetes Control and Complications Trial (DCCT) has shown that intensive diabetes management can prevent or delay complications (5Go). Subsequent changes in care may have reduced rates and severity of retinopathy in persons with type 1 diabetes. A recent report from Sweden indicated a lower prevalence of diabetic retinopathy during the first 10 years of diabetes duration (6Go). The authors ascribed the decline to improved treatment and achievement of better blood glucose control in the decade following the DCCT findings. Current, post-DCCT reports on retinopathy in type 1 diabetes from population-based US studies have been called for (7).

We examined the prevalence and severity of retinopathy in a population-based cohort followed from diabetes diagnosis through 4–14 years' duration during the period 1990–2002. We also investigated possible modifiable and nonmodifiable risk factors for developing diabetic retinopathy, such as diabetes self-care practices, glycemic control, and demographic factors, to determine whether factors related to the risk of retinopathy may also have changed during this period.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Population
The Wisconsin Diabetes Registry Study is a prospective study of the natural history of long-term diabetes complications and their risk factors in children, adolescents, and young adults with type 1 diabetes. Details of case recruitment and ascertainment have been published previously (8Go). In brief, from May 1987 through April 1992, all persons newly diagnosed with type 1 diabetes who were ≤30 years of age and lived within a defined area (28 counties) in central and southern Wisconsin were eligible. People were referred to the study by a physician, nurse, diabetes educator, or family member or by self-report. Hospitals and clinics in the study area were called every 3 months for ascertainment of any missed cases. A total of 733 persons with diabetes were identified (overall case ascertainment rate = 82 percent), of whom 597 enrolled, with fewer adults and non-Whites participating.

Follow-up examinations, including fundus photographs, were conducted from August 1990 through June 2002. Among the 597 enrolled participants from the Registry, 474 (79 percent) provided gradable fundus photographs from one or more examinations undergone during 4–14 years' duration of diabetes. Examinations were offered to continuing participants at 4, 7, 9, and 14 years postonset. Of those entering the cohort, 90 percent, 63 percent, 79 percent, and 23 percent reached 4, 7, 9, and 14 years' duration, respectively, during the study. Among these persons, 80 percent, 73 percent, 64 percent, and 54 percent participated in the respective examinations. Participation in the 7-, 9-, and 14-year examinations was lowered by the ending of grant cycles, leading to not all eligible persons' being contacted. Estimated participation rates excluding such persons were 75 percent, 70 percent, and 62 percent for these examinations. In total, 1,053 observations were included in the initial analysis: 420 persons participated in fundus photography at 4 years, 275 at 7 years, 290 at 9 years, and 68 at 14 years. The institutional review board for the protection of human subjects at the University of Wisconsin–Madison approved the informed consent process and study methods.

Data and specimen collection
Diabetes management and demographic data.
Data on demographic characteristics, including years of mother's education and parental occupation, were collected by telephone interview with participants or parents (if participants were under 18 years of age) 2–3 months after diagnosis. An occupation-based socioeconomic score for subjects' parents was assigned using the scheme of Stevens and Cho (9Go), where higher scores indicate higher socioeconomic status. The higher of the two levels was used for households with two working parents. Diabetes self-management data, including insulin dose (taken by injection or insulin pump) in units per kilogram and numbers of daily injections and blood glucose checks, were also collected during the initial interview and every 6 months throughout follow-up by mailed questionnaire.

Glycosylated hemoglobin testing.
Blood samples for determination of total glycosylated hemoglobin levels were collected routinely throughout follow-up. Participants were mailed kits every 4 months to take to their physician's office or clinic for sample collection. Anticoagulated whole blood samples were hand-delivered or mailed to a central laboratory and were analyzed for total glycosylated hemoglobin within 7 days. Details on the assay and the rationale for its use in this study were published previously (10Go, 11Go). Nondiabetic pediatric and young adult subjects had a mean glycosylated hemoglobin concentration of 5.5 percent (standard deviation, 0.77). To more readily make comparisons, we calculated a DCCT equivalent hemoglobin A1c value from glycosylated hemoglobin values, following split-sample testing (n = 154) for total glycosylated hemoglobin (GHb) at the central laboratory and for hemoglobin A1c at the core DCCT laboratory (5Go), using the following regression equation: hemoglobin A1c = 0.786 + 0.797(GHb) – 0.006(GHb2).

Fundus photographs.
During their fourth, seventh, ninth, and/or 14th year of diabetes duration, participants were invited to one of three regional clinic sites for an examination. Anthropometric data collection included height, weight, and waist and hip circumference measurements. Blood pressures were measured with a random-zero sphygmomanometer according to the Hypertension Detection and Follow-up Program protocol (12Go). Eyes were dilated and color stereoscopic 30° fundus photographs of seven standard fields were taken. Retinopathy status was determined using a severity scale developed for the WESDR (2Go). The scale classifies people as having no, minimal, mild, moderate, or severe nonproliferative retinopathy, as well as treated (panretinal photocoagulation) or proliferative retinopathy, on an 11-step scale ranging from both eyes' having no retinopathy (10/10) to both eyes' having treated or proliferative retinopathy (60+/60+). Detailed protocols have been described previously (13Go).

Statistical methods
Analyses were performed using the SAS statistical package (14Go). Mean values, standard deviations, and percentages were used to characterize the sample and to compare persons with and without retinopathy. Tests of significance were two-sided. The prevalence of retinopathy was estimated at each duration time point, with and without weighting by the inverse of the probability of participation in the corresponding examination (15Go–17Go). This probability (of participation in each examination) was modeled on the basis of 522 of 597 cohort members who provided baseline characteristics and glycosylated hemoglobin measures in the first 3 years of diabetes duration. The probability was obtained by logistic regression of participation status on age at diagnosis, occupation-based socioeconomic score and mother's years of education at baseline, sex, race, and mean glycosylated hemoglobin level in the first 3 years.

The prevalence of retinopathy observed at the WESDR baseline examination in 1979–1980 was compared by age-and-duration group with the prevalence observed in the current study. Duration was grouped in the two studies as follows: 3–7 years' duration (7- or 4-year examination), 8–11 years' duration (9-year examination), and 12–15 years' duration (14-year examination).

Data on glycosylated hemoglobin level, daily insulin dose, and number of blood glucose checks per day were summarized for each subject, and examination time was calculated as the cumulative mean from the start of the study to the respective examination. Mean values were also calculated starting from 1 year preceding each examination to evaluate effects closer to the examination times and during the first 2 years of diabetes to evaluate the effects of early glycemic control. The cumulative proportion of follow-up time spent using three or more daily insulin injections or using an insulin pump was similarly summarized for each subject.

The risk of being found to have retinopathy for the first time at a given examination was modeled on demographic and diabetes-care characteristics by regression analysis with the complementary log-log link for interval-censored data (18Go, 19Go). This approach includes indicator variables for the examination time points (4, 7, 9, or 14 years) as covariates. The risk sets were formed from the initial data set by removing prevalent cases following the first finding of retinopathy; the sets included 420 subjects at the 4-year examination, 261 subjects at the 7-year examination, 245 subjects at the 9-year examination, and 41 subjects at the 14-year examination.

A basic model included the examination time-point indicator variables, age at examination, and sex and race as predictors. Model-building proceeded by evaluating the strength of the association between variously summarized measures of glycosylated hemoglobin over time and risk of retinopathy, followed by stepwise inclusion of the following variables in the model, keeping those that were significant (p ≤ 0.05) at each step: occupation-based socioeconomic score, mother's education, daily blood glucose checking, daily insulin injections, body mass index (weight (kg)/height (m)2), waist:hip ratio, systolic and diastolic blood pressure, and current cigarette smoking status. On the basis of graphical evidence, age at examination was modeled by a piecewise linear approach to capture the nonlinear relation between age and risk of retinopathy. Terms for interaction between each significant variable and the time-point indicator variables representing diabetes duration were also tested in a stepwise fashion. A second model for comparison was formed by including blood glucose checking and number of insulin injections per day in the basic model without mean glycosylated hemoglobin.

The impact of compliance with glycosylated hemoglobin kit returns on the risk of retinopathy in the final model was investigated using a pattern-mixture model approach (20Go, 21Go). A summary measure of glycosylated hemoglobin kit compliance, formed by calculating the fraction of a participant's kit returns out of the expected number of returns up to each examination time point, was checked for significance and for possible modification of the effect of mean glycosylated hemoglobin level.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Study population and diabetes management history
The 474 participants providing fundus photographs were representative of the entire cohort; sex (51 percent male), race (95 percent White), mother's education (14 years) and parents' occupational level (45) at study enrollment, age at diagnosis (11 years; range, 7 months–30 years), and mean glycosylated hemoglobin level in the first 3 years (10.3 percent) were very similar to those of the enrolled cohort, with slightly more people who were White and younger at diagnosis participating in the examinations (data not shown). These characteristics were included as predictors in determining the probability of participation for evaluations of the effect of participation on retinopathy prevalence.

In the first 4 years, the average proportion of time spent using three or more insulin injections per day was 37 percent and the average proportion of time spent using three or more blood glucose checks per day was 68 percent, with only one person using an insulin pump. In the year preceding the 9-year examination, these proportions were 72 percent and 70 percent, respectively. By 9 years, there were 28 persons (10 percent) using a pump.

Prevalence and severity of retinopathy
Minimal, mild, or moderate–severe nonproliferative retinopathy was identified in 202 subjects (43 percent) throughout follow-up. One person was treated with panretinal photocoagulation during the seventh year. The prevalence and severity of retinopathy at each examination are presented in table 1. The prevalence of retinopathy increased with duration, from 6 percent in the fourth year to 73 percent in the 14th year. The prevalence determined after weighting by the inverse of the probability of participation in the examinations was very similar (1–3 percent) to the unadjusted prevalence (data not shown). The majority of participants showed minimal retinopathy. Only 2 percent of participants during the ninth year and 10 percent during the 14th year had evidence of moderate or severe nonproliferative retinopathy. Once identified, retinopathy was observed for 79 percent of participants during subsequent follow-up.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Prevalence and severity of diabetic retinopathy (DR) and characteristics of persons with and without DR, by examination, Wisconsin Diabetes Registry Study, 1990–2002{dagger}

 
Direct comparison with the prevalence of retinopathy observed at the WESDR baseline examination showed retinopathy prevalence to increase by duration across all ages and to be highest among adults (≥20 years of age) in both studies (figure 1). In the current study, the prevalence of retinopathy was consistently lower at all ages in the duration groups 3–7 years and 8–11 years. Smaller numbers in the duration group 12–15 years in both studies resulted in larger confidence intervals, but a tendency toward lower prevalence of retinopathy in the current study persisted. Among persons aged 15–40 years, prevalence in the current study was lower than in the WESDR by 13–40 percent in the first two duration groups (data not shown).


Figure 1
View larger version (15K):
[in this window]
[in a new window]
  		FIGURE 1. Prevalence of diabetic retinopathy by duration of type 1 diabetes and age at examination in the Wisconsin Diabetes Registry Study (WDRS) (1990–2002) and the Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR) (1979–1980). Duration groups: 3–7 years (7- or 4-year examination), 8–11 years (9-year examination), and 12–15 years (14-year examination). T-shaped bars show 95% confidence intervals. The number of persons in each age-and-duration group is noted above the bar.

 
Risk of retinopathy
A cross-sectional comparison of characteristics between persons with (prevalent) retinopathy and persons without (prevalent) retinopathy at each examination is also presented in table 1. Persons with retinopathy were similar to those without retinopathy with respect to sex and race but were older at diagnosis, were older at the time of the examination, had a higher mean glycosylated hemoglobin (or hemoglobin A1c) concentration, and had a history of less frequent mean daily blood glucose checking at each examination. Mean glycosylated hemoglobin level calculated from the beginning of the study to the examination had a stronger relation with retinopathy risk than glycosylated hemoglobin level determined close to the duration interval or during the first few years of diabetes only (data not shown). The proportions of time spent using three or more insulin injections per day did not differ significantly in persons with retinopathy and persons without retinopathy at the 4- and 7-year examinations but did differ for the longer durations. Occupational level at study enrollment tended to be lower in persons who developed retinopathy, but this was not statistically significant at all durations. Blood pressure and body mass index tended to be higher in persons with retinopathy but not statistically significantly so at all durations.

Multivariable complementary log-log regression analyses showed that the risk of developing retinopathy increased greatly with duration. In the final model, with adjustment for age at examination, glycemic control, sex, and race and in comparison with the 4-year examination, hazard ratios were 4.1 at the 9-year examination and 7.9 at the 14-year examination (table 2). Glycosylated hemoglobin was also strongly related to the risk of retinopathy, and the association was stronger at longer durations. For a 1 percent increase in mean glycosylated hemoglobin level, at 4 years, the hazard of retinopathy increased by 14 percent, and at 7 and 9 years, the hazards increased by 44 percent and 42 percent, respectively. Age was also significantly related to the development of retinopathy in multivariable analysis. For every 1-year increase in age up to age 20 years, the hazard of retinopathy increased by 20 percent (hazard ratio = 1.2, 95 percent confidence interval: 1.1, 1.3). After age 20, however, the hazard associated with age showed no further increase (hazard ratio = 1.0, 95 percent confidence interval: 1.0, 1.0).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Hazard ratios for the development of diabetic retinopathy from multivariable analyses, Wisconsin Diabetes Registry Study, 1990–2002

 
In evaluating the relation between retinopathy and diabetes care, with adjustment for age, sex, race, and duration, an increase of one blood glucose check per day was associated with a decreased risk of retinopathy (hazard ratio = 0.9, 95 percent confidence interval: 0.7, 1.0). However, blood glucose checking was no longer significantly related once glycosylated hemoglobin was introduced into the model. Insulin use, evaluated as the proportion of time spent using three or more daily insulin injections and as daily insulin dose, was not related to risk of retinopathy in univariate or multivariable models. Socioeconomic indicators had no impact in the multivariable model after adjustment for the other factors described. There was also no relation between the development of retinopathy and blood pressure, body mass index, cigarette smoking, or year of study enrollment. An interaction between mean cumulative glycosylated hemoglobin and kit compliance in the pattern-mixture model analysis suggested that the relation between glycosylated hemoglobin and retinopathy risk may have been slightly underestimated in the final model.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
The Wisconsin Diabetes Registry Study is unique in providing contemporary longitudinal data from a population-based incident type 1 diabetes cohort of children, adolescents, and young adults in the United States. In this cohort, followed through the first 4–14 years of diabetes duration, we observed a much lower prevalence of retinopathy than expected. The population-based WESDR, which recruited persons with diabetes of up to 30 or more years' duration in 1979–1980, reported prevalences of 74 percent at 9–10 years' duration and 95 percent at 13–14 years' duration among persons with type 1 diabetes (2Go). We found 47 percent and 73 percent prevalences of retinopathy, respectively, at similar durations among persons recruited from many of the same counties as WESDR participants with diabetes onset 1–2 decades later. In direct comparisons by age-and-duration group, we found the prevalence of retinopathy to be consistently lower in the current study.

The lower prevalence observed in our cohort may be a result of changes in diabetes care since 1980. Our cohort was followed before and after the release of findings from the DCCT that intensive diabetes management could prevent or delay complications (5Go). Indeed, we observed an increase in the use of three or more daily insulin injections, daily blood glucose monitoring, and pump use during follow-up. In the WESDR, nearly all persons with type 1 diabetes used one or two insulin injections per day, only 12 percent monitored their blood glucose level three or more times per day, and the mean DCCT equivalent hemoglobin A1c concentration was 10.1 percent at baseline (22Go). In our study, as early as 4 years' duration, people were more likely to be using three or more insulin injections and blood glucose checks per day (45 percent and 66 percent, respectively), and the mean hemoglobin A1c concentration was 9.0 percent. The prevalence of retinopathy that exists today may thus be overestimated by relying on the prevalence in previous reports (23Go, 24Go), especially for younger persons, given the apparent improvements in diabetes management and glycemic control.

Recent population-based studies carried out in Sweden and Finland also showed a lower prevalence of retinopathy than did previous reports (6Go, 25Go, 26Go). Falck et al. (25Go) reported a prevalence of 59 percent at 9–12 years' duration in Finland, and Kernell et al. (26Go) reported a prevalence of 32 percent at 10–12 years' duration in Sweden; these prevalences are similar to our 47 percent at 9–10 years. Diabetes treatment was not discussed in detail in these reports; however, intensive diabetes control may have been better achieved in these populations. In the most recent population-based report, also from Sweden, "intensive diabetes management with multiple insulin injections and home glucose monitoring" (6, p. 352) was routine practice during the study period. In that study, Henricsson et al. (6Go) found 37 percent of persons with type 1 diabetes to have retinopathy after 8–10 years' duration. Although it is difficult to compare levels of glycosylated hemoglobin between that report and our study because of differing methods (27Go), it appears that glycemia was lower among people in the latter Swedish study (e.g., an approximate mean DCCT equivalent hemoglobin A1c concentration of 7.5–8.0 percent as compared with our 9 percent at 9 years' duration). Therefore, more intensive treatment leading to lower glycosylated hemoglobin levels could account for the lower prevalence reported in Sweden.

We also observed less severe retinopathy than expected, with a very low prevalence of moderate–severe nonproliferative retinopathy (10 percent) and only one person with treated or proliferative retinopathy by 14 years' duration. In contrast, at the baseline evaluation in WESDR, moderate–severe nonproliferative retinopathy was found in 35 percent of persons and proliferative retinopathy in 25 percent of persons at 13–14 years' duration. By age, moderate–severe nonproliferative retinopathy was found in 21 percent of persons aged 20–29 years and proliferative retinopathy in 20 percent of persons aged 20–29 years (2Go). Again, the more recent population-based studies in Sweden and Finland reported lower levels of advanced nonproliferative or preproliferative retinopathy, ranging from approximately 2 percent to 5 percent (6Go, 25Go, 26Go). Moderate-to-severe forms of retinopathy could be more sensitive to improvements in diabetes management and glycemic control, because nearly all people with diabetes may eventually develop minor changes in the microvasculature of the eye (e.g., retinal microaneurysms) (28Go). Slowed progression with more intensive therapy (29Go) could magnify an innate slowing of the process of retinopathy progression that may occur after 10–15 years' duration (30Go). In recent articles on persons with longer durations of diabetes, Klein et al. (30Go) and Hovind et al. (31Go) also reported a decrease in the rate of progression to or incidence of proliferative retinopathy, citing improved diabetes treatment.

We noted in multivariable analyses that longer duration, higher glycosylated hemoglobin level, and older age continue to be the most important factors in the development of retinopathy among younger persons during early diabetes. This is consistent with some previous reports citing age and duration as both being important and independent factors in the development of retinopathy (2Go, 3Go, 32Go). Our findings of a low prevalence of retinopathy among children less than 10 years of age were also consistent with previous reports in children (3Go, 13Go, 25Go, 26Go). The prevalence data suggested possible modification of the duration effect by age; however, in the multivariable analysis, the increasing risk associated with duration was not modified by age. The risk of retinopathy did flatten after 20 years of age. This may represent physiologic or behavioral changes occurring after this age.

The increase in retinopathy risk associated with higher glycosylated hemoglobin levels after 7 or more years was stronger than the increase associated with higher glycosylated hemoglobin levels at durations up to 4 years. This agrees with reports that cumulative glycemic exposure is important for the development of retinopathy (5Go, 13Go). Other factors, such as blood pressure, may contribute to the development or progression of retinopathy at later durations as well (2Go). In Denmark, improved management of blood pressure, in addition to better glycemic control, was credited with contributing to a decreased incidence of severe retinopathy after 20 or more years of diabetes (31Go).

This study was limited by the fact that we did not have fundus photographs for all eligible persons. Sensitivity analyses indicated that this had only a marginal influence on the estimated prevalence of retinopathy (1–3 percent) and that the association of retinopathy with glycemic control may have been underestimated. Nevertheless, it remains possible that unmeasured factors could have affected both participation and retinopathy status.

In summary, we observed a lower prevalence of retinopathy and less severe retinopathy in an incident cohort of children, adolescents, and young adults than we anticipated from an earlier population-based study carried out in the same region of Wisconsin (23Go). The fortuitous historical control, combined with similar recent reports from Europe, lends support to the conclusion that the prevalence and severity of retinopathy in type 1 diabetes have decreased notably. This has occurred in association with an increase in the intensity of diabetes management, which we have extensively documented in our population. We observed this decrease among persons receiving care across a continuum of providers and services. This speaks to the widespread dissemination and effectiveness of intensive diabetes management, the efficacy of which was demonstrated in the DCCT. The observed decreased prevalence of diabetic retinopathy has important implications for this serious microvascular complication. It emphasizes the usefulness of continuing observational studies of the natural history of type 1 diabetes as management of this difficult illness continues to improve.


   ACKNOWLEDGMENTS
 
This research was supported by National Institutes of Health grant DK36904.

The authors gratefully acknowledge the efforts of all those involved in the Wisconsin Diabetes Registry Study. The authors thank the Wisconsin Fundus Photograph Reading Center (Madison, Wisconsin) and Dr. Dennis Han and the staff of the Eye Institute (Milwaukee, Wisconsin).

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Bode BW, ed. Medical management of type 1 diabetes. 4th ed. Alexandria, VA: American Diabetes Association, 2004.
  2. Klein R, Klein BBK, Moss SE, et al. The Wisconsin Epidemiologic Study of Diabetic Retinopathy II. Prevalence and risk of diabetic retinopathy when age at diagnosis is less than 30 years. Arch Ophthalmol 1984;102:520–6.[Abstract/Free Full Text]
  3. Burger W, Hövener G, Düsterhus R, et al. Prevalence and development of retinopathy in children and adolescents with type 1 (insulin-dependent) diabetes mellitus. A longitudinal study. Diabetologia 1986;29:17–22.[CrossRef][Web of Science][Medline]
  4. Goldstein DE, Blinder KJ, Ide CH, et al. Glycemic control and development of retinopathy in youth-onset insulin-dependent diabetes mellitus. Results of a 12-year longitudinal study. Ophthalmology 1993;100:1125–31.[Web of Science][Medline]
  5. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993;329:977–86.[Abstract/Free Full Text]
  6. Henricsson M, Nyström L, Blohmé G, et al. The incidence of retinopathy 10 years after diagnosis in young adult people with diabetes: results from the nationwide population-based Diabetes Incidence Study in Sweden (DISS). Diabetes Care 2003;26:349–54.[Abstract/Free Full Text]
  7. Klein R. Has the frequency of proliferative diabetic retinopathy declined in the U.S.? Diabetes Care 2003;26:2691–2.[Free Full Text]
  8. Palta M, LeCaire T, Daniels K, et al. Risk factors for hospitalization in a cohort with type 1 diabetes. Am J Epidemiol 1997;146:627–36.[Abstract/Free Full Text]
  9. Stevens G, Cho JH. Socioeconomic indexes and the new 1980 Census occupational classification scheme. Soc Sci Res 1985;14:142–68.[CrossRef][Web of Science]
  10. Duck SC, Lee M, D'Alessio D. 24–42 month stability of internal blood standards for glycated hemoglobin analysis. Diabetes Res Clin Pract 1990;9:195–9.[CrossRef][Web of Science][Medline]
  11. Palta M, Shen G, Allen C, et al. Longitudinal glycosylated hemoglobin patterns from diagnosis in a population based cohort of IDDM. Am J Epidemiol 1996;144:954–61.[Abstract/Free Full Text]
  12. Writing Committee on behalf of the HDFP Cooperative Group. The Hypertension Detection and Follow-up Program. Prev Med 1976;5:207–15.[CrossRef][Web of Science][Medline]
  13. Klein R, Palta M, Allen C, et al. Incidence of retinopathy and associated risk factors from time of diagnosis of insulin-dependent diabetes. Arch Ophthalmol 1997;115:351–6.[Abstract/Free Full Text]
  14. SAS Institute, Inc. SAS/STAT user's guide, version 8. Cary, NC: SAS Institute, Inc, 1999.
  15. Robins JM, Rotnitzky A, Zhao LP. Analysis of semiparametric regression models in the presence of missing data. JAMA 1995;90:106–21.
  16. Palta M. Application of weighting with probability sampling and nonresponse. In: Quantitative methods in population health: extensions of ordinary regression. Hoboken, NJ: John Wiley and Sons, Inc, 2003:97–109.
  17. Hogan JW. GEE and missing data in longitudinal studies. In: Lenderking WR, Revicki DA, eds. Advancing health outcomes research methods and clinical applications. McLean, VA: Degnon Associates, 2005:139–54.
  18. Prentice PL, Gloeckler LA. Regression analysis of grouped survival data with applications to breast cancer data. Biometrics 1978;34:57–67.[CrossRef][Web of Science][Medline]
  19. Allison PD. Discrete-time methods for the analysis of event histories. In: Leinhardt S, ed. Sociological methods and research. San Francisco, CA: Jossey-Bass Publishers, 1982:61–98.
  20. Wu MC, Follman DA. Use of summary measures to adjust for informative missingness in repeated measures data with random effects. Biometrics 1999;55:75–84.[CrossRef][Web of Science][Medline]
  21. Park S, Palta M, Shao J, et al. Bias adjustment in analyzing longitudinal data with informative missingness. Stat Med 2002;21:277–91.[CrossRef][Web of Science][Medline]
  22. Klein R, Moss S. A comparison of the study populations in the Diabetes Control and Complications Trial and the Wisconsin Epidemiologic Study of Diabetic Retinopathy. Arch Intern Med 1995;155:745–54.[Abstract/Free Full Text]
  23. Roy MS, Klein R, O'Colmain BJ, et al. The prevalence of diabetic retinopathy among adult type 1 diabetic persons in the United States. Arch Ophthalmol 2004;122:546–51.[Abstract/Free Full Text]
  24. Kostraba JN, Klein R, Dorman JS, et al. The Epidemiology of Diabetes Complications Study. IV. Correlates of diabetic background and proliferative retinopathy. Am J Epidemiol 1991;133:381–91.[Abstract/Free Full Text]
  25. Falck A, Käär M-L, Laatikainen L. A prospective, longitudinal study examining the development of retinopathy in children with diabetes. Acta Paediatr 1996;85:313–19.[CrossRef][Web of Science][Medline]
  26. Kernell A, Dedorsson I, Johansson B, et al. Prevalence of diabetic retinopathy in children and adolescents with IDDM. A population-based multicentre study. Diabetologia 1997;40:307–10.[CrossRef][Web of Science][Medline]
  27. Larsson L-I, Alm A, Bergenheim T, et al. Retinopathy in diabetic patients aged 15–50 years in the county of Umeå, Sweden. Acta Ophthalmol Scand 1999;77:430–6.[CrossRef][Web of Science][Medline]
  28. Danne T, Kordonouri O, Enders I, et al. Monitoring for retinopathy in children and adolescents with type 1 diabetes. Acta Paediatr Suppl 1998;425:35–41.[Medline]
  29. The Diabetes Control and Complications Trial Research Group. The relationship of glycemic exposure (HbA1c) to the risk of development and progression of retinopathy in the Diabetes Control and Complications Trial. Diabetes 1995;44:968–83.[Abstract]
  30. Klein R, Klein BEK, Moss SE, et al. The Wisconsin Epidemiologic Study of Diabetic Retinopathy: XVII. The 14-year incidence and progression of diabetic retinopathy and associated risk factors in type 1 diabetes. Ophthalmology 1998;105:1801–15.[CrossRef][Web of Science][Medline]
  31. Hovind P, Tarnow L, Rossing K, et al. Decreasing incidence of severe diabetic microangiopathy in type 1 diabetes. Diabetes Care 2004;26:1258–64.[CrossRef][Web of Science]
  32. Olsen BS, Johannesen J, Sjolie AK, et al. Metabolic control and prevalence of microvascular complications in young Danish patients with type 1 diabetes mellitus. Diabet Med 1999;16:79–85.[CrossRef][Web of Science][Medline]

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Arch OphthalmolHome page
J. B. Saaddine, A. A. Honeycutt, K. M. V. Narayan, X. Zhang, R. Klein, and J. P. Boyle
Projection of Diabetic Retinopathy and Other Major Eye Diseases Among People With Diabetes Mellitus: United States, 2005-2050
Arch Ophthalmol, December 1, 2008; 126(12): 1740 - 1747.
[Abstract] [Full Text] [PDF]

Home page
 BMJHome page
P. H. Gallego, M. E Craig, S. Hing, and K. C Donaghue
Role of blood pressure in development of early retinopathy in adolescents with type 1 diabetes: prospective cohort study
BMJ, August 26, 2008; 337(aug26_1): a918 - a918.
[Abstract] [Full Text] [PDF]

Home page
 Diabetes CareHome page
J. H. Vallance, P. J. Wilson, G. P. Leese, R. McAlpine, C. J. MacEwen, and J. D. Ellis
Diabetic Retinopathy: More Patients, Less Laser: A longitudinal population-based study in Tayside, Scotland
Diabetes Care, June 1, 2008; 31(6): 1126 - 1131.
[Abstract] [Full Text] [PDF]

Home page
 CJASNHome page
N. Iqbal and A. H. Rubenstein
Does Lowering of Blood Glucose Improve Cardiovascular Morbidity and Mortality?
Clin. J. Am. Soc. Nephrol., January 1, 2008; 3(1): 163 - 167.
[Full Text] [PDF]

Home page
 Diabetes CareHome page
R. M. Cohen, T. J. LeCaire, C. J. Lindsell, E. P. Smith, and D. J. D'Alessio
Relationship of Prospective GHb to Glycated Serum Proteins in Incident Diabetic Retinopathy: Implications of the glycation gap for mechanism of risk prediction
Diabetes Care, January 1, 2008; 31(1): 151 - 153.
[Full Text] [PDF]

Home page
 Diabetes CareHome page
N. K. Sweitzer, M. Shenoy, J. H. Stein, S. Keles, M. Palta, T. LeCaire, and G. F. Mitchell
Increases in Central Aortic Impedance Precede Alterations in Arterial Stiffness Measures in Type 1 Diabetes
Diabetes Care, November 1, 2007; 30(11): 2886 - 2891.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
164/2/143    most recent
kwj166v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (4)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by LeCaire, T.
Right arrow Articles by D'Alessio, D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by LeCaire, T.
Right arrow Articles by D'Alessio, D.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?