A Review of the Evidence for the U.S. Preventive Services Task Force
May 2008
Prepared by: Susan L. Norris, M.D., M.P.H.; Devan Kansagara, M.D.; Christina Bougatsos, B.S.; and Rongwei Fu, Ph.D.
Corresponding Author: Susan L. Norris, M.D., MPH, Department of Medical Informatics and Clinical Epidemiology, Oregon Health & Science University, 3181 SW Sam Jackson Park Road, Mail Stop B1CC, Portland, OR 97239; E-mail: norriss@ohsu.edu.
This report was conducted by the Oregon Evidence-based Practice Center under contract to the Agency for Healthcare Research and Quality (contract no. 290-02-0024, Task order no. 2 for the U.S. Preventive Services Task Force. The investigators involved have declared no conflicts of interest with objectively conducting this research. The findings and conclusions in this document are those of the authors, who are responsible for its content, and do not necessarily represent the views of AHRQ. No statement in this report should be construed as an official position of AHRQ or of the U.S. Department of Health and Human Services.
The information in this report is intended to help clinicians, employers, policymakers, and others make informed decisions about the provision of health care services. This report is intended as a reference and not as a substitute for clinical judgment.
This report may be used, in whole or in part, as the basis for the development of clinical practice guidelines and other quality enhancement tools, or as a basis for reimbursement and coverage policies. AHRQ or U.S. Department of Health and Human Services endorsement of such derivative products may not be stated or implied.
This report was first published in the Annals of Internal Medicine on June 3, 2008 (Ann Intern Med 2008;148:855-68; http://www.annals.org).
Contents
Abstract
Introduction
Methods
Results
Discussion
Appendix Figure 1: Analytic framework
Appendix Figure 2: Diabetes incidence in lifestyle trials
Appendix Figure 3: Diabetes incidence in drug trials
Appendix Table 1: Randomized, Controlled Trials of Hypertension Treatment in Diabetic Populations
Appendix Table 2: Randomized, Controlled Trials of Lipid Interventions in Diabetic and Nondiabetic Populations
Appendix Table 3: Randomized, Controlled Trials of Interventions in Prediabetes
References
Abstract
Background: More than 19 million Americans are affected by type 2 diabetes mellitus, which is undiagnosed in one third of these persons. In addition, it is estimated that more than 54 million adults have prediabetes. Debate continues over the benefits and harms of screening and then treating adults who have asymptomatic diabetes or prediabetes.
Purpose: To update the 2003 U.S. Preventive Services Task Force review on the evidence for potential benefits and harms of screening adults for type 2 diabetes and prediabetes in primary care settings.
Data Sources: MEDLINE and the Cochrane Library for relevant studies and systematic reviews published in English between March 2001 and July 2007.
Study Selection: Trials and observational studies that directly addressed the effectiveness and adverse effects of screening interventions were included. Randomized, controlled trials were used to assess the effectiveness of diabetes and prediabetes treatments. For diabetes interventions, trials of patients with disease for 1 year or less were included, as well as trials comparing outcomes among diabetic and nondiabetic patients.
Data Extraction: Relevant data were abstracted in duplicate into a standardized template.
Data Synthesis: Data were synthesized in a qualitative manner, and a random-effects meta-analysis of the effects of interventions in prediabetes on the incidence of diabetes was performed.
Limitations: Most of the data on diabetes treatment were not from primary trial data but from subgroup analyses. Participants in intensive lifestyle interventions for prediabetes may not be representative of general prediabetic populations.
Conclusion: Direct evidence is lacking on the health benefits of detecting type 2 diabetes by either targeted or mass screening, and indirect evidence also fails to demonstrate health benefits for screening general populations. Persons with hypertension probably benefit from screening, because blood pressure targets for persons with diabetes are lower than those for persons without diabetes. Intensive lifestyle and pharmacotherapeutic interventions reduce the progression of prediabetes to diabetes, but few data examine the effect of these interventions on long-term health outcomes.
Return to Contents
Introduction
The 2002 National Health and Nutrition Examination Survey estimated that 19.3 million U.S. adults (9.3% of the total U.S. population) had diabetes mellitus, one third of whom had undiagnosed diabetes 1. In addition, 26.0% had impaired fasting glucose and impaired glucose tolerance was even more prevalent 2. The prevalence of diagnosed diabetes is increasing, particularly among obese
individuals 3,5. The risk for death among persons with diabetes is about twice that of persons without diabetes, and cardiovascular events account for more than three fourths of these deaths 4.
Type 2 diabetes often goes undiagnosed for many years because hyperglycemia develops gradually and may not produce symptoms 3,5. Persons with diabetes are at increased risk for microvascular and macrovascular complications, and duration of diabetes and degree of hyperglycemia are associated with an increased risk for microvascular complications 6-9. The prevalence of macrovascular complications is elevated in persons with prediabetes (defined as impaired fasting glucose, impaired glucose tolerance, or both) and in persons with newly diagnosed diabetes 10-18. A substantial proportion of persons presenting with a new cardiovascular event have undiagnosed diabetes or prediabetes 10,19-23. Several recent observational studies and a meta-analysis suggest an association between chronic hyperglycemia and cardiovascular disease and stroke 24-27.
Diabetes has a long preclinical phase, estimated at 10 to 12 years on the basis of the progression of microvascular complications 28, and valid and reliable tests can detect type 2 diabetes during this asymptomatic period 29. The previous evidence review for the U.S. Preventive Services Task Force (USPSTF) 29,30 suggested that screening is justified if the ensuing treatments offer incremental net benefits compared with treatment at the time of clinical presentation. In 2003, the USPSTF concluded that the evidence was insufficient to recommend for or against routinely screening asymptomatic adults for type 2 diabetes or prediabetes, but it did recommend screening for adults with hypertension or hyperlipidemia 31.
This review summarizes evidence that has become available since the previous report to inform an update of the 2003 USPSTF recommendations on screening for type 2 diabetes and prediabetes.
Return to Contents
Methods
The methods of the USPSTF evidence reviews are fully detailed elsewhere 32. The analytic framework (Appendix Figure 1) focuses on decreasing the risk for complications from type 2 diabetes as a result of screening for diabetes. We did not consider secondary prevention studies that exclusively enrolled persons with known cardiovascular disease, because we considered those persons to have a potential preexisting diabetes complication.
We searched MEDLINE and the Cochrane Library for relevant English-language systematic reviews and studies published between March 2001 (6 months before the end date of the previous search) and July 2007 related to 5 key questions. We also examined the reference lists of included studies and ClinicalTrials.gov for relevant trials. We evaluated all studies included in relevant systematic reviews for potential inclusion. We included randomized, controlled trials (RCTs) and observational studies that examined the effectiveness or adverse effects of screening and diagnosis of type 2 diabetes. We used RCTs to assess the effectiveness of diabetes and prediabetes treatments. For diabetes interventions, we included trials with patients who had disease for 1 year or less, as well as trials comparing outcomes among diabetic and nondiabetic populations. We used good-quality systematic reviews to assess the adverse effects of treatment. Search strategies are available in
the full evidence report, which can be found at www.ahrq.gov/clinic/uspstf08/type2/type2art.pdf.
An investigator screened titles and abstracts, and a random sample of 1500 titles and abstracts was dual reviewed. Two reviewers examined the full text of potentially relevant articles to achieve consensus on inclusion (Figure 1). Data were abstracted by one investigator and checked by another. We assessed internal validity of individual trials by examining factors that might introduce bias: adequate randomization, allocation concealment, baseline comparability of participants, blinding, and loss to followup. We rated studies as good, fair, or poor quality by using standard USPSTF criteria 32. We rated systematic reviews on the basis of established criteria, and we included only good-quality reviews 33,34. We assessed potential applicability of individual studies to primary care practice on the basis of the methods of participant recruitment and selection.
We identified studies that modeled screening interventions from our main search, as well as from a recent, good-quality systematic review of screening for type 2 diabetes by the National Health Service Research and Development Health Technology Assessment Programme 35. We independently abstracted the relevant studies included in that report and relied on their extensive assessments of model quality.
We performed a qualitative synthesis of abstracted data that were generally too heterogeneous for quantitative pooling, except for estimates of the effect of pharmacotherapeutic or lifestyle interventions on diabetes incidence in prediabetic populations. We calculated these pooled estimates
by using a hazard ratio and its SE from Cox regression; either a rate ratio or a risk ratio was calculated when a hazard ratio was not reported 36-38. We tested for statistical heterogeneity with the standard chi-square test and obtained the overall estimates of relative risk by using a random-effects model 39.
Role of the Funding Source
This study was funded by the Agency for Healthcare Research and Quality (AHRQ) under a contract to support the work of the USPSTF. Agency staff and USPSTF members participated in the initial scope of this work and reviewed interim analyses and the final report. A draft version was distributed to content experts for review. Agency approval was required before this manuscript could be submitted for publication, but the authors are solely responsible for the content and the decision to submit it for publication.
Return to Contents
Results
Key Question 1
Is there direct evidence that systematic screening for type 2 diabetes, impaired fasting glucose, or impaired glucose tolerance among asymptomatic adults improves health outcomes?
We identified no RCTs examining the effectiveness of a screening program for type 2 diabetes. A small, good-quality, case-control study did not find benefit from screening when microvascular complications were considered 40. Limited data from 2 cross-sectional studies did not provide good-quality, direct evidence of the effectiveness of screening for type 2 diabetes in either targeted or general populations 41,42,.
Of modeling studies identified 35,43-48, 2 recent high-quality studies suggested that targeted screening for type 2 diabetes among persons with hypertension may be relatively cost-effective when macrovascular benefits of optimal blood pressure control are considered 35,47, older persons benefitted more than younger persons 35,47, and screening obese persons was more cost-effective than mass screening 35.
The ADDITION (Anglo-Danish-Dutch Study of Intensive Treatment in People with Screen Detected Diabetes in Primary Care) study 49, currently in progress, may shed light on differences in baseline characteristics and long-term health outcomes between persons with screening-detected diabetes and those who present with symptoms.
Key Question 2
Does beginning treatment of type 2 diabetes early as a result of screening provide an incremental benefit in health outcomes compared with initiating treatment after clinical diagnosis?
We identified no studies that directly explored this question by comparing treatment effects between persons with screening-detected versus clinically detected diabetes, nor did we identify new studies reporting treatment effects in an exclusively screening-detected or recently diagnosed diabetes cohort. Because of the lack of direct evidence, we examined intervention studies comparing treatment effects in diabetic versus nondiabetic populations 50-61 to address the question: "Would early knowledge of a diabetes diagnosis prompt a change in clinical management?"
Tight Glycemic Control
No new, completed studies have examined the effect of glycemic control strategies in persons with newly diagnosed type 2 diabetes since the previous USPSTF review 29. The UKPDS (United Kingdom Prospective Diabetes Study) 62 remains the largest and most influential trial of intensive glycemic control in persons with newly diagnosed, mainly clinically detected, type 2 diabetes. In the UKPDS, persons assigned to intensive glycemic control had a 25% reduction (95% CI, 7% to 40%) in microvascular complications, mostly due to a reduced need for retinal photocoagulation, as well as a nonsignificant 16% relative risk reduction (CI, 71% to 100%) of myocardial infarction 62. The UKPDS investigators estimated that 19.6 persons (CI, 10 to 500 persons) would need to be intensively treated for 10 years to prevent 1 person from developing any single clinical end point 62. A recent meta-analysis combined results from the UKPDS and other older trials examined in the last USPSTF review, and it concluded that tight glycemic control resulted in a modest reduction of macrovascular events, particularly peripheral vascular and cerebrovascular events, in persons with type 2 diabetes (combined incidence rate ratio for any macrovascular event, 0.81 [CI, 0.73 to 0.91]) 27. Examination of the individual trials, however, showed largely nonsignificant results, and it was unclear how overlapping populations from the UKPDS were accounted for in the metaanalysis.
It is unlikely that good-quality trial evidence of the final health benefits of early glycemic control in a screening-detected population will ever be available because withholding treatment from persons with known diabetes is unethical and the length of follow-up required might be prohibitive. The ADDITION study 49 should provide valuable information, although it will be assessing the incremental benefit of very aggressive glycemic control over current standards for glycemic control in a screened population.
Specific Antihypertensive Treatment
There is no clear evidence that persons with diabetes detected by screening would respond differently to specific antihypertensive regimens compared with persons without diabetes. We found no new studies involving antihypertensive agents in screening-detected individuals; however, we identified 2 new trials comparing the effect of different antihypertensive regimens in persons with and those without diabetes (Appendix Table 1) 51,52. The ALLHAT (Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial) 51 was an effectiveness trial showing no demonstrable advantage of either a calcium-channel blocker or an angiotensin-converting enzyme inhibitor over a thiazide diuretic in reducing deaths or cardiovascular events in both the diabetes and nondiabetes subgroups. A second study compared verapamil with either a ß-blocker or a thiazide diuretic and found no evidence of a differential effect on cardiovascular outcomes between those with and those without diabetes 52. However, neither trial was originally powered to detect differences between the diabetes and nondiabetes subgroups. A third trial (included in the previous USPSTF review 29) examined persons with hypertension and left ventricular hypertrophy. It showed that persons with diabetes had lower cardiovascular mortality with losartan compared with atenolol, and those without diabetes experienced a reduction in stroke with losartan 53,54.
We identified 1 meta-analysis of antihypertensive trials that compared outcomes between persons with and those without diabetes 63. Angiotensin-receptor blockers provided significantly greater protection against congestive heart failure for those with diabetes than for those without diabetes. All of the studies of angiotensin-converting enzyme inhibitors compared with placebo were secondary prevention trials, except for the HOPE (Heart Outcomes Prevention Evaluation) trial, which was a combination of primary and secondary prevention 64,65 and was included in the previous USPSTF review 29. The HOPE trial showed that persons with type 2 diabetes and at least moderate cardiovascular risk (age > 55 years and 1 additional cardiovascular risk factor) experienced a 25% relative risk reduction (CI, 12% to 36%) in cardiovascular events, cardiovascular deaths, and stroke with ramipril treatment—a similar benefit to that achieved in persons with a history of ischemic heart disease and no diabetes 64,65.
Intensity of Antihypertensive Treatment
As discussed in the previous USPSTF review 29, 1 trial (the HOT [Hypertension Optimal Treatment] trial 66) provided evidence that aggressive blood pressure control in persons with diabetes reduces cardiovascular morbidity. In that trial, the diabetes subgroup experienced a 51% relative risk reduction in cardiovascular events from more aggressive blood pressure control, a greater benefit than that observed in nondiabetic patients 29,66. We did not identify new trials comparing intensive and less intensive blood pressure treatment targets in persons with and without diabetes. A recent meta-analysis presented limited evidence that higher-intensity antihypertensive treatment reduces the risk for major cardiovascular events in persons with diabetes (relative risk, 0.64 [CI, 0.46 to 0.89]) but not in those without diabetes 63; the differential effect on cardiovascular mortality was less clear. The ACCORD (Action to Control Cardiovascular Risk in Diabetes) trial, currently in progress, will examine the relative benefits of very intensive blood pressure control compared with more moderate standards (target systolic blood pressure <120 mm Hg vs. <140 mm Hg) 67.
Initiation of Lipid-Lowering Treatment
Studies of intensive lipid-lowering treatment suggest that persons with diabetes benefit to a similar extent as those without diabetes. For this update, we identified 4 trials (Appendix Table 2) 50,55,57,58 and 1 meta-analysis 68 examining the effects of 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors on primary prevention of cardiovascular events and deaths in persons with and without diabetes. In 1 trial, neither the diabetes group nor the nondiabetes subgroup benefited from statin treatment in reducing mortality or cardiovascular event rates, but the rate of nonstudy statin use was high in the control group and
the differential reduction in low-density lipoprotein cholesterol between study groups was relatively small 50. In 2 fair-quality trials, statin therapy did not significantly reduce the primary end point (coronary events in 1 trial and coronary and/or stroke events in the other) in the diabetes subgroup, but it did benefit the nondiabetes subgroup 55,58. Comparisons between persons with and those without diabetes were hampered by a relatively low absolute number of events in the diabetes subgroup.
The Heart Protection Study 57 was a large, good-quality RCT examining the efficacy of an HMG-CoA reductase inhibitor in primary and secondary prevention of cardiovascular events and death. Persons with diabetes had a similar reduction in cardiovascular events (relative risk reduction, 27% [CI, 7% to 40%]) as did persons without diabetes who had known vascular disease, and the benefit was independent of initial low-density lipoprotein cholesterol levels. Many in the diabetes subgroup had additional cardiovascular risk factors, including smoking, hypertension, dyslipidemia (high triglyceride and low high-density lipoprotein cholesterol levels), or a combination of these. Although persons with shorter diabetes duration seemed to benefit to a similar extent as those with much longer-duration diabetes, power was insufficient to determine whether participants with newly diagnosed diabetes (that is, <1 year) benefited to a significant extent.
A recent meta-analysis of 6 primary prevention trials—the 4 just discussed, an older trial using a fibric acid derivative, and an older statin trial—reported that lipid-lowering drug treatment seemed to be equally efficacious in persons with and those without diabetes 68.
Aspirin for Primary Prevention
The previous USPSTF review 29 included several trials of aspirin for primary prevention of cardiovascular disease. The Antithrombotic Trialists’ Collaborative metaanalysis showed a nonsignificant 7% relative risk reduction in the incidence of vascular events in the high-risk diabetic population 69, a result mainly driven by the results of the ETDRS (Early Treatment Diabetic Retinopathy Study), in which the incidence of fatal and nonfatal coronary events decreased in the treatment group (relative risk, 0.83 [CI, 0.66 to 1.04]) 70. In the Physicians' Health Study 71, aspirin was associated with significant cardiovascular risk reduction in persons with diabetes, and the benefit seemed greater in those with diabetes than in those without.
We identified 2 new studies of low-dose aspirin for primary prevention of cardiovascular events in persons with and without diabetes 59,60. In the Primary Prevention Project 59, the subgroup with diabetes did not experience any benefit, whereas the subgroup without diabetes experienced a reduction in the incidence of major cardiovascular and cerebrovascular events (relative risk, 0.59 [CI, 0.37 to 0.94]). This fair-quality study was stopped early, with a resultant low event rate in both groups. Given the small size of the group with diabetes, the trial was probably underpowered to detect a difference in this subgroup. The Women's Health Study 60, a large, good-quality trial, showed that aspirin reduced the incidence of ischemic stroke (relative risk, 0.42 [CI, 0.22 to 0.82]), but not cardiovascular events, in women with diabetes. There was no evidence that the effect of aspirin was significantly more pronounced in women with diabetes than in those without. The difference in results between the Primary Prevention Project 59 and the Women’s Health Study 60 may be due to differences in the populations considered or to the differential risks for stroke versus those for myocardial infarction (the rate of stroke was higher than that of myocardial infarction in the Women's Health Study).
Key Question 3
Does beginning treatment of impaired fasting glucose or impaired glucose tolerance early as a result of screening provide an incremental benefit in final health outcomes compared with initiating treatment after clinical diagnosis of type 2 diabetes?
Three studies reported cardiovascular outcomes with intensive lifestyle interventions in persons with prediabetes 36,72,73. In the DPP (Diabetes Prevention Program) 36, neither the cumulative incidence of cardiovascular disease nor the event rate differed among treatment groups; however, the study was not adequately powered to examine these outcomes 74. The STOP-NIDDM (Study to Prevent
Non–Insulin-Dependent Diabetes Mellitus) trial 72, in which patients with impaired glucose tolerance were randomly assigned to placebo or acarbose, showed a reduction in cardiovascular events of any type (hazard ratio, 0.51 [CI, 0.28 to 0.95]; absolute risk reduction, 2.5%). However, this study was limited by an attrition rate of 24% overall, with a much higher rate in the treatment group. In the DREAM (Diabetes Reduction Assessment with Ramipril and Rosiglitazone Medication) trial 73,75, the incidence rate of the primary composite outcome of cardiovascular events did not significantly differ between the rosiglitazone and placebo groups (hazard ratio, 1.37 [CI, 0.97 to 1.94]) 72 (Appendix Table 3).
Many studies have examined the effect of lifestyle interventions on the incidence of type 2 diabetes among persons with prediabetes 36,38,76-81, several of which 36,76,80,81 were included in the previous review 29. In the DPP 36, the incidence of diabetes was reduced at 3-year follow-up with an intensive lifestyle intervention (reduction in incidence, 58% [CI, 48% to 66%]) and with treatment with metformin (reduction in incidence, 31% [CI, 17% to 43%]), both compared with placebo. The Finnish Diabetes Prevention study 76 examined a lifestyle intervention, and the incidence rate of diabetes was significantly reduced at mean follow-up of 3.2 years (hazard ratio, 0.4 [CI, 0.3 to 0.7]). This was maintained 3 years after completion of the intervention (hazard ratio, 0.57 [CI, 0.43 to 0.76]) 82. A Chinese study also reported a significant decrease in the incidence of type 2 diabetes at 6-year follow-up with an intensive lifestyle intervention 80. Two smaller, more recent trials examined the effect of lifestyle interventions on incidence rates of diabetes among persons with prediabetes and found a significant decrease in incidence compared with usual care 38,77.
Several recent studies examined the effect of pharmacotherapeutic interventions on diabetes incidence. In the DREAM trial 73,75, rosiglitazone reduced the incidence of diabetes among persons with prediabetes when it was administered for a median of 3.0 years (hazard ratio, 0.38 [CI, 0.33 to 0.44]) 75, whereas ramipril was not effective in reducing the incidence of diabetes 73. In the STOPNIDDM trial 72, the incidence rate of type 2 diabetes was reduced significantly in the acarbose treatment group over the 3.3-year intervention (hazard ratio, 0.75 [CI, 0.63 to 0.90]).
In the XENDOS (XENical in the Prevention of Diabetes in Obese Subjects) study 83, which was rated fair-to-poor quality because of high attrition, orlistat reduced the incidence of type 2 diabetes over 4 years in patients with impaired glucose tolerance (hazard ratio, 0.55 [CI not reported]) 83. A meta-analysis of 3 other studies of orlistat produced similar results 37. Acarbose 84 and metformin 77 also decreased diabetes incidence at up to 3-year follow-up. Two studies of interventions in persons with prediabetes are in progress, and published results are not yet available 85,86.
Results from our meta-analyses showed that the incidence of type 2 diabetes was decreased with lifestyle interventions (pooled hazard ratio, 0.48 [CI, 0.40 to 0.58]) (Appendix Figure 2 and Appendix Figure 3). Pharmacotherapeutic interventions also reduced diabetes incidence (pooled hazard ratio, 0.65 [CI, 0.51 to 0.83]), although the data were statistically heterogeneous largely due to the effect of the rosiglitazone group of the DREAM trial 73.
We did not identify any data to address the question of whether there should be different treatment targets for lipid levels and blood pressure for persons with prediabetes compared with normoglycemic persons.
We identified only 1 study examining the comparative effectiveness of different medications for treating hyperlipidemia, hypertension, and cardiovascular disease among persons with prediabetes versus those with normoglycemia. The ALLHAT 51 examined various antihypertensive therapies among persons with diabetes, impaired fasting glucose, and normoglycemia and failed to demonstrate superiority for an angiotensin-converting enzyme inhibitor or a calcium-channel blocker compared with a thiazide-type diuretic across the 3 glycemic strata for the composite outcome of coronary heart disease death and nonfatal myocardial infarction.
Modeling studies have been used to examine the treatment of prediabetes 35,87-94. The health technology assessment by Waugh and colleagues 35 recommended screening for glucose intolerance because strategies for reducing cholesterol and blood pressure are effective and because type 2 diabetes can be prevented. Waugh and colleagues seem to assume that the effects of treating persons with screening-detected diabetes are the same as those of treating persons with clinically detected diabetes and that there are proven linkages between treating dysglycemia and final health outcomes. They also systematically reviewed published economic models and noted that, despite the variable quality, structure, and assumptions of the models, all predicted that delaying the onset of diabetes would substantially reduce the incidence of vascular complications, improve quality of life, and avoid future medical costs. They concluded that if a screening program was implemented to target persons at risk for diabetes, subsequent treatment for persons with impaired glucose tolerance with lifestyle or pharmacologic interventions was a good use of resources.
Herman and associates 90 examined the lifetime utility and cost-effectiveness of the DPP lifestyle intervention 36 and found the intervention to be relatively cost-effective (cost per quality-adjusted life-year, $8800 [from a societal perspective]), with a 0.5-year gain in life expectancy and 20% decrease in diabetes incidence. Results were somewhat less marked with metformin, which was still relatively cost-effective.
Eddy and colleagues 87,88 examined the DPP interventions and also predicted large absolute reductions in the proportion of persons developing type 2 diabetes and a delay of 7 to 8 years in onset of diabetes, as well as that the DPP lifestyle intervention will lead to fewer complications and improved quality-adjusted life-years 95. They, however, estimated much higher marginal cost-effectiveness ratios than did Herman and associates 96.
Several other models recently evaluated primary prevention of type 2 diabetes among persons with impaired glucose tolerance 91,92,94,97, and all demonstrated relative cost-effectiveness of lifestyle interventions. Two models examined metformin and found it to be cost-saving under many conditions 92,97.
Key Question 4
What adverse effects result from screening a person for type 2 diabetes, impaired fasting glucose, or impaired glucose tolerance?
Data are sparse on the psychological effects of screening for type 2 diabetes, and none of the data that we identified suggested significant adverse effects at up to 1-year follow-up 98-110. In the ADDITION study 103, stepwise screening had limited effects on anxiety levels at up to 1-year follow-up. In a cross-sectional study, Skinner and colleagues 109 did not find that screening high-risk patients for type 2 diabetes with an oral glucose tolerance test was associated with significant anxiety. Other included studies also did not report any serious psychological effects of a new diagnosis of type 2 diabetes 98-104,107,108,110.
Several studies compared persons with screening-detected diabetes with persons without diabetes. Using Hoorn observational data, Adriaanse and colleagues 100 found no significant differences in well-being and health-related quality of life between patients with newly diagnosed diabetes and those at high risk but without diabetes at 2-week and 1-year follow-ups. Poorer quality-of-life scores at 6-month follow-up in the group with diabetes may suggest a temporary effect. Similar results were found in several other studies 98,104,107. In the ADDITION study 102,103,110, persons with screening-detected diabetes generally reported low emotional distress, with some differences in distress and self-efficacy noted between groups treated intensively compared with usual care.
We identified no studies that addressed the effects of a false-positive result from any of the tests used to screen for dysglycemia. We identified no studies that directly addressed labeling of persons with screening-detected diabetes and no studies that examined the effect of a diagnosis of
prediabetes.
Key Question 5
What adverse effects result from treating a person with type 2 diabetes, impaired fasting glucose, or impaired glucose tolerance detected by screening?
Recent systematic reviews of the adverse effects of drugs used in treating type 2 diabetes and prediabetes 111-136 reveal some important new data related to the safety to thiazolidinediones. An association between rosiglitazone and increased risk for myocardial infarction 134,137 and heart failure 134 was noted recently. For other drugs examined in the studies included in this review, we identified no new data on severe adverse effects compared with data available at the time of the previous USPSTF review 29. Intensive glucose control in the UKPDS was not associated with high rates of hypoglycemia (0.55% annual incidence of major hypoglycemia) 138. Relatively common side effects, such as cough with angiotensin-converting enzyme inhibitors and gastrointestinal effects with acarbose, should be considered when prescribing these drugs, but they are not associated with increased deaths or adverse cardiovascular outcomes.
Return to Contents
Discussion
No direct evidence clearly determines whether screening asymptomatic individuals for diabetes or prediabetes alters health outcomes (Table 1). Evidence shows that persons with diabetes benefit from control of blood pressure and lipid levels, but studies have not included persons with screening-detected diabetes. Persons with hypertension and type 2 diabetes benefit from lower blood pressure targets than persons with hypertension but without diabetes 66. Persons with newly diagnosed, largely clinically detected, diabetes benefit from intensive glycemic control, largely because of a reduction in microvascular events 62. Evidence shows that intensive lifestyle modification in persons with prediabetes—an implicitly screening-detected population—delays the progression to clinical diabetes, but whether treatment alters final health outcomes is unknown because studies were not powered for those outcomes or were not of sufficient duration.
Table 2 and Table 3 show the numbers needed to screen to prevent an outcome of interest in different theoretical populations. These outcomes have not changed from the estimates of the previous USPSTF review 29 because we identified no new data on the effectiveness of these interventions. As noted elsewhere 29, interventions that target cardiovascular events produce greater effects than those that target microvascular complications occurring later in the disease process.
On the basis of the DPP 36 and the Finnish Diabetes Prevention Study 76, screening 1000 persons with prediabetes will delay 44 cases of type 2 diabetes over 3.0 years. Pharmacotherapy with metformin (on the basis of DPP data 36) produced a somewhat less favorable number needed to screen. Many important assumptions underlying number-needed-to-screen estimates remain, including length of the asymptomatic period, prevalence of undiagnosed diabetes or prediabetes, incidence rates of diabetes complications, and treatment effect.
Screening targeted to populations at risk for diabetes would probably increase the yield and economic efficiency of screening, and risk scores have been developed to identify those at high risk for diabetes 139-144. In the DPP, older age and higher body mass index increased the yield of screening across ethnic groups 145. On the other hand, the prevalence of diagnosed diabetes in certain high-risk groups, such as non-Hispanic black persons and Mexican-American persons, has increased, whereas the proportion of persons with undiagnosed disease in those groups has decreased,
suggesting that opportunistic screening targeted to populations at high risk may already be occurring. This trend reduces the prevalence of undiagnosed diabetes and increases the number needed to screen to prevent adverse events in the remaining unscreened group 1.
A diabetes population of significant interest to a screening program would be individuals who would benefit from aggressive interventions to reduce macrovascular complications in persons who would not have been otherwise identified through recommended hypertension and hyperlipidemia screening 31. Many persons with diabetes are hypertensive or have additional cardiovascular disease risk factors, and those with the highest cardiovascular risk profiles are likely to benefit most from treatment 57,62,146-148. As shown in the Heart Protection Study 63, elevated low-density lipoprotein cholesterol levels alone may not identify many persons with diabetes and dyslipidemia
who might benefit from lipid-lowering treatment, but this population had higher-than-average cardiovascular risk profiles. The benefit of identifying and treating asymptomatic diabetes in normotensive, nondyslipidemic persons at average cardiovascular risk is unclear.
The potential yield of diabetes and prediabetes screening must be weighed carefully against the potential harms of screening and diagnosis. We did not identify evidence suggesting serious adverse effects of screening for type 2 diabetes. The literature does, however, have important limitations.
Included studies examined persons at high risk for diabetes, and thus the results may not be applicable to mass screening programs that are not targeted 98-100. Theoretical concerns include the effects of labeling 149 on anxiety and insurability, but available evidence is insufficient
to support or refute these concerns.
Several limitations deserve mention. First, we restricted our review of diabetes treatment to studies with mean diabetes duration of 1 year or less, because we felt that these patient populations would most closely resemble screening-detected populations. Individuals with longstanding type 2 diabetes will likely show greater benefits from treatment, so focusing on treatment of early disease,
in the absence of trials with extended follow-up, may underestimate the effectiveness of treatment and therefore screening interventions. For studies comparing a given treatment among persons with and persons without type 2 diabetes, we included studies of any duration of disease, and the applicability of these data to populations with screening-detected disease is uncertain. Second, attempts to divide patients with diagnosed diabetes into those with a "clinical diagnosis" based on symptoms and those deemed to be "screened" because of alleged asymptomatic status does not truly compare "not screened" with "screened" patients. Third, participants with prediabetes in studies of intensive lifestyle interventions may not be representative of general prediabetic populations. For example, the level of physical inactivity in the DPP cohort was less than that reported in the Third National Health and Nutrition Examination Survey 150. Fourth, most of the data on diabetes treatment were from prespecified subgroup analyses of large trials that included both diabetic and nondiabetic populations. The diabetes and nondiabetes subgroups had important differences, and subgroup analyses were often underpowered to demonstrate significant changes in primary outcomes. Prevention trials among persons with prediabetes were powered to examine the primary outcome of new cases of diabetes and not to examine long-term health outcomes, such as cardiovascular events.
Models rely on data from trials and observational studies and are only as good as the data and assumptions underlying them. All 7 models that we identified that examined the effect of screening interventions 35,43-48 lack transparency to some degree, and all have had 1 or more of their important underlying assumptions criticized 35.
Further research is needed to define the benefits and harms of screening average-risk individuals for type 2 diabetes. We must learn whether early, aggressive glycemic control in persons with diabetes produces improvements in clinical outcomes after many years of follow-up 151. An extension of the largest study of an initial strategy of sustained tight glycemic control in type 1 diabetes 152 suggested that participants originally randomly assigned to tight glycemic control had a significant reduction in cardiovascular events at long-term follow-up despite similar glycemic control in the control group during the postrandomization period 153. To date, similar data are unavailable for type 2 diabetes. We also need studies to define the duration of the prediabetes phase and identify measurable risk factors for progression to diabetes and its complications, particularly cardiovascular disease.
The cost-effectiveness of diabetes screening programs is considered to be mainly determined by the long-term health benefits rather than the cost of detection and treatment of diabetes 154. Thus, intervention research needs to continue focusing on long-term, sustainable interventions that affect health outcomes in real-world settings. Further work is also needed to examine the effect of screening and diagnosis on patient self-efficacy, motivation for lifestyle change, and the potential psychological effects of labeling.
Direct evidence is lacking on the health benefits of detecting type 2 diabetes by either targeted or mass screening, and indirect evidence also fails to demonstrate health benefits for screening general populations or persons at high risk for diabetes complications without hypertension.
Persons with hypertension do benefit from knowing their diagnosis of diabetes, because blood pressure targets are lower than for nondiabetic persons. Although intensive lifestyle interventions delay or prevent diabetes onset in persons with prediabetes, positive effects of this delay on long-term health outcomes have not been adequately demonstrated.
From the Oregon Evidence-based Practice Center of the Oregon Health & Science University and Portland Veterans Administration Medical Center, Portland, Oregon.
Return to Contents
Acknowledgment: The following people provided valuable guidance
and insights: Mark Helfand, MD, MPH, Evelyn Whitlock, MD, MPH, and Peggy Nygren, MA, of the Oregon Evidence-based Practice Center; Tracy Wolff, MD, MPH, and Mary Barton, MD, MPP, at the Agency
for Healthcare Research and Quality; and the U.S. Preventive Services Task Force members Ned Calonge, MD, Russ Harris, MD, MPH, George Isham, MD, MS, and Virginia Moyer, MD, MPH. The authors thank Andrew Hamilton, MLS, MS, for assistance in developing and running search strategies; Peggy Nygren, MA, and Tracy Dana, MLS, for assistance with data abstraction; and Sarah Baird, MS, for technical assistance.
Grant Support: This report was conducted by the Oregon Evidence-based Practice Center under contract to the Agency for Healthcare Research and Quality (contract no. 290-02-0024, Task order no. 2 for the U.S. Preventive Services Task Force.
Potential Financial Conflicts of Interest: None disclosed.
Requests for Single Reprints: Reprints are available from the Agency for Healthcare Research and Quality Web site (www.preventiveservices.ahrq.gov).
AHRQ Publication No. 08-05116-EF-4
Current as of June 2008
Internet Citation:
Norris SL, Kansagara D, Bougatsos C, et al. Screening Adults for Type 2 Diabetes: A Review of the Evidence for the U.S. Preventive Services Task Force. AHRQ Publication No. 08-05116-EF-4, June 2008. Agency for Healthcare Research and Quality, Rockville, MD. http://www.ahrq.gov/clinic/uspstf08/type2/type2art.htm
540 Gaither Road Rockville, MD 20850