• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information

Table B3Screening Evidence Table (KQ1)

Author, Year
Quality rating
Study objectiveCountry; SettingStudy designLength of follow-upInclusion criteriaOutcomesAdherence Withdrawals (%)ConclusionsComments
Agarwal et al, 200686
To compare the occurrence of diabetic retinopathy in targeted screening diabetic patients with newly-diagnosed diabetic patients in general practiceIndia, rural communities and urban clinicsCross-sectional with comparison groupNAGroup I (targeted diabetes screening): N=173; >30 years who attended rural or urban diabetes screening clinics, who screened (+) for DM2, and who then reported for eye examination
Group II (newly diagnosed in general practice): N=128; diagnosed with DM in last 1 month and reported for eye examination
Diagnosis of diabetic retinopathy:
Group I: 6.4%
Group II: 11.7% (between-group p-value =0.22)
NADiabetic retinopathy was found in both screen-detected and newly-diagnosed in general practice, with no significant difference in prevalence between the 2 groups.Group I: only 15% reported for eye examination; 100% for Group II
Study performed in urban and rural India; may not be applicable to United States populations
Olafsdottir et al, 200785
To establish a gold standard for prevention of blindness in DM2 populations by comparing a DM-screened population to a nonDM population for visual acuitySweden, using national register for diabetes and control group from population registerCohort with comparison groupNAAll inhabitants of Laxa with DM2; this community has a systematic screening program
Age- and sex-matched controls from national register
No significant difference in visual acuity between DM and control groups; except more control subjects had acuity ≥ 1.0 (p=0.027)NAIn a population that had been screened for DM2 and for diabetic eye disease, the prevalence of visual impairment and blindness was no greater than in the control groupDM group was considered ‘screened’ but likely some were detected clinically
Schellhase et al, 200384
To determine if glucose screening reduces the risk of diabetic complicationsUnited States, HMOCase control10 yCases: diagnosed with DM2 after age 3y, had developed 1+ microvascular complications attributable to DM2, enrolled in health plan for 10+y

Control subjects: randomly selected and matched to cases

Exclusion criteria NR
Number of screening BG tests over 10y period: cases 6.3, controls 4.8
88% of testing was random BG; 81% of BG tests occurred without symptoms (i.e. were screened)
OR for BG screening at least once vs no screening: (adjusted) 0.87 (0.38–1.98)
NAPersons who had 1+ screening events in the 10y period had a 13% decreased the risk of developing severe microvascular complications from DM2 after adjusting for multiple confounding factorsThe study included persons tested without symptoms of diabetes, persons with HTN, or other incidental screening with other chronic illnesses that could include CVD

Abbreviations: BG, blood glucose; CVD, cardiovascular disease; DM, diabetes; DM2, type 2 diabetes; HMO, Health Maintenance Organization; HTN, hypertension; N, number of participants, NA, not applicable; NR, not reported; OR, odds ratio; y, years.

From: Appendix B Evidence Tables

Cover of Screening for Type 2 Diabetes Mellitus: Update of 2003 Systematic Evidence Review for the U.S. Preventive Services Task Force
Screening for Type 2 Diabetes Mellitus: Update of 2003 Systematic Evidence Review for the U.S. Preventive Services Task Force [Internet].
Evidence Syntheses, No. 61.
Norris SL, Kansagara D, Bougatsos C, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.