Current issues of ACP Journal Club are published in Annals of Internal Medicine


Diagnosis

Fasting plasma glucose was a good test to screen for non-insulin-dependent diabetes mellitus in older adults

ACP J Club. 1992 May-June;116:83. doi:10.7326/ACPJC-1992-116-3-083


Source Citation

Blunt BA, Barrett-Connor E, Wingard DL. Evaluation of fasting plasma glucose as screening test for NIDDM in older adults. Rancho Bernardo Study. Diabetes Care. 1991 Nov;14:989-93.


Abstract

Objective

To examine fasting plasma glucose (FPG) as a screening test for non-insulin-dependent diabetes mellitus.

Design

Different levels of FPG were compared with the results of oral glucose tolerance tests (OGTT).

Setting

An upper middle-class, white community in California.

Patients

801 men and 1050 women between 50 and 79 years old, who did not have insulin-dependent diabetes, were included. 63 participants had non-insulin-dependent diabetes based on previous diagnosis and diabetes medication use.

Description of test and diagnostic standard

Sensitivity, specificity, positive and negative predictive values, and the proportion of the population to be recalled for a confirmatory OGTT (those with a FPG > the cut point) were calculated for FPG cut points ranging between 4.4 and 7.8 mM (at 0.6 mM increments) for different age groups. Diabetes was defined by a 2-hour postchallenge plasma glucose level ≥ 11.1 mM during an OGTT.

Main results

197 participants (10.6%) had non-insulin-dependent diabetes mellitus. Prevalence increased with increasing age, but FPG remained stable. Sensitivity was higher for participants 50 to 64 years old than for those 65 to 79 years old at all FPG levels from 5.5 mM to 7.2 mM (P < 0.05). Specificity ranged from 2% at 4.4 mM to 99% at 7.8 mM and did not differ by age. The positive predictive value increased with increasing FPG and was higher for the older age group at all FPG cut points except 5 mM and 7.2 mM (P < 0.04). The negative predictive value was similar for all FPG cut points and for both age groups. Sensitivity, specificity, {likelihood ratio of a positive and negative test}*, positive and negative predictive values, and the proportion to be recalled for a confirmatory OGTT at a FPG cut point of 6.1 mM were 88%, 87%, {6.7, and 0.14}*, 26%, 99%, and 17% for the younger group and 60%, 80%, {3.0, and 0.25}*, 44%, 93%, and 18% for the older group, respectively.

Conclusions

The sensitivity of fasting plasma glucose as a screening test for non-insulin-dependent diabetes mellitus decreased with age. In patients who were 50 to 64 years of age, a fasting plasma glucose ≥ 6.1 mM was a sensitive and specific test for non-insulin-dependent diabetes mellitus.

Source of funding: National Institute of Diabetes and Digestive and Kidney Diseases.

Address for article reprint: Dr. E. Barrett-Connor, Department of Community and Family Medicine, 0607, University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0607, USA.

*Numbers calculated from data in article.


Commentary

The diagnostic value of the FPG and OGTT has been documented for young patients, especially pregnant women (1). Whereas the focus of this article is the FPG, the key issue is the clinical significance and variability of the OGTT—the reference test or "gold standard"—in this group of older patients. Because the sensitivity and specificity of cut points for the FPG varied within the 2 age strata described, the authors properly question the validity of OGTT as a "gold standard" for elderly patients. Knowing the HbA1c or 2-hour postprandial glucose levels for some patients may have aided assessment of FPG in this setting.

The authors point out that prevalence or pretest probability of disease directly affects the predictive values of diagnostic tests. To further guide clinical decision making, the diagnostic cut points for subsets of this population could have also been determined from receiver operator characteristic curves, and the corresponding likelihood ratios (which are more stable than sensitivity and specificity [2]) could have been calculated.

Present World Health Organization criteria for diabetic screening and diagnosis using the plasma glucose and OGTT are still under discussion (3). In this study, negative predictive values were high and stable in both age strata described. Ruling out diabetes in the elderly using the suggested FPG cut points, therefore, seems feasible. As the writers rightly point out, however, the value to this population of early diagnosis is still subject to debate and further research.

Robert Bloomfield, MD
Carolyn Pedley, MD
Wake Forest University School of Medicine Bowman Gray CampusWinston-Salem, North Carolina, USA

Robert Bloomfield, MD
Wake Forest University School of Medicine Bowman Gray Campus
Winston-Salem, North Carolina, USA

Carolyn Pedley, MD
Wake Forest University School of Medicine Bowman Gray Campus
Winston-Salem, North Carolina, USA


References

1. Spallacy WN. Understanding and managing of pregnancy in women with diabetes mellitus. In: Gold JJ, Josimovich JB: eds. Gynecologic Endocrinology. 4th ed. New York: Plenum Publishing Corp.; 1987:529-40.

2. Sackett DL, Haynes RB, Tugwell P. In: Clinical Epidemiology. Boston: Little Brown & Co.; 1985:101-7.

3. World Health Organization. Diabetes mellitus: Report of the WHO Study Group. Geneva: World Health Organization; 1985; Tech. Rep. Ser. no. 727.

1997 Editorial Update: The recent American Diabetes Association recommendation for a diagnosis of diabetes requries only 2 FPGs ≥ 7 for a positive test. Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 1997; 20:1183-97.