Limits...
Use of glucometer and fasting blood glucose as screening tools for diabetes mellitus type 2 and glycated haemoglobin as clinical reference in rural community primary care settings of a middle income country.

Muktabhant B, Sanchaisuriya P, Sarakarn P, Tawityanon W, Trakulwong M, Worawat S, Schelp FP - BMC Public Health (2012)

Bottom Line: Determinations of CBG and VPG resulted in suspected T2DM cases, with 7.3% when assessed by CBG and 6.4% by VPG using a cutoff point of 7 mmol/L (126 mg/dl).The low sensitivity indicates that using fasting CBG or VPG as a screening tool in the field results in a high proportion of diseased individuals remaining undetected.Further implications of the results and the controversial discussion related to the use of HbA1c as clinical evidence for suffering from T2DM are also discussed.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Nutrition, Faculty of Public Health, Khon Kaen University, Khon Kaen 40002, Thailand.

ABSTRACT

Background: Thailand is considered to be a middle income country, and to control and prevent type 2 diabetes mellitus (T2DM) is one of the main concerns of the Thai Ministry of Public Health (MoPH). Screening for T2DM and care for T2DM patients has been integrated into the primary health care system, especially in rural areas. The intention of this investigation is to link public health research at the academic level with the local health authorities of a district of a north-eastern province of the country.

Methods: Epidemiological methods were applied to validate the screening tools fasting capillary blood glucose (CBG), measured by glucometer and venous blood for the determination of plasma glucose (VPG), used for screening for T2DM among asymptomatic villagers. For assessing the validity of these two methods glycated haemoglobin (HbA1c) values were determined and used as the 'clinical reference'.

Results: All together 669 villagers were investigated. Determinations of CBG and VPG resulted in suspected T2DM cases, with 7.3% when assessed by CBG and 6.4% by VPG using a cutoff point of 7 mmol/L (126 mg/dl). Taking HbA1c determinations with a cutoff point of 7% into account, the proportion of T2DM suspected participants increased to 10.4%. By estimating sensitivity, specificity and the positive predictive value of CBG and VPG against the 'clinical reference' of HbA1c, sensitivity below 50% for both screening methods has been observed. The positive predictive value was determined to be 58.5% for CBG and 56.8% for VPG. The specificity of the two screening tests was over 96%.

Conclusions: The low sensitivity indicates that using fasting CBG or VPG as a screening tool in the field results in a high proportion of diseased individuals remaining undetected. The equally low positive predictive values (below 60%) indicate a high working load for the curative sector in investigating suspected T2DM cases to determine whether they are truly diseased or false positive cases according to the screening method. Further implications of the results and the controversial discussion related to the use of HbA1c as clinical evidence for suffering from T2DM are also discussed.

Show MeSH

Related in: MedlinePlus

ROC curve with CBG and VPG as screening variables and HBA1c as reference (positive actual state 7%). Original calculation has been done with glucose concentration measured in mg/dl
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3403882&req=5

Figure 2: ROC curve with CBG and VPG as screening variables and HBA1c as reference (positive actual state 7%). Original calculation has been done with glucose concentration measured in mg/dl

Mentions: The ROC curve with CBG values as the screening variable and VPG values as the reference (positive actual state 7.0 mmol/L (126 mg/dl)) is shown in Figure 1 and the ROC curves with CBG and VPG as the screening variables and HbA1c as the reference (positive actual state 7%) are provided in Figure 2. Statistical indicators for the interpretation of the three ROC curves are displayed in Table 3. The largest area under these three ROC curves is 0.898, indicating that statistically the best combination is VPG as reference with the CBG results as the screening tool. The smaller area under the curve for the case where HbA1c is the reference and VPG is the screening tool is 0.733. Comparing the lower and upper bounds of the 95% C.I., CBG as a screening tool has a more narrow range (0.620 to 0.888) in comparison to VPG (0.529 to 0.938), taking HbA1c as reference. According to the result of the ROC curves, optimal results would be achieved with a cutoff point of 5.6 mmol/L (101 mg/dl) taking CBG as the screening tool and the VPG as the reference. Sensitivity in this case would be 75% and specificity 78.8%. Optimal cutoff point for CBG would be 5.03 mmol/L (90.6 mg/dl) when taking HbA1c as the reference (with a cutoff point of 7%). Sensitivity would be then 85.7% and specificity 50%. Using VPG as the screening tool, the optimal cutoff point would be 4.6 mmol/L (82.9 mg/dl) when HbA1c is the reference (cutoff point of 7%). The sensitivity in this case would also be 85.7%, but the specificity would decrease to 26.2%.


Use of glucometer and fasting blood glucose as screening tools for diabetes mellitus type 2 and glycated haemoglobin as clinical reference in rural community primary care settings of a middle income country.

Muktabhant B, Sanchaisuriya P, Sarakarn P, Tawityanon W, Trakulwong M, Worawat S, Schelp FP - BMC Public Health (2012)

ROC curve with CBG and VPG as screening variables and HBA1c as reference (positive actual state 7%). Original calculation has been done with glucose concentration measured in mg/dl
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3403882&req=5

Figure 2: ROC curve with CBG and VPG as screening variables and HBA1c as reference (positive actual state 7%). Original calculation has been done with glucose concentration measured in mg/dl
Mentions: The ROC curve with CBG values as the screening variable and VPG values as the reference (positive actual state 7.0 mmol/L (126 mg/dl)) is shown in Figure 1 and the ROC curves with CBG and VPG as the screening variables and HbA1c as the reference (positive actual state 7%) are provided in Figure 2. Statistical indicators for the interpretation of the three ROC curves are displayed in Table 3. The largest area under these three ROC curves is 0.898, indicating that statistically the best combination is VPG as reference with the CBG results as the screening tool. The smaller area under the curve for the case where HbA1c is the reference and VPG is the screening tool is 0.733. Comparing the lower and upper bounds of the 95% C.I., CBG as a screening tool has a more narrow range (0.620 to 0.888) in comparison to VPG (0.529 to 0.938), taking HbA1c as reference. According to the result of the ROC curves, optimal results would be achieved with a cutoff point of 5.6 mmol/L (101 mg/dl) taking CBG as the screening tool and the VPG as the reference. Sensitivity in this case would be 75% and specificity 78.8%. Optimal cutoff point for CBG would be 5.03 mmol/L (90.6 mg/dl) when taking HbA1c as the reference (with a cutoff point of 7%). Sensitivity would be then 85.7% and specificity 50%. Using VPG as the screening tool, the optimal cutoff point would be 4.6 mmol/L (82.9 mg/dl) when HbA1c is the reference (cutoff point of 7%). The sensitivity in this case would also be 85.7%, but the specificity would decrease to 26.2%.

Bottom Line: Determinations of CBG and VPG resulted in suspected T2DM cases, with 7.3% when assessed by CBG and 6.4% by VPG using a cutoff point of 7 mmol/L (126 mg/dl).The low sensitivity indicates that using fasting CBG or VPG as a screening tool in the field results in a high proportion of diseased individuals remaining undetected.Further implications of the results and the controversial discussion related to the use of HbA1c as clinical evidence for suffering from T2DM are also discussed.

View Article: PubMed Central - HTML - PubMed

Affiliation: Department of Nutrition, Faculty of Public Health, Khon Kaen University, Khon Kaen 40002, Thailand.

ABSTRACT

Background: Thailand is considered to be a middle income country, and to control and prevent type 2 diabetes mellitus (T2DM) is one of the main concerns of the Thai Ministry of Public Health (MoPH). Screening for T2DM and care for T2DM patients has been integrated into the primary health care system, especially in rural areas. The intention of this investigation is to link public health research at the academic level with the local health authorities of a district of a north-eastern province of the country.

Methods: Epidemiological methods were applied to validate the screening tools fasting capillary blood glucose (CBG), measured by glucometer and venous blood for the determination of plasma glucose (VPG), used for screening for T2DM among asymptomatic villagers. For assessing the validity of these two methods glycated haemoglobin (HbA1c) values were determined and used as the 'clinical reference'.

Results: All together 669 villagers were investigated. Determinations of CBG and VPG resulted in suspected T2DM cases, with 7.3% when assessed by CBG and 6.4% by VPG using a cutoff point of 7 mmol/L (126 mg/dl). Taking HbA1c determinations with a cutoff point of 7% into account, the proportion of T2DM suspected participants increased to 10.4%. By estimating sensitivity, specificity and the positive predictive value of CBG and VPG against the 'clinical reference' of HbA1c, sensitivity below 50% for both screening methods has been observed. The positive predictive value was determined to be 58.5% for CBG and 56.8% for VPG. The specificity of the two screening tests was over 96%.

Conclusions: The low sensitivity indicates that using fasting CBG or VPG as a screening tool in the field results in a high proportion of diseased individuals remaining undetected. The equally low positive predictive values (below 60%) indicate a high working load for the curative sector in investigating suspected T2DM cases to determine whether they are truly diseased or false positive cases according to the screening method. Further implications of the results and the controversial discussion related to the use of HbA1c as clinical evidence for suffering from T2DM are also discussed.

Show MeSH
Related in: MedlinePlus