Initial impact and cost of a nationwide population screening campaign for diabetes in Brazil: A follow up study

Table 2 Parameters considered in base case analysis and range for selected parameter estimates included in sensitivity analysis

Parameters	Base Case Estimate	Range
Proportion of population ≥ 40 years who participated in the screening program	73%	-
Percentage of tests in screenees who reported to be under treatment for diabetes prior to screening	5.4%	4.8 – 6.1^a
Percentage of subjects in fasting state when screened^b	46.7%	30 – 50%^c
Percentage of positive screening tests^d	16.4%	-
Percentage of positive screenees who reported having diabetes diagnosis prior to the screening program^e	16%	14 – 18.1%^a
Percentage of positive screenees who returned for confirmatory testing	37.1%	0 – 100%^c
Percentage of positive screenees who were diagnosed as having diabetes mellitus	10.1%	-
Percentage of positive screenees diagnosed with diabetes and incorporated into the healthcare system	9.4%	-
Additional local costs	0%	10 – 25%^f
Estimated labor costs	US$ 5.99 million	US$ 4.65 – 8.98 million
Percentage of positive screenees who returned for confirmatory testing in the public system (as opposed to the private health sector)	100%	75 – 25%^c

National Campaign to Detect Diabetes Mellitus. Brazil, 2001.
^a Range corresponds to the 95% confidence interval of each estimate from the probabilistic sample.
^b Fasting was defined as absence of food ingestion 4 hours prior to capillary glucose test.
^c Range corresponds to an arbitrary estimate of the authors.
^d Fasting glucose ≥ 5.6 mmol/l or a casual glucose ≥ 7.8 mmol/l.
^e These subjects knew their previous diabetes diagnosis; however, they had not provided that information when asked about it during screening program.
^f Range expressed as an additional percentage of national costs.

ISSN: 1472-6963