Skip to main content

Table 1 Definitions for automated database algorithm to identify diabetes

From: Positive predictive value of a case definition for diabetes mellitus using automated administrative health data in children and youth exposed to antipsychotic drugs or control medications: a Tennessee Medicaid study

 

Inpatient-Primary

Inpatient-Secondary

Outpatient †

Prescription ‡

Diabetes-Related Medical Care Encounter *

Definition

Primary discharge diagnosis of diabetes (ICD-9-CM codes of 250, 250.0, 250.1, 250.2, 250.3, 250.9)§

Inpatient stay with 1) a secondary or admission diagnosis for diabetes; or 2) a physician encounter with a primary diagnosis of diabetes during the hospital stay period, defined as the day prior to admission through the day following discharge.

Outpatient visit with a primary diagnosis of diabetes, excluding those during the hospital stay period.

Filled prescription for any diabetes medication, including insulin, insulin adjuncts (pramlintide), and oral hypoglycemics. There can be no diagnosis, primary or secondary, of polycystic ovarian syndrome in the interval [tx-120,tx + 120]

Index date (tx) initial

Admission date or prior day if ED/ outpatient visit with diabetes diagnosis on that day

As for inpatient-primary

Day of visit

Day of prescription fill

Additional Criteria for Diabetes Case

Exclusion||

None

   

Confirmation||, primary definition

None

1. Diabetes medication prescription, or 2. Outpatient or inpatient (any) diagnosis

1. Diabetes medication prescription, or 2. Inpatient (any) diagnosis

1. Outpatient or inpatient (any) diagnosis, or

2. Subsequent prescription, and procedure indicating diabetes management¶, and no diagnosis absent/irregular menses.

Confirmation||, secondary definition

None

None

Glycosylated hemoglobin test (indicating possible diabetes management).

As above

Index date, final

If diabetes-related procedure# in the interval [tx-29, tx-1] tx is set to procedure date.

  1. *Does not include deaths as there were none with diabetes coded as an underlying cause of death for cohort members during the study period.
  2. †Includes ED visits, but excludes case management services because these may not include patient assessment. Only primary outpatient diagnoses considered because secondary diagnoses occurred very infrequently in the absence of a primary diagnosis and a preliminary study showed they added little predictive value.
  3. ‡If both a prescription and other encounter on the same day, classified as a prescription encounter. Prescriptions with a concurrent diagnosis of polycystic ovarian syndrome were not considered as diabetes-related medical care encounters, given that this disease is frequently treated with oral hypoglycemics.
  4. §Does not include 250.4-250.8, which are chronic complications of diabetes and thus unlikely to be present for newly diagnosed cases, particularly in a population of children/youth.
  5. ||Period for exclusion or confirmation is [tx-120, tx + 120].
  6. ¶Diabetes management: HbA1c (glycated hemoglobin), glucose test strips, glucose monitor, insulin pump.
  7. #Diabetes-related procedure: HbA1c, islet cell antibody test, insulin RIA, or metabolic panel.
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us