Retrospective derivation of a causal pathway for diabetic ketoacidosis in adult patients with type 2 diabetes mellitus
Authors' conclusion
in Detroit, MI, prior ketosis and private or Medicare health insurance were significantly associated with new or recurrent DKA in T2KPDM. Medication non-adherence had a mediating role.
Study context
- Population
- Adults (>18 years) seeking emergency care at four hospitals in greater Detroit, identified by a laboratory order for ß-hydroxybutyrate; Diabetic ketoacidosis (DKA) defined as ß-hydroxybutyrate >20 mg/dL326 non-type 1
- Participants
- 450 adults
- Exposure
- Risk factors for DKA among adults with non-type 1 DM
- Outcome
- DKA (ß-hydroxybutyrate >20 mg/dL)
Methodological detailsmulticentre, directed acyclic graph causal pathway, Two research coordinators abstracted 119 data fields
- Multicentre
- Directed acyclic graph causal pathway
- Two research coordinators abstracted 119 data fields
Main finding
As reported in the abstract
Non-type 1 patients with DKA
37%
Unadjusted odds for DKA, non-adherence
1.74
95% CI 1.08 to 2.21
Unadjusted odds for DKA, arrival by Emergency Medical Services
0.54
95% CI 0.33 to 0.86
Unadjusted odds for DKA, private or Medicare insurance
6.80
95% CI 4.00 to 11.60
Observational data: associations, not effects.
Further findings
Median
| Outcome | DKA | No DKA | P value |
|---|---|---|---|
| HbA1c by DKA statusHigher with DKA | 11.3% | 9.5% | Mann-Whitney U p<0.001 |
HbA1c by DKA status
- HbA1c also higher in patients with a history of non-adherence
ß-hydroxybutyrate, glucose, bicarbonate: non-type 1 vs T1D
- Not different between non-type 1 and T1D patients
ICU admission and hospital LOS among the four phenotypes with DKA
- Similar between the four phenotypes with DKA
Original publication
Composed by AI from the published abstract and checked against it. How Briefs are made →