. Author manuscript; available in PMC: 2015 Jul 1.

Published in final edited form as: Pharmacoepidemiol Drug Saf. 2014 Mar 24;23(6):628–635. doi: 10.1002/pds.3609

Table 3.

Weighted sensitivity, specificity, and positive predictive value (PPV) of the antipsychotic polypharmacy claims-based measure in a random sample of Medicaid mental health clinic clients (n = 312)

		Validation criterion: medical record-based measure
		Antipsychotic polypharmacy	No antipsychotic polypharmacy	Total	Sensitivity (95%CI)	Specificity (95%CI)	PPV (95%CI)
Medicaid claims-based measure	Antipsychotic polypharmacy	36	3	39	81.8% (67.7-91.8)	98.8% (96.7-99.7)	92.3% (79.1-98.4)	Raw
	No antipsychotic polypharmacy	8	265	273	79.4% (70.4-86.6)	99.1% (98.2-99.6)	90.9% (83.1-95.9)	Weighted
Total	44	267	312

Notes:

95% exact binomial confidence intervals were calculated for each proportion. Weighted results reflect the underlying sampling probability.

Antipsychotic polypharmacy: two or more antipsychotics for longer than 90 days, as of 1/4/2008.

No antipsychotic polypharmacy: did not meet criteria for antipsychotic polypharmacy, as of 1/4/2008.