Table 3.

Multivariate regression results for all study outcomes by drug class cohort

	High adherence			Hospital inpatient ER visit			Non‐office based Evaluation & Management Costs			Office Based Part B Costs
	Diabetes	RASA	Statins	Diabetes	RASA	Statins	Diabetes	RASA	Statins	Diabetes	RASA	Statins
NP	−0.011*	0.001	−0.009**	−0.031***	−0.029***	−0.026***	−87.745***	−100.072***	−83.173***	−195.436***	−193.244***	−187.077***
	(0.006)	(0.003)	(0.004)	(0.004)	(0.002)	(0.003)	(17.842)	(10.398)	(11.236)	(8.995)	(5.611)	(5.736)
Indepentent SOP	−0.009	0.013**	0.013*	−0.000	0.009	0.009*	−93.465**	−49.225**	−38.952*	18.424	39.234***	31.522***
	(0.011)	(0.006)	(0.007)	(0.010)	(0.005)	(0.005)	(40.767)	(20.178)	(21.911)	(19.210)	(9.467)	(10.779)
High adherence				−0.094***	−0.099***	−0.072***	−384.209***	−368.170***	−294.048***	−40.369***	−57.162***	−33.154***
				(0.004)	(0.002)	(0.003)	(22.004)	(12.062)	(12.286)	(6.250)	(3.884)	(4.226)
N	75, 649	195, 950	190, 087

Notes: All models adjusted for beneficiaries’ sex, race, age, dual eligibility status, provider type, rural region, CMS regions, year, NP independence, and Elixhauser comorbidities.

Cost and utilization models also adjusted for high adherence.

Coefficients for binary outcomes are reported in probabilities.

Results based on an attribution threshold of 30%.

Standard errors in parentheses; *P < 0.10, **P < 0.05, ***P < 0.01.

Propensity score ranges: anti‐diabetics: ≥0.0296622 and ≤0.5590561; RASA: ≥0.0274983 and ≤0.5630087; statins: ≥0.0322372 and ≤0.5937169.