. 2022 May 6;3(5):e220900. doi: 10.1001/jamahealthforum.2022.0900

Table 3. Differences in the Trajectories of the Percentage of New Monthly Enrollees With ADRD, Monthly Patient Census With ADRD, and Care Days Provided to Patients With ADRD Coinciding With Recent Policy Changes Affecting the Medicare Hospice Benefit as Estimated by Fully Adjusted Models and Stratified by For-Profit Hospices vs Nonprofit/Government-Owned Hospices^a.

Characteristic	For-profit^a		Nonprofit/government-owned^a		P value for-profit vs nonprofit^b
Characteristic	Estimate (95% CI)	P value	Estimate (95% CI)	P value	P value for-profit vs nonprofit^b
Percentage of new monthly enrollees with ADRD
Pre-FTT/debility exclusion slope^c	0.07 (0.06 to 0.08)	<.001	0.03 (0.02 to 0.04)	<.001	<.001
Change during FTT/debility exclusion notice	4.21 (3.70 to 4.72)	<.001	3.10 (2.60 to 3.61)	<.001	.002
Post-FTT/debility exclusion slope^c	0.14 (0.08 to 0.19)	<.001	−0.01 (−0.06 to 0.04)	.74	<.001
Change during IMPACT Act passage	−1.86 (−2.84 to −0.88)	<.001	−0.59 (−1.49 to 0.31)	.20	.18
Change during IMPACT Act implementation	−1.88 (−2.87 to -0.89)	<.001	−2.13 (−3.05 to −1.21)	<.001	.75
Post–IMPACT Act slope^c	0.13 (0.08 to 0.18)	<.001	0.07 (0.02 to 0.12)	.01	.06
Change during TTP implementation	0.07 (−0.41 to 0.55)	.77	0.36 (−0.12 to 0.83)	.14	.26
Post TTP-slope^c	0.01 (0.00 to 0.02)	.14	0.01 (0.00 to 0.02)	.105	.64
Percentage of monthly patient census with ADRD
Pre-FTT/debility exclusion slope^c	0.15 (0.14 to 0.16)	<.001	0.08 (0.08 to 0.09)	<.001	<.001
Change during FTT/debility exclusion notice	3.15 (2.86 to 3.45)	<.001	2.79 (2.47 to 3.11)	<.001	.33
Post-FTT/debility exclusion slope^c	0.2 (0.17 to 0.23)	<.001	−0.01 (−0.04 to 0.02)	.46	<.001
Change during IMPACT Act passage	−0.79 (−1.36 to −0.22)	.007	−0.53 (−1.13 to 0.07)	.08	.78
Change during IMPACT Act implementation	−1.31 (−1.88 to −0.73)	<.001	−1.80 (−2.42 to −1.19)	<.001	.59
Post–IMPACT Act slope^c	0.02 (0.00 to 0.05)	.104	0.10 (0.07 to 0.13)	<.001	.02
Change during TTP implementation	−0.38 (−0.66 to −0.11)	.006	−0.23 (−0.55 to 0.08)	.14	.96
Post TTP-slope^c	0.02 (0.01 to 0.02)	<.001	0.03 (0.02 to 0.04)	<.001	<.001
Percentage of monthly care days provided to patients with ADRD
Pre-FTT/debility exclusion slope^c	0.16 (0.16 to 0.17)	<.001	0.10 (0.10 to 0.11)	<.001	<.001
Change during FTT/debility exclusion notice	3.21 (2.89 to 3.54)	<.001	2.94 (2.57 to 3.30)	<.001	.72
Post-FTT/debility exclusion slope^c	0.20 (0.17 to 0.23)	<.001	−0.01 (−0.05 to 0.03)	.55	<.001
Change during IMPACT Act passage	−0.50 (−1.12 to 0.12)	.113	−0.57 (−1.27 to 0.12)	.11	.89
Change during IMPACT Act implementation	−1.27 (-1.9 to −0.64)	<.001	−1.69 (−2.40 to −0.98)	<.001	.57
Post–IMPACT Act slope^c	0.00 (−0.03 to 0.03)	.98	0.11 (0.07 to 0.15)	<.001	<.001
Change during TTP implementation	−0.38 (−0.67 to −0.08)	.01	−0.35 (−0.71 to 0.01)	.06	.37
Post TTP-slope^c	0.01 (0.01 to 0.02)	<.001	0.04 (0.03 to 0.05)	<.001	<.001

Abbreviations: ADRD, Alzheimer disease and related dementias; FTT, failure-to-thrive and debility exclusion as a primary hospice diagnosis; IMPACT, Improving Medicare Post-Acute Care Transformation Act; TTP, 2-tier payment.

^{^a}

Adjusted for month indicators, (absorbed) hospice fixed effects and time-varying hospice-level covariates (hospice ownership, total registered nurses, total employees, new enrollee demographic distributions [age, sex, race and ethnicity], percentage of care days provided at routine home care and continuous home care levels, and percentage of care days provided in the community/home setting, nonskilled long-term care setting, and skilled nursing home/inpatient setting) and time-varying county-level covariates (population, percent population older than 65 years and older than 85 years, sex and race and ethnicity distributions, number of active physicians, rural-urban setting, percentage living below the poverty level, median household income).

^{^b}

For each policy indicator/slope, P values indicate significance of differences between for-profit vs nonprofit/government-owned hospices. They are P values associated with the interaction term between hospice ownership and each corresponding policy indicator/slope term in analyses on the full sample of all hospices, excluding absorbed hospice fixed effects, and fully adjusted for covariates identical to those used in the fully stratified analyses.

^{^c}

Slope coefficients are direct estimates of the slope during each specified period.