JAMDA xxx (2017) 1e2
JAMDA journal homepage: www.jamda.com
Completion of an Outpatient Visit After Skilled Nursing Facility Discharge and Readmission Risk Ernest Shen PhD a, Angelika Alem MS, MPH a, Peter Khang MD, MPH a, Heather L. Watson MBA a, Jing Li MD, MS b, Huong Q. Nguyen PhD, RN a, * a b
Kaiser Permanente Southern California, Pasadena, CA Department of Internal Medicine, Center for Health Services Research, University of Kentucky, Lexington, KY
a b s t r a c t Keywords: SNF readmission outpatient visit
Objectives: Examine the association between completion of an outpatient visit with a physician or advanced practice provider (PCP) within 7 days of discharge from a short skilled nursing facility (SNF) stay and 30-day readmission and determine if functional status at discharge moderates visit effectiveness. Design: Retrospective cohort study. Setting: Large integrated health care system. Participants: Adults 65 years and older, discharged home from a short SNF stay (n ¼ 4073). Intervention: None. Measurements: Exposure is completion of an outpatient visit with a PCP within 7 days of discharge from an SNF. Primary outcome is readmission within 30 days of SNF discharge. Covariates included gender, risk score for readmission or early death, medical or surgical hospitalization, SNF facility, SNF length of stay, SNF stay in the previous 12 months, discharge to home or home health, and discharge functional independence measures (FIM). Results: A total of 476 (11.6%) patients were readmitted within 30 days of SNF discharge. Patients who completed an outpatient visit with a PCP within 7 days of SNF discharge had a 23% higher risk of being readmitted compared to patients who did not complete any visit (hazard ratio [HR] 1.23, 95% conﬁdence interval [CI] 1.01e1.50). Patients who had FIM scores 80 and completed a visit had an increased readmission risk (HR 1.37, 95% CI 1.04e1.79); the increased risk was not seen for those with worse functional impairment, FIM <80 (HR 1.11, 95% CI 0.85e1.46). Conclusion: The ﬁnding of increased risk of readmission post SNF discharge with completion of an outpatient visit likely reﬂects inadequate adjustment for selection bias in this observational study, which strongly argues for the need to design prospective studies to test transitional care services post SNF discharge. Ó 2017 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
Hospital readmission within 30 days of a skilled nursing facility (SNF) discharge is the primary measure of quality under the Improving Medicare Post-Acute Care Transformation Act of 2014.1,2 Although numerous studies have focused on testing the effectiveness of different clusters of transitional care services from hospital to home,3 as well as interventions to reduce rehospitalizations during an SNF stay,4,5 the evidence base for transitional care services post SNF discharge is minimal.6 We recently found that completion of an outpatient follow-up visit within 7 days of hospital discharge to home was associated with lower 30-day readmission risk in a large Medicare The authors declare no conﬂicts of interest. * Address correspondence to Huong Q. Nguyen, PhD, RN, Department of Research and Evaluation, Kaiser Permanente Southern California, 100 S. Los Robles Avenue, 2nd Floor, Pasadena, CA 91101. E-mail address: [email protected]
(H.Q. Nguyen). http://dx.doi.org/10.1016/j.jamda.2017.05.023 1525-8610/Ó 2017 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
Advantage population.7 We aimed to examine if completion of any outpatient visit with a physician or advanced practice provider (PCP) within 7 days of discharge from a short SNF stay was associated with 30-day readmission compared with no visit and whether functional status at discharge moderates visit effectiveness. Methods Kaiser Permanente members ages 65 and older, who were discharged alive from a short SNF stay (30 days) between January 1, 2011, and December 31, 2014, to home or home health from 72 SNFs, had a completed functional independence measure (FIM)8 at SNF discharge, and remained enrolled in the health plan for at least 30 days postdischarge, were included (n ¼ 4073). Only the ﬁrst SNF episode during this period was examined. An outpatient visit could have been
E. Shen et al. / JAMDA xxx (2017) 1e2
scheduled before or after SNF discharge by the patient or a provider for any reason related to or unrelated to the hospitalization and SNF stay. The primary outcomes of readmission within 30 days after SNF discharge were obtained from the electronic medical records and claims. We used Cox proportional hazards regression and treated visit completion as a time-dependent variable that could change in the ﬁrst 7 days. Patients who died before 30 days were censored. Covariates that were either meaningfully associated with visit completion or outcome (gender, risk for readmission or early death [LACE score9]), hospitalization for a medical or surgical condition, SNF facility, SNF length of stay, SNF stay in the previous 12 months, discharge to home or home health, and discharge FIM scores were included. Inverse probability of treatment weights were used to adjust for differences in these covariates.10 We also stratiﬁed the analyses by FIM scores (<80 or 80), which indicates level of assistance needed post SNF discharge.8 Analyses were performed with SAS v.9.3 (SAS Institute Inc, Cary, NC). A P < .05 was considered signiﬁcant. Results A total of 476 (11.6%) patients were readmitted within 30 days of being discharged alive from SNF (Table 1). Patients who were readmitted were slightly older; more likely to be men; on the medicine Table 1 Sample Characteristics 30-Day Readmission From SNF Discharge* Total, n ¼ 4073 Outpatient visit 7 days No visit 2411 (59%) Visit with providery 1662 (41%) Sociodemographics Age 79.9 (7.8) Women 2685 (66%) Race: white 2638 (65%) Partnered 1661 (41%) Education: college or morez 31% (17%) z Income: $20,000 or more 84% (9%) Characteristics of index hospitalization Service line Medicine 1813 (45%) Surgical 2260 (56%) LACE readmission risk score 9.7 (3.1) LACE 11þ 1859 (46%) IQR: 10e12 Charlson comorbidity index 2.8 (2.0) IQR: 3e4 Length of stay 5.3 (5.2) IQR: 4e6 Characteristics of SNF Stay Length of stayx <13 days 1823 (45%) 13 days 2250 (55%) Functional impairment measurek At discharge 81.8 (17.1) <80 1545 (38%) 80 2528 (62%) Discharge disposition Home 1111 (27%) Home health 2962 (73%)
No, n ¼ 3597
Yes, n ¼ 476
2113 (59%) 1484 (41%)
298 (63%) 178 (37%)
79.8 2404 2341 1448 31% 84%
81.1 281 297 213 30% 84%
(7.8) (67%) (65%) (40%) (17%) (9%)
(7.7) (59%) (62%) (45%) (17%) (9%)
<.001 <.001 .93 .04 .12 .68 <.001
1511 (42%) 2086 (58%) 9.6 (3.2) 1564 (43%) IQR: 10e12 2.8 (2.0) IQR: 2e4 5.1 (4.9) IQR: 4e6
302 (63%) 174 (37%) 11.0 (2.7) 295 (62%) IQR: 11e13 3.4 (2.1) IQR: 3e5 6.83 (7.0) IQR: 5e8
1632 (45%) 1965 (55%)
191 (40%) 285 (60%)
<.001 <.001 <.001 <.001
82.6 (16.7) 75.1 (18.5) 1290 (36%) 255 (54%) 2307 (64%) 221 (46%)
<.001 <.001 <.01
955 (27%) 2642 (74%)
156 (33%) 320 (67%)
IQR, interquartile range; SNF, skilled nursing facility. Values are presented as either mean (SD) or n (column %), unless otherwise indicated. *Only short SNF stays were included (30 days). y Physician and advanced practice providers included medical doctors, doctor of osteopathic medicine, nurse practitioners, and physician assistants. z Census-based. x SNF length of stay median cutoff. k FIM cutoff of 80 indicates minimal assistance needed from a family caregiver.
service; had more comorbidities, longer hospital stay, and worse FIM score at SNF discharge; and were less likely to have received home health after discharge (all, P < .01). Patients who completed an outpatient visit with any physician or PCP within 7 days of SNF discharge had a 23% higher risk of being readmitted within 30 days compared with patients who did not complete any visit (hazard ratio [HR] 1.23, 95% conﬁdence interval [CI] 1.01e1.50). Functional independence scores at SNF discharge modiﬁed the effects of the outpatient visit, whereby patients who had FIM scores 80 had an increased readmission risk (HR 1.37, 95% CI 1.04e1.79); however, completion of an outpatient visit was not signiﬁcantly associated with readmission risk in those with worse functional impairment, FIM <80, (HR 1.11, 95% CI 0.85e1.46). Discussion The ﬁnding that completion of an outpatient visit with a physician or other PCP within 7 days of SNF discharge was associated with a higher risk of readmission within 30 days of SNF discharge was unexpected given our recent positive ﬁndings in a large cohort of patients discharged directly home from hospital.7 However, the fact that patients who were less functionally impaired at SNF discharge were at higher readmission risk when they completed an outpatient visit within 7 days compared with no increased risk for those who were more functionally impaired likely reﬂects confounding by indication and our inability to fully adjust for other unmeasured confounders (eg, social factors, burden of geriatric syndromes, or account for the quality of care provided during the visits). It is also possible that transitional care services that work for a general older population well enough to be discharged home may not generalize to frail individuals with multiple physical and psychological comorbidities that is characteristic of a SNF population. More prospective research is critically needed to evaluate the effects of different post-acute care transitional services for this vulnerable older adult population. Acknowledgments We thank Ms Janet Lee, Jianjin Wang, and Adeline Wong for their assistance in acquiring the data. Written permission was obtained from all persons named in the acknowledgment section. References 1. Centers for Medicare and Medicaid Services. Medicare program; prospective payment system and consolidated billing for skilled nursing facilities for FY 2017, SNF value-based purchasing program, SNF quality reporting program, and SNF payment models research. Final rule. Fed Regist 2016;81: 51969e52053. 2. Stefanacci RG, Riddle A. Delivering on quality measures: Six new CMS SNF quality measures. Geriatr Nurs 2016;37:231e234. 3. Hansen LO, Young RS, Hinami K, et al. Interventions to reduce 30-day rehospitalization: A systematic review. Ann Intern Med 2011;155:520e528. 4. Ouslander JG, Lamb G, Tappen R, et al. Interventions to reduce hospitalizations from nursing homes: Evaluation of the INTERACT II collaborative quality improvement project. J Am Geriatr Soc 2011;59:745e753. 5. Ingber MJ, Feng Z, Khatutsky G, et al. Initiative to reduce avoidable hospitalizations among nursing facility residents shows promising results. Health Aff (Millwood) 2017;36:441e450. 6. Toles M, Colon-Emeric C, Asafu-Adjei J, et al. Transitional care of older adults in skilled nursing facilities: A systematic review. Geriatr Nurs 2016;37:296e301. 7. Shen E, Koyama SY, Huynh DN, et al. Association of a dedicated post-hospital discharge follow-up visit and 30-day readmission risk in a medicare advantage population. JAMA Intern Med 2017;177:132e135. 8. Uniform Data System for Medical Rehabilitation. The FIM Instrument: Its Background, Structure, and Usefulness. Buffalo, NY: UDSMR; 2012. 9. van Walraven C, Dhalla IA, Bell C, et al. Derivation and validation of an index to predict early death or unplanned readmission after discharge from hospital to the community. CMAJ 2010;182:551e557. 10. McCaffrey DF, Grifﬁn BA, Almirall D, et al. A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Stat Med 2013;32:3388e3414.