Your capstone change project begins this week when you identify a practice issue that you believe needs to change. The practice issue must pertain to a systematic review that you must choose from a List of Approved Systematic Reviews for the capstone proj
Tysha625
readmission-capstone.pdf
Research
Original Investigation
Preventing 30-D ay Hospital Readmissions A Systematic Review and Meta-analysis of Randomized Trials Aaron L. Leppin. MD; Michael R. Gionfriddo, PharmD; Maya Kessler, MD; Juan Pablo Brito, MBBS; Frances S. Mair, MD; Katie Gallacher, MBChB; Zhen Wang, PhD; Patricia J. Erwin. MLS; Tanya Sylvester, BS; Kasey Boehmer, BA; Henry H. Ting, MD, MBA; M. Hassan Murad, MD; Nathan D. Shippee, PhD; V ictor M. Montori, MD
IMPORTANCE Reducing early (<30 days) hospital readmissions is a policy priority aimed at improving health care quality. The cumulative complexity model conceptualizes patient context. It predicts that highly supportive discharge interventions will enhance patient capacity to enact burdensome self-care and avoid readmissions.
OBJECTIVE To synthesize the evidence o f the efficacy o f interventions to reduce early hospital readmissions and identify intervention features-including their impact on treatment burden and on patients' capacity to enact postdischarge self-care-that might explain their varying effects.
DATA SOURCES We searched PubMed, Ovid MEDLINE, Ovid EMBASE, EBSCO CINAHL, and Scopus (1990 until April 1,2013), contacted experts, and reviewed bibliographies.
s t u d y SELECTION Randomized trials that assessed the effect o f interventions on all-cause or unplanned readmissions within 30 days o f discharge in adult patients hospitalized for a medical or surgical cause for more than 24 hours and discharged to home.
DATA EXTRACTION a n d SYNTHESIS Reviewer pairs extracted trial characteristics and used an activity-based coding strategy to characterize the interventions; fidelity was confirmed with authors. Blinded to trial outcomes, reviewers noted the extent to which interventions placed additional work on patients after discharge or supported their capacity for self-care in accordance w ith the cumulative complexity model.
m a in outcomes a n d measures Relative risk o f all-cause or unplanned readmission w ith or w ithout out-of-hospital deaths at 30 days postdischarge.
RESULTS In 42 trials, the tested interventions prevented early readmissions (pooled random-effects relative risk, 0.82 [95% Cl, 0.73-0.91]; P < .001; I2 = 31%), a finding that was consistent across patient subgroups. Trials published before 20 0 2 reported interventions that were 1.6 times more effective than those tested later (interaction P = .01). In exploratory subgroup analyses, interventions with many components (interaction P = .001), involving more individuals in care delivery (interaction P = .05), and supporting patient capacity for self-care (interaction P = .04) were 1.4,1.3, and 1.3 times more effective than other interventions, respectively. A post hoc regression model showed incremental value in providing comprehensive, postdischarge support to patients and caregivers.
CONCLUSIONS AND relevance Tested interventions are effective at reducing readmissions, but more effective interventions are complex and support patient capacity for self-care. Interventions tested more recently are less effective.
JAMA Intern Med. 2014:174(7):1095-1107. doLlO.lOOl/jamainternmed.2014.1608 Published online May 12,2014.
Author Affiliations: Author affiliations are listed at the end of this article.
Corresponding Author: Victor M. Montori, MD, Knowledge and Evaluation Research Unit, Department of Medicine, Mayo Clinic, 200 First StSW, Plummer Bldg, Rochester. MN 55905 (montori.victor @mayo.edu).
[J Supplemental con tent at jamainternalmedicine.com
1095
R e s e a rc h O r ig in a l In v e s tig a t io n P r e v e n tin g 3 0 - D a y H o s p ita l R e a d m is s io n s
E arly hospital readm issions have been recognized as a com m on and costly occurrence, particularly am ong el derly and high-risk patients. One in 5 Medicare benefi
ciaries is readm itted w ithin 30 days, for example, at a cost of m ore th an $26 billion per year.1 To encourage im provem ent in th e quality o f care and a reduction in unnecessary health ex pense, policymakers, reim bursem ent strategists, and the US governm ent have m ade reducing 30-day hospital read m is sions a national priority.2"4 Achieving th is goal, however, re quires a more complete understanding of the underlying causes o f readm ission.
The cum ulative com plexity m odel (CuCoM)5 is a fram e work developed by our research group th at conceptualizes p a tie n t co n tex t as a balance b etw e en w orkload an d capacity (Figure 1). Workload consists o f all th e work of being a patient and includes efforts to u n d erstan d and plan for care, to enroll th e s u p p o rt o f o th e rs , an d to access an d u se h e a lth care services.6,7 Capacity is determ ined by th e quality and avail ability o f resources th a t patients can mobilize to carry out this work (physical an d m ental health, social capital, financial re sources, and environm ental assets). The CuCoM is novel in its consideration of th e effects o f treatm ent burden on patient con text, and it illustrates how infeasible, u n su p p o rted , and con- text-irreverent care can lead to poor health outcom es and re du ced h ea lth care effectiveness. Because p atien ts recently discharged from th e hospital are in a state o f extrem e physi ologic and psychological vulnerability,8 their capacity for en acting self-care is low. The CuCoM predicts th at, unless suffi cient support is given to enhance patient and caregiver capacity to carry o u t th e work o f patienthood, placing highly b u rd en some discharge dem ands on these patients will lead to poor outcom es and hospital readm ission.
To evaluate th e validity o f th e CuCoM and provide hy pothesis-generating work in th e understanding of patient con text, we chose to synthesize th e evidence on the efficacy of interventions to reduce early hospital readm issions. In p ar ticular, we sought to determ ine th e degree to w hich a n u m ber o f intervention ch aracteristics-in clu d in g their im pact on p a tie n t cap acity an d w o rk lo a d -m ig h t acco u n t for d iffer ences in their effectiveness.
F ig u r e 1. T h e C u m u l a t i v e C o m p l e x i t y M o d e l
P a tie n t c o n t e x t is r e p r e s e n te d as a b a la n c e b e t w e e n w o r k lo a d a n d c a p a c ity .
T h is b a la n c e m u s t b e o p t im iz e d t o e n s u r e c a re e ffe c tiv e n e s s a n d im p r o v e
o u t c o m e s . In t u r n , t h e o u t c o m e s a c h ie v e d fe e d b a c k t o a f f e c t t h e
w o r k lo a d - c a p a c ity b a la n c e .
Methods
A registered protocol (PROSPERO CRD42013004773) guided the conduct o f this review,9 w hich we report in adherence to th e Preferred Reporting Item s for Systematic Reviews and M eta analyses (PRISMA) S tatem ent.10
Eligibility Criteria Eligible studies were random ized trials reported in English or Spanish, since 1990, th a t assessed th e effectiveness o f peri- discharge interventions vs any com parator on th e risk of early (ie, w ithin 30 days o f discharge) all-cause or u n p lan n ed read m issions w ith or w ith o u t out-of-hospital deaths. The in te r vention had to focus its efforts on th e hospital-to-hom e tra n sition, perm it patients across arm s to have otherw ise similar in p atien t experiences, and be generalizable to contexts b e yond a single p atien t diagnosis. Adult patients had to be a d m itted from th e com m unity to an inpatient ward for at least 24 hours w ith a medical or surgical cause. Studies including o b ste tric or p sy ch iatric ad m issio n s or only in clu d in g d is charges to skilled n u rsin g or reh a b ilita tio n facilities w ere excluded.
Info rm ation Sources In co lla b o ra tio n w ith an e x p e rie n c e d re s e a rc h lib ra ria n (P.J.E.), w e se a rc h e d in April o f 2013 th e follow ing d a ta b a s e s : P u bM ed, Ovid MEDLINE, Ovid EMBASE, EBSCO CINAHL, a n d S co p u s. T h e c o m p le te s e a rc h s tr a te g y is re p o rte d in eA ppendix 1 (in S u p plem ent). Two review ers (T.S. a n d A .L.L.) h a n d - s e a r c h e d th e b ib lio g ra p h ie s o f in c lu d ed stu d ie s an d rec en t review s. E xperts in th e field w ere asked to identify additional references.
Study Selection F our rev ie w ers (A.L.L., M.R.G., J.P.B., a n d T.S.) w o rk ed in d e p en d e n tly an d co n sid ered th e eligibility o f c an d id ate articles by exam ining th eir titles and abstracts, and th en th e full version o f articles identified as potentially eligible by at least 1 reviewer. Conflicts about th e eligibility of full articles w ere resolved by discussion an d consensus. Eligibility was delayed for studies reporting outcom es incompletely, p e n d ing au th o r contact.
Data Collection After creating and piloting a standardized form, the review ers (A.L.L., M.R.G., and J.P.B.), working independently and in duplicate an d using a w eb-based program (DistillerSR), ab stracted details about th e p atien t population, th e in te rv en tions com pared, and th e outcom es reported.
We abstracted details o f th e interventions tested verba tim from either th e trial report or a cited protocol, lim iting our focus to th e period o f hospitalization until 30 days after dis charge, and identifying th e “n et intervention” by selecting out activities th a t occurred in th e intervention arm b u t n o t in th e control arm . These activities w ere coded using a taxonom y adapted from H ansen e t al11 (Table 1). We also noted th e n u m ber o f m eaningfully involved individuals participating in th e
1 0 9 6 J A M A I n t e r n a l M e d ic in e J u ly 2 0 1 4 V o lu m e 174, N u m b e r 7 ja m a in t e r n a lm e d ic in e .c o m
Preventing 30-Day Hospital Readmissions Original Investigation Research
intervention’s delivery and the number of meaningful inter actions these individuals had with patients. Meaningfully in volved individuals played a structured and requisite function in the delivery of central aspects of the intervention (eg, a phy sician who might be contacted only as needed would not be considered meaningfully involved). Similarly, meaningful pa tient interactions were defined as those that were the pro posed sources of the intervention’s effectiveness (eg, a nurse visiting a patient only to deliver educational materials but not to actually engage in educational activity would not be con sidered a meaningful interaction). Two team members (A.L.L. and M.R.G.) created summary descriptions of the interven tions in a standardized format; these were shared with each author to confirm their fidelity to what happened in the trial.
After calibrating judgments on a pilot sample, 2 raters fa miliar with the CuCoM (F.S.M. and K.G.), not involved in data collection and blinded to trial results, evaluated each stan dardized intervention description on a scale of l (substan tially decrease) to 4 (no effect) to 7 (substantially increase) to reflect the degree to which the intervention was likely to affect patient workload and patient capacity for self-care. The im pact on patient capacity was rated with perfect agreement 50% of the time and within l point of difference in 42% of cases (8% differed by 2 points). Because no interventions were rated to decrease patient capacity and all mean ratings fell within the range of 4.0 to 5.5, we elected to dichotomize the variable (threshold of >5 for increasing capacity) for analysis. Work load was more difficult to assess reliably: perfect agreement and minor disagreement (±1 point) were seen in 29% and 44% of cases, respectively, with 27% of cases differing by 2 or more points. This variable was divided into 3 categories (increase, decrease, no change).
For each included trial, we extracted or computed the risk of early readmission for each arm, analyzing patients as ran domized (intention to treat analysis). We used the number ran domized as the denominator except when the number of pa tients discharged was reported and differed from the number randomized. We selected the outcome to extract on the basis of an ad hoc hierarchy of outcomes of interest, with priority given to unplanned readmissions, then to all-cause readmis sions, and finally to the composite end points of unplanned and all-cause readmissions plus out-of-hospital deaths, re spectively. Outcomes were extracted and analyzed at the lon gest period of follow-up, up to 30 days from discharge. Exami nation of trials reporting the effect of interventions on more than 1 of these outcomes revealed that treatment effects were consistent across them (data not shown).
Risk o f Bias Two raters (A.L.L. and M.K.) worked independently and in du plicate to determine the extent to which each trial was at risk of bias using a standardized form based on the Cochrane Col laboration’s tool.12 The assessment considered the quality of the randomization sequence generation, allocation conceal ment, blinding of outcome assessors, the potential for miss ing outcomes (ie, likelihood of missing readmissions to other hospitals), and the proportion of patients lost to follow-up. For missing outcomes, “high risk of bias” was assigned when the
Table 1. A c tiv ity -B a s e d C odin g F ra m e w o rk f o r D ischarge In te rv e n tio n s
Label A ctivity Observed
Discharge planning
Simply thinking about and formalizing an approach to prepare for discharge when this did not occur in any way in the control arm
Case management
Logistical coordination of care and/or resources not specifically focused on self-management and either not occurring in control arm or occurring to lesser degree
Telephone follow-up
Use of a telephone or videophone for provider-initiated communication after discharge that does not occur in the control arm
Telemonitoring Use o f remote technology designed fo r the patient to transm it objective measures of health status w ith or w ithout connected subjective assessment
Patient education
Patient-directed education related to diagnosis or treatm ent rationale but not focused on encouraging self management and not occurring in control arm
Self-management Patient-directed education or coaching directly focused on improving patient's ability to self-manage care needs that does not happen in control arm
Medication intervention
Medication reconciliation or special education aimed at improving medication understanding or adherence; often conducted by a pharmacist but need not be
Home visits Physical visitation by intervention provider to patient's place of residence when this does not happen in control arm
Follow-up scheduled
Scheduling of a follow -up visit prior to discharge when this is not done in the control arm or is done less reliably
Patient-centered discharge instructions
Some difference in the form at or usability of discharge materials to make them more accessible or relevant compared w ith control
Clinician continuity
Increased provider presence on both sides o f the hospital- to-hom e transition compared w ith control; may include involvement o f PCP in inpatient care or strategic follow -up w ith inpatient clinician after discharge or "bridging" clinician
Timely follow-up Postdischarge follow -up visit or communication with patient when this either does not occur or occurs at a later date in the control arm
Timely PCP communication
Engagement with PCP in communication about patient status when this either does not occur or occurs at a later date in the control arm
Patient hotline Presence of an open line fo r patient-initiated communication when this either does not exist in the control arm or is more restricted in availability or usefulness
Rehabilitation intervention
Patient-directed rehabilitation efforts that are not entirely diagnosis specific but aimed at improving functional status and do not exist in the control arm
Streamlining A general streamlining o f services provided, often with dedicated assignment of responsibility, when this does not occur in the control arm
Making requisite Increasing the use or quality of services currently available but underutilized compared w ith the situation in the control arm
Other Special situations unique to the intervention (eg, caregiver education, peer mentoring)
Abbreviation: PCP, prim ary care provider.
readmissions data came from internal health system records only. To assess for publication bias, we examined a funnel plot for asymmetry and conducted asymmetry regression accord ing to Sterne and Egger13 and determined the associated P value.
Data Synthesis We used random-effects meta-analyses to estimate pooled risk ratios and 95% confidence intervals for early readmission.14,15 We tested for heterogeneity of effect on this outcome using the Cochran Q x2 te st16 and estim ated between-trial inconsis tency not due to chance using the P statistic.17
jamainternalmedicine.com JAM A In te rn a l M edicine July 2014 Volume 174, Number 7 1097
R e s e a rc h O r ig in a l I n v e s t ig a t io n P r e v e n tin g 3 0 - D a y H o s p ita l R e a d m is s io n s
To explore th e effects o f p atien t, intervention, and o u t come characteristics on th e im pact o f m easured intervention effectiveness, we conducted planned subgroup analyses, te st ing variables 1 at a tim e.
Patient characteristics tested were age (m ean >65 years or not), diagnosis (h eart failure or o th er), an d h o sp ital w ard (general medical or other). Intervention characteristics tested included th e num ber o f u n iq u e activities involved in th e in tervention, th e n um ber o f unique individuals or roles m ean ingfully involved in its delivery, the minimum num ber of m ean in g fu l p a tie n t in te ra c tio n s o ccu rrin g w ith in 30 days, th e location o f th e intervention activity (ie, w h eth er it occurred entirely during th e inpatient stay, after discharge, or as a com bination th a t “bridged” th e transition), w h eth er th e interven tion was rated to increase or decrease p atien t w orkload, and w h eth er th e intervention w as rated to increase p atien t capac ity (no intervention was found th a t decreased patient capac ity for self-care). Ad hoc variables tested were year o f publi c a tio n a n d ty p e o f o u tc o m e r e p o r te d (ie , u n p la n n e d readm issions vs other).
In form ed by th e findings o f th e ex p lo rato ry subgroup analyses and our initial hypotheses, we constructed a post hoc m etaregression m odel to te st a variable th a t reflected th e d e gree to w hich discharge interventions provided com prehen sive p atien t and caregiver support. This “com prehensive su p p o rt” variable could re tu rn values w ith in a range o f 0 to 4 “p oints” on th e basis o f w h eth er th e intervention (1) was rated to increase p atien t capacity, (2) had at least 5 (75th percentile o f distribution) unique intervention activities, (3) had at least 5 (75th percen tile o f d istrib u tio n ) m eaningful p atien t co n tacts, and (4) h ad at least 2 (75th percentile o f distribution) in dividuals involved in its delivery. We created 3 categories for this variable: interventions w ith zero points (category 1), in terventions w ith 1 or 2 points (category 2), and interventions w ith 3 or 4 points (category 3). To control for changes in s ta n dard care delivery over tim e, we adjusted on th e basis o f the year o f publication variable.
Results
Study Selection Our initial database search generated 1128 reports (eFigure 1 in Supplement). Through abstract and title screening, 256 re p o rts w ere id e n tified for fu ll-tex t review. D uring fu ll-tex t screening (agreement, 89%), 24 were selected for inclusion and 39 were set aside for au th o r contact prior to making a d eci sion. Of 7 potentially eligible studies identified from bibliog raphies and expert consultation, 2 were included and 1 was set aside for au th o r contact. Of th e 40 trials requiring au th o r con ta ct for a final eligibility decision, 21 were deem ed eligible. Of th e 48 apparently eligible trials, 1 was found ineligible after the au th o r confirm ed th a t readm ission data were collected only for readm issions related to th e index diagnosis.18 The final sam ple therefore com prised 47 trials from 46 reports.19"64
Of th e 47 eligible trials, 42 con trib u ted data for th e pri m ary meta-analysis, and 5 (those th a t reported num bers o f re adm issions rath er th a n th e n um ber o f p atien ts readm itted)
w ere analyzed separately.31,45,50,55,61 A com plete list o f ex cluded full-text studies w ith rationale for exclusion is avail able in eAppendix 2 (in Supplement).
Study Characteristics Table 2 d es crib es th e in c lu d e d tria ls . M any w ere sin g le center trials taking place in academ ic m edical centers, enroll ing few patients (eg, 22 trials enrolled <200 patients), and re porting 30-day readm issions. Most interventions tested took place in b oth the inpatient and o u tp atien t settings. The coded activity analysis is reported in eTable 1 (in Supplement). In gen eral, interventions included anyw here from 1 to 7 unique ac tivities. Case m anagement, patient education, hom e visits, and self-m anagem ent support were com m only present in n et ac tivity descriptions (eTable 1 in Supplement). Trial authors re sponded to confirm ation requests for 34 of th e 47 n et in ter v e n tio n d e s c r ip tio n s . T h re e a u th o r s r e q u e s te d m in o r modifications and 1 au th o r m ade major modifications to these descriptions.
Most studies were at low risk o f bias (eTable 2 and eFig ure 2 in Supplement). The m ost com m on methodological limi tation of these trials was th e lack o f a reliable m ethod for deal ing w ith missing data.
Meta-analysis In th e 42 trials reporting readm ission rates, th e overall pooled relative risk (RR) o f readm ission w ithin 30 days w as O.82 (95% Cl, 0.73-0.91; P < .001) (Figure 2). Inconsistency across trials was low (P = 31%). Funnel plot exam ination show ed asym m etry suggestive o f publication bias in th e context o f smaller studies (eFigure 3 in Supplem ent), and the Egger test was sig nificant (P = .02). The 5 trials reporting n u m ber o f readm is sions (rather th an num ber o f patients w ith readm issions) had a pooled relative risk o f readm ission o f 0.93 (95% Cl, 0.72- 1.20; P = 23% ; P = .59). Although this result was consistent with th e risk found in trials reporting readm ission rates (interac tio n P = .38), we opted n o t to include th ese trials in subgroup analyses.
Subgroup analyses failed to find an interaction betw een trial results and p atien t characteristics or outcom e m easured (Table 3). A num ber o f intervention characteristics, however, d id in te ra c t w ith m e a su re d effec tiv e n ess. T h ese in c lu d e w hether th e intervention was rated to augm ent patient capac ity for self-care (RR, 0.68 [95% Cl, 0.53-0.86] w hen it was and RR, 0.88 [95% Cl, O.8O-O.97] w h en it w as n o t; in te ractio n P = .04), w h eth er th e intervention had at least 5 unique, com p o n en t activities (RR, 0.63 [95% Cl, 0.53-0.76] w h en it did and RR, 0.91 [95% Cl, 0.81-1.01] w h e n it d id n o t; in te ra c tio n P = .001), an d w h eth er th e in terv en tio n had at least 2 in d i viduals involved in delivery (RR, 0.69 [95% Cl, O.57-O.84] w hen it did and RR, 0.87 [95% Cl, 0.77-0.98] w hen it did not; in ter action P = .05). Studies testin g in terventions m ore recently were associated w ith reduced effectiveness (RR, 0.89 [95% Cl, O.81-O.97] w hen published in 2002 or later and RR, 0.56 [95% Cl, 0.40-0.79] w h en p u b lis h e d p rio r to 2002; in te ra c tio n P = .01). Other characteristics of th e interventions, such as their rated effect on patient w orkload and th e site o f delivery, had no significant interaction w ith th e intervention effect.
1098 JAMA Internal Medicine J u ly 2 0 1 4 V o lu m e 174, N u m b e r 7 ja m a in t e r n a lm e d ic in e .c o m
Preventing 30-Day Hospital Readmissions Original Investigation Research
Table 2. Study Characteristics3
Source Setting Population Added Intervention Program
Baseline Control Activity
Outcome Time Reported
Patients Discharged,
No.
Activities6/ People'/
Interactions8 Included in
Intervention, No. Capacity' Workload' Location9
Melton e ta l,19 2012
48 US states
Commercially insured for 3 acute DRGs
Risk-prioritized telephone follow-up
Nonprioritized telephone follow-up
UR, 1 mo 3988 1/1/1 4.0 3.5 OP
Marusicetal,20 2013
Academic hospital in Croatia
Elderly patients receiving >2 medications for chronic disease
Specialized pharmacotherapeutic counseling
Standard discharge planning and usual information about drugs from physician
UR, 1 mo 160 2/1/1 4.0 3.5 IP
Altfeldetal,21 2012
US Elderly patients receiving >7 medications with psychosocial need
Targeted telephone follow-up program
Standard discharge planning without any follow-up contact
ACR, 1 mo 906 2/1/1 5.0 2.0 OP
Davis eta l,22 2012
Academic center, US
Patients with HF with mild cognitive impairment
Self-management focused education program
Standard discharge HF teaching and booklet
ACR, 1 mo 125 5/1/2 5.0 3.5 Both
Bowles et al,23 2011
Urban community, US
Mostly black patients with HF under specific home care agency
Telehomecare substitution of traditional home care
Clinical pathway for HF and home care
ACR, 1 mo 218 1/2/5 4.0 5.5 OP
Finn etal,24 201 l h
Academic center, US
General medical patients of academic center; approximately 25% discharged toSNF
Embedded nurse practitioner into academic team to improve discharge process
Standard discharge planning with follow-up scheduled by resident
ACR, 1 mo 646 2/1/1 4.0 2.5 IP
Wong e ta l,25 2011
Large general hospital, Hong Kong
Elderly general medical population
Use of nurse case managers and trained volunteers to improve transition through health-social partnership
Usual discharge planning for follow-up and support
ACR, 1 mo 686 4/3/5 4.5 3.5 Both
Leventhal e ta l,26 2011
University hospital, Switzerland
Elderly patients with HF
Outpatient, interdisciplinary education and support program
Standard care with HF consultation, education booklet
ACR, 1 mo 34 5/1/4 4.5 4.0 OP
Rytter et al,27 2010
Single center, Denmark
Elderly patients from medical or geriatric ward
Use of mandatory home visits to improve follow-up from PCP and district nurses
Standard discharge procedures and letters
ACR, 1 mo 331 4/2/2 4.0 2.5 OP
Koehler etal,28 2009
Academic center, US
Elderly general medical patients expected to return to home or assisted living
Supplemental care bundle that shifted responsibilities from nurses to care coordinators and added follow-up
Nursing staff do medication reconciliation, discharge medication teaching, and education; pharmacist review
UR, 1 mo 41 6/2/5 4.5 4.0 Both
Braun,29 2009 Medical center, Israel
General medical patients
Use of tight telephone Standard follow-up, especially discharge to improve adherence planning and
summary
ACR, 1 mo 400 1/1/2 4.5 3.0 OP
Courtney eta l,30 2009
Tertiary center, Australia
Elderly general medical patients at high risk
Individualized, exercise-based care plan for elderly
Routine discharge and rehabilitation advice, planning
UR, 1 mo 128 6/2/5 5.0 5.0 Both
Jack e ta l,31 2009
Academic, urban, safety net center, US
General medical patients; 51% black
Standardized discharge package to minimize failures using discharge planners and pharmacists
Similar to intervention but uncoordinated
ACRE, 1 mo 738 6/2/3 5.0 2.5 Both
(continued)
jamainternalmedidne.com JAMA Internal Medicine July 2014 Volume 174, Number 7 1099
Research Original Investigation Preventing 30-Day Hospital Readmissions
Table 2. Study Characteristics3 (continued)
Activities3/ People1/
Interactions3
Source Setting Population Added Intervention Program
Baseline Control Activity
Outcome Time Reported
Patients Discharged,
No.
Included in Intervention,
No. Capacity' Workloadf Location9
Wakefield e ta l,32 2008
VA medical center, Iowa, US
Men w ith HF; mean age, 69 y
Telehealth-facilitated Usual discharge postdischarge support teaching and program follow -up
ACR, 1 mo 148 2 /1 /5 5.0 5.0 OP
Balaban e ta l,33 2008
Small Culturally and community linguistically teaching diverse general hospital, US medical or
surgical patients
Program to promptly reconnect patients to medical home through discharge form
Standard discharge planning and instruction
ACR, 1 mo 96 3 /2 /2 5.0 2.5 Both
Wong e ta l,34 2 0 08
3 Regional hospitals, Hong Kong
Elderly patients readmitted to department of medicine
Preventive, postdischarge home visits fo r high-risk patients
Standard discharge planning and instruction
UR, 1 mo 354 2 /1 /2 4.0 4.0 Both
Coleman e ta l,35 2006
Single center, US
Elderly medical patients who were in capitated delivery system; approximately 20% discharged to SNF
Use o f transition coaches and personal health record to equip patients and caregivers to be more active in care
Not well reported
UR, 1 mo 750 5 /1 /5 4.5 5.0 Both
Linne et al,3e 2006
Multiple community hospitals, Sweden
Patients w ith HF; discharge disposition not reported
Use computer-based education session in discharge process
Standard HF education and materials
ACR, 1 mo 230 2 /0 /0 4.0 4.5 Both
Casas et al,37 2006
2 Tertiary centers, 1 in Spain and l i n Belgium
Elderly patients w ith COPD
Integrated care plan Standard to generate synergy discharge and avoid redundancy protocol w ithout between inpatient and support o f nurse outpatient care teams or call center fo r patients with COPD
ACR, 1 mo 155 7 /3 /5 5.0 4.5 Both
Riegel et al,38 2006
2 Mexican Community Americans with hospitals, HF who were southern old, ill, and California, poorly US acculturated
Telephone case management program to improve discharge transition in Mexican Americans
Nonstandardized, ACR, 1 mo HF education often in English
135 5 /1 /1 5.5 3.0 OP
Koelling e ta l,39 2 005
University Selected hospital, US patients w ith HF;
mean age, 65 y
Single predischarge education session
Standard discharge information and education, booklet
ACR, 1 mo 223 2 /1 /1 4.5 4.5 IP
Mejhert e ta l,40 2 0 04
University hospital, Sweden
Elderly patients w ith HF
Nurse-driven, protocol-based outpatient management program
Standard discharge care; usual follow -up
ACR+D, 1 mo
196 3 /2 /1 5.0 4.5 OP
Kwok et al,41 2004
2 Acute hospitals, Hong Kong
Elderly patients w ith chronic lung disease at high risk
Community nurse-supported program based on weekly home visits
Standard follow -up with home visits as needed
UR, lm o 157 4 /1 /5 5.5 4.5 Both
Doughty et a l 42 2002"
Single Patients w ith HF; center, New dispositions not Zealand reported
Outpatient, integrated management program for HF
Usual care under PCP
ACR, 1 mo 197 6 /3 /1 4.5 5.0 OP
Jaarsma e ta l,43 1999
University Elderly patients Hospital, w ith HF; mean Netherlands age, 73 y
Nurse-led education and support program w ith follow -up home visit
Usual care; no structured education, follow -up call, or home visit
ACR, 1 mo 179 3 /1 /3 4.0 3.5 Both
Naylor et al,44 1999
2 Urban academic hospitals, US
Elderly medical and surgical patients; 45% black
Advanced practice nurse-directed program that stressed continuity, home and telephone follow -up
Routine discharge planning and home care
ACR, 1 mo 363 5 /1 /4 5.5 2.5 Both
Stewart et al,45 1998
Tertiary referral center, Australia
General medical and surgical patients; 83% considered high
Risk-targeted, home- based intervention by nurse and pharmacist
Discharge planning with follow -up w ithin 2 wk
UR+DE, 1 mo
762 5 /2 /3 5.0 2.5 Both
risk
(continued)
1100 JAMA Internal Medicine July 2014 Volume 174, Number 7 jamainternalmedicine.com
Preventing 30-Day Hospital Readmissions Original Investigation Research
Table 2. Study Characteristics3 (continued)
Activities11/ People11/
Interactions11 Outcome Patients Included in
Added Intervention Baseline Control Time Discharged, Intervention, Source Setting Population Program Activity Reported No. No. Capacity' Workload1 Location9 Dunn et al,46 1995
Geriatric hospital, England
Geriatric ward patients; mean age, 83 y
Single home visit from Usual discharge public health nurse process
ACR, 1 mo 204 2 /1 /1 4.5 3.0 OP
Rich et al,47 1995
Single academic center, US
High-risk, elderly patients w ith HF
Nurse-directed, m ultidisciplinary intervention w ith home visit follow -up
Conventional care by PCP
ACR, 1 mo 274 6 /4 /5 5.5 2.0 Both
Naylor et al,48 1994
Single university hospital, US
Elderly patients w ith or w ithout caregiver for medical cardiac diagnosis
Individualized, comprehensive program directed by clinical nurse specialists, including home follow -up
Robust but not individualized routine discharge plan
ACR, 2 wk 142 5 /1 /4 4.5 2.0 Both
Naylor et at,48 1994
Single university hospital, US
Elderly patients w ith or w ithout caregiver for surgical cardiac diagnosis
Individualized, comprehensive program directed by clinical nurse specialists, including home follow -up
Robust but not individualized routine discharge plan
ACR, 2 wk 134 5 /1 /4 4.5 2.0 Both
Naylor et al,49 1990
Urban medical center, US
Elderly general medical or surgical patients
Comprehensive, individualized discharge planning protocol w ith home follow -up directed by nurse specialists
Nurse-directed routine discharge planning
ACR, 2 wk 40 4 /1 /4 4.0 3.0 Both
Kulshreshtha e ta l,50 2010
Urban Patients w ith HF; teaching could enter hospital, US study up to 2
weeks after discharge
Remote monitoring Not well follow -up program fo r described ambulatory patients
ACRE, 1 mo 150 2 /1 /5 5.5 5.5 OP
Graumlich e ta l,512009h
Tertiary General medical teaching patients at high hospital, US risk of
readmission
Discharge software to improve communication and address deficiencies
Usual care with handwritten discharge forms
ACR, 1 mo 631 3 /1 /0 4.0 3.0 IP
Atienza et al,52 2004
3 Tertiary university hospitals, Spain
Patients w ith HF; mean age, 68 y
Hospital discharge and outpatient disease management program
Variable and nonstructured; PCP follow -up
ACR, 1 mo 338 5 /2 /2 4.5 4.0 Both
R ie g e le ta l,53 2004
2 Hospitals in suburban Southwest, US
Patients w ith HF in integrated health system; mean age, 73 y
Use of peer mentors to improve self-care in recently discharged patients
Inpatient HF education; support groups available
ACR, 1 mo 88 1 /1 /2 4.5 3.0 OP
Stowasser e ta l,54 2 0 02
2 Large hospitals, Australia
General medical and surgical patients
Medication liaison service to improve communication of medication-related issues through discharge process
Routine care and pharmacist medication review, discharge planning
UR, 1 mo 240 3 /1 /0 4.0 3.0 IP
Li e ta l,55 2012 Academic Elderly patients center, New and their fam ily York, US caregivers
Training of fam ily caregivers to prepare fo r anticipated postdischarge role
Routine care w ith practical information given to caregivers
ACRE, 2 wk 407 1 /0 /0 4.5 3.5 IP
Shyu e ta l,56 2005
Large, single center, Taiwan
Elderly patients w ith hip fracture
Interdisciplinary program of geriatric consultation, rehab, and discharge planning service
Routine care and inpatient physical therapy w ithout home visits
ACR, 1 mo 137 4 /3 /5 5.5 3.5 Both
Angermann e ta l,57 2012
9 Centers, Germany
Patients w ith HF; mean age, 69 y
Nurse-coordinated disease management program that emphasized a "call and care center"
Standard discharge planning and follow -up
ACR, 1 mo 715 4 /1 /5 4.5 4.5 Both
Naylor e ta l,58 2004
5 Academic and
Elderly patients with HF; 36%
Advanced practice nurse-directed care
Standard, site- specific HF
ACR+D, 1 mo
239 6 /1 /5 5.0 3.0 Both
community black program with discharge hospitals, emphasis on comorbid planning and US and chronic condition follow -up
management; included home follow -up
(continued)
jamainternalmedicine.com JAMA Internal Medicine July 2014 Volume 174, Number 7 1101
R e s e a rc h O r ig in a l I n v e s t ig a t io n P r e v e n tin g 3 0 - D a y H o s p ita l R e a d m is s io n s
T a b le 2 . S t u d y C h a r a c t e r i s t i c s 3 ( c o n t i n u e d )
S o u rce S e ttin g P o p u la tio n A d d e d I n te r v e n tio n P ro g ra m
B a s e lin e C o n tr o l A c t i v i t y
O u tc o m e T im e R e p o rte d
P a tie n ts D is c h a rg e d ,
N o.
A c t iv it ie s 6/ P e o p le 1/
In te r a c tio n s 6 In c lu d e d in
I n te r v e n tio n , N o. C a p a c ity " W o r k lo a d ' L o c a tio n 9
S tro m b e rg e t a l , 59 2 0 0 3
1 U n iv e r s ity E ld e rly p a tie n ts a n d 2 w it h HF c o u n ty h o s p ita ls , S w e d e n
R e q u is ite f o l lo w - u p in s p e c ia liz e d , p r o t o c o l- d r iv e n , n u r s e -le d HF c lin ic
C o n v e n tio n a l p r im a r y ca re f o llo w - u p
ACR+D, 1 m o
1 0 6 5 / 1 / 1 5 .0 3 .5 OP
H ansen e t a l , 60 1 9 9 5
U n iv e r s ity H ig h ly s e le c te d H o m e v is it f o llo w - u p h o s p ita l, p a tie n ts f r o m p ro g ra m f o r h ig h ly D e n m a rk s u b a c u te g e r ia tr ic ta r g e te d e ld e r ly
w a rd n e e d in g p o p u la tio n h o m e r e h a b ilita tio n and m e d ic a l a nd s o c ia l s u p p o r t
D is c h a rg e s u m m a ry s e n t a n d s ta n d a rd s u p p o r t a rra n g e d
ACR, 1 m o 1 9 3 4 / 2 / 2 5 .0 3 .5 OP
M a slo ve e t a t,61 2 0 0 9 h
S in g le G e n e ra l m e d ic a l a c a d e m ic p a tie n ts ; c e n te r, a p p r o x im a te ly Canada 8 0 % d is c h a rg e d
h o m e
D e v e lo p m e n t o f m o re u s e fu l a nd s ta n d a rd iz e d d is c h a rg e s u m m a ry
S ta n d a rd , a tte n d in g p h y s ic ia n - g e n e ra te d d is c h a rg e s u m m a rie s a nd p la n n in g
ACRE, 1 m o 2 0 9 2 / 1 / 0 4 .0 3 .5 IP
F o rs te r e t a l , 62 2 0 0 5
2 C am p u se s G e n e ra l m e d ic a l o f a p a tie n ts ; m e an te a c h in g a g e , 6 6 y h o s p ita l, Canada
In te g r a tio n o f d e d ic a te d c lin ic a l n u rs e s p e c ia lis t in to c a re te a m t o f a c ilit a t e d is c h a rg e p la n n in g p roce ss
R e g u la r d is c h a rg e c a re p la n n in g
ACR+D, 1 m o
3 6 1 3 / 1 / 1 4 .0 3 .0 B o th
D udas e t a l,63 2 0 0 1
S in g le G e n e ra l m e d ic a l a c a d e m ic s e rv ic e p a tie n ts c e n te r, US
P h a rm a c y s e rv ic e f o l lo w - u p c a ll
R e g u la r p h a rm a c y - f a c ilita te d d is c h a rg e p roce ss
ACR, 1 m o 2 2 1 2 / 1 / 1 4 .0 3 .0 OP
P a rry e t a l , 64 2 0 0 9
2 F e e - fo r - s e r v ic e Use o f t r a n s it io n C o m m u n ity M e d ic a re p a tie n ts co a c h e s a nd p e rs o n a l h o s p ita ls , in s in g le h e a lth h e a lth r e c o rd t o e q u ip US s y s te m ; in c lu s io n p a tie n ts and
c r ite r ia d e s ire d t o c a re g iv e rs t o a s s e rt c a tc h p a tie n ts m o re a c tiv e r o le in d is c h a r g in g t o c a re tr a n s it io n S N F - d id n o t r e p o r t d is p o s itio n s
N o t w e ll re p o r te d
UR, 1 m o 9 8 5 / 1 / 5 5 .0 2 .5 B o th
A b b r e v ia tio n s : AC R , a ll-c a u s e r e a d m is s io n r a te ; A C R +D , a ll-c a u s e r e a d m is s io n c N u m b e r o f in d iv id u a ls m e a n in g f u lly in v o lv e d in d e liv e r y o f t h e in t e r v e n tio n ,
a n d o u t - o f- h o s p ita l d e a t h ra te ; A C R E , a ll-c a u s e r e a d m is s io n e v e n t c o u n t; B o th , a |y|jn j m u m n u m b e r o f m e a n in g f u l h u m a n in t e r a c t io n s in i n t e r v e n t io n d e liv e ry . a c t i v it y o c c u r r e d in b o t h in p a t ie n t a n d o u t p a t ie n t e n v ir o n m e n ts ; C O P D , c h r o n ic „ . , ,,
. . , ,. ,. . . . . . . . . e R a te d lik e lih o o d o f i n t e r v e n t io n t o a f f e c t p a t ie n t c a p a c ity f o r s e lf- c a r e o n a o b s t r u c t iv e p u lm o n a r y d is e a s e ; DRG , d ia g n o s is - r e la te d g r o u p ; HF, h e a r t f a ilu r e ; , , , , . „ . , .
,, .. .. . . . . . . . . „ .. .. ,. s c a l e o f l ( s u b s t a n t i a l ly d e c r e a s e ) t o 4 ( n o e f f e c t ) t o 7 ( s u b s t a n t i a l l y in c r e a s e ) . IP, a ll a c t i v it y o c c u r r e d in i n p a t ie n t e n v ir o n m e n t ; OP, a ll a c t iv it y o c c u r r e d in o u t p a t ie n t e n v ir o n m e n t ; SNF, s k ille d n u r s in g f a c ility ; U R, u n p la n n e d f R a te d lik e lih o o d o f in t e r v e n t io n t o im p o s e w o r k o r b u r d e n o n p a t ie n t o n a
r e a d m is s io n ra te ; U R + D E , u n p la n n e d r e a d m is s io n a n d o u t - o f- h o s p ita l d e a th s s c a le o f 1 ( s u b s t a n t ia lly d e c re a s e ) t o 4 ( n o e f f e c t ) t o 7 ( s u b s t a n t ia lly in c re a s e ),
e v e n t c o u n t ; US, U n ite d S ta te s ; V A . V e te r a n s A ffa ir s . s S e ttin g ( in p a tie n t , o u t p a t ie n t , o r b o t h ) w h e r e in t e r v e n t io n a c t i v it y o c c u r r e d .
3 I n t e r v e n t io n a n d b a s e lin e /c o n tr o l a c tiv it ie s w e r e s y s te m a tic a lly c o d e d in h C lu s te r - r a n d o m iz e d s tu d y .
g r e a t e r d e ta il th a n c a n b e e x p r e s s e d in t h is t a b le (s e e e T a b le 1 in S u p p le m e n t) .
b N u m b e r o f a c t iv it ie s in t h e i n t e r v e n t io n as e v a lu a te d b y c o d in g s tr a te g y f r o m
T a b l e !
P o s t H o c M e t a r e g r e s s io n A n a ly s is
D e sp ite p o te n tia l c o lin e a rity o f th e c o n tr ib u tin g v a ria b le s , m e ta re g re ssio n sh o w e d a sig n ifican t a n d in c re m e n ta l effect o f “c o m p r e h e n s i v e s u p p o r t ” o n r e d u c i n g r e a d m is s io n s (Table 4). C ategory 3 com prised 7 in terv en tio n s.28,3037'47'56,58,64 C om pared w ith cate g o ry 1 in te rv e n tio n s, th e s e w ere a sso c i a te d w ith a relativ e risk o f re a d m iss io n o f 0.63 (95% Cl, 0.43- 0 .91; P = .02). C ategory 3 in te rv e n tio n s u s e d a c o n siste n t an d c o m p le x s tra te g y th a t e m p h a s iz e d th e a s s e s s m e n t a n d a d d re ssin g o f facto rs re la te d to p a tie n t c o n te x t a n d ca p a c ity for self-care (including th e im p act o f com orbidities, fu nctional s ta tu s , caregiver capabilities, socioeconom ic factors, p o te n tia l for
self-m a n a g e m e n t, a n d p a tie n t a n d caregiver goals for care). T h ese in te rv e n tio n s co o rd in a te d care across th e in p a tie n t-to - o u tp a tie n t tr a n s itio n a n d in v o lv ed m u ltip le p a tie n t in te ra c tio n s; all b u t l 28 inv o lv ed p a tie n t h o m e visits.
Discussion
O u r F in d in g s
T he b o d y o f ra n d o m iz e d trial ev id en c e show s a c o n siste n t and b en eficial effect o f te s te d in te rv e n tio n s o n th e risk o f 30-day read m issio n s. E x p lo rato ry su b g ro u p a n aly ses su g g est th a t ef-
1 1 0 2 J A M A I n t e r n a l M e d ic in e J u ly 2 0 1 4 V o lu m e 174, N u m b e r 7 ja m a in t e r n a lm e d ic in e .c o m
P r e v e n t in g 3 0 - D a y H o s p ita l R e a d m is s io n s O r ig in a l I n v e s t ig a t io n Research
Figure 2. Results o f Prim ary Meta-analysis
Study RR (95% Cl) Naylor e ta l,49 1990 0.33 (0 .0 4 -2 .9 4 )
Naylor e ta l,48 1994 0 .2 7 (0 .0 8 -0 .9 1 ) Naylor et a l 48 1994 0.69 (0 .2 3 -2 .0 8 )
Dunn et al,46 1995 1.13 (0 .6 1 -2 .08 ) Rich et a l,47 1995 0.69 (0 .4 2 -1 .1 3 )
Hansen e ta l,60 1995 0 .3 0 (0 .1 6 -0 .5 7 ) Jaarsma e ta l,43 1999 0.89 (0 .4 3 -1 .8 5 ) Naylor e ta l,44 1999 0 .3 5 (0 .1 9 -0 .6 5 ) Dudas e ta l,63 2 001 0.61 (0 .3 6 -1 .0 5 ) Doughty et al,42 2002 1 .1 4 (0 .6 4 -2 .0 4 ) Stowasser et al,54 2002 0 .8 4 (0 .3 7 -1 .9 3 ) Stromberg et a l,59 2003 0 .6 1 (0 .3 5 -1 .0 9 ) M ejhert et al,40 2004 0.91 (0 .4 5 -1 .8 3 ) Kwok et al,41 2004 1 .1 8 (0 .8 0 -1 .7 4 )
Atienza et a l,52 2 0 04 0.57 (0 .3 1 -1 .06 ) Riegel and Carlson,53 2004 1.53 (0 .5 4 -4 .31 ) Naylor et a l,58 2004 0.51 (0 .3 0 -0 .88 ) Koelling et a l,39 2 0 05 0 .7 0 (0 .4 3 -1 .1 5 ) Shyu e ta l,56 2005 0 .6 1 (0 .1 5 -2 .4 5 )
Forster e ta l,62 2 0 05 1.35 (0 .8 4 -2 .16 ) Coleman e ta l,35 2006 0.70 (0 .4 5 -1 .07 )
Linne and Liedholm ,36 2006 1.05 (0 .6 3 -1 .77 ) Casas et a l,37 2006 0 .5 9 (0 .2 4 -1 .4 6 )
Riegel et a l,38 2 0 06 0 .7 9 (0 .3 8 -1 .6 3 ) Balaban e t a l, 33 2008 1 .0 4 (0 .2 8 -3 .9 3 ) Wong et a l,34 2008 0 .9 8 (0 .7 3 -1 .3 1 ) Wakefield e ta l,32 2008 0.66 (0 .3 0 -1 .4 6 ) Koehler et a l,28 2009 0.42 (0 .0 9 -1 .9 2 ) Braun et a l,29 2009 0.92 (0 .4 1 -2 .0 3 ) Courtney e t a l, 30 2009 0 .3 0 (0 .0 9 -1 .0 4 ) Graumlich e ta l,51 2009 0.97 (0 .6 6 -1 .45 ) Parry et a l,64 2 0 09 0.44 (0 .1 5 -1 .3 5 ) R y tte r e ta l,27 2010 0 .7 1 (0 .4 5 -1 .1 3 ) Bowles et a l,23 2011 0.83 (0 .4 6 -1 .49 ) Finn e ta l,24 2011 1 .1 6 (0 .8 5 -1 .6 0 ) Wong e ta l,25 2011 0.78 (0 .5 3 -1 .16 ) Leventhal e ta l,26 2011 0 .5 6 (0 .0 6 -5 .6 3 ) Melton e ta l,19 2012 0 .7 8 (0 .6 2 -0 .9 9 ) A ltfeld et a l,212012 1.05 (0 .7 7 -1 .43 ) Davis et a l,22 2012 1.15 (0 .5 8 -2 .28 ) Angermann et al,57 2012 1.09 (0 .7 1 -1 .66 )
Marusic e ta l,20 2013 1 .2 0 (0 .3 8 -3 .7 7 ) Overall effect (P<.001) 0.82 (0 .7 3 -0 .9 1 )
Heterogeneity: l 2 = 31%; P = .03
Favors Intervention
Favors Control
j
0.1 1.0 RR (95% Cl)
Weight, %
0.24
0.71
0.88
2.30
3.08
2.14
1.75
2.28
2.77
2.49
1.44
2.57
1.88 4.17
2.31
0.97
2.75
3.11
0.57
3.33
3.72
2.93
1.24
1.77
0.62
5.46
1.54
0.48
1.54
0.70
4.09
0.86
3.38
2.47
5.06
4.16
0.22 6.30
5.18
1.95
3.77
0.81
100.00
T T 1 10
S ize o f t h e d a ta m a r k e r c o r r e s p o n d s
t o t h e r e la tiv e w e ig h t a s s ig n e d in t h e
p o o le d a n a ly s is u s in g r a n d o m - e ffe c ts
m o d e ls . RR in d ic a te s r e la tiv e ris k .
fective interventions are more complex and seek to enhance patient capacity to reliably access and enact postdischarge care. In addition, interventions tested more recently are, in gen eral, less efficacious when compared with controls.
Our findings are consistent with the CuCoM in their sug gestion that providing comprehensive and context-sensitive support to patients reduces the risk of early hospital readmis sion; however, we could not identify an effect of rated inter vention workload on this risk.
Limitations and Strengths of This Review Many studies in this review were conducted in single, aca demic centers; this raises questions about applicability. Also,
the scales that we used to evaluate intervention effects on pa tient workload and capacity relied on global judgments (rather than criterion-based judgments) and are original to this work. To our knowledge, no validated scale exists to assess the po tential of an intervention to impose patient workload or treat ment burden and/or affect a patient’s capacity for self-care. Al though our raters were consistent in their assessm ents of interventions’ effect on patient capacity, their judgment of im pact on patient workload was less reliable. Particularly, raters believed that some burdensome interventions could be ben eficial if the patient had the capacity and resources to access and enact the care. Because the experience of treatment bur den is not constant between patients, an ideal analysis of its
ja m a in t e r n a lm e d ic in e .c o m JAMA Internal Medicine J u ly 2 0 1 4 V o lu m e 174, N u m b e r 7 1103
R e s e a rc h O r ig in a l I n v e s t ig a t io n P r e v e n tin g 3 0 - D a y H o s p ita l R e a d m is s io n s
T a b le 3 . S u b g r o u p A n a ly s e s
S tu d y S u b g ro u p C h a r a c te r is tic (N o . o f S tu d ie s in S u b g r o u p ) 3
R e a d m is s io n , R e la tiv e R isk ( 9 5 % C l) P V a lu e f o r In te r a c tio nS u b g ro u p C o m p a ris o n G ro u p
P a tie n t c h a r a c te r is tic s
HF ( 1 6 ) 0 .8 2 ( 0 . 7 0 - 0 . 9 5 ) 0 . 8 0 ( 0 . 6 9 - 0 . 9 3 ) .8 3
A g e > 6 5 y ( 3 6 ) 0 .7 9 ( 0 . 6 9 - 0 . 9 0 ) 0 .9 1 ( 0 . 7 4 - 1 . 1 0 ) .2 4
F ro m g e n e r a l m e d ic a l w a r d ( 1 8 ) 0 . 8 0 ( 0 . 6 7 - 0 . 9 5 ) 0 . 8 3 ( 0 . 7 2 - 0 . 9 5 ) .7 9
I n te r v e n tio n c h a r a c te r is tic s
R a te d t o in c re a s e p a t ie n t c a p a c ity ( 1 6 ) 0 . 6 8 ( 0 . 5 3 - 0 . 8 6 ) 0 . 8 8 ( 0 . 8 0 - 0 . 9 7 ) .0 4
R a te d t o in c re a s e p a t ie n t w o r k lo a d ( 5 ) b 0 .7 7 ( 0 . 5 7 - 1 . 0 3 ) 0 .8 2 ( 0 . 7 1 - 0 . 9 6 ) .6 8
R a te d t o d e c re a s e p a t ie n t w o r k lo a d ( 1 9 ) b 0 .8 1 ( 0 . 6 7 - 0 . 9 8 ) 0 .8 2 ( 0 . 7 1 - 0 . 9 6 ) .9 0 A b b r e v ia tio n : HF, h e a r t fa ilu r e .
D e liv e re d b y 2 o r m o re in d iv id u a ls ( 1 3 ) c 0 . 6 9 ( 0 . 5 7 - 0 . 8 4 ) 0 .8 7 ( 0 . 7 7 - 0 . 9 8 ) .0 5 a C o m p a r e d w it h t h e r e m a in d e r o f
a n a ly z e d s tu d ie s ( ie , 4 2 - N ) u n le s s
o th e r w is e n o te d . in v o lv e d > 5 m e a n in g f u l p a t ie n t in t e r a c t io n s ( 1 3 ) c
0 . 7 7 ( 0 . 6 4 - 0 . 9 2 ) 0 . 8 4 ( 0 . 7 3 - 0 . 9 6 ) .4 3
C o m p ris e d > 5 u n iq u e a c t iv it ie s ( 1 6 ) c 0 .6 3 ( 0 . 5 3 - 0 . 7 6 ) 0 .9 1 ( 0 . 8 1 - 1 . 0 1 ) .0 0 1 b C o m p a r e d w it h " n o c h a n g e ”
( n = 1 8).
c C u t o f fs c h o s e n b e c a u s e t h e y H ad b o th a n in p a t ie n t a n d o u t p a t ie n t c o m p o n e n t ( 2 2 )
0 .7 8 ( 0 . 6 5 - 0 . 9 2 ) 0 . 8 4 ( 0 . 7 4 - 0 . 9 7 ) .4 6
S tu d y p u b lis h e d 2 0 0 2 o r la t e r ( 3 3 ) d 0 .8 9 ( 0 . 8 1 - 0 . 9 7 ) 0 . 5 6 ( 0 . 4 0 - 0 . 7 9 ) .0 1 r e p r e s e n te d t h e 7 5 t h p e r c e n tile o f
O u tc o m e c h a r a c te r is tic s
O u tc o m e m e a s u re d w a s u n p la n n e d r e a d m is s io n s ( 9 )
0 . 8 4 ( 0 . 6 9 - 1 . 0 2 ) 0 . 8 0 ( 0 . 7 0 - 0 . 9 1 ) .7 0 r e p r e s e n te d t h e m i d - p o in t o f s tu d y
e lig ib ilit y f o r t h is r e v ie w .
T a b le 4 . E f f e c t s o f C o m p r e h e n s i v e S u p p o r t in M e t a r e g r e s s i o n A n a ly s i s
S tu d y C h a r a c te r is tic S tu d ie s , N o. R e a d m is s io n , R e la tiv e R isk ( 9 5 % C l)a P V a lu e
C o m p re h e n s iv e s u p p o r t c a te g o r y 5
1 ( 0 p o in ts ) 1 5 1 [R e fe r e n c e ]
2 (1 o r 2 p o in ts ) 2 0 0 .8 2 ( 0 . 6 6 - 1 . 0 2 ) .0 7
3 (3 o r 4 p o in ts ) 7 0 .6 3 ( 0 . 4 3 - 0 . 9 1 ) .0 2
P u b lic a tio n in 2 0 0 2 o r a fte r 3 3 1 .4 7 ( 1 . 1 0 - 1 . 9 6 ) .0 1
a T h is r e p r e s e n ts t h e a d ju s te d e f f e c t o f e a c h c h a r a c te r is tic o n e a r ly r e a d m is s io n t h a t (1) w e r e r a te d t o in c re a s e p a t ie n t c a p a c ity , ( 2 ) h a d > 5 u n iq u e in t e r v e n t io n
in m e ta r e g r e s s io n . a c tiv it ie s , (3 ) h a d s 5 m e a n in g fu l p a t ie n t in t e r a c t io n s , a n d ( 4 ) h a d > 2
b T h e c o m p r e h e n s iv e s u p p o r t v a r ia b le r e tu r n e d 1 p o in t e a c h f o r in t e r v e n tio n s in d iv id u a ls in v o lv e d in its d e liv e ry .
effects would be based on patient-reported assessm ents o f in te rv e n tio n w o rk lo ad . In d e e d , m an y eligible p a tie n ts d e clined en ro llm en t in to som e s tu d ie s , 2 3 , 2 8 ,4 4 , 5 0 ,5 3 o ften b e cause th ey did not w ish to take on th e perceived burden o f the intervention; evaluating th e effect o f intervention-im posed w orkload in such sam ples is o f lim ited applicability. In gen eral, th e se assessm en ts should be regarded as h y p o th e sis generating and th e inferences m ade on th e basis o f subgroup analyses m u st be view ed as tentative (given th e potential for chance findings from testing multiple hypotheses and the pos sibility th a t some variables are correlated). Finally, despite ro b u st efforts to obtain unpublished data, there was evidence o f publication bias. The overall effect o f this on our findings is n o t known.
This review also has m any strengths. First, it provides, to our knowledge, th e largest a n d m ost com prehensive assess m ent o f discharge interventions and their effect on 30-day re adm issions, including 47 random ized trials at low risk o f bias. This is a stronger and less heterogeneous body o f evidence than previously assem bled , 11,65 and it includes u npublished data from 18 trials. Our study used an activity-based coding m ethod designed to en su re appropriate characterization o f each in te rv en tio n and th e n et difference in activity betw een in te r
v en tio n and control arm s. This m e th o d co n trib u tes to th e field an d can be applied to fu tu re assessm en ts o f com plex interventions. To our knowledge, this is also th e first use of th e CuCoM5 to analyze th e im pact o f health care delivery in terv en tio n s on p atien ts as an explanation for th e ir relative efficacy.
C o m p a r is o n W i t h O t h e r S t u d ie s
We identified 31 more random ized clinical trials th a n were accu m u lated in th e m o st recen t review o f discharge in te r ven tio n effects on 30-day readm ission ra te s , 11 an d w e p ro vide th e first m eta-analysis on th is topic. Although previous stu d ies an d review s have suggested th a t “b u n d le d ” in te r v en tio n s are o f g rea ter v a lu e , 11,65 th is m e ta-an a ly sis p ro vides objective support for this claim. In addition, our study ad d s to an d en h an ces th e body o f evidence rela ted to th e im portance o f p atien t contextual factors in affecting health outcom es. 6 5
Im p lic a t io n s f o r P o lic y a n d P r a c t ic e
In this analysis, interventions th a t used a com plex and su p portive strategy to assess and address contextual issues and lim itations in patient capacity w ere m ost effective at reduc-
1 1 0 4 J A M A I n t e r n a l M e d ic in e J u ly 2 0 1 4 V o lu m e 174, N u m b e r 7 ja m a in t e r n a lm e d ic in e .c o m
Preventing 30-Day Hospital Readmissions Original Investigation Research
ing early hospital readm issions. Many o f these contacted th e p atien t frequently, used hom e visits, and reported cost sav ings. This inform ation can be used to guide the design and te st ing o f fu tu re interventions. The CuCoM may also have value in helping to conceptualize th e effects o f health care in terv en tions across diverse patient contexts, b u t we w ere unable to characterize a co n sisten t effect o f rated in terv en tio n w ork load on outcom es. Finally, we found th a t more recently tested interventions were less effective. We hypothesize th at this may rep rese n t (l) a general im provem ent over tim e in th e s ta n dard o f care th a t was no t fully appreciated in control descrip tions, (2) an increased effort over tim e to test sim pler and less comprehensive interventions, (3) a higher likelihood over tim e o f more diverse interventions to m easure and report 30-day readm ission rates (eg, including th o se less focused on red u c ing early readm issions), and/or (4) a general shift away from interventions stressing hum an interaction tow ard those more
high tech in nature. Additional stu d y is need ed to d eterm ine th e im plications o f this finding.
Conclusions
Our results suggest th a t m ost interventions tested are effec tiv e in red u cin g th e risk o f early read m issio n s. Some fe a tures, however, m ay enhance th e effect o f these programs. In particular, we found value in interventions th a t supported p a tie n ts’ capacity for self-care in their transition from hospital to hom e. F uture work in ten d ed to improve th e effectiveness o f health care delivery may benefit from consideration o f the dem ands th a t h ealth care interventions place on recently dis charged patients and th eir caregivers and th e extent to w hich these dem ands are offset by comprehensive support for im ple m entation.
ARTICLE INFORMATION
Accepted for Publication: March 8,2014.
Published Online: May 12,2014. doi:10.1001/jamainternmed.2014.1608.
Author Affiliations: Knowledge and Evaluation Research Unit, Mayo Clinic, Rochester, Minnesota (Leppin, Gionfriddo, Kessler, Brito, Wang, Boehmer, Ting, Murad, M ontori); Mayo Graduate School. Mayo Clinic, Rochester, Minnesota (Gionfriddo); Departm ent o f Medicine, Mayo Clinic, Rochester, Minnesota (Kessler, Brito, M ontori); Mayo Clinic Center fo r the Science o f Healthcare Delivery, Mayo Clinic, Rochester, Minnesota (Brito, Wang, Murad, M ontori); General Practice and Primary Care, Institute o f Health and Wellbeing, University o f Glasgow, Glasgow, Scotland, United Kingdom (Mair, Gallacher); Mayo Clinic Libraries, Mayo Clinic, Rochester, Minnesota (Erwin); medical student at St Louis University School o f Medicine, St Louis, Missouri (Sylvester); graduate student at University o f Minnesota School o f Public Health, Minneapolis (Boehmer); Division o f Health Policy and Management, School o f Public Health, University o f Minnesota. Minneapolis (Shippee).
Author Contributions: Drs Leppin and M ontori had full access to all o f th e data in the study and take responsibility fo r th e integrity o f the data and the accuracy o f th e data analysis. Study concept and design: Leppin, Gionfriddo, Mair, Gallacher, Erwin, Murad, Shippee, M ontori. Acquisition, analysis, o r interpretation o f data: Leppin, Gionfriddo, Kessler, Brito, Mair, Gallacher, Wang, Sylvester, Boehmer, Ting, Murad, M ontori. Drafting o f the manuscript: Leppin, Mair, Gallacher, Boehmer, Murad, Montori. Critical revision o f the manuscript fo r im portant intellectual content: All authors. Statistical analysis: Wang, Murad, Montori. Administrative, technical, or m aterial support: Leppin, Kessler, Brito, Mair, Gallacher, Erwin, Sylvester, Boehmer, M ontori. Study supervision: Leppin, M ontori.
Conflict o f Interest Disclosures: None reported.
Funding/Support: This publication was made possible by Clinical and Translational Science Award grant UL1TR000135 from th e National Center for Advancing Translational Sciences, a com ponent o f th e National Institutes o f Health.
Role o f the Sponsors: The funding source had no role in the design and conduct o f th e study; collection, management, analysis, and interpretation o f the data; preparation, review, or approval o f the manuscript; and decision to submit the manuscript fo r publication.
Disclaimer: The contents are solely the responsibility o f the authors and do n ot necessarily represent the official view o f the National Institutes o f Health.
Additional Contributions: The follow ing individuals provided unpublished data, conducted secondary analyses, assisted w ith study identification, and/or provided guidance and support: Agneta Bjorck Linne, MS, PhD, and Hans Liedholm, MD, PhD (Malmo University Hospital, Sweden); Marcia E. Leventhal, RN, MSN, Sabina De Geest, PhD. RN. and Kris Denhaerynck, PhD, RN (Institute o f Nursing Science, University o f Basel, Switzerland); Lars Rytter, MD (Glostrup University Hospital, Denmark); Gillian A. Whalley, PhD (University o f Auckland, New Zealand); David Maslove, MD, FRCPC (University o f Toronto, Canada); Judith Garcia-Aymerich, MD, PhD (Universitat Pomeu Fabra-Barcelona, Spain); Bonnie J. Wakefield, PhD, RN (Iowa City Veterans Affairs Healthcare System); Kathleen Finn, MD (D epartm ent o f Medicine, Massachusetts General Hospital, Boston); Jon C. Tilburt, MD, MPH (Mayo Clinic); Christiane E. Angermann, MD (Universitatsklinikum Wurzburg, Denmark); Felipe Atienza, MD, PhD (Hospital General Universitario Gregorio Maranon-Madrid, Spain); Dan Gronseth, BS (Mayo Clinic); Michael W. Rich, MD (Washington University, St Louis); A ndrew Masica, MD, MSCI (Baylor Health Care System); Karen B. Hirschman, PhD, and Mary D. Naylor, PhD (University o f Pennsylvania School o f Nursing); James F. Graumlich, MD (University o f Illinois College o f Medicine at Peoria); Anna Stromberg, RN, PhD (Linkoping University Hospital, Sweden). These contributors were not compensated fo r th e ir contributions.
REFERENCES
1. Jencks SF, Williams MV, Coleman EA. Rehospitalizations among patients in th e Medicare fee-for-service program. N Engl J Med. 2 0 0 9 :3 6 0 (14):1418-1428.
2. Committee on Redesigning Health Insurance Performance Measures, Payment, and Performance Im provem ent Programs. Rewarding Provider Performance: Aligning Incentives in Medicare. Washington, DC: National Academies Press; 2 0 0 6 .
3. Medicare Payment Advisory Commission. Report to the Congress: Promoting Greater Efficiency in Medicare. Washington, DC: Medicare Payment Advisory Commission; 2007.
4 . Joynt KE, Jha AK. A path forw ard on Medicare readmissions. N Engl J Med. 2013;368(13):1175-1177.
5. Shippee ND, Shah ND, May CR, Mair FS, M ontori VM. Cumulative complexity: a functional, patient-centered model o f patient com plexity can improve research and practice. J Clin Epidemiol. 2012;65(10):1041-1051.
6. Gallacher K, May CR, M ontori VM, Mair FS. Understanding patients' experiences o f treatm ent burden in chronic heart failure using normalization process theory. Ann Earn Med. 2011;9(3):235-243.
7. Sav A, Kendall E, McMillan SS, et al. 'You say treatm ent, I say hard w ork': treatm ent burden among people w ith chronic illness and th e ir carers in Australia. Health Soc Care Community. 2013;21 (6):665-674.
8. Krumholz HM. Post-hospital syndrom e-an acquired, transient condition o f generalized risk. N Engl J Med. 2013;368(2):100-102.
9. Leppin AL, Ting HH, Gionfriddo MR, et al. Describing the usefulness and efficacy o f discharge interventions: predicting 30 day readmissions through application o f the cumulative com plexity model (protocol). 2013. http://kerunit.fMes .wordpress.com/2011/07/ccm-readmissions -protocolll.docx. Accessed September 1,2013.
10. Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: th e PRISMA statement. PLoSMed. 2 0 0 9 ;6 (7 ):e 1 00 0 09 7 .
It. Hansen LO, Young RS, Hinami K, Leung A, Williams MV. Interventions to reduce 30-day rehospitalization: a systematic review. Ann Intern Med. 2011:155(8):520-528.
12. Higgins JP. Altman DG, Gatzsche PC, e t al; Cochrane Bias Methods Group; Cochrane Statistical Methods Group. The Cochrane Collaboration's tool
jamainternalmedicine.com JAMA Internal Medicine July 2014 Volume 174, Number 7 1105
Research Original Investigation Preventing 30-Day Hospital Readmissions
fo r assessing risk o f bias in randomised trials. BMJ. 2011;343:d5928.
13. Sterne JA, Egger M. Funnel plots fo r detecting bias in meta-analysis: guidelines on choice o f axis. J Clin Epidemiol. 2001;54(10):1046-105S.
14. DerSimonian R, Kacker R. Random-effects model fo r meta-analysis o f clinical trials: an update. Contemp Clin Trials. 2007;28(2):105-114.
15. Mantel N, Haenszel W. Statistical aspects o f the analysis o f data from retrospective studies o f disease. J Natl Cancer Inst. 1959:22(4):719-748.
16. Cochran WG. Some m ethods fo r strengthening th e common x 2 tests. Biometrics. 1954;10:417-451.
17. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557-560.
18. Herrera-Espineira C, Rodriguez del Aguila MdelM, Navarro Espigares JL, et al. Effect o f a telephone care program after hospital discharge from a trauma surgery unit [in Spanish], GacSanit. 2011;25(2):133-138.
19. M elton LD, Foreman C, Scott E, McGinnis M, Cousins M. Prioritized post-discharge telephonic outreach reduces hospital readmissions fo r select high-risk patients. Am J Manag Care. 2012:18(12): 838-844.
20. Marusic S, Gojo-Tomic N, Erdeljic V, e t al. The effect o f pharmacotherapeutic counseling on readmissions and emergency departm ent visits. Int J Ciin Pharm. 2013;35(1):37-44.
21. Altfeld SJ, Shier GE, Rooney M, e t al. Effects o f an enhanced discharge planning intervention for hospitalized older adults: a randomized trial. Gerontologist. 2013;53(3):430-440.
22. Davis KK, Mintzer M, Dennison Him melfarb CR, Hayat MJ, Rotman S, Allen J. Targeted intervention improves knowledge b ut not self-care or readmissions in heart failure patients w ith mild cognitive im pairm ent. EurJ Heart Fail. 2012;14(9): 1041-1049.
23. Bowles KH, Hanlon AL, Glick HA, et al. Clinical effectiveness, access to, and satisfaction w ith care using a telehomecare substitution intervention: a randomized controlled trial. In tJ Telemed Appl. 2011:2011:540138.
24. Finn KM, Heffner R, Chang Y, et al. Improving the discharge process by embedding a discharge facilitator in a resident team. J Hasp Med. 2011:6(9): 4 9 4 -5 0 0 .
25. Wong FK, Ho MM, Yeung S, Tam SK, Chow SK. Effects o f a health-social partnership transitional program on hospital readmission: a randomized controlled trial. SocSci Med. 2011;73(7):960-969.
26. Leventhal ME, Denhaerynck K, Brunner-La Rocca HP, e t al. Swiss Interdisciplinary Management Programme fo r Heart Failure (SWIM-HF): a randomised controlled trial study o f an outpatient inter-professional management programme for heart failure patients in Switzerland. Swiss Med Wkly. 2011;141:W13171.
27. Rytter L, Jakobsen HN, Ronholt F, e t al. Comprehensive discharge follow -up in patients' homes by GPs and district nurses o f elderly patients: a randomized controlled tria l. Scand J Prim Health Care. 2010;28(3):146-153.
28. Koehler BE, Richter KM, Youngblood L, e t al. Reduction o f 30-day postdischarge hospital readmission or emergency departm ent (ED) visit
rates in high-risk elderly medical patients through delivery o f a targeted care bundle. JHosp Med. 2009;4(4):211-218.
29. Braun E, Baidusi A, Alroy G. Azzam ZS. Telephone follow -up improves patients satisfaction follow ing hospital discharge. EurJ Intern Med. 2009;20(2):221-225.
3 0. Courtney M, Edwards H, Chang A, Parker A, Finlayson K, Hamilton K. Fewer emergency readmissions and better quality o f life fo r older adults at risk o f hospital readmission: a randomized controlled trial to determine the effectiveness o f a 24-week exercise and telephone follow -up program. JA m GeriatrSoc. 2009;57(3):395-402.
31. Jack BW, Chetty VK, Anthony D, et al. A reengineered hospital discharge program to decrease rehospitalization: a randomized trial. Ann Intern Med. 2009;150(3):178-187.
32. Wakefield BJ, Ward MM, Holman JE .etal. Evaluation o f home telehealth follow ing hospitalization fo r heart failure: a randomized trial. Telemed JE Health. 2008;14(8):753-761.
33. Balaban RB, Weissman JS, Samuel PA, Woolhandler S. Redefining and redesigning hospital discharge to enhance patient care: a randomized controlled study. J Gen Intern Med. 2 00 8 :2 3 (8): 1228-1233.
34. Wong FK, Chow S, Chung L, e t al. Can home visits help reduce hospital readmissions? randomized controlled trial. JAdvNurs. 2 0 0 8 :6 2 (5):585-595.
35. Coleman EA, Parry C, Chalmers S, Min SJ. The care transitions intervention: results o f a randomized controlled trial. Arch Intern Med. 2 0 0 6 ; 166(17):1822-1828.
36. Linne AB, Liedholm H. Effects o f an interactive CD-program on 6 m onths readmission rate in patients w ith heart fa ilu re -a randomised, controlled trial [NCT00311194], BMC Cardiovasc Disord. 2 0 0 6 :6 :3 0 .
37. Casas A, Troosters T, Garcia-Aymerich J, et al; members o f the CHRONIC Project. Integrated care prevents hospitalisations fo r exacerbations in COPD patients. EurRespirJ. 2006;28(1):123-130.
38. Riegel B, Carlson B, Glaser D. Romero T. Randomized controlled trial o f telephone case management in Hispanics o f Mexican origin w ith heart failure. J Card Fail. 2006:12(3):211-219.
39. Koelling TM, Johnson ML, Cody RJ, Aaronson KD. Discharge education improves clinical outcomes in patients w ith chronic heart failure. Circulation. 2005;111(2):179-185.
40. M ejhert M, Kahan T, Persson H, Edner M. Lim ited long term effects o f a management programme fo r heart failure. Heart. 2 0 0 4 :9 0 (9 ): 1010-1015.
41. Kwok T, Lum CM, Chan HS, Ma HM, Lee D, Woo J. A randomized, controlled trial o f an intensive com m unity nurse-supported discharge program in preventing hospital readmissions o f older patients w ith chronic lung disease. JAm GeriatrSoc. 2004;52(8):1240-1246.
42. Doughty RN, W right SP, Pearl A, e t al. Randomized, controlled trial o f integrated heart failure management: the Auckland Heart Failure Management Study. Eur Heart J. 2002;23(2):139-146.
43. JaarsmaT, HalfensR, Huijer Abu-Saad H .e ta l. Effects o f education and support on self-care and
resource utilization in patients w ith heart failure. Eur Heart J. 1999;20(9):673-682.
44. Naylor MD, Brooten D, Campbell R, e t al. Comprehensive discharge planning and home follow -up o f hospitalized elders: a randomized clinical trial. JAMA. 1999;281(7):613-620.
45. Stewart S, Pearson S, Luke CG, H orow itz JD. Effects o f home-based intervention on unplanned readmissions and out-of-hospital deaths. JAm GeriatrSoc. 1998;46(2):174-180.
46. Dunn RB, Guy PM, Hardman CS, Lewis PA, V etter NJ. Can a house call by a public health nurse improve th e quality o f th e discharge process for geriatric patients? Clin Perform Qual Health Care. 1995;3(3):151-155.
47. Rich MW, Beckham V, W ittenberg C, Leven CL, Freed land KE, Carney RM. A m ultidisciplinary intervention to prevent th e readmission o f elderly patients w ith congestive heart failure. N Engl J Med. 1995;333(18):1190-1195.
48. Naylor M, Brooten D, Jones R, Lavizzo-Mourey R, MezeyM, Pauly M. Comprehensive discharge planning fo r the hospitalized elderly: a randomized clinical trial. Ann Intern Med. 1994;120(12):999-1006.
4 9. Naylor MD. Comprehensive discharge planning fo r hospitalized elderly: a pilot study. Nurs Res. 1990;39(3):156-161.
50. Kulshreshtha A, Kvedar JC, Goyal A, Halpern EF, Watson AJ. Use o f remote m onitoring to improve outcomes in patients w ith heart failure: a p ilo t trial. In tJ Telemed Appl. 2010:2010:870959.
51. Graumlich JF, Novotny NL, Nace GS, e t al. Patient readmissions, emergency visits, and adverse events after software-assisted discharge from hospital: cluster randomized trial. J Hasp Med. 2009;4(7):11-19.
52. Atienza F, Anguita M, Martinez-Alzamora N, e t al; PRICE Study Group. M ulticenter randomized trial o f a comprehensive hospital discharge and outpatient heart failure management program. Eur J Heart Fail. 2004;6 (5 ):6 4 3-6 5 2.
53. Riegel B, Carlson B. Is individual peer support a promising intervention fo r persons w ith heart failure? J Cardiovasc Nurs. 2004;19(3):174-183.
54. Stowasser DA, Collins DM, Stowasser M. A randomised controlled trial o f medication liaison service s-p a tien t outcomes. J Pharm Pract Res. 2002;32(2):133-140.
55. Li H, Powers BA, Melnyk BM, et al. Randomized controlled trial o f CARE: an intervention to improve outcomes o f hospitalized elders and family caregivers. Res Nurs Health. 2012;35(5):533-549.
56. Shyu Yl, Liang J, Wu CC, e t al. A pilot investigation o f th e short-term effects o f an interdisciplinary intervention program on elderly patients w ith hip fracture in Taiwan. JAm GeriatrSoc. 2005;53(5):811-818.
57. Angermann CE, Stork S, Gelbrich G, e t al; Competence Network Heart Failure. Mode o f action and effects o f standardized collaborative disease management on m ortality and m orbidity in patients w ith systolic heart failure: the Interdisciplinary Network fo r Heart Failure (INH) study. Circ Heart Fail. 2012;5(1):25-35.
58. Naylor MD, Brooten DA, Campbell RL, Maislin G, McCauley KM, Schwartz JS. Transitional care of older adults hospitalized w ith heart failure: a randomized, controlled trial [published correction
1106 JAMA Internal Medicine July 2014 Volume 174, Number 7 jamainternalmedicine.com
Preventing 30-Day Hospital Readmissions Original Investigation Research
appears in JA m GeriatrSoc. 2004;52(7):1228]. JA m GeriatrSoc. 2 004;52(5):675-684.
59. Stromberg A, Martensson J, Fridlund B. Levin LA, Karlsson JE, Dahlstrom U. Nurse-led heart failure clinics improve survival and self-care behaviour in patients w ith heart failure: results from a prospective, randomised trial, Eur Heart J. 2003;24(11):1014-1023.
6 0 . Hansen FR, Poulsen H, Sorensen KH. A model o f regular geriatric follow -up by home visits to selected patients discharged from a geriatric ward: a randomized controlled trial. Aging (Milano). 1995; 7(3):202-206.
61. Maslove DM, Leiter RE, Griesman J, et al. Electronic versus dictated hospital discharge summaries: a randomized controlled trial. J Gen Intern Med. 2009;24(9):995-1001.
62. Forster AJ. Clark HD, Menard A, e ta l. Effect o f a nurse team coordinator on outcomes for hospitalized medicine patients. Am J Med. 2005;118 (10):1148-1153.
63. Dudas V, Bookwalter T, Kerr KM, Pantilat SZ. The im pact o f follow -up telephone calls to patients after hospitalization. Am J Med. 2001;111(9B): 26S-30S.
6 4. Parry C, Min SJ, Chugh A, Chalmers S, Coleman EA. Further application o f the care transitions
intervention: results o f a randomized controlled trial conducted in a fee-for-service setting. Home Health CareServQ. 200 9;28(2-3):84-99.
65. Hesselink G, Schoonhoven L, Barach P, et al. Improving patient handovers from hospital to prim ary care: a systematic review. Ann intern Med. 2012;l57(6):417-428.
6 6. Payne RA. Abel GA, Guthrie B, Mercer SW. The effect o f physical m ultim orbidity, mental health conditions and socioeconomic deprivation on unplanned admissions to hospital: a retrospective cohort study. CMAJ. 2013;185(5):221-228.
jam ainternalmedicine.com JAMA Internal Medicine July 2014 Volume 174, Number 7 1107
Copyright 2014 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
