1. Exploration of Mobility in Hospitalized Older Adults
Cynthia J. Brown, MD, MSPH
Associate Professor of Medicine,
Director, Geriatric Medicine Section University of Alabama at Birmingham
Birmingham/Atlanta VA GRECC

2. Funding The John A. Hartford Foundation
Veterans Administration (VA) Rehabilitation Research and Development
National Institutes of Health (NIH)
Financial Disclosures: None

3. Determining the Scope of the Problem

4. - Dr. Tinsley Harrison, JAMA 1944
“Rest of injured body parts and of diseased bodies is probably the oldest and most valuable of all methods of treatment… Nevertheless we seem from time to time to forget that this therapeutic method like all others may lead to untoward results when utilized either injudiciously or excessively.”
- Dr. Tinsley Harrison, JAMA 1944
Almost 70 years ago in a JAMA editorial, Dr. Tinsley Harrison noted that rest of injured body parts may lead to untoward results when used injudiciously or excessively.

5. Creditor MC Ann Intern Med 1995
Almost 20 years ago, Dr. Creditor published this well-known figure in Annals of Internal Medicine, describing the interaction between aging and bed rest.
Yet as a geriatric medicine fellow at Yale in 2001, I found that while there were several review papers on the subject of mobility during hospitalization, there was little actual research on the subject.
Creditor MC Ann Intern Med 1995

6. Prevalence and Outcomes
Brown CJ, Friedkin RJ, Inouye SK. Prevalence and Outcomes of Low Mobility in Hospitalized Older Patients. J Am Geriatr Soc 52: , 2004.

7. Prevalence and Outcomes
Prevalence and Outcomes of Low Mobility in Hospitalized Older Patients
Prevalence and Outcomes
498 hospitalized medical patients, age ≥ 70 years
Mobility scale based on nurse report:
degree of assistance needed
number of times transferred and ambulated
Average of mobility observations for each patient, scores trichotomized
Low mobility: bed rest or bed to chair
Intermediate mobility
High mobility
Using a data set from Dr. Inouye’s delirium work, we examined the prevalence and outcomes associated with low mobility

8. Prevalence of Low Mobility
Bed rest present at some point for 33% of hospitalized older patients
16% patients experienced low mobility throughout hospitalization

9. Intermediate Mobility
Risk of Adverse Outcomes
by Mobility Level
Outcomes
Low Mobility
Intermediate Mobility
Any decline in ADLs
5.6
2.5
New Institutionalization at Discharge
6.0
2.9
Death
34.3
10.1
Death or New Institutionalization
7.2
3.3
Patients categorized as having low mobility throughout their hospitalization had 5.6 times higher odds of having ADL decline and 6 times higher odds of being discharged to a new institution. Among those in the intermediate mobility group, they had a 2.5 and 2.9 times higher odds of ADL decline and new institutionalization, respectively when compared to the high mobility group. Importantly, we adjusted for illness severity and comorbidity.
Adjusted for ADLs, Demographics, APACHE II, Charlson and ICU/CCU stay; Odds Ratio compared to High mobility group (P < .006)

10. Conclusions
Low mobility common and associated with adverse outcomes even after controlling for illness severity and comorbidities
However, little known about barriers to mobility

11. Barriers to Hospital Mobility
Using qualitative methods, we set out to identify the barriers to hospital mobility.
Brown CJ, Williams BR, Woodby LL, Davis LL, Allman RM.
Barriers to mobility during hospitalization from the perspective of older patients, their nurses and physicians.
J Hosp Med 2(5): , 2007.

12. Model of Potential Barriers
The first step was to create a model of the factors we thought might impact hospital mobility. We conceived of the barriers as being in one of four domains:

13. Methods Participants: Questionnaire Development:
10 patients, age > 75 years admitted to medical wards at UAB Hospital
Patient’s nurse & physician also recruited (n=29)
Questionnaire Development:
Semi-structured interview guide
New themes incorporated into interview
Interviews audiotaped, transcribed and examined for common themes

15. Barrier: Lack of Importance
"I don’t believe they are going to get me out of bed while I am here. If I said I really needed to get out of bed, they try to do what you want them to do. But evidently they don’t think it is that important.“ - a Patient

16. Barrier: Lack of Time
“We try to encourage the doctors to order physical therapy because we don’t have time to ambulate patients in the hallway like the doctor expects.” - a Nurse

17. Barrier: Environment
“I think it is just that patients, when they are in the hospital, they feel they are supposed to be in bed. And they are more comfortable there and a lot of times they can see the TV better.” - a Doctor

18. Implications
Suggests modifiable and non-modifiable reasons for low mobility
Important step in development of successful interventions to minimize low mobility

19. Beyond Functional Decline
At this point we know low mobility is common and associated with adverse outcomes, and that there are modifiable barriers. But we don’t know if the impact of hospitalization goes beyond ADL decline
Brown CJ, Roth DL, Allman RM, Sawyer P, Ritchie CS, Roseman JM. Trajectories of Life-Space Mobility after Hospitalization. Ann Intern Med 150(6): , 2009.

20. UAB Study of Aging
1000 Subjects, stratified, random sample of Medicare beneficiaries living in 5 counties in central Alabama Study over-sampled males, African Americans, and rural residents
Autauga
Baldwin
Barbour
Bibb
Blount
Bullock
Butler
Calhoun
Chambers
Cherokee
Chilton
Choctaw
Clarke
Clay
Cleburne
Coffee
Colbert
Conecuh
Coosa
Covington
Crenshaw
Cullman
Dale
Dallas
De Kalb
Elmore
Escambia
Etowah
Fayette
Franklin
Geneva
Greene
Hale
Henry
Houston
Jackson
Jefferson
Lamar
Lauderdale
Lawrence
Lee
Limestone
Lowndes
Macon
Madison
Marengo
Marion
Marshall
Mobile
Monroe
Montgomery
Morgan
Perry
Pickens
Pike
Randolph
Russell
Saint Clair
Shelby
Sumter
Talladega
Tallapoosa
Tuscaloosa
Walker
Washington
Wilcox
Winston
I had the opportunity to use data collected through the UAB Study of Aging, Dr. Allman’s R01.
Urban Counties
Rural Counties
ALABAMA

21. Measuring Life-Space
Bedroom
18.7 (15.1)
Home
19.0 (10.1)
Neighborhood
44.1 (14.0)
Town
59.3 (15.6)
Out of Town
81.8 (17.1)
Yard
32.1 (10.4)
Mean (standard deviation) for baseline composite life-space score among all UAB Study of Aging participants by LSA achieved without help from another person. Scores range
Bowling CB, et al. 2013
In the past 4 weeks have you gotten out of the room where you sleep-yes no. If yes, how frequently did you do this and did you need help either from another person or from equipment. Life space levels progress from the bedroom to going out of town.

22. Methods 211 hospitalizations among 687 participants over 4 years
Surgical admissions = 44;
Non-surgical admissions = 167
Life-Space Assessment every 6 months
Using multilevel change model to determined trajectory of Life-Space before and after hospitalization.

23. Life-Space Trajectories after Hospitalization
= Surgical admissions = Non-surgical admissions
Both green and blue--Going into town regularly
This study demonstrated that the adverse effects of hospitalization were even greater, causing a decrease in community mobility that was not recovered even after up to 2 years of follow-up.

24. Conclusions
Adverse consequences exceed functional decline, with older adults experiencing decrease in community mobility after hospitalization without recovery
Have clearly shown scope of problem, but need validated method to measure hospital mobility

25. Validating a Measurement Tool

26. Validation of Accelerometers
Brown CJ, Roth DL, Allman RM. Validation of the Use of Wireless Monitors to Measure Levels of Mobility During Hospitalization.
J Rehabil Res Dev 45(4): , 2008.

27. Methods
49 patients, age > 65 years admitted to medical wards at Birmingham VA Medical Center.
Inclusion Criteria:
Ambulatory 2 weeks prior to admission
Cognitively intact
English-speaking

28. Wireless Monitor Protocol
Monitors attached to thigh and ankle
Record position in space and range of motion every 20 seconds
A priori lying, sitting, and standing/walking defined based on pilot data
Direct observation of mobility to validate; up to six 2-hour periods of time over 2 days (total 12 hours)

30. Results 49 patients with 187 two-hour direct behavioral observations
Results obtained by monitors highly correlated with direct behavioral observations
Median Time-locked kappa =
Time-locked kappa, across all observations =
You use kappa when you are comparing 2 things for agreement (or disagreement) like how 2 radiologist read a chest x-ray.
Kappa is the measure of agreement beyond chance.
Anything > than 0.8 is considered excellent.
So the data collected by the wireless monitors very closely matched the data collected by the observer who is considered to be the gold standard.

31. Brown CJ, Redden DT, Flood KL, Allman RM
Brown CJ, Redden DT, Flood KL, Allman RM. The underrecognized epidemic of low mobility during hospitalization of older adults. J Am Geriatr Soc 57(9): , 2009

32. Epidemic of Low Mobility
45 hospitalized VA medical patients
Added exclusion criteria: Monitors attached within 48 hours of admission
Mean proportion of time spent lying, sitting, and standing/walking determined for each hour after hospital admission

33. Results Mean length of stay 5.1 days
Generated 2592 one-hour periods of data
No patient in bed entire hospital stay
83% of hospital stay spent lying in bed
Time spent standing/walking
Ranged from 0.2% to 21%
Median time was 3% or 43 minutes/day
Non-Exercise Activity (Thermogenesis)

34. Hourly Mobility Levels%
On Y axis is the proportion of veterans doing that activity; On the X axis is admission day and time. The number of patients included for each hour varies as patients are consented and discharged at different times.
Day Day Day Day Day Day Day 4
12pm am pm am pm am pm
Admission Day and Time

35. Conclusions
First study to document mobility continuously over initial 7 days of hospitalization
Found hospital patients spending at least 80% of time in bed
On average, less than 43 minutes a day standing or walking
This result has been duplicated by Dr. Steve Fisher at Galveston who found 57 minutes/day and a group from Europe who found standing/walking activity occurred XX minutes a day.

36. Developing an Intervention
At this point we have demonstrated the scope of the problem and we have a validated instrument to measure mobility, so it was time to try an intervention.

37. Previous Out of Bed Protocols
Transporters used to walk patients during quiet periods, especially nights, week-ends1
Pilot study, demonstrated feasibility only
Nurse driven protocol of progressive ambulation among patients with pneumonia2
No functional outcomes assessed
1 Tucker, 2 Mundy, et al. Chest, 2003

38. Safety and Efficacy of a Hospital Mobility Program
Specific Aim:
To examine impact of hospital mobility program on activities of daily living (ADL) and community mobility measured by Life-Space Assessment (LSA) compared to usual care

39. Methods 100 patients from Birmingham VAMC
Not delirious or demented, walking 2 weeks PTA
Randomly assigned to Mobility Program (MP) or Usual Care (UC)
Assessments by blinded assessors
One month telephone follow-up
Physicians blinded to assure no change in usual care (e.g. activity orders, PT consults)

40. Methods (cont.) Mobility Program (MP) Usual Care (UC)
Twice daily walks with assistance
Provision of rolling walker, if needed & safe
Provision of folder; document goals; track sitting, walking
Daily motivational interviewing; focus on goals & barriers
Twice daily friendly visits
Provision of folders; document friendly messages and track visitors

41. Assessments and Analyses
In-Hospital
One month follow-up
ADL ability
Baseline LSA
Depression
APACHE II
Charlson Comorbidity
Chart review for LOS, PT consults
ADL ability
Post-hospital LSA
Analyses
Missing values imputed
Missing values were imputed using several different methods. What is shown on the following slide is random number generation.

42. Baseline Characteristics
Usual Care
Walking Program
P value
Age
73.4 ± 7.0
74.4 ± 6.9
0.48
Gender, male
49 (98%)
48 (96%)
0.56
Race, black
8 (16%)
11 (22%)
0.44
LOS, mean
3.6 ± 2.4
4.6 ± 4.0
0.13
median
3.0
GDS
5.0 ± 3.0
4.7 ± 3.2
0.63
Charleson Comorbidity
4.1 ± 2.6
4.4 ± 2.4
0.55
APACHE
15.3 ± 11.8
14.3 ± 10.6
0.67
PT Ordered
17 (34%)
24 (48%)
0.15

43. Results In hospital, 3 falls in 2 patients reported – all in UC group
8 participants did not complete study;
2 UC and 6 MP
Death (n=3; 2MP, 1UC)
Medical complications (n=4, 4MP)
Patient refusal (n=1, 1UC)

44. Pre-Post Hospital Function
Usual Care
Mobility Program
P value
Baseline ADL
8.8 ± 2.3
8.5 ± 1.9
0.4
Post-Hospital ADL
8.3 ± 2.2
8.2 ± 1.9
0.9
P-values for group differences between pre and post hospital outcomes adjusted for baseline, age, gender, race.

45. Pre-Post Hospital Life-Space Assessment
Usual Care
Mobility Program
P value
Baseline LSA
51.5 (21.1)
53.9 (29.4)
0.4
Post-Hospital LSA
41.6 (21.5)
52.5 (29.0)
.0096
P-values for group differences between pre and post hospital outcomes adjusted for baseline, age, gender, race

46. Conclusions
Participants in UC group experienced mean 10-point decline in LSA scores; MP group experienced1-point decline.

47. Take Home Points
Older adults spend significant proportion of hospital stay in bed
Many barriers to hospital mobility modifiable
Our small RCT demonstrates feasibility, safety and efficacy of a hospital mobility program
Next steps include replication in larger cohort

48. Thank You Richard M. Allman, MD Kenneth Covinsky, MD, MPH
Thomas M. Gill, MD
Sharon K. Inouye, MD, MPH
Theodore M. Johnson II, MD, MPH
C. Seth Landefeld, MD
Kenneth Schmader, MD
Ronald I. Shorr, MD, MS
Stephanie A. Studenski, MD, MPH
Mark A. Supiano, MD
Mary E. Tinetti, MD
I’d like to end by thanking the group of geriatricians who nominated me for this award. I greatly appreciate your support.

49. Thank You The Team Tanya Brecht, BA Richard M. Allman, MD
Ellen Porter, BA
Angel Watson
Aaron Weeks, BS
Kathy T. Foley, PhD, OTR/L
John D. Lowman, Jr. PhD, PT
The Veterans and Nurses at the Birmingham VAMC
Richard M. Allman, MD
Sharon K. Inouye, MD, MPH
Stephanie A. Studenski, MD, MPH
Mary E. Tinetti, MD
Robert and Jean Brown
David T. Gadbois
I’d also like to thank my team who makes this work happen, especially Tanya and Ellen my project coordinators as well as the veterans and nurses at the Birmingham VA.
My mentors, Dr. Richard Allman who has helped me nurture my career both as a clinician investigator as well as a leader in geriatrics. And my female triumvirate, Drs. Inouye, Studenski and Tinetti who got me started on this road and have helped me stay the course even when it got rocky.
Last, I would like to thank my parents Bob and Jean Brown, for teaching me to work hard and never take myself too seriously. And my beloved husband Dave for his love and support which have made all things possible.

50. Exploration of Mobility in Hospitalized Older Adults
Cynthia J. Brown, MD, MSPH
Associate Professor of Medicine,
Director, Geriatric Medicine Section University of Alabama at Birmingham
Birmingham/Atlanta VA GRECC