OUTCOME OF STROKE AND HEALTH CARE RESOURCE UTILIZATION

1
OUTCOME OF STROKE AND HEALTH CARE RESOURCE
UTILIZATION

• SANDHYA SAMAVEDAM PGY3
• INTERNAL MEDICINE CATHOLIC HEALTH SYSTEM

2
Introduction

• Current standards for stroke care - without
  regard to age or functional status.
• Little data to guide adjustments to the medical
  and functional needs of the elderly.
• Unnecessary tests can only lead prolonged stay ,
  less time for rehabilitation.

3
Aim

• To look at the health care utilization and its
  effect on outcomes, taking into account the
  severity of stroke among patients who were 80 and
  above.
• Health care resources investigations,
  treatment, length of stay
• Outcome measures- functional independence,
  mortality, PEG dependence for feeding.
• Cutoff time for rehab. Centers was 90 days and
  cutoff for nursing home was 60 days.

4
Methods

• Under HIPAA
• Retrospective data collection from hospital
  records, nursing homes, rehabilitation centers.
• Data included 207 patients.
• Exclusion criteria Hemorrhagic stroke (2),
  advanced malignancy (2), brain tumors, TIAs (16),
  HIV(0).
• Additional exclusion included missing charts (1),
  repetition (6), nursing homes not being able to
  provide with outcome data (2), miscoding (1),
  thus total remaining 175 patients.

5
OUTCOME

• Favorable outcome
• patient being discharged home
• becomes independent or mildly dependent
• does not get a PEG.
• Poor outcome
• functional level remains moderate to severe
  dependency at cutoff point of time.
• If patient gets PEG tube or goes to hospice or
  deceases, then it is poor outcome.
• Functionality measured by FIM. The cutoff point
  of FIM equivalent rankin scores for defining
  moderate/severe dependency was 4 or above.

6
Statistics

• average age 86, max - 100
• 35.6 - males , 64.4 - females.
• Origin 85 - home and 15 - nursing home,
  assisted living or rehabilitation center.
• 37 - moderate to severe dementia at
  presentation.

7
CONTINUOUS INDEPENDENT VARIABLES
Minimum Maximum Mean Standard deviation
AGE 80 100 85.6 4.3
NIH 1 26 9.1 6.7
FIM AT D/C 13 91 46 28
LENGTH OF STAY 0 98 7.7 8.5
8
DISCRETETE INDEPENDENT VARIABLES
YES NO
DYSPHAGIA 33 67
DEMENTIA (mod-severe) 37 63
ATRIAL FIBRILLATION 46 (35 old, 11 newly diag.) 54
FEMALE 64.4 35.6
HOME 85 15
9
Discharge destination
DISCHARGE DESTINATION
HOME/ASSISTED LIVING 28
REHABILITATION CENTER 29
NURSING HOME 15
HOSPICE/DEATH 26
HOME VNA/ OTHERS 2

• discharges

10
DISCRETE VARIABLES CONTINUED

• TIME AFTER ONSET OF SYMPTOMS

• TREATMENT DISTRIBUTION

11
MORBIDITY DATA

• MORBIDITY LIST

• MORBIDITY DISTRIBUTION

• STROKE WITH HEMIPLEGIA OR RESIDUAL DEFECTS
  IMPAIRING MOBILITY
• CHF
• DM WITH ENDORGAN DAMAGE
• MOD-SEVERE PULMONARY DISEASES
• SEVERE ARTHRITIS/FRACTURE CAUSING IMMOBILITY
• PVD IMPAIRING MOBILITY

12
RISK FACTORS

• RISK FACTORS LIST

• RISK FACTORS

• HTN
• DYSLIPIDEMIA
• CAD/PVD/MI/AAA
• TIA
• A.FIB
• DM WITHOUT END ORGAN DAMAGE
• PFO

Risk factors No. subject
0 1 0.6
1 16 9.1
2 55 31.4
3 65 37.5
4 31 17.7
5 7 4
13
Investigations used per protocolin acute
situation

• CT scan brain
• CT STROKE PROTOCOL
• Doppler of carotids
• Further investigations as per need (if treatment
  could be changed with further investigations)

14
Per protocol investigations not followed
criteria

• When patient presents with stroke more than 6 hrs
  or with unknown time both CT brain or stroke
  study as well as MRI
• When patients present after 3 hrs and CT and CTA
  already shows ischemic stroke and did correspond
  to clinical presentation, also had MRI/MRA
• When already CT angio or doppler showed arterial
  block, patient had MRA

15
Investigations

• No correlation between NIH with the number of
  investigations a subjects received.
• No correlation between time of presentation with
  the number of investigations a subjects received.
• Mean NIH was similar in both the groups as seen
  from t-test

• No of subjects who had investigations that did
  not yield extra information that changed
  management in this study was 58.

16
OUTCOME STATS

• 64 SURVIVED STROKE
• 36 DIED AFTER CUTOFF TIME (DEATH
  DIRECTLY/INDIRECTLY RELATED TO STROKE OR MAY NOT
  RELATE TO STROKE)
• FAVORABLE OUTCOME SEEN IN 86 PATIENTS
• POOR OUTCOME SEEN IN 89 PATIENTS

17
Outcome stats

• There was no gender differences in the outcome
  groups
• More patients with dementia were in poor outcome
  group
• More patients with dysphagia were in poor outcome
  groups
• There was no significant correlation with atrial
  fibrillation
• Mean length of stay was about 2.6 days higher for
  patients with poor outcome. P-value 0.04 (6.4
  vs 9)

18
OUTCOME VS STROKE SEVERITYP-VALUE OF T-TEST
0.000SIGNIFICANT DIFFERENCE IN OUTCOME
OUTCOME N MEAN STD. DEVIATION
NIH score FAVORABLE 86 5.25 3.84
NON FAVORABLE 89 12.89 6.8

19
Outcome vs severity of stroke

• At about NIH 0f 8-9, data had more subjects with
  poor outcome than favorable outcome
• NIH correlates with outcome even after adjusting
  for time after onset of symptoms till
  presentation to ER.
• NIH also correlates when adjusted for type of
  treatment

20
Outcome vs comorbidities

• Chi square test was used to associate number of
  co-morbidities and outcome
• There was a significant correlation between the
  two.
• P-value was 0.039
• There was no significant relation between risk
  factors and outcome.

21
OUTCOME VS ONSET OF PRESENTATION

• Time of onset of symptoms did not correlate well
  with outcome.
• The above correlation was true even after
  adjusting for severity of stroke.
• Used chi square test.

22
Outcome vs investigations

• There was no correlation between outcome and
  investigations
• Used chi square test and logistic regression.
• This is true even after accounting for severity
  of stroke (NIH score) and co-morbid conditions.

23
LENGTH OF STAY AND INVESTIGATIONS

• Mean length of stay among those who got more
  investigational tests was 9 days and among those
  who had right amount of tests was about 6 days.
• There was a statistically significant difference
  between the two groups.
• After correcting for severity of stroke, the LOS
  was still statistically different between the two
  groups.

24
TREATMENT VS OUTCOME

• There was no statistically significant
  correlation between mode of treatment and
  outcome.
• Even after adjusting for severity of stroke,
  there was no significant correlation.

25
TREATMENT VS OUTCOME
26
Outcome vs functionality
FAVORABLE OUTCOME POOR OUTCOME
FIM (MEAN) 68 24
T-VALUE 16.01 (P-VALUE 0.0000) 16.01 (P-VALUE 0.0000)

• There is significant correlation with functional
  independence and outcome

27
conclusion

• Outcome of stroke among patients more than 80
  depended on NIH, Comorbid conditions, dysphagia,
  dementia.
• Outcome depended on functional independence.
• Investigational studies did not decide outcome
• Length of stay was more among those with poor
  outcome (difference of 3 days)
• Cutting down on investigation could save more on
  length of stay and could be used for functional
  improvement of patient

28
conclusion

• Patients presenting with NIH more than 9 mostly
  had worse outcome.
• Patients presenting after 3 hrs of onset of
  symptoms or after unknown time, there may be no
  requirement for more investigational studies than
  just CT head, perfusion study and carotid
  doppler.

29
strengths

• Simple retrospective study
• Data from a good stroke center
• Well defined outcome criteria
• Well defined functional level of patients were
  available in charts

30
weaknesses

• Sample size
• Needs further definition of investigational tests
  that did not help in change of treatment(Based on
  stroke protocol, which was designed for all
  stroke patients irrespective of age)
• Need further analysis with regards above.

31
references

• Guidelines for the Early Management of Patients
  With Ischemic Stroke A Scientific Statement From
  the Stroke Council of the American Stroke
  Association Adams et al., Stroke.
  2003341056-1083
• Recommendations for Imaging of Acute Ischemic
  Stroke A Scientific Statement From the American
  Heart Association. Latchaw et al., Stroke
  2009403646-3678 originally published online
  Sep 24, 2009

32
references

• Shaw TG, Mortel KF, Meyers JS et al Cerebral
  blood flow changes in benign aging and
  cerebrovascular disease. Neurology 1984 34
  855-862
• Falconer JA, Naughton BJ, Dunlop DD, Roth EJ,
  Strasser DC, Sinacore JM Predicting stroke
  inpatient rehabilitation outcome using a
  classification tree approach.
• Mauthe R., Haaf D., Hayn P., Krau J. Predicting
  discharge destination of stroke patients using a
  mathematical model based on six items from
  Functional independent measure. Archives of
  physical medicine and rehabilitation 1996 77
  10-13.
• Kelly-Hayes M, Robertson JT, Broderick JP,
  Duncan PW, Hershey LA, Roth EJ, Thies WH, Trombly
  CA. The American Heart Association Stroke
  Outcome Classification Executive summary.
  Circulation 1998 97 2474-2478.