An Evidence-based Pressure Ulcer Monitoring Tool for Spinal Cord Injury/Disease (SCI-PUMT)Gail Powell-Cope PhD, ARNP, FAANActing Director, HSR&D/RR&D Center of ExcellenceTampa, FLGail.powell-cope@va.gov

Monitoring Pressure Ulcer Healing in Persons with Spinal Cord Impairment

Funded by VA Health Services Research and

Development Service (HSR&D)

Nursing Research Initiative 03-245, IRB#:

104145, 2006 – 2008

These findings and conclusions do not

necessarily represent the

Department of Veteran Affairs or HSR&D

Investigators

Co-Principal Investigators

◦Susan S. Thomason DNP, MN, RN

◦Audrey Nelson PhD, RN, FAAN

(Retired)

Co-Investigators

◦Steven Luther PhD

◦Jeffrey J. Harrow MD, PhD, FACP

Study StaffPolly Placios, MS (Project/Data

Manager)Data Collectors

◦Stephanie McGovern, RN◦Francis Hernandez, RN◦Suk Tomlinson, RN◦Olivia Monteso-Smithson, RN◦Linda Smith, RN

Mary Reeder, BIS (Program Assistant)

Conclusion

This study found that the SCI-PUMT was a reliable, valid, and sensitive

instrument for measuring PrU healing in persons with SCI in a

100 bed VHA SCI/D Center.

“Problems” (or challenges)Clinical Problem

◦Pressure ulcers are a high volume, high cost condition in Spinal Cord Impairment

Implementation◦Translating consistent and quality

pressure ulcer monitoring into clinical practice, across all 32 SCI/D Centers, is a challenge. It takes 17 years from for new knowledge

generated by a randomized controlled trial to be incorporated into practice, and even then, the application is highly uneven (Balas & Boren, 2000).

Options for Implementing Changes

1. Dissemination Alone (journal articles, distribution of printed materials, CME) (not effective)

2. Educational Outreach (Academic Detailing and Local Opinion Leaders) (promising and mixed evidence)

3. Computer-based decision support systems (mixed)

4. Audit and Feedback (mixed evidence)

5. Patient-mediated Interventions such as education, reminders (promising)

6. Patient-specific clinical reminders (promising)

PARiHS FrameworkPromoting Action on Research

Implementation in Health Services

Successful implementation is a function of: ◦the nature and type of evidence ◦qualities of the context in which the

evidence is being introduced, and ◦the way implementation is facilitated

Evidence (Strong)

Evidence(Weak)

Context(Strong)

Context(Weak)

Ideal Situation for

Implementation

Innovation

Kitson, A. L., et al., (2008). Evaluating the successful implementation of evidence into practice using the PARiHS framework: theoretical and practical challenges. Implementation Science: IS, 3, 1. doi:10.1186/1748-5908-3-1  

Innovation

Innovation

Faci

litat

ion

Facilit

ation

Facilitation

Clinical ProblemPersons with spinal cord impairment

(SCI) are at extreme risk for PrU due to immobility, lack of sensation, collagen degradation, moisture, nutritional status, transfers, decreased ability to self-perform pressure redistribution, pain, and other risk factors.

PrU prevalence is 14-32%. PrU affect morbidity, mortality,

function, quality of life, and economics.

Clinical Practice Guidelines

Consortium for Spinal Cord Medicine (2000) Recommendations:Modify the treatment plan if the

ulcer shows no evidence of healing within 2-4 weeks.

Evaluate healing progress using an instrument or other quantitative measurements.

Clinical Practice Guidelines

National Pressure Ulcer Advisory Panel (NPUAP) European Pressure Ulcer Advisory Panel (EPUAP) (2009) Recommendations:Assess progress toward healing…use a

validated tool…Re-evaluate the PrU, the plan of care,

and individual if the PrU does not show progress toward healing within 2 weeks…

SettingMichael Bilirakis

Spinal Cord Injury/Disorders Center

James A. Haley Veterans HospitalTampa, Florida

100 inpatient beds CARF-accredited)Large outpatient patient populationHome Care (CARF-accredited)Long Term Care

However…Variations in how PrU healing is

measured varies across sites.◦Bates-Jensen Wound Assessment Tool

(BWAT)◦Pressure Ulcer Scale for Healing (PUSH)◦Hybrid tools with little psychometric

evaluationThese variations limit the ability to

conduct comparable trials of interventions

Research Questions

1. Is the SCI-PUMT valid for measuring PrU healing?

2. Is the SCI-PUMT reliable for measuring PrU healing?

3. How sensitive is the SCI-PUMT for measuring healing over time?

SCI-PUMT Phases

I. Development of item pool

II. Development and testing of SCI-PUMT

III. Analysis and SCI-PUMT refinement

IV. Assessment of SCI-PUMT reliability

DEVELOPMENT OF ITEM POOL

Phase 1

Development of Item PoolExpert Panel #1

◦ Aim: Identify measures and variables important and/or specific to PrU healing in SCI population

◦ 1 day on-site, Tampa, Florida◦ 9 interdisciplinary experts (MDs, RNs, OT, PT,

RD)◦ Variables then sent to EP for comment

Expert Panel #2

◦ Aim: Obtain content validity (all relevant concepts) for item pool

◦ 11 interdisciplinary experts (MDs, RNs, RD)◦ Aggregated variables sent to EP for comment

Item Pool

Consisted of 30 items◦Items from two established PrU

healing assessment tools (PUSH, BWAT)

◦Additional items identified by Expert Panels

DEVELOPMENT AND TESTING OF SCI-PUMT

Phase II

Subjects

Recruited from Inpatient, Outpatient, Home Care

3-year longitudinal cohort study

Assessed 30 PrU variablesPrU unit of analysis

Inclusion CriteriaEnrolled in SCI/D Registry and

receiving primary care from JAHVA SCI primary physician

Primary or secondary diagnosis of

Stages II-IV PrUSCI duration more than 12

months

Exclusion Criteria

Immune compromised

Severely mentally ill or cognitively impaired

Terminally ill

Subject Profile

Sample Size 66 Unique Patients167 Pressure Ulcers

Age 60 years (mean)

Gender Male 98%

Level of Injury Tetra 49%; Para 46%

ASIA A 58%; B 20%; Other 23%

Years since SCI onset

23 Years (mean)

High School Graduate

80%

Pressure Ulcer Characteristic

Findings

Number PrU / subject 1 – 9 (mean 2.5)

Previous PrU 77%

Prior PU surgery 53%

Location Ischia 43% Sacrococcygeal 26% Trochanter 8%Heel 8%

Stage II 20%; III 38%; IV 42%

Ulcer Pain 18%

Chronic Osteomyelitis 33%

Co-Morbidity Incidence

Diabetes Mellitus 26%

Anemia 24%

Peripheral Vascular Disease

11%

Chronic Obstructive Pulmonary Disease

9%

Congestive Heart Failure

8%

Heterotopic Ossification

8%

Other Baseline Factors

Incidence

Immunosuppressant Medications

12%

Spasticity Interference with Function(1 = none; 5 = maximum)

2.4 mean (SD 1.6)

SpasticityModified Ashworth Scale(1 = slight ↑ tone; 5 =

rigidity)

2.1 mean (SD 1.2)

Pain(0 = none; 10 = severe)

3.9 (SD = 2.6)

Mean Body Weight 175 lb (SD 36.4)

Nicotine Preceding WeekSubstance Abuse

29%6%

Data Collection6 Registered Nurse Data Collectors13 time points: 30 variables +

VeV PhotographBaseline and 12 weeks or

◦Complete healing◦Patient withdrawal◦Hospital discharged and lived >40

miles◦Plastic surgery intervention

VeV Measurement Documentation Software

Digital images used to calculate:• Volume

Intra- and Inter-Rater Reliability Ranges4 RN Data CollectorsIntra-Rater Reliability1 DCSame PrUTwice 1 ½ hours

apart

Inter-Rater Reliability4 DCSame PrUConsecutively

TOOL Intra-RaterICC

Inter-Rater ICC

PUSH 0.880.996

0.76 – 0.96

BWAT 0.87 – 0.99

0.69 – 0.91

ANALYSIS AND SCI-PUMT REFINEMENT

Phase III

Statistical Analysis

Construct validity

Predictive validity

Sensitivity to change

Internal consistency reliability

Construct ValidityExploratory Factor Analysis (EFA)

N = 167 PrUPrincipal factor extraction with

Promax (orthogonal rotation)Items removed from analysis based

on: ◦Values in correlation matrix◦Factor loadings of similar items (from 2

tools)◦Items not well defined by factors (low

communalities)

Variable Source

Geometric Factor

Substance

Factor Depth PUMT .82 -

Tunneling PUMT .77 -

Edges PSST .55 -

Undermining PUMT .48

Surface area PUSH .35 .51

Necrotic amount PSST - .52

Exudate type PUMT - .40

* Factor loading < |.30| have been replaced with “-“for ease of reading

Factor Analysis Results

Predictive ValidityOutcome variables to represent

PrU healing: Surface Area & Volume

Criterion Validity - VeV MD Software (within limits)

Correlates with the gold standard

Regression analyses – SCI-PUMT at baseline

Predictive ValidityExplains outcome variationsDependent variable:

Volume (VeV Camera)Predictor variables: Factor

analysis itemsSCI PUMT explained an

estimated59% of the variance in volume over the course of the study

Comparison of Scales:Volume (by VeV) Regression

SCI-PUMT

PUSH(Pressure

Ulcer Scale for Healing)

BWAT (Bates-Jensen Wound

Assessment Tool)

R2

(estimated based on proportional reduction in mean squared prediction error as per Snijders &

Bosker, 1994)

59% 57% 24%

ASSESSMENT OF SCI-PUMT RELIABILITY

Phase IV

Internal Consistency Reliability

Cronbach’s alpha = 0.74 (using study data)

SCI-PUMT ReliabilityAim: Evaluate intra- and inter-

rater reliability of in a clinical setting

26 Nurses trained in SCI-PUMT at Tampa VA SCI/D Center

Two months later, two sets of 3 SCI RNs evaluated 16 ulcers twice with an interval of 1½ hours between assessments

ResultsClinician Reliability

Intra-rater reliability 0.81 – 0.99

Inter-rater reliability 0.79

All reliability measures found to be above our established acceptability threshold

VARIABLES AND SCORING

SCI-PUMT

Pressure Ulcer

Site:

Sacrum or

Coccyx

Trochanter

Ischium

Heel

Other

______

Body Side: Right

Left

Midline

Orientation:

Medial Lateral

Positioning

Upper Leg Flexed When

Turned:

Yes

No

Surface Turned Onto: Right Left Back

Abdomen

Spinal Cord Impairment Pressure Ulcer Monitoring Tool (SCI-PUMT)

Patient ___________________SS#______________________ Ulcer # ______

Variable Score Options Score Geometric Factor

Surface Area(L x W)

1 2 3 4 5<1 cm2 >1 - <2.5 cm2 >2.5 - <5 cm2 >5 - <10 cm2 >10 - <15 cm2

6 7 8 9 10>15 - <25 cm2 >25 - <35 cm2 >35 - <55 cm2 >55 - <85 cm2 >85 cm2

Depth 0 1 2 3 40 cm >0 - <1 cm >1 - <2 cm >2 - <3 cm >3 cm

Edges 1 Indistinct, diffuse, none

clearly visible Distinct, outline clearly

visible, attached, even with ulcer base

Well-defined, not attached to ulcer base

2 Well-defined, not attached to base, rolled

under, thickened Well-defined, fibrotic, scarred, or hyperkeratotic

Tunneling 0None

1≤ 2 cm

2> 2 - ≤ 4 cm

3>4 cm

Undermining 0None

1≤ 2 cm

2> 2 - ≤ 4 cm

3>4 cm

Sub-total Score Geometric Factor

Substance Factor

Exudate Type 0None

1Serous/Sanguineous

2Green/Purulent

Necrotic Tissue Amount 0None

1<25%

2>25%

Sub-total Score Substance Factor

TOTAL SCORE (Total of Geometric and Substance Sub-totals)

_______________________________ ____________________

Maximum score = 26 The HIGHER the score, the more severe the ulcer. Evaluator: _________________________________________ Date ___________________________

SCI-PUMT ScoringEach variable assigned ordinal value

Data & clinical judgment to develop cut-points and weights for individual items and total scales score

Determined total score for SCI-PUMT = 26

Assigned proportion of total score to each sub-scale

Surface Area 40%

Depth 14%

Tunneling and Undermining

12% each

Edges, Exudate, Necrotic Tissue

8% each

SCI-PUMT Scoring

Study Limitations

Sample stratification excluded patients who had multiple etiologies of SCI; differentiation of ulcer etiology and ulcer stages were too small for computation

Healing process could be altered by tissue type and ulcer depth

Sample included persons with SCI from one SCI/D Center.

Continuing Psychometric Analysis

Can results of regression model be replicated over time

Does weighting of items improve the SCI-PUMT’s predictive value?

Do subscale scores have clinical utility?

Implications

SCI-PUMT can:◦Help to improve communication

among SCI healthcare providers.◦Form basis for outcomes monitoring

of PrU healing in persons with SCI.◦Assist clinician in critical decisions

affecting overall PrU management.

Implications

◦Allow for comparisons of healing rates within facilities and across sites.

◦Contribute to performance improvement initiatives and local and national performance measures.

◦Provide foundation to conduct treatment effectiveness studies of PrUs in multi-site VA SCI/D Center studies.

Conclusions

This study found that the SCI-PUMT was a reliable, valid, and sensitive

instrument for measuring PrU healing in persons with SCI in a

100 bed VHA SCI/D Center.

The Challenge—Full Implementation of the SCI-PUMT in the VA!