Title: Research Utilization: and its many determinants
1Research Utilization and its (many) determinants
Carole A. Estabrooks Professor, Faculty of
Nursing Canada Research Chair in Knowledge
Translation University of Alberta
University of Melbourne Melbourne, AU October,
2005
2(No Transcript)
3Edmonton, Alberta
4University of Alberta
5(No Transcript)
6PARiHS Framework for Research Implementation
Evidence-based practice
Context
Facilitation
Evidence
PARiHS Model (Kitson, et al., 1998, QSHC
7Policy Framework
Policy Analysis Framework
Lomas, 2000
8First what do we know about the determinants of
research utilization?
9- Common barriers encountered in knowledge transfer
in hospital settings
- Commonly identified barriers include
- Time
- Resources
- Support
- Knowledge skills
- Adequate evidence
- Culture, context, environment.
- Less commonly identified barriers include
- Disincentives
- Irrationality of decision-making environment
- Decision-making errors
- Attitudes beliefs
- Inability to see consequences
10Best strategies for transferring research
findings?
- Imperfect evidence base for decision makers
- Many current rigorous evaluations have
methodological weaknesses - Poor reporting of study settings, barriers to
change, content and rationale of intervention - Generalisability of study findings is frequently
uncertain - Reminders most consistently observed to be
effective - Educational outreach only led to modest effects
- Dissemination of educational materials may lead
to modest but potentially important effects
(similar effects to more intensive interventions) - Multifaceted interventions not necessarily more
effective than single interventions
Grimshaw JM, Thomas RE, MacLennan G, Fraser C,
Ramsay C, Vale L et al. Effectiveness and
efficiency of guideline dissemination and
implementation strategies. Health Technol Assess
2004. (Available from http//www.hta.nhsweb.nhs.u
k/)
11From some of our work in KUSP
- In nursing
- Most common interventions tend to be education,
protocol guideline implementation - Little audit feedback, this is emphasized more
among physicians - Individual patient information, personal
experience, and social interactions are main
information sources - Limited use of journals, textbooks, and other
media, including the internet
1
12Scott-Findlays doctoral work on culture and
RU (in progress)
- Structure of Authority
- Nature of nurses work
- location of work
- expected level of activity
- routinized work
- expected ways to get work done
- Workplace climate
- ethical issues
- the response to errors
- the receptivity to innovation
- Types of knowledge valued
- clinical knowledge (practical, often tacit)
- specialized knowledge (e.g., PICU)
- research
13Scott-Findlay (in progress)
14From work in KUSP
For interventions we also need to consider (from
ethnographic case studies)
- Information needs and how these relate to the
sources of knowledge used in practice - How the knowledge sources used bear on the
packaging and dissemination of evidence - How dissemination strategies need to focus on
what is needed and useful regardless of what we
perceive may be the most relevant approach - A focus on those determinants of research use
with the potential to be manipulated in order to
develop interventions with potential utility
4
15From work in KUSP
Organizationally (causal and multi-level
modeling)
- nursing control over practice
- opportunity for nurse-to-nurse collaboration
- support for innovation
- facilitation of research use
- positive culture (characterized by good
leadership, evaluation and performance feedback) - increased time to nurse (-ve)
- emotional exhaustion (-ve)
- Higher nurse education levels
- Greater control over practice
- Greater freedom to make decisions
- Having a good nursing leader on unit
- Low levels of emotional exhaustion
- A sense of personal accomplishment
5
16KUSP Studies
- Theorizing the determinants of research
utilization (2000 2005) - KUPI - The knowledge utilization policy
implementation program (2002 2007) - The Bibliometric study (2004 2006)
- Measuring research utilization (2005 2008)
- New and future studies
17- Theorizing the determinants of research
utilization (2000 2005)
Research Utilization Studies
The Determinants of Research Utilization Pain
Management in Adults Children
Purpose To understand how nurses use research in
their practice and what influences their ability
to use or not use research.
18METHODS Ethnographic case studies
- Four hospitals (two in Ontario, two in Alberta)
- Adult and pediatric surgical units
- Qualitative and quantitative data collected
- Six months of participant observation each unit
- Nurses, patients, physicians, social workers,
physiotherapists, and administrators participated
19Data Sources
- Documents and records
- Interviews (individual group)
- Participant observations
- Physical artifacts
- Measures (e.g., Research utilization, unit
culture, critical thinking, pain, workload,
environmental complexity)
20Quantitative findings (selected - sources only)
- In this study the patient care unit was unit of
analysis - Our goal to understand the role of context in
shaping RU - We mapped relationship of contextual factors
using correspondence analysis
21Quantitative findings (contd)
- Units with highest RU clustered together on
factors such as - Critical thinking
- Unit culture
- Workload and people support clustered more
closely to units with lowest RU scores
22The 7 Units on knowledge sources
23(No Transcript)
24Number of Sources Used by Nurses
- 20 of nurses use 1 5 sources
- 50 of nurses uses 6 10 sources
- 25 of nurses use 11-15 sources
-
257 Unit Comparison
- Overall, few differences in the sources used by
nurses across units
26Sources of Knowledge Patterns Over Time
Note. Mean scores were transformed to fit a
five-point likert scale (0-4), with 0 as never
and 4 as always.
27Qualitative findings (selected - sources only)
Main Categories of Knowledge Sources
- Social interactions
- Experience
- Documents
- Intra-Personal
28Taxonomy of nurses sources of knowledge
29Why?
- Time
- Nurses typically work not only on multiple
schedules high and frequent levels of
resequencing are required - Context
- Nurses tend to be motivated to seek knowledge
when they have context specific patient care
situations, problems to deal with - Trust
- Nurses tend to seek information , knowledge from
those they trust (knowledgeable, empathetic) - Hierarchy
- An environment where norms of organizational
hierarchy are strictly adhered to may negatively
affect knowledge transfer between nurses and
CNSs, clinical leaders, and physicians.
30Summary Conclusions
- Individual patient information, personal
experience, and social interactions are main
information sources - Limited use of journals, textbooks, and popular
media, including the internet - The structure and organization of nurses work
has a major influence on their choice of
knowledge sources - Researchers need to merge nurses
information-seeking behavior with dissemination
strategies
31 Summary Conclusions
- The nurses in this study overwhelmingly relied on
social interactions and experience as their
primary sources of practice knowledge - Time, trust, and unit environment played an
important role in determining the nurses sources
of knowledge -
32- KUPI The knowledge utilization policy
implementation program (2002 2007)
33Evidence-based practice
- PARISH
- Promoting Action on Research Implementation in
Health Services - SI f (E,C,F)
- Successful implementation is a function of
evidence, context and facilitation - Kitson et al, 1998
- Rycroft-Malone et al, 2002
Context
Facilitation
Evidence
34 Knowledge utilization and policy implementation
(KUPI)
A five year research program whose purpose is to
develop theory in the knowledge utilization field
that increases (research) knowledge use by
clinicians and decision-makers across multiple
levels - by using existing data
35Objectives of the larger KUPI program
- To study
- how predictors of research use vary across levels
of decision-making - how research use is accomplished in
organizations, and - how research use shapes policy implementation
-
36Modeling Sub-Projects
Designed to study factors influencing knowledge
use, specifically research utilization (RU), at
different levels of decision-making among nurses
in Alberta hospitals.
37Modeling sub-project Five studies
- DV derivation outcome validation
- A structural equation (SEM) assessment of the
PARiSH (Kitson) model - Multi-level (HLM) modeling
- Hierarchical structural equation modeling (H-SEM)
- A methodological evaluation of HLM vs. H-SEM and
their usefulness for the study of research
utilization
38Example Multi-level (HLM) study objectives
- To determine which nursing factors at the
individual, unit, and hospital levels predict
research utilization - To measure the contribution of the significant
factors to variation in research utilization
across units and hospitals
39The Alberta Nurse Survey (1998)
A census population of all 12,345 registered
nurses from 129 hospitals in Alberta, Canada.
Useable returns 6526 (52.8)
A Employment Characteristics B Nursing Work
Index (NWI) C Maslach Burnout Inventory(MBI)
D Job characteristics E Last shift F
Demographics G Site specific questions
40The Research Utilization Survey (1996)
A randomly selected sample of staff nurses in
Alberta, Canada. Useable returns 600 (42)
- Research utilization
- Background and daily demands
- Individual and professional factors
- The research itself
- Your organization
-
-
41Final Sample
Hospitals Nurses 109
6,526 90 5,228 Selection criteria
5 nurses per hospital
42The 3 level model
Organization factors (hospital level)
Specialty factors (unit level)
Nurse factors (individual level)
43Significant univariate predictors Individual,
specialty (unit), and hospital levels
- Individual level
- Higher nurse education
- Control over practice
- Freedom to make decisions
- Emotional exhaustion
- Lack of personal accomplishment
- Specialty (unit) level
- Leadership
- Relational capital
- Consulting opportunity
- Participate in policy
- Organizational level
- Climate
- Innovativeness
- Organizational slack
-
44Summary
Greater research utilization among nurses was
predicted in the final model by
- Higher nurse education levels
- Greater control over practice
- Greater freedom to make decisions
- Having a good nursing leader on unit
- Low levels of emotional exhaustion
- A sense of personal accomplishment
45Components () of research explained at the
individual, specialty (unit), and hospital levels
4.5
3.7
91.8
46Summary
Variation in research utilization was mainly due
to differences in individual characteristics of
nurses, with organizational factors contributing
less. Of the latter, however, having good
leadership on the unit was the most significant
determinant of research utilization
47Structural equation (SEM) modeling objective
- to develop a theoretical model of hospital
characteristics that predict research use by
nurses
48The Structural Equation Model (SEM)
49Structural Equation Modeling Results
- Positive impact on research use
- nursing control over practice
- opportunity for nurse-to-nurse collaboration
- support for innovation
- facilitation of research use
- positive culture (characterized by good
leadership, evaluation and performance feedback) - Negative impact on research use
- Increased time to nurse
- emotional exhaustion
50- The Bibliometric study (2004 2006)
51- The Bibliometric study (2004 2006)
- Objectives
- Map the development of the field of knowledge
utilization - Identify the structure of the scientific
community - Provide direction to investigators
decision-makers in the field
52Terminology Tangle
53Pilot study A bibliometric analysis of the RU
literature in nursing
- 600 articles
- Findings/Conclusions
- Limited collaborations
- Primarily opinion pieces with little research or
theoretical work - UK US most prolific in the field
- Increased productivity since early 1990s with
change in terminology
Estabrooks, C.A., Winther, C., Derksen, L.
(2004). Mapping the field. A bibliometric
analysis of the research utilization literature
in nursing. Nursing Research, 53(5), 293-303.
54(No Transcript)
55Larger Study Methods
- Overview
- Web of Science 1945-2004
- Articles related to KU in general
- Three case studies related to KU
- Guidelines
- Policy management
- Nursing
56Preliminary findings
Key authors 15 most cited authors
57Key journals
- Journal of Evaluation in Clinical Practice
- Journal of Advanced Nursing
- British Medical Journal
- Knowledge Creation, Diffusion, Utilization
- Lancet
- Journal of General Internal Medicine
- Research Policy
- Journal of the American Medical Association
- Science Communication
- Social Science Medicine
58- Measuring research utilization (2005
2008)
59Systematic Review Findings
Existing measures
- Nursing Practice Questionnaire (NPQ)
- NPQ (Modified)
- Research Utilization Questionnaire (RUQ)
- Edmonton Research Orientation Survey (EROS)
- Other multi-item measures (N3)
- Single item measures (N19)
60Systematic Review Findings
- Poor construct clarity
- Lack of theoretical framing
- Lack of measurement theory
- Lack of psychometric assessment
- Presumption of linearity
- Absence of longitudinal work
- Self-report
- Recall
- Social desirability
- Scaling
- Unit of analysis
61Instrument Development Study
- Objectives
- to clarify and validate the construct
- to develop observable indicators
- to develop a set of items that measure research
use - to conduct a pilot test of the instrument
62Instrument Development study
- Approach
- Construct clarity ? Series of focus groups with
three sets of experts - Local (research team) expert panel (complete)
- International panel (May 2005)
- Managers and educators
- Providers (RNs, LPNs, assistants, aides)
- Indicator and item development ? Series of focus
groups with two sets of experts - Managers and educators
- Providers of nursing care (RNs, LPNs, RPNs,
assistants, aides) - Pilot test the draft instrument(s)
63Action in Practice
Action
Thinking (i.e., Clinical Decision
Making Clinical Reasoning )
Instrumental Use
Task completion
Research Use
Instrumental Use
Task completion Relational Assessment Screening
Research Non-Use
Non-use
A. Informed (non-defensible)
Non-use
B. Informed (defensible)
Non-use
Uninformed
64HI
Research utilization
Conceptual
Instrumental
HI
LO
LO
65- Health research utilization in hospitals (start
up) - Under review
- SHARP (Share and House Accessible Research
Products) - TROPIC team grant
- (KUSP) Team grant (letter of intent stage) The
influence of context on research use (long term
care settings)
66Health research utilization in hospitals (start
up) (2nd in a series of 3-4 studies)
Objectives of the first (completed) study
- To describe two groups (researchers and users)
- To compare research dissemination behaviour among
the researcher sub-groups - To compare research utilization behaviour among
the users sub-groups - To explore which factors contribute to research
dissemination (researchers) and research use
(users)
67Key Messages (from the first study)
- About Researchers
- Research infrastructure matters
- Not all academic researchers are equal
- Researchers do contribute to research use
- About Users
- Organizational capacity matters
- Intermediaries matter
- Talking with researchers matters
68Engaged Dissemination By Group
Plain Dissemination By Group
Number of Publication By Group
69Correlations
Correlation is significant at the 0.01 level
(2-tailed). Correlation is significant at the
0.05 level (2-tailed).
70Research Utilization Index
- Summary of Comparisons
- There is significant difference across groups.
- Multiple comparison test revealed 4 homogeneous
subsets nurse providers, nurse decision-makers
and other decision-makers, other decision-makers
and intermediaries and intermediaries and
physician providers. - The physicians are significantly different from
all nurses and other decision-makers.
71Factors Influencing Dissemination and Publication
Plain Dissemination Contribution by users,
Perceived impact, Perceived importance of
dissemination activities (B), Recognition of
expertise, Research focus users need Engaged
Dissemination Perceived impact, Perceived
importance of dissemination activities (B),
Research focus users need Publication Number
of research personnel, Research focus scholarly
advancement, Academic rank, Cost of utilization
72Present Study
- Objectives
- to determine the feasibility of undertaking a
larger scale study in Western Canada - To validate a typology of providers To validate a
revised survey instrument assessing individual
and organizational characteristics - To identify predictors of research utilization
- To assess provider group differences and
similarities
73The SHARP grant (infrastructure)
74Why SHARP?
Preserves data for future analysesFacilitates
researcher collaborationIncreases effectiveness
of fundingProvides opportunities for mentoring
and education in data managementReduces
duplication of data collection
75TROPIC
Purpose to ameliorate pain in children by
narrowing the gap between clinical practice and
the research evidence supporting optimal patient
care. Using the PARIHS framework
Main applicants Bonnie Stevens (PI), University
of Toronto Celeste Johnston, McGill
University Shoo Lee, University of Alberta Carole
Estabrooks, University of Alberta Shannon
Scott-Findlay, University of Alberta Patrick
McGrath, Dalhousie University Christine Chambers,
Dalhousie University
76Figure 1. Working model of research
utilization (Estabrooks, Scott-Findlay
Cummings, 2005)
77Three major TROPIC projects
- The creation of a centralized Canadian Pediatric
Pain Research Network Database (e.g., collect
demographics and information about pain) - Assess organizational context and determine its
impact on the implementation of pain research in
pediatric clinical practice - Evaluate the efficacy and effectiveness of (a)
Evidence-based Practice In Change (EPIC Lee et
al., 2002) and (b) Audit and Feedback
78- Culture
- OCI, Cooke Lafferty (1987)
- Leadership
- Leadership Practice Inventory (LPI) Kouzes,
Posner (2002) - Leadership Empowering Behaviours (LEB) Hui (1994)
- Evaluation
- Focused qualitative data
- Other overall contextual measures
- Maslach Burnout Inventory
- Nursing Work Index - Revised (NWI R)
- Environmental Complexity Scale (ECS)
- Project Research in Nursing (PRN) Workload
Measurement Instrument - Research Utilization (revised)
- Research Utilization Survey (Estabrooks 1997)
79- A main goal of the analyses will be to assess the
role of unit contextual variables on nurse
research utilization behaviors - Additionally, to estimate the independent effects
of these contextual variables, we have proposed
Hierarchical Linear Modeling (HLM) and Structural
Equation Modeling (SEM)
80(KUSP) Team LOI (tentative and evolving)
- Carole Estabrooks (PI, Canada Research Chair)
- Lesley Degner (CHSRF Chair)
- Michael Leiter (Canada Research Chair)
- Heather Laschinger (UWO, leadership)
- Greta Cummings (UA, leadership)
- Joanne Profetto-McGrath (UA, facilitation)
- Malcolm Smith (Business, Univ of Manitoba)
- Laurel Strain (Gerontology, sociology)
- UK
- Joanne Rycroft-Malone
- Sue Dopson
- USA
- Anne Sales
- And others tba
81Examining the context dimension of the PARIHS
framework in LTC settings
82www.ualberta.ca/kusp