Development and assessment of the Alberta Context Tool
© Estabrooks et al; licensee BioMed Central Ltd. 2009
Received: 20 June 2009
Accepted: 15 December 2009
Published: 15 December 2009
The context of healthcare organizations such as hospitals is increasingly accepted as having the potential to influence the use of new knowledge. However, the mechanisms by which the organizational context influences evidence-based practices are not well understood. Current measures of organizational context lack a theory-informed approach, lack construct clarity and generally have modest psychometric properties. This paper presents the development and initial psychometric validation of the Alberta Context Tool (ACT), an eight dimension measure of organizational context for healthcare settings.
Three principles guided the development of the ACT: substantive theory, brevity, and modifiability. The Promoting Action on Research Implementation in Health Services (PARiHS) framework and related literature were used to guide selection of items in the ACT. The ACT was required to be brief enough to be tolerated in busy and resource stretched work settings and to assess concepts of organizational context that were potentially modifiable. The English version of the ACT was completed by 764 nurses (752 valid responses) working in seven Canadian pediatric care hospitals as part of its initial validation. Cronbach's alpha, exploratory factor analysis, analysis of variance, and tests of association were used to assess instrument reliability and validity.
Factor analysis indicated a 13-factor solution (accounting for 59.26% of the variance in 'organizational context'). The composition of the factors was similar to those originally conceptualized. Cronbach's alpha for the 13 factors ranged from .54 to .91 with 4 factors performing below the commonly accepted alpha cut off of .70. Bivariate associations between instrumental research utilization levels (which the ACT was developed to predict) and the ACT's 13 factors were statistically significant at the 5% level for 12 of the 13 factors. Each factor also showed a trend of increasing mean score ranging from the lowest level to the highest level of instrumental research use, indicating construct validity.
To date, no completely satisfactory measures of organizational context are available for use in healthcare. The ACT assesses several core domains to provide a comprehensive account of organizational context in healthcare settings. The tool's strengths are its brevity (allowing it to be completed in busy healthcare settings) and its focus on dimensions of organizational context that are modifiable. Refinements of the instrument for acute, long term care, and home care settings are ongoing.
Organizational context can be defined as "...the environment or setting in which people receive healthcare services, or in the context of getting research evidence into practice, the environment or setting in which the proposed change is to be implemented" [, p. 299]. Organizational context is widely considered to be an important influence on the successful implementation of research evidence in healthcare settings [1–4]. However, relatively little empirical evidence exists to support this claim. Further, its measurement has not been adequately addressed. In this paper, we report the first major assessment of a newly developed instrument, the Alberta Context Tool (ACT), designed to parsimoniously measure organizational context as perceived by healthcare providers working in complex healthcare settings.
The Alberta Context Tool (ACT) was developed with a specific purpose in mind and this shaped the approach taken to expanding our understanding of the construct of 'context'. Given our belief that organizational context is a central influence on the effective use of clinically relevant research evidence by healthcare providers, we sought to develop a tool that would allow us to assess context validly and reliably within complex healthcare settings where care is provided to patients. The resulting context measure was intended for administration at the level of the individual healthcare provider to determine their perception of context as it applies to a patient care unit or organization (e.g., hospital), depending on the individual's context of care delivery.
In the development of the ACT we tried to balance, to the extent possible, three principles: a substantive theory, brevity, and modifiability. We used the Promoting Action on Research Implementation in Health Services (PARiHS) framework to conceptualize organizational context. When the framework did not provide direction, we operationalized concepts from related literature (e.g., [5–8]). The PARiHS framework provides a broad conceptualization of how research implementation occurs in organizational settings. In the index paper for PARiHS  successful research implementation was proposed to result from the interplay and interdependence of three core elements: (1) evidence, (2) facilitation, and (3) context. We were interested in the context domain.
Context, in the PARiHS framework, is construed generally as the work setting and more specifically embodies three domains: culture, leadership and evaluation. Culture is defined as "the forces at work, which give the physical environment a character and feel" [9, 10]. Subsequent exploration into the concept of 'culture' by McCormack and colleagues  resulted in further refinement of the definition of culture to encompass the prevailing beliefs and values, as well as consistency in these values and a receptivity to change, among members of an organizational setting.
The PARiHS framework defines leadership as the "nature of human relationships" [, p.98] with effective leadership giving rise to clear roles, effective teamwork and organizational structures, and involvement by organizational members in decision making and learning. This closely resembles 'transformational leadership', a broad term reflecting leaders thought to be among the most effective leaders because they are able to transpose their ideas and beliefs into collective beliefs which eventually become assumptions and part of a unit's culture [11–13]. Emotionally intelligent leadership styles are one form of leadership consistent with transformational styles because they focus on how leaders manage their own emotions and their relationships with others both individually and in larger social settings .
Evaluation is described in the PARiHS framework as feedback mechanisms (individual and system level), sources, and methods for evaluation . Audit (data gathered about the processes and/or outcomes of patient care) coupled with feedback (data provided to members of the organization) is one of the most commonly applied evaluation methods in healthcare organizations. Research implementation is hypothesized by the PARiHS developers to be most successful when evaluation occurs routinely.
A recent paper by the PARiHS group suggests that a fourth contextual component, resources, is important to the implementation of research findings. In 2004, Rycroft-Malone and colleagues  interviewed staff nurses, nurse managers, and other implementation 'experts' at two acute care agencies in the United Kingdom and identified time, equipment, and clinical skills as resources needed to implement research findings. They also identified the complexity of relationships among these resources.
Development of the ACT
Concepts in the ACT survey
The actions of formal leaders in an organization (unit) to influence change and excellence in practice, items generally reflect emotionally intelligent leadership
H1: Care providers who perceive more positive (emotionally intelligent) unit leadership report higher research use
Calmly handles stressful situations
The way that 'we do things' in our organizations and work units, items generally reflect a supportive work culture
H2: Care providers who perceive a more positive unit culture report higher research use
My organization effectively balances best practice and productivity
The process of using data to assess group/team performance and to achieve outcomes in organizations or units
H3: Care providers who perceive a larger number of unit feedback mechanisms report higher research use
Our team routinely monitors our performance with respect to the action plans
Social Capital 1
The stock of active connections among people. These connections are of three types: bonding, bridging, and linking
H4: Care providers who perceive more positive unit social capital activities report higher research use
People in the group share information with others in the group
Formal Interactions 2
Formal exchanges that occur between individuals working within an organization (unit) through scheduled activities that can promote the transfer of knowledge
H6: Care providers who perceive a larger number of formal unit interactions report higher research use
How often do these activities occur?
Informal Interactions 2
Informal exchanges that occur between individuals working within an organization (unit) that can promote the transfer of knowledge
H5: Care providers who perceive a larger number of informal unit interactions report higher research use
How often do you interact with people in the following roles or positions?
- Someone who champions research and its use in practice
Structural/Electronic Resources 3
The structural and electronic elements of an organization (unit) that facilitate the ability to assess and use knowledge
H7: Care providers who perceive a larger number of unit structural and electronic resources report higher research use
How often do you use/attend the following?
- A Library
The cushion of actual or potential resources which allows an organization (unit) to adapt successfully to internal pressures for adjustments or to external pressures for changes
• Human Resources (staffing)1
H8: Care providers who perceive sufficient unit staffing levels report higher research use
Enough staff to deliver quality care
• Time 1
H9: Care providers who perceive having sufficient time on their unit report higher research use
Time to do something extra for patients
• Space 1
H10: Care providers who perceive having sufficient space on their unit report higher research use
Use of designated space
The reduced (56-item) version of the ACT was pretested for feasibility and completion time with pediatric nurses in two hospitals in Alberta (Canada). In the pretest, the 56-item ACT was embedded in a larger survey consisting of 135 items and administered to 249 nurses. The mean time to complete the entire survey was 22 minutes for those who completed it online (n = 209) and 33 minutes for those who completed it using paper (n = 40), resulting in average item to completion times of 9.8 seconds for the online administration and 14.7 seconds for the paper administration. Using this average item time, we estimated a mean time to completion for the 56-item ACT of 9.1 minutes (when administered online) and 13.7 minutes (when administered by paper), both significantly less than time to completion for the original (76-item) version. Based on these completion times, we decided to administer the ACT in the larger multi-site study (reported in this paper) in online format only.
Design, sample, and data collection
Seven pediatric hospitals in six Canadian provinces provided the sampling pool for the administration of the English version of the refined 56-item ACT (henceforth simply the ACT). Five healthcare professional subgroups were eligible to participate: nurses, physicians, allied professionals, educators/clinical specialists, and managers. Inclusion and exclusion criteria for the professional subgroups are summarized in Additional File 2. Data were collected using an on-line survey. Eligible participants were provided with a survey package containing a letter introducing the study, and a business card providing a Uniform Resource Locator (URL) and unique password to access the survey on-line.
Ethical approvals for the study were obtained from the appropriate universities and hospital review boards in the respective Canadian provinces.
Data analyses (except aggregation statistics) were carried out using the Statistical Package for the Social Sciences for Windows (SPSS, v. 16.0) on data collected from professional nurses (n = 752); aggregation statistics were carried out using the SAS 9.2 statistical program. Data analyses included a missing-values analysis; items with missing values greater than 10% were considered for removal and/or imputation . Descriptive statistics (variance, mean, histograms) were generated for each item and examined for amount of variance and middle range mean scores as well as sufficient endorsement frequency. Items with a very high or low frequency (endorsement frequency) were considered for elimination since answers can be predicted with frequencies greater than 80% accuracy and below 20% . These frequencies would also have no influence on the scale's psychometric properties and may increase burden by making it longer.
Since this was the first major field assessment of the ACT, our assessment was largely exploratory rather than confirmatory in nature. Therefore, to examine the underlying dimensional structure of the ACT, we performed factor analysis using principal component analysis (PCA) with orthogonal (Varimax) rotation rather then other factor-analytic methods such as 'principal axis factoring' or 'common factor analysis'. Missing values, which were limited, were treated as such with no substitution or imputation of estimated values. Factors were identified using the 1.0 eigenvalue cutoff rule and the Scree test. Item retention was based on coefficient values (factor loadings ≥ 0.35). Items that cross loaded (factor coefficients ≥ 0.35) on two or more factors were examined on a case-by-case basis and were either re-conceptualized or eliminated from the scale to achieve a balance of good estimation and avoidance of overcapitalizing on sampling error . The Varimax rotation with Kaiser normalization, as recommended by Kline , was used to enhance interpretability of the principal component analysis. Following factor analysis, corrected item-total correlations were reviewed for items within the factors identified; items that correlated with the total score below 0.30 were considered for deletion . Items were also considered for deletion if they: (1) caused a significant increase in scale alpha values if they were deleted (item-total statistics), or (2) were highly correlated (> .70 from item-to-item correlations) with each other .
The ACT was developed to measure organizational context and was motivated by a need to build a better understanding of how to design effective interventions that result in better research uptake. Several studies examining the impact of context on research implementation in both the nursing [2, 23–26] and organizational behaviour literature  support the importance of contextual factors to research use. We assessed construct validity of the ACT by examining associations between each of its factors and instrumental research utilization. Instrumental research utilization was defined to study participants as the use of observable research-based practices when caring for patients and was scored on a 5-point frequency scale from 1 (use less then 10% of the time) to 5 (use almost 100% of the time). This item has been used in several previous studies [28–30]. Items within each ACT concept were averaged (culture, leadership, evaluation, social capital, organizational slack-human resources, organizational slack-time, organizational slack-space) or recoded as existing or not existing and then counted (informal interactions, formal interactions, structural and electronic resources) to calculate one derived score for each factor.
While research utilization and the ACT variables were measured and analyzed at the individual level in the study reported in this paper, individual scores on the ACT can be aggregated to obtain unit scores by calculating group means. Therefore, we also calculated a set of indices to assess each identified factor's performance when aggregated. One-way analysis of variance (ANOVA) was performed for each variable using the unit as the group variable. The source table from the one-way ANOVA was used to calculate the following indices: (1) interclass correlation ICC (1) = (BMS - WMS)/(BMS + [K - 1] WMS), where BMS is the between-group mean square, WMS is the within-group mean square, and K is the number of subjects per group. The average K for unequal group size was calculated as K = (1/[N - 1]) (ΣK - [ΣK2/ΣK]); (2) interclass correlation ICC (2) = (BMS - WMS)/BMS; (3) η2 = SSB/SST, where SSB is the sum of squares between groups and SST is the sum of squares total; and (4) ω2 = (SSB - [N - 1]WMS)/(SST + WMS). For each variable analyzed, there is strong agreement among nurses in each given unit when ICC (1) is greater than 0.1. Aggregated data are considered reliable when the F statistic from the ANOVA table is statistically significant (p < 0.05) and/or ICC (2) is greater than 0.60 . An indicator of effect size is η2, the proportion of variance in the individual factor accounted for by group membership . Omega squared (ω2) is a measure of the relative strength of the aggregated variable at the group level . Both η2 and ω2are measures of validity of the aggregated data at the patient care unit level.
Reliability of the factors within the ACT instrument was examined using Cronbach's alpha (α). Factors below the acceptable standard (0.70) for scales intended to compare groups were considered for revision [21, 34].
Characteristics of Study Sample (n = 752)
Gender [n, (%)]
Age [n, (%)]
Highest Education Level [n, %]
Number of Years Worked in Current Position [Mean (SD)]
Number of Years Worked on Unit [Mean (SD)]
Overall instrumental research utilization score (mean, SD)
Missing values and descriptive statistics
We used listwise deletion to deal with missing data. No individual ACT items were missed by greater than 10% of respondents. Item distribution of the 56 ACT items showed acceptable variance and middle range mean scores.
Validity - Internal structure
ACT Domains: Psychometric Validity and Reliability
Factor Analysis1 (n = 704)
No. Completed Responses
Factor Loadings (Range)
0.753 - 0.842
0.389 - 0.701
0.774 - 0.864
0.584 - 0.684
0.369 - 0.702
Type 1- Non-direct Care providers
0.550 - 0.768
Type 2- Direct Care providers
0.339 - 0.798
Structural and Electronic Resources*
Type 1-Formal Resources
0.439 - 0.806
Type 2- Traditional Resources
0.510 - 0.720
Type 3- Electronic Resources
0.602 - 0.710
0.606 - 0.717
0.636 - 0.807
0.759 - 0.788
Culture, Leadership, Evaluation, Structural and Electronic Resources
Culture, leadership, evaluation, and resources constitute organizational context according to the PARiHS framework, which guided the development of the ACT. In our 13-factor solution, the first two factors (eigenvalues 9.806 and 3.825) included all items in the evaluation and leadership subscales and accounted for most of the variance at 17.51% and 6.83%, respectively. The culture items constituted the seventh factor, (eigenvalue 1.725), accounting for 3.08% of the variance. Structural and electronic resources were represented in the eighth, ninth, and thirteenth factors (eigenvalues 1.441, 1.406, and 1.042 respectively) and accounted for 2.57%, 2.51%, and 1.86% of the variance respectively in organizational context. Items with the highest factor coefficients were used to name the three 'types' of structural and electronic resources; type 1 (factor 8) represented formal resources (e.g., policies and procedures, clinical practice guidelines); type 2 (factor 9) represented traditional resources (e.g., textbooks, journals), and type 3 (factor 13) represented electronic resources (e.g., reminder systems, computerized decision support). Together, culture, leadership, evaluation, and structural and electronic resources (context according to the PARiHS framework) accounted for 34.36% of the variance in organizational context as measured by the ACT.
The third factor (eigenvalue 3.412) represented social capital and accounted for 6.09% of the variance in organizational context as measured by the ACT.
Informal interactions were represented in the fourth and fifth factors (eigenvalues 2.519 and 2.460) and accounted for 4.50%, and 4.39% of the variance respectively for a total explained variance of 8.89%. Items with the highest factor coefficients were used to name the two 'types' of informal interactions; type 1 (factor 4) represented interactions with non-direct care providers (e.g., interactions with a clinical educator, interactions with a quality improvement specialist) while type 2 (factor 5) represented interactions with direct care providers (e.g., interactions with other nurses, hallway talk). Formal interactions (e.g., team meetings, patient rounds) were represented in the tenth factor (eigenvalue 1.286) and accounted for an additional 2.30% of the variance in organizational context as measured by the ACT.
The sixth, eleventh, and twelfth factors (eigenvalues 1.836, 1.232, and 1.195 respectively) represented the three sub dimensions of organizational slack - time, space, and human resources (staffing). These sub dimensions accounted for 3.28% (time), 2.20% (space), and 2.13% (human resources) of the variance for a combined variance of 7.61% in organizational context as measured by the ACT.
Corrected item-total correlations for items within each of the 13 factors, with the exception of one item, (continuing education in the formal interaction factor, item-total correlation = .231) were greater than the predetermined cutoff of .30 indicating items within each factor were related to the overall scale for that factor. Item-total statistics (alpha when item deleted) for each factor also remained stable, providing further internal structure validity evidence for the ACT.
Assessment of Construct Validity: Correlation of derived ACT factors by increasing levels of instrumental research utilization (IRU)
Mean value (or relative ▽ %) of ACT factors by increasing levels of instrumental research utilization
Bivariate correlation with IRU*
P-value for means differences
0.090* 1/0.098** 2
0.000 (0.000) **
0.009 (0.027) **
0.000 (0.000) **
(Type 1- Non-Direct Care Providers)
0.000 (0.000) **
(Type 2- Direct Care Providers)
0.000 (0.000) **
Structural and Electronic Resources
(Type 1-Formal Resources)
0.000 (0.000) **
Structural and Electronic Resources
(Type 2- Traditional Resources)
0.003 (0.006) **
Structural and Electronic Resources
(Type 3- Electronic Resources)
0.000 (0.000) **
Organizational Slack-Human resources
0.000 (0.001) **
0.000 (0.001) **
We also assessed the relative percent difference in mean score of each of the ACT's 13 factors from the sample average (Table 4). The results showed a positive incremental relationship with increasing levels of instrumental research utilization (i.e., the higher the contextual scores for each factor assessed relative to the sample average, the better the level of research utilization).
Internal reliability estimations
Table 3 lists the Cronbach alpha coefficients for each of the 13 factors within the ACT. Coefficients ranged from a low of .54 (for structural and electronic resources - type 3 [electronic resources]) to a high of .91 (for leadership and evaluation factors). With the exception of four factors (structural and electronic resources - type 2 [traditional resources] and type 3 [electronic resources], formal interactions and organizational slack-space) all exceeded the acceptable standard (> 0.70) for scales intended to compare groups recommended by Nunnally and Bernstein  and Altman and Bland .
Aggregation of the measures to unit level
The ACT is intended to provide responses that can be aggregated to the level of the patient care unit or to higher organizational levels depending on the context of care delivery for the group(s) completing the instrument. Therefore, we also assessed the performance of the ACT factors when aggregated to the unit level. When developing the ACT, items within the various dimensions were constructed to direct respondents' attention to common experiences on a particular patient care unit in order to ensure that the ACT was meaningful at the unit level. For example, lead-in instructions to the various ACT dimensions asked respondents to assume the shared perspective of their patient care unit (e.g., on my unit we....). Designing items in this way brings about less within-group variability and more between-group variability compared to traditional survey items that focus on individual experiences and perceptions [35, 36].
Aggregation of Data to Patient Care Unit Level
(Type 1- Non-Direct Care Providers)
(Type 2- Direct Care Providers)
Structural and Electronic Resources
(Type 1-Formal Resources)
Structural and Electronic Resources
(Type 2- Traditional Resources)
Structural and Electronic Resources
(Type 3- Electronic Resources)
Organizational Slack-Human Resources
ICC(1): The range of ICC(1) values (all greater than 0.00) indicate a degree of perceptual agreement among the nurses about the mean values on the ACT factors within each unit. That is, the nurses' perceptions about context within a particular patient care unit were similar.
ICC(2): All ACT factors showed statistically significant (p < .05) F statistics and ICC(2) values greater than 0.60 (with the exception of structural and electronic resources type 2 [traditional resources]), that is, were we to draw repeated subsequent samples from the same groups (units) we would obtain similar mean scores.
η 2 and ω 2 : However, the relative effect sizes for both η2 and ω2 values were smaller, suggesting that, as scores on the ACT factors were aggregated, our ability to assign the same meaning for the factor at the unit level as we had at the individual level lessened.
Experts generally disagree on the precise benchmarks that should be applied to psychometric measures such as alpha coefficients. In part, these benchmarks depend on the application. For example, lower Cronbach alpha coefficients (.70 - .80) are deemed acceptable for scales intended to compare groups, while for scales used to measure change within individuals, higher Cronbach alpha coefficients (> .90) are desired [21, 34]. The internal consistency of the ACT, in terms of the Cronbach's alpha coefficients of its core dimensions, is for the most part, consistent with usual practice for measures intended to be used at the level of the group, or in our case, the patient care unit [34, 38]. Only 4 of the 13 ACT factors identified in the factor analysis (structural and electronic resources - type 2 [traditional resources], structural and electronic resources - type 3 [electronic resources], formal interactions and organizational slack-space), had alpha coefficients less than this standard. These lower coefficients may be due to these items addressing concepts that are broader, and perhaps more subject to individual interpretation, than the items in the remaining context concepts.
Validity - Internal Structure
Factor loadings for all 56 items, with the exception of one item (informal teaching sessions, factor loading = .339) in the ACT exceeded the minimum cut-off of 0.35, indicating that items were representative of underlying factors. Some items did not, however, load as expected. The items on how often respondents participate in 'hallway talk' and 'informal teaching sessions' originally part of the dimension of formal interactions loaded with informal interactions. Further, the item on how often respondents participate in 'continuing education', originally part of the structural and electronic resources dimension, loaded with the formal interaction dimension. After careful consideration of these findings the team decided that these loadings were actually a more accurate reflection of the ACT dimensions (as defined in Table 1) that they loaded with and thus we relabelled the item groupings to align with the factor analysis findings.
In developing the ACT we originally hypothesized a 10-factor solution (eight contextual dimensions: culture, leadership, evaluation, social capital, informal interactions, formal interactions, structural and electronic resources, and organizational slack (representing three sub-concepts - time, space, human resources)) with items designated for each concept loading onto a single factor. However, two of our contextual concepts turned out to be multidimensional, loading onto more than one factor (thus yielding a 13-factor solution). Informal interactions had two factors and structural and electronic resources had three factors indicating more complexity to the construct of organizational context than we had originally proposed.
The validation process in this study demonstrated beginning empirical support for the construct validity of the ACT. Statistically significant bivariate relationships were found between all but one of the ACT's 13 factors at varying levels of instrumental research utilization. That is, higher levels of research utilization were aligned with more positive contextual conditions. Further analyses showed that the mean scores for each of the ACT's 13 factors varied consistently with a positive incremental association between them and reported research utilization levels. These findings are consistent with the PARiHS framework's assertions (see Table 1) and provide additional empirical support for the construct validity of the ACT.
Aggregation of individual level data to the unit level
Our aggregation statistics indicate that the ACT (when used with professional nurses) can reliably be aggregated to obtain a unit-level assessment of organizational context. We ran the same aggregation statistics with the allied healthcare professionals (n = 209) who completed the ACT survey in the study reported in this paper to explore this further. As we had anticipated, the aggregation statistics did not support aggregation of the ACT with the allied professionals at the unit level; ICC(2) was <.60 for all ACT dimensions indicating low reliability of group means when aggregated to the patient care unit level. Given the differences in how work is constructed for nurses and allied professionals this made sense to us. Most nurses perform their work on a single unit, are aligned with that unit and therefore are able to assess and report on common unit practices, beliefs and values causing them to respond similarly on items examining their patient care unit. Allied professionals such as physiotherapists and respiratory therapists, on the other hand, often work across programs (which consist of several units) and therefore should (and do) display greater within unit variability decreasing reliability of their aggregated response. Therefore, at this point in time we only recommend aggregating responses of professional nurses to obtain unit-level scores on the concepts contained within the ACT.
Although the validation results presented in this manuscript are promising, this initial assessment of the ACT was conducted in one country, with one professional group, and with a moderate size sample. Validation of a newly developed instrument such as the ACT is a longitudinal and multi-step process, requiring numerous positive findings, across a variety of applications and settings. Test-retest reliability was not assessed, so the ACT's stability is unknown. Cross-validation studies are needed to confirm the factor structure obtained in this initial field test and to establish the reliability and validity of the scales in other samples and settings. Additional validation studies using larger sample sizes will be undertaken as additional data are available; these will permit us to extend our assessment to include confirmatory factor analyses and hierarchical linear modeling.
Additional and much longer term investigation is also needed to explore whether overall scores for the multidimensional ACT concepts (e.g., structural and electronic resources, informal interactions, and organizational slack) can be derived from the instrument. At present we are using overall derived scores for each factor as supported by the factor analysis reported in this paper.
The findings from this initial validation of the ACT must be interpreted with caution and are not generalizable beyond the sample of nurses reported in this manuscript. Although the overall pattern of the data was consistent with the structure hypothesized in the development of the ACT, some items loaded onto their respective factors less strongly than others. This may indicate that respondents are conceptualizing the individual items within a particular concept somewhat differently than we anticipated. This was not totally unexpected, as for the ACT to be applicable to healthcare professionals across a variety of settings, its items were designed to address the respective dimensions as generically as possible. Preliminary work also suggests the instrument in its current form produces its best aggregated results at the unit level among professional nurses.
Follow-up studies are in progress in which we are assessing the ACT with nurses, allied healthcare professionals, physicians, educators and specialists, and managers in long-term care (nursing home) settings, as well as with unregulated (healthcare aide) workers in long-term care settings. Validation of the instrument within the home care sector is also planned. Additional information on the ACT is available from the lead author of this paper.
- Commonly used abbreviations in this manuscript include:
(1) (ACT): Alberta Context Tool
- and (2) (PCA):
Principal Components Analysis.
The Alberta Context Tool was developed as part of an Alberta-based study funded by the Alberta Heritage Foundation for Medical Research (AHFMR) (Co-Principal Investigators: CAE and PGN). The authors acknowledge the CIHR Team in Children's Pain (Dr Bonnie Steven's, Principal Investigator) for contributions to this study. Funding for the data used in the validation assessment of the ACT reported in this manuscript was provided by the Canadian Institutes of Health Research (CTP-79854 and MOP-86605). Dr Estabrooks is supported by a CIHR Canada Research Chair in Knowledge Translation. Ms Squires is supported by CIHR, AHFMR, and Killam Fellowships. Dr Cummings holds CIHR New Investigator and AHFMR Population Health Investigator awards. We would like to thank Sung Hyun Kang, MSc (University of Alberta, Canada) for his assistance with statistical analysis, Dr William Midodzi (University of Alberta, Canada) for his thoughtful input on statistical analyses and drafts, and Dr. Mandi Newton (University of Alberta, Canada) for her assistance with the development of the ACT.
- Rycroft-Malone J: The PARIHS framework - A framework for guiding the implementation of evidence-based practice. J Nurs Care Qual. 2004, 19 (4): 297-304.View ArticlePubMed
- Wallin L, Estabrooks CA, Midodzi WK, Cummings GG: Development and validation of a derived measure of research utilization by nurses. Nurs Res. 2006, 55 (3): 149-160. 10.1097/00006199-200605000-00001.View ArticlePubMed
- Dopson S, FitzGerald L, Ferlie E, Gabbay J, Locock L: No magic targets! Changing clinical practice to become more evidence based. Health Care Manage Rev. 2002, 27 (3): 35-47.View ArticlePubMed
- Meijers JM, Janssen MA, Cummings GG, Wallin L, Estabrooks CA, R H: Assessing the relationships between contextual factors and research utilization in nursing: Systematic literature review. J Adv Nurs. 2006, 55 (5): 622-635. 10.1111/j.1365-2648.2006.03954.x.View ArticlePubMed
- Grol R, Berwick D, Wensing M: On the trail of quality and safety in health care. BMJ. 2008, 336 (7635): 74-76. 10.1136/bmj.39413.486944.AD.PubMed CentralView ArticlePubMed
- Fleuren M, Wiefferink K, Paulussen T: Determinants of innovation within health care organizations: Literature review and Delphi study. Int J Qual Health Care. 2004, 16 (2): 107-123. 10.1093/intqhc/mzh030.View ArticlePubMed
- Greenhalgh T, Robert G, Macfarlane F, Bate P, Kyriakidou O: Diffusion of innovations in service organizations: Systematic review and recommendations. Milbank Q. 2004, 82 (4): 581-629. 10.1111/j.0887-378X.2004.00325.x.PubMed CentralView ArticlePubMed
- Grol R, Grimshaw J: From best evidence to best practice: Effective implementation of change in patients' care. Lancet. 2003, 362 (9391): 1225-1230. 10.1016/S0140-6736(03)14546-1.View ArticlePubMed
- Kitson A, Harvey G, McCormack B: Enabling the implementation of evidence based practice: A conceptual framework. Qual Saf Health Care. 1998, 7 (3): 149-158. 10.1136/qshc.7.3.149.View Article
- Rycroft-Malone J, Kitson A, Harvey G, McCormack B, Seers K, Titchen A, Estabrooks C: Ingredients for change: Revisiting a conceptual framework. Qual Saf Health Care. 2002, 11 (2): 174-180. 10.1136/qhc.11.2.174.PubMed CentralView ArticlePubMed
- McCormack B, Kitson A, Harvey G, Rycroft-Malone J, Titchen A, Seers K: Getting evidence into practice: The meaning of 'context'. J Adv Nurs. 2002, 38 (1): 94-104. 10.1046/j.1365-2648.2002.02150.x.View ArticlePubMed
- Schein EH: Organizational Culture and Leadership. 1985, San Francisco: Jossey-Bass Publishers
- Wright J, McCormack B, Coffey A, McCarthy G: Evaluating the context within which continence care is provided in rehabilitation units for older people. International Journal of Older People Nursing. 2007, 2 (1): 9-19. 10.1111/j.1748-3743.2007.00046.x.View ArticlePubMed
- Goleman D, Boyatzis RE, McKee A: The New Leaders: Transforming the Art of Leadership into the Science of Results. 2002, London: Little, Brown
- Rycroft-Malone J, Harvey G, Seers K, Kitson A, McCormack B, Titchen A: An exploration of the factors that influence the implementation of evidence into practice. J Clin Nurs. 2004, 13 (8): 913-924. 10.1111/j.1365-2702.2004.01007.x.View ArticlePubMed
- Estabrooks CA, Squires JE, Adachi AM, Kong L, Norton PG: Utilization of Health Research in Acute Care Settings in Alberta Technical Report. 2008, Edmonton: Faculty of Nursing, University of Alberta
- Acock AC: Working with missing data. Family Science Review. 1997, 10 (1): 76-102.
- Streiner DL, Norman GR: Health Measurement Scales: A Practical Guide to their Development and Use. 2003, Oxford: Oxford University Press, 3
- Thompson B: Exploratory and Confirmatory Factor Analysis: Understanding Concepts and Applications. 2004, Washington, DC: American Psychological AssociationView Article
- Kline P: An Easy Guide to Factor Analysis. 1994, London: Routledge
- Nunnally J, Bernstein I: Psychometric Theory. 1994, New York: McGraw-Hill, 3
- Betz NE: Test construction. The Psychology Research Handbook: A Guide for Graduate Students and Research Assistants. Edited by: Leong FTL, Austin JT. 2000, Thousand Oaks, CA: Sage Publications, 239-250.
- Angus J, Hodnett E, O'Brien-Pallas L: Implementing evidence-based nursing practice: A tale of two intrapartum nursing units. Nurs Inq. 2003, 10 (4): 218-228. 10.1046/j.1440-1800.2003.00193.x.View ArticlePubMed
- Hunt M: The process of translating research findings into practice. J Adv Nurs. 1987, 12: 101-110. 10.1111/j.1365-2648.1987.tb01308.x.View ArticlePubMed
- Davies S, McDonnell A, Brown J, Shewan J: Practice nurses' use of evidence-based research. Nurs Times. 1999, 95 (4): 57-60.PubMed
- Estabrooks CA, Midodzi WK, Cummings GG, Wallin L: Predicting research use in nursing organizations: A multilevel analysis. Nurs Res. 2007, 56 (4 Suppl): S7-23. 10.1097/01.NNR.0000280647.18806.98.View ArticlePubMed
- Ferlie E, Fitzgerald , Wood M: Getting evidence into clincial practice: An organisational behaviour perspective. J Health Serv Res Policy. 2000, 5 (2): 96-102.PubMed
- Estabrooks CA: The conceptual structure of research utilization. Res Nurs Health. 1999, 22 (3): 203-216. 10.1002/(SICI)1098-240X(199906)22:3<203::AID-NUR3>3.0.CO;2-9.View ArticlePubMed
- Estabrooks CA, Kenny DJ, Cummings GG, Adewale AJ, Mallidou AA: A comparison of research utilization among nurses working in Canadian civilian and United States Army healthcare settings. Res Nurs Health. 2007, 30 (3): 282-296. 10.1002/nur.20218.View ArticlePubMed
- Estabrooks CA, Scott S, Squires JE, Stevens B, O'Brien-Pallas L, Watt-Watson J, Profetto-McGrath J, McGilton K, Golden-Biddle K, Lander J, et al: Patterns of research utilization on patient care units. Implement Sci. 2008, 3: 31-10.1186/1748-5908-3-31.PubMed CentralView ArticlePubMed
- Glick WH: Conceptualizing and measuring organizational and psychological climate: Pitfalls in multilevel research. Acad Manage Rev. 1985, 10 (3): 601-616. 10.2307/258140.
- Rosenthal R, Rosnow RL: Essentials of Behavioural Research: Methods and Data Analysis. 1991, New York: McGraw Hill, 2
- Keppel G: Design & Analysis: A Researcher's Handbook. 1991, Englewood Cliffs, NJ: Prentice-Hall, 3
- Altman DG, Bland JM: Statistics notes: Units of analysis. BMJ. 1997, 314 (7098): 1874.PubMed CentralView ArticlePubMed
- Klein KJ, Conn AB, Smith DB, Sorra JS: Is everyone in agreement? An exploration of within-group agreement in employee perceptions of the work environment. J Appl Psychol. 2001, 86 (1): 3-16. 10.1037/0021-9010.86.1.3.View ArticlePubMed
- Vogus TJ, Sutcliffe KM: The Safety Organizing Scale: Development and validation of a behavioral measure of safety culture in hospital nursing units. Med Care. 2007, 45 (1): 46-54. 10.1097/01.mlr.0000244635.61178.7a.View ArticlePubMed
- Bliese PD: Within-group agreement, non-independence, and reliability: Implications for data aggregation and analysis. Multilevel Theory, Research and Methods in Organizations: Foundations, Extensions and New Directions. Edited by: Klein KJ, Kozlowski SWJ. 2000, San Francisco: Jossey-Bass, 349-381.
- Young TL, Kirchdoerfer LJ, Osterhaus JT: A development and validation process for a disease-specific quality of life instrument. Drug Inf J. 1996, 30 (1): 185-193.
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1472-6963/9/234/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.