Systematic review
Effect of the organisational development tool Appreciative Inquiry
Systematic review
|Updated
The aim of this review is to answer whether AI was more effective than other organizational development methods during a process of change in an organization.
Key message
The Norwegian Knowledge Centre for the Health Services was commissioned by The Regional Health Authority of South-Eastern Norway, Unit for Service Development and Cooperation to summarize available research on the effect of the organisational change methodology Appreciative Inquiry (AI). The aim of this review is to answer whether AI was more effective than other organizational development methods during a process of change in an organization. Even though we wished to focus on changes in the health services, we did not restrict the outcomes, where the intervention had taken place or what kind of organisational change that was studied.
We searched for controlled studies of effect both in medical and social electronic databases and identified 367 references. We included the six studies that had a control group. All were controlled before and after studies.
The included studies were conducted in different enterprises, a ward in a hospital in England, US Postal Services, a chain of fast food restaurants, a manufacturer of freight elevator doors, a trucking company, all in the USA, and a group of students in Canada. Several of the studies had more than one outcome, but none had measured an outcome in the same way. The outcomes comprised absence due to sickness, turnover, attitudes toward colleagues that make mistakes, conflict management, task quality, trust in the recourses of the group and a wish for future cooperation.
We assessed the studies to have an unclear or high risk of bias. The quality of documentation for effect of AI was very low, and we cannot draw clear conclusions. Some of the included studies reported that AI seemed to be more efficient than other organisational development tools, others not. Some of the included studies also reported that AI sometimes was not more efficient than not using any development tools. Future studies on the effect of AI should be larger and of better quality than the identified studies. The elements of AI that is the focus of research should be clearly specified and the outcomes more precisely defined.
Summary
Background
Appreciative Inquiry (AI) as a developmental tool is presented as a way of thinking, an organizational change methodology and management instrument which focuses on elements that works well in the organisation and which inspire the employees to work there. By focusing on the positive elements, it is claimed that the development of the organization would be better than if a more traditional organizational development tool had been used.
The aim of this review was to answer if the organizational development tool AI is more effective than other organizational development tools in a process of organizational change. Organizational change in this project was defined very wide and includes quality work and mergers in addition to change. The experiences of the consultants managing the organizational change about their own situation were not included in this project. The outcomes we sought comprise cooperation, quality of task or use of resources. We wanted to include studies performed in the health sector. However, we did not have limitations as to where the study was conducted, sector or type of enterprise or what type of intervention that was performed.
The Norwegian health system has for a long period of time been subject to a number of organizational changes, large and small, local and national. During the last ten years there have been several fundamental health reforms in Norway. If there was an organizational development tool that can be used when implementing these reforms, this could simplify the implementation of the various reforms. Resources might be saved and the process might be more predictable. A common general tool would imply that a larger number of people and institutions in the health system could implement the reform, and it could also facilitate the achievement of the goals of the reforms.
Methods
We searched systematically for literature in the following data bases; The Cochrane Library, MEDLINE, EMBASE, PsychINFO, Ovid AMED, Ovid British Nursing Index, CINAHL, CSA ERIC, CSA Social Services, CSA Sociological Abstracts. The search for literature was performed 17.10.2008 and updated 28.6.2009.
The inclusion criteria were:
· Population: Organisations that goes through a development process
· Intervention: Appreciative Inquiry used as an organizational development tool. Only the concept Appreciative Inquiry should be used, not other tools based on the same understanding of organizational change.
· Comparison: All other organizational development tools, or no intervention.
· Outcome: All that can express results of organizational change i.e. task quality, work environment, cooperation and productivity.
· Study design: Systematic reviews, randomised controlled studies, controlled before and after studies, interrupted time series
The results of the searches for literature were assessed by two persons independently. The risk of bias of the included studies was assessed and the quality of documentation was assed using GRADE.
Results
We found 367 references, none of them were systematic reviews. We included the six primary studies that had a control group. All were controlled before and after studies. The six included studies were assessed to have unclear or high risk of bias. The quality of documentation for the outcomes, assessed using GRADE, was very low. The quality of available documentation is too low to conclude whether AI is more or less efficient than other organizational development tools or if the organization did not use any such tool.
A short description of the results:
· At a ward in a hospital in England where the staff was interviewed according to the AI principles, it was reported similar levels for recruitment and possibly fewer absences due to sickness compared to a ward that had no intervention.
· In a trucking firm in the USA there was no statistically significant difference in attitudes towards their colleagues that facilitate forgiveness toward colleagues that make mistakes among the employees of a department four months and two years after an AI-conference in comparison with a department that did not participate in an AI-conference.
· Students in Canada that received an intervention more similar to action research scored higher on task performance than students that used AI, and both scored higher than students that received placebo, i.e. an expert presentation, on conflict management. There were no statistically significant difference between the group that received AI and expert presentation concerning participation in the group.
· There was no statistically significant difference between groups of employees in the US Postal Service that received AI compared to groups that used teambuilding or groups which received no intervention.
· Fast food restaurants in the USA that used AI did not retain a statistically significant larger part of its leaders than restaurants that used a problem solving approach or no intervention.
Groups in a company manufacturing freight elevator doors in USA that used AI achieved better task quality and more identification with the group than groups that used a more traditional organisational development tool did after two weeks after the intervention.
Discussion
In this project we have limited inclusion to interventions that have used the concept Appreciative Inquiry. Any other similar tools have not been included.
The authors of articles on AI criticise traditional organisational development tools and action research on a general basis, but other organisational development tools achieved better or similar results compared to AI for some outcomes. The critique was not confirmed by the results of the studies included in this review. Nor could we conclude that any of the tools do better than other tools.
AI comes across as complex and vague, and to evaluate the importance of single elements of it was difficult. Weaknesses that were reported by some authors when using AI was that difficult personal problems were overlooked, dissatisfied members withdrew, challenges are not addressed and in hindsight one appeared as naïve. It was also difficult to connect changes in the enterprise to an internally motivated change and not to external factors. This applies to all organizational development tools, not only AI. The developers of AI claim that the first question asked in a process of change is very significant. However, it was unclear how information can be collected to be able to ask this question.
How distribution of power in the enterprise will influence the choice of goals of an organizational change were not addressed, neither was the possible significance of a clever and enthusiastic consultant regardless of which tool that is being used. AI emphasizes the used of the positive factors in the enterprise, but none of the studies did clarify which positive factors the enterprises wished to further develop.
Conclusion
Current published literature cannot answer whether AI achieve better or worse results during an organisational development process compared to other organisational development tools or when no such tool is being used.