Developing a taxonomy for choosing which system development methodologies or strategies an organization will use to decide of which information system to implement is much like deciding which equations one might choose to solve an algebra problem. While there are many ways to arrive at a solution, there are many rules to choose from. Many variables must be collected and evaluated before deciding on an optimal methodology or strategy.
According to Whitten & Bentley (2007, p. 92), three considerations need to be analyzed prior to deciding on a system development methodology or strategy. The three considerations are:
- If the methodology is product-driven, is a prototype to be built or is code going to be written?
Organizations using this evaluation method will be better positioned to decide on which system development methodology or strategy to use when implementing an information system. Where this strategy fails is that is does not consider the information system’s return on investment (ROI), employee usability, alignment with corporate culture, or reasoning behind the purchase. It is also important to conduct a thorough investigation of the viability of the information system companies that are under consideration, by evaluating publicly available information.
Whitten & Bentley (2007, pp. 77-78) also suggest conduction the PIECES Problem-Solving Framework Checklist, prior to deciding on a methodology. The PIECES method includes:
Whitten & Bentley (2007, p. 94-104), suggest the following additional methodologies:
System Development Methodologies for a Large Community Hospital
Model-Driven Development Strategy
As a large community hospital is likely to have a host of existing systems, the model-driven development strategy would be a good systems development approach. According to Whitten & Bentley (2007, p. 94), the model-driven development strategy includes more commonly known methods that may make implementation easier than other methods. This method requires a good deal of planning and when implementing a new system, this is a good thing. A model-driven development strategy also allows for more thorough documentation, easier validation and conceptualization as models are visualized, and better built systems as they are more thorough diagrammed the first time compared to other strategies (Whitten & Bentley, 2007, p. 96).
System Development Methodologies for a Private Physician Practice
The Commercial Application Package Implementation Strategy
Regardless of the EHR System a private physician practice decides to implement, in most cases some degree of customization is likely to occur. But for the most part, existing commercial EHR systems will be ready for small practices right out of the box. A CAPIS strategy would be an ideal solution as this allows for the submission of requests for proposal (RFPs) to reviewed by the physician practice (Whitten & Bentley, 2007, p. 101). CAPIS allows for rapid implementation as the heavy programming has already been done by the EHR vendor, requires less “in-house” development knowledge and expertise, and allows for a more cost effective solution as the system can be duplicated to multiple users (Whitten & Bentley, 2007, p. 103).
Agile Method/Spiral Model
EHR Vendors may also benefit from the use of the spiral model to decrease the risks associated with software development (Houston, 2011, p. 45). As prototypes are developed, risk analysis occurs iteratively, so the product develops on an evolving basis (Shelly & Rosenblatt, 2012, p.26). The agile methodology allows for rapid development and testing, and also allows for flexibility and responsiveness by developers (Shelly & Rosenblatt, 2012, p.26). The agile methodology may pose some challenges to EHR Vendors who are not familiar with the methodology. The Project Management Institute (PMI) offers an agile certification through its PMI Agile Certified Practitioner (PMI-ACP) certification (PMI, 2012). EHR vendors that utilize agile methodologies would benefit from acquiring professionals who possess this certification.
Other barriers to effectively implementing an agile methodology include a lack of structure that can introduce greater risk, increased scope creep, and an increased level of communication by all team members (Shelly & Rosenblatt, 2012, p. 26). Houston (2011, p. 45) argues that increased risk is not associated with an agile method/spiral model, as risk analysis can be conducted iteratively (see Figure 5-2).
EHR Vendors would also benefit from incorporating this methodology. As the various objects emerge when planning a new technology, having an O-O methodology will help when breaking down responsibilities to various in-house team members.
Implementing a new technology is a big undertaking for any organization. For some leaders compiling too much data can result in a “decidophobia,” a clinical fear of reaching a decisions (Useen, 2010, p. 514). Leaders that suffer from decidophobia can take advice from the U.S. Marine Corp and adopt a 70 percent solution, which states that if a leader “has 70 percent of the information, has performed 70 percent of the analysis, and feels 70 percent confident, he or she is instructed to decide” (Useen, 2010, p. 514).
Houston, S. M. (2011). The project manager’s guide to health information technology implementation. Chicago, IL: HIMSS.
Project Management Institute. (2012). PMI-Agile Certified Practitioner (PMI-ACP)SM Pilot Program. Retrieved August 22, 2013 from http://www.pmi.org/certification/new-pmi-agile-certification/pmi-agile-certification-pilot-program.aspx
Shelly, G. B., & Rosenblatt, H. J. (2012). System analysis and design (9th ed.). Boston, MA: Course Technology - Cengage Learning.
Useem, M. (2010). Decision making as leadership foundation. In N. Nohria & R. Khurana (Eds.), Handbook of leadership theory and practice (pp. 507-525). Boston, MA: Harvard Business Press.
Whitten, J. L., & Bentley, L. D. (2007). Systems analysis and design methods. (7th ed.). New York, NY: McGraw-Hill/Irwin.
This work is licensed under a Creative Commons Attribution-NonCommercial- ShareAlike 3.0 United States License.
Christopher M. Bell