OJNI

Information

This article was written on 30 Jun 2013, and is filled under Volume 17 Number 2.

Current post is tagged

, , , , , ,

A Current Review of the Benefits, Barriers, and Considerations for Implementing Decision Support Systems

by

Candice McCool, BSN, RN, CCRN

Citation

McCool, C. (June 2013). A Current Review of the Benefits, Barriers, and Considerations for Implementing Decision Support Systems. Online Journal of Nursing Informatics (OJNI), vol.17 (2), Available at  http://ojni.org/issues/?p=2673

 

Abstract

The overwhelming quantity of clinical practice guidelines (CPGs) requires a method of access that is efficient and effective.  Adherence to CPGs results in improved patient outcomes as measured by decreased morbidity, mortality, and health care costs while improving patient care and safety (Stenberg & Wann-Hansson, 2011).  Decision support systems (DSSs) provide a database for CPGs.  The use of DSSs improves adherence to evidence-based recommendations and decreases morbidity (Wright, Sittig, Ash, Sharma, Pang, & Middleton, 2009).  Barriers to DSS implementation include limited access to adequate technology, absence of encoding standardization between systems. Concerns regarding credibility, validity, reliability, and maintaining current, up to date CPGs will be examined.  Methods to overcome barriers will be presented.  The use of CPGs as a performance measure will be evaluated.  Successful implementation of DSSs will require early involvement of stakeholders, information technology support, ongoing progress reports and feedback, and an inter-operable system offering specialized support.  Ongoing research to determine the most successful DSS system will be necessary as use becomes more prevalent.

Key Words

informatics, clinical practice guidelines, decision support systems, evidence-based practice, patient outcomes

 Introduction

The use of evidence-based protocols in the form of clinical practice guidelines (CPGs) have been shown to improve quality of care and patient safety while decreasing medication errors and health care costs (Middleton, 2009).  Decision support systems (DSSs) contain a database of CPGs to support the nurse, researcher, or practitioner in formulation of an individualized plan of care that will result in the best outcomes for patients.  This topic is of interest because CPGs are not widely utilized or accepted by healthcare providers despite associated positive outcomes.  The use of a DSS improves adherence to CPGs (Boxwala, Rocha, Maviglia, Kashyap, Meltzer, Kim, et al., 2011).  The purpose of this paper is to express the issues associated with advancing healthcare informatics through progressive development and implementation of DSSs.

Analysis of Findings

Development of DSSs

Improved access to the abundant number of CPGs available can be accomplished through the use of a computerized DSS.  A DSS provides guidance in diagnosing and managing illness through its library of CPGs.  CPGs are formulated based on strategies that have been rigorously tested for validity, reliability, reproducibility, and cost effectiveness by practitioners and researchers who are experts in the particular area of focus (Donovan, 2011).  A challenge of maintaining current and up to date information within CPGs remains.

The development and use of DSSs remains rudimentary.  The computerized clinical guidance provided by DSSs performs most optimally when used with electronic health records, electronic medication administration systems, and computerized physician order entry (Wright et al., 2009).  DSS recommendations may be triggered by any clinical indicator that has been programmed into the system.  One very basic example: a lab result that is out of reference range in the electronic health record can prompt an alert to the provider through the decision support system.  Programming may be more complex: A drug search may be prompted based on the lab that is out of reference range to check for interactions or adverse reactions and a page sent out to the responsible physician for notification and to offer treatment options (Wright et al., 2009).

An on-demand DSS provides evidence-based guidelines derived from CPGs when sought out by the nurse, researcher, or practitioner.  This type of system also serves as an educational source to keep health care providers up to date in current practice standards (Genuis, 2005).  In contrast, there are critiquing DSSs that alert practitioners when their computerized order entry does not follow the recommendations in a CPG.  There are DSSs used to exclusively guide pharmacotherapy and address drug interactions and detection of adverse events.  Another modality of a DSS can provide support for treatment of acute and chronic disease processes as well as preventative care (Wright et al., 2009).  This can be achieved through “alerts, reminders, order sets, drug-dose calculations … or performance feedback on quality indicators” (Bright, Wong, Dhurjati, Bristow, Bastlan, Coeytaux, et al., 2012).  As the use of CPGs by health care providers increases due to the proliferation of DSSs, continuing research will determine the most efficient and effective system.

An insurmountable number of topics are detailed by CPGs.  DSSs have provided an efficient method of data management to improve accessibility to CPGs.  Development of DSSs will be continuous as further research and analysis evolve.  Development will need to focus on specializing DSSs to health care providers’ unique specialties.  DSSs can be utilized by nurses, researchers, and practitioners alike to align practice methods to evidence-based recommendations outlined by CPGs.  Bloomrosen and Detmer (2010) suggest that many stakeholders will need to be involved in the formulation of CPGs to improve acceptance and adherence of use by all health care providers at various levels of education.  Stakeholders should include users of CPGs as well as “computer programmers, behavioral scientists, usability experts, health informaticians, workflow experts, quality improvement specialists, and patients” (Bloomrosen & Detmer, 2010, p. 120).

Standardization

The process of incorporating CPGs into a DSS requires data to be modified into a format that can be used by the computer’s software systems.  CPGs must be encoded to trigger the DSS to complete the appropriate action in the appropriate situation.  One literature review found that some DSSs use numerical data while others used internationally encoded data of which there are two different varieties: International Classification of Disease (ICD) and Logical Observation Identifiers Names and Codes (LOINC) (Ahmadian, van Engen-Verheul, Bakhshi-Raiez, Peek, Cornet, & de Keizer, 2010).  Inconsistent encoding hinders the progressive development and use of DSSs.  A questionnaire distributed to authors of 77 controlled clinical trial articles that were included in a literature review revealed that 92% of respondents had previously decided not to start or to stop a DSS due to data standardization required for development (Ahmadian et al., 2010).  Standardization of encoding is fundamental to streamlining DSS use into today’s healthcare industry.

For a DSS to function effectively, the organization’s electronic health record system and supporting software must use compatible encoding (Boxwala et al., 2011).  Without standardization, each CPG will have to be encoded according to the individual programming needs of a particular facilities’ electronic health records system.  Creating this type of system requires the finite knowledge of experts and programmers to consistently update ever-changing CPGs within the facilities’ DSS.  The cost of creating and continually updating the DSS may outweigh the benefit for some facilities (Boxwala et al., 2011).

Encoding a CPG to be used within a DSS requires rigor.  The DSS must be encoded using a broad to specific approach to be most efficacious to its variety of users (Boxwala et al., 2011).  One effective method of formatting information into a functional database is a layered framework comprised of four different layers to achieve the broad to specific structure.  The layered framework is constructed using a “team comprising clinicians, programmers, analysts, knowledge engineers, and others” (Boxwala et al., 2011, p. 135).

Measuring Performance Based on DSS Recommendations

As CPGs are more consistently used through the improved accessibility provided by DSSs, health care providers’ performance may be measured by comparing their actions against guideline recommendations.  More research is needed to identify whether measuring performance based on adherence to DSS recommendations will be appropriate.  This is because some institutions may adopt the most recent evidence-based practice derived from well-designed controlled trials or systematic reviews prior to a CPG being updated (Lin, Redberg, Anderson, Shaw, Milford-Beland, Peterson, et al., 2010).  This may result in an inappropriate negative reflection of the provider’s or institution’s performance.

Additionally, a CPG recommended by the DSS may not be a perfect fit for each patient’s individual situation.  Individualized healthcare decisions, based on demographic data such as age for example, may require an alteration of the guideline recommendations.  Creating an individualized plan of care may again result in an inappropriate negative reflection of the provider or institution when measuring performance by recommendation adherence (Genuis, 2005).  This is an important issue when considering that performance improvement measurements, such as incentives and reporting, are purported to become obligatory in the healthcare industry (Lin et al., 2010).

The Institute of Medicine has endorsed reimbursement plans that parallel the quality of performance.  Additionally, “Medicare and Medicaid Services have initiated pay for performance initiatives to encourage improved quality of care” (Bloomrosen & Detmer, 2010, p. 118).  Medicare and Medicaid use the internet to report the quality of care provided by health care institutions to the public, and programs to compare the quality of care provided by individual physicians is currently begin developed (Baker, Qaseem, Reynolds, Gardner, Schneider, 2013).

Barriers to DSS Implementation

Some health care providers have contended that DSSs are unnecessary, that current practice methods are adequate and do not need to be changed, and that CPG research is inadequate (Genuis, 2005).  There are several causes for nurses, researchers, and practitioners to be reluctant about the use of CPGs.  First, there is difficulty ensuring that CPGs reflect the most current and reliable evidence.  This is because of the vast quantity of guidelines and the absence of a governing body.  Endorsement of a guideline by a professional organization, such as the Department of Health and Human Services’ Agency for Healthcare Research and Quality (AHRQ), may prompt acceptance for use (Abrahamson, Fox, & Doebbeling, 2012; Woo & Wynne, 2012b).

AHRQ is just one example of a leading professional organization.  AHRQ initiated the National Guideline Clearinghouse (NGC) which is an organization that appraises CPGs and shares their archive of endorsed guidelines in a pdf format provided to the public for free (Woo & Wynne, 201ba).  The NGC is sponsored by prominent organizations such as the American Medical Association and the American Association of Health Plans (National Guideline Clearinghouse, N.D.).  Although the NGC has reputable creators and sponsors, the organization has broad inclusion criteria allowing CPGs of variable levels of evidence to be endorsed and included in its archive (Development, identification, and evaluation, 2011).  Alternatively, the publicly funded health system of England, the National Health Service, has begun granting accreditation to organizations developing guidelines in an effort to allow users of CPGs quick appraisal and identification of high quality data (Mayor, 2009).  Ultimately, nurses, researchers, and practitioners must use their expertise and critical appraisal when using any CPG whether recommended by a DSS or endorsed by a professional organization due to the variability of standards for inclusion and endorsement.

For a CPG to be credible, it must be unbiased and rely on evidence.  One study revealed that less than half of organizations who listed at least five CPGs in the NGC had policies regarding conflict of interest (COI).  These organizations’ policies did not meet the standards of the Institute of Medicine for COI whereas the remaining organizations did not have a COI policy at all (Norris, Holmer, Burda, Ogden, & Fu, 2012).  Creators of CPGs must take steps to ensure that no conflict of interest exists in order for the CPG to be credible and validated for use in clinical practice.  One source suggests the use of an international registry with specific criteria for inclusion of CPGs to ensure credibility, validity reliability, and frequent updating to ensure the most current evidence is reflected (Genuis, 2005).

There have been occurrences of CPGs recommendations being withdrawn due to poor outcomes which has contributed to the reluctancy of nurses, researchers, and practitioners to use CPGs.  Genuis (2005) brings attention to two CPGs. The first focused on hormone replacement therapy post-menopause which has been shown to promote development of cancer.  The second CPG outlines anti-depressant use in adolescence which has been associated with worsening depression and increased risk of suicidality.  Both CPGs were supported by experts in the specialty area and were recommended for use in current practice.  HRT therapy was deemed to be safe and effective for long term negative effects at the time of recommendation based on numerous “large, well designed randomized controlled trials” (Genuis, 2005, p. 420).  Both CPGs were consequently withdrawn due to the negative effects of therapy.  Due to these types of occurrences, a lack of endorsement by an accrediting agency for each individual CPG will make the use of guidelines less widely accepted by healthcare providers (Abrahamson et al., 2012).

One source suggests that if CPG developers accept specific standards for guideline development such as the standards outlined by the Institute of Medicine, convergence of guideline recommendations would be achieved by organizations working together to use the best evidence to formulate one superior CPG (Development, identification, and evaluation, 2011).  Organizations creating inferior CPGs would no longer be able to perform to the specific standards and would either collaborate and contribute to one superior CPG or allow a larger organization capable of funding such an effort complete the task.  Successful convergence of guidelines will purportedly be achieved passively after organizations become increasingly aware of specific standards (Development, identification, and evaluation, 2011).

Another barrier that may hamper the acceptance of use of CPGs is variance in recommendations between organizations.  For example, rosiglitazone has been removed from CPGs detailing the treatment of type II diabetes authored by the American Diabetes Association and the European Association for the Study of Diabetes due to associated cardiovascular risk.  However, the Canadian Diabetes Association still recommends rosiglitazone for the treatment of type II diabetes stating that there is insufficient evidence to remove this medication (Woo & Wynne, 2012a).  When several organizations propose different approaches, professional judgment should be used in deciding if the CPG recommendations are appropriate for the individual patient’s circumstances (Woo & Wynne, 2012b).

The Endocrine Society includes a disclosure statement at the beginning of their endorsed guidelines.  The disclaimer states that intended outcomes are not guaranteed, guidelines are neither all inclusive nor do they represent the standard of care, and that judgment must be used to determine the needs of each individual provider and patient (Evaluation and treatment, 2012).  Similar disclaimers are found when reviewing CPGs of other organizations.  Such statements can prevent litigation when guidelines are appropriately chosen for use and negative outcomes occur.

Methods of Successful Implementation of DSSs

            Once an organization has decided to employ the use of a DSS, established methods for successful implementation should be incorporated in the strategy of initiation.  It has been identified that initial and ongoing success of DSSs requires the “knowledge, skill, and motivation” of influential users (Trivedi, Daly, Kern, Grannemann, Sunderajan, & Claassen, 2009, p. 8).  Early involvement of users, stakeholders, and administrators is essential to maintaining participation during and after implementation of DSSs. When changing from a previously used system to a DSS, it is recommended to abruptly shift rather than convert in phases; when users were asked to use both paper/pencil and electronic systems, they reported “frustration and confusion” (Trivedi et al., 2009, p. 5).  An adequate information technology support team is necessary to support users during the transition.  Other factors that offered predictable success of DSS implementation were progress reports and feedback, specialized decision support systems, and inter-operability between the DSS and the electronic health record system (Trivedi et al., 2009).

Implications for Practice

The effect of using protocols in the form of CPGs to manage acute and chronic healthcare conditions is improved patient outcomes (Stenberg & Wann-Hansson, 2011; Kollef & Micek, 2010).  The use of DSSs, which use the recommendations from CPGs, increases the frequency of guideline use by healthcare providers (Middleton, 2009).

Another benefit of increasing CPG usage is cost effectiveness of healthcare delivery (Genuis, 2005).  A defensive method of health care delivery may be used to curb litigation and is a cause of increased healthcare expenditure that can be decreased by the use of CPGs (Genuis, 2005).  CPGs have been used in the legal process and served as expert opinion.  This has led to irrefutable resolution of malpractice claims (Bovbjerg & Berenson, 2012).  By using CPGs, healthcare providers’ concerns of litigation should decrease and the care delivered should be more appropriate for the patient’s clinical condition.  Costs will decrease and quality of care will improve.

Additional benefits resulting from CPG usage include safety, accuracy of diagnosis, and “measurable improvements in patient care” (Middleton, 2009, p. 26).  Depending on the CPG chosen, different outcomes are measured.  For example, a CPG for fall prevention will measure the number of occurrences of patient falls and incurred injuries.  A CPG for mechanical ventilator weaning will measure days on the ventilator, intensive care unit length of stay, and incidence of ventilator associated pneumonia.  The surmounting evidence of benefits from using CPGs reveals improved quality of care measured also by “mortality, morbidity, and cost effectiveness” (Stenburg & Wann-Hansson, 2011, p. 88).

In one study, an institution formulated the cost of creating and implementing a shared DSS and assessed the impact it may have on a chronic disease, type 2 diabetes. The study resulted in slight improvements in short term risk and moderate risk reduction related to long term sequelae of type 2 diabetes. However, the DSS would need to be more efficient for cost savings to overcome expenditure further emphasizing the need for continued DSS development and standardization (O’Reilly, Holbrook, Blackhouse, Troyan, & Goeree, 2012).

Conclusions

            After reviewing the literature, it is implicitly obvious why it is important to advance informatics technology in the area of DSSs.  Continued development of DSSs will support users of CPGs by providing a more efficient method of delivery with the ultimate goal of improving patient outcomes and decreasing healthcare costs.  Many stakeholders must be involved to create useful, efficient, and specialized support systems.

Barriers to DSS implementation focus on limited access to adequate technology and a lack of acceptance of CPGs by users and their peers.  Health care providers’ reluctance to use CPGs is partially due to concerns regarding credibility, validity, reliability, and infrequent updating resulting in poor reflection of the most current evidence.  However, it may be improved by the creation of an international CPG registry and strictly using guidelines that are endorsed by accredited agencies.  The harm in measuring health care provider performance based on CPG adherence will require further evaluation as the use of guidelines advances.

The implementation of DSSs will be most effective by educating users regarding the effects of improved adherence to CPGs.  The implications of widespread CPG use include improved patient outcomes, cost effectiveness, safety, diagnosis, mortality, and morbidity.  Due to the overwhelming benefits of guidelines use, adherence can only be expected to increase.

References

Abrahamson, K.A., Fox, R.L., & Doebbeling, B.N. (2012). Facilitators and barriers to clinical practice guideline use among nurses. American Journal of Nursing, 112. Advance online        publication. doi: 10.1097/01.NAJ.0000415957.46932.bf

Ahmadian, L., van Engen-Verheul, M., Bakhshi-Raiez, F., Peek, N., Cornet, R., de Keizer, N.F.   (2011). The role of standardized data and terminological systems in computerized clinical  decision support systems: Literature review and summary. International Journal of Medical Informatics, 80, 81-93. doi: 10.1016/j.ijmedinf.2010.11.006

Baker, D.W., Qaseem, A., Reynolds, P., Gardner, L.A., Schneider, E.C. (2013). Design and use   of performance measures to decerase low-value services and achieve cost-conscious care.    Annals of Internal Medicine, 158, 55-59. doi:10.7326/0003-4819-158-1-201301010-00560

Bloomrosen, M., & Detmer, D.E. (2010) Informatics, evidence-based care, and research;   implications for national policy: A report of an American Medical Informatics Association health policy conference. Journal of American Medical Informatics Association, 17, 115-123. doi: 10.1136/jamia.2009.001370

Bovbjerg, R.R., & Berenson, R.A. (2012). The value of clinical practice guidelines as malpractice “safe harbors.” Quick Strike Series, 3. Urban Institute. Retrieved from             http://rwjf.org/files/research/74237.quickstrike.safeharbors04.20.12.pdf

Boxwala, A.A., Rocha, B.H., Maviglia, S., Kashyap, V., Meltzer, S., Kim, J., Tsurikova, R., Wright, A., Paterno, M.D., Fairbanks, A., and Middleton, B. (2011). A multi-layered   framework for disseminating knowledge for computer-based decision support. Journal of American Medical Informatics Association, 18, 132-139. doi; 10.1136/amiajnl-2011-000334

Bright, T.J., Wong, A., Dhurjati, R., Bristow, E., Bastian, L., Coeytaux, R.R., Samsa, G., Hasselblad, V., Williams, J.W., Musty, M.D., Wing, L., Kendrick, A.S., Sanders, G.D., Lobach, D. (2012). Effect of clinical decision-support systems: A systematic review.  Annals of Internal Medicine, 157, 29-43. Retrieved from http://annals.org/article.aspx? articleid=1206700

Development, identification, and evaluation of trustworthy clinical practice guidelines. (2011). Clinical Practice Guidelines We Can Trust. Washington, DC: The National    Academies      Press. Retrieved from http://www.nap.edu/openbook.phprecord_id=13058&page=187

Donovan, N. (2011). Everyone benefits from guidelines. Journal of Women’s Health Physical Therapy, 35, 43-44. doi: 10.1097/JWH.0b-013e318225f734. doi: Genuis, S.J.

Evaluation and treatment of hypertriglyceridemia: An endocrine society clinical practice   guideline. (2012). Journal of Clinical Endocrinology and Metabolism, 97, 2969-2989. Retrieved from http://www.endo-society.org/guidelines/upload/082312_Hypertriglyceridemia_FinalA.PDF

Genuis, S.J. (2005). The proliferation of clinical practice guidelines: Professional development or medicine by numbers? The Journal of the American Board of Family Practice, 18,          419-425. doi: 10.3122/jabfm.18.5.419

Kollef, M.H., Micek, S.T. (2010). Using protocols to improve patient outcomes in the  intensive care unit: Focus on mechanical ventilation and sepsis. Seminars in     Respiratory and Critical Care Medicine, 31, 19-30. doi: 10.1055/s-0029-1246286

Lin, G.A., Redberg, R.F., Anderson, H.V., Shaw, R.E., Milfort-Beland, S., Peterson, E.D., Rao, S.V., Werner, R.M., Dudley, R.A. (2010). Impact of changes in clinical practice guidelines on assessment of quality of care. Medical Care, 48, 733-738. doi: 10.1097/MLR.0b013e3181e35b3a

Mayor, S. (2009). NICE and SIGN given accreditation as guideline developers. British Medical   Journal, 339, 4388-4389. doi: 10.1136/bmj.b4388

Middleton, B. (2009). The clinical decision support consortium. Medical Informatics in a   United and Healthy Europe. Advance online publication. doi: 10.3233/978-1-60750-044-5-26

Norris, S.L., Holmer, H.K., Burda, B.U., Ogden, L.A., & Fu, R. (2012). Conflict of interest policies for organizations producing a large number of clinical practice guidelines. PLoS           ONE, 7(5), 1-12. doi: 10.1371/journal.pone.003741

O’Reilly, D., Holbrook, A., Blackhouse, G., Troyan, S., & Goeree, R. (2012). Cost-effectiveness of a shared computerized decision support system for diabetes linked to electronic       medical records. Journal of American Medical Informatics Association, 19, 341-345. doi:10.1136/amiajnl-2011-000371

Stenberg, M., Wann-Hansson, C. (2011). Health care professionals’ attitudes and compliance to   clinical practice guidelines to prevent falls and fall injuries.    Worldviews on Evidence-   Based Nursing, 8, 87-95. doi: 10.1111/j.1741-6787.2010.00196.x

The National Guideline Clearinghouse. (n.d.). About. Retrieved from http://guideline.gov/about/   index.aspx

Trivedi, M.H., Daly, E.J., Kern, J.K., Grannemann, B.D., Sunderajan, P., & Claassen, C.A. (2009). Barriers to implementation of a computerized decision support system for        depression: An observational report on lessons learned in “real world” clinical settings.  BMC Medical Informatics and Decision Making, 9, 1-9. doi: 10.1186/1472-6947-9-6

Wright,A. Sittig, D.F., Ash, J.S., Sharma, S., Pang, J.E., & Middleton, B. (2009). Clinical decision support capabilities and commercially-available clinical decision support systems. Journal of American Medical Informatics Association, 16, 637-344. doi:10.1197/jamia/M3111

Woo, T.M., & Wynne A.L. (2012a). Drugs affecting the endocrine system. In M. Robinson & A.L. Wynne (Eds.), Pharmacotherepeutics for nurse practitioner prescribers (pp. 571-644). Philadelphia, PA: F.A. Davis Company.

Woo, T.M., & Wynne A.L. (2012b). Information technology and pharmacotherepeutics. In D.L. Ewing (Ed.), Pharmacotherepeutics for nurse practitioner prescribers (pp. 127-142).         Philadelphia, PA: F.A. Davis Company.

Author Bio

Candice McCool is a registered nurse with certification in critical care. She is pursuing an advanced practice nursing degree.  Candice has worked in leading institutions to assist them in implementing informatics technologies and clinical practice guidelines.  She also serves on committees for evidence based practice, performance improvement, staffing methodology, and participates in root cause analysis for improved patient safety and systems redesign.  Her career has provided her with the opportunity to assess the benefits of and barriers to implementing informatics technologies to support patient care. 

Back to Issue Index

Back to Issue Index

Be Sociable, Share!

Comments are closed.