Abstract
Introduction
Despite evidence of the effectiveness of preventative services and the development of published national guidelines1,2 rates of preventative health care services nationwide remain low. 3 Electronic clinical reminders (ECRs) have been shown to be effective in primary care to increase certain preventative services.4,5 However, studies have shown that ECRs have had uneven success, generally because of human factors rather than technical reasons, including provider workload, overwhelming number of reminders for each patient, lack of coordination between doctors and nurses, decrease in quality of doctor–patient interaction, and reminder inflexibility.6-11 In an environment where provider time is limited, health care systems must be streamlined to increase preventative screening measures during visit time. In a small case study in Alaska, one health care system has been able to increase screening on a number of preventative health care measures using ECR.
In 2009, the Kodiak Area Native Association (KANA)* adopted the use of ECR. As of 2012, KANA had deployed more than 40 reminders. From documentation aids (pregnancy education, pain agreement) to grant-based documentations (elder fall risk) to evidence based screening recommendations (colorectal cancer screening, HIV screening), KANA has used ECRs to efficiently broaden its services of appropriate preventative care.
Methods
KANA uses the Resource and Patient Management System as its health information technology platform, with support from Indian Health Service (IHS). The Resource and Patient Management System Electronic Health Record (EHR), is intended to help providers manage all aspects of patient care electronically by providing a full range of functions for data retrieval and capture to support patient review, encounter, and follow-up.
KANA has an in-house Clinical Applications Coordinator (CAC)/Medical Informaticist who is responsible for maintaining a comprehensive electronic health information and reimbursement system through ownership and coordination of health information technology such as EHR and other programs. The CAC also provides training to facility staff as needed.
Electronic clinical reminder in an EHR can be been designed and deployed at the local or national level. National-level reminders must be activated by the facility, and must ensure that local taxonomies and codes are integrated into the reminder, but generally they represent an “off the shelf” resource for IHS facilities. Of the 40 ECRs in use, 27 are evidence based,12,13 of which KANA prioritized a subset of 5 among the indicators that IHS tracks nationally: tobacco use, alcohol use, depression, intimate partner violence (IPV), and a comprehensive cardiovascular exam (documentation of blood pressure, body mass index, smoking status, low-density lipoprotein cholesterol level, and education of exercise/nutrition). Patient eligibility for screening and completed screenings are defined with standardized International Classification of Diseases, Logical Observation Identifier Names and Codes, and Current Procedural Terminology codes. 14
Results
Implementation of Clinical Reminders
As part of a national improvement collaborative sponsored by IHS and facilitated by the Institute for Healthcare Improvement, KANA uses a 5-step process for implementation of ECR through the use of plan–do–study–act cycles.15,16
1. Establish Clinical Demand for the Reminder
KANA uses 2 steps to gauge and foster provider demand. First, share the evidence-based guidelines so providers understand the rationale for the screening. KANA only uses ECR for established screenings that are backed by a credible national body such as the US Preventive Services Task Force. Second, share local data that shows that current screening rates need to be improved. KANA generates their screening data from their own patient data. Establishing clinical demand is part of building a will for change, and is a necessary but not sufficient factor for the successful rollout of an ECR.
2. Pilot Test Reminder
KANA develops local ECRs or deploys ECRs from the national IHS reminder “library.” After pilot testing, some reminders need the age range and screening intervals to be modified based on local data, whereas other reminders can be used locally without modification. All 5 prioritized indicators were targeted using national reminders that needed minor or no local modifications.
The CAC is the lead for all technical aspects of ECR. The CAC is based within the clinic, which allows for timely modifications to the reminders as needed, often same-day. The CAC deploys the new ECR for a small group of providers who have agreed to pilot test the reminder. Providers give feedback to the CAC via e-mail or informal updates; the CAC is often on the medical floor and will seek out providers who are piloting a new ECR for feedback. This process ensures that the ECR has been field-tested and reduces the false positives of a flawed reminder that results in an ECR being ignored for perceived inaccuracy.
KANA has all ECRs in one centralized location on the EHR, under the graphical user interface of an alarm clock that is red if any reminders are due for that patient. No ECRs are delivered using an active pop-up strategy. A provider may exit all reminders; compliance is not required.
3. Expand Reminder to All Providers in Health Facility
There is a scaling up of the process in step (Table 1), whereby all relevant providers are included. Feedback and modification are still needed or the ECR risks losing momentum.
Key Steps and Timeline in ECR Reminder Deployment, KANA Clinic.
Abbreviations: ECR, electronic clinical reminder; KANA, Kodiak Area Native Association.
4. Measure Outcomes and Share Results
The CAC keeps providers informed about results on the indicator associated with the new ECR. After the ECR has been deployed for a period, the screening results are broken down from aggregate to provider-specific rates. This can spur healthy competition and sharing of ideas among providers. It is a judgment call at what point the results are individually identifiable—if it is too early, it will produce distress rather than adaptive change, but if it is too late it will have lost momentum.
5. Delegation of Clinical Reminders to Other Staff
To the extent possible, responsibility for responding to ECRs and completing all due preventive screenings should be delegated to the appropriate staff level. This prevents the provider from being overwhelmed with reminders. KANA currently has 40 reminders deployed. Most are delegated to the nursing or medical assistant level, and only referred to the provider if the screen result is positive. When an ECR is due, it is clear who on the medical team (medical assistant, nurse, or provider) is responsible to ensure the screening is completed. This step was usually initiated 4 to 8 weeks after the beginning of the process.
Evaluation of Reminders
KANA deployed the national ECR for tobacco, alcohol, depression, and IPV screening in October 2009 and made local modifications based on outcomes of the pilot period. The cardiovascular disease reminder was deployed in January 2010.
An audit of the cardiovascular disease measure found that almost all incomplete screenings was due to only 1 of the 5 requirements to meet that screening (patient education), and the reminder targeted that element.
An analysis of results from reporting year 2007 to reporting year 2011 showed that for all 5 measures, KANA started with significantly lower rates and ended with significantly higher rates than IHS sites nationwide (Figure 1).
Screening rates for KANA and all IHS by year and test.
Discussion
Electronic clinical reminders have been a key part of improving screening for depression, tobacco cessation, IPV, alcohol use, and cardiovascular disease. They are valuable tools in identifying patients who are overdue for preventive screenings. While many ECRs are in use at KANA, overload is avoided by delegating responsibility for reminder follow-up away from providers. In addition, KANA has experienced a virtuous cycle: reminders increase screening rates, which in turn reduce the number of screenings that are due for any given patient. Providers anecdotally report that their reduced screening burden gives them more time during a consultation to build relationships with patients.
The KANA model for deployment concentrates on both technical soundness as well as the “human factor” of provider acceptance and use. The model relies on an ongoing improvement process: establishing clinical demand with local development, pilot testing with provider feedback and modification, expanding to all providers, measuring outcomes and sharing information with staff, and delegation of clinical reminders. KANA’s success is also partly attributable to ease of use; all ECRs are in one spot on the EHR for the provider to use or bypass as needed.
These data have some limitations. Although the successful deployment of ECR has been instrumental in screening improvements in these 5 measures, other variables are important. KANA is a setting with less than 10 providers, limited turnover, and a clearly defined user population. The facility hired its full complement of providers in 2008, reducing the patient load for each provider, and reducing the wait time for an appointment from 1 to 2 weeks to same day. The improvements in continuity of clinical staff, patient access to care, and team-based care sharing responsibility for preventive care are conditions that must be met prior to implementation of this ECR deployment process. The process may be difficult to replicate in health facilities that are understaffed, have high provider turnover, or it may need to be more formalized in a larger setting.
The cycle of measuring outcomes and sharing results is ongoing for continuous improvement in the delivery of care. Next steps and directions have been identified at staff to continue to strengthen and improve primary health care. KANA continues to seek ways for further enhance preventative health screenings to >90%.
Footnotes
Acknowledgements
The authors would like to thank Tom Sequist, Howard Hays, Erika Wolter, and Juanita Lee for their contributions to this article.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Indian Health Service National HIV Program.