Best Practices for Problem Lists in an EHR

Problem lists can be a healthcare organization’s best friend or worst nightmare. Electronic record systems can extend the benefits or compound the issues. The same forces that make problem lists valuable for patient care and secondary data uses can also create barriers to clinical efficiency.

This practice brief explores the use and maintenance of problem lists in health records with a special focus on how electronic environments support additional functionality for sharing information and supporting continuity of care.

What Is a Problem List?

Computer-based hospital information systems emerged in the late 1960s. At that time they were primarily used for collecting and routing orders and accessing laboratory test results. These early systems collected clinical information, but their major purpose was to capture charges to comply with reimbursement requirements, not assist physicians with delivering patient care.

Dr. Lawrence Weed introduced the problem-oriented medical record more than 40 years ago.¹ It was a new concept for health record keeping that reflected the physician’s logical thinking for delivering patient care. Weed suggested that the primary organization of the medical record should be by medical problem and that all diagnostic and therapeutic plans be linked to a specific problem.

Historically, the information retrieved from a patient-physician relationship centered on what is termed the “problem list.” In many settings the problem list has evolved into a virtual table of contents in an EHR presenting a holistic view of the patient. In the problem-oriented medical record model all data associated with the patient can be linked to a list of problems.²

A variety of authoritative sources offer problem list definitions (see appendix A in the online version of this article in the HIM Body of Knowledge). For the sake of this practice brief, a problem list is defined as a compilation of clinically relevant physical and diagnostic concerns, procedures, and psychosocial and cultural issues that may affect the health status and care of patients. This information should identify the date of occurrence or discovery and resolution, if known.

Safe and efficient patient care relies on a clinical workflow that assesses problems, documents interventions, and evaluates the effects of treatment. The problem list is expected to support these activities in an effective and concise fashion. Technology offers the opportunity to achieve this goal while retaining information across the healthcare continuum and reducing redundant processes.

Problem List Challenges

There is no shortage of challenges when implementing and maintaining a problem list in the EHR. However, multiple resources are available to provide solutions for success. (A resource list of additional research and guidance is available in appendix C online.) The following issues and barriers to the effective use of problem lists within EHR systems are consistent throughout literature review and informal anecdotal information.

Time and Effort

The amount of time and resources required to document and maintain a problem list presents a substantial barrier. Studies dating back to 1996 cite concerns with the speed of entry and limited physician acceptance of electronic healthcare documentation systems.³ Clinicians frequently complain that a system requires too many clicks to get to the problem list and that using the documentation system takes too much effort and cuts into patient time.

The amount of human resources required to create and maintain the patient problem list is often significant. Healthcare providers and organizations should plan ahead to include the number of hours required to support this feature in clinical and administrative workflows in EHR systems.

Search Functionality for Providers

Physicians who are accustomed to writing down their diagnosis on paper problem lists in some cases are now required to electronically search through thousands of options to retrieve the specific diagnosis needed to accurately describe the patient’s disease process. This frustrating process may lead to incorrect or nonspecific data in a patient’s medical record due to inefficient search capability.

For example, a physician may see a patient for an acute sore throat and type “sore throat” into the search field. The system retrieves 54 options, including everything from acute gangrenous pharyngitis to streptococcal sore throat. The physician may try to limit the search to include only phrases that link to ICD-9-CM codes but still retrieves more than 33 options. EHR search functionality varies widely, and the retrieval and storage of problems remains an issue.

In another example, a physician sees a child with otitis media and types “otitis media” into the search field. The system retrieves 106 options, including acute allergic mucoid otitis media and tubotymphanic otitis media of the right ear. Once again, when the physician limits the search to diagnoses linked to an equivalent ICD-9-CM code, the system retrieves 99 options.

These inefficiencies can be remedied by placing a common list of acute illnesses into a favorites list. These lists can be populated by each physician personally or at an administrative level. However, users must still consider issues surrounding misspellings, acronyms, and abbreviations. Healthcare organizations must balance the ability for physicians to choose the diagnosis with the most specificity while still completing documentation in a timely manner.

Multiple Uses and Needs

The problem list serves a variety of uses for clinicians and nonclinicians in diverse healthcare settings. It can provide a succinct view of a patient’s health status and therefore must be used and maintained to meet different needs. A primary care physician is concerned with chronic and acute conditions. A specialty provider may focus only on a subset of problems relevant to that area of medicine. An emergency provider may address only the critical acute presenting problems. Other clinicians may use the problem list for tracking conditions that should be addressed for specific care delivery goals.

Coding professionals use the problem list to confirm or clarify documentation found in other parts of the health record. To address billing needs, the problem list may be used as a source of diagnostic information. Problem list entries may be linked or integrated with other parts of the EHR to minimize duplication and improve documentation. Key information about a patient (e.g., drug-seeking behavior or other pertinent facts affecting care or treatment) may be added to the problem list as a means of communication to all providers, since the problem list is intended for review at every patient encounter.

Although problem lists have typically been used as a source of information for clinical care, in an electronic environment this section of the record is considered a source for secondary data use including quality measurement initiatives and research. Disparate needs present challenges when creating and maintaining the problem list.

Specialty and emergency providers may address and update only problems relevant to a single episode of care, placing accountability and maintenance of the problem list on the primary care provider. Clinicians may not always specify problems at the diagnostic level necessary for accurate billing, yet problems may be used as the basis for diagnosis codes on claims to communicate reason for care and justification of medical necessity.

Including psychosocial, cultural, and other key information about the patient may lengthen the problem list and compromise its utility as a quick reference about the patient’s overall health condition.

Administrative Maintenance

Administrative management issues surrounding an EHR problem list must be resolved during the mass customization of a data entry platform. The primary challenge is to provide tools that balance patient care with the taxonomy of data standardization. The flexibility required by a variety of clinicians to document complex, patient-oriented care is at odds with the structure required by a standardized set of terminology describing discrete conditions, events, and measurable outcomes.

Multiple coding schemes can satisfy some administrative issues so that secondary data use cases are satisfied. Encoded results can be cross-referenced with frequently used entries placed at the front of the look-up table, resulting in faster identification of acceptable entries. The choices presented by the variety of clinical settings and their array of problems and diagnoses require that administrators work with application developers and clinical users to develop the cross-references needed to increase the relevance of returned entries to populate the list.

Incomplete problem entry selection with unspecified or vague results when linked to codes for reporting presents additional issues without an application-based solution. Personnel resources are often committed to conduct quality reviews of patient information, providing an opportunity to educate users in all departments, increase overall data quality, and review workflow to leverage the value of the EHR.

As the number of problem list entries increases, so do the data integrity risks, which can jeopardize the potential patient safety benefits of an EHR. Strict review of workflow and a shift in responsibilities from clinicians to order-receiving departments can ensure the problem list is updated through direct edits by personnel from the laboratory or pharmacy or automated import from these systems.

The most costly administrative issue is the lack of software application interoperability in list management. Custom application interfaces are required for data exchange between disparate systems. Legacy databases further increase the complexity and cost of EHR implementation. Competing developers may limit the interactivity of systems, increasing the cost of adopting new systems.

Cross-mapping entries speeds identification of an entry and identifies synonyms for intended entries, but the work involved may delay adoption of new systems. Additional work is often required when a new system is implemented due to the lack of standardized data dictionaries and cross-mapping guidelines. Local review of technical requirements prior to implementation limits delays and minimizes expense, though it increases the costs associated with technical reviews.

An independent technical review by qualified IT staff must be part of the assessment where complex EHR systems are used. Quality of care and patient safety are affected by software interoperability conflicts. Dual purpose software applications satisfying patient care requirements in addition to providing a source for billing codes should not limit the ability of the clinician to document clinical care appropriately.

Clinical Management Maintenance

A significant barrier facing the clinical end user is problem list clutter. Without careful management, the shared problem list accumulates multiple diagnoses and symptoms that may or may not be accurate to the patient’s true condition. Eventually, the list may become useless due to length and inaccuracy, and often the list is either appended to or displayed with each patient encounter. If the list is filled with inaccuracies, the clinical documentation is viewed as suspect and becomes a potential liability.

There are multiple clutter sources. First, many of the items on the problem list are symptoms and health factors that are not condition-specific. These symptom diagnoses may equate to a specific ICD-9-CM code but may not be important problems for the long-term management of the patient. An example would be dysuria. This diagnosis is important for the purpose of ordering diagnostic evaluation, but it is an improper entry to the patient’s active (long-term) problem list after the condition has been resolved with treatment.

Other entries contributing to the clutter barrier are acute care diagnoses. These are important for the purpose of diagnostic evaluation and clinical management of a patient’s condition at that particular encounter; however, they don’t belong on the active (long-term) problem list.
Continuing the above example, acute cystitis would be a patient problem identified following the evaluation of dysuria. Linking this problem or diagnosis to a clinical encounter is important; however, acute cystitis is by definition a self-limited problem, unlikely to require multiple encounters for resolution. If promoted to the active problem list, the problem becomes clutter when the patient presents months later with a different chief complaint.

ICD-9-CM V codes are used in problem list search engines but rarely belong on an active problem list. The supplementary classification from ICD-9-CM is for classification of factors influencing health status and contact with health services. Though a patient may present to the office for screening for malignant disease of the cervix, adding this diagnosis to the active problem list is of little clinical utility, despite being critical to the encounter for the purpose of ordering and billing.

On the other hand, problems affecting health status expressed as problems linked to V codes add important clinical facts to the record. Examples include the V44 category for artificial opening status and V42 category for organ transplant status or problems linked to codes similar to V60.4, No other household member able to render care.

As providers in different specialties add diagnoses to a patient’s list, multiple diagnostic statements describing a single condition will be entered into the same problem list. An example might include “diabetes mellitus” versus “type II diabetes mellitus.” As more specific diagnoses are added, the less specific entries remain in the list, adding length but not clinical or administrative detail.

The accumulation of entries that fail to define the patient’s current condition concisely frequently occurs in systems where the problem list is expressed using ICD-9-CM descriptions (a classification approach), but it is not limited to this code set. Use of a clinical terminology standard or controlled vocabulary for problem list expression is a contributing factor to the success of a well-designed problem list.⁴

Health Language and Terminology Standards

An effective problem list should provide a clear, descriptive summary of a patient’s health history. To accomplish this, the problem list must use a terminology that provides sufficient detail. Ideally, it allows the transfer of this information across healthcare systems and within the various portions of the EHR. Codification of problem lists enables interoperability and data mining for other purposes, such as quality of care measurement and administrative use, including claims submission for reimbursement.

A number of terminologies and classification systems have features that facilitate their use in problem lists. (Appendix B provides a list of these candidate terminologies, along with information pertaining to ownership, content, and means of distribution.) Some of these candidates, such as ICD-9-CM, are more familiar to clinicians and therefore are often viewed as the first choice. However, a classification system is not the best choice for a number of reasons.

ICD-9-CM codes often lack the granularity necessary to fully describe a health condition. For a complex condition such as muscular dystrophy, ICD-9 offers very limited codes. Specific types of muscular dystrophy, including Erb’s, Gower’s, and Duchenne’s, all fall under ICD-9-CM code 359.1, Hereditary progressive muscular dystrophy.

In contrast, SNOMED CT offers unique codes for each of these conditions. A clinician might be puzzled to see the ICD-9-CM description “Foreign body in the larynx” on a problem list when the condition they are looking for is “Aspiration.” Other terminologies, such as SNOMED CT, offer more detailed, descriptive codes for these conditions. UMLS, which incorporates both ICD-9-CM and SNOMED, may seem like the perfect compromise. Unfortunately, it is too large and unwieldy to function effectively for problem list use.

Of the candidate terminologies reviewed, SNOMED CT has the best mix of features and benefits for problem list use.⁵ SNOMED CT is the recommended clinical terminology standard for documenting clinical care for the problem list by Consolidated Healthcare Informatics (CHI) initiative and the National Committee on Vital and Health Statistics (NCVHS), and it is now endorsed by the Healthcare Information Technology Standards Panel.^6-8 SNOMED CT was developed for use in the healthcare system and selected based on its overall comprehensive clinical domain coverage, reference information, and demonstrated good update and maintenance practices.

In addition, CHI and NCVHS also recommend that healthcare organizations implement accurate cross mappings between clinical vocabularies such as SNOMED CT and the HIPAA code sets, including ICD-9-CM, to satisfy administrative purposes. This includes external reporting or data aggregation for measuring quality of care, monitoring resource utilization, or processing of claims for reimbursement.⁹

SNOMED CT is readily available, frequently updated, and currently being used successfully in many organizations. It provides a comprehensive, detailed listing of more than 300,000 concepts and 900,000 descriptions (compared to only about 13,000 codes in ICD-9-CM). Although the comprehensive nature of SNOMED CT makes it attractive for problem list use, it may be necessary to limit searches to the clinical findings hierarchy when adding a new item to a problem list. This would prevent clinicians from having to search through an overwhelming list of search results. Subsets of SNOMED concepts are useful for problem list development.¹⁰

EHR Functionality and Data Standards for Problem Lists

Problem list functionality in EHR systems must be designed to improve care, data quality, and timeliness, and it must support interoperable exchange of problem list content. Health Level Seven’s EHR System Functional Model recognizes the need to create and maintain problem lists in function DC.1.4.3. Conformance criteria for this function identify EHR system functionality needed to manage electronic problem lists over time, supporting the documentation of current and historical problems and tracking the changing character of these problems and their priority.

In addition, the Certification Commission for Healthcare Information Technology defines EHR system requirements in both its ambulatory and inpatient functionality criteria that address the need to create and maintain patient-specific problem lists. Ambulatory care software products certified in 2006 and 2007 have been required to meet nine criteria addressing the management of problems lists. Current inpatient functionality requirements related to problem lists will not be part of product certification testing until 2009 or later.

The Continuity of Care Record standard developed by ASTM includes a data segment for problem lists. The Continuity of Care Document standard from Health Level Seven applies the organization’s clinical document architecture messaging standard required for portability and interoperability to the ASTM data content requirements. These standards facilitate transfer of essential data from one EHR system to another wherever care is delivered.

Best Practices for Managing Problem Lists

Problem lists require standard approaches to managing data quality. The best practice is to avoid entering bad data, then to audit and remove any bad data found. Clinicians must be committed to entering only appropriate, supported diagnoses and patient problems and able to accept the careful removal of unnecessary or irrelevant data and resolved problems. Best practice considerations for problem lists are organized in table form in appendix D, also available online.

Organizations realize the benefits of improved quality care and patient safety when the EHR can cross-reference problems and treatment factors such as allergies and medication use in an active list and support clinical decisions. Potentially harmful clinical decisions and treatment plans can be avoided, or an interaction with a resolved problem may be signaled and identified for correction if the list is inaccurate or out of date.

Workflow solutions can address accuracy, especially when the patient or caregiver is involved in the process. The waiting patient can be part of his or her care by reviewing the problem list, which enables support staff to remove resolved problems from the list pending physician confirmation during the encounter.

Care should be taken when removing active problems identified by another provider when problem lists are shared between organizations or physicians. The patient may be able to provide enough information to support removing an entry. If unable to collaborate with other providers to verify accuracy, it is advisable to leave a problem on the list. Diagnostic and problem statements are critical for the appropriate documentation of a specific encounter, even if not important for long-term care.

EHRs should provide a means to enter diagnostic statement entries linked to specific encounters without promotion to the active problem list. EHR design considerations include the capability to separate acute problems from chronic ones or the functionality to set a date for automatic resolution of the two. An “inactive” problem location, separate from the active list, can retain these interim issues. For example, it is important to document a transient diagnosis (problem) of hypokalemia and subsequent treatment with potassium, but it may not be necessary to add it to a long-term care list.

A Shared Problem List

The EHR is truly a shared and interdisciplinary record—more so than any previous incarnation of patient health record keeping. This presents new challenges for management of shared problem lists, and benefiting from new opportunities for quality enhancement requires carefully considered organizational guidelines.

A concise list is necessary to enhance readability and usability. Problem list accuracy is of paramount importance in order to take full advantage of safety measures, document encounters precisely, and maximize informatics opportunity. All users in the organization must contribute to list maintenance to realize improvements in healthcare delivery promised by EHRs. The organization is responsible for providing guidelines for problem list creation and maintenance to achieve this goal.

The use of an electronic problem list is a change predicated on patient safety and improved patient care. It may help control costs, and it has the potential to increase efficiency, depending on how it is used. The benefits realized from the use of this tool rely on the dedication of professionals to change for the benefit of the patients they serve.

Notes

1. Weed, Lawrence L. “Medical Records that Guide and Teach.” New England Journal of Medicine 278, no. 11 (1968): 593–600.

2. Salmon, P., et al. “Taking the Problem Oriented Medical Record Forward.” AMIA Annual Fall Symposium (1996): 463–67.

3. Institute of Medicine. The Computer-Based Patient Record: An Essential Technology for Health Care. Washington, DC: National Academy Press, 1997.

4. Salmon, P., et al. “Taking the Problem Oriented Medical Record Forward.”

5. Campbell, J.R. “Strategies for Problem List Implementation in a Complex Clinical Enterprise.” Proceedings/AMIA Annual Symposium (1998): 285–89.

6. Elkin, P.L., et al. “Evaluation of the Content Coverage of SNOMED CT: Ability of SNOMED Clinical Terms to Represent Clinical Problem Lists.” Mayo Clinic Proceedings 81 no. 6 (2006): 741–48.

7. Department of Health and Human Services, Office of the National Coordinator for Health Information Technology. “Federal Initiatives.” Available online at www.hhs.gov/healthit/chiinitiative.html.

8. National Committee on Vital and Health Statistics. Letter to Secretary Thompson. November 3, 2003. Available online at www.ncvhs.hhs.gov/031105lt3.pdf.

9. Healthcare Information Technology Standards Panel. Available online at www.ansi.org/hitsp.

10. Tang, P.C., et al. “Comparison of Methodologies for Calculating Quality Measures Based on Administrative Data versus Clinical Data from an Electronic Health Record System: Implications for Performance Measures.” Journal of the American Medical Informatics Association 14, no. 1 (2007): 10–15.

Appendixes

The following appendixes appear in the online version of this practice brief, available in the FORE Library: HIM Body of Knowledge at www.ahima.org.

• Appendix A: Definitions of Problem Lists from Authoritative Sources
• Appendix B: Candidate Terminologies for Encoding Problem Lists
• Appendix C: Resource List for Additional Research and Guidance
• Appendix D: Best Practice Considerations for Problem Lists

Prepared By

AHIMA Best Practices for Problem Lists in an
EHR Work Group
Raeanna Bonetti, RHIT, CPC
Joe Castelli, MD
Jennifer L. Childress, RHIT
Joan Cohen, RHIT
Lisa Hanson, MHA, CCS
Maribeth Hernan, MA, RHIA, CHP
Teonna Ingram, RHIA
Daniel A. Kowalczyk, RN, BSN, MBA
Mary Lambert, RHIA
Melanie Loucks, RHIT
Katherine Maddox, RHIA
Sue Mitchell, RHIA
Deborah Neville, RHIA, CCS-P
Susan Penders, RHIT
Clarice P. Smith, RHIA, CHP
Sharon Sprenger, RHIA
Kimberly Suggs, RHIA, CCS
Anne Tegen, MHA, RHIA, HRM
Patricia S. Wilson, RT(R), CPC, PMP

AHIMA Staff

Kathy Giannangelo, MA, RHIA, CCS, CPHIMS
Crystal Kallem, RHIT
Rita A. Scichilone, MHSA, RHIA, CCS, CCS-P, CHC

This work was supported in part by a grant to the Foundation of Research and Education from 3M Health Information Systems.