Update on Logical Observation Identifier Names and Codes (LOINC)

Susan Matney, RN, MS, Intermountai n Health Care, Salt Lake City, Utah

Judy G. Ozbolt, Ph.D. R.N., Vanderbilt University, Nashville, Tennessee

Suzanne Bakken, DNSc., R.N., Columbia University

Citation:

Matney, S., Ozbolt, J. & Bakken, S. (February 2004). Update on Logical Observation Identifier Names and Codes (LOINC) . Online Journal of Nursing Informatics (OJNI). Vol. 8, No. 1. [Online]. Available at http://ojni.org/8_1/nurslangup.htm

This update gives an overview of what has happened with the nursing Logical Observation Identifier Names and Codes (LOINC) terminology since its recognition on December 18, 2002 by the American Nurses Association (ANA) Committee for Nursing Practice Information (Thede, 2003) . A new nursing subcommittee of Clinical LOINC has developed a statement of mission and action items and has submitted new nursing content for inclusion into the LOINC database. The main content focus for this update will be nursing goals.

Before the ANA recognition, nursing codes in LOINC were being submitted on a “piece-meal” basis. LOINC is still far from having a comprehensive set of nursing measurements (Matney, 2003; Matney, Bakken, & Huff, 2003). To add nursing content to LOINC at a faster rate, participants at the June 2003 Clinical LOINC meeting accepted a proposal to create a nursing subcommittee of LOINC. The executive committee for the nursing LOINC subcommittee consists of Susan Matney (Chair), Suzanne Bakken, and Judy Ozbolt. The mission of the nursing subcommittee is to meet needs for machine-readable clinical, administrative, and regulatory nursing data by providing LOINC codes for observations at key stages of the nursing process, including assessments, goals, and outcomes. The nursing subcommittee will fulfill the following responsibilities:

  1. Act as a point of entry for those who wish to submit nursing data to LOINC, assisting as needed with understanding LOINC methods and processes.
  2. Triage all nursing content submitted to LOINC.
  3. Identify “gaps” in the content and solicit input from experts who can assist in the creation of content for those gaps
  4. Create task forces as needed to work on specific terminology areas.
  5. Submit content to the Clinical LOINC committee in a “fit” condition to use.
  6. Disseminate starter sets, such as vital signs assessment data, to nurses to review and critique.
  7. Disseminate new content in the literature.

Two types of gaps identified so far are goals and nursing assessments. This update will focus on nursing goals and the following update will focus on new nursing assessment measurements in LOINC.

LOINC Structure

Let's briefly review the LOINC structure before looking at goals. LOINC models the semantic structure of aggregate or precoordinated expressions using the multi-axial representational approach. Each LOINC record corresponds to a single observation, measurement, or test result. The code for the single LOINC record can be used in the observation field of a Health Level Seven (HL7) message (Huff et al., 1998). The LOINC axes include component, kind of property, time aspect, system, precision, and type of method.

  1. Component (analyte) - the substance or entity that is measured, evaluated, or observed . Examples include systolic blood pressure, pain onset, and sodium .
  2. Kind of Property - The characteristic or attribute of the analyte that is measured, evaluated, or observed. Examples include length, volume, time stamp, mass, ratio, number, and temperature.
  3. Time Aspect - The interval of time over which the observation or measurement was made. Examples include point in time and 24 hour.
  4. System - The system (context) or specimen type within which the observation was made. Examples include urine, serum, fetus, patient, and family.
  5. The type of Scale – The scale of the measure. A scale may be quantitative (a true measurement), ordinal (a ranked set of options), nominal (e.g., E. coli; Staphylococcus aureus), or narrative (e.g., dictation results from x-rays).
  6. Type of Method - The procedure used to make the measurement or observation. The LOINC record includes Method only when the method of measurement makes an important distinction in sensitivity or specificity. Method is the only axis that is optional.
Goals

Therapeutic goals specify the desired state, usually a condition or a behavior of a patient or a family member. It is useful to identify goals that are short-term, observable markers of patient progress to be assessed at particular points in a plan of care or a clinical pathway. When a clinician receives for each patient a list of goals not met in the previous 24 hours, the clinician can see in what areas the patient is not progressing as expected and can fine tune the plan of care accordingly. When a clinical team receives for each pathway the aggregate goal achievement status for all patients on each pathway in the previous month or the previous quarter, the team members can identify pathways where performance is less than optimal. By examining the rate of achievement of each goal on the pathway, the team can identify particular areas where the pathway needs improvement.

Participants at the Nursing Terminology Summit Conferences assessed the relationship of goals to other kinds of assessment data, such as observations and outcomes. They determined that goals share similar semantic structures to other types of findings (Bakken et al., 2002; Ozbolt, 2003). For the purposes of HL7 messages, goals will be differentiated from other findings or observables by the use of the mood code of Goal to denote that it is a statement about the future rather than a current observation (mood code of Event). In content, goals differ from many other observations by describing desirable conditions or behaviors rather than pathologies.

Any LOINC code in the LOINC database could be considered a goal statement if the HL7 Mood were identified as “Goal” instead of “Event.” The LOINC committee, however, has found it useful to model “goal evaluation statements,” real events assessed as impressions of the goal achievement status at some designated time. With that in mind, the codes from the Home Healthcare Classification System and the Omaha system already included in LOINC (Martin & Norris, 1996; Saba & Zuckerman, 1992) can be used as goal evaluation statements. See the examples below.

Examples of HHCC and Omaha Nursing Codes included LOINC

COMPONENT

PROP

TIMING

SAMPLE

SCALE

METHOD

VALUES

KNOWLEDGE DEFICIT

IMP

PT

^PATIENT

ORD

OBSERVED.HHCC

IMPROVED, STABILIZED, DETERIORATED

ANXIETY

IMP

PT

^PATIENT

ORD

OBSERVED.HHCC

IMPROVED, STABILIZED, DETERIORATED

BODY IMAGE DISTURBANCE

IMP

PT

^PATIENT

ORD

OBSERVED.HHCC

IMPROVED, STABILIZED, DETERIORATED

BOWEL INCONTINENCE

IMP

PT

^PATIENT

ORD

OBSERVED.HHCC

IMPROVED, STABILIZED, DETERIORATED

CHRONIC PAIN

IMP

PT

^PATIENT

ORD

OBSERVED.HHCC

IMPROVED, STABILIZED, DETERIORATED

COGNITION.BEHAVIOR

IMP

PT

^PATIENT

ORD

OBSERVED.OMAHA

1=NOT APPROPRIATE, 2=RARELY APPROPRIATE, 3=INCONSISTENTLY APPROPRIATE, 4=USUALLY APPROPRIATE, 5=CONSISTENTLY APPROPRIATE

GRIEF.KNOWLEDGE

IMP

PT

^PATIENT

ORD

OBSERVED.OMAHA

1=NOT APPROPRIATE, 2=RARELY APPROPRIATE, 3=INCONSISTENTLY APPROPRIATE, 4=USUALLY APPROPRIATE, 5=CONSISTENTLY APPROPRIATE

At Vanderbilt University Medical Center (VUMC), precoordinated goal statements from the Patient Care Data Set (PCDS-VU) are integrated into clinical pathways that guide the care of every patient. Informatics methods and organizational procedures are being put into place to use daily feedback about goal achievement to manage the care of each patient and monthly feedback about goal achievement to learn from clinical data how to improve the pathways (Ozbolt et al., under review). These goal data are proving useful for quality improvement at VUMC and therefore have been approved by the Clinical LOINC Committee. They will be available this year in a future release of the LOINC database so that they may have a standardized representation and be made available to others. The examples below show some precoordinated phrases from VUMC pathways (Component) and their representation in LOINC.

Examples of PCDS Nursing Goals Submitted to LOINC

COMPONENT

PROPERTY

TIMING

SAMPLE

SCALE

METHOD

VALUES

PERFORMS EXERCISES

IMP

PT

^PATIENT

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

EXERCISES SAFELY

IMP

PT

^PATIENT

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

INDEPENDENT WITH EXERCISE

IMP

PT

^PATIENT

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

MOBILE WITH ASSISTANCE

IMP

PT

^PATIENT

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

MOBILITY PROGRESSING

IMP

PT

^PATIENT

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

MOBILITY RESTORED

IMP

PT

^PATIENT

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

ANXIETY RELIEF

IMP

PT

^PATIENT

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

ANXIETY RELIEF

IMP

PT

^FAMILY

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

ANXIETY RELIEF

IMP

PT

^SO

ORD

OBSERVED.PCDS

Met, In Progress, Not Met

FLUID BALANCE ACHIEVED

IMP

PT

INTAKE AND OUTPUT

ORD

PCDS

Met, In Progress, Not Met

ELECTROLYTE BALANCE ACHIEVED

IMP

PT

ELECTROLYTES

ORD

PCDS

Met, In Progress, Not Met

FLUID BALANCE ACHIEVED

IMP

PT

ELECTROLYTES, INTAKE AND OUTPUT

ORD

PCDS

Met, In Progress, Not Met

Conclusion

Providing a comprehensive set of nursing measurements in LOINC will require the creation of a great deal of LOINC content. Work is currently in progress to translate nursing goals into LOINC format. The next update will focus on nursing assessment data. The open development process associated with the creation of the LOINC database supports the submission of codes from those with content expertise. A new nursing subcommittee within LOINC is now working to speed the addition of nursing content to LOINC.

References

Bakken, S., Warren , J. J., Casey, A., Konicek, D., Lundberg, C., & Pooke, M. (2002). Information model and terminology model issues related to goals. Proceedings of the AMIA 2002 Annual Symposium , 17-21.

Huff, S. M., Rocha, R. A., McDonald, C. J., De Moor, G. J. E., Fiers, T., Bidgood, W. D., Jr., Forrey, A. W., Francis, W. G., Tracy, W. R., Leavelle, D., Stalling, F., Griffin, B., Maloney, P., Leland, D., Charles, L., Hutchins, K., & Baeziger, J. (1998). Development of the LOINC (Logical Observation Identifier Names and Codes) Vocabulary. Journal of the American Medical Informatics Association , 5 (3), 276-292.

Martin, K. S., & Norris, J. (1996). The Omaha System: A model for describing practice. Holistic Nursing Practice, 11 (1), 75-83. Matney, S. (2003). Logical Observation Identifier Names and Codes (LOINC) ANA Recognition Commentary. Online Journal of Nursing Informatics, 7 (3).

Matney, S., Bakken, S., & Huff, S. M. (2003). Representing nursing assessments in clinical information systems using the logical observation identifiers, names, and codes database. Journal of Biomedical Informatics, 36 (4-5), 287-293.

Ozbolt, J. (2003). Reference Terminology for Therapeutic Goals: A New Approach. Reference Terminology for Therapeutic Goals: A New Approach , 504-508.

Ozbolt, J., Özdas, A, Waitman, L.R., Smith, J. B., Brennan, G. V., Miller, R.A. (Under review). Decision Support for Patient Care: Implementing Cybernetics.

Saba , V. K., & Zuckerman, A. E. (1992). A new home health class ification method. Caring Magazine, 11 (9), 27-34.

Thede, L. Q. (2003). Informatics and Nursing: Opportunities & Challenges (2nd ed. ed.). Philadelphia : Lippincott Williams & Wilkins Publishers.