DIFFERENCE IN NURSING DOCUMENTATION
BEFORE AND AFTER COMPUTERIZATION:
A PILOT STUDY (Part 1)
Sandra Boldreghini, MSN, RN, CS
Family Nurse Practitioner
Desoto Family Medical Practice, P.C
Olive Branch, MS
June H. Larrabee, PhD, RN
Center for Nursing Research
Camcare Health Education & Research Institute
Revolutionary changes are reshaping the health care arena across the USA into a
managed competition market in response to pressure from payers concerned with curbing
escalating health care costs (Larrabee, 1996b). Accompanying this trend have been
escalating initiatives in technology assessment, development of clinical practice guidelines, and development of meaningful outcome indicators of quality. Use of these guidelines is intended to direct cost-effective quality improvements that yield better patient outcomes. In addition, using indicators of quality as comparative benchmarks for decision-making about provider selection and reimbursement is increasing market competition (Adams & Biggerstaff, 1995; Batalden, Nelson & Roberts, 1994; Field & Lohr, 1992; Luce & Brown, 1995). Nurses, like other health professionals, are engaged in activities
demonstrating their unique contribution to health care and outcomes (Maas, Johnson & Moorhead, 1996).
Historically, the patient chart is the primary source of information about health care quality Donabedian, 1988). Documented evidence of the nursing process has been used as an indicator of quality in both quality assessment (Goldmann, 1990; Mize, Bentley & Hubbard, 1991) and research (Ehnfors & Smedby, 1993; Hegyvary & Haussmann, 1976; Larrabee, Engle & Tolley, 1995). Yet, handwritten narrative notes often lack important patient data, are not consistent in documenting problems or plan of
care, and may also lack nursing interventions and patient teaching (Gropper & Dicapo, 1995). Nursing
documentation most often addresses interventions and less frequently addresses outcomes (Ehnfors & Smedby, 1993). Incomplete documentation compromises the reliability and validity of chart data (Aaronson & Burman, 1994; Bjornstad, Farr, Vernon, Welkie & Witte, 1995) for patient care decision-making, quality assessment, and research.
Fortunately, computerized nursing documentation has been demonstrated to improve documentation comprehensiveness (Barnoud, 1994; Kahl, Ivancin & Fuhrmann, 1991; Larrabee et al., 1992; Pabst, Scherubel & Minnick, 1996), including documentation of nursing interventions (Pryor, 1989), while improving patient satisfaction (Barnoud, 1994) and user satisfaction (Churgin, 1994; Pabst et al., 1996). In comparing documentation comprehensiveness pre- and postimplementation of a nursing information
system (NIS), Pryor found that documentation of mandatory interventions increased from 40% to 91% and optional interventions increased from 27% to 90%. Because computerized nursing documentation is more comprehensive than handwritten documentation, it is a more valid data source for investigations into the influence of nursing interventions upon patient outcomes.
When care planning, nurses select expected outcomes that are anticipated to promote the individual patient's well-being, and they select interventions anticipated to achieve those outcomes. Theoretically, achieving outcomes can be viewed as beneficial or a form of beneficence (Larrabee, 1996a), and implementing appropriate interventions can be viewed as evidence of care quality (Mize et al., 1991). Although the relationship between goal achievement (beneficence) and intervention implementation (quality) has been evaluated in a setting with a computerized NIS (Larrabee, Engle & Tolley, 1995), to date no study has specifically investigated the influence that implementing an NIS has on the
documentation of patient outcomes or on the relationship between patient outcomes and nursing interventions. Information about such relationships could guide quality improvement efforts to give nursing services higher quality and make them more cost-effective. Therefore, the purposes of this study were, first, to determine the difference in nursing documentation comprehensiveness of nursing care quality and nurse goal achievement before and after implementing an NIS. The second purpose was
to determine the relationship between nursing care quality and nurse goal achievement before and after
implementing an NIS.
An emerging model of quality (Larrabee, 1996a) was the conceptual framework for this study. In this model, quality is defined as "the presence of socially acceptable, desired attributes within the multifaceted wholistic experience of being and doing" and includes four interrelated concepts: value, beneficence, prudence, and justice. Quality and beneficence are the two concepts of focus in this study. "Beneficence is
the potentiality or actuality of (a) producing good and (b) promoting well-being" and includes harmlessness. This model proposes that quality and beneficence are interrelated, and
empirical evidence supports this relationship. Low pain score (beneficence) was a predictor
of patient-perceived quality (quality) in a group of 199 hospitalized adults (Larrabee et al., 1995).
Postpartal women's perceived quality (quality) and women's perceived benefit (beneficence) were correlated in a group of 60 women (Hunter, 1994). Expected outcomes in health care are goals for the well-being of patients and are, therefore, potentially beneficial. Appropriately implemented interventions in health care are designed to actualize goal achievement and are, therefore, indicators of quality.
This study used a comparative descriptive design to describe differences in quality and beneficence before and after implementing a computerized NIS. A descriptive correlational design was used to examine the relationship between quality and beneficence before and after implementing the NIS.
Site and Sample
The setting was a 100-bed university-affiliated teaching hospital predominately providing acute medical and surgical care to adults in a metropolitan area in the midsouth of the United States. Financing of care at this hospital comes from Medicare (63%); TennCare (13%), the state's alternative to Medicaid; commercial insurance (12%); or self-pay (12%).
The sample consisted of charts of 40 hospitalized patients admitted during the months of August 1995, the month prior to implementing the NIS, and April 1996, six months postimplementation. Twenty charts were retrospectively evaluated from both months. The 6-month delay in collecting data allowed for nurses to develop skills after training in the use of the system.
Go to Part 2 of the article.