Approaches to and uses of patient classification systems.
Motivated by escalating costs, efforts to restructure and downsize the industry have focused on decreasing expensive inpatient hospital care. Outpatient, home care, and a variety of less expensive treatment modalities are being substituted for or used to supplement inpatient care. The trend towards less inpatient and more ambulatory care has accelerated since the passage of Medicare's Prospective Payment System (PPS) in 19c,3. Further, this trend continues today as rapidly increasing numbers of persons receive their health care within a managed care framework which provides strong financial incentives to avoid hospitalization in favor of ambulatory and preventive care.
The volume and variety of health care services that fall under the ambulatory care umbrella today are vastly more complex and diverse. The continued escalation of these trends, even without enactment of health care reform legislation, will pose many challenges for health care managers. For example, identifying the resource use and needs of groups of patients is an important part of matching patient needs with appropriate personnel. Care managers need tools and methods to assist in planning for cost-effective staffing levels and staff mix for the diverse clinics and outpatient services. In accord with this need, the purpose of this article is to review tools and approaches to measuring patient needs for care in ambulatory settings and to present a new measure of nursing intensity for use in ambulatory care.
Approaches to Patient Classification
Patient Classification Systems (PCSs), which measure patients' needs for care and care activities, have been used since the 1960s in hospitals as the basis of workload monitoring systems that determine staffing. While these PCSs have been criticized on a number of grounds (Gioventti & Mayer, 1984; Prescott, 1991), they can be helpful adjuncts to seasoned managers if they are constructed and used appropriately. Systems developed for use with hospital inpatients do not easily, or directly, transfer to the ambulatory setting as patient needs and ambulatory care differ from inpatient care in important ways.
1. The usual patient visit is brief.
2. A single visit may involve multiple distinct encounters.
3. Care often is discontinuous with patients returning for visits at irregular intervals.
4. Patients have widely different care needs from one visit to another.
5. Telephone management and screening are important dimensions of some ambulatory services.
6. Many visits involve no specific medical treatments or procedures and have a limited focus or purpose.
Given these differences in patients and care settings, health care managers need classification systems developed and tested for reliability and validity in ambulatory care. The need is particularly acute for nursing as few systems currently exist to account for patients' needs for, and use of, nursing resources. Staffing (in terms of both level and mix of personnel) and also nursing care roles vary widely in ambulatory care. In some settings, nursing personnel escort patients to and from examining rooms, and most direct patient care is delivered by physicians. In other settings, nurse practitioners substitute for physicians and provide the majority of the patient care.
To date most PCSs developed and tested in ambulatory care have been created from a medical perspective to capture physician resource use. As with the development of the DRG system to account for inpatient resource use, nursing services have not been included as an important professional resource to be accounted for within these systems. Despite the largely medical focus, it is important that those interested in measuring nursing intensity be aware of the classification systems that have been and currently are being tested in ambulatory settings.
Medical Approaches to Patient Classification in Ambulatory Care
In 1990 Georgoulakis and colleagues identified 35 medically oriented PCSs for ambulatory care in varying states of development. The motivation behind many of these PCSs is to create, for ambulatory care, a system analogous to the DRGs, that can be used as the basis for prospective payment. A number of efforts have been made to form clinically meaningful groups. However, the systems are designed primarily for payment as opposed to clinical management of patients or day-to-day deployment of personnel, which is the focus of PCSs developed in nursing. Despite these differences, many of the factors that make patients resource intensive from a medical point of view also influence nursing intensity.
Diagnostic clusters developed by Schneeweiss, Rosenblatt, Cherkin, Kirkwood, and Hart (1983) and Schneeweiss et al. (1986) use the patient's primary medical diagnosis during an ambulatory encounter to form 110 groups or clusters of similar patients by combining medical diagnoses (ICD-9-CM) and the International Classification of Health Problems in Primary Care (ICHPPC). The diagnostic clusters account for almost 90% of all diagnoses recorded by family physicians in a variety of ambulatory settings.
Building on the work of diagnostic clusters, Fetter and the other architects of the DRG system at Yale developed Ambulatory Visit Groups (AVGs). In this system, 571 groups are formed from 19 major diagnostic categories, types of procedures performed, patient status, age, and sex (Fetter, Averill, Lichtenstein, & Freeman, 1984; Schneider et al., 1988). Evaluation of the AVGs by the Brandeis group (Lion, Malbon, & Bergman, 1987) revealed a correlation of r = .31 between provider time and total resource use based on 7,531 cases.
Ambulatory Patient Groups (APGs) represent another system currently under development by Averill and Goldfield who are working with the Health Care Financing Administration (HCFA) to modify the AVGs. According to the final report of this project released in 1991, there are a total of 297 APGs (comprising 145 procedure, 80 medical, and 72 ancillary service APGs) which account for the facility cost of providing ambulatory services. Facility cost refers to room charges and medical/surgical supplies, but excludes physician professional services. Based on a test of the APGs with 1988 Medicare outpatient data, the APGs accounted for 79% of the variance in trimmed charges, and this contrasts favorably with the inpatient DRG system which accounts for 50%. Also, in contrast with the DRG system, patients may be assigned to multiple APGs (Averill et al., 1990).
Another approach, developed and used in New York, classifies ambulatory care patients and services into the Products of Ambulatory Care (PAC) (Tenan et al., 1988). The PAC is built by packaging together all the labor and ancillary services typically delivered within an ambulatory visit. Patient demographic characteristics are one factor in forming the 24 diagnostic service categories, but the system is not dependent on diagnosis. Twenty-four PACs are formed by combining type of patient with diagnostic management services provided by all ambulatory care personnel. Preliminary data reported by Tenan et al. (1988) are promising as the PAC explained 65% of the variance in resource use among the 10,000 cases in the data set.
Ambulatory Care Groups (ACGs) developed at Johns Hopkins consist of 51 categories based on demographic characteristics and disease patterns over time. The ACGs are formed using the general process of grouping ICD-9-CM diagnoses used by DRGs, and the dependent variable is the number of ambulatory care visits made during a year's time. The ACG case mix system was tested and accounted for between 32% and 50% of the variance in number of visits over a year and 5% of total charges. The authors suggest ACGs are not appropriate for use at the level of individual patient visits but rather for HMO rate setting and utilization review (Starfield, Weiner, Mumford, & Steinwachs, 1991; Weiner, Starfield, Steinwachs, & Mumford, 1991).
The Ambulatory Severity Index (ASI) is intended to measure severity of illness by two major dimensions: a biophysical dimension related to number and type of diagnoses and a behavioral dimension which includes physical and psychosocial functioning, social factors, and patient compliance (Horn, Buckle, & Carver, 1988). The ASI, in contrast with other systems, is rated by the provider at the time of the visit. No data are reported regarding the relationship of the ASI to resource use.
One limitation of these systems is that they either exclude physician professional services or have focused predominantly or solely on physician time and activity to the exclusion of other providers such as nurses, social workers, and others (Georgoulakis et al., 1990). As pointed out by Lion et al. (1987), physician time correlates poorly with total provider time and total resource use. Only 81% of the visits for all primary care involved any physician time at all. Almost 16% of the visits in the Brandeis study, 23% of pediatric, 30% of medical oncology, and 50% of ob/gyn and radiation visits did not involve any physician time. Goldfield (1993) summarized in tabular form eight ambulatory classification systems under varying degrees of development and examines their differences using two case histories for illustrative purposes.
Increasingly primary services are being performed by nurses, and this trend is likely to accelerate in the future (Division of Nursing, 1994). The appropriateness of existing medically based systems for capturing resource consumption other than physician time is unknown but an important consideration for future study.
Nursing Approaches to Patient Classification in Ambulatory Care
In 1984, Richards and Tracy indicated that there were no published models for determining nurse staffing in ambulatory care. In a beginning effort they analyzed nurse staffing patterns in a single setting where nursing time associated with direct patient care activities was divided into three phases of a clinic visit: checking in, examination room functions, and checking out. Regression analyses were performed to associate nurse reported time with specific tasks and procedures.
Subsequently, Hoffman and Wakefield (1986) used a Delphi method for developing a set of nursing care factors for classifying patients in ambulatory care. Nurse self-reports were used to establish initial estimates of time and frequency parameters for nursing care tasks. They emphasized differences between inpatient and outpatient care which are important for workload determination, such as workflow and volume patterns as well as variability in patients.
Also in 1986, Verran proposed a model of six "responsibility areas" which included 44 nursing care activity categories representative of ambulatory nursing practice. She suggested that both time and complexity associated with performance of nursing care are important dimensions of a PCS. The Ambulatory Care Client Classification Instrument (ACCCI) developed and tested by Verran is an important step in developing classification systems used by nursing. It is conceptually derived and includes complexity weights which are assigned to specific nursing care activities.
In discussing PCSs for ambulatory care, Hastings (1987) identified many of the important differences between inpatient and outpatient nursing care which make it difficult to use inpatient tools in ambulatory settings. Further, she suggested that the type of encounter or reason for a clinic visit may be as, or more, important than patients' care needs in determining nursing resource use in ambulatory care.
Johnson (1989) described an adaptation of the inpatient Allocation Resource Identification and Costing system (ARIC). The six-category system is formed by combining dependent nursing activities (physician driven) with independent care needs which include nurse-initiated psychosocial support and patient teaching.
In 1987 Parrinello developed a four-level PCS based on the Strong Memorial Hospital inpatient classification system. The distribution of 150 patients in the study across the four time categories revealed that the measure was not highly discriminatory for there were no level one patients and two-thirds of the cases were classified at level three.
In a later study Parrinello, Brenner, and Vallone (1988) adapted and tested the ACCCI developed by Verran (1986a & b). Distribution problems were noted with 74% of the 5,022 patient visits being classified at level one. In addition, although the median values for the four time categories differed, the ranges overlapped significantly. Also, while the investigators began with the ACCCI, it appears that they dropped the concept of complexity from their determination of activity weights which were based on nurse estimates of relative nursing time required.
Also based on Verrans work, Hastings and Muir-Nash (1989) demonstrated high levels of agreement among ambulatory care nurses regarding the Verran taxonomy, and Schade and Austin (1992) developed and tested the Ambulatory Care Patient Classification Tool (ACPCT) for use in pediatric ambulatory care. Mean time and complexity weights for the ACPCT were developed by Delphi surveys for each of the 44 activity categories. Predicted time was compared with actual worked time across clinics and was highly variable. Joseph (1990) built upon the Tighe, Fisher, Hastings, and Heller (1985) adaptation of Verran's work. Work sampling methods were applied to identify times associated with 47 nursing activities.
Finally, Kirsch and Talbott (1990) described an approach to outpatient and short stay patient classification which is department specific in terms of patient descriptors. Time is related to patient descriptors through relative value units by usual methods.
The tremendous growth in ambulatory care services over the past decade has stimulated efforts to develop PCSs appropriate for managing nursing resources in these diverse settings. To date there is no widely used and methodologically adequate approach to measuring nursing intensity for ambulatory care services. The majority of PCSs developed for ambulatory settings use the task-based approach common to inpatient systems. With the notable exception of Verran and those who have built upon her work, few deal with the complexity of care which is an important dimension of nursing intensity.
Verran's effort to apply Perrow's definition of technological complexity is only partially successful at a conceptual level. In Perrow's framework, complexity derives from "the nature of the raw materials" and also the uncertainties surrounding the "search process" (Perrow, 1967). Applying these concepts to nursing, the patient is the "raw material," and the "search process" is the performance of the nursing process (assessment, intervention, evaluation). Complexity is associated with applying the nursing process in the assessment, management, and evaluation of patient care needs. Using this definition, complexity is not inherent in the task or activity alone. Instead it relates to decision making that surrounds performance of activities. Patient and situational factors combine with task or activity performance to determine complexity, and these dimensions are not included in Verran's approach to complexity. For example, turning a patient may be a low-complexity activity with an uncomplicated surgical patient. The same activity can be very complicated with a critically ill trauma patient who is hemodynamically unstable.
New Uses for PCSs in Ambulatory Care
In the past, PCSs in nursing have been used exclusively for administrative purposes (to project staffing and in some hospitals for billing for nursing care). In medicine PCSs have been used largely for payment as witnessed by the inpatient DRG system and the analogous systems currently under development for ambulatory care. While these uses remain relevant, PCSs have many other potential uses as well. It is these other uses that offer exciting new opportunities for both clinicians and administrators (see Figure 1).
Uses for Patient Classification Systems
1 Clinical Subsystem
* Framework for practice
* Individual patient outcomes
2 Traditional Administrative Subsystem
3 Newer Administrative Subsystem
Administrative Research/ Evaluation and Planning
* Skill mix and outcomes
4 Quality Assessment/Improvement Subsystem
* Data for use in conjunction with clinical paths and protocols
5 Research Subsystem
* Patient outcomes for special populations
* Patient outcomes associated with changes in clinical practice
In Figure 1, the second arrow illustrates the traditional administrative use of data from PCSs (staffing and budgeting). Many existing systems are adequate for this purpose, but not for the other purposes outlined. To be useful for other administrative purposes shown by arrow 3, quality assurance and quality improvement (arrow 4), research (arrow 5), and guiding clinical practice (arrow 1), the PCS must provide clinically meaningful information. Most PCSs designed for staffing or payment purposes alone do not meet these criteria and thus can not be used for these newer applications.
To effectively and efficiently deploy and manage resources in today's highly competitive and cost-driven environment, clinicians and managers alike must link patient needs with provider activities, resource use, and patient outcomes. PCSs that collect clinically meaningful patient data in a reliable and valid manner and link those data with the patient record in a clinically relevant time frame can be used for all five purposes identified in Figure 1.
To further illustrate the idea, most nursing-based PCSs deal with the concept of patient mobility. Traditional systems often employ a checklist of nursing activities weighted by the average time needed to perform the activity; for example, turning a patient. The amount of time needed to turn a patient then is combined with the time needed to perform other tasks and procedures associated with a particular patient's care. Hours of care and associated staffing are then projected.
An alternative approach, and the one employed in the PCS to be introduced in the next section of this article, is to rate the degree to which the patient has limitations in mobility. For example, in the Patient Intensity for Nursing Index (PINI) the patient is scored from a low of 1 (indicating the patient is independently fully mobile) to a high of 5 (indicating total immobility and dependence on nursing personnel to accomplish mobility functions). This approach produces a score (mobility in this instance) that is meaningful to clinicians in that it can be used to assess the patient's need for care and to determine the status of his limitations before and after interventions to yield outcome information for both individual patients and groups of patients. A further discussion of how PCS data can be used for the multiple clinical research and administrative purposes in Figure 1 will be continued after first presenting the conceptual basis of the PINI and the Patient Intensity for Nursing Ambulatory Care (PINAC).
Conceptual Basis and Development of the PINAC
The PINI and PINAC include selected concepts from both medical and nursing approaches to classifying patients in ambulatory care. The specific conceptual model of the PINAC derives from that used to develop the PINI, an inpatient measure of nursing intensity, which has been tested for reliability and validity over 10 years (Prescott & Phillips, 1988; Prescott et al., 1991; Prescott, 1991; Soeken & Prescott, 1991).
As shown in Figure 2, the PINI has three conceptual components: complexity, dependency, and severity. Complexity refers to the knowledge, skill, and critical thinking needed to implement the assessment, intervention, and evaluation phases of the nursing process. The approach to measuring complexity in decision making associated with the nursing process is adapted from the framework of Beach and Mitchell (1978). They identify factors that make decisions more or less complex. For example:
1. Ambiguity of goals.
2. Instability of the situation.
3. Consequence of a mistake.
4. Number of variables that must be considered in arriving at a course of action.
5. Degree to which the problem has a known solution and others.
PINI Conceptual Model
Severity of Illness Complexity of Care Patient Dependency Needs
The PINI is unique in its direct measurement of complexity. Complexity of care has been included in other PCSs only to the degree that it is related to time. While often it is true that more complex things take more time to perform, complexity can be distinctly different from time and should be measured separately in its own right. For example, a stable bedridden patient may require large amounts of low-complexity care associated with activities of daily living while another patient, such as a teenage primipara with no prenatal care and multiple family and school problems, is likely to require fewer hours of care than the first patient. While the second patient requires fewer hours of care, she is likely to need the skills of an advanced clinician to assess and manage her complex health care/psycho/social/and developmental needs. In this example, the first patient requires more care time, but the second patient has more complex care needs. Most existing PCSs would score the first patient higher than the second. Because the PINI measures complexity directly, the second patient would receive a PINI score as high or higher than the first patient, thus assuring appropriate identification of the care needs of both patients.
Dependency includes patient needs for assistance in performing basic activities of daily living such as feeding, bathing, toileting, grooming, mobility, and protection from injury due to cognitive and/or communication deficits. The need for assistance in performing activities of daily living is one of the primary reasons that patients are hospitalized, and items measuring these needs are found in some fashion in all PCSs used in acute care.
Severity of illness, the third PINI dimension, refers to the patient's illness in terms of the abnormality and instability of physiological parameters. While not typically included in PCSs used for projecting nurse staffing, research has shown that severity of illness is related to consumption of hospital resources (Horn & Sharkey, 1983). Since nursing resources are a major component of hospital resources, it seems reasonable to include severity of illness as a dimension in a basic measure of patient need for hospital and nursing service.
Like the PINI, the PINAC shown in Figure 3, has three conceptual dimensions: severity of illness, patients' psycho/social needs, and complexity of care (Prescott, 1991).
PINAC Conceptual Dimensions
Severity of Illness Complexity of Care Patient Psycho/Social Needs
Severity of illness has the same meaning as in the PINI, that is the abnormality, instability and seriousness of the patient's medical condition. Psycho/social needs refer to patient/family needs for teaching and emotional support. In the PINI these items were a part of the complexity dimension. In the PINAC they are isolated in a separate dimension given their large importance in ambulatory care. Complexity has the same meaning as in the PINI and refers to the knowledge and skill needed to perform tasks and procedures and to make decisions regarding implementation of the nursing process. The dependency dimension of the PINI was not included in the PINAC as the majority of ambulatory patients can meet basic dependency needs.
In addition to the three conceptual dimensions, the PINAC also contains items describing type of visit. The first item categorizes visits in terms of three variables: (a) nature of the services delivered (limited vs. extensive), (b) stability of the patient (stable vs. unstable), and (c) newness of the patient to the clinic (new vs. returning patient). The second item categorizes visits in terms of six descriptors:
3. Walk in, non-emergency
4. Inpatient consultation
5. Emergency visit
Given the high volume of patient visits of short duration, a very abbreviated classification, such as contained in these visit descriptors, would be both practical and desirable if they were adequately sensitive to distribute patient visits by volume and complexity of resource consumption. Hastings (1987) argued that visit, rather than patient characteristics, may provide meaningful classification in ambulatory care. For this reason, the visit descriptors were added to the other eight items of the PINAC which measure patient care needs.
The PINAC and PINI from which it is derived, can be used for all of the purposes previously identified in Figure 1. For example, scores on individual items can be used pre and post visit as a way of measuring patient status prior to treatment and patient outcome as a result of intervention. Knowledge deficit scores might be high on a patient's initial visit for a new condition. Such scores would indicate a need for patient teaching, the effect of which would be indicated by a lower knowledge deficit score following intervention.
In addition to determining outcomes of care for an individual patient as a guide to the clinician providing care, outcomes for groups of patients can be examined using PINAC item level scores. For example, one could look at the knowledge deficit scores of a group of patients with myocardial infarctions before and after implementing a new teaching protocol or critical pathway.
Another potential use of PINAC data involves using subscores associated with conceptual components of the measure. A complexity subscore, for example, could be useful in making decisions about staff assignment. Clinics with high-complexity scores may need a different staff skill mix than those serving patients with low-complexity needs. Similarly, areas with high psycho/social subscores probably need staff with expertise in patient counseling and teaching to a greater degree than do areas with low or average scores on this dimension.
Finally total PINAC scores can be related to hours of care through work sampling studies to yield the traditional algorithms for projecting staffing. Data from PCSs are useful to the nurse administrator to establish trends which assist in planning and budgeting activities. They should not be used to replace the sound clinical judgment of an experienced nurse manager, and they should not be used to micro-manage a patient care area because sudden changes in patient condition, physician practices, and many other factors affect workload adequacy. PCSs can provide reliable and valid indicators of patient care needs. This information is necessary but not sufficient to effectively manage work in a busy and rapidly changing clinical environment.
As the health care delivery system continues its evolution from the horse and buggy days of fee for service, inpatient-based episodic care to integrated managed care networks based on capitated payment, health care managers and clinicians will face many challenges. One important challenge is to use data wisely and economically. We can no longer afford duplicative, profession-specific systems which are limited in their uses and highly redundant in the patient data solicited. Especially as we move toward an electronic patient record within vertically integrated delivery systems, we need tools which provide valid and reliable, real time, generic patient data which can be collected once but used for a variety of purposes by a variety of different types of practitioners.
To meet the challenges of a multi-use system while simultaneously creating a tool that is feasible for daily use in a busy practice is not an easy task. Pragmatic considerations regarding ease of use, training time, and time associated with use must be considered and balanced against the desire to capture all potentially relevant information.
The PINAC is one effort to create a multiple use PCS that is short and easy to use as well as conceptually based. While no single measure can include data sufficient for all potential uses by all types of providers, both the PINI and PINAC provide clinically meaningful data about basic patient care needs. These data can be used by nurses and other providers interested in and responsible for the dimensions of care measured by these tools. Staff nurses will find PINAC useful to guide assessment of patient needs for care. For example, the tool directs the nurse to assess the patient and/or family for knowledge deficit or needs for emotional support. Subsequent to determining a patient's level of need and providing appropriate nursing interventions to meet the identified need(s), the nurse can use the PINAC to determine the patient response to treatment. In the example given, a patient with a high initial knowledge deficit score should, after effective teaching, have a lower PINAC score on this dimension. Post-intervention scores can thus be used as outcome measures which determine the patient's status at a given point in time and provide information to help the nurse evaluate the effectiveness of the care provided.
Nurse managers and nurse executives can use data from PINAC and PINI for a variety of research, quality assurance, and administrative purposes as illustrated in Figure 1. The ability to express nursing intensity in component parts and match staff mix to patient needs based on the types of patient need is a particularly important administrative use of nursing intensity tools in today's health care environment.
In the second part of this two part series, the development and testing of the PINAC will be described in greater detail.$
American Hospital Association. (1993-1994). Hospital statistics. Chicago: Author.
Averill, R., Goldfield, N., McGuire, J., Bender, J., Mullin, R., & Gregg, L. (1990). Design and development of a prospective payment system for ambulatory care. Wallingford, CT: 3M Health Information Systems.
Beach, L., & Mitchell, T. (1978). A contingency model for selection of decision strategies. Academy of Management Review, 438-449.
Division of Nursing. (1994). The registered nurse population 1992. Health Resources Administration, U.S. Public Health Service, USDHHS. Washington, DC: U.S. Government Printing Office #1994-515-025/03519.
Fetter, R., Averill, R.E, Lichtenstein, J.L., & Freeman, J.L. (1984). Ambulatory visit groups: A framework for measuring productivity in ambulatory care. Health Services Research, 19(3),415-437.
Georgoulakis, J.M., Akins, S.E., Richards, J.D., Gullien, A.C., Gaffney, C.L., Bolling, D.R., Austin, V.R., & Moon, J.P (1990). A comparison of ambulatory classification systems: A preliminary report. Journal of Ambulatory Care Management, 13(3), 39-49.
Gioventti, R, & Mayer, G. (1984). Building confidence in patient classification systems. Nursing Management, 15(8), 31-34.
Goldfield, N. (1993). Ambulatory encounter systems: Implications for payment and quality. Journal of Ambulatory Care Management, 16(2),33-49.
Hastings, C. (1987). Classification issues in ambulatory care nursing. Journal of Ambulatory Care Management, 10(3), 50-64.
Hastings, C., & Muir-Nash, J. (1989). Validation of a taxonomy of ambulatory nursing practice. Nursing Economic$, 7(3),142-149.
Hoffman, F., & Wakefield, D. (1986). Ambulatory care patient classification. Journal of Nursing Administration, 16(4), 23-30.
Horn, S., Buckle, J., & Carver, C. (1988). Ambulatory severity index: Development of an ambulatory case mix system. Journal of Ambulatory Care Management, 11(4), 53-62.
Horn, S., & Sharkey, R (1983). Measuring severity of illness to predict patient resource use within DRGs. Inquiry, 20, 314-321.
Johnson, J. (1989). Quantifying an ambulatory care patient classification instrument. Journal of Nursing Administration, 19(11), 36-42.
Joseph, A. C. (1990). Ambulatory care: An objective assessment. Journal of Nursing Administration, 20(2), 27-33.
Kirsch, E., & Talbott, J. (1990). Outpatient and short-stay patient classification systems. Nursing Management, 21(9), 118-122.
Lion, J., Malbon, A., & Bergman, A. (1987). Ambulatory visit groups: Implications for hospital outpatient departments. Journal of Ambulatory Care Management, 10(1), 56-59.
Parrinello, K. (1987). Accounting for patient acuity in an ambulatory surgery center. Nursing Economic$, 5(4),167-172.
Parrinello, K., Brenner, R, & Vallone, B. (1988). Refining and testing a nursing patient classification instrument in ambulatory care. Nursing Administration Quarterly 13(1), 54-65.
Perrow, C. (1967). A framework for the comparative analysis of organizations. American Social Review, 32, 194-208.
Prescott, P. (1991). Nursing intensity: Needed today for more than staffing. Nursing Economic$, 9(6),409-414.
Prescott, R, & Phillips, C. (1988). Gauging nursing intensity to bring costs to light. Nursing and Health Care, 9(1), 17-22.
Prescott, P.A., Ryan, J.W., Soeken, K.L., Castoor, A.H., Thompson, K.O., & Phillips, C.Y. (1991). The patient intensity for nursing index: A validity assessment. Research in Nursing and Health, 14(3), 213-221.
Richards, J., & Tracy, R. (1984). An assessment process for nursing staff patterns in ambulatory care. Journal of Ambulatory Care Management, 69-79. Schade, J., & Austin, J. (1992). Quantifying ambulatory care activities by time and complexity. Nursing Economic$, 10(3), 183-192.
Schneeweiss, R., Cherkin, D.C., Hart, L.G., Revicki, D.A., Wollstadt, L.J., Stephenson, M.J., Froom, J., Dunn, E.V., Tindall, H.L., Rosenblatt, R.A. (1986). Diagnosis clusters adapted for ICD-9-CM and ICHPPC-2. Journal of Family Practice, 22, 69-72.
Schneeweiss, R., Rosenblatt, R., Cherkin, D., Kirkwood, C.R., & Hart, G. (1983). Diagnosis clusters: A new tool for analyzing the content of ambulatory care. Medical Care, 21(1),105-122.
Schneider, K.C., Lichtenstein, J.L., Freeman, J.L., Newbold, R.C., Fetter, R.B., Gottleib, L., Leaf, RJ., & Portlock, C.S. (1988). Ambulatory visit groups: An outpatient classification system. Journal of Ambulatory Care Management, 11(3), 1-12.
Soeken, K.L., & Prescott, RA. (1991). The patient intensity for nursing index: The measurement model. Research in Nursing and Health, 14(4), 297-304.
Starfield, B., Weiner, J., Mumford, L., & Steinwachs, D. (1991). Ambulatory care groups: A categorization of diagnoses for research and management. Health Services Research, 26(1),53-74.
Tenan, P.M., Fillmore, H.H., Caress, B., Kelly, W.R, Nelson, H., Graziano, D., & Johnson, S.C. (1988). PACs: Classifying ambulatory care patients and services for clinical and financial management. Journal of Ambulatory Care Management, 11(3), 36-53.
Tighe, M.G., Fisher, S.G., Hastings, C., & Heller, B. (1985). A study of the oncology nurse role in ambulatory care. Oncology Nursing Forum, 12(6), 23-27.
Verran, J. (1986a). Patient classification in ambulatory care. Nursing Economic$, 4(5),147-251.
Verran, J. (1986b). Testing a classification instrument for the ambulatory care setting. Research in Nursing and Health, 9, 279-287.
Weiner, J.P., Starfield, B.H., Steinwachs, D.M., & Mumford, L.M. (1991). Development and application of a population-oriented measure of ambulatory care case mix. Medical Care, 29(5), 452-472.
PATRICIA A. PRESCOTT, PhD, RN, FAAN, is Professor, University of Maryland School of Nursing, Baltimore, MD.
KAREN L. SOEKEN, PhD, is Associate Professor, University of Maryland, School of Nursing, Baltimore, MD.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Measuring Nursing Intensity in Ambulatory Care, part 1|
|Author:||Prescott, Patricia A.; Soeken, Karen L.|
|Date:||Jan 1, 1996|
|Previous Article:||Case management: implementing the vision.|
|Next Article:||Nurses' use and delegation of indirect care interventions.|