Susan L. Jones

A comparison of imputation techniques for handling missing data.

Researchers are commonly faced with the problem of missing data. This article presents theoretical and empirical information for the selection and application of approaches for handling missing data on a single variable. An actual data set of 492 cases with no missing values was used to create a simulated yet realistic data set with missing at random (MAR) data. The authors compare and contrast five approaches (listwise deletion, mean substitution, simple regression, regression with an error term, and the expectation maximization [EM] algorithm) for dealing with missing data, and compare the effects of each method on descriptive statistics and correlation coefficients for the imputed data (n = 96) and the entire sample (n = 492) when imputed data are included. All methods had limitations, although our findings suggest that mean substitution was the least effective and that regression with an error term and the EM algorithm produced estimates closest to those of the original variables.

Structural equation modeling and its relationship to multiple regression and factor analysis

Using a conceptual and nontechnical approach, the meaning of structural equation modeling (SEM) and the similarities to, and differences from, more commonly used procedures such as correlation, regression, path analysis, and factor analysis are explained. Application of the statistical technique is presented using data from a study of the relationships among stresses, strains, and physical health in a random sample of 492 community-dwelling elders aged 65 and older. Advantages of each statistical procedure are described. Theoretical issues related to the use of each procedure are presented with emphasis on the need for a sound theoretical model and match between the statistical procedure and the aims of the analysis.

OSTEOPOROSIS PREVENTION IN YOUNG WOMEN

PURPOSE To assess whether young women who participate in an osteoporosis prevention program based on the Health Belief (Rosenstock, 1966) and Self-Efficacy Models (Bandura, 1977) demonstrate higher levels of knowledge regarding osteoporosis prevention than young women who do not participate in such a program. DESIGN A classic experimental design with one treatment group and one control group was used to test the efficacy of the osteoporosis prevention program. Pretest data on knowledge, health belief attitudes (7 subscales), and self-efficacy (2 subscales) were collected in the treatment and control groups. SAMPLE A convenience sample of 31 young college women were randomly assigned to an experimental group or to a control group to receive an osteoporosis prevention program. METHOD Subjects in the experimental and control groups completed the Osteoporosis Knowledge Test, the Osteoporosis Health Belief Scale, and the Osteoporosis Self-Efficacy Scale (Kim et al., 1991) at two times. The experimental group received an osteoporosis prevention program. MAIN RESEARCH CLASSIFICATIONS Osteoporosis, Health Belief, Self-Efficacy, Womens Health. FINDINGS Subjects in the experimental group had significantly higher knowledge and health belief scores after receiving the intervention than their pretest scores while subjects in the control group had no change in scores. CONCLUSION The osteoporosis program was effective in increasing awareness of osteoporosis prevention in this group of young women. IMPLICATIONS FOR NURSING RESEARCH The results may be useful for developing young womens awareness and knowledge of osteoporosis prevention. Future research could include developing osteoporosis prevention programs at an earlier age when girls are in grade school and junior high school. Teaching health promotion strategies for bone health is essential to all women across the life span.

Health belief model intervention to increase compliance with emergency department patients.

The effects on compliance of clinical and telephone intervention, based on the Health Belief Model (HBM), were investigated for 842 Emergency Department (ED) patients. The influence of mediating variables on compliance was also examined. Compliance was defined operationally as follow-through on a recommended referral originating in the ED. The study design was a 2 X 2 X11 factorial design, in which the first factor was the HBM clinical intervention, the second was the HBM telephone intervention, and the third was the type of presenting problem. Patients were randomly assigned to one of four intervention groups, with all nursing care, interventions, and follow-up telephone calls being done by the research nurse. The HBM clinical, telephone, and combination clinical/telephone interventions were strongly associated with increased compliance in the 11 presenting problems. Availability of child care, knowledge of presenting problem, nature and duration of the illness, and demographic variables (such as age of the patient) were also related to compliance

Improving compliance for asthmatic patients visiting the emergency department using a health belief model intervention.

The sample included 74 asthmatic patients presenting to an emergency department with an acute asthmatic attack. The purposes of the study were: to assess demographic and situational variables associated with compliance, and to test the impact of a Health Belief Model (HBM) intervention to increase compliance. Compliance was operationalized as making and/or keeping a follow-up referral appointment for the patients asthmatic condition. Demographic and situational variables associated with compliance included age, sex, marital status, previous treatment experience, seriousness of the asthmatic condition, and need for child care. Most important, the HBM intervention was very effective in increasing compliance in both making and keeping a follow-up referral appointment.

A Randomized Trial to Improve Compliance in Urinary Tract Infection Patients in the Emergency Department

A randomized trial was used to evaluate two forms of a health belief model (HBM) intervention aimed at increasing compliance among 139 patients with urinary tract infections who presented to the emergency department. Patients who received an HBM clinical intervention administered in the ED, HBM telephone intervention two to four days after the ED visit, or both interventions were much more likely than control patients to both schedule and keep a follow-up referral appointment. Both the clinical and telephone interventions were designed to increase perceived susceptibility to complications of the urinary tract infection, seriousness of the complications, and benefits and costs of action. Other factors predicting compliance include age of the patient, urgency of the urinary tract infection as rated subjectively by the emergency physician, need for child care, whether transportation was available, and questioning the nurse about the referral appointment.

Compliance in acute and chronic patients receiving a health belief model intervention in the emergency department.

We compared compliance behavior in acute (n = 670) and chronic (n = 172) patients visiting the Emergency Department (E.D.) of an acute care, private hospital; compliance was operationalized as scheduling or keeping a follow-up referral appointment originating in the Emergency Department. The purposes of the study were to assess demographic and illness experience differences among patients and to assess the relative impact of a Health Belief Model Intervention on compliance in acute versus chronic patients. Within the acute and chronic categories patients were randomized into four experimental groups to test the efficacy of the HBM intervention. Although chronic patients were generally more compliant than acute patients, the HBM interventions were equally effective in increasing compliance in both groups. Chronic patients were older, more likely to have had previous treatment for an illness, and more likely to regard the potential consequences of their illness as serious; these factors were associated with greater compliance.

Advantages and disadvantages of collaborative research: a university and behavioral health care provider's experience.

This article describes the issues involved in a collaborative research effort between a university and a behavioral health care provider. We describe the collaborative process and highlight advantages and disadvantages of such collaboration. Specific examples are used to show how the collaborative process between a university team and clinical team can integrate theory, research, and clinical practice.

The psychiatric clinical nurse specialist/nurse practitioner: An example of a combined role

The purpose of this article is to provide an historical overview of the clinical nurse specialist and nurse practitioner roles in psychiatric nursing. This article is written as a follow-up to that of Dyer, Hammill, Regan-Kubinski, Yurick, & Kobert, (1997), Archives of Psychiatric Nursing, 11, 2-11, in which a new paradigm for delivering comprehensive mental and general health services is conceptualized, i.e., the psychiatric-primary care nurse practitioner role. In this article an alternative model is presented: the psychiatric clinical nurse specialist/nurse practitioner role. Similarities and differences in the two roles are discussed.

Evaluation of television consultations between a large neonatal care hospital and a community hospital.

Two-way television consultations between community hospital nurses and neonatologists at a nearby teaching hospital were conducted over a two- and one-half year period of time and were evaluated with respect to a baseline time period in which the television was not available. Screening for illness and prematurity in neonates occurred in a high risk population residing in a black, economically deprived, innercity area. Outcomes including transfer of sick babies from the community hospital to the large teaching hospital are analyzed in relation to prenatal maternal risk characteristics, Apgar scores, birth weight and gestational age. Evidence available suggests that television consultations facilitated formation of appropriate criteria for interhospital transfer and that routine clinical screening tests were performed more consistently following initiation of interhospital consultations.