Psychometrics of Halpin Nausea and Vomiting Scales for Use in Clinical Practice

Tuesday, 23 July 2013: 3:30 PM

Angela P. Halpin, RN, MN, CNS
School of Nursing, University of California, Los Angeles, Los Angeles, CA
Loucine M. Huckabay, PhD, PNP, RN, FAAN
School of Nursing, California State University Long Beach, Long Beach, CA

Learning Objective 1: A methodological study description of psychometric properties for 0-5, Halpin Nausea and Vomiting (HNV) scales with descriptors, so learners can replicate the study.

Learning Objective 2: This is a presentation to identify HNV findings and results on reliability, validity, and sensitivity, so learners will be confident in use of scales.

Purpose:  To communicate psychometric properties of the new 0-5 Halpin Nausea and Vomiting (HNV) scales with descriptors, so learners can use scales in clinical practice. 

Methods: A methodological design was applied to establish psychometrics of HNV scales using three groups of patients, N = 163.  Group I was admitted with a diagnosis of NV, n=54.  Group II consisted of cancer patients scheduled for chemotherapy, n=52.  Group III was a control group with medical-surgical diagnoses and no expected NV, n=57.   

Results: HNV scales had a high inter-rater reliability (Kappa test=.851, p<.001). Concurrent validity was established between the HNV scales and Morrow’s worst nausea ratings, significant (r=.318, p=.03). HNV was found to measure fine differences between and within groups, establishing sensitivity.

Chi square tests were non-significant for ethnicity, gender, coronary artery disease, hypertension, diabetes, congestive heart failure, other medical surgical diagnosis and risk for PONV.  Group I had more nausea patients (X2=63.94, p<.001) and higher degree of the diagnosis of pancreatitis (X2=8.72, p=.01).  Group II had more cancer (X2=76.00, p<.001), and chemo patients (X2=93.06, p<.001).  In terms of age, one way analysis of variance indicated that the groups are significantly different (F=6.12, p=.003).  Scheffe’s test indicates that group I (the NV group) were significantly younger (p<.05) than the control and the cancer group.  Group II and III were homogenous with no significantly difference in age.

Conclusion: Accurate assessment of NV with valid and reliable scales can lead to more accurate and timely treatment.  Alleviation of patient’s discomfort is improved by scales with descriptors because the specific patient response has an individual measurement.  Patients can rank and use the descriptors to communicate their symptom severity. Knowing the intensity of NV predictors in surgical and chemotherapy patients supports use of HNV scales to promote symptom management and improves quality of life.