Open AccessThe Lymphatic Malformation Function (LMF) Instrument
Author(s) -
Kirkham Erin M.,
Edwards Todd C.,
Weaver Edward M.,
Balakrishnan Karthik,
Perkins Jonathan A.
Publication year - 2015
Publication title -
otolaryngology–head and neck surgery
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.232
H-Index - 121
eISSN - 1097-6817
pISSN - 0194-5998
DOI - 10.1177/0194599815594776
Subject(s) - cronbach's alpha , intraclass correlation , exploratory factor analysis , criterion validity , construct validity , psychology , reliability (semiconductor) , medicine , statistics , psychometrics , clinical psychology , mathematics , power (physics) , physics , quantum mechanics
Objective The Lymphatic Malformation Function (LMF) instrument is a preliminary parent‐report assessment designed to measure outcomes in children with cervicofacial lymphatic malformation (LM). This study aimed to assess the measurement properties of the LMF, refine it, test criterion validity, and evaluate the test‐retest reliability. Study Design Cross‐sectional. Setting Two pediatric tertiary referral centers. Subjects Parents of 60 children from 6 months to 15 years old with cervicofacial LM. Methods Parents were recruited via mail and online. The LMF was administered on paper or online initially and again within 21 days. Response distributions and interitem correlations were examined for item reduction. Exploratory factor analysis was conducted on retained items. Cronbach’s α, Spearman correlation, and intraclass correlation (ICC) coefficients were calculated to test internal consistency, criterion validity (compared to stage), and test‐retest reliability, respectively. Results One item was removed due to a floor effect. The response scale was collapsed from a 5‐point scale to a 3‐point scale due to skewness. Six items were discarded due to redundancy (interitem correlations >0.7); 2 items were discarded due to factor loadings <0.4. Exploratory factor analysis revealed a 2‐factor structure explaining 84% of variance, and the domains Signs and Impacts had good internal consistency (all Cronbach’s α >0.80 and <0.90), significant association with stage ( P <. 05), and good overall test‐retest reliability (ICC, 0.82). Conclusion The LMF has been refined into a 12‐item, 2‐domain instrument measuring LM‐specific signs and impacts with internal consistency, criterion validity, and test‐retest reliability.