Soft classification and regression analysis of audiometric phenotypes of age-related hearing loss.

Ce Yang, Benjamin Langworthy, Sharon Curhan, Kenneth I Vaden, Gary Curhan, Judy R Dubno, Molin Wang
Author Information
  1. Ce Yang: Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA 02115, United States. ORCID
  2. Benjamin Langworthy: Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA 02115, United States.
  3. Sharon Curhan: Harvard Medical School, Boston, MA 02115, United States.
  4. Kenneth I Vaden: Hearing Research Program, Department of Otolaryngology-Head and Neck Surgery, Medical University of South Carolina, Charleston, SC 29425, United States.
  5. Gary Curhan: Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA 02115, United States.
  6. Judy R Dubno: Hearing Research Program, Department of Otolaryngology-Head and Neck Surgery, Medical University of South Carolina, Charleston, SC 29425, United States. ORCID
  7. Molin Wang: Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA 02115, United States. ORCID

Abstract

Age-related hearing loss has a complex etiology. Researchers have made efforts to classify relevant audiometric phenotypes, aiming to enhance medical interventions and improve hearing health. We leveraged existing pattern analyses of age-related hearing loss and implemented the phenotype classification via quadratic discriminant analysis (QDA). We herein propose a method for analyzing the exposure effects on the soft classification probabilities of the phenotypes via estimating equations. Under reasonable assumptions, the estimating equations are unbiased and lead to consistent estimators. The resulting estimator had good finite sample performances in simulation studies. As an illustrative example, we applied our proposed methods to assess the association between a dietary intake pattern, assessed as adherence scores for the dietary approaches to stop hypertension diet calculated using validated food-frequency questionnaires, and audiometric phenotypes (older-normal, metabolic, sensory, and metabolic plus sensory), determined based on data obtained in the Nurses' Health Study II Conservation of Hearing Study, the Audiology Assessment Arm. Our findings suggested that participants with a more healthful dietary pattern were less likely to develop the metabolic plus sensory phenotype of age-related hearing loss.

Keywords

References

  1. J Assoc Res Otolaryngol. 2017 Apr;18(2):371-385 [PMID: 27830350]
  2. Arch Intern Med. 2011 Nov 14;171(20):1851-2 [PMID: 22083573]
  3. Sci Rep. 2020 Apr 21;10(1):6704 [PMID: 32317648]
  4. Stat Med. 2022 Nov 20;41(26):5335-5348 [PMID: 36125070]
  5. J Assoc Res Otolaryngol. 2013 Oct;14(5):687-701 [PMID: 23740184]
  6. Front Med (Lausanne). 2021 Mar 25;8:662045 [PMID: 33842516]
  7. Stat Med. 1994 Jun 15;13(11):1149-63 [PMID: 8091041]
  8. Hypertension. 2006 Feb;47(2):296-308 [PMID: 16434724]
  9. Hear Res. 2010 Jun 1;264(1-2):10-20 [PMID: 20144701]
  10. J Am Stat Assoc. 2011 Mar 1;106(493):166-177 [PMID: 22162896]
  11. Laryngoscope Investig Otolaryngol. 2018 Oct 30;3(6):478-485 [PMID: 30599033]
  12. Biometrics. 1993 Dec;49(4):1033-44 [PMID: 8117899]
  13. Laryngoscope. 2006 Sep;116(9 Pt 3 Suppl 112):1-12 [PMID: 16946668]
  14. Am J Epidemiol. 2020 Mar 2;189(3):204-214 [PMID: 31608356]
  15. Bull Math Biol. 1990;52(1-2):99-115; discussion 73-97 [PMID: 2185863]
  16. Br J Nutr. 2013 Jun;109(11):2079-88 [PMID: 23200201]
  17. J Nutr. 2018 Jun 1;148(6):944-951 [PMID: 29757402]
  18. Trends Hear. 2018 Jan-Dec;22:2331216518797848 [PMID: 30198420]

Grants

  1. R01 DC017717/NIH HHS

MeSH Term

Humans
Causality
Regression Analysis
Hearing Loss
Phenotype

Word Cloud

Created with Highcharts 10.0.0hearingphenotypeslossaudiometricclassificationpatternage-relatedanalysisestimatingdietarymetabolicsensoryphenotypeviaquadraticdiscriminantsoftplusdataStudyAge-relatedcomplexetiologyResearchersmadeeffortsclassifyrelevantaimingenhancemedicalinterventionsimprovehealthleveragedexistinganalysesimplementedQDAhereinproposemethodanalyzingexposureeffectsprobabilitiesequationsreasonableassumptionsequations areunbiasedleadconsistentestimatorsresultingestimatorgoodfinitesampleperformancessimulationstudiesillustrativeexampleappliedproposedmethodsassessassociationintakeassessedadherencescoresapproachesstophypertensiondietcalculatedusingvalidatedfood-frequencyquestionnairesolder-normaldeterminedbasedobtainedNurses'HealthIIConservationHearingAudiologyAssessmentArmfindingssuggestedparticipantshealthfullesslikelydevelopSoftregressionclusteredgeneralizedequation

Similar Articles

Cited By