Academic Journal
Προβολή σε EDS

Identification of cognitive subtypes and risk networks in Chinese older adults: based on latent profile analysis and explainable machine learning.

Λεπτομέρειες βιβλιογραφικής εγγραφής
Τίτλος: Identification of cognitive subtypes and risk networks in Chinese older adults: based on latent profile analysis and explainable machine learning.
Συγγραφείς: Lai L; The Institute of Infectious Disease and Vaccine, School of Public Health, Zhejiang Chinese Medical University, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China.; Center for Vaccine Impact Assessment, Key Laboratory for Quality Monitoring and Evaluation of Vaccines and Biological Products, National Medical Products Administration, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China., Dong B; The Institute of Infectious Disease and Vaccine, School of Public Health, Zhejiang Chinese Medical University, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China.; Center for Vaccine Impact Assessment, Key Laboratory for Quality Monitoring and Evaluation of Vaccines and Biological Products, National Medical Products Administration, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China., Fu C; The Institute of Infectious Disease and Vaccine, School of Public Health, Zhejiang Chinese Medical University, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China. fuchuanxi@zcmu.edu.cn.; Center for Vaccine Impact Assessment, Key Laboratory for Quality Monitoring and Evaluation of Vaccines and Biological Products, National Medical Products Administration, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China. fuchuanxi@zcmu.edu.cn.
Πηγή: BMC geriatrics [BMC Geriatr] 2026 Aug 03; Vol. 26 (1). Date of Electronic Publication: 2026 Aug 03.
Τύπος έκδοσης: Journal Article
Γλώσσα: English
Στοιχεία περιοδικού: Publisher: BioMed Central Country of Publication: England NLM ID: 100968548 Publication Model: Electronic Cited Medium: Internet ISSN: 1471-2318 (Electronic) Linking ISSN: 14712318 NLM ISO Abbreviation: BMC Geriatr Subsets: MEDLINE
Imprint Name(s): Original Publication: London : BioMed Central, [2001]-
Ιατρικοί όροι (MeSH): Cognitive Dysfunction*/diagnosis , Cognitive Dysfunction*/psychology , Cognitive Dysfunction*/epidemiology , Cognitive Dysfunction*/classification , Cognition*/physiology , Cognition*/classification , Machine Learning*, China/epidemiology ; Humans ; Female ; Aged ; Male ; Longitudinal Studies ; Risk Factors ; Middle Aged ; Random Forest ; Predictive Learning Models ; Aged, 80 and over ; East Asian People
Περίληψη: Background: Traditional cognitive screening often relies on total scores, which can easily mask asynchronous impairment patterns across different cognitive domains. In particular, calculation ability, a sensitive indicator of executive function, is frequently misattributed to normal aging and overlooked clinically. This study aims to utilize data-driven methods to identify specific concealed cognitive phenotypes in the older population and deconstruct their underlying sociodemographic and clinical driving mechanisms.
Methods: This study included 7,159 older participants from the China Health and Retirement Longitudinal Study (CHARLS). Latent Profile Analysis (LPA) was employed to identify latent cognitive subtypes based on three dimensions: orientation, calculation, and memory. Prediction models were constructed using Boruta feature selection and the Random Forest algorithm. Furthermore, SHAP (SHapley Additive exPlanations) and Gaussian Graphical Models (GGM) were utilized to quantify the nonlinear contributions and network interactions of risk factors.
Results: LPA identified a unique cognitive impairment subtype termed Calculation-Dominant Impairment, accounting for 24.6% of the sample population. Although the majority of respondents were classified as cognitively intact (55.0%), this specific subgroup exhibited a disproportionately severe decline in calculation function (V-shaped deficit), while episodic memory was relatively preserved. The Random Forest model demonstrated superior performance in identifying this subtype (AUC = 0.781), significantly outperforming traditional linear models, and achieved high specificity (specificity = 0.960). Interpretability analysis revealed a pattern dominated by significant social determinants: education level and internet use were the primary predictors, with importance weights significantly exceeding those of somatic comorbidities such as hypertension and diabetes. Network analysis further confirmed that internet use acts as a critical "bridge node" connecting age and education level, exerting a potential cognitive reserve effect in mitigating calculation decline.
Conclusion: This study defines a cognitive impairment subtype primarily driven by education poverty and digital exclusion, which is prone to be missed in routine memory-centric screenings. The findings suggest that biomedical interventions alone may be insufficient to prevent such decline; promoting digital inclusion among older adults should be regarded as a critical public health strategy to compensate for early-life educational deficits and prevent the decline of calculation and executive functions.
(© 2026. The Author(s).)
Competing Interests: Declarations. Ethics approval and consent to participate: The data used in this study were derived from the China Health and Retirement Longitudinal Study (CHARLS). The original CHARLS survey protocol was approved by the Biomedical Ethics Review Committee of Peking University (Approval No. IRB00001052-11015), and all respondents signed informed consent forms prior to the survey. All methods were carried out in accordance with relevant guidelines and regulations (e.g., Declaration of Helsinki). Consent for publication: Not applicable. Competing interests: The authors declare no competing interests.
References: Bioinformatics. 2001 Jun;17(6):520-5. (PMID: 11395428)
Lancet. 2015 Oct 24;386(10004):1698-706. (PMID: 26595643)
Psychol Methods. 2018 Dec;23(4):617-634. (PMID: 29595293)
Behav Res Methods. 2018 Feb;50(1):195-212. (PMID: 28342071)
Neuroimage. 2011 Feb 1;54(3):2382-93. (PMID: 20946958)
Am J Epidemiol. 2012 Feb 1;175(3):210-7. (PMID: 22262640)
Psychol Aging. 2019 Feb;34(1):17-24. (PMID: 30211596)
J Gerontol A Biol Sci Med Sci. 2018 Oct 8;73(11):1519-1524. (PMID: 29211879)
Front Public Health. 2025 Feb 18;13:1510418. (PMID: 40041182)
J Gerontol B Psychol Sci Soc Sci. 1997 May;52 Spec No:37-48. (PMID: 9215356)
Int J Epidemiol. 2014 Apr;43(2):576-85. (PMID: 24671021)
Neurology. 2010 Mar 23;74(12):951-5. (PMID: 20308679)
Brain. 2015 Sep;138(Pt 9):2732-49. (PMID: 26141491)
Biometrics. 1988 Sep;44(3):837-45. (PMID: 3203132)
Int J Public Health. 2023 Jan 19;68:1605322. (PMID: 36798738)
Int J Epidemiol. 2014 Feb;43(1):61-8. (PMID: 23243115)
Lancet. 2024 Aug 10;404(10452):572-628. (PMID: 39096926)
Med Decis Making. 2006 Nov-Dec;26(6):565-74. (PMID: 17099194)
Int J Geriatr Psychiatry. 2000 Aug;15(8):677-9. (PMID: 10960879)
BMJ. 2009 Jun 29;338:b2393. (PMID: 19564179)
Contributed Indexing: Keywords: Calculation-dominant impairment; Cognitive phenotypes; Cognitive reserve; Digital inclusion; Explainable machine learning; Latent Profile Analysis
Entry Date(s): Date Created: 20261009 Date Completed: 20261010 Latest Revision: 20261011
Update Code: 20261011
DOI: 10.1186/s12877-026-08079-1
PMID: 42855675
Βάση Δεδομένων: MEDLINE
FullText Text:
  Availability: 0
CustomLinks:
  – Url: https://resolver.ebsco.com/c/fiv2js/result?sid=EBSCO:cmedm&genre=article&issn=14712318&ISBN=&volume=26&issue=1&date=20260803&spage=&pages=&title=BMC geriatrics&atitle=Identification%20of%20cognitive%20subtypes%20and%20risk%20networks%20in%20Chinese%20older%20adults%3A%20based%20on%20latent%20profile%20analysis%20and%20explainable%20machine%20learning.&aulast=Lai%20L&id=DOI:10.1186/s12877-026-08079-1
    Name: Full Text Finder (for New FTF UI) (ns324271)
    Category: fullText
    Text: Full Text Finder
    MouseOverText: Full Text Finder
Header DbId: cmedm
DbLabel: MEDLINE
An: 42855675
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Identification of cognitive subtypes and risk networks in Chinese older adults: based on latent profile analysis and explainable machine learning.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AU" term="%22Lai+L%22">Lai L</searchLink>; The Institute of Infectious Disease and Vaccine, School of Public Health, Zhejiang Chinese Medical University, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China.; Center for Vaccine Impact Assessment, Key Laboratory for Quality Monitoring and Evaluation of Vaccines and Biological Products, National Medical Products Administration, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China.<br /><searchLink fieldCode="AU" term="%22Dong+B%22">Dong B</searchLink>; The Institute of Infectious Disease and Vaccine, School of Public Health, Zhejiang Chinese Medical University, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China.; Center for Vaccine Impact Assessment, Key Laboratory for Quality Monitoring and Evaluation of Vaccines and Biological Products, National Medical Products Administration, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China.<br /><searchLink fieldCode="AU" term="%22Fu+C%22">Fu C</searchLink>; The Institute of Infectious Disease and Vaccine, School of Public Health, Zhejiang Chinese Medical University, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China. fuchuanxi@zcmu.edu.cn.; Center for Vaccine Impact Assessment, Key Laboratory for Quality Monitoring and Evaluation of Vaccines and Biological Products, National Medical Products Administration, 548 Binwen Road, Binjiang District, Hangzhou, Zhejiang, 310053, China. fuchuanxi@zcmu.edu.cn.
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22100968548%22">BMC geriatrics</searchLink> [BMC Geriatr] 2026 Aug 03; Vol. 26 (1). <i>Date of Electronic Publication: </i>2026 Aug 03.
– Name: TypePub
  Label: Publication Type
  Group: TypPub
  Data: Journal Article
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: TitleSource
  Label: Journal Info
  Group: Src
  Data: <i>Publisher: </i><searchLink fieldCode="PB" term="%22BioMed+Central%22">BioMed Central </searchLink><i>Country of Publication: </i>England <i>NLM ID: </i>100968548 <i>Publication Model: </i>Electronic <i>Cited Medium: </i>Internet <i>ISSN: </i>1471-2318 (Electronic) <i>Linking ISSN: </i><searchLink fieldCode="IS" term="%2214712318%22">14712318 </searchLink><i>NLM ISO Abbreviation: </i>BMC Geriatr <i>Subsets: </i>MEDLINE
– Name: PublisherInfo
  Label: Imprint Name(s)
  Group: PubInfo
  Data: <i>Original Publication</i>: London : BioMed Central, [2001]-
– Name: SubjectMESH
  Label: MeSH Terms
  Group: Su
  Data: <searchLink fieldCode="MM" term="%22Cognitive+Dysfunction%22">Cognitive Dysfunction*</searchLink>/<searchLink fieldCode="MM" term="%22Cognitive+Dysfunction+diagnosis%22">diagnosis</searchLink> <br /><searchLink fieldCode="MM" term="%22Cognitive+Dysfunction%22">Cognitive Dysfunction*</searchLink>/<searchLink fieldCode="MM" term="%22Cognitive+Dysfunction+psychology%22">psychology</searchLink> <br /><searchLink fieldCode="MM" term="%22Cognitive+Dysfunction%22">Cognitive Dysfunction*</searchLink>/<searchLink fieldCode="MM" term="%22Cognitive+Dysfunction+epidemiology%22">epidemiology</searchLink> <br /><searchLink fieldCode="MM" term="%22Cognitive+Dysfunction%22">Cognitive Dysfunction*</searchLink>/<searchLink fieldCode="MM" term="%22Cognitive+Dysfunction+classification%22">classification</searchLink> <br /><searchLink fieldCode="MM" term="%22Cognition%22">Cognition*</searchLink>/<searchLink fieldCode="MM" term="%22Cognition+physiology%22">physiology</searchLink> <br /><searchLink fieldCode="MM" term="%22Cognition%22">Cognition*</searchLink>/<searchLink fieldCode="MM" term="%22Cognition+classification%22">classification</searchLink> <br /><searchLink fieldCode="MM" term="%22Machine+Learning%22">Machine Learning*</searchLink><br /><searchLink fieldCode="MH" term="%22China%22">China</searchLink>/<searchLink fieldCode="MH" term="%22China+epidemiology%22">epidemiology</searchLink> ; <searchLink fieldCode="MH" term="%22Humans%22">Humans</searchLink> ; <searchLink fieldCode="MH" term="%22Female%22">Female</searchLink> ; <searchLink fieldCode="MH" term="%22Aged%22">Aged</searchLink> ; <searchLink fieldCode="MH" term="%22Male%22">Male</searchLink> ; <searchLink fieldCode="MH" term="%22Longitudinal+Studies%22">Longitudinal Studies</searchLink> ; <searchLink fieldCode="MH" term="%22Risk+Factors%22">Risk Factors</searchLink> ; <searchLink fieldCode="MH" term="%22Middle+Aged%22">Middle Aged</searchLink> ; <searchLink fieldCode="MH" term="%22Random+Forest%22">Random Forest</searchLink> ; <searchLink fieldCode="MH" term="%22Predictive+Learning+Models%22">Predictive Learning Models</searchLink> ; <searchLink fieldCode="MH" term="%22Aged%2C+80+and+over%22">Aged, 80 and over</searchLink> ; <searchLink fieldCode="MH" term="%22East+Asian+People%22">East Asian People</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Background: Traditional cognitive screening often relies on total scores, which can easily mask asynchronous impairment patterns across different cognitive domains. In particular, calculation ability, a sensitive indicator of executive function, is frequently misattributed to normal aging and overlooked clinically. This study aims to utilize data-driven methods to identify specific concealed cognitive phenotypes in the older population and deconstruct their underlying sociodemographic and clinical driving mechanisms.<br />Methods: This study included 7,159 older participants from the China Health and Retirement Longitudinal Study (CHARLS). Latent Profile Analysis (LPA) was employed to identify latent cognitive subtypes based on three dimensions: orientation, calculation, and memory. Prediction models were constructed using Boruta feature selection and the Random Forest algorithm. Furthermore, SHAP (SHapley Additive exPlanations) and Gaussian Graphical Models (GGM) were utilized to quantify the nonlinear contributions and network interactions of risk factors.<br />Results: LPA identified a unique cognitive impairment subtype termed Calculation-Dominant Impairment, accounting for 24.6% of the sample population. Although the majority of respondents were classified as cognitively intact (55.0%), this specific subgroup exhibited a disproportionately severe decline in calculation function (V-shaped deficit), while episodic memory was relatively preserved. The Random Forest model demonstrated superior performance in identifying this subtype (AUC = 0.781), significantly outperforming traditional linear models, and achieved high specificity (specificity = 0.960). Interpretability analysis revealed a pattern dominated by significant social determinants: education level and internet use were the primary predictors, with importance weights significantly exceeding those of somatic comorbidities such as hypertension and diabetes. Network analysis further confirmed that internet use acts as a critical "bridge node" connecting age and education level, exerting a potential cognitive reserve effect in mitigating calculation decline.<br />Conclusion: This study defines a cognitive impairment subtype primarily driven by education poverty and digital exclusion, which is prone to be missed in routine memory-centric screenings. The findings suggest that biomedical interventions alone may be insufficient to prevent such decline; promoting digital inclusion among older adults should be regarded as a critical public health strategy to compensate for early-life educational deficits and prevent the decline of calculation and executive functions.<br /> (© 2026. The Author(s).)
– Name: Abstract
  Label: Competing Interests
  Group: Ab
  Data: Declarations. Ethics approval and consent to participate: The data used in this study were derived from the China Health and Retirement Longitudinal Study (CHARLS). The original CHARLS survey protocol was approved by the Biomedical Ethics Review Committee of Peking University (Approval No. IRB00001052-11015), and all respondents signed informed consent forms prior to the survey. All methods were carried out in accordance with relevant guidelines and regulations (e.g., Declaration of Helsinki). Consent for publication: Not applicable. Competing interests: The authors declare no competing interests.
– Name: Ref
  Label: References
  Group: RefInfo
  Data: Bioinformatics. 2001 Jun;17(6):520-5. (PMID: <searchLink fieldCode="PM" term="%2211395428%22">11395428)</searchLink><br />Lancet. 2015 Oct 24;386(10004):1698-706. (PMID: <searchLink fieldCode="PM" term="%2226595643%22">26595643)</searchLink><br />Psychol Methods. 2018 Dec;23(4):617-634. (PMID: <searchLink fieldCode="PM" term="%2229595293%22">29595293)</searchLink><br />Behav Res Methods. 2018 Feb;50(1):195-212. (PMID: <searchLink fieldCode="PM" term="%2228342071%22">28342071)</searchLink><br />Neuroimage. 2011 Feb 1;54(3):2382-93. (PMID: <searchLink fieldCode="PM" term="%2220946958%22">20946958)</searchLink><br />Am J Epidemiol. 2012 Feb 1;175(3):210-7. (PMID: <searchLink fieldCode="PM" term="%2222262640%22">22262640)</searchLink><br />Psychol Aging. 2019 Feb;34(1):17-24. (PMID: <searchLink fieldCode="PM" term="%2230211596%22">30211596)</searchLink><br />J Gerontol A Biol Sci Med Sci. 2018 Oct 8;73(11):1519-1524. (PMID: <searchLink fieldCode="PM" term="%2229211879%22">29211879)</searchLink><br />Front Public Health. 2025 Feb 18;13:1510418. (PMID: <searchLink fieldCode="PM" term="%2240041182%22">40041182)</searchLink><br />J Gerontol B Psychol Sci Soc Sci. 1997 May;52 Spec No:37-48. (PMID: <searchLink fieldCode="PM" term="%229215356%22">9215356)</searchLink><br />Int J Epidemiol. 2014 Apr;43(2):576-85. (PMID: <searchLink fieldCode="PM" term="%2224671021%22">24671021)</searchLink><br />Neurology. 2010 Mar 23;74(12):951-5. (PMID: <searchLink fieldCode="PM" term="%2220308679%22">20308679)</searchLink><br />Brain. 2015 Sep;138(Pt 9):2732-49. (PMID: <searchLink fieldCode="PM" term="%2226141491%22">26141491)</searchLink><br />Biometrics. 1988 Sep;44(3):837-45. (PMID: <searchLink fieldCode="PM" term="%223203132%22">3203132)</searchLink><br />Int J Public Health. 2023 Jan 19;68:1605322. (PMID: <searchLink fieldCode="PM" term="%2236798738%22">36798738)</searchLink><br />Int J Epidemiol. 2014 Feb;43(1):61-8. (PMID: <searchLink fieldCode="PM" term="%2223243115%22">23243115)</searchLink><br />Lancet. 2024 Aug 10;404(10452):572-628. (PMID: <searchLink fieldCode="PM" term="%2239096926%22">39096926)</searchLink><br />Med Decis Making. 2006 Nov-Dec;26(6):565-74. (PMID: <searchLink fieldCode="PM" term="%2217099194%22">17099194)</searchLink><br />Int J Geriatr Psychiatry. 2000 Aug;15(8):677-9. (PMID: <searchLink fieldCode="PM" term="%2210960879%22">10960879)</searchLink><br />BMJ. 2009 Jun 29;338:b2393. (PMID: <searchLink fieldCode="PM" term="%2219564179%22">19564179)</searchLink>
– Name: SubjectMinor
  Label: Contributed Indexing
  Group:
  Data: <i>Keywords: </i>Calculation-dominant impairment; Cognitive phenotypes; Cognitive reserve; Digital inclusion; Explainable machine learning; Latent Profile Analysis
– Name: DateEntry
  Label: Entry Date(s)
  Group: Date
  Data: <i>Date Created: </i>20261009 <i>Date Completed: </i>20261010 <i>Latest Revision: </i>20261011
– Name: DateUpdate
  Label: Update Code
  Group: Date
  Data: 20261011
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1186/s12877-026-08079-1
– Name: AN
  Label: PMID
  Group: ID
  Data: 42855675
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=cmedm&AN=42855675
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1186/s12877-026-08079-1
    Languages:
      – Code: eng
        Text: English
    Subjects:
      – SubjectFull: China epidemiology
        Type: general
      – SubjectFull: Humans
        Type: general
      – SubjectFull: Female
        Type: general
      – SubjectFull: Aged
        Type: general
      – SubjectFull: Male
        Type: general
      – SubjectFull: Longitudinal Studies
        Type: general
      – SubjectFull: Risk Factors
        Type: general
      – SubjectFull: Middle Aged
        Type: general
      – SubjectFull: Random Forest
        Type: general
      – SubjectFull: Predictive Learning Models
        Type: general
      – SubjectFull: Aged, 80 and over
        Type: general
      – SubjectFull: East Asian People
        Type: general
      – SubjectFull: Cognitive Dysfunction diagnosis
        Type: general
      – SubjectFull: Cognitive Dysfunction psychology
        Type: general
      – SubjectFull: Cognitive Dysfunction epidemiology
        Type: general
      – SubjectFull: Cognitive Dysfunction classification
        Type: general
      – SubjectFull: Cognition physiology
        Type: general
      – SubjectFull: Cognition classification
        Type: general
      – SubjectFull: Machine Learning
        Type: general
    Titles:
      – TitleFull: Identification of cognitive subtypes and risk networks in Chinese older adults: based on latent profile analysis and explainable machine learning.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Lai L
      – PersonEntity:
          Name:
            NameFull: Dong B
      – PersonEntity:
          Name:
            NameFull: Fu C
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 03
              M: 08
              Text: 2026 Aug 03
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-electronic
              Value: 1471-2318
          Numbering:
            – Type: volume
              Value: 26
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: BMC geriatrics
              Type: main
ResultId 1