PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
 
Neurobiol Aging. Author manuscript; available in PMC Aug 1, 2010.
Published in final edited form as:
PMCID: PMC2740902
NIHMSID: NIHMS128973
No cross-sectional influence of APOE ε4 dose on clinical tests in Alzheimer's disease
Rochelle E. Tractenberg, Ph.D., M.P.H.,1,2 Paul S. Aisen, M.D.,1 Gregory R. Hancock, PhD.,3 and G. William Rebeck, Ph.D.4
1 Department of Neurology, Georgetown University School of Medicine, Washington, D.C.
2Departments of Biostatistics, Bioinformatics & Biomathematics, and Psychiatry, Georgetown University School of Medicine, Washington, D.C.
3Center for Integrated Latent Variable Research and Department of Measurement, Statistics and Evaluation, University of Maryland, College Park
4 Department of Neuroscience, Georgetown University School of Medicine, Washington, D.C.
Address for correspondence and reprint requests: Rochelle E. Tractenberg 207 Building D 4000 Reservoir Rd., NW Washington, DC 20057 TEL: 202.687.2247 FAX: 202.784.3504 ; ret7/at/georgetown.edu
Abstract
This study sought to determine if there are detectible influences on the symptoms of Alzheimer's disease (AD) from the genetic risk factor for AD, the ε4 allele of apolipoprotein-E (APOE). Using data from two cohorts of AD patients, a cross-sectional latent variable model of AD was tested with three symptom factors explaining variability in the observed variables after taking a general neurological factor into account. No significant influence of ε4 was detected. APOE's effect in AD may occur prior to clinical symptoms, or may simply be more subtle than these instruments can detect.
Keywords: Factor analysis, statistical; Alzheimer's disease; genetic risk factor; cross-sectional analysis.
The association between the apolipoprotein-E (APOE) allele and its genotypes and the presence of AD has been established. Of greater controversy is whether there is an important effect of APOE on the common cognitive, functional and behavioral symptoms that are the clinical hallmarks of AD. Tests reflecting an association are rarely, if ever, used in clinical trials for AD. The present report sought to determine if the APOE-εD allele has detectible effects on AD clinical trial instrumentation through latent factors in a cross-sectional model of AD (Tractenberg et al. 2006). Control of the measurement error in the model can circumvent the lack of sensitivity that the clinical trial tests exhibit relative to more comprehensive neuropsychological tests.
The data were collected in two NIH-funded clinical trials (Prednisone study (PR), Aisen et al. 2000; Non-steroidal study (NS), Aisen et al. 2003; descriptive statistics/modeling, Tractenberg et al. 2006).
Materials
The cognitive outcome measures in both trials were the Alzheimer's Disease Assessment Scale – Cognitive, the Clinical Dementia Rating (CDR box scores reflecting cognition (memory, orientation, judgment & problem solving) and function (community affairs, home & hobbies, personal care) were created), and the Mini Mental State Exam. In one study, function was assessed with the Dependency Scale. In the other study, function was assessed with an Activities of Daily Living Inventory, and both cohorts also had CDR functional box scores. Both cohorts had one indicator of general behavioral disturbance and three indicators reflecting psychosis, agitation, and depression. (see Tractenberg et al. 2006 for details).
Procedure
Structural equation modeling was carried out using EQS 6.1 (Multivariate Software, Inc., 2005; Bentler and Wu, 1995). Four models were fit: a model with no effect of ε4; a full effects model where ε4 dose is hypothesized to influence all latent variables; one partial effects model where ε4 dose is hypothesized to influence only the latent general neurological status variable and one model with a causal influence of ε4 dose on each of the three symptom-level latent variables. Four robust (Bentler & Wu, 1999) and one standard fit statistics were assessed for each model.
Results
The two cohorts have been described elsewhere (Aisen et al. 2000, Aisen et al. 2003) as has the modeling (Tractenberg et al. 2006). The model with no influence of ε4 fit the data in both cohorts well, and better than any of the other models. Thus, no evidence was found in either cohort supporting effects of APOE in the model of AD symptoms.
Discussion
The present study tested for influence of ε4 dose (0, 1 or 2 alleles) on latent variables; this represents indirect but error free effects (of the allele) on observed scores. Our results suggest that any influence of ε4 dose on general neurological or symptom-specific factors driving performance on instruments that are common in studies of AD (and particularly in clinical trials/studies) is not significant – or detectable - in a cross-sectional model. Appropriate longitudinal versions of the models described here might be able to detect an effect.
01
1
No cross-sectional influence of APOE ε4 dose on clinical tests in Alzheimer's disease: Supplementary Material
Rochelle E. Tractenberg, Ph.D., M.P.H.1,2, Paul S. Aisen, M.D. 1 Gregory R. Hancock, PhD.3, and G. William Rebeck, Ph.D. 4
Address for correspondence and reprint requests:
Rochelle E. Tractenberg 207 Building D 4000 Reservoir Rd., NW Washington, DC 20057 TEL: 202.687.2247 FAX: 202.784.3504 ret7/at/georgetown.edu
1 Department of Neurology; 2Departments of Biostatistics, Bioinformatics & Biomathematics, and Psychiatry, Georgetown University School of Medicine, Washington, D.C. 3Center for Integrated Latent Variable Research and Department of Measurement, Statistics and Evaluation, University of Maryland, College Park; 4 Department of Neuroscience, Georgetown University School of Medicine, Washington, D.C.
1. Introduction
A series of studies recently published describe the association of having at least one ε4 allele (relative to having none) with poorer performance on several cognitive measures argued to be important preclinical indicators of incipient AD. Although no associations were reported between presence of the ε4 allele and typical clinical trial instrumentation, there were differences between persons with at least one ε4 allele and those with no ε4 allele on some highly specific neuropsychological or neuroimaging measures tested (Bondi et al., 1995; Lange et al., 2002; Bondi et al., 2005; Houston et al., 2005; Jacobson, Delis, Lansing, et al., 2005; Jacobson, Delis, Bondi, & Salmon, 2005; Wetter et al., 2005; Han et al., 2006). Importantly, these tests are rarely, if ever, used in clinical trials for AD. These analyses used a latent variable model established earlier (Tractenberg et al., 2006) to evaluate the role of ε4 dose on test scores.
2. Methods
The emphasis in these analyses was on the role of the genetic risk factor APOE4 dose, the instruments that were modeled are only briefly described.
2.1 Data sources
2.1.1 Prednisone study (PR).
Aisen et al. (2000) conducted a multi-center randomized placebo-controlled clinical trial to determine the effect of steroid anti-inflammatory drug prednisone (PR) to slow the cognitive decline in 138 persons who met National Institute of Neurological and Communicative Diseases and Stroke/Alzheimer Disease and Related Disorders Association (NINCDS-ADRDA; McKhann et al., 1984) criteria for probable AD. Use of cholinesterase inhibitor was not allowed.
2.1.2 Non-steroidal study (NS).
Aisen et al. (2003) conducted a multi-center randomized placebo-controlled clinical trial to determine the effect of the nonsteroidal anti-inflammatory drugs (NSAIDs) refecoxib, a selective cyclooxygenase −2 (COX-2) inhibitor and naproxen (a non-selective NSAID) on cognitive decline in 351 persons with NINCDS-ADRDA diagnoses of probable AD. At baseline, 68% of this sample was on a stable dose of a cholinesterase inhibitor.
2.2 Materials.
Cognitive impairment was evaluated with the same three cognitive outcomes at baseline in the two cohorts. The primary outcome measure in both trials was the Alzheimer's Disease Assessment Scale – Cognitive (ADAS-Cog, Mohs, Rosen, & Davis, 1983; Rosen, Mohs, & Davis, 1984). The ADAS-Cog, designed specifically for use with persons with AD, measures memory, attention, reasoning, language, orientation and praxis. Scores range from 0−70 with higher scores reflecting higher impairment. The Clinical Dementia Rating (CDR, Hughes, Berg, Danziger, Coben, & Martin, 1982) was a secondary outcome measure in both studies. The CDR is based on a semi-structured interview of both the patient and a knowledgeable informant, assessing/grading overall dementia severity. The box scores representing these two domains at baseline were used as individual indicators. Higher box scores reflect higher impairment in terms of cognition (memory, orientation, judgment & problem solving) and function (community affairs, home & hobbies, personal care), respectively.
The Mini Mental State Exam (MMSE, Folstein et al., 1976) was part of the screening process for both studies. Full-scale MMSE scores range from 0−30 (worst to best).
Functional impairment was evaluated with slightly different instruments in the two cohorts, although the fit of the latent variable model was not sensitive to this. The CDR box scores representing function is common to both studies. In the Prednisone (PR) study, function was assessed with the Dependency Scale (DS, Stern et al., 1994) which assesses various physical activities of daily living including eating, dressing and toileting. In the Nonsteroidal (NS) study, function was assessed with an Activities of Daily Living Inventory (ADL, Galasko et al., 1997) measuring the level of assistance required to carry out 23 ADLs and instrumental ADLs (IADLs). This instrument includes all the activities that the DS includes, plus additional activities. Functional scores were converted, if necessary, so that higher functional ‘scores’ reflected worse performance; both cohorts had four functional indicators.
The behavioral instruments were different for both studies. Both cohorts had one indicator of general behavioral disturbance and three indicators reflecting key behavioral symptom types: psychosis, agitation, and depression. In the PR study, two measures of behavioral symptoms were employed. General behavioral symptoms were assessed with the Brief Psychiatric Rating Scale (BPRS, Overall & Gorham, 1962); the total BPRS score was used as an indicator of ‘general behavioral disturbance’, while indicators of agitation and psychosis were derived as simple sums of BPRS frequency ratings for items reflecting agitation or psychosis. In the PR study, depression was assessed with the Hamilton Rating Scale for Depression (HRSD, Hamilton, 1961).
In the NS study, behavioral symptoms were assessed using the Neuropsychiatric Inventory (NPI, Cummings et al., 1994). To generate specific depression scores for NS, the NPI sub-items from across domains that most closely match the Hamilton items were identified. These sub-item ratings were summed to create a depression score for the NS cohort; specific indicators for agitation and psychosis were created based on relevant sub-items (see Tractenberg et al., 2006 for details of the scoring).
2.3 Procedure:
The models (Figures 1A-1D) were run separately for the cohorts’ baseline responses and fit indices (described below) were recorded for each model run. Model fit statistics were compiled in order to determine which configuration resulted in the best fit. To ensure interpretability, the “best” fit was defined as that model fitting best in both of the two independent cohorts; otherwise, it would suggest that the influence of ε4 dose was not consistently exerted (or not detectible) in the cross-sectional residualized model of symptomatology and underlying neurological disease that was studied here. The “best fitting” model would be the one where all fit indices were consistent in indicating the best (most desirable) value.
In an earlier study (Tractenberg et al., 2006) we found that a second-order model, hypothesizing a direct causal effect of a general neurological factor (GN) on the observed variables with no direct effect of GN on the three latent symptom factors (cognitive, functional and behavioral; COG, FUN, BEH respectively), fit the data from the same two independent cohorts described in this paper. In this model, the three latent symptom factors (COG, FUN, BEH) are correlated, and exert direct effects on the observed variables above and beyond, i.e., residualized after quantifying, the effects of GN. The interpretation of this model is that the variability in observed variables (scores) is due to the combination of the direct effects of the general factor plus those of the symptom factors after controlling for the general factor. Tractenberg et al. 2006 reported that this residualized model fit data from these two independent cohorts well, and the fit was the best of four models considered (described in Tractenberg et al., 2006). Therefore, the residualized model (Figure 1A) was taken as the “null” model, against which the models in the present analysis were to be compared. In it, APOE ε4 dose has no influence on the observed scores.
As noted, the latent variable model approach provides an ‘error free’ representation of the constructs that instruments used in clinical trials for AD are intended to represent. That is, the MMSE score is intended to represent ‘cognitive functioning’ but it does so with some degree of measurement error; cognitive function is an unobservable (latent) factor. We are interested in the effects of ε4 dose on cognitive function, and not on MMSE score per se; thus, the models shown in Figures 1B-1D have arrows (representing hypothesized causal influence) from ε4 dose, an observed variable, to latent factors (in circles).
The fit of the null model to the data in these independent cohorts was compared to the fit of three other models (depicted in Figures 1B-1D): a full effects model and two partial effects models. Figure 1B shows the full effects where ε4 dose is hypothesized to exert influence on the latent constructs of general neurological status and each of the symptom-level latent factors. In this model, ε4 dose contributes to observed scores through both pathology and symptomatology. Figure 1C shows one partial effects model where ε4 dose is hypothesized to exert influence only on the latent construct of general neurological status (“general neurological effects only”); ε4 dose is hypothesized to have no influence on the symptom-level latent factors. Conversely, the model shown in Figure 1D hypothesizes a causal influence of ε4 dose on each of the three symptom-level latent factors, but not on the general neurological status factor (“symptom effects only”).
2.3.1 Model Fit.
The models were fit to the data from each cohort's baseline visits separately. Five different aspects of fit were assessed for each run in each cohort, reflecting general data-model fit (Satorra-Bentler model chi square, X2 - lower is better), assessment of the fit of the model to data in hypothetical replications (Akaike's Information Criterion, AIC –lower is better), incremental model fit relative to an independence model (comparative fit index, CFI – between .95 and 1.0 is desirable), error in approximation of the data by the model (root mean squared error of approximation, RMSEA- smaller and <0.08 is ideal), and the mean absolute value of the covariance residuals (standardized root mean square residual, SRMR - smaller and <0.09 is best) (criteria for fit indices are based on standard, and not robust, versions; see Hu & Bentler, 1999). These indices describe different aspects of the fit of the model; we would consider a model that is superior in all indices to be the “best fitting”. Robust fit statistics were computed as well as the SRMR, which has no robust counterpart but which summarizes the fit in a way the other (robust) indices do not.
3. Results
Table 1 presents descriptive statistics of the cohorts, including genotype data.
Table 2 presents the fit statistics from the four models (null model; full effects; general neurological effects only; symptom effects only) across the two cohorts.
The fit of the original residualized model (“null”), which hypothesizes no effect of APOE on the observed test scores, is summarized in the first pair of rows in Table 2. This model fits the NS data well, and better than any of the four evaluated according to all five indices. The null model fits the PR data well, and best according to all criteria, although the SRMR index was ideal for every model in the PR cohort. Thus, no support was obtained for hypotheses of effects of APOE on the observed scores.
Acknowledgements
This manuscript was based on an internship report completed to partially fulfill the requirements for a Doctoral-level Graduate Certificate in Gerontology at the University of Maryland, College Park Center on Aging (RET). Supported by a Research Starters Grant in Health Outcomes from the Pharmaceutical Researchers and Manufacturers Association of America (PhRMA) Foundation to RET. The data were collected by the Alzheimer's Disease Cooperative Study under NIA Grant U01 AG10483.
Disclosure Statement:
No actual or potential conflicts of interest for any authors.
References
Aisen PS, Davis KL, Berg JD, Schafer K, Campbell K, Thomas RG, Weiner MF, Farlow MR, Sano M, Grundman M, Thal LJ. (2000). A randomized controlled trial of prednisone in Alzheimer's disease. Alzheimer's Disease Cooperative Study. Neurology. 54(3): 588−593.
Aisen PS, Schafer KA, Grundman M, Pfeiffer E, Sano M, Davis KL, Farlow MR, Jin S, Thomas RG, Thal LJ; Alzheimer's Disease Cooperative Study. (2003). Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial. JAMA 289(21): 2819−26.
Arnold SE, Hyman BT, Flory J, Damasio AR, Van Hoesen GW. (1991). The topographical and neuroanatomical distribution of neurofibrillary tangles and neuritic plaques in the cerebral cortex of patients with Alzheimer's disease. Cereb Cortex 1(1):103−16
Bentler P M. and Wu E. (1995). EQS structural equations program manual. Encino, CA: Multivariate Software Inc.
Bondi MW (2002). Genetic and brain imaging contributions to neuropsychological functioning in preclinical dementia. J Int Neuropsychol Soc. 8(7):915−7.
Bondi MW, Houston WS, Eyler LT, Brown GG. (2005). fMRI evidence of compensatory mechanisms in older adults at genetic risk for Alzheimer disease. Neurology. 64(3):501−8.
Bondi MW, Salmon DP, Galasko D, Thomas RG, Thal LJ. (1999). Neuropsychological function and apolipoprotein E genotype in the preclinical detection of Alzheimer's disease. Psychol Aging. 14(2):295−303.
Bondi MW, Salmon DP, Monsch AU, Galasko D, Butters N, Klauber MR, Thal LJ, Saitoh T. (1995). Episodic memory changes are associated with the APOE-epsilon 4 allele in nondemented older adults. Neurology. 45(12):2203−6.
Cummings JL. (2003). Alzheimer's disease: From molecular biology to neuropsychiatry. Seminars in Clinical Neuropsychiatry 8(1): 31−36.
Cummings JL, Mega M, Gray K, Rosenberg-Thompson S, Carusi DA and Gornbein J. (1994). The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in Dementia. Neurology 44:2308−2314.
Folstein M, Folstein S, McHugh P. (1975). “Mini-Mental State”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189−98.
Galasko D, Bennett D, Sano M, Ernesto C, Thomas R, Grundman M, Ferris S. (1997). An inventory to assess activities of daily living for clinical trials in Alzheimer's disease. The Alzheimer's Disease Cooperative Study. Alzheimer Dis Assoc Disord 11(Suppl 2): S33−39.
Ghebremedhin E, Schultz C, Thal DR, Rüb U, Ohm TG, Braak E, Braak H. (2001). Gender and age modify the association between APOE and AD-related neuropathology. Neurology 56(12):1696−701.
Gomez-Isla T, West HL, Rebeck GW, Harr SD, Growdon JH, Locascio JJ, Perls TT, Lipsitz LA, Hyman BT. (1996). Clinical and pathological correlates of apolipoprotein E epsilon 4 in Alzheimer's disease. Annals of Neurology 39(1):62−70.
Hamilton M. (1960). A rating scale for depression. J Neurol Neurosurg Psychiatry 23:56−62.
Han SD, Houston WS, Jak AJ, Eyler LT, Nagel BJ, Fleisher AS, Brown GG, Corey-Bloom J, Salmon DP, Thal LJ, Bondi MW. (2006). Verbal paired-associate learning by APOE genotype in non-demented older adults: fMRI evidence of a right hemispheric compensatory response. Neurobiol Aging [Epub ahead of print]
Houston WS, Delis DC, Lansing A, Jacobson MW, Cobell KR, Salmon DP, Bondi MW. (2005). Executive function asymmetry in older adults genetically at-risk for Alzheimer's disease: verbal versus design fluency. J Int Neuropsychol Soc. 11(7):863−70.
Hughes CP, Berg L, Danziger WL, Coben LA, Martin RL. (1982). A new clinical scale for the staging of dementia. Br J Psychiatry 140: 566−572.
Hu L. & Bentler PM. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling 6(1): 1−55.
Ingelsson M, Fukumoto H, Newell KL, Growdon JH, Hedley-Whyte ET, Frosch MP, Albert MS, Hyman BT, Irizarry MC. (2004). Early Abeta accumulation and progressive synaptic loss, gliosis, and tangle formation in AD brain. Neurology 62(6):925−31.
Jacobson MW, Delis DC, Lansing A, Houston W, Olsen R, Wetter S, Bondi MW, Salmon DP. (2005). Asymmetries in global-local processing ability in elderly people with the apolipoprotein e-epsilon4 allele. Neuropsychology. 19(6):822−9.
Jacobson MW, Delis DC, Bondi MW, Salmon DP. (2005). Asymmetry in auditory and spatial attention span in normal elderly genetically at risk for Alzheimer's disease. J Clin Exp Neuropsychol. 27(2):240−53.
Kline RB. (2005). Principles and Practice of Structural Equation Modeling, 2E. New York NY: Guilford Press.
Lane RM and Farlow MR. (2005). Lipid homeostasis and apolipoprotein E in the development and progression of Alzheimer's disease. Journal of Lipid Research 46: 949−68.
Lange KL, Bondi MW, Salmon DP, Galasko D, Delis DC, Thomas RG, Thal LJ. (2002). Decline in verbal memory during preclinical Alzheimer's disease: examination of the effect of APOE genotype. J Int Neuropsychol Soc 8(7):943−55.
Loehlin JC. (2004). Latent Variable Models: An introduction to factor, path, and structural equation analysis, 4E. Mahwah, NJ: Lawrence Earlbaum Associates.
Mahley RW, Weisgraber KH and Huang Y. (2006). Alipoprotein E4: A causative factor and therapeutic target in neuropathology, including Alzheimer's disease. Proceedings of the National Academy of Sciences (PNAS) Early Edition; downloaded from www.pnas.org/cgi/doi/10.1073/pnas.0600549103 on 10 March 2006.
McKhann G, Drachman D, Folstein M, et al. Clinical diagnosis of Alzheimer's disease: report of the NINCDS-ADRDA Work Group under the auspices of the Department of Health and Human Services Task Force on Alzheimer's disease. Neurology 1984;34:939−944.
Mohs RC, Rosen WG and Davis KL. (1983). The Alzheimer's Disease Assessment Scale (ADAS): An instrument for assessing treatment efficacy. Psychopharmacology Bulletin; 18: 448−450.
Marin DB, Green CR, Schmeidler J, et al. (1997). Noncognitive disturbances in Alzheimer's Disease: Frequency, longitudinal course, and relationship to cognitive symptoms. J Am Geriatr Soc 45:1331−8.
Olichney JM, Hansen LA, Galasko D, Saitoh T, Hofstetter CR, Katzman R, Thal LJ. (1996). The apolipoprotein E epsilon 4 allele is associated with increased neuritic plaques and cerebral amyloid angiopathy in Alzheimer's disease and Lewy body variant. Neurology 47(1):190−6.
Overall JE and Gorham DR. (1962). Brief Psychiatric Rating Scale. Psychol Rep 10:799−812.
Pirttilä T, Soininen H, Mehta PD, Heinonen O, Lehtimäki T, Bogdanovic N, Paljärvi L, Kim KS, Kosunen O, Winblad B, Riekkinen P, Wisniewski HM. (1997). Apolipoprotein E genotype and amyloid load in Alzheimer disease and control brains. Neurobiology of Aging 18(1):121−7.
Rebeck GW, Reiter JS, Strickland DK, Hyman BT. (1993). Apolipoprotein E in sporadic Alzheimer's disease: allelic variation and receptor interactions. Neuron.11(4):575−80.
Rosen WG, Mohs RC and Davis KL. (1984). A new rating scale for Alzheimer's disease. Am J Psychiatry 141: 1356−1364.
Satorra A & Bentler PM. (2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika 66(4): 507−514.
Schmechel DE, Saunders AM, Strittmatter WJ, Crain BJ, Hulette CM, Joo SH, Pericak-Vance MA, Goldgaber D, Roses AD. (1993). Increased amyloid beta-peptide deposition in cerebral cortex as a consequence of apolipoprotein E genotype in late-onset Alzheimer disease. Proc Natl Acad Sci U S A 90(20):9649−53.
Selkoe DJ and Lansbury Jr, PL. (1999). Biochemistry of Alzheimer's disease. In GJ Siegel et al. (Eds.) Basic Neurochemistry: Molecular, Cellular and Medical Aspects, 6th Edition. Philadelphia: Lippincott Williams and Wilkins. Pp. 949−968.
Stern Y, Albert SM, Sano M, et al. (1994). Assessing patient dependence in Alzheimer's disease. J Gerontol 49(5):M216−22.
Thal L. J. (1997). Development of the Alzheimer's Disease Cooperative Study. Int J Geriatr Psychopharmacol 1: 6−9.
Tiraboshi P, Hansen LA, Masliah E, Alford BA, Thal LJ and Corey-Bloom J. (2004). Impact of APOE genotype on neuropathologic and neurochemical markers of Alzheimer's disease. Neurology 62: 1977−1983.
Tractenberg R.E., Aisen P. S., Weiner M. F., Cummings J. L. and Hancock, G. R. (2006). Independent contributions of neural and ‘higher order’ deficits to symptoms in Alzheimer's disease: a latent variable approach. Alzheimer's and Dementia. 2(4):303−313
Tractenberg RE, Weiner MF, Cummings JL, and Thal LJ. (2005). Independence of changes in behavior from changes in cognition and function in persons with Alzheimer's disease: A factor-analytic approach. J Neuropsychiatry and Clin Neurosci17(1):51−60.
Tractenberg RE, Weiner MF, Patterson MB, Schneider L, and Thal LJ. (2002). Emergent psychopathology in Alzheimer's disease patients over 12 months associated with functional, not cognitive, changes. J Geriatric Psychiatry Neurol, 15(2): 110−117.
Wetter SR, Delis DC, Houston WS, Jacobson MW, Lansing A, Cobell K, Salmon DP, Bondi MW. (2005). Deficits in inhibition and flexibility are associated with the APOE-E4 allele in nondemented older adults. J Clin Exp Neuropsychol. 27(8):943−52.
Table 1. Descriptive statistics for scores reflecting cognition, function, and behavior at baseline in two clinical trial samples of persons with Alzheimer's disease.
Table 2. Model-data fit for residualized model with ε4 dose (0,1,2) having no effects(null), full effects, effects on general neurological factor only, or on symptoms factors only. Results are shown by study sample. Bolded values per column represent the best supported model according to that criterion; where bolded values appear for the NS cohort, these values are also underlined.
FIGURES LEGEND: Figures 1A-1D are all based on the same information, described here: All scores were from the pre-treatment visit (baseline, BL); boxes are observed variables and circles are latent factors. e4 = this is the genetic test result reflecting the number of ε4 alleles (0, 1, 2). ADAS = Alzheimer's Disease Assessment Scale; MMSE = Mini Mental State Examination; MEM, ORIEN, JUDGE= CDR box scores for cognitive function (memory; orientation; judgment/problem solving); DEPEND = % dependence, derived from either the Dependency Scale or the ADL Inventory; COMM, HOME, PERS= CDR box scores for function (community, home/hobbies; personal care); BEH = General behavioral measure, derived as either the NPI or BPRS total score; DEPR = depression score, derived as either the Hamilton total score or the sum of frequency ratings on NPI items most closely matching the Hamilton; PSYC = psychosis score, derived from BPRS or NPI items; AGIT = agitation score, derived from BPRS or NPI items. In this model, all latent variables derive their scales from standardization of the factors’ variances (all =1). E1-E13 terms represent the error associated with the observed variables’ reflection of the underlying ‘true’ state. Where causal influences are hypothesized on latent variables (Figures 1B-1D), D1-D4 terms represent error associated with the causal influence of the e4 dose on the latent factors. Curved arrows between symptom-level factors represent correlations between the factors (Model 1C) or between the factor's disturbance terms (Models 1B and 1D).
Figure 1A. “Null” model: ε 4 dose has no effects on any variable: this is the residualized model, hypothesizing no effect of ε 4.
Figure 1B. Full effects model: ε 4 dose has effects on both the general neurological and symptom specific latent factors.
Figure 1C. ε 4 dose effects hypothesized on only the general neurological latent factor.
Figure 1D. ε 4 dose effects on just the symptom latent factors.
Acknowledgement
This manuscript was based on an internship report completed to partially fulfill the requirements for a Doctoral-level Graduate Certificate in Gerontology at the University of Maryland, College Park Center on Aging (RET). Supported by a Research Starters Grant in Health Outcomes from the Pharmaceutical Researchers and Manufacturers Association of America (PhRMA) Foundation to RET. The data were collected by the Alzheimer's Disease Cooperative Study under NIA Grant U01 AG10483.
Footnotes
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
  • Aisen PS, Davis KL, Berg JD, Schafer K, Campbell K, Thomas RG, Weiner MF, Farlow MR, Sano M, Grundman M, Thal LJ. A randomized controlled trial of prednisone in Alzheimer's disease. Alzheimer's Disease Cooperative Study. Neurology. 2000;54(3):588–593. [PubMed]
  • Aisen PS, Schafer KA, Grundman M, Pfeiffer E, Sano M, Davis KL, Farlow MR, Jin S, Thomas RG, Thal LJ., Alzheimer's Disease Cooperative Study Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial. JAMA. 2003;289(21):2819–26. [PubMed]
  • Bentler PM, Wu E. EQS structural equations program manual. Multivariate Software Inc.; Encino, CA: 1995.
  • Hu L, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling. 1999;6(1):1–55.
  • Tractenberg RE, Aisen PS, Weiner MF, Cummings JL, Hancock GR. Independent contributions of neural and ‘higher order’ deficits to symptoms in Alzheimer's disease: a latent variable approach. Alzheimer's and Dementia. 2006;2(4):303–313. [PMC free article] [PubMed]