Abstract
Objective: To assess the possible association between tacrine (Cognex, manufactured by Parke-Davis, Morris Plains, NJ) dose and likelihood of nursing home placement (NHP) or death in patients with AD. Design: A 30-week, randomized, double-blind, placebo-controlled, parallel-group multicenter clinical trial involving 663 patients, after which patients were treated openly and followed up a minimum of 2 years later. Patients: At baseline, outpatients were at least 50 years of age, met criteria for probable AD, with baseline Mini-Mental State Examination scores between 10 and 26 (inclusive), were otherwise healthy, and had a caregiver who could provide assessments and ensure medication compliance. Interventions: Randomized assignment to placebo or one of three ascending dosage regimens of tacrine over 30 weeks, followed by open label treatment for all patients who began the double-blind trial. Outcome Measures: NHP and death were examined using logistic regression. Results: Patients who remained on tacrine and were receiving doses > 80 mg/d or > 120 mg/d were less likely to have entered a nursing home than patients on lower doses (odds ratios > 2.7, 2.8, respectively.) There was a trend for lower mortality for patients receiving > 120 mg/d (p = 0.063). Conclusions: Treatment with tacrine at doses > 80 mg/d was associated with a reduced likelihood of NHP. These data demonstrate that tacrine's 30-week effects on cognitive function and clinicians' global ratings may generalize to effects on a major milestone of AD. Future studies should attempt to replicate these findings prospectively.
NEUROLOGY 1996;47: 166-177
The cholinesterase inhibitor tacrine has been demonstrated to have statistically and clinically significant effects on cognition, functional activities, and clinicians' global assessments in patients with AD over the course of 12- [1] and 30-week [2] randomized placebo-controlled trials. There is no information, however, on the effectiveness of tacrine over a longer period of time or whether tacrine treatment may delay the appearance of milestones of disease progression. The relationship between the symptomatic acute effects of tacrine observed over 30 weeks and long-term out-comes such as nursing home placement (NHP) remains speculative.
NHP is an important milestone in the progression of AD. It is a proxy for severe disability in AD, although an imperfect one. [3-10] Most AD patients in the United States, at least 75% in one study, eventually end up in nursing homes and remain there on average for 3 years. [11] At an estimated cost of more than $40,929 per year in 1991, NHP imposes a huge financial burden on families. [11] In addition, the emotional toll of NHP on both patient and family is enormous. [3,4] Delay of NHP would be a more tangible beneficial effect of an antidementia agent than a mere change in mental status scores.
To use NHP as an outcome measure in patients whose initial presentation is mild to moderate dementia requires multiyear observation. The 30-week tacrine study [2] was not long enough to observe sufficient numbers of instances of NHP to assess this outcome. Given that tacrine is now approved and marketed in the United States, it would be perhaps unethical [12] and logistically difficult to conduct a multiyear controlled trial of high-dose tacrine based on randomized treatment assignment to tacrine or placebo. Consequently, extension of knowledge about tacrine's effects to key events in the long-term course of AD must be addressed in an uncontrolled setting. [13]
We had the opportunity to follow up 90% of the 663 AD patients who were randomized in the 30-week tacrine trial [2] 2 years after the double-blind portion was completed. We specifically examined time to NHP and mortality in this large cohort of patients on the basis tacrine dose.
Methods.
Study design.
The 30-week study [2] was a randomized, double-blind, placebo-controlled, parallel-group clinical trial conducted at 33 centers in the United States. (See [2] for a complete description of methodology.) Patients randomized to tacrine treatment received an initial dose level of 40 mg/d (10 mg qid) and were force-titrated at 6-week intervals in increments of 40 mg to maximum dose levels of 80 mg/d for group II, 120 mg/d for group III, and 160 mg/d for group IV. Group I received placebo for the entire 30 weeks.
Patients who completed the 30-week study and those who finished early were eligible to receive long-term, open-label, tacrine treatment at the discretion of the study physician in consultation with the family. In the open-label phase, the initial dose for all patients was 40 mg/d (10 mg qid), which could be increased every 4 weeks in 40-mg increments to a maximum dose of 160 mg/d. Protocol-specified visits in the open-label phase were every 3 months once a patient reached a stable dose. Dosage records were available for all patients.
Approximately 2 years after the last patient completed the double-blind phase of this study, the protocol was amended to allow collection of additional information, and attempts were made, through the study centers, to contact the families of all 663 patients who took study medication.
Patient population.
Patients eligible for the 30-week study were men and women, at least 50 years of age, meeting National Institute of Neurological and Communicative Disorders and Stroke criteria for a diagnosis of probable AD. [14] Patients were otherwise healthy and did not have significant extrapyramidal, cerebrovascular, cardiac or hepatic disease, insulin-dependent diabetes mellitus, or chronic renal insufficiency. Disease severity was defined as mild to moderate based on Mini-Mental State Examination [15] (MMSE) scores between 10 and 26 inclusive at the time of study entry. Written informed consent was obtained from caregivers and patients (or their legal representatives). Medications with known CNS effects likely to interfere with assessments of efficacy and medications likely to mask cholinergic side effects were prohibited during double-blind treatment. Once patients completed or terminated from the double-blind phase of the study, there were no restrictions on concurrent medications, even if the patient continued on tacrine.
Baseline assessments included the MMSE, Global Deterioration Scale [16] (GDS), Instrumental Activities of Daily Living Scale [17] (IADL), and Physical Self-Maintenance Scale [17] (PSMS). The Alzheimer's Disease Assessment Scale [18] (ADAS) was also completed at baseline (and was the primary outcome measure in the 30-week trial). Reference will be made to one item from the ADAS, the rating of delusions, because delusions are a feature of AD that has been associated with rate of progression. [19]
Outcome measures.
NHP and mortality were selected as the primary endopints for this study. Each site was requested to provide data on dates of death or NHP of patients, if either occurred. Additional assessments were also obtained on patients who returned for clinic visits at the 2-year follow-up. Between 260 and 292 patients (depending on the measure) had follow-up MMSE, GDS, IADL, and PSMS assessments. These data will be reported separately.
Statistical methods.
Thirty-week double-blind phase.
Patients randomized to treatment groups II (40 to 80 mg/d), III (40 to 120 mg/d), and IV (40 to 160 mg/d) were compared with patients randomized to group I (placebo). The SAS procedure PROC LOGISTIC [20] was used for performing logistic regressions [21] with occurrence of NHP or death during the 30-week double-blind phase as the dichotomous dependent variable. Because of the small number of events, NHP and deaths were combined for this analysis. Logistic regression is used when the dependent variable can have two discrete values. It models the risk (or probability) of NHP or death as a linear function of the independent variables; the key ones here are the treatment group assignments, by including indicator variables for comparisons between patients randomized to the placebo treatment group and each of the three tacrine-treated groups. The logistic regression approach also adjusted for the covariates of baseline IADL, age, and gender.
Estimated treatment differences between placebo and the three treatment groups were computed using ratios of odds. For example, if NHP or death occurred with probability p1 in the patients exposed to placebo and p2 in patients in treatment group IV (40 to 160 mg/d), the odds of NHP or death are p1/(1 - p1) and p2/(1 - p2). The ratio of the odds of NHP or death in the placebo patients to the odds of NHP or death in the group IV patients is the odds ratio (OR), where an OR > 1 indicates that there was a greater risk of NHP or death in the placebo patients.
Long-term tacrine treatment follow-up.
Patients were grouped into one of four categories according to the last daily dose of tacrine they received: 0 to <or=to40 mg/d, >40 to <or=to80 mg/d, >80 to <or=to120 mg/d, and >120 mg/d with a maximum dose of 160 mg/d. Patients who were initially randomized to placebo and then never took tacrine in the open-label phase (26 in analyses of NHP and 25 in analyses of mortality) were considered to have a last dose of 0 mg/d. (When these patients were excluded from the analyses, similar results were obtained.) For analyses limited to patients who remained on tacrine, the lower two dosage groups were combined.
Last tacrine dose was used because it was representative of the dose used in long-term treatment. The last dose actually taken, rather than dose at the time of NHP or death, was chosen because the latter grouping method biases toward tacrine (almost all instances of NHP or death would be expected to be preceded by withdrawal of treatments). Analyses were also conducted using other means of subject grouping: average tacrine dose over the course of treatment or average dose over the 60 days before NHP, death, end of study, or discontinuation of tacrine. Similar results were obtained with these alternative methods of grouping patients by dose, although both other methods have methodologic flaws.
NHP and mortality were analyzed separately using logistic regression and comparing the low-dose group to the other groups. Logistic regression was feasible because all groups were followed-up for roughly the same period of time. Because of the need to titrate tacrine, life-table methods such as Cox proportional hazards modeling might have been overly sensitive to early NHP or mortality; logistic regression, by contrast, is insensitive to the time that an event occurred. The logistic regression analysis adjusted for IADL score obtained at the observation point immediately before tacrine treatment, age, gender, and the use of other investigational agents for dementia (an issue during open-label treatment only). IADL score was used as the covariate for severity because it was the disease-severity covariate that differed the most across the fourdose categories. The logistic regression included indicator variables to compare outcome of patients in the four-dose groups. Estimated treatment differences between dosage groups were evaluated using ORs. Patients lost to follow-up and patients in an assisted care facility before day 1 were excluded from the NHP analyses. All patients initially randomized were included in the analyses of mortality.
Life-table method summaries of time to NHP or death and Kaplan-Meier product-limit estimates [22] of the 25th percentiles of time to NHP were also computed, using PROC LIFETEST, [20] to illustrate the time dependence of NHP and death but not to determine statistical significance. The time to an endpoint was calculated using the first day of treatment with tacrine as day 1. For patients who were in the placebo group of the double-blind trial, this day was their first day of open-label treatment. For all other patients, it was the first day of the double-blind trial. For patients who were randomized to placebo in the double-blind trial and who then never received tacrine, day 1 was the day of randomization in the double-blind trial. Patients were right-censored if, at the 2-year follow-up point of ascertainment, they were alive and living at home.
Results.
Thirty-week double-blind phase: analyses for nursing home placement or mortality.
The results of the 30-week study's primary outcome measures have been reported previously. [2] Statistically significant differences in favor of group IV (40 to 160 mg/d tacrine) over placebo were observed for the ADAS (cognitive portion) and a clinician's global assessment. Seven percent (n = 12) of 184 patients randomized to the placebo group were in a nursing home or had died by week 30, compared with 7% (n = 4) of 61 patients randomized to titrate to tacrine 80 mg/d, 4% (n = 8) of 179 patients randomized to titrate to tacrine 120 mg/d, and 3% (n = 6) of 239 patients randomized to titrate to tacrine 160 mg/d. Statistically significant differences in NHP/death between patients randomized to placebo and tacrine 160 mg/d were observed (OR = 2.8, 95% CI = 1.0 to 7.8, p = 0.046).
Patient disposition at follow-up.
Data on NHP were available for 595 of the 663 patients (90%) randomized to treatment in the 30-week study Figure 1. Of the 68 patients excluded, 43 were lost to follow-up, 12 were in an assisted care facility before randomization in the 30-week study, 8 were in nursing homes but missing exact dates of nursing home entry, and 5 had unknown status regarding nursing home placement. There were 81 deaths among the 663 patients initially randomized.
Figure 1. Diagrammatic representation of patient disposition.
Analyses for NHP and mortality at follow-up: all patients by tacrine dose regardless of treatment duration.
NHP and death were analyzed separately. Gender differences and differences in IADL and PSMS were present across groups before the start of tacrine treatment. MMSE scores, age, and GDS scores were similar across groups Table 1. Although there were no statistically significant differences across groups in the distribution of ratings from the ADAS of delusions, the lower dose groups had slightly larger proportions of patients who were rated as having moderate or moderately severe delusions at baseline.
Table 1. Patient characteristics for analyses of NHP and mortality: All patients by last tacrine dose (mg/d)
The number of days of exposure to tacrine differed significantly between groups Table 2. Patients in the lower dose groups had far shorter mean and median exposure times than patients in the higher dose groups. The shorter exposure times to tacrine in the lower dose groups were due to censoring by NHP or death and early withdrawal from tacrine treatment.
Table 2. Follow-up data for analyses of NHP and mortality: All patients by last tacrine dose (mg/d)
Results from the logistic regression analysis indicated that the probability of NHP showed a statistical trend toward reduction in the patients whose last tacrine dose was >120 mg/d Table 3, but the probability of death was significantly reduced. The logistic regressions excluded an additional 11 patients who did not state whether or not they were taking other investigational medications for dementia and 2 patients with missing baseline IADL scores. Appendix 1 Table 7
Table 3. Logistic regression analysis of NHP and mortality: All patients by last tacrine dose (mg/d)
We also performed similar analyses Appendix 2 Table 8 substituting MMSE, PSMS, and GDS scores for IADL as covariates for severity, which gave comparable results in terms of magnitude of the association of high-dose tacrine and reduction in NHP and death. We performed an analysis in which we excluded patients with delusions and found a trend for delay of NHP and death by high-dose tacrine Appendix 2.
Table 7. (Appendix 1) Additional results (see text).
Table 8. (Appendix 2) Additional results (see text).
Because of the baseline group differences in several features, we also performed analyses in which we stratified by baseline MMSE scores (10 to 15, 16 to 20, and 21 to 26), IADL scores (21 to 30, 11 to 20, and 0 to 10), PSMS scores (>or=to10, 7 to 9, and 0 to 6), and GDS ratings (6, 5, and 2 to 4). For stratification by MMSE, IADL, and PSMS scores, the mildest and midrange severity strata in each instance showed higher ORs than the lowest strata, with trends toward association of high-dose tacrine treatment and NHP delay. The two milder GDS strata both showed similar trends. In addition, we found that men and women and different age strata had similar associations between high-dose tacrine and reduction in NHP.
A further analysis of patients who took tacrine for at least 180 days was performed. This analysis excluded patients who may have reached endpoints before receiving higher doses of tacrine. The analysis showed a trend for NHP in favor of high-dose tacrine that did not reach the level of statistical significance and a slightly stronger effect on mortality that achieved p = 0.033 Appendix 2.
Analyses for NHP and mortality at follow-up: patients on tacrine by dose.
We also performed an analysis that was restricted to patients who were taking tacrine at the time of an event (NHP or death) or at ascertainment or had ceased taking tacrine within 60 days of an event or ascertainment. The purpose was to examine dose effects in patients at different daily doses who did not discontinue tacrine until immediately before NHP or death. Baseline characteristics Table 4 and follow-up data Table 5 are shown for patients in analyses of NHP and mortality by final daily dose of tacrine. Because of the small number of patients receiving <or=to40 mg/d, the <or=to40-mg/d and >40- to <or=to80-mg/d groups were combined. The patients in this analysis included almost all (122/147) of the patients with last daily doses of >120 mg/d from the previous analysis, but only a fraction of patients in the lower dosage groups (96/142 of the >80- to <or=to120-mg/d patients and 102/306 of the combined groups receiving <or=to80 mg/d).
Table 4. Patient characteristics for analyses of NHP and mortality: Patients on tacrine or off <60 days by last dose (mg/d)
Table 5. Follow-up data for analyses of NHP and mortality: Patients on tacrine or off <60 days by last dose (mg/d)
Patients who remained on tacrine and who were receiving doses of >80 to <or=to120 mg/d or >120 mg/d had significantly reduced likelihood of NHP Table 6. Use of other covariates for severity and exclusion of patients with delusions gave similar results Appendix 2. Mortality was not significantly reduced, although there was a trend in the highest dose group (see Table 6). Because there were only seven deaths in the highest dose group, the estimates of ORs were highly sensitive to the inclusion or exclusion of one or two patients, so that the mortality data for this population subset should be interpreted with caution. Figure 2 illustrates the time to NHP and Figure 3 illustrates the time to death in the three final dose groups. The figures are useful in showing the temporal pattern of NHP and death but should be interpreted with the caveat that dose titration at the initiation of tacrine dosing precluded events of NHP or death occurring during titration from being attributed to the higher dose groups.
Table 6. Logistic regression analysis of NHP and mortality: Patients who were on tacrine or off <60 days at the time of censoring, by last dose (mg/d)
Figure 2. Probability of remaining at home according to final daily tacrine dose for patients who continued on tacrine up to or within 60 days of NHP or death, for whom exact date of NHP was available. Effective sample sizes at each interval are given below the graph.
Figure 3. Probability of survival according to final daily tacrine dose for patients who continued on tacrine up to or within 60 days of death or the end of the study. Effective sample sizes at each interval are given below the graph.
We performed an analysis of patients from this subset who took tacrine for at least 180 days and then remained on medication. This showed significant effects for both higher tacrine doses compared with the lower dose group for NHP but not mortality, confirming that the titration phase of treatment did not bias the results Appendix 2 Table 8.
Analyses of NHP and mortality in patients off tacrine.
Patients who had not taken tacrine for at least 60 days before an endpoint or ascertainment were grouped according to their last dose while receiving tacrine. No evidence of a dose effect on either endpoint was identified in patients who had been off tacrine for more than 60 days.
Discussion.
The results of this 2-year follow-up study showed that AD patients who titrated to doses of tacrine >80 mg/d and remained on the medication had lower risk of NHP than those who either discontinued the drug or continued on lower doses. The effect appeared to be clinically important in that high-dose tacrine-treated patients were 2.8 times less likely to be institutionalized than low-dose treated patients. These findings should not be dismissed peremptorily because of the methodologic limitations of the study. The results should be interpreted as an association rather than a causal relationship between high-dose tacrine treatment and outcome. Based on the analyses presented here and other evidence supporting a beneficial effect of tacrine on AD, the association is most plausibly the result of a salutory effect of tacrine on the symptoms of AD and not due to selection bias. To establish causality, however, would require a prospective randomized trial.
Analyses of the double-blind phase of the 30-week study allowed a straightforward comparison of patients randomized to tacrine treatment or placebo. The results in the first 30 weeks were consistent with the subsequent unblinded period of observation. Despite the small number of events, the results strongly suggested a beneficial causal effect at the higher tacrine doses on reduction of events of NHP and death. Because treatment assignment was randomized, causality can be inferred from this portion of the results.
There were a number of features of the analyses that mitigated the bias introduced by nonrandomized grouping [23,24] in the open-label follow-up. The present study used rigid entry criteria for probable AD, thus avoiding diagnostic errors and excessive patient variability. Through a variety of means, we equated baseline characteristics of groups. Use of covariates in the logistic regression analyses and use of stratified analyses both gave similar results favoring high-dose tacrine despite the baseline differences. Analyses with different subsets of patients divided according to MMSE, IADL, and PSMS scores showed a consistent pattern in which high-dose tacrine-treated patients had better outcomes than lower dose-treated patients. We also were able to minimize the effect of known predictors of AD progression. [19] Parkinsonism was highly likely to be absent in patients selected for the original 30-week study because the presence of the diagnosis of Parkinson's disease was exclusionary. There were small group differences in the presence of delusions, but analyses excluding patients with delusions still showed that high doses of tacrine were associated with a better outcome.
The use of logistic regression as our primary analytic tool was justified because all patient groups were followed for roughly the same amount of time. Logistic regression analysis was a more conservative approach than life-table methods, because it was insensitive to the time that an event of NHP or death occurred during the period of observation. Because of the need to titrate tacrine dosage at the initiation of treatment, events that occurred in the titration phases of treatment were likely to be attributed to lower dose groups with either analysis. By examining the subset of patients who remained on tacrine and took it for at least 180 days, we eliminated the biasing effect of events that occurred during the titration phase of treatment. The high-dose tacrine effect was still statistically significant in logistic regression when patients who reached endpoints during titration were excluded.
Dose effects support a causal relationship but do not prove one. [25] Data from the 30-week study [2] and other randomized controlled trials [1,26,27] supported a strong dose dependence of tacrine effects. The dose-effect on NHP in the present study was entirely consistent with the prior observations from short-term studies of cognitive function.
The analysis of all subjects suggested the association of highest doses of tacrine with improved outcome. The strength of the analysis was that it included all available patients and made no assumptions about relationships between reaching endpoints and anticipatory discontinuation of tacrine. This analysis had problems, however. The group of patients that received <80 mg/d of tacrine included many patients who remained on the medication for only a short period of time.
The analysis including patients who remained on tacrine up to or within 60 days of censoring controlled for different tacrine exposures. Again, there was a strong effect of higher doses on improved outcome. Individuals who remained on tacrine and whose last dose was <or=to80 mg/d served as a comparison group for this analysis. Patients in the low-dose group in this analysis continued medication under the same circumstances as the higher-dose groups and discontinued it before the end of the observation period for the same reasons (i.e., NHP or death). The low- and high-dose groups were probably comparable in terms of treatment-specific caregiver demands and capabilities. The major limitation of this analysis is that a large number of patients in the low-dose group were excluded from this analysis.
For patients who had discontinued tacrine more than 60 days before an event, there was no evidence of a tacrine effect on NHP or survival, suggesting that tacrine may be exerting a symptomatic treatment effect rather than an enduring effect on disease progression.
The present data are the strongest to data addressing the long-term effects of tacrine. Other studies [28-32] were also observational and, with one exception, [32] involved no control groups. Several of the studies [28-31] concluded that tacrine had beneficial long-term effects. The one negative study, [32] a placebo-controlled 9-month study of 32 patients, used a low dose (80 mg/d or less) of the active drug. The 30-week study [2] and the present results showed conclusively that doses of tacrine in excess of 120 mg/d are much more likely to be effective than 80 mg/d or less.
The major interpretative difficulties with the present study are that the treatment assignment was not random nor was it blinded after 30 weeks. Thus, it cannot be proved that there was a causal relationship between higher dose tacrine treatment and more favorable outcomes. It is possible that these results occurred as a result of segregating patients on the basis of tolerance to tacrine, which in turn was positively correlated with disease progression. A similar interpretive problem would occur in a randomized trial of comparable duration, if differential attrition across treatment groups occurred. Because high-dose tacrine cannot be tolerated by a large number of patients (over half in this study), there is a basis for this claim. To accept the hypothesis that tacrine tolerance does predict a patient's vigor implies that gastrointestinal and hepatic function are important predictors of NHP. Although tacrine tolerance seems superficially to be an explanation for the findings, the specific processes by which tacrine induces transaminase elevations or gastrointestinal toxicity do not seem to have anything to do with risk of NHP. The lower death rate in the high-dose tacrine-treated patients could be used to support the contention that hardier patients were more likely to remain on higher doses, if one assumed that tacrine itself could not logically exert an effect on survival. Tacrine currently has not been associated with interrupting an underlying mechanism of AD (but see [33]). Yet, the difference in death rate was small. Finally, the contention that patients in the higher dose groups indicated that they had more resourceful caregivers seems unlikely to account for the results. Remaining on any dose of tacrine is probably what took resourcefulness, and we effectively controlled for that in the analysis of patients who remained on the drug. Based on the multiple analyses performed in this study, the most credible explanation for the delay in NHP is the symptomatic benefit of high-dose tacrine treatment. Recent observations of beneficial effects of tacrine on several behavioral symptoms of AD provide a possible mechanism for tacrine's effect on delay of NHP. [34]
The fact that the investigators and patients were not blinded to treatment during the open-label follow-up phase might also tend to increase spuriously the size of the high-dose tacrine effect. [35] On the other hand, death and NHP are not subjective outcomes. In addition, an intention to carry out an outcome study was not known to study physicians virtually until they were contacted to collect the data. It seems implausible that there could have been any overt or covert bias toward NHP in low-dose treated patients.
In conclusion, the present study has shown an association between high-dose tacrine treatment and reduced NHP and mortality. Given the cognitive effects of tacrine in several randomized placebo-controlled trials, [1,26,27] a plausible and testable explanation of the association between high-dose tacrine and outcome is tacrine's direct beneficial effects in AD. Future prospective randomized placebo-controlled or casecontrol studies should attempt to replicate the findings.
Appendix 1 *Table 7*.
The following investigators comprised the Tacrine Study Group and contributed data to this study: Jeffrey T. Apter, MD, Princeton Biomedical Research, Princeton; Michael Barnett, MD, Falmouth, MA; Barry Baumel, MD, Larry S. Eisner, MD, Neuro/Medical Research Associates, Miami Beach, FL; David Bennett, MD, Concetta Forchetti, MD, Rush Alzheimer's Disease Center, Chicago, IL; John P. Blass, MD, PhD, Burke Medical Research Institute, Cornell University Medical College, White Plains, NY; Richard L. Borison, MD, PhD, Psychiatry Service, Downtown VA Medical Center, Augusta, GA; Gastone G. Celesia, MD, Loyola Medical Center, Maywood, IL; James Dexter, MD, Department of Neurology, University of Missouri Health Sciences Center, Columbia; Rachelle Doody, MD, PhD, Department of Neurology, Baylor College of Medicine, Houston, TX; Eugene A. DuBoff, MD, Center for Behavioral Medicine, Wheat Ridge, CO; Nancy L. Earl, MD, Memory Disorders Clinic, Durham, NC; Martin Farlow, MD, Hugh C. Hendrie, MD, Indiana University Center for Alzheimer's Disease and Related Disorders, Indianapolis; Mildred Farmer, MD, Clinical Studies Florida (G.E.R.I.), St. Petersburg, FL; James Ferguson, MD, Pharmacology Research Corporation, Salt Lake City, UT; Norman L. Foster, MD, Department of Neurology, University of Michigan Medical Center, Ann Arbor; Maria Greenwald, MD, Southwest Institute of Clinical Research, Rancho Mirage, CA; Adrian Groenendyk, MD, Encino Medical Plaza, Albuquerque, NM; Claire Jurkowski, MD, Geriatric Assessment Program, Memorial Hospital of Burlington County, Mt. Holly, NJ; Ira Katz, MD, PhD, Department of Psychiatry, Medical College of Pennsylvania, Philadelphia; Charles Kellgaret Prodner, MD, Medical University of South Carolina, Charleston; David S. Knopman, MD, Department of Neurology, University of Minnesota Medical School, Minneapolis; Richard A. Margolin, MD, Vanderbilt University Medical Center, Nashville, TN; John C. Morris, MD, Eugene H. Rubin, MD, Washington University School of Medicine, Alzheimer's Disease Research Center, St. Louis, MO; Patricio F. Reyes, MD, Department of Neurology, Jefferson Medical College, Philadelphia, PA; Marilyn M. Rymer, MD, Center for Clinical Neurologic Studies, L.C., Kansas City, MO; Carl H. Sadowsky, MD, West Palm Beach Neurological Group, West Palm Beach, FL; Lon Schneider, MD, Sonia Pawluczyk, MD, University of Southern California, Los Angeles; Ward T. Smith, MD, Pacific Northwest Research Center, Portland, OR; Paul R. Solomon, PhD, William W. Pendlebury, MD, The Memory Disorders Clinic, Southwestern Vermont Medical Center, Bennington, VT; John Taylor, MD, Department of Neurology, Richmond, VA; Stephen G. Thein, Jr, PhD, Pacific Research Network, San Diego, CA; Paul Tuttle, MD, Carolina Neurological Clinic, Charlotte, NC; Robert J. Tyndall, MD, Kelley Clinical Trials Center, Springfield, MO.
Acknowledgments
We thank all of the participants and investigators who contributed to the completion of this study. We also acknowledge Margaret Knapp, PharmD, whose energy and commonsense guided the 30-week study.
- Copyright 1996 by Advanstar Communications Inc.
REFERENCES
Letters: Rapid online correspondence
REQUIREMENTS
If you are uploading a letter concerning an article:
You must have updated your disclosures within six months: http://submit.neurology.org
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
You May Also be Interested in
Use of Whole-Genome Sequencing for Mitochondrial Disease Diagnosis
Dr. Robert Pitceathly and Dr. William Macken
► Watch
Alert Me
Recommended articles
Efficacy and safety of eptastigmine for the treatment of patients with Alzheimer’s disease
Effects of estrogen replacement therapy on response to tacrine in patients with Alzheimer's disease
Metrifonate treatment of the cognitive deficits of Alzheimer's disease
Metrifonate benefits cognitive, behavioral, and global function in patients with Alzheimer's disease