The Integrative Personalized Functional Medicine Approach to Reverse Cognitive Decline: Academic Experience of the First 51 Patients Case Series

Abstract

Background: Alzheimer’s disease (AD) and related dementias remain major public health challenges with limited treatment options. Personalized, multimodal integrative approaches have emerged as potential strategies for patients with mild cognitive impairment (MCI) and early-stage AD. Purpose: To describe outcomes from the first 51 participants enrolled in an academic integrative functional medicine program for cognitive decline. Research Design: Retrospective case series, chart analysis. Study Sample: Participants with MCI or AD enrolled at the GW Center for Integrative Medicine (n=51, 2017-2025); 22 remained in the program for ≥2.5 years for long-term cognitive analyses. Data Collection and/or Analysis: Participants underwent serial cognitive assessments [Montreal Cognitive Assessment (MoCA) and/or Boston Cognitive Assessment (BOCA), subset also completing CNS-Vitals and SLAMs at variable intervals]. MoCA was the prespecified primary outcome. Interventions included dietary modification, time-restricted eating, exercise, cognitive training, sleep optimization, individualized supplementation, hormonal management, and 40 Hz gamma-frequency sensory stimulation. Exploratory measures included volumetric brain MRI and Alzheimer's blood biomarkers (Aβ42/Aβ40, p-tau217, NfL) in a subset. Outcomes were summarized descriptively without formal statistical analyses. Results: Among participants with ≥2.5 years of follow-up, mean MoCA increased from 21.7 to 22.5, and 73% (16/22) demonstrated improvement (≥2-point increase) or stabilization (−1 to +1 points). Attrition was high; 29/51 discontinued within the first year, most commonly due to cost, program complexity, and lack of caregiver support. Conclusions: This program demonstrated encouraging preliminary findings of cognitive stabilization or improvement in participants with MCI or AD who remained enrolled. Controlled prospective trials are needed to determine efficacy and mechanisms.

Keywords

multimodal therapy Alzheimer’s disease mild cognitive impairment cognitive function geriatrics integrative medicine functional medicine lifestyle medicine patient-centered care

Introduction

Alzheimer’s disease (AD) and related dementias are significant public health challenges with limited treatment options. Several major memory centers have been investigating approaches to address suspected underlying drivers of cognitive decline through personalized, multi-modal interventions.^1-5 Through this personalized approach, several patient-specific factors are evaluated and targeted for intervention, including lifestyle, hormonal, and neurological adjustments. It is hypothesized that by targeting these areas through personalized medicine, there is an opportunity to slow and in some cases even reverse new-onset late-life cognitive decline. This case series presents outcomes of the first 51 patients who participated in one such program and aims to add to preliminary evidence regarding the efficacy of a personalized multimodal integrative functional medicine approach to cognitive decline in the context of prior uncontrolled and ongoing controlled trials.^1-5

Methods

This retrospective chart analysis includes 51 patients enrolled between 2017 and 2025 in the program at the Center for Integrative Medicine (affiliated with The George Washington University School of Medicine and Health Sciences). Participants were selected based on their clinical diagnosis of mild cognitive impairment (MCI) or AD. All participants underwent a battery of standardized cognitive tests (Montreal Cognitive Assessment [MoCA] and/or Boston Cognitive Assessment [BOCA])⁶ with a subset completing imaging (volumetric brain MRI and/or amyloid PET scan). MoCA was prespecified as the primary cognitive outcome for this case series. Volumetric MRI and Alzheimer’s disease blood biomarkers were obtained in a subset of participants as exploratory measures and are reported descriptively. The majority of participants also underwent standard neuropsychiatric assessment at our affiliated academic memory centers, where diagnosis was established by an MD specializing in memory disorders. Alternatively, patients presented with pre-existing diagnosis that was confirmed by chart review and included documented MCI or AD diagnosis established by qualified provider such as neurologist or memory expert. In a subset of patients, Alzheimer’s disease blood biomarkers were obtained as part of routine specialty evaluation, including Aβ42/Aβ40 ratio, phosphorylated tau (eg, p-tau217) within an amyloid/tau/neurodegeneration (‘ATN’) panel, and neurofilament light (NfL). The Aβ42/Aβ40 ratio is a validated proxy for amyloid pathology, p-tau reflects tau tangle biology, and NfL provides an index of neurodegeneration; however, these tests were introduced into the clinic only in the last ∼24 months and were obtained at limited, non-standardized time points. Data on 4 patients from the program described herein have been previously published.¹

The comprehensive approach included diet (mild ketogenic diet and, if not possible, Mediterranean); all patients were on daily time-restricted eating (no caloric intake from PM to AM for at least 14 and sometimes up to 20 h)^7-9; structured exercises (yoga, Tai Chi, or other aerobic/strength activities for 30-60 minutes/day, 6 days/week)^10-12; individualized supplementation (high DHA:EPA fish oil, choline, B vitamins, and others; a detailed list is provided in Appendix A); and hormonal adjustments. “Mild ketogenic” refers to a Ketoflex-style dietary pattern that is keto-leaning rather than strictly ketogenic. Patients are encouraged to limit refined carbohydrates and emphasize non-starchy vegetables, healthy fats, and adequate protein, but they are not instructed to track ketones or maintain sustained nutritional ketosis.¹³ As part of the program, participants also underwent comprehensive endocrine evaluation, including assessment of thyroid function, sex hormones, adrenal axis markers, and other trophic hormones as clinically indicated. Identified abnormalities were managed with lifestyle interventions and, when appropriate, hormone replacement aimed at restoring values toward physiologic, age-appropriate ranges, consistent with the precision-medicine principles of established protocols.¹³ Detoxification from heavy metals was performed using common functional medicine protocols in close affiliation with one of several local biologic dentists when amalgam removal was required.¹⁴ Patients with suspected chronic inflammatory response syndrome (CIRS) or other biotoxin-related illness underwent evaluation and management informed by the Shoemaker Protocol, a structured, sequential clinical approach developed by Ritchie Shoemaker, MD, for patients with illness attributed to mold and water-damaged buildings. The protocol emphasizes removal of ongoing exposure, targeted laboratory-guided intervention to bind and clear circulating biotoxins, stepwise correction of associated inflammatory and physiologic abnormalities, and serial monitoring to guide progression through defined treatment stages.¹⁵ Sleep optimization and sleep apnea correction were done through an affiliated academic medicine sleep center and/or private practice sleep dentist. Neurostimulation was provided using commercially available 40 Hz gamma light and sound therapy via a free app.^16-19 Cognitive training was provided using various free and commercial tools, including programs such as BrainHQ,²⁰ supported by evidence that computerized cognitive training benefits memory and global cognition in individuals with MCI or dementia.^21,22 The program has required at least 2 h of daily engagement between exercise, cognitive training, and light/sound therapy. All patients were required to demonstrate capacity to adhere to the program or have a caregiver who was willing and able to assure adherence. The program drew conceptually from previously described protocols.¹³

Cognitive function was measured using standardized neuropsychological tests: all participants had MoCA testing at baseline and every 6 months, with >50% of participants also doing other tests (eg, BOCA, CNS-Vitals, SLAMs) at variable intervals; a minority did BOCA weekly. Overall health data (comorbid condition control, laboratory results, weight, and clinician-documented functional status) were collected at baseline and routine follow-up visits. Additionally, most participants had at least one MRI with brain volumetric assessment using NeuroQuant, an FDA-cleared (510[k]) automated tool for T1-weighted brain MRI that segments major structures (including left and right hippocampi), calculates absolute volumes and age/sex/intracranial-volume–adjusted normative percentiles (1-99), and compares each patient with a large reference database.²³ For descriptive purposes, participants were categorized as “improved,” “unchanged,” or “progressed” based on change in MoCA from baseline to last follow-up: improved (≥2-point increase), unchanged (−1 to +1 points), and progressed (≤−2-point decrease).

Lifestyle, comorbidity management, and quality-of-life observations were based on clinician documentation rather than standardized instruments.

The clinical team consisted of an MD specializing in functional medicine with training in geriatrics and memory disorders, a nutritionist with specialized training in Alzheimer’s disease dietary approaches, a naturopathic doctor specializing in bio-identical hormonal replacement and managing of metabolic disorders, a chiropractor specializing in mold detoxification protocol, and a licensed social worker (LSW) specializing in cognitive training and family therapy around issues of patients with Alzheimer’s disorder. An additional program coordinator with master’s in integrative medicine served as program coordinator and coach fully versed in all parts of the program.

Not all patients had to see all providers; this depended on what drivers were present. However, all patients had to see the MD overseeing entire program at least once per year, and often quarterly. The LSW led weekly virtual support group for those interested in additional one-on-one visits for either help with cognitive training or psychological counseling.

Management of vascular risk factors (hypertension, hyperlipidemia, diabetes), tobacco and alcohol counseling, and hearing evaluation were conducted through participants’ primary clinicians, consistent with standard dementia prevention guidelines.

Results

Patient Demographics

The analyzed study population of those remaining in the program for at least 2.5 years (n = 22) had a mean age of approximately 74 years (range early 60s to late 80s). All participants underwent APOE4 allele testing; approximately 50% were positive for at least one allele. The sex distribution in this analyzed group was approximately 60% women and 40% men (about 13/22 and 9/22, respectively). Comorbidities were present in the majority; baseline hippocampal volume and serum markers (Ptau217, ATN) were collected but not in all patients due to various practical clinical factors. Approximately half of all patients were co-managed by our affiliated academic memory center where care was provided as usual. Of note, only 1 patient in the cohort was on newer anti-amyloid medications while most patients were on older Alzheimer’s medications. A detailed breakdown of MCI vs AD subtype and specific medical comorbidities was not consistently available from chart abstraction and is therefore not reported. Of the 51 enrolled participants, 30 (59%) were female.

Table 1.

Baseline Characteristics and MoCA Outcomes by Cognitive Trajectory Among Participants With ≥2.5 Years of Follow-Up (n = 22)

Category	n (%)	Mean age in years	APOE4 ≥1	Average baseline MoCA (range)	Average follow-up duration in years, (range)	Most recent MoCA mean (range)
Improved	7 (32%)	80.2	75%	22 (19-26)	2.9 (2.5-5.5)	26 (23-30)
Unchanged	9 (41%)	73.1	33.3%	24 (22-26)	2.5 (2.5-4.5)	24 (22-26)
Progressed	6 (27%)	69.2	20%	18 (5-27)	2.7 (2.5-4.5)	16 (2-23)

Values are mean (range) unless otherwise indicated. APOE4 ≥1 = proportion of participants with at least one APOE4 allele. MoCA = Montreal Cognitive Assessment.

Outcomes

• Cognitive function (MoCA): Among participants who remained in the program, average baseline MoCA was 21.7 and average at 2.5 years was 22.5. Overall, 73% (16/22) of participants demonstrated either improvement (≥2-point increase) or stabilization (−1 to +1 points) in MoCA. As shown in Table 1, participants in the “improved” group were older and more likely to carry at least one APOE4 allele, whereas those in the “progressed” group were younger and started with lower baseline MoCA scores.

• Imaging: Longitudinal volumetric MRI with NeuroQuant was available for a pragmatically selected subset of participants, often with only 1 or 2 scans at variable time points. Because of this incomplete and heterogeneous sampling, we did not attempt cohort-level quantitative analysis analogous to Table 1. Qualitatively, several participants showed increases in age- and sex-adjusted hippocampal percentile; in one illustrative case, the hippocampal percentile increased from a single-digit value at baseline to the 32nd percentile after 2 years, moving from markedly reduced to within the normal range.

• Serum markers: Alzheimer’s disease blood biomarkers (Aβ42/Aβ40, p-tau217/ATN panels, and NfL) were obtained in a subset of participants, typically at 1 or 2 time points, after these assays became clinically available. Three participants who improved or stabilized on MoCA also showed improvement in their ATN profile. Given the small sample size, non-standardized timing, and incomplete longitudinal data, we report these biomarker findings as exploratory observations only and do not perform cohort-level quantitative analyses.

• Adherence challenges: 29 participants discontinued within the first 12 months; common reasons: cost, program complexity, lack of caregiver support at home.

• Cost considerations: Cost varied widely; from under $3000 to over $30,000 in the first year; average ∼$10,000 in first year, with approximately 30% cost reduction in subsequent years. Largest part of the year 1 cost depended on dental care, patients who had to have many amalgams removed had over 50% of the cost of year one being due to dental work. The relationship between cost and outcome was not analyzed.

• Attrition: Due to the pragmatic, observational nature of the program, we were not able to assess any outcomes in drop-out patients due to lack of follow-up, but the average age was 74 years and APOE4 ≥1 copy was present in 40%.

• Statistical analysis: Due to uncontrolled design and heterogeneity of the cohort, no formal statistical analyses were performed.

Discussion

The preliminary results from this case series suggest that this novel integrative functional medicine program may offer a promising approach to reversing or slowing cognitive decline in patients with MCI or AD. The comprehensive nature of the intervention, targeting multiple pathways involved in cognitive decline, appears to contribute to the observed outcomes.

These findings are broadly consistent with prior multidomain and lifestyle-based interventions, including precision-medicine and intensive lifestyle protocols, which have demonstrated stabilization or improvement of cognition in MCI and early AD,^1-5 as well as dietary,^7-9 exercise,^10-12 cognitive training,^21,22 and gamma sensory stimulation interventions.^16-19 Together, these data support the biological and clinical plausibility of a comprehensive, personalized program that simultaneously targets diet, physical activity, sleep, detoxification, neuromodulation, and cognitive training. In addition to the integrative program, patients continued to receive standard medical management of comorbidities and lifestyle interventions through their primary clinicians.

However, the observational nature of this project and the lack of a randomized control group limit the strength of causal inference. We explicitly acknowledge potential sources of bias — including regression to the mean, placebo or Hawthorne effects, selection bias (patients highly motivated and willing to commit to intensive integrative protocol), and absence of a standardized comparator arm. The cost of the intervention, variability in adherence, and requirement for caregiver support further limit generalizability.

Additionally, several participants survived many years beyond what was predicted. One home-hospice participant lived 9 additional years (for a total of 21 years after diagnosis) and passed from progressive frailty without ever declining further cognitively, reflecting evolving ethical, legal and financial challenges as Alzheimer’s becomes a prolonged disease-trajectory context rather than terminal illness with expected accelerated phase of decline in last 1-2 years of illness. While we cannot infer that the program contributed to longevity; rather, this case illustrates the complexities that may arise when cognitive decline stabilizes despite progressive frailty.

Future Directions

Prospective randomized controlled trials in senior-living communities and broader geriatric populations are needed. Mechanistic investigations (imaging, biomarkers, neuroplasticity changes) should accompany these trials to elucidate pathways of cognitive improvement and reversal. As of the time this publication is undergoing, a separate randomized controlled trial of a similar program has been concluded and peer reviewed results expected to be published shortly.³

Conclusion

This program represents a novel, personalized approach to the treatment of cognitive decline. The encouraging outcomes observed in this initial cohort of 51 participants suggest that it could be an effective strategy for managing MCI and Alzheimer’s disease, but validation through controlled clinical research is necessary to support broader implementation and understand underlying mechanisms.

Footnotes

ORCID iD

Leigh A. Frame

Author Contributions

Dr. Mikhail Kogan and Mehrshad Fahim Devin led manuscript writing and preparation. Drs. Leigh Frame and Christina Prather provided manuscript editing and guidance. All other co-authors provided clinical care for program participants and offered editing of the manuscript. Dr Kogan is an advisor and co-owner of Alz.life and CMO/co-owner of the Center for Integrative Medicine.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. Kogan is an advisor and co-owner of Alz. life (one of many elements in the protocol) and CMO/co-owner of the clinic, the GW Center for Integrative Medicine.

Appendix

Appendix A.

Typical Supplementation List With Common Daily Doses Ranges, Typically Divided into Two Daily Doses

Vitamin A, mixed carotenoids	2500-5000 IU
Vitamin C (as ascorbic acid)	500-1000 mg
Vitamin D3 (as cholecalciferol)	0-5000 IU
Thiamin (vitamin B-1) (as thiamin HCL 25 mg and benfotiamine 50 mg)	50-150 mg
Riboflavin (vitamin B-2)	25-100 mg
Niacin (vitamin B-3)(as niacinamide)	50-100 mg
Vitamin B-6 (as pyridoxine HCL and pyridoxal-5-phosphate)	10-50 mg
Folate (only methylated form used)	500-5000 mcg
Vitamin B-12 (methylcobalamin or adenosylcobalamine)	500-5000 mcg
Biotin (as d-biotin)	1-5 mg
Pantothenic acid (as d-calcium pantothenate)	100-250 mg
Iodine (as potassium iodide)	0-150 mcg
Magnesium (elemental) (from 2000 mg Magtein™ Magnesium L-threonate)	144 mg
Zinc (TRAACS® zinc glycinate chelate)	10-30 mg
Selenium (as selenium glycinate complex)	0-100 mcg
Manganese (TRAACS® manganese glycinate chelate)	0-5 mg
Chromium (TRAACS® chromium nicotinate glycinate chelate)	0-500 mcg
Molybdenum (TRAACS® molybenum glycinate chelate)	0-100 mcg
Potassium (as potassium glycinate complex)	0-100 mcg
R- alpha lipoic acid	0-200 mg
Taurine	0-1000 mg
Green tea (camellia sinensis)(leaves) [standardized to contain 45% EGCg, 98% polyphenols]	200 mg
High gamma mixed tocopherols (as d-gamma, d-delta, d-alpha, d-beta)	165 mg
Vitamin K2 (MK7)	100-200 mcg
Meriva brand Curcumin	1000 mg
Resveratrol	250 mg
CoQ10 – Ubiquinol	0-100 mg
Acetyl-L-Carnitine	1000-1500 mg
GPC-Choline	400-600 mg
Wild blueberry extract	150 mg
Bacopa monniera extract (leaf) (standardized to contain 20% bacosides A & B) (CDRI 08®)	300-500 mg
Lemon balm (melissa officinalis) extract (leaf) (standardized to contain 5% rosemarinic acid)	250-500 mg
Ginkgo (Ginkgo biloba) leaf extract	120-360 mg
N-Acetyl-Cysteine	600 mg
Lion’s Mane	1500 mg
Creatine (added to all patients as of mid 2024)	10,000 mg
Lithium Orotate (switched to Lithium Ascorbate in mid 2024)	5-20 mg
Copper	0-1 mg

References

Bredesen

Sharlin

Jenkins

, et al. Reversal of cognitive decline: 100 patients. J Alzheimers Dis Parkinsonism. 2018;8:450. doi:10.4172/2161-0460.1000450.

Bredesen

Amos

Canick

, et al. Precision medicine approach to Alzheimer’s disease: successful pilot project. J Alzheimers Dis. 2022;88(4):1411-1421. doi:10.3233/JAD215707.

LPWell . Dementia reversal trial: ReCODE randomized clinical trial. 2023. Available from. https://www.lpwell.com/dementia-reversal-trial-2023

Ornish

Madison

Kivipelto

, et al. Effects of intensive lifestyle changes (diet, exercise, stress management, support) on progression of MCI or early AD. Alzheimers Res Ther. 2024;16(1):122. doi:10.1186/s13195-024-01482-z.

Bereczki

Mangialasche

Barbera

, et al. Clinical trials combining multidomain interventions to prevent cognitive decline: systematic review including FINGER and related studies. J Prev Alzheimers Dis. 2025;12(10):100367. doi:10.1016/j.tjpad.2025.100367.

Vyshedskiy

Netson

Fridberg

, et al. Boston cognitive assessment (BOCA) - a comprehensive self-administered smartphone- and computer-based at-home test for longitudinal tracking of cognitive performance. BMC Neurol. 2022;22:92. doi:10.1186/s12883-022-02620-6.

Devranis

Margarity

Angelopoulou

, et al. Mediterranean diet, ketogenic diet or MIND diet for aging, cognition, and dementia: a systematic review. Life (Basel). 2023;13(1):154. doi:10.3390/life13010154.

Fekete

Szabó

Szabó-Salfay

, et al. The role of the Mediterranean diet in reducing the risk of cognitive impairment, dementia, and Alzheimer’s disease: a meta-analysis. GeroScience. 2025;47(1):153-167. doi:10.1007/s11357-024-01488-3.

Rong

Peng

Shen

, et al. Effects of ketogenic diet on cognitive function of patients with AD or MCI: meta-analysis/systematic review. Nutr Health Aging. 2024;28(10):100306. doi:10.1016/j.jnha.2024.100306.

10.

Lopez-Ortiz

Valenzuela

Seisdedos

, et al. Exercise interventions in Alzheimer’s disease: a systematic review and meta-analysis of randomized controlled trials. Ageing Res Rev. 2021;72:101479. doi:10.1016/j.arr.2021.101479.

11.

Liu

Zhang

Wang

, et al. Effect of physical exercise on cognitive function of older patients with Alzheimer’s disease or dementia: a meta-analysis. Front Psychiatry. 2022;13:927128. doi:10.3389/fpsyt.2022.927128.

12.

Andrade

Siqueira

D'Oliveira

Dominski

. Effects of exercise in the treatment of Alzheimer’s disease: an umbrella review of systematic reviews and meta-analyses. J Aging Phys Activ. 2022;30(3):535-551. doi:10.1123/japa.2021-0033.

13.

Apollo Health . ReCODE Protocol. Apollo Health; 2023. https://www.apollohealthco.com/recode-protocol. Accessed April 29, 2025.

14.

Lerner

. General guidelines for the patient undergoing removal of mercury-amalgam fillings. Yorktown Heights, NY: The Center for Holistic Dentistry. [cited 2025 Aug 18]. Available from. https://holisticdentist.com/wp-content/uploads/2020/03/General-Guidelines-for-the-Patient-Undergoing-Removal-of-Mercury-Amalgam-Fillings.pdf

15.

Dooley

Vukelic

Jim

. Chronic inflammatory response syndrome: a review of the evidence of clinical efficacy of treatment. Ann Med Surg. 2024;86(12):7248-7254. doi:10.1097/MS9.0000000000002718. PMID: 39649915; PMCID: PMC11623837.

16.

AlzLife . Multi-Sensory Gamma Stimulation Ameliorates Alzheimer’s-associated Pathology and Improves Cognition. AlzLife; 2019. https://www.alz.life/gamma-news/multi-sensory-gamma-stimulation-ameliorates-alzheimers-associated-pathology-and-improves-cognition/

17.

McNett

Vyshedskiy

Savchenko

, et al. A feasibility study of AlzLife 40 Hz sensory therapy in patients with MCI and early AD. Healthcare. 2023;11(14):2040. doi:10.3390/healthcare11142040

18.

Chan

Suk

Jackson

, et al. Gamma frequency sensory stimulation in mild probable Alzheimer’s dementia: phase 1 and 2A trials. PLOS ONE. 2022;17(12):e0278412. doi:10.1371/journal.pone.0278412

19.

Manippa

Palmisano

Filardi

, et al. An update on the use of gamma (multi)sensory stimulation for Alzheimer’s disease treatment. Front Aging Neurosci. 2022;14:1095081. doi:10.3389/fnagi.2022.1095081.

20.

Posit Science . BrainHQ Exercises. BrainHQ; 2023. https://www.brainhq.com/exercises/

21.

Chan

ATC

RTF

Tran

JYS

Chan

JYC

Tsoi

KKF

. Computerized cognitive training for memory functions in mild cognitive impairment or dementia: a systematic review and meta-analysis. Npj Digit Med. 2024;7:1. doi:10.1038/s41746-023-00987-5.

22.

Belleville

Cuesta

Bier

, et al. Five-year effects of cognitive training in individuals with mild cognitive impairment. Alzheimer’s Dement. 2024;16:e12626. doi:10.1002/dad2.12626.

23.

Persson

Barca

Cavallin

, et al. Comparison of automated volumetry of the hippocampus using NeuroQuant^® and visual assessment of the medial temporal lobe in Alzheimer’s disease. Acta Radiol. 2018;59(8):997-1001. doi:10.1177/0284185117743778.