# Using Informatics to Evaluate and Predict Cataract Surgery Impact on Alzheimer's Disease and Related Dementias and Mild Cognitive Impairment Outcomes

> **NIH NIH R01** · STANFORD UNIVERSITY · 2023 · $769,326

## Abstract

PROJECT SUMMARY/ABSTRACT
Background. Visual impairment has been strongly associated with Alzheimer’s disease and related
dementias (ADRD) in numerous cross-sectional and longitudinal studies, and we have found that worse
baseline vision is tied to increasingly higher risk of subsequent dementia. Neurosensory deprivation from
visual impairment may place greater demands on cognitive resources, accelerating cognitive decline and
increasing the incidence of cognitive impairment. Conversely, improving vision could improve cognitive
outcomes by increasing neurosensory input and reducing cognitive demand for processing visual
information. Cataracts are the most common cause of visual impairment—fortunately reversible with
surgery, however, we have found that ADRD patients are only half as likely to undergo cataract surgery as
those without ADRD. This may reflect concerns regarding less potential benefit and greater perceived risks.
Objectives. Our long-term goal is to evaluate cataract surgery as a potential intervention to “bend the
curve” for risk of ADRD onset and progression, including optimizing patient selection and timing for surgery.
The objective of this proposal is to investigate how cataract surgery may affect incidence and progression of
mild cognitive impairment (MCI) and ADRD, develop models to predict individual patients’ ADRD/MCI
outcomes following cataract surgery, and identify key confounders, mediators, and effect modifiers. We
hypothesize that cataract surgery is associated with (1) reduction in incidence of new MCI and ADRD and
(2) reduced cognitive decline and impairment progression among patients with baseline MCI or ADRD, and
that (3) we will be able to predict individual patient outcomes. We propose to use methods our group has
developed to archive and analyze electronic health record (EHR) data, to develop a curated data set and
achieve three Aims: (1) Determine impact of cataract surgery on ADRD and MCI incidence; (2) Determine
impact of cataract surgery on cognitive decline and impairment among patients with baseline ADRD or MCI,
and (3) Develop patient-level predictive models for ADRD and MCI outcomes after cataract surgery.
Impact. EHR-based machine learning analysis has not been applied to ADRD research to date, and the
influence of cataract surgery on cognitive outcomes is not yet known. Finding that a widely-available cataract
surgery intervention improves cognitive outcomes would be transformative. We estimate a potential unmet
need for cataract surgery affecting almost 350,000 patients annually—just among the subset of patients with
existing Alzheimer’s disease. Results from this work will directly inform discussion of cataract surgery risks
and benefits and will also facilitate future research, including pragmatic clinical trial design. By developing and
disseminating open source EHR-based algorithms to identify and classify cognitive and visual impairment,
this proposal will enable investigation of other ADRD risk fa...

## Key facts

- **NIH application ID:** 10688255
- **Project number:** 5R01AG072582-02
- **Recipient organization:** STANFORD UNIVERSITY
- **Principal Investigator:** Suzann Pershing
- **Activity code:** R01 (R01, R21, SBIR, etc.)
- **Funding institute:** NIH
- **Fiscal year:** 2023
- **Award amount:** $769,326
- **Award type:** 5
- **Project period:** 2022-09-01 → 2027-05-31

## Primary source

NIH RePORTER: https://reporter.nih.gov/project-details/10688255

## Citation

> US National Institutes of Health, RePORTER application 10688255, Using Informatics to Evaluate and Predict Cataract Surgery Impact on Alzheimer's Disease and Related Dementias and Mild Cognitive Impairment Outcomes (5R01AG072582-02). Retrieved via AI Analytics 2026-05-24 from https://api.ai-analytics.org/grant/nih/10688255. Licensed CC0.

---

*[NIH grants dataset](/datasets/nih-grants) · CC0 1.0*