Predictors of mortality shortly after entering a long-term care facility.

OBJECTIVE: Moving into a long-term care facility (LTCF) requires substantial personal, societal and financial investment. Identifying those at high risk of short-term mortality after LTCF entry can help with care planning and risk factor management. This study aimed to: (i) examine individual-, facility-, medication-, system- and healthcare-related predictors for 90-day mortality at entry into an LTCF and (ii) create risk profiles for this outcome. DESIGN: Retrospective cohort study using data from the Registry of Senior Australians. SUBJECTS: Individuals aged ≥ 65 years old with first-time permanent entry into an LTCF in three Australian states between 01 January 2013 and 31 December 2016. METHODS: A prediction model for 90-day mortality was developed using Cox regression with the purposeful variable selection approach. Individual-, medication-, system- and healthcare-related factors known at entry into an LTCF were examined as predictors. Harrell's C-index assessed the predictive ability of our risk models. RESULTS: 116,192 individuals who entered 1,967 facilities, of which 9.4% (N = 10,910) died within 90 days, were studied. We identified 51 predictors of mortality, five of which were effect modifiers. The strongest predictors included activities of daily living category (hazard ratio [HR] = 5.41, 95% confidence interval [CI] = 4.99-5.88 for high vs low), high level of complex health conditions (HR = 1.67, 95% CI = 1.58-1.77 for high vs low), several medication classes and male sex (HR = 1.59, 95% CI = 1.53-1.65). The model out-of-sample Harrell's C-index was 0.773. CONCLUSIONS: Our mortality prediction model, which includes several strongly associated factors, can moderately well identify individuals at high risk of mortality upon LTCF entry.

Deep learning for prediction of colorectal cancer outcome: a discovery and validation study.

BACKGROUND: Improved markers of prognosis are needed to stratify patients with early-stage colorectal cancer to refine selection of adjuvant therapy. The aim of the present study was to develop a biomarker of patient outcome after primary colorectal cancer resection by directly analysing scanned conventional haematoxylin and eosin stained sections using deep learning. METHODS: More than 12 000 000 image tiles from patients with a distinctly good or poor disease outcome from four cohorts were used to train a total of ten convolutional neural networks, purpose-built for classifying supersized heterogeneous images. A prognostic biomarker integrating the ten networks was determined using patients with a non-distinct outcome. The marker was tested on 920 patients with slides prepared in the UK, and then independently validated according to a predefined protocol in 1122 patients treated with single-agent capecitabine using slides prepared in Norway. All cohorts included only patients with resectable tumours, and a formalin-fixed, paraffin-embedded tumour tissue block available for analysis. The primary outcome was cancer-specific survival. FINDINGS: 828 patients from four cohorts had a distinct outcome and were used as a training cohort to obtain clear ground truth. 1645 patients had a non-distinct outcome and were used for tuning. The biomarker provided a hazard ratio for poor versus good prognosis of 3·84 (95% CI 2·72-5·43; p<0·0001) in the primary analysis of the validation cohort, and 3·04 (2·07-4·47; p<0·0001) after adjusting for established prognostic markers significant in univariable analyses of the same cohort, which were pN stage, pT stage, lymphatic invasion, and venous vascular invasion. INTERPRETATION: A clinically useful prognostic marker was developed using deep learning allied to digital scanning of conventional haematoxylin and eosin stained tumour tissue sections. The assay has been extensively evaluated in large, independent patient populations, correlates with and outperforms established molecular and morphological prognostic markers, and gives consistent results across tumour and nodal stage. The biomarker stratified stage II and III patients into sufficiently distinct prognostic groups that potentially could be used to guide selection of adjuvant treatment by avoiding therapy in very low risk groups and identifying patients who would benefit from more intensive treatment regimes. FUNDING: The Research Council of Norway.

Computerized image analysis of the Ki-67 proliferation index in mantle cell lymphoma.

AIMS: Manual counting of the fraction of Ki-67-positive cells (the Ki-67 index) in 1000 tumour cells is considered the 'gold standard' to predict prognosis in mantle cell lymphoma (MCL). This time-consuming method is replaced by the faster, but less accurate, semiquantitative estimation in routine practice. The aim of this study was to investigate the use of computerized image analysis software for scoring of Ki-67 in MCL. METHODS AND RESULTS: We developed an automated method for determining the Ki-67 index by computerized image analysis and tested it using a cohort of 62 MCL patients. The data were compared to Ki-67 scores obtained by semiquantitative estimation and image-based manual counting. When using the Ki-67 index as a continuous parameter, both image-based manual counting and computerized image analysis were related inversely to survival (P = 0.020 and P = 0.025, respectively). Ki-67 index obtained by semiquantitative estimation was not associated significantly with survival (P = 0.093). The results were validated in a second patient cohort with similar results. CONCLUSION: Computerized image analysis of the Ki-67 index in MCL is an attractive alternative to semiquantitative estimation and can be introduced easily in a routine diagnostic setting for risk stratification in MCL.

Hospital and out-of-hospital services provided by public geriatric medicine departments in Australia and New Zealand.

OBJECTIVE: To describe the types of hospital and out-of-hospital services provided by public geriatric medicine departments in Australia and New Zealand, and to explore head of department (HOD) views on issues in current and future service provision. METHODS: An electronic survey was sent to HODs of public geriatric medicine departments. RESULTS: Seventy-six (89%) of 85 identified HODs completed the survey. Seventy-one (93%) departments admit inpatients and 51 (67%) admit acute inpatients, with variable admission criteria. Sixty-four (84%) have hospitals with an inpatient general medicine service, and 58 (91%) of these admit older patients with acute geriatric issues. Sixty (79%) departments provide inpatient rehabilitation. Forty (53%) have beds for behavioural symptoms of dementia and/or delirium. Seventy (92%) provide a proactive orthogeriatric service. In terms of out-of-hospital services, 74 (97%) departments have outpatient clinics, 59 (78%) have telehealth and 68 (89%) perform home visits. Forty-five (59%) provide an inreach/outreach service to nursing homes. The most frequent gaps in service provision identified by HODs were acute geriatrics, surgical liaison, a designated dementia/delirium behavioural management unit, geriatricians in Emergency, outreach/inreach to residential care and shared care with some medical specialities. Increasing staff numbers and government policy change were the most frequently identified ways to address these gaps. CONCLUSIONS: Geriatric medicine service provision is variable across Australia and New Zealand, with key gaps identified. These findings will inform future directions in implementation of geriatric medicine models of care and discussions with various levels of government about the ongoing development of geriatric medicine services.

Translational artificial intelligence-led optimization and realization of estimated discharge with a supportive weekend interprofessional flow team (TAILORED-SWIFT).

Weekend discharges occur less frequently than discharges on weekdays, contributing to hospital congestion. Artificial intelligence algorithms have previously been derived to predict which patients are nearing discharge based upon ward round notes. In this implementation study, such an artificial intelligence algorithm was coupled with a multidisciplinary discharge facilitation team on weekend shifts. This approach was implemented in a tertiary hospital, and then compared to a historical cohort from the same time the previous year. There were 3990 patients included in the study. There was a significant increase in the proportion of inpatients who received weekend discharges in the intervention group compared to the control group (median 18%, IQR 18-20%, vs median 14%, IQR 12% to 17%, P = 0.031). There was a corresponding higher absolute number of weekend discharges during the intervention period compared to the control period (P = 0.025). The studied intervention was associated with an increase in weekend discharges and economic analyses support this approach as being cost-effective. Further studies are required to examine the generalizability of this approach to other centers.

The pharmacokinetics and pharmacodynamics of single dose (R)- and (S)-warfarin administered separately and together: relationship to VKORC1 genotype.

AIMS: 1) To determine the pharmacokinetics and pharmacodynamics of (R)- and (S)-warfarin given alone and in combination and 2) to determine whether the relative potency of (R)- and (S)-warfarin is dependent on VKORC1 genotype. METHODS: A three way crossover study was conducted in which 17 healthy male subjects stratified by VKORC1 1173 C>T genotype and all CYP2C9 1*/1* received (R)-warfarin 80 mg, (S)-warfarin 12.5 mg and rac-warfarin sodium 25 mg. Plasma (R)- and (S)-warfarin unbound and total concentrations and prothrombin time were determined at multiple time points to 168 h. RESULTS: Pharmacokinetic parameters for (R)- and (S)-warfarin were similar to the literature. (R)-warfarin 80 mg alone resulted in a mean AUC(PT) (0,168 h) of 3550 s h (95% CI 3220, 3880). Rac-warfarin sodium 25 mg containing (S)-warfarin 11.7 mg produced a greater effect on AUC(PT) (0,168 h) than (S)-warfarin 12.5 mg (mean difference 250 s.h, 95% CI 110, 380, P < 0.002) given alone. In a mixed effects model the ratio of response between (R)- and (S)-warfarin (AUC(PT((R)-warfarin)) : AUC(PT((S)-warfarin))) was 1.21 fold higher (95% CI 1.05, 1.41, P < 0.02) in subjects of VKORC1 TT genotype compared with the CC genotype. CONCLUSIONS: (R)-warfarin has a clear PD effect and contributes to the hypoprothrombinaemic effect of rac-warfarin. VKORC1 genotype is a covariate of the relative R/S potency relationship. Prediction of drug interactions with warfarin needs to consider effects on (R)-warfarin PK and VKORC1 genotype.

Can a personalised music listening intervention decrease agitation in hospitalised patients with dementia? A feasibility trial.

Introduction: Agitation is a common manifestation of the behavioural and psychological symptoms of dementia (BPSD). Pharmacotherapy is not the first-line management because of its potential harms, particularly in the elderly. Music as a non-pharmacological intervention for agitation has been explored in residential aged-care facilities, but few studies have been situated in hospitals. This pilot aims to evaluate the feasibility of a personalised music listening intervention for reducing agitation in hospitalised patients with dementia in a metropolitan Geriatric Evaluation and Management (GEM) unit. Methods: Two-arm randomised control feasibility trial. Eligible patients were assigned to the music intervention or control group, with the intervention group receiving music daily between 15:00-16:00, and agitation levels measured in both groups hourly based on the Pittsburgh Agitation Score (PAS) over 5 days of hospitalisation. Post-trial semi-structured interviews assessed feasibility of the intervention. Results: Twenty-one patients were recruited over 8 months. Interviews with staff involved indicated that the music intervention was manageable to deliver, assisted engagement with patients which increased efficiency of some clinical tasks, and challenged staff mindset around using psychotropic medication to address agitation. PAS results were inconclusive, because of underpowered numbers in this pilot study. Conclusion: It is feasible for nursing staff to deliver a personalised music listening intervention to patients with dementia in a geriatric unit of a tertiary hospital, without compromising on usual clinical care.

A retrospective observational study on the types of anaesthesia in hip fracture surgery.

BACKGROUND: There is no specific recommendation regarding the type of anaesthesia in hip fracture surgery. OBJECTIVES: This study sought to examine the current local anaesthetic practice (general anaesthesia versus regional anaesthesia (RA)) in hip fracture surgery and to analyse their associations with perioperative outcomes. METHODOLOGY: A retrospective observational study of hip fracture patients from April to December 2017 was undertaken. Patient characteristics and perioperative outcomes were analysed against the types of anaesthesia using multiple logistic regression. RESULTS: One hundred and twelve out of 154 patients (72.7%) had a general anaesthesia. Patients from residential care facilities were more likely to receive general anaesthesia (OR = 2.9, 95% CI: 1.1, 7.4; P = 0.03). There was no significant association between type of anaesthesia and specific postoperative outcomes; however, patients with postoperative delirium and hypotension were more likely to have received general anaesthesia [OR = 1.7, 95% CI: 0.68, 4.38; P = 0.25] and [OR = 1.6, 95% CI: 0.67, 4.04; P = 0.27] respectively). Subgroup analysis showed increased length of stay with patients who underwent general anaesthesia (OR = 1.26, 95% CI:1.04, 1.54; P = 0.02). CONCLUSION: Regional anaesthesia may be considered in patients without contraindications in view of increased risk of postoperative delirium and hypotension, and longer length of stay with general anaesthesia. A larger prospective study is needed to confirm these findings.

Detecting retinal cell stress and apoptosis with DARC: Progression from lab to clinic.

DARC (Detection of Apoptosing Retinal Cells) is a retinal imaging technology that has been developed within the last 2 decades from basic laboratory science to Phase 2 clinical trials. It uses ANX776 (fluorescently labelled Annexin A5) to identify stressed and apoptotic cells in the living eye. During its development, DARC has undergone biochemistry optimisation, scale-up and GMP manufacture and extensive preclinical evaluation. Initially tested in preclinical glaucoma and optic neuropathy models, it has also been investigated in AMD, Alzheimer's, Parkinson's and Diabetic models, and used to assess efficacy of therapies. Progression to clinical trials has not been speedy. Intravenous ANX776 has to date been found to be safe and well-tolerated in 129 patients, including 16 from Phase 1 and 113 from Phase 2. Results on glaucoma and AMD patients have been recently published, and suggest DARC with an AI-aided algorithm can be used to predict disease activity. New analyses of DARC in GA (Geographic Atrophy) prediction are reported here. Although further studies are needed to validate these findings, it appears there is potential for the technology to be used as a biomarker. Much larger clinical studies will be needed before it can be considered as a diagnostic, although the relatively non-invasive nature of the nasal as opposed to intravenous administration would widen its acceptability in the future as a screening tool. This review describes DARC development and its progression into Phase 2 clinical trials from lab-based research. It discusses hypotheses, potential challenges, and regulatory hurdles in translating technology.

Salivary inflammatory biomarkers are predictive of mild cognitive impairment and Alzheimer's disease in a feasibility study.

Alzheimer's disease (AD) is an insidious disease. Its distinctive pathology forms over a considerable length of time without symptoms. There is a need to detect this disease, before even subtle changes occur in cognition. Hallmark AD biomarkers, tau and amyloid-ß, have shown promising results in CSF and blood. However, detecting early changes in these biomarkers and others will involve screening a wide group of healthy, asymptomatic individuals. Saliva is a feasible alternative. Sample collection is economical, non-invasive and saliva is an abundant source of proteins including tau and amyloid-ß. This work sought to extend an earlier promising untargeted mass spectrometry study in saliva from individuals with mild cognitive impairment (MCI) or AD with age- and gender-matched cognitively normal from the South Australian Neurodegenerative Disease cohort. Five proteins, with key roles in inflammation, were chosen from this study and measured by ELISA from individuals with AD (n = 16), MCI (n = 15) and cognitively normal (n = 29). The concentrations of Cystatin-C, Interleukin-1 receptor antagonist, Stratifin, Matrix metalloproteinase 9 and Haptoglobin proteins had altered abundance in saliva from AD and MCI, consistent with the earlier study. Receiver operating characteristic analysis showed that combinations of these proteins demonstrated excellent diagnostic accuracy for distinguishing both MCI (area under curve = 0.97) and AD (area under curve = 0.97) from cognitively normal. These results provide evidence for saliva being a valuable source of biomarkers for early detection of cognitive impairment in individuals on the AD continuum and potentially other neurodegenerative diseases.

Multi-Omics, an Integrated Approach to Identify Novel Blood Biomarkers of Alzheimer's Disease.

The metabolomic and proteomic basis of mild cognitive impairment (MCI) and Alzheimer's disease (AD) is poorly understood, and the relationships between systemic abnormalities in metabolism and AD/MCI pathogenesis is unclear. This study compared the metabolomic and proteomic signature of plasma from cognitively normal (CN) and dementia patients diagnosed with MCI or AD, to identify specific cellular pathways and new biomarkers altered with the progression of the disease. We analysed 80 plasma samples from individuals with MCI or AD, as well as age- and gender-matched CN individuals, by utilising mass spectrometry methods and data analyses that included combined pathway analysis and model predictions. Several proteins clearly identified AD from the MCI and CN groups and included plasma actins, mannan-binding lectin serine protease 1, serum amyloid A2, fibronectin and extracellular matrix protein 1 and Keratin 9. The integrated pathway analysis showed various metabolic pathways were affected in AD, such as the arginine, alanine, aspartate, glutamate and pyruvate metabolism pathways. Therefore, our multi-omics approach identified novel plasma biomarkers for the MCI and AD groups, identified changes in metabolic processes, and may form the basis of a biomarker panel for stratifying dementia participants in future clinical trials.

Implementing patient direct access to musculoskeletal physiotherapy in primary care: views of patients, general practitioners, physiotherapists and clinical commissioners in England.

PURPOSE: Musculoskeletal problems are the leading cause of chronic disability. Most patients in the UK seek initial care from general practitioners (GPs), who are struggling to meet demand. Patient direct access to National Health Service physiotherapy is one possible solution. The purpose of this study was to understand the experiences of patients, GPs, physiotherapists and clinical commissioners on direct access in a region in England with it commissioned. METHODS: The study was informed by Normalisation Process Theory (NTP). Data collection was via semi-structured individual face-to-face and telephone interviews with 22 patients and 20 health care professionals (HCPs). Data were analysed thematically using NPT. RESULTS: Three themes emerged: understanding physiotherapy and the direct access pathway; negotiating the pathway; making the pathway viable. HCPs saw direct access as acceptable. Whilst patients found the concept of direct access, those with complex conditions continued to see their GP as first point of contact. Some GPs and patients reported a lack of clarity around the pathway, reflected in ambiguous paperwork and inconsistent promotion. Operational challenges emerged in cross-disciplinary communication and between HCPs and patients, and lack of adequate resources. CONCLUSION: Direct access to NHS musculoskeletal physiotherapy is acceptable to patients and HCPs. There is need to ensure: effective communication between HCPs and with patients, clarity on the scope of physiotherapy and the direct access pathway, and sufficient resources to meet demand. Patient direct access can free GPs to focus on those patients with more complex health conditions who are most in need of their care.

Predicting wet age-related macular degeneration (AMD) using DARC (detecting apoptosing retinal cells) AI (artificial intelligence) technology.

OBJECTIVES: To assess a recently described CNN (convolutional neural network) DARC (Detection of Apoptosing Retinal Cells) algorithm in predicting new Subretinal Fluid (SRF) formation in Age-related-Macular-Degeneration (AMD). METHODS: Anonymized DARC, baseline and serial OCT images (n = 427) from 29 AMD eyes of Phase 2 clinical trial (ISRCTN10751859) were assessed with CNN algorithms, enabling the location of each DARC spot on corresponding OCT slices (n = 20,629). Assessment of DARC in a rabbit model of angiogenesis was performed in parallel. RESULTS: A CNN DARC count >5 at baseline was significantly (p = 0.0156) related to development of new SRF throughout 36 months. Prediction rate of eyes using unique DARC spots overlying new SRF had positive predictive values, sensitivities and specificities >70%, with DARC count significantly (p < 0.005) related to the magnitude of SRF accumulation at all time points. DARC identified earliest stages of angiogenesis in-vivo. CONCLUSIONS: DARC was able to predict new wet-AMD activity. Using only an OCT-CNN definition of new SRF, we demonstrate that DARC can identify early endothelial neovascular activity, as confirmed by rabbit studies. Although larger validation studies are required, this shows the potential of DARC as a biomarker of wet AMD, and potentially saving vision-loss.

Using data linkage for national surveillance of clinical quality indicators for dementia care among Australian aged care users.

Care quality has important implications for people with dementia. We examined trends and geographical variation of four clinical quality indicators (CQIs) in Australia. This retrospective cohort study included all people with dementia using Australian government-subsidised aged care in 2008-2016 (n = 373,695). Quality indicator data were derived from linked national aged care, health, and pharmaceutical datasets. Negative binomial regression modelling assessed trends in CQI performance over time (2011-2016) and funnel plots examined geographical variation in performance. The incidence rate of antipsychotic medicine dispensing decreased slightly from 1.17/1000 person-days to 1.07/1000 person-days (adjusted incidence rate ratio (aIRR) = 0.98, 95%CI 0.98-0.99). Cholinesterase inhibitors and memantine dispensing did not change (aIRR = 1.02, 95%CI 1.00-1.04), while exposure to high sedative load increased slightly from 1.39/1000 person-days to 1.44/1000 person-days (aIRR = 1.01, 95%CI 1.00-1.01). Dementia and delirium-related hospitalisations increased slightly from 0.17/1000 person-days to 0.18/1000 person-days (aIRR = 1.02, 95%CI 1.01-1.03). There was marked variation in cholinesterase inhibitor and memantine dispensing by geographical area (0-41%). There has been little change in four indicators of dementia care quality in Australian aged care users over time. Cholinesterase inhibitor and memantine dispensing varied substantially by geographical region. Existing strategies to improve national performance on these indicators appear to be insufficient, despite the significant impact of these indicators on outcomes for people with dementia.

Cohort profile: Dementia in the Registry of Senior Australians.

PURPOSE: Clinical quality registries (CQRs) are being established in many countries to monitor, benchmark, and report on the quality of dementia care over time. Case ascertainment can be challenging given that diagnosis occurs in a variety of settings. The Registry of Senior Australians (ROSA) includes a large cohort of people with dementia from all Australian states and territories identified using routinely collected aged care assessment data. In ROSA, assessment data are linked to information about aged and health service use, medicine dispensing, hospitalisations and the National Death Index. The ROSA dementia cohort was established to capture people for the Australian dementia CQR currently in development who may not be identified elsewhere. PARTICIPANTS: There were 373 695 people with dementia identified in aged care assessments from 2008 to 2016. Cross-sectional analysis from the time of cohort entry (e.g. when first identified with dementia on an aged care assessment) indicates that individuals were 84.1 years old on average, and 63.1% were female. More than 44% were first identified at entry to permanent residential aged care. The cohort recorded more severe cognitive impairment at entry than other international dementia registries. FINDINGS TO DATE: The cohort has so far been used to demonstrate a declining prevalence of dementia in individuals entering the aged care sector, examine trends in psychotropic medicine prescribing, and to examine the impact of dementia on aged care service use and outcomes. FUTURE PLANS: The ROSA dementia cohort will be updated periodically and is a powerful resource both on its own and as a contributor to the Australian dementia CQR. Integration of the ROSA dementia cohort with the dementia CQR will ensure that people with dementia using aged care services can benefit from the ongoing monitoring and benchmarking of care that a registry can provide.

Predictors of hospitalisations and emergency department presentations shortly after entering a residential aged care facility in Australia: a retrospective cohort study.

OBJECTIVES: To: (1) examine the 90-day incidence of unplanned hospitalisation and emergency department (ED) presentations after residential aged care facility (RACF) entry, (2) examine individual-related, facility-related, medication-related, system-related and healthcare-related predictors of these outcomes and (3) create individual risk profiles. DESIGN: Retrospective cohort study using the Registry of Senior Australians. Fine-Gray models estimated subdistribution HRs and 95% CIs. Harrell's C-index assessed risk models' predictive ability. SETTING AND PARTICIPANTS: Individuals aged ≥65 years old entering a RACF as permanent residents in three Australian states between 1 January 2013 and 31 December 2016 (N=116 192 individuals in 1967 RACFs). PREDICTORS EXAMINED: Individual-related, facility-related, medication-related, system and healthcare-related predictors ascertained at assessments or within 90 days, 6 months or 1 year prior to RACF entry. OUTCOME MEASURES: 90-day unplanned hospitalisation and ED presentation post-RACF entry. RESULTS: The cohort median age was 85 years old (IQR 80-89), 62% (N=71 861) were women, and 50.5% (N=58 714) had dementia. The 90-day incidence of unplanned hospitalisations was 18.0% (N=20 919) and 22.6% (N=26 242) had ED presentations. There were 34 predictors of unplanned hospitalisations and 34 predictors of ED presentations identified, 27 common to both outcomes and 7 were unique to each. The hospitalisation and ED presentation models out-of-sample Harrell's C-index was 0.664 (95% CI 0.657 to 0.672) and 0.655 (95% CI 0.648 to 0.662), respectively. Some common predictors of high risk of unplanned hospitalisation and ED presentations included: being a man, age, delirium history, higher activity of daily living, behavioural and complex care needs, as well as history, number and recency of healthcare use (including hospital, general practitioners attendances), experience of a high sedative load and several medications. CONCLUSIONS: Within 90 days of RACF entry, 18.0% of individuals had unplanned hospitalisations and 22.6% had ED presentations. Several predictors, including modifiable factors, were identified at the time of care entry. This is an actionable period for targeting individuals at risk of hospitalisations.

Salivaomics as a Potential Tool for Predicting Alzheimer's Disease During the Early Stages of Neurodegeneration.

BACKGROUND: The metabolomic and proteomic basis of mild cognitive impairment (MCI) and Alzheimer's disease (AD) is poorly understood and the relationships between systemic abnormalities in metabolism and AD/AMCI pathogenesis are unclear. OBJECTIVE: The aim of the study was to compare the metabolomic and proteomic signature of saliva from cognitively normal and patients diagnosed with MCI or AD, to identify specific cellular pathways altered with the progression of the disease. METHODS: We analyzed 80 saliva samples from individuals with MCI or AD as well as age- and gender-matched healthy controls. Saliva proteomic and metabolomic analyses were conducted utilizing mass spectrometry methods and data combined using pathway analysis. RESULTS: We found significant alterations in multiple cellular pathways, demonstrating that at the omics level, disease progression impacts numerous cellular processes. Multivariate statistics using SIMCA showed that partial least squares-data analysis could be used to provide separation of the three groups. CONCLUSION: This study found significant changes in metabolites and proteins from multiple cellular pathways in saliva. These changes were associated with AD, demonstrating that this approach might prove useful to identify new biomarkers based upon integration of multi-omics parameters.