Return to search

Reliability of Cognitive Assessment for Older Adults via Video Consultation

Abstract Older adults with complex memory problems benefit when they have the opportunity to consult with a specialist for a comprehensive cognitive assessment. Specialists such as geriatricians, psychogeriatricians or neurologists often work in major cities or large metropolitan towns. Travelling to see a specialist either alone or with a carer is complicated for an older person because of medical issues or disability. The use of videoconferencing may provide a way to link a specialist with a patient without the need for the patient, or the specialist, to travel long distances. Two literature reviews were carried out. The initial review identified studies of the diagnosis of cognition via telemedicine. Thirty-two studies were identified which assessed cognition via telephone or video conference. The focus of the study was either the administration of a standardised cognitive assessment tool (n=30) or an unstructured comprehensive cognitive assessment interview (all via video conference) (n=2). The sample sizes were small but the levels of agreement were high, suggesting that further work in this area may identify that diagnosis of dementia via video conference is reliable. There has been limited work in the area of diagnostic agreement when a specialist is assessing a patient for the first time via video conference, even less work in the area of mental health assessment of older people. The second literature review identified 19 studies of diagnostic agreement using video conference with a sample size of 20 or more. The fields of research were: Dermatology (n=10); Mental Health (n=4); Minor Injuries (n=2); Neurology (n=2); and Rheumatology (n=1). Of the four studies in the area of mental health, one focused on the diagnosis of dementia with the publication of a protocol for assessing Alzheimer’s disease (AD) via video conference. The review highlighted that diagnosis via video conference in other medical fields had been shown to be reliable but that limited work was evident regarding the reliability of diagnosing dementia via video conference. A range of statistical analyses have been used to measure agreement in studies of diagnosis via VC. Overall Proportional Agreement (Po) and Cohen’s kappa (K) are the two most common calculations. There was little uniformity of reporting in the studies identified in the literature review. The variation in reporting made it difficult to compare results or provide data for a meta-analysis of similar studies. Consideration of the methods for analysing diagnostic agreement was undertaken using the approaches identified in the literature review as a starting point. The aim of this research was to identify if a diagnosis of the presence of dementia in an older adult by a specialist is reliable when the assessment interview occurs via video conference. A pilot study was carried out prior to the implementation of a National Health and Medical Research Council (NHMRC) funded multi-site project to test inter-rater agreement for the diagnosis of dementia and subsidiary questions. The candidate is a Chief Investigator (CI) on the NHMRC project grant and participated in writing the grant submission. A pilot study was completed for the purpose of refining the research protocol and establishing preliminary data for the calculation of sample size. A geriatrician carried out a cognitive assessment via video conference with the patient and the carer. The doctor had access to the patient’s chart and the results of a battery of standardised cognitive assessments administered face-to-face (FTF) by the clinic nurse earlier in the day. A second interview was carried out, face-to-face, by a second doctor on the same day. Inter-rater reliability was assessed between doctors. To place the level of agreement in context, inter-rater reliability between paired face-to-face assessments was also measured. Forty-two participants were divided into two groups: paired face-to-face assessments (FF, n=22) or paired video conference and face-to-face assessments (FV/VF, n=20). Twenty-two participants were male. Their average age was 70 years (SD=11.1, Range 50-90). The mean Standardised Mini-Mental State Examination (SMMSE) score was 23.93 (SD=5.42, Range 8-30)[1]. The outcome of agreement was measured using PO and Cohen’s K. FF group (PO=0.636; K=0.430, p=0.005) and the FV/VF group (PO=0.650; K=0.650, p=0.004) showed similar levels of agreement. The main study was a non-inferiority, prospective cohort study following a similar format as the pilot. Patients were randomised to two groups. The first group participated in two interviews; one via VC, the other was face-to-face (FV/VF). The second group was subjected to dual face-to-face assessments (FF). The video conference interview consisted of real-time video conference with the doctor interacting with both the carer and the patient. A battery of standardised assessments, a medical history, imaging and blood tests were prepared before hand and available to the doctor for use in the interview. One hundred and fifty-five participants were divided into two groups: paired face-to-face assessments (FF, n=73) or paired video conference and face-to-face assessments (FV or VF, n=82). Seventy-five of the participants were male. The average age was 76-years (SD=9, Range 54-95). The mean SMMSE was 23.8 (SD=4.4, Range 8-30). Overall proportional agreement (Po) and Weighted K were calculated as a measure of agreement for the presence of dementia. The FF group (Po=0.740; K=0.57, p<0.0001) and the FV/VF group (Po=0.780; K=0.64, p<0.0001) showed similar levels of agreement. Using the same study cohort, additional data were collected to identify the significance of the physical examination (PE) for diagnosing dementia. One of two doctors was allocated the task of completing a physical examination of the patient following initial assessment interview and after the diagnosis decisions had been recorded. Following the physical examination a second record of diagnosis decisions were recorded by the same doctor. The doctor was asked to identify if the diagnosis, formulation, treating options or additional investigations had altered as a result of the in-person physical examination. The physical examination supported clarity of the diagnosis particularly if a vascular element was involved. This dissertation provides evidence of the reliability of a diagnosis of dementia obtained via video conference. Furthermore, the work detailed in this dissertation represents the largest international study on assessing diagnostic accuracy of cognition and makes a significant contribution to the work in telemedicine in the area of mental health. This dissertation provides generalisations that can impact the use of video conference for diagnosis across a range of specialities, with the ultimate goal of improving access to specialist advice for people living in rural areas.

Identiferoai:union.ndltd.org:ADTP/279149
CreatorsMelinda Martin-khan
Source SetsAustraliasian Digital Theses Program
Detected LanguageEnglish

Page generated in 0.0023 seconds