Objective: To characterise and analyse beliefs, knowledge, and attitudes regarding dementia three years after the first Portuguese survey (2015), in order to inform mental health research and policy-making, and establish the baseline to assess “Dementia Friends” campaign impact. Method: An advisory group, including academics and representatives of Alzheimer Portugal and Alzheimer Society UK, drafted the survey based on previous international research and the first national survey. The survey was uploaded to the Directorate-General of Health website and disseminated through relevant stakeholders and social networks. The survey was cascaded by inviting recipients to further share it. Data were collected during July 2018. Results: 1716 individuals completed the survey: median age 43y, 83.4% female, 25.1% professionals working in the field of dementia. 31.7% of respondents know someone close with dementia, and 42.5% have/had relatives with dementia. 14.7% consider people with dementia must leave/stop to attend ceremonies and social events, but this percentage increased to over one-third (37.3%) when asked about the opinion of most people living in their community. Regarding the key messages of “Dementia Friends” campaign, results reveal that 25.6% of respondents consider dementia is a natural part of ageing, 21.3% that it is not caused by diseases of the brain and 17.4% that it is just about losing memory. Conclusions: The second survey results reinforce the need for initiatives to change people’s perceptions of dementia, that build up on evidence-based decision-making and international learning and cooperation as part of “Dementia Friends” global movement.

No abstract

1573 Background: Clinical studies and new drug approvals are delayed by slow data transfer and transcription errors from site entered data. These delays have been compounded by a shortage of data entry personnel at sites such that data entry approaches 10-30 days post visit. Data transfer is largely performed by manual keyboard entry from electronic medical records (EMR) into electronic case report forms (ECRF). Methods: We conducted two separate measure-of-work time studies to compare a commercially available interface software product, ProXimity to the current manual data entry. The clinical trial data from two different EMRs (ARIA and IKM G2) to an EDC (Medidata Rave). The EDC mirrored an IRB approved clinical study. Time to transfer data and error rates were the primary and secondary endpoints, respectively. For study 1 Aria EMR data from 3 subjects and 1497 data fields including demographics, vital signs, ECOG PS, physical findings, adverse events, and lab results including CBC, CMP, urinalysis, coagulation, serology were selected for visits from Screening and C2D1. For Study 2 IKM G2 data from 6 subjects and 834 data fields included demographics, vitals, and lab results. The data entry personnel were aware of the timed nature of the study. Results: Study 1 ProXimity took 13.2 min to transfer the data compared to 73.4 min for manual entry, with error rates of 0.8% compared to 3.5%, respectively. In Study 2 Proximity took 6.5 min compared to 29 min for manual entry, with error rates of 1.4% each due to non- conformant data (text). Conclusions: Software data transfer interfaces can markedly shorten the time for data entry, reduce error rates and reduce operational costs.