Knowledge and attitudes in dementia held by general practitioners in the primary care setting of Botucatu, São Paulo, Brazil

ABSTRACT Background: Alzheimer's disease (AD) is the leading cause of dementia worldwide. Despite alarming evidence on dementia prevalence, the condition is still underdiagnosed by general practitioners (GPs) in primary care. Early detection of the disease is beneficial for patients and relatives, who should be provided comprehensive guidance on dealing with dementia complications, covering medical, family and social aspects, thereby providing an opportunity to plan for the future. Objective: The objective of this study was to assess the knowledge of and attitudes toward dementia held by GPs from a city in the interior of São Paulo State, Brazil. Methods: A non-randomized intervention study was conducted involving six lectures about dementia. Before and after the intervention, the participating physicians completed two quizzes about knowledge of and attitudes towards dementia. The study was carried out in the primary care services of the town and a total of 34 GPs participated in the study. Results: The mean age of the sample was 33.9 (±10.2) years and the majority (76.5%) of the sample had not undertaken medical residency training. The mean number of correct answers on the Knowledge Quiz about dementia before and after the training intervention was 59.6 and 71.2% (p<0.001), respectively. The comparison of the mean responses on the Attitude Quiz revealed no statistically significant difference between the two applications of the instrument, before and after intervention (p=0.059). Conclusions: More training for GPs on dementia should be provided.


INTRODUCTION
The aging of the population has led to a substantial rise in the number of older people with cognitive decline and dementia. By 2050, there will be an estimated 152 million persons with dementia globally 1 . Every year, another 10 million people will develop dementia, of which 6 million are from low-to middle-income countries, such as Brazil 1,2 . The estimated global cost of the condition in 2015 was 818 billion USD, a figure set to rise to 2 trillion USD by 2030 3,4 .
Dementia is a syndrome whose main risk factor is age. In most cases, the disease is progressive and marked by cognitive decline, which impacts the individual's independence and autonomy 5 . Alzheimer's disease (AD) accounts for 50-70% of all dementia cases 6 . Although AD dementia is a clinical condition with major personal, family and social impacts and no curative treatment, early detection is believed important because diagnosis allows patients and their carers to discuss future care plans, helping to reduce psychic distress and costs 7,8,9,10 .
In many countries, including Brazil, general practitioners (GPs) play a central role in the health of older people and especially in detecting individuals with dementia. However, the literature shows that dementia is underdiagnosed in primary care, delaying diagnosis and leading to unfavourable outcomes for patients and their relatives 11,12,13 . A systematic review showed that the main barrier to GPs establishing a diagnosis is lack of training and education on dementia. Moreover, lack of training leads to doubts and limited confidence among clinicians in confirming a dementia diagnosis and managing symptoms 14 . Other studies have shown that unclear guidelines on screening for cognitive decline, misunderstandings regarding drug therapy, and difficulties talking with patients and their relatives about the disease and its complications, can preclude early diagnosis of dementia in primary care 15,16,17 .
Since 1988, following the creation of the Brazilian National Health System (SUS), primary care has become the main entry point to the public health system. Currently, 73% of older people are treated solely by the SUS and most of them receive care provided by GPs 18,19 . Studies on dementia detection involving medical students and physicians on the first year of medical residency programs in Brazil have shown the need to provide more in-depth education and training in the detection and diagnosis of cognitive impairments in older people 20,21,22 . This lack of training may be due to the teaching method used by medical schools, which until recently, focused on training students about serious illnesses involving hospital inpatients, with little emphasis on outpatient treatment, i.e., primary care. However, recent curriculum models now engage students in primary care practice. Nevertheless, studies point to the need for continuous teaching and training of newly graduated physicians in the practice of medicine, given the rapid demographic and epidemiological shifts that have taken place in the last few years. This limited professional training in internal medicine has prompted the present study assessing the knowledge and attitudes in dementia held by GPs from a primary care service of a city in the interior of São Paulo State, Brazil.

Study design and setting
A quasi-experimental intervention study was conducted involving six lectures (60 minutes each) about dementia and/ or a presentation script. The lectures were delivered, and the scripts were written by a geriatrician with experience in dementia (VFSM). The lectures/presentation script covered the definition, epidemiology, diagnosis and management of dementia syndromes. Before and after the intervention, the participating physicians completed two quizzes about knowledge of and attitudes towards dementia 18 . The coordination of the health system of Botucatu held monthly meetings from 2 pm to 5 pm with GPs from the primary care service to deliver updates on health topics to these professionals. The primary care service of the city of Botucatu comprises 4 basic health units (UBS), 2 health centres (CS), 2 polyclinics and 12 family health units (USF). All 40 GPs of the city of Botucatu primary health service were invited to take part in the study, most of whom were newly graduated physicians.

Data collection
Data collection took place between August 2016 and January 2017, gathering sociodemographic data (name, age, gender), responses to questions about time since graduation from medical school, completion of medical residency and specialty, and also about previous training in dementia during medical school: ("How long have you been a practicing physician?", "Did you have any classes on dementia during your medicine course?", "Have you done a Medical Residency program?"; If so, in which specialty?", "Would you like to have refresher courses on dementia?").

Participants
All GPs of the primary health service of Botucatu (n=40) commenced the study. Six physicians were excluded because they began classes but failed to complete the content and thus did not take part in the 2 nd application of instruments. Physicians wishing to take part in the study, but who could not be away from their posts at health units, were provided with a script containing the topics presented in the lectures. After completing the 2 quizzes in the presence of the researcher, 10 physicians received the intervention based on this script. Thirty days later, the same group completed the quizzes a second time.

Instruments
The instrument Knowledge and Attitudes Quiz about Dementia was developed in the United Kingdom in 2004 with the purpose of measuring GPs' knowledge of, confidence with and attitudes to the diagnosis and management of dementia in primary care 16 . In 2015, these instruments were translated and culturally adapted for use in the Brazilian setting 20 .The Knowledge Quiz contains 14 multiple-choice questions, each with five possible answers, only one of which is correct, with all questions including the response option "I don´t know". The instrument is divided into 3 sub-items addressing aspects of dementia related to epidemiology (3 questions), diagnosis (8 questions) and management (3 questions).
The Attitude Quiz contains 10 sentences about physicians' thoughts on the management of patients with dementia, scored on a Likert-type scale with the following possible answers: "strongly agree", "agree", "neither agree nor disagree", "disagree" and "strongly disagree" 16 16 .

Statistical analysis
The data were analysed using the IBM-SPSS 21 statistical software. Numerical variables were expressed as mean and standard deviation, while categorical variables were expressed as frequency and percentage. Student's t-test was used to compare means. The level of significance adopted was 0.05.

Ethics committee
This study and the free and informed consent form were analysed and approved by the Research Ethics Committee of the Clinical Hospital of the Botucatu School of Medicine-UNESP.

RESULTS
The mean age of the sample was 33.9 (±10.2) years, where the majority (18 subjects -52.9%) were 20-30 years old. Race consisted of thirty-two self-declared white, one black and one yellow. Sixteen (47%) of the physicians were female and 21 (61.8%) had been practicing for up to 5 years. Six (17.6%) physicians aged 47-59 years had been practicing for 18 years or longer. Of the 8 (23.5%) GPs who had undertaken medical residency training in the overall sample, 5 were >47 years old, 1 was aged 28 years and 2 aged 37-39 years. The majority of the sample had not undertaken medical residency training (76.5%). With regard to the question on having received classes on dementia during medical training, 29 (85.3%) stated "yes". Of the participants who reported "no", 3 had been practicing for 1-5 years and 1 for 18 years. All of the physicians (100%) stated that they wished to have a refresher course on dementia.

Knowledge about dementia
The Knowledge Quiz about dementia with response options and number of correct answers, before and after the training intervention of presentation class and/or script is given in Table 1. The mean number of correct answers before Table 1. Knowledge Quiz about Dementia with response options and number of correct answers, before and after training intervention, of general practitioners from the primary care system of Botucatu city, São Paulo state.

Questions
Response options Before After and after the training intervention was 8.35 (59.6%) and 9.97 (71.2%), respectively. The percentage of correct answers on the epidemiology, diagnosis and management subscales before intervention were 44.1, 68.3, and 50.9% versus 57.7, 79.7, and 61.7% after intervention, respectively. Results on Student's paired t-test for comparisons of mean correct answers on the quizzes before and after the training showed a statistically significant difference only on the diagnosis subscale (p<0.001) ( Table 2).

Attitudes to dementia
The comparison of the mean responses on the Attitude Quiz is presented in Table 3, revealing no statistically significant difference between the two applications of the instrument, before and after intervention (p=0.059).

DISCUSSION
This study employed two quizzes assessing the knowledge and attitudes of physicians about dementia. The mean total correct answers on knowledge before and after the intervention was 8.35 (59.6%) and 9.97 (71.2%) out of 14 points, respectively. Turner et al. 16 , used the same instruments in a single application to 127 GPs who had volunteered to join a randomized clinical trial of educational interventions to improve the detection and management of dementia. The GPs scored 67% overall on the quiz about knowledge of dementia and 48, 74 and 73% on the epidemiology, diagnosis and management subscales, respectively. In the present study, the GPs exhibited a lower performance, scoring 59.6 versus 67% by British GPs. Brazilian scores on the epidemiology, diagnosis and management subscales also proved to be lower, at 44.1, 68.3 and 50.9%, respectively. The most common mistakes in the UK study involved questions on epidemiology, whereas Brazilian GPs erred most on epidemiology and management of dementia. However, both groups of GPs performed best on questions about diagnosis in dementia.
A previous study by Jacinto et al. of 152 medical residents of a federal university hospital in São Paulo state applied the same two quizzes used in the present study 21 . The sample comprised 40.8% (n=61) residents in neurology, psychiatry, and geriatrics and internal medicine, and therefore probably had contact with older people with dementia during training. The remaining residents (59.2%) specialized in surgery and clinical subspecialties. Around 59.7% of residents embarking on their programs reported having received good training on cognitive impairments during medical school. Another study where 155 final-year medical students from two different universities participated, 92 (59.7%) considered that they had good training in cognitive alterations during their undergraduate medical course and 67 (58.8%) declared having only theoretical training 22 . However, this group of students scored an average of 6.9 (49.2%) on the Knowledge Quiz, pointing to the need for continued education, given that the workforce in the public health system in Brazil, akin to the UK, comprises very young doctors dealing with the population of older people users seeking public health services 23 .
The study by Downs et al. 24 , assessing the effectiveness of educational interventions in improving detection and management of dementia, concluded that GPs have difficulty detecting and managing dementia symptoms. Two interventions showed positive results: a decision-support system built into the electronic medical records and practice-based workshops. There were improvements in detection and confidence of GPs in the diagnosis and management of dementia. A similar result was seen in the present study, where primary care GPs showed improved performance, post-training, on the epidemiology diagnosis and management subscales. These results confirm that continued education may be effective for improving detection of cognitive impairment and dementia in older adults in primary care.
Several barriers, for both doctors and patients, to early diagnosis of dementia in primary care were outlined in the European Carers' Report 2018 25 : GPs believed forgetfulness was part of normal aging and so gave little importance in diagnosis; and also due to the fact that no curative treatment exists, regarding patients and/or relatives, delay in seeking help at  the early stage of the disease is explained by fears of prejudice among friends, at work and also amongst health professionals 26 , where clear information is lacking on all aspects of dementia for physicians and for individuals who are aging 26,27 .
Regarding attitudes towards dementia held by GPs, the present study showed that most of them agreed much can be done to improve the quality of life of people with dementia (82.3%) and of their carers (97.1%), but 50% believed dementia is best diagnosed by a specialist. Similar results were found in the study by Turner et al. 16 . These findings are alarming because individuals are unable to get an appointment to see a specialist following the onset of symptoms because neither patient nor relatives took the cognitive and functional impairments seriously, believing them to be part of normal aging. Another important finding of the study was that 20.6% strongly agreed with the statement that treating individuals with dementia is frustrating.
In the study by Ahmad et al., 28 assessing the attitudes, awareness and practice regarding early diagnosis of dementia of 1011 GPs, they found that older physicians were more confident diagnosing and advising about dementia. However, although these physicians believed early diagnosis of dementia was beneficial, they were more likely to feel that patients with dementia can be a drain on resources with little positive outcome. By contrast, younger physicians were more positive and felt that much can be done to improve the quality of life of patients with dementia 27 .
The World Alzheimer Report 2017-2025 warns of the global impact of AD dementia in the near future and emphasizes the urgent need for awareness of this problem by society, public authorities and healthcare professionals 29 .
Banerjee et al. 30 showed that programs that encourage the individual with dementia to share their experiences with students result in the deepening of knowledge about the cognitive decline by the student, favouring a change in attitude to deal with the manifestations of dementia.
The study by Alzheimer's Disease International (ADI) "World Alzheimer Report 2019 -Attitudes to dementia" 31 showed that in Brazil, there is still a huge stigma and negative attitudes towards people with dementia by health professionals and people in the community. The authors believe that changes in attitudes to eliminate stigma about dementia should have the direct participation of individuals with dementia. In addition, the inclusion of students in the health field from the beginning of graduation so that they share the early stages of the disease can be the path to ethical management in the management of dementia.
The limitations of this study included the convenience sample and small sample size, all derived from a single location in Sao Paulo State. Future research should explore whether the same findings can be replicated in a larger, representative sample of GPs in Brazil.