Knowledge and attitudes about dementia of primary care physicians in Southern Brazil

Abstract

Background

Primary Care Physicians have a central role in assisting individuals with dementia and evaluating their preparedness to care these patients is fundamental. Our aim is to evaluate the knowledge and attitudes regarding dementia of the Primary Care Physicians (PCP) in Rio Grande do Sul (RS) state, Southern Brazil.

Methods

We collected sociodemographic data, volume of patients with dementia treated/referred and perception of difficulties in caring for these patients. A previously validated questionnaire was sent: "Quiz on Knowledge and Attitudes in Dementia".

Results

From March/2022 to June/2023, 296 PCP responded to the questionnaire. They were mostly women (52.7%, 156), with a median [IQR] age of 35 [29–44] years, mostly were White (82.1%, 243) and had 7 (4–16) years of experience as a physician. Less than half of the physicians performed cognitive screening (43.9%) and Mini Mental State Examination was the most screening (63.5%) test used. The mean percentage of correct answers in the Knowledge Quiz was 46.4%. In the attitude quiz, we identified 3 factors: 1) frankly positive attitudes; 2) perceive primary care as important but have a pessimistic attitude towards them; 3) see primary care as important for patient care.

Conclusion

Knowledge about dementia is low among PCP in RS; however, most have positive attitudes towards these patients or think primary care is important to these patient's care.

Background

The increase in life expectancy in Brazil and globally seen in recent decades has brought about the urgent need to know the most prevalent diseases among the elderly. Although dementia syndromes are common in this age group and the risk of developing dementia for a person over 60 years of age is around 6% [1], early diagnosis faces several barriers [2].

The percentage of undetected dementia is more than 50% in Asia, North America and Europe [3]. In Brazil, a study estimated the frequency of undiagnosed cases in a medium-sized city at 77% [4]. The role of primary care physicians (PCP) in the early diagnosis of dementia is fundamental, especially in low- and middle-income countries, where the access to specialized health services is still restricted [5]. PCPs face several difficulties when caring for patients with dementia. Lack of time for a detailed evaluation, need for updating in different areas of medical knowledge and stigmas related to the prognosis of patients with dementia are some of the factors that can hinder both early dementia diagnosis and care of already diagnosed patients [6]. Therefore, assessing knowledge regarding the diagnostic and therapeutic process of dementia in primary care physicians is important to calibrate continuing education activities in each region [7].

As a way of contributing to future educational policies in the context of primary health care in our region, the objective of this study is to evaluate the knowledge and attitudes of primary care physicians in relation to dementia in the state of Rio Grande do Sul, Southern Brazil.

Methods

We performed a cross-sectional study with primary care physicians (PCP) from Rio Grande do Sul (RS) state, Brazil. The state of RS is the southernmost state in the country and one of the largest states, with a current population of almost 11 million people [8]. RS is one of the states with highest development index and per capita income in the country; in addition to being the state with highest percentage of self-identified White population [8]. Primary health care in the state of RS is organized through "Estratégia Saúde da Família" (Family Health Strategy), formed by around 2500 basic health units [9]. This study was approved by the Institutional Review Board of the Hospital de Clínicas de Porto Alegre (Office for Human Research Protections: 00000921) and all participants agreed to participate in the study through an electronically signed informed consent form.

This project was performed entirely electronically, by sending emails to PCP to invite them to participate in the study. As we did not have direct access to the physicians' emails, we have sent the message to the health departments in all regions of the state, and we asked them to resend it to the physicians' personal email addresses. For each health department, we emailed the message about 3–4 times, with 3-month intervals between each email sending. This email contained basic information about the project and a link to an electronic form. In this form, in addition to the informed consent form, there were questions about socio demographic characteristics, volume of patients with dementia treated and perception of difficulties in these services. In addition, a previously validated questionnaire was sent: "Quiz on Knowledge and Attitudes in Dementia” [10, 11]. The Knowledge Quiz consists of 14 multiple choice questions, including the answer "I don't know" and with only one correct answer. This questionnaire covers concepts about the epidemiology (3 questions), diagnosis (8 questions) and treatment (3 questions) of dementia. The Attitudes Quiz contains 10 sentences about the physician's perception regarding the management of patients with dementia. Responses are in Likert format, from "strongly agree" to "strongly disagree". The questions of this questionnaire address several aspects that may arise when caring for patients with dementia, such as feelings of frustration, reward and satisfaction. All PCP were asked to rate their own knowledge of dementia on a scale of 0 to 10. Furthermore, we asked PCPs for possible suggestions to improve the care of patients with dementia in primary care. We offered some suggestions and left space for free comments.

Categorical variables were described by frequency and continuous variables by mean and standard deviation or median and interquartile range (IQR), according to their distributions. The evaluation of the normality of the distribution was carried out by inspecting the histogram and the Shapiro–Wilk test. The association between categorical variables was assessed using the chi-square test and the association between continuous variables with normal distribution was performed using the Student's t test and the Wilcoxon-Mann–Whitney test for those with a distribution other than normal. We performed a principal component analysis in the scores obtained from the Attitude Quiz to determine groups of "positive" and "negative" attitudes towards the care of people with dementia. For this analysis, only principal components with eigenvalues > 1 was retained. For the interpretation of each component, only variables with rotation factor > 0.4 was included. The level used in significance tests was 5%. All analyzes were performed in R, version 4.2.2 [12].

Results

Between March 2022 and June 2023, we received responses from 296 PCPs, mostly (52.7%, 156) women, with a median [IQR] age of 35 [29–44] years and mostly were White (82.1%, 243). The PCPs had 6.5 [3-16] years of experience as physician and mostly had a Family Medicine specialty, 41.6% (123) or no specialty, 40.2% (119). The sample characteristics are presented in Table 1. Considering that there is at least one PCP in each of the 2500 basic health units (that is, 2500 PCPs throughout the state), the response rate was 11.8%. These responses came from 175/497 (35.2%) municipalities in the RS state. A choropleth map of the municipalities with at least one response show that visually there was no geographic concentration of responses (Fig. 1). Supplementary Table 1 shows the number of responses from each municipality.

Table 1 Characteristics of the study participants

Fig. 1

Choropleth map of the municipalities with at least one response

Most physicians assist fewer than 5 patients with dementia per month and most of them refer < 5 patients to tertiary centers. Most of them do not perform cognitive screening (44.3%, 131) and the Mini Mental State Examination was the most used test among those who routinely do it. In the Knowledge Quiz, the percentage of correct answers was 42.9%, with the treatment section having the higher percentage of correct answers (66.7%), followed by epidemiology (50%) and diagnosis (42.9%). The median (IQR) percentage of correct answers was higher in those PCPs who routinely performed screening tests (50 [42.9,57.1] vs 42.9 [35.7,50]) (p = 0.001). Sex and years of experience as a physician were not associated with Knowledge Quiz.

The principal components (PC) analysis of the Attitude Quiz questionnaire showed three main vectors with eigenvalues > 1, explaining 54% of the variation. The first two (PC1 and PC2) indicated a more positive attitude towards the patients with dementia and the third (PC3), a more negative attitude (Table 2 and Fig. 2). There were mild negative correlations between the two optimistic factors and the Knowledge Quiz total score (Rho = -0.138 and -0.161, respectively). No such correlation was found with PC3. The time as a physician did not correlate with any of the vectors.

Table 2 Composition of the optimistic and pessimistic factors structure

Fig. 2

Biplot of the optimistic principal factors structure from the attitude quiz

Among the possibilities of suggestions to improve care for patients with dementia, the majority of responses were "Existence of a continued training platform" (66.6%) and "Longer consultation time for patients with cognitive complaints" (59.1%) (Supplementary Table 3).

Discussion

We conducted a cross-sectional assessment of a sample of primary care physicians from the southernmost state of Brazil on their knowledge and attitudes about dementia. In this evaluation, we found that the percentage of correct answers in the Knowledge Quiz was less than 50% in our sample. Additionally, we found that in the principal components analysis of the Attitudes Quiz, most of the variation in responses pointed to factors that we interpreted as optimistic. Taken together, these results point to possibly deficient training in ​​cognition and behavior of these primary care physicians, but at the same time, an open attitude towards these patients.

Because the intrinsic difficulties of conducting a study exclusively via email, we are not sure if our sample is representative of PCP in our state (Rio Grande do Sul, RS), as we do not have information about the demographic characteristics of this population. Although there is no exact number of PCPs in RS, we estimate that we included about 11% of this population and the spatial distribution of the municipalities where that was at least one response did show such large concentrations. Previous studies showed a heterogeneous percentage of responses. In the study by Stewart et al. 2014, a questionnaire was sent by mail to PCPs in the United States and Canada and the response percentage was higher than ours, 60% [13]. More recently, Wangler and Jansky obtained only 22% responses to a mailed questionnaire to German primary care physicians [14]. It is possible that our response rate was lower because we did not send the questionnaire directly to each physician's personal email, but rather to the local health department, which then forwarded the email. Therefore, this process may not have reached all physicians.

The results regarding the knowledge quiz show that most participants got less than 50% correct in the questionnaire. Although low, this finding is very similar to what was found by Jacinto et al. 2018 in another Brazilian city, using the same questionnaire (~ 50%) [15]; but lower than what was found by Turner et al. in English PCPs, in the original questionnaire validation article (67%) [11]. Although the percentage of correct answers is not associated with the length of experience as a physician, we found a higher percentage of answers among physicians who routinely used screening tests for cognitive impairment. Although we did not find a similar association in the literature, this finding is quite intentional, as physicians who perform screening probably spend more time with their patients and seek more information in the literature.

Regarding the attitudes quiz, in the principal components analysis, we found two main factors that pointed to a positive attitude of PCPs towards patients with dementia. Furthermore, we found a third factor, which explained a smaller percentage of the variation in responses, which indicated a more pessimistic attitude. Although the total variation explained by these 3 factors is 54%, the findings indicate a tendency towards a more optimistic attitude towards dementia. These findings are similar to those found by Jacinto et al. 2020; In this study, the authors found two main factors, which they called "heartsink" and "heartfelt" factors [15]. Similar finding seen by Turner et al., in the original English study of the attitudes quiz [11]. Although these results point to a favorable response to a training program for PCPs [16], it suggests that there are still barriers to training these physicians, many probably related to stigma regarding dementia and possibly a feeling of futility in treating these patients.

The primary care physicians we evaluated suggested several actions that, in their view, could improve care for patients with dementia. The frequency of responses points to an interesting result. The majority suggested the existence of an ongoing training platform and more time to care for patients. This suggests that the majority of PCPs in RS are willing to qualify and appear motivated to care for patients with dementia. This finding is similar to that found by Stewart et al. 2014 in a study of 1500 North American PCPs. In this study, 78% of PCPs indicated that lack of time was a barrier to caring for patients with dementia [13].

Our study has some strengths, such as covering most regions in the state of RS. In addition, this kind of assessment has never been performed in our state, and in fact, in a country as large as Brazil, few studies exist on the knowledge and attitudes of PCPs in relation to dementia. Therefore, it seems to us that we were able to contribute with some data that may help to construct public policies for dementia in our country. However, some weaknesses deserve mention. As we did not have the physician's personal emails, we depended on the health departments in each region to forward the questionnaire to the physicians. Therefore, the response rate was low, 11%. Furthermore, as the questionnaire was exclusively answered electronically, we cannot rule out that the most motivated physicians were those who answered the questionnaires. In any case, the attitude quiz profile suggests that it was not just those with more positive attitudes who responded to the questionnaire.

Conclusion

Despite knowledge about dementia was low among primary care physicians in RS state, most have positive attitudes towards these patients or think primary care is important to these patient's care. Long-term educational strategies could improve the capacity of primary care physicians of assist patients.