Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

How are people with mild cognitive impairment or subjective memory complaints managed in primary care? A systematic review

How are people with mild cognitive impairment or subjective memory complaints managed in primary... Background: Primary care is typically the first point of contact in the health care system for people raising concerns about their memory. However, there is still a lack of high-quality evidence and understanding about how primary care professionals (PCPs) currently manage people at higher risk of developing dementia. Objectives: To systematically review management strategies provided by PCPs to reduce cognitive decline in people with mild cognitive impairment and subjective memory complaints. Method: A systematic search for studies was conducted in December 2019 across five databases (EMBASE, Medline, PsycInfo, CINAHL and Web of Science). Methodological quality of included studies was independently assessed by two authors using the Mixed Methods Appraisal Tool. Results: An initial 11  719 were found, 7250 were screened and 9 studies were included in the review. Most studies were self-reported behaviour surveys. For non-pharmacological strategies, the most frequent advice PCPs provided was to increase physical activity, cognitive stimulation, diet and social stimulation. For pharmacological strategies, PCPs would most frequently not prescribe any treatment. If PCPs did prescribe, the most frequent prescriptions targeted vascular risk factors to reduce the risk of further cognitive decline. Conclusion: PCPs reported that they are much more likely to provide non-pharmacological strategies than pharmacological strategies in line with guidelines on preventing the onset of dementia. However, the quality of evidence within the included studies is low and relies on subjective self-reported behaviours. Observational research is needed to provide an accurate reflection of how people with memory problems are managed in primary care. Lay summary People will typically go to their general practitioners, also known as primary care professionals (PCPs), to raise concerns about their memory. However, there is no clear understanding of what advice or treatment PCPs provide to people with memory concerns who are at high risk of dementia. This review aims to summarize the findings from research that studied what advice or treatments PCPs would give to a person with memory concerns. Nine studies were included in the review after screening through 11 719 studies. The current review found that PCPs were more likely to provide advice rather than prescribe any drug treatment. The most common advice that © The Author(s) 2021. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 670 Family Practice, 2021, Vol. 38, No. 5 Key Messages • Review of primary care professionals’ (PCPs) management of memory concerns. • The review included a wide range of quantitative and qualitative study designs. • Most frequent advice was to increase physical activity. • Most common drug response was to not prescribe any treatment. • Majority of PCPs reported strategies that followed prevention guidelines. • Future research needs more observational studies to observe real-life practice. PCPs provided was to increase physical activity, cognitive stimulation and social stimulation. If PCPs decided to prescribe drugs, the most common prescriptions were to improve blood flow. Improving blood flow has been linked with reducing the risk of developing dementia. However, the quality of the studies included in this review is low because many relied on PCPs answering questionnaires on their intentions to manage people with memory concerns. Therefore, future research needs to observe PCPs’ real-life practice to provide an accurate reflection of how people with memory problems are managed in primary care. Key words: Cognitive dysfunction, dementia, memory, primary health care, primary prevention, systematic review protective factors for dementia. Alcohol misuse (21) and dementia Introduction has a complex J-shaped relationship with excessive alcohol use and Background non-consumption being associated with greater risk than moderate An estimated 50 million people are expected to be living with de- consumption. However, this research addressed all risk factors indi- mentia worldwide, with this projected to rise to 152 million in the vidually rather than the effectiveness of a behavioural health inter- next 30  years (1). Dementia is the seventh leading cause of death vention that combines strategies for multiple risk factors. Evidence across the world (2) and the leading cause of death within England from trials of time-intensive behavioural health interventions and Wales (3). Dementia is the only condition within the top 10 targeting the lifestyle risk factors aiming to reduce cognitive decline causes of death without a treatment to slow or cure its progression and onset of dementia in people with memory concerns is mixed (3). However, it is believed that up to 40% of dementia cases could (5,22). Further investigations of lifestyle interventions, such as Active be prevented if the following risk factors were addressed: low level Prevention in People at risk of dementia through Lifestyle, bEhav- of education, hearing loss, traumatic brain injury, hypertension, al- iour change and Technology to build REsiliEnce (APPLE-Tree) (23) cohol misuse, obesity, smoking, depression, physical inactivity, social and the Systematic Multi-domain Alzheimer’s Risk Reduction Trial isolation, air pollution and diabetes (4). (SMARTT) (24) are ongoing. SMARRT will recruit older adults People defined as high risk of developing dementia have been with subjective cognitive complaints from primary care and be ran- operationalized in various ways. For example, the FINGER trial (5) domly assigned to the intervention or a health education control. used the CAIDE dementia risk score, whilst other studies may use The intervention will be to develop a personalized plan for risk fac- the Framingham vascular risk scores (6). However, the one indicator tors hypertension, hyperglycaemia, depressive symptoms, poor sleep, that often leads to consultation due to concerns about the risk of polypharmacy, physical inactivity, low cognitive stimulation, social developing dementia is memory concerns (7). The term ‘memory isolation, poor diet and smoking. All of these factors are associated concerns’ refers to people with subjective memory complaint (SMC) with an increased risk of dementia and strategies addressing these and mild cognitive impairment (MCI). SMC is defined as a form issues provide the most likely approach to delay the onset of de- of complaint that an individual makes regarding his or her cogni- mentia. However, the efficacy of dementia prevention interventions tion, but no clear impairment is found by objective psychometric in delaying incident dementia is still mixed and inconclusive (5,22). testing (8). In contrast, people with MCI do show a noticeable de- Therefore, there are no current specific treatment recommenda- cline in cognition using objective testing, which is not severe enough tions provided by the national health governing bodies for people to interfere with daily activities and be defined as dementia (9). SMC with memory problems (SMC and MCI) due to the lack of strong affects half of people over 65 years old (10) and MCI affects 20% of current evidence (25–27). Consequently, the current guidelines for people over 65 (11). Reviews have indicated that people with SMC health professionals to delay the onset of dementia is to provide gen- are twice as likely to develop dementia as individuals without SMC eric non-pharmacological recommendations to all people in mid-life (12), highlighting the need for health care professionals to effectively (25). This includes encouraging healthy behaviours, such as smoking manage people with SMC and MCI in order to reduce the risk of cessation, increasing physical activity and reducing alcohol con- developing dementia. sumption (25). There is low-to-moderate quality evidence that addressing hyper- Primary care is typically the first point of contact in the health tension (13), diabetes (14), physical activity (15), tobacco cessation care system for people raising concerns about their memory (28). (16), cognitive stimulation (17) and social isolation (17) has been Therefore, primary care is critically placed to play a greater role demonstrated to reduce dementia risk in low-to-moderate quality in providing preventive treatments to delay the onset of dementia evidence. Treatment addressing hearing loss (18), obesity (19) and in adults with memory problems (28). Despite this, dementia pre- depression (20) requires further research and has yet to demonstrate vention advice or even recognition of cognitive impairment by Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 671 general practitioners (GPs) is variable, often with failure to respond BH and JR completed 100% of the full-text screening independently to memory loss symptoms (29). Godbee et  al. (30) have recently with any discrepancies resolved by a third independent reviewer. published a preliminary conceptual model on how to implement From the studies included in the systematic review, a pre-piloted dementia risk reduction practice in primary care, providing five data collection form was used by BH and JR to extract the necessary implementation strategies, which were (i) identifying ‘champions’ data. Extracted data included: author (year), study design, setting, to promote brain health to patients, (ii) conducting educational professionals, service users, key findings/themes, type of pharmaco- meetings, (iii) conducting local consensus discussions, (iv) altering logical recommendations and type of non-pharmacological recom- incentive structure and (v) capturing and sharing local knowledge. mendations. Study authors were contacted for any missing data or However, there is still a lack of high-quality evidence and under- any additional data that might be deemed relevant to the review. standing about how primary care professionals (PCPs) currently A  narrative analysis of studies was conducted using a data-driven manage people at higher risk of developing dementia. Therefore, this integrated synthesis approach. Quantitative and qualitative studies systematic review will investigate what management strategies are were synthesized applying a transformation process known as offered by PCPs in response to managing cognitive decline and risk quantitizing. Quantitizing is a method validated for mixed-method of dementia in people with MCI or SMC. The review will aim to reviews whereby qualitative data are quantified. (33) bridge the gap within the literature by exploring both pharmaco- logical and non-pharmacological strategies recommended to people Quality assessment with MCI or SMC in a primary care setting. Two authors independently assessed the methodological quality of each study using the mixed-methods appraisal tool (MMAT) (34). The use of MMAT in mixed-method reviews has been validated, Methods which then allows quality appraisal for the variety of study de- This review was performed in accordance with the PRISMA guide- signs to be completed using one tool (35,36). Therefore, the MMAT lines (31) and the protocol was registered with Prospero (ID: was chosen to appraise both qualitative and quantitative study de- CRD42020170804). signs included in the current review. Similar to data extraction, the interrater reliability was deemed acceptable with Kappa equal or Search strategy above 0.8, and any disagreements were discussed with a third inde- pendent reviewer. The systematic review was conducted on 11 December 2019 using five online bibliographic databases (EMBASE, Medline, PsycInfo, CINAHL and Web of Science). See Supplementary Figures 1–5 for Results full search terms used. No limits were set for time or language and authors were contacted to acquire missing or further information Study selection if needed. Forward selection and reference lists from the final in- The search yielded 11 719 papers. After de-duplication and the add- cluded papers were manually searched to identify potentially rele- ition of one extra paper identified through other sources, 7250 title vant studies that may not have been captured in the literature search. and abstracts were screened. A second independent reviewer screened 10% (n = 725) of the title and abstracts with a high interrater re- Inclusion and exclusion liability (a  =  0.89). Of 275 full-text papers retrieved, 9 were in- cluded in the final systematic review with high interrater agreement To be included, studies were required to assess pharmacological or (a = 0.85). Figure 1 summarizes the study selection process (31). non-pharmacological management options provided by any profes- sional (GPs, practice nurses, pharmacists, etc.) in a primary care set- ting to people over 50  years old with MCI or cognitive complaint Characteristics and quality of included studies without dementia. The threshold of 50 years old was selected as ac- We included seven quantitative studies: one descriptive naturalistic quired memory concerns are increasing and starting to be treated study (37), one structured interview (38) and five cross-sectional more seriously (32). The study could be quantitative or qualita- surveys (39–43) of PCPs’ self-reported management strategies. tive. Non-English language papers were accepted during initial Additionally, two qualitative studies were included, one study using screening. However, non-English papers were excluded during full- semi-structured interviews (44) and one case report (45). The in- text screening if an English version was not be obtained. Exclusion cluded studies are set across seven countries (Canada, Germany, criteria included only people with a confirmed diagnosis of dementia Israel, Malaysia, Spain, UK and USA), with four studies including or healthy older adults. Intervention-based studies were excluded in data from the USA. A total of 2756 primary care physicians partici- order to capture real-life management practices. Additionally, inter- pated across eight of the included studies, with Argimon-Pallas et al. ventional studies, reviews, book chapters and dissertations were also (37) reporting the number of primary care practices participating excluded. Finally, if the study focussed on diagnosis or screening ra- rather than the number of physicians. Six of the studies focussed on ther than treatment or management, it was also excluded. the management of people with MCI (37–40,44,45). Three studies focussed on SMC and memory concerns (41–43). Data extraction The methodological quality of the study designs included was Two reviewers were responsible for the screening process. The second of low-to-moderate quality overall. Aspects of methodology and reviewer (JR) completed a random 10% of the initial screening that analysis for several of the studies were unclear. None of the studies was blinded to the first reviewer (BH). If interrater reliability was included healthy control groups to allow comparisons between below 0.80 for Kappa, then another 10% of the papers would be managements strategies of PCPs for both cognitively healthy older screened by JR. However, if Kappa was above 0.80, then this would adults versus people with memory problems. Argimon-Pallas et  al. be deemed satisfactory and reviewers would progress to full-text (37) was the only study using comparison groups, comparing treat- screening. If either reviewer considered a paper potentially relevant, ments received for groups with memory problems against group it was retrieved and included for the full-text screening process. Both with confirmed diagnosis of dementia. Another concern for each Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 672 Family Practice, 2021, Vol. 38, No. 5 Figure 1. PRISMA flowchart describing the process of study selection. of the survey-based designs was the lack of clarity on accounting memory problems (SMC or MCI). Three of the five studies were for the potential bias in response rates and investigating any dif- survey based, one was a case report and one was semi-structured ference in characteristics between responders and non-responders interviews and a focus group. of the survey. The quality appraisal for all studies can be found in Supplementary Table 1 (a = 0.80). Subjective memory concern Two studies investigated primary care physician’s non- Non-pharmacological management pharmacological management intentions in response to a patient Two thousand one hundred and sixty-nine primary care phys- presenting with SMC (41,42). Both studies used the DocStyles icians were recruited across five studies that investigated non- survey measure. DocStyles is a web-based survey with a range of pharmacological management for people presenting with either questions, including how to reduce cognitive decline in people Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 673 Table 1. Study characteristics of studies included in systematic review Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Quantitative Physician-reported management strategies Banjo et al. Survey USA, 212 primary care 1–10 years N/A Case vi- Yes: Likert scale (1–7 with 7 Modafinil N/A (43) Canada physicians (45%) gnette—65 year three types of being highest comfort) of Methylphenidate 25–40 years old 11–20 old with SMC neurotropics how comfortable physician Sildenafil (36%) (24%) (cognitive would feel prescribing cog- (all drugs fit the 41–59 years old 20+ years enhancers) nitive enhancers: criteria for noo- (45%) (31%) M = 4.8 (SD N/A) tropics, otherwise 60+ years old known as cogni- (19%) tive enhancers) 55% males Medication (not Physical activity Day et al. Survey USA 493 primary care 3–19 years N/A Memory Yes: % of physicians that would specified) Social activity (41) physicians (73%) concerns (not 10 options provide advice to patient Avoid Diet 479 internist 20+ years specified) (6 non- on: polypharmacy Cognitive stimu- <50 years old (27%) pharmacological; Physical activity n = 892 Nutritional sup- lation (69%) 3 pharmaco- (91.8%) plements Limit alcohol 50+ years old logical; 1 no Intellectual stimulation (not specified) Weight/BMI (31%) treatment option) n = 829 (85.3%) Healthy diet n = 809 77% males (83.2%) Socially activity n = 775 (79.7%) Limiting alcohol n = 626 (64.4%) Maintaining a healthy weight n = 511 (52.6%) Avoiding polypharmacy n = 434 (44.7%) Taking nutritional supple- ments n = 332 (34.2%) Taking certain new medica- tions (not specified) n = 164 (16.9%) Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 674 Family Practice, 2021, Vol. 38, No. 5 Table 1. Continued Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Friedman Survey USA 1,000 family phys- 14.7 years N/A Memory Yes: % of physicians that would Medication (not Physical activity et al. (42) icians and internist (SD = n/a) concerns (not 10 options provide advice to patient specified) Social activity 72% males 16.4 years specified) (6 non- on: Avoid Diet 250 nurse practi- (SD = n/a) pharmacological; No recommendations polypharmacy Cognitive stimu- tioners 3 pharmaco- n = 40 (4%) Nutritional sup- lation 13% males logical; 1 no Medication (not specified) plements Limit alcohol treatment option) n = 116 (11.6%) (not specified) Weight/BMI Take vitamins n = 293 (29.3%) Avoid polypharmacy n = 411 (41.1%) Healthy weight n = 457 (45.7%) Limit alcohol n = 591 (59.1%) Healthy diet n = 609 (60.9%) Socially active n = 667 (66.7%) Intellectual stimulation n = 802 (80.2%) Physically active n = 861 (86.1%) Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 675 Table 1. Continued Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Maeck Survey Ger- 159 family N/A N/A Case vignette— No: Survey asks if family phys- Ginkgo biloba N/A—not dis- et al. (38) (structured many physicians MCI Open-ended ques- icians would prescribe any (natural rem- cussed interview) (year = 1993) tions but would dementia related medi- edies) 70% males categorize an- cation to case vignette of Pentoxiphylline 122 fam- swers to facilitate someone with MCI who is (vascular man- ily physicians analysis at high risk of dementia: agement) (year = 2001)  = 1993 results Piracetam 56% males = 2001 results (nootropic) Yes (any treatment) Nimodipine (70.4%) (vascular man- (43.4%) agement) Ginkgo Biloba (34.0%) Memantine (23.0%) (anti-dementia Pentoxiphylline (13.2%) drug) (2.5%) Cholinesterase Piracetam (39%) inhibitors (3.3%) (anti-dementia Nimodipine (22.0%) drug) N/A Medication (not Memantine* N/A specified) (12.3%) Cholinesterase inhibitors N/A (8.2%) Other medication (44.7%) (0.8%) Suribhatla Survey UK 65 GPs N/A N/A Vascular cogni- Yes: Patient with vascular cogni- Statins N/A—not dis- et al. (39) % of sex not re- tive impairment Only discussed tive impairment (vascular man- cussed ported (VCI) prescription of 26% of GPs (16 out of agement) statins and no 61) would prescribe statins other strategies. to help manage vascular and cognitive risks Patients at risk of VCI 42% of GPs (27/64) felt that statins have a role in preventing VCI in at risk people Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 676 Family Practice, 2021, Vol. 38, No. 5 Table 1. Continued Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Werner Survey Israel 197 family phys- 21.9 years N/A MCI Yes % of family physicians Pharmacological Physical activity et al. (40) icians (SD = 10.4) 11 pharmaco- (n = 168) preferences for (not specified) Social activity 50.1 years old logical and non- treatment of MCI: Natural medi- Diet (SD = 9.2) pharmacological Physical activities (88%) cation (not spe- Cognitive 49% Male therapies Social activities (88%) cified) training Cognitive training (88%) Vitamins Relaxation/ Engagement in support (not specified) meditation group (80%) therapy Relaxation exercise (47%) Change in diet (44%) Psychotherapy (35%) Yoga or meditation (34%) Vitamins (34%) Pharmacological treatment (13%) Natural Medications (10%) Physician-observed management strategies Argimon- Descriptive Spain 105 general prac- N/A 921 pa- 45 diagnosed No preset list of % of service users receiving Nootropics (not N/A Pallas et al. naturalistic tices tients re- with MCI strategies treatment in response to specified) (37) study N/A ported to 157 cognitive cognitive impairment: Calcium (12 months) GP with impairment not After initial visit: antagonists memory dementia Any type of treatment (antihypertensive concerns Other groups: in response to cognitive drugs) Male 145 dementia impairment (type not speci- (74.9, 52 vascular fied) (76%) ±6.5) dementia Nootropics (24%) Female 73 Alzheimer’s Calcium antagonists (10%) (74.0, disease At 12 months: ±6.9) 137 psycho- Any type of treatment pathological in response to cognitive disorder impairment (type not speci- 25 other fied; 76%) 126 not stated Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 677 Table 1. Continued Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Qualitative Physician-reported management strategies Ambigga Case report Ma- 1 primary care N/A N/A Case vignette— No preset list Preferences for treatment: Vascular medi- Physical activity et al. (45) laysia physician MCI Provided recom- promote independence in cation Social activity mendations based communication and activ- Diet on evidence ities of daily living Cognitive stimu- Mental exercise (e.g. puz- lation zles) Sleep Healthy lifestyle including Limit alcohol physical activity and diet Getting enough sleep Limit alcohol intake Control vascular risk fac- tors (e.g. hypertension) Regular follow up 3–6 months 16.7 years N/A Case vignette— No preset list Participants provided range Blood pressure Physical activity Hochhalter Focus USA 28 primary care (SD = n/a) MCI of pharmacological and (vascular man- Social activity et al. (44) groups, physicians 26.2 years non-pharmacological strat- agement) Diet semi- <44 years old (SD = n/a) egies. However, key findings Cholesterol Cognitive stimu- structured (53.6%) from study also outlined (vascular man- lation interviews 45–64 years old that some participants felt agement) (39.3%) that the management op- Reassessment of >65 years old tions are too generic and diabetes man- (7.1%) that in some cases dementia agement (not 79% male is not preventable. specified) 21 advanced prac- tice provider <44 years old (9.5%) 45–64 years old (90.5%) >65 years old (0%) 19% male M, mean; SD, standard deviation; N/A, not applicable (not provided). 1993 results. 2001 results. Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 678 Family Practice, 2021, Vol. 38, No. 5 with memory concerns using a preset list of pharmacological and non-pharmacological strategies. Across both studies, the top two recommendations were increasing physical activity and increasing cognitive stimulation (41,42). For physicians surveyed in Day et al., the third most common recommendation was for the patient to improve their diet. However, in Friedman et  al., physicians’ third highest recommendation to patients was to increase social stimula- tion. A small proportion, 40 physicians (4%) from Friedman et al., indicated that they would provide no advice for any treatment or strategies in preventing cognitive decline. Day et al. did not report if any physicians would not provide advice to patients with subjective memory concerns (please see Table 2). Mild cognitive impairment For patients presenting with MCI, three studies investigated primary care physicians’ intentions to provide non-pharmacological man- agement strategies. Werner et  al. (40) used a survey-based measure with 11 preset pharmacological and non-pharmacological strategies, which largely overlapped with DocStyles, but had some different strategies listed. Ambigga et al. (45) provide a case report and vignette on how primary care physicians should manage a patient with MCI. The final study, Hochhalter et al. ( 44), conducted a qualitative study using case vignettes in focus groups and semi-structured interviews. Across all three studies (40,44,45), four recommendations were high- lighted: physical activity, cognitive stimulation, social stimulation and diet. PCPs who participated in semi-structured interviews outlined the importance of recommending physical activity for a key reason ‘Vigorous daily exercise… because it improves, basically, all the vas- cular risks which people in this age group face’ [(44), p. 3]. The min- imum requirement of what is deemed enough physical activity, or for any of the other recommendations, is not outlined across any of the studies. Hochhalter et al. (44) also identified a small number of PCPs who did not provide any sort of management strategies because they felt that cognitive impairment, specifically Alzheimer’s disease, is not preventable ‘Stuff like Alzheimer’s, we can’t do anything about. Either you get it, or you don’t. You can’t prevent it’ [(44), p. 3]. Pharmacological management Pharmacological management for people presenting with either memory problems (SMC or MCI) was investigated by all nine studies, which has been outlined above in the Characteristics and quality of included studies section (please see Table 1 for study characteristics). Subjective memory concern For patients presenting with SMC, three studies investigated PCPs’ in- tentions for pharmacological management strategies (41–43) (please see Table 3). Both Day et  al. (41) and Friedman et  al. (42) used the Docstyles measure. Banjo et  al. (43) utilized a different method by using a case vignette of a patient with memory concerns and then asking how comfortable PCPs would be prescribing cognitive enhan- cers. Banjo et al. averaged the PCPs response to how comfortable they felt prescribing a cognitive enhancer (a preset list of sildenafil, methyl- phenidate and modafinil) on a Likert scale with 1 being ‘Less comfort- able’ and 7 being ‘More comfortable’. The PCPs felt most comfortable prescribing sildenafil. The only management response that appeared across all three studies was to provide no pharmacological response (41–43). Banjo et al. did not report the specific number of physicians providing advice but did report that some physicians did not provide any pharmacological response. A minimum of 1 in 5 physicians within Table 2. Physician behaviour of managing MCI and SMCs using a non-pharmacological response Non-pharmacological treatments - % of family practitioners who would provide treatment advice Study Meditation Sleep Reduce BMI Physical activity Limit alcohol Diet Social stimulation Cognitive No treatment stimulation MCI studies Ambigga et al. (45) N/R N/R N/R N/R N/R N/R – Hochhalter et al. (44) – – – N/R – N/R N/R N/R N/R Werner et al. (40) 57/168 – – 148/168 (88.0%) – 74/168 (44.0%) 148/168 (88.0%) 148/168 (88.0%) – (34.0%) SMC studies Day et al. (41) – – 550/972 (56.6%) 892/972 (91.8%) 626/972 (64.4%) 809/972 (83.2%) 775/972 (79.7%) 829/972 (85.3%) – Friedman et al. (42) – – 457/1000(45.7%) 861/1000 (86.1%) 591/1000 (59.1%) 609/1000 (60.9%) 667/1000 (66.7%) 802/1000 (80.2%) 40/1000 (4.0%) N/R = Numbers not reported but treatment advice still provided;  = treatment advice provided. BMI, body mass index. Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 679 Table 3. Physician behaviour of managing MCI and SMCs using a pharmacological response Pharmacological treatments—% of family practitioners who would provide treatment advice Study Anti-dementia Disease Natural remedies Vitamins/ New drugs Reduce Nootropics Vascular No drug treatment drugs management supplements (not specified) polypharmacy management MCI studies Ambigga et al. (45) – – – – – – – N/R – Hochhalter et al. – N/R – – – – – N/R N/R (44) a a a a a a Maeck et al. (38) 0/159 (0%) – 54/159 (34.0%) – 71/159 (44.7%) – 62/159 (39.0%) 56/159 (35.2%) 47/159 (29.6%) b b b b b b 25/122 (20.5%) 28/122 (23.0%) 1/122 (0.8%) 4/122 (3.3%) 3/122 (2.5%) 69/122 (56.6%) Suribhatla et al. (39) – – – – – – – 16/61 (26.0%) 45/61 (74.0%) Werner et al. (40) – – 17/168 (10.1%) 57/168 (33.9%) 22/168 (13.1%) – – – 72/168 (42.9%) Argimon-Pallas et al. – – – – 53.0% – 24.0% 10.0% 47.0% (37) SMC studies Banjo et al. (43) – – – – – – N/R – N/R Day et al. (41) – – – 332/972 (34.2%) 164/972 (16.9%) 435/972 (44.7%) – – 41/972 (4.2%) Friedman et al. (42) – – – 293/1000 (29.3%) 116/1000 (11.6%) 411/1000(41.1%) – – 180/1000 (18.0%) N/R = numbers not reported but treatment advice still provided;  = treatment advice provided. 1993 survey data. 2001 survey data. No figures given within study but is minimum amount of people who did not receive treatment. Figure based on adding all treatments options up; then taking that total away from 100. Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 680 Family Practice, 2021, Vol. 38, No. 5 Friedmann et al. and 1 in 20 physicians within Day et al. reported that type II diabetes) (44), natural remedies (such as Gingko Biloba) they would not provide any pharmacological response at all. These (38,40) and even anti-dementia drugs (38). In 2001, 122 PCPs in are minimum estimates as these figures are based on adding all treat- Germany (38) were given a case vignette of a patient with MCI who ment options up, then taking that total away from the study popu- has an increased risk of developing dementia. At that time, 12% of lation. However, within two studies, pharmacological response was PCPs (n = 15) would prescribe memantine and 8% (n = 10) would more frequent among physicians than no treatment at all. Reducing prescribe cholinesterase inhibitors to improve cognitive symptoms in polypharmacy was a management response to SMC being reported people with MCI (38). that just under half of physicians highlighted across two studies (41,42). Additionally, approximately a third of physicians in two Discussion studies also reported that they recommended the initiation of supple- ments and vitamins (41,42). However, the specific type of vitamins The review-highlighted PCPs were reporting that they were more and supplements were not specified. likely to provide non-pharmacological strategies than pharmaco- logical treatments. The three most common non-pharmacological Mild cognitive impairment strategies reported as being used to reduce cognitive decline and Five studies investigated PCPs’ intentions and one study investigated dementia risk in people with memory problems were (i) physical PCPs’ observed behaviour for pharmacological management strat- activity, (ii) cognitive stimulation and (iii) social stimulation (40– egies for patients presenting with MCI (please see Table 3). Across 42,44,45). Particular types of physical activity or cognitive and four of the five studies investigating reported management strategies, social stimulation were not specified. However, current evidence sug- physicians would not provide any pharmacological treatment in re- gests that not all types of physical activity are equally effective. For sponse to managing a patient with MCI. Maeck et al. surveyed phys- example, in a recent review, 4–6 months of aerobic exercise twice a icians in 1993 and 2001. In 1993, just under one in three physicians week or one to three times a week combining cognitive and motor reported that they would not typically provide any pharmacological challenges (Tai Chi, dance or dumbbell training) works to improve treatment. In comparison to 2001, just over one in two physicians memory and global cognitive functioning, but short-term resistance would not provide any pharmacological treatment. In a more recent training for less than 4 months did not improve memory or cognitive survey, Werner et al. also indicated that just under one in two phys- functioning (46–48). While there is less evidence in the arenas of cog- icians reported that they would not provide any pharmacological nitive and social activities, it appears that, in these domains too, not treatment. For physicians surveyed over the last 20  years, 43% to all activity types are equally effective (17,46–48). Other key strat- 74% would not prescribe any form of medication (38–40,44). If egies that physicians reported that they used included improving diet physicians were to advise on the use of pharmacological treatment, (40–42,44,45) and reducing alcohol intake (41,42,45). However, it is vascular management appeared the most common, being highlighted important to consider that all studies on non-pharmacological man- across four of the five studies (38,39,44,45). Vascular management agement evaluated self-reported (hypothetical) behaviours and none included any treatments aimed at lowering cholesterol, blood pres- observed actual behaviours. Additionally, three of the five studies sure and blood glucose in order to improve blood flow. One in four investigating non-pharmacological strategies used preset survey lists. physicians in Suribhatla et  al. reported that they would prescribe Therefore, these studies did not provide opportunity for physicians statins to manage vascular-related MCI. This was supported by a to outline other strategies they may implement. similar response rate of using vascular treatment management for For pharmacological treatment offered by PCPs for people with MCI by physicians surveyed in 1993 within the Maeck et al. study. memory problems, the most common across eight of the nine studies However, by 2001, this treatment strategy was reported by only was to provide no drug treatment. This appears to be in line with 3 physicians out of 122 surveyed. Two studies did not report the guidance for MCI management (49), which does not recommend number of physicians as one was a case report and the other was any drug treatments. Additionally, treatment for memory problems a qualitative study (44,45). Physicians within the focus groups out- is typically assessed and initiated by specialists in memory clinics or lined the importance of managing vascular risk factors not just for other secondary care services, which is common practice in coun- risk of conversion to dementia but also other health conditions that tries in North America, Europe and Oceania (50–52). However, it is could occur as a result of vascular disease (44). Only one study in important to consider that, within two studies investigating SMCs, the review (37) investigated observed natural behaviour rather than physicians were more frequently providing some pharmacological physicians’ reported management strategies. Argimon-Pallas et  al. responses, the most common responses being vascular risk man- (37) conducted a 12-month naturalistic descriptive study of 105 agement and vitamins. As for non-pharmacological approaches, the primary care centres across Spain and 202 patients who presented studies did not report the specific vascular management strategies with cognitive impairment. Of these patients, one in four were pre- used, and not all are equally effective. For example, insulin therapy scribed nootropics, which are drugs aimed at enhancing cognition has been associated with an increased risk of developing dementia, and can include piracetam (38), methylphenidate (43) and modafinil whereas thiazolidinedione exposure is associated with protective ef- (43). However, the type of nootropics prescribed in Argimon-Pallas fects and reduces the risk of dementia (14). Some evidence has indi- were not specified. One in 10 patients was prescribed calcium ant- cated that all classes of antihypertensives may have protective effects agonists, which are primarily used for treating hypertension but can for dementia with minimal difference in effect between classes (53). also be used for heart arrhythmia and headaches. This is a similar For vitamin or supplement management, low levels of vitamin D rate to the patients diagnosed with dementia within this study, but (54) or B vitamins (55) (B6, folate and B12) are typically associated Argimon-Pallas et  al. (37) did not provide analysis of any other with increased risk of dementia and are specific vitamin deficiencies comparator groups. that PCPs could address with minimal adverse effects. Other pharmacological strategies that PCPs reported they would Despite mixed evidence, the World Health Organization (48) use included prescription of vitamins (40), new drugs (type not spe- has set out a list of strategies for managing people at high risk of cified) (38,40), review of disease management medication (such as developing dementia that are appropriate for PCPs across the world Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 681 to deliver. This review has demonstrated that most PCPs’ reported Future research management strategies are adhering to most of the generic recom- Though self-report measures may provide some correspondence to mendations outlined in the WHO report. However, within the observed behaviour, there are still large discrepancies between self- included studies, there were some important omissions of manage- reported attitudes and actual observed behaviours (58,59). To gain a ment strategies that PCPs did not report as offering to people with more accurate reflection of primary care current management strat- memory problems. Depression, smoking and hearing loss are asso- egies for people with MCI or SMC, high-quality longitudinal observa- ciated with an increased risk of developing dementia, yet no study tional studies are needed. Observational studies can provide an insight or PCPs acknowledged this as an important strategy. Additionally, it into if people with memory problems are actively being managed is important to note that most of the included studies are reported differently than people who are cognitively healthy. Future research strategies from PCPs and, therefore, may not accurately portray be- should monitor both pharmacological and non-pharmacological de- haviours in observed practice. The only study to use a descriptive mentia prevention strategies offered by primary care. Research should naturalistic design, which was conducted in 2007, demonstrated that also capture the specific types of management strategies offered, such neurotropics (cognitive enhancers) were being prescribed more than as aerobic exercise or weight training for physical activity. is being recommended (37). This is perhaps surprising given the lack of evidence to suggest the effectiveness of neurotropics or acetyl- Conclusion cholinesterase inhibitors in people with MCI and SMC (56,57). In particular, acetylcholinesterase inhibitor prescription in MCI should The current review highlighted that when people are presenting with not be recommended due to many safety issues and minimal im- memory problems, primary care physicians will suggest that the pa- provement in cognition (57). tient can mitigate cognitive decline by improving physical activity, Primary care is in an optimal position to not only first identify cognitive stimulation, social stimulation and diet. Addressing hearing people with memory concerns and problems but also to coordinate loss, smoking and depression were not mentioned as strategies. For the management of risk after the patient is screened as having SMC MCI, most physicians report that they will not intend to prescribe or MCI. Therefore, it is important that PCPs advise people with any pharmacological treatments; but if they did, it would most likely memory problems on the modifiable health and lifestyle factors as- be to manage vascular risk factors. For SMC, there were physicians sociated with dementia, such as hypertension, depression, hearing across all three studies that provided no pharmacological treatment loss and the other nine factors identified in the Lancet commission at all. However, in two studies, physicians were more likely to reduce (4). By informing patients of these strategies, people with memory polypharmacy and increase vitamins than to provide no treatment at problems could reduce the risk of further cognitive decline or delay all. Most studies were surveys of subjective self-reported behaviours the onset of dementia. and there is a lack of strong evidence to accurately answer what are the current treatment responses for people with memory problems Limitations provided by PCPs. Future research using observational study designs is needed to obtain a more accurate reflection of actual current prac- There are some limitations to consider when interpreting the find- tice rather than reported practice. By understanding current prac- ings of this study. Due to heterogeneity in location, population and tices, research can optimize the management of cognitive decline and methods across different studies, we did not pool data across the dementia prevention in primary care. studies for a meta-analysis. We employed inclusive eligibility cri- teria in terms of study design, which allowed survey-based studies, qualitative interviews and observational studies to be synthesized to- Supplementary material gether. The included studies were conducted across a range of coun- Supplementary material is available at Family Practice online. tries, with different guidelines for practice, which may have impacted on the strategies reported by the PCPs. A  major limitation of all studies was that control groups were not used to compare how treat- Acknowledgements ment for an older patient at high risk of developing dementia might The authors are extremely grateful for the support provided in the development differ from an older patient with no memory problems. Therefore, of the search strategy by Jenni Ford, Evidence Services Librarian at Royal Free the percentage of people with memory problems who receive non- Library. The authors would also like to thank the numerous authors who were pharmacological recommendations, such as diet, physical activity contacted and kindly provided additional information regarding their studies. and social stimulation, may be the same percentage of older people who would anyway receive non-pharmacological recommendations Declaration as part of general health promotion advice or to treat other condi- Funding: Brendan Hallam is a PhD student funded by the Economic & Social tions. The lack of description, especially for pharmacological treat- Research Council’s London (UBEL) Doctoral Training Partnership, embed- ments, made it difficult to know the specific types of drugs used. For ded within the APPLE-TREE programme (Project reference: ES/S010408/1). example, Argimon-Pallas et al. (37) used the term nootropics, which Ethics approval: none. is a generic term for substances that aim to improve cognition, and Conflicts of interest: none declared. can range from caffeine to Ritalin. Other limitations in relation to the methodology of the current References review are only selecting English language studies. The current re- 1. World Health Organization. Dementia Factsheet: World Health Organ- view did not have the capacity or resources to translate non-English ization. 2020. Updated 21 September 2020. https://www.who.int/news- articles, which could introduce bias if potential key data from room/fact-sheets/detail/dementia (accessed on 14 January 2021). non-English articles are missed. Additionally, due to limited re- 2. World Health Organization. The Top 10 Causes of Death: World Health sources, the review also prioritized peer-reviewed articles to main- Organization. 2020. Updated 9 December 2020. https://www.who.int/ tain the scientific standard of the literature included in the review news-room/fact-sheets/detail/the-top-10-causes-of-death (accessed on 14 and excluded grey literature. January 2021). Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 682 Family Practice, 2021, Vol. 38, No. 5 3. Office for National Statistics. Deaths Registered in England and Wales: 25. National Institute for Health and Care Excellence. Dementia, Disability 2018. Office for National Statistics. 2019. https://www.ons.gov.uk/ and Frailty in Later Life-Mid-Life Approaches to Delay or Prevent Onset peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bul- (NG16). NICE Guideline. 2015. https://www.nice.org.uk/guidance/ng16 letins/deathsregistrationsummarytables/2018 (accessed on 23 June 2020). (accessed on 23 June 2020). 4. Livingston G, Huntley J, Sommerlad A et al. Dementia prevention, inter- 26. Petersen RC, Lopez O, Armstrong MJ et  al. Practice guideline update vention, and care: 2020 report of the Lancet Commission. Lancet 2020; summary: Mild cognitive impairment: report of the Guideline Develop- 396(10248): 413–46. ment, Dissemination, and Implementation Subcommittee of the American 5. Ngandu T, Lehtisalo J, Solomon A et al. A 2 year multidomain intervention Academy of Neurology. Neurology 2018; 90(3): 126–35. of diet, exercise, cognitive training, and vascular risk monitoring versus 27. Royal Australian College of General Practitioners. Guidelines for Pre- control to prevent cognitive decline in at-risk elderly people (FINGER): a ventive Activities in General Practice. 9th edn, updated. East Melbourne, randomised controlled trial. Lancet 2015; 385(9984): 2255–63. Australia: RACGP; 2018. 6. Viticchi G, Falsetti L, Buratti L et al. Framingham risk score and the risk 28. Wells  CE, Smith  SJ. Diagnostic care pathways in dementia: a review of of progression from mild cognitive impairment to dementia. J Alzheimers the involvement of primary care in practice and innovation. J Prim Care Dis 2017; 59(1): 67–75. Community Health. 2017; 8(2): 103–11. 7. Newson RS, Kemps EB. The nature of subjective cognitive complaints of 29. Ahmad S, Orrell M, Iliffe S, Gracie A. GPs’ attitudes, awareness, and prac- older adults. Int J Aging Hum Dev 2006; 63(2): 139–51. tice regarding early diagnosis of dementia. Br J Gen Pract 2010; 60(578): 8. Steinberg SI, Negash S, Sammel MD et al. Subjective memory complaints, e360–5. cognitive performance, and psychological factors in healthy older adults. 30. Godbee K, Gunn J, Lautenschlager NT, Curran E, Palmer VJ. Implementing Am J Alzheimers Dis Other Demen 2013; 28(8): 776–83. dementia risk reduction in primary care: a preliminary conceptual model 9. Gauthier S, Reisberg B, Zaudig M et al.; International Psychogeriatric As- based on a scoping review of practitioners’ views. Prim Health Care Res sociation Expert Conference on Mild Cognitive Impairment. Mild cogni- Dev 2019; 20: e140. tive impairment. Lancet 2006; 367(9518): 1262–70. 31. Moher  D, Liberati  A, Tetzlaff  J, Altman  DG; PRISMA Group. Preferred 10. Lopez OL, Kuller LH, Becker JT et al. Incidence of dementia in mild cog- reporting items for systematic reviews and meta-analyses: the PRISMA nitive impairment in the cardiovascular health study cognition study. Arch statement. PLoS Med 2009; 6(7): e1000097. Neurol 2007; 64(3): 416–20. 32. Vaportzis E, Gow AJ. People’s beliefs and expectations about how cogni- 11. Stewart R. Subjective cognitive impairment. Curr Opin Psychiatry 2012; tive skills change with age: evidence from a UK-wide aging survey. Am J 25(6): 445–50. Geriatr Psychiatry 2018; 26(7): 797–805. 12. Mitchell AJ, Beaumont H, Ferguson D, Yadegarfar M, Stubbs B. Risk of de- 33. Noyes  J, Booth  A, Moore  G, Flemming  K, Tunçalp  Ö, Shakibazadeh  E. mentia and mild cognitive impairment in older people with subjective memory Synthesising quantitative and qualitative evidence to inform guidelines on complaints: meta-analysis. Acta Psychiatr Scand 2014; 130(6): 439–51. complex interventions: clarifying the purposes, designs and outlining some 13. Tully  PJ, Hanon  O, Cosh  S, Tzourio  C. Diuretic antihypertensive drugs methods. BMJ Glob Health 2019; 4(Suppl 1): e000893. and incident dementia risk: a systematic review, meta-analysis and meta- 34. Hong QN, Fàbregues S, Bartlett G et  al. The Mixed Methods Appraisal regression of prospective studies. J Hypertens 2016; 34(6): 1027–35. Tool (MMAT) version 2018 for information professionals and researchers. 14. McMillan JM, Mele BS, Hogan DB, Leung AA. Impact of pharmacological Educ Inf 2018; 34(4): 285–91. treatment of diabetes mellitus on dementia risk: systematic review and 35. Pace  R, Pluye  P, Bartlett  G et  al. Testing the reliability and efficiency of meta-analysis. BMJ Open Diabetes Res Care 2018; 6(1): e000563. the pilot Mixed Methods Appraisal Tool (MMAT) for systematic mixed 15. Dupré C, Bongue B, Helmer C et al. Physical activity types and risk of de- studies review. Int J Nurs Stud 2012; 49(1): 47–53. mentia in community-dwelling older people: the Three-City cohort. BMC 36. Souto RQ, Khanassov V, Hong QN, Bush PL, Vedel I, Pluye P. Systematic Geriatr 2020; 20(1): 132. mixed studies reviews: updating results on the reliability and efficiency 16. Choi D, Choi S, Park SM. Effect of smoking cessation on the risk of de- of the Mixed Methods Appraisal Tool. Int J Nurs Stud 2015; 52(1): mentia: a longitudinal study. Ann Clin Transl Neurol 2018; 5(10): 1192–9. 500–1. 17. Fratiglioni  L, Marseglia  A, Dekhtyar  S. Ageing without dementia: can 37. Argimón  Pallàs  JM, Riu  Subirana  S, Lizán  Tudela  L, Badia  Llach  X, stimulating psychosocial and lifestyle experiences make a difference? Martínez Lage JM. [Initial treatment of patients with subjective memory Lancet Neurol 2020; 19(6): 533–43. complaints and/or cognitive impairment in primary care: ISSEA Study]. 18. Amieva H, Ouvrard C. Does treating hearing loss in older adults improve Aten Primaria 2007; 39(4): 171–7. cognitive outcomes? A review. J Clin Med 2020; 9(3): 805. 38. Maeck L, Haak S, Knoblauch A, Stoppe G. Primary care physicians’ at- 19. Danat  IM, Clifford  A, Partridge  M et  al. Impacts of overweight and titudes related to cognition enhancers in early dementia: a representa- obesity in older age on the risk of dementia: a systematic literature review tive eight-year follow-up study in Lower Saxony, Germany. Int J Geriatr and a meta-analysis. J Alzheimers Dis 2019; 70(s1): 87–99. Psychiatry 2008; 23(4): 415–21. 20. Chan JY, Yiu KK, Kwok TC, et al. Depression and antidepressants as po- 39. Suribhatla S, Dennis MS, Potter JF. A study of statin use in the prevention tential risk factors in dementia: a systematic review and meta-analysis of of cognitive impairment of vascular origin in the UK. J Neurol Sci 2005; 18 longitudinal studies. J Am Med Dir Assoc 2019; 20(3): 279–86. e1. 229–230: 147–50. 21. Sabia S, Fayosse A, Dumurgier J et  al. Alcohol consumption and risk of 40. Werner  P, Heinik  J, Kitai  E. Familiarity, knowledge, and preferences of dementia: 23 year follow-up of Whitehall II cohort study. BMJ 2018; 362: family physicians regarding mild cognitive impairment. Int Psychogeriatr k2927. 2013; 25(5): 805–13. 22. Andrieu  S, Guyonnet  S, Coley  N et  al.; MAPT Study Group. Effect of 41. Day KL, Friedman DB, Laditka JN, et al. Prevention of cognitive impair- long-term omega 3 polyunsaturated fatty acid supplementation with or ment: physician perceptions and practices. J Appl Gerontol. 2012; 31(6): without multidomain intervention on cognitive function in elderly adults 743–54. with memory complaints (MAPT): a randomised, placebo-controlled trial. 42. Friedman DB, Rose ID, Anderson LA et  al. Beliefs and communication Lancet Neurol 2017; 16(5): 377–89. practices regarding cognitive functioning among consumers and primary 23. Cooper C, Aguirre E, Barber JA, et al. APPLE‐Tree (Active Prevention in care providers in the United States, 2009. Prev Chronic Dis 2013; 10: E58; People at risk of dementia: Lifestyle, bEhaviour change and Technology quiz 8–13. to REducE cognitive and functional decline) programme: protocol. Int J 43. Banjo OC, Nadler R, Reiner PB. Physician attitudes towards pharmaco- Geriatr Psychiatry 2019. logical cognitive enhancement: safety concerns are paramount. PLoS One 24. Yaffe  K, Barnes  DE, Rosenberg  D et  al. Systematic multi-domain 2010; 5(12): e14322. Alzheimer’s risk reduction trial (SMARRT): study protocol. J Alzheimers 44. Hochhalter AK, Bryant LL, Hunter R et al. Multisite qualitative study of Dis 2019; 70(s1): 207–20. primary care physicians’ and midlevel providers’ self-reported practices Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 683 and perceptions about maintaining cognitive health. Prev Chronic Dis 52. OECD. Care Needed: Improving the Lives of People With Dementia. 2012; 9: E169. Paris: OECD Publishing; 2018. 45. Ambigga D, Suthahar A, Ramli A, Ng K, Radziah A, Marymol K. Diagnosis 53. Murray  MD, Hendrie  HC, Lane  KA et  al. Antihypertensive medication and management of mild cognitive impairment in the community: what is the and dementia risk in older adult African Americans with hypertension: a role of primary care physician? Malays Fam Physician 2011; 6(2–3): 74–8. prospective cohort study. J Gen Intern Med 2018; 33(4): 455–62. 46. Whitty E, Mansour H, Aguirre E et  al. Efficacy of lifestyle and psycho- 54. Sommer I, Griebler U, Kien C et al. Vitamin D deficiency as a risk factor social interventions in reducing cognitive decline in older people: system- for dementia: a systematic review and meta-analysis. BMC Geriatr 2017; atic review. Ageing Res Rev 2020; 62: 101113. 17(1): 16. 47. Alty  J, Farrow  M, Lawler  K. Exercise and dementia prevention. Pract 55. Lefèvre-Arbogast S, Féart C, Dartigues J-F et al. Dietary B vitamins and a Neurol 2020; 20(3): 234–40. 10-year risk of dementia in older persons. Nutrients 2016; 8(12): 761. 48. World Health Organization. Risk Reduction of Cognitive Decline and De- 56. Tricco AC, Soobiah C, Berliner S et al. Efficacy and safety of cognitive en- mentia: WHO Guidelines. France: World Health Organization; 2019. https:// hancers for patients with mild cognitive impairment: a systematic review apps.who.int/iris/bitstream/handle/10665/312180/9789241550543-eng. and meta-analysis. CMAJ 2013; 185(16): 1393–401. pdf?ua=1 (accessed on 17 July 2020). 57. Matsunaga S, Fujishiro H, Takechi H. Efficacy and safety of cholinesterase 49. NICE CKS. Dementia, Scenario: Suspected Dementia. 2019. [2 April 2020]. inhibitors for mild cognitive impairment: a systematic review and meta- https://cks.nice.org.uk/dementia#!scenario (accessed on 25 June 2020). analysis. J Alzheimers Dis 2019; 71(2): 513–23. 50. Jolley D, Benbow SM, Grizzell M. Memory clinics. Postgrad Med J 2006; 58. Eccles MP, Hrisos S, Francis J et al. Do self-reported intentions predict 82(965): 199–206. clinicians’ behaviour: a systematic review. Implement Sci 2006; 1(1): 51. Gruters  AAA, Ramakers  IHGB, Kessels  RPC et  al. Development of 28. memory clinics in the Netherlands over the last 20  years. Int J Geriatr 59. Sheeran P, Webb TL. The intention–behavior gap. Soc Pers Psychol Com- Psychiatry 2019; 34(8): 1267–74. pass. 2016; 10(9): 503–18. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Family Practice Oxford University Press

How are people with mild cognitive impairment or subjective memory complaints managed in primary care? A systematic review

Download PDF
15 pages

Loading next page...
 
/lp/oxford-university-press/how-are-people-with-mild-cognitive-impairment-or-subjective-memory-v0apmU0VSV

References (62)

Publisher
Oxford University Press
Copyright
Copyright © 2022 Oxford University Press
ISSN
0263-2136
eISSN
1460-2229
DOI
10.1093/fampra/cmab014
Publisher site
See Article on Publisher Site

Abstract

Background: Primary care is typically the first point of contact in the health care system for people raising concerns about their memory. However, there is still a lack of high-quality evidence and understanding about how primary care professionals (PCPs) currently manage people at higher risk of developing dementia. Objectives: To systematically review management strategies provided by PCPs to reduce cognitive decline in people with mild cognitive impairment and subjective memory complaints. Method: A systematic search for studies was conducted in December 2019 across five databases (EMBASE, Medline, PsycInfo, CINAHL and Web of Science). Methodological quality of included studies was independently assessed by two authors using the Mixed Methods Appraisal Tool. Results: An initial 11  719 were found, 7250 were screened and 9 studies were included in the review. Most studies were self-reported behaviour surveys. For non-pharmacological strategies, the most frequent advice PCPs provided was to increase physical activity, cognitive stimulation, diet and social stimulation. For pharmacological strategies, PCPs would most frequently not prescribe any treatment. If PCPs did prescribe, the most frequent prescriptions targeted vascular risk factors to reduce the risk of further cognitive decline. Conclusion: PCPs reported that they are much more likely to provide non-pharmacological strategies than pharmacological strategies in line with guidelines on preventing the onset of dementia. However, the quality of evidence within the included studies is low and relies on subjective self-reported behaviours. Observational research is needed to provide an accurate reflection of how people with memory problems are managed in primary care. Lay summary People will typically go to their general practitioners, also known as primary care professionals (PCPs), to raise concerns about their memory. However, there is no clear understanding of what advice or treatment PCPs provide to people with memory concerns who are at high risk of dementia. This review aims to summarize the findings from research that studied what advice or treatments PCPs would give to a person with memory concerns. Nine studies were included in the review after screening through 11 719 studies. The current review found that PCPs were more likely to provide advice rather than prescribe any drug treatment. The most common advice that © The Author(s) 2021. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/ by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 670 Family Practice, 2021, Vol. 38, No. 5 Key Messages • Review of primary care professionals’ (PCPs) management of memory concerns. • The review included a wide range of quantitative and qualitative study designs. • Most frequent advice was to increase physical activity. • Most common drug response was to not prescribe any treatment. • Majority of PCPs reported strategies that followed prevention guidelines. • Future research needs more observational studies to observe real-life practice. PCPs provided was to increase physical activity, cognitive stimulation and social stimulation. If PCPs decided to prescribe drugs, the most common prescriptions were to improve blood flow. Improving blood flow has been linked with reducing the risk of developing dementia. However, the quality of the studies included in this review is low because many relied on PCPs answering questionnaires on their intentions to manage people with memory concerns. Therefore, future research needs to observe PCPs’ real-life practice to provide an accurate reflection of how people with memory problems are managed in primary care. Key words: Cognitive dysfunction, dementia, memory, primary health care, primary prevention, systematic review protective factors for dementia. Alcohol misuse (21) and dementia Introduction has a complex J-shaped relationship with excessive alcohol use and Background non-consumption being associated with greater risk than moderate An estimated 50 million people are expected to be living with de- consumption. However, this research addressed all risk factors indi- mentia worldwide, with this projected to rise to 152 million in the vidually rather than the effectiveness of a behavioural health inter- next 30  years (1). Dementia is the seventh leading cause of death vention that combines strategies for multiple risk factors. Evidence across the world (2) and the leading cause of death within England from trials of time-intensive behavioural health interventions and Wales (3). Dementia is the only condition within the top 10 targeting the lifestyle risk factors aiming to reduce cognitive decline causes of death without a treatment to slow or cure its progression and onset of dementia in people with memory concerns is mixed (3). However, it is believed that up to 40% of dementia cases could (5,22). Further investigations of lifestyle interventions, such as Active be prevented if the following risk factors were addressed: low level Prevention in People at risk of dementia through Lifestyle, bEhav- of education, hearing loss, traumatic brain injury, hypertension, al- iour change and Technology to build REsiliEnce (APPLE-Tree) (23) cohol misuse, obesity, smoking, depression, physical inactivity, social and the Systematic Multi-domain Alzheimer’s Risk Reduction Trial isolation, air pollution and diabetes (4). (SMARTT) (24) are ongoing. SMARRT will recruit older adults People defined as high risk of developing dementia have been with subjective cognitive complaints from primary care and be ran- operationalized in various ways. For example, the FINGER trial (5) domly assigned to the intervention or a health education control. used the CAIDE dementia risk score, whilst other studies may use The intervention will be to develop a personalized plan for risk fac- the Framingham vascular risk scores (6). However, the one indicator tors hypertension, hyperglycaemia, depressive symptoms, poor sleep, that often leads to consultation due to concerns about the risk of polypharmacy, physical inactivity, low cognitive stimulation, social developing dementia is memory concerns (7). The term ‘memory isolation, poor diet and smoking. All of these factors are associated concerns’ refers to people with subjective memory complaint (SMC) with an increased risk of dementia and strategies addressing these and mild cognitive impairment (MCI). SMC is defined as a form issues provide the most likely approach to delay the onset of de- of complaint that an individual makes regarding his or her cogni- mentia. However, the efficacy of dementia prevention interventions tion, but no clear impairment is found by objective psychometric in delaying incident dementia is still mixed and inconclusive (5,22). testing (8). In contrast, people with MCI do show a noticeable de- Therefore, there are no current specific treatment recommenda- cline in cognition using objective testing, which is not severe enough tions provided by the national health governing bodies for people to interfere with daily activities and be defined as dementia (9). SMC with memory problems (SMC and MCI) due to the lack of strong affects half of people over 65 years old (10) and MCI affects 20% of current evidence (25–27). Consequently, the current guidelines for people over 65 (11). Reviews have indicated that people with SMC health professionals to delay the onset of dementia is to provide gen- are twice as likely to develop dementia as individuals without SMC eric non-pharmacological recommendations to all people in mid-life (12), highlighting the need for health care professionals to effectively (25). This includes encouraging healthy behaviours, such as smoking manage people with SMC and MCI in order to reduce the risk of cessation, increasing physical activity and reducing alcohol con- developing dementia. sumption (25). There is low-to-moderate quality evidence that addressing hyper- Primary care is typically the first point of contact in the health tension (13), diabetes (14), physical activity (15), tobacco cessation care system for people raising concerns about their memory (28). (16), cognitive stimulation (17) and social isolation (17) has been Therefore, primary care is critically placed to play a greater role demonstrated to reduce dementia risk in low-to-moderate quality in providing preventive treatments to delay the onset of dementia evidence. Treatment addressing hearing loss (18), obesity (19) and in adults with memory problems (28). Despite this, dementia pre- depression (20) requires further research and has yet to demonstrate vention advice or even recognition of cognitive impairment by Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 671 general practitioners (GPs) is variable, often with failure to respond BH and JR completed 100% of the full-text screening independently to memory loss symptoms (29). Godbee et  al. (30) have recently with any discrepancies resolved by a third independent reviewer. published a preliminary conceptual model on how to implement From the studies included in the systematic review, a pre-piloted dementia risk reduction practice in primary care, providing five data collection form was used by BH and JR to extract the necessary implementation strategies, which were (i) identifying ‘champions’ data. Extracted data included: author (year), study design, setting, to promote brain health to patients, (ii) conducting educational professionals, service users, key findings/themes, type of pharmaco- meetings, (iii) conducting local consensus discussions, (iv) altering logical recommendations and type of non-pharmacological recom- incentive structure and (v) capturing and sharing local knowledge. mendations. Study authors were contacted for any missing data or However, there is still a lack of high-quality evidence and under- any additional data that might be deemed relevant to the review. standing about how primary care professionals (PCPs) currently A  narrative analysis of studies was conducted using a data-driven manage people at higher risk of developing dementia. Therefore, this integrated synthesis approach. Quantitative and qualitative studies systematic review will investigate what management strategies are were synthesized applying a transformation process known as offered by PCPs in response to managing cognitive decline and risk quantitizing. Quantitizing is a method validated for mixed-method of dementia in people with MCI or SMC. The review will aim to reviews whereby qualitative data are quantified. (33) bridge the gap within the literature by exploring both pharmaco- logical and non-pharmacological strategies recommended to people Quality assessment with MCI or SMC in a primary care setting. Two authors independently assessed the methodological quality of each study using the mixed-methods appraisal tool (MMAT) (34). The use of MMAT in mixed-method reviews has been validated, Methods which then allows quality appraisal for the variety of study de- This review was performed in accordance with the PRISMA guide- signs to be completed using one tool (35,36). Therefore, the MMAT lines (31) and the protocol was registered with Prospero (ID: was chosen to appraise both qualitative and quantitative study de- CRD42020170804). signs included in the current review. Similar to data extraction, the interrater reliability was deemed acceptable with Kappa equal or Search strategy above 0.8, and any disagreements were discussed with a third inde- pendent reviewer. The systematic review was conducted on 11 December 2019 using five online bibliographic databases (EMBASE, Medline, PsycInfo, CINAHL and Web of Science). See Supplementary Figures 1–5 for Results full search terms used. No limits were set for time or language and authors were contacted to acquire missing or further information Study selection if needed. Forward selection and reference lists from the final in- The search yielded 11 719 papers. After de-duplication and the add- cluded papers were manually searched to identify potentially rele- ition of one extra paper identified through other sources, 7250 title vant studies that may not have been captured in the literature search. and abstracts were screened. A second independent reviewer screened 10% (n = 725) of the title and abstracts with a high interrater re- Inclusion and exclusion liability (a  =  0.89). Of 275 full-text papers retrieved, 9 were in- cluded in the final systematic review with high interrater agreement To be included, studies were required to assess pharmacological or (a = 0.85). Figure 1 summarizes the study selection process (31). non-pharmacological management options provided by any profes- sional (GPs, practice nurses, pharmacists, etc.) in a primary care set- ting to people over 50  years old with MCI or cognitive complaint Characteristics and quality of included studies without dementia. The threshold of 50 years old was selected as ac- We included seven quantitative studies: one descriptive naturalistic quired memory concerns are increasing and starting to be treated study (37), one structured interview (38) and five cross-sectional more seriously (32). The study could be quantitative or qualita- surveys (39–43) of PCPs’ self-reported management strategies. tive. Non-English language papers were accepted during initial Additionally, two qualitative studies were included, one study using screening. However, non-English papers were excluded during full- semi-structured interviews (44) and one case report (45). The in- text screening if an English version was not be obtained. Exclusion cluded studies are set across seven countries (Canada, Germany, criteria included only people with a confirmed diagnosis of dementia Israel, Malaysia, Spain, UK and USA), with four studies including or healthy older adults. Intervention-based studies were excluded in data from the USA. A total of 2756 primary care physicians partici- order to capture real-life management practices. Additionally, inter- pated across eight of the included studies, with Argimon-Pallas et al. ventional studies, reviews, book chapters and dissertations were also (37) reporting the number of primary care practices participating excluded. Finally, if the study focussed on diagnosis or screening ra- rather than the number of physicians. Six of the studies focussed on ther than treatment or management, it was also excluded. the management of people with MCI (37–40,44,45). Three studies focussed on SMC and memory concerns (41–43). Data extraction The methodological quality of the study designs included was Two reviewers were responsible for the screening process. The second of low-to-moderate quality overall. Aspects of methodology and reviewer (JR) completed a random 10% of the initial screening that analysis for several of the studies were unclear. None of the studies was blinded to the first reviewer (BH). If interrater reliability was included healthy control groups to allow comparisons between below 0.80 for Kappa, then another 10% of the papers would be managements strategies of PCPs for both cognitively healthy older screened by JR. However, if Kappa was above 0.80, then this would adults versus people with memory problems. Argimon-Pallas et  al. be deemed satisfactory and reviewers would progress to full-text (37) was the only study using comparison groups, comparing treat- screening. If either reviewer considered a paper potentially relevant, ments received for groups with memory problems against group it was retrieved and included for the full-text screening process. Both with confirmed diagnosis of dementia. Another concern for each Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 672 Family Practice, 2021, Vol. 38, No. 5 Figure 1. PRISMA flowchart describing the process of study selection. of the survey-based designs was the lack of clarity on accounting memory problems (SMC or MCI). Three of the five studies were for the potential bias in response rates and investigating any dif- survey based, one was a case report and one was semi-structured ference in characteristics between responders and non-responders interviews and a focus group. of the survey. The quality appraisal for all studies can be found in Supplementary Table 1 (a = 0.80). Subjective memory concern Two studies investigated primary care physician’s non- Non-pharmacological management pharmacological management intentions in response to a patient Two thousand one hundred and sixty-nine primary care phys- presenting with SMC (41,42). Both studies used the DocStyles icians were recruited across five studies that investigated non- survey measure. DocStyles is a web-based survey with a range of pharmacological management for people presenting with either questions, including how to reduce cognitive decline in people Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 673 Table 1. Study characteristics of studies included in systematic review Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Quantitative Physician-reported management strategies Banjo et al. Survey USA, 212 primary care 1–10 years N/A Case vi- Yes: Likert scale (1–7 with 7 Modafinil N/A (43) Canada physicians (45%) gnette—65 year three types of being highest comfort) of Methylphenidate 25–40 years old 11–20 old with SMC neurotropics how comfortable physician Sildenafil (36%) (24%) (cognitive would feel prescribing cog- (all drugs fit the 41–59 years old 20+ years enhancers) nitive enhancers: criteria for noo- (45%) (31%) M = 4.8 (SD N/A) tropics, otherwise 60+ years old known as cogni- (19%) tive enhancers) 55% males Medication (not Physical activity Day et al. Survey USA 493 primary care 3–19 years N/A Memory Yes: % of physicians that would specified) Social activity (41) physicians (73%) concerns (not 10 options provide advice to patient Avoid Diet 479 internist 20+ years specified) (6 non- on: polypharmacy Cognitive stimu- <50 years old (27%) pharmacological; Physical activity n = 892 Nutritional sup- lation (69%) 3 pharmaco- (91.8%) plements Limit alcohol 50+ years old logical; 1 no Intellectual stimulation (not specified) Weight/BMI (31%) treatment option) n = 829 (85.3%) Healthy diet n = 809 77% males (83.2%) Socially activity n = 775 (79.7%) Limiting alcohol n = 626 (64.4%) Maintaining a healthy weight n = 511 (52.6%) Avoiding polypharmacy n = 434 (44.7%) Taking nutritional supple- ments n = 332 (34.2%) Taking certain new medica- tions (not specified) n = 164 (16.9%) Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 674 Family Practice, 2021, Vol. 38, No. 5 Table 1. Continued Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Friedman Survey USA 1,000 family phys- 14.7 years N/A Memory Yes: % of physicians that would Medication (not Physical activity et al. (42) icians and internist (SD = n/a) concerns (not 10 options provide advice to patient specified) Social activity 72% males 16.4 years specified) (6 non- on: Avoid Diet 250 nurse practi- (SD = n/a) pharmacological; No recommendations polypharmacy Cognitive stimu- tioners 3 pharmaco- n = 40 (4%) Nutritional sup- lation 13% males logical; 1 no Medication (not specified) plements Limit alcohol treatment option) n = 116 (11.6%) (not specified) Weight/BMI Take vitamins n = 293 (29.3%) Avoid polypharmacy n = 411 (41.1%) Healthy weight n = 457 (45.7%) Limit alcohol n = 591 (59.1%) Healthy diet n = 609 (60.9%) Socially active n = 667 (66.7%) Intellectual stimulation n = 802 (80.2%) Physically active n = 861 (86.1%) Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 675 Table 1. Continued Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Maeck Survey Ger- 159 family N/A N/A Case vignette— No: Survey asks if family phys- Ginkgo biloba N/A—not dis- et al. (38) (structured many physicians MCI Open-ended ques- icians would prescribe any (natural rem- cussed interview) (year = 1993) tions but would dementia related medi- edies) 70% males categorize an- cation to case vignette of Pentoxiphylline 122 fam- swers to facilitate someone with MCI who is (vascular man- ily physicians analysis at high risk of dementia: agement) (year = 2001)  = 1993 results Piracetam 56% males = 2001 results (nootropic) Yes (any treatment) Nimodipine (70.4%) (vascular man- (43.4%) agement) Ginkgo Biloba (34.0%) Memantine (23.0%) (anti-dementia Pentoxiphylline (13.2%) drug) (2.5%) Cholinesterase Piracetam (39%) inhibitors (3.3%) (anti-dementia Nimodipine (22.0%) drug) N/A Medication (not Memantine* N/A specified) (12.3%) Cholinesterase inhibitors N/A (8.2%) Other medication (44.7%) (0.8%) Suribhatla Survey UK 65 GPs N/A N/A Vascular cogni- Yes: Patient with vascular cogni- Statins N/A—not dis- et al. (39) % of sex not re- tive impairment Only discussed tive impairment (vascular man- cussed ported (VCI) prescription of 26% of GPs (16 out of agement) statins and no 61) would prescribe statins other strategies. to help manage vascular and cognitive risks Patients at risk of VCI 42% of GPs (27/64) felt that statins have a role in preventing VCI in at risk people Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 676 Family Practice, 2021, Vol. 38, No. 5 Table 1. Continued Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Werner Survey Israel 197 family phys- 21.9 years N/A MCI Yes % of family physicians Pharmacological Physical activity et al. (40) icians (SD = 10.4) 11 pharmaco- (n = 168) preferences for (not specified) Social activity 50.1 years old logical and non- treatment of MCI: Natural medi- Diet (SD = 9.2) pharmacological Physical activities (88%) cation (not spe- Cognitive 49% Male therapies Social activities (88%) cified) training Cognitive training (88%) Vitamins Relaxation/ Engagement in support (not specified) meditation group (80%) therapy Relaxation exercise (47%) Change in diet (44%) Psychotherapy (35%) Yoga or meditation (34%) Vitamins (34%) Pharmacological treatment (13%) Natural Medications (10%) Physician-observed management strategies Argimon- Descriptive Spain 105 general prac- N/A 921 pa- 45 diagnosed No preset list of % of service users receiving Nootropics (not N/A Pallas et al. naturalistic tices tients re- with MCI strategies treatment in response to specified) (37) study N/A ported to 157 cognitive cognitive impairment: Calcium (12 months) GP with impairment not After initial visit: antagonists memory dementia Any type of treatment (antihypertensive concerns Other groups: in response to cognitive drugs) Male 145 dementia impairment (type not speci- (74.9, 52 vascular fied) (76%) ±6.5) dementia Nootropics (24%) Female 73 Alzheimer’s Calcium antagonists (10%) (74.0, disease At 12 months: ±6.9) 137 psycho- Any type of treatment pathological in response to cognitive disorder impairment (type not speci- 25 other fied; 76%) 126 not stated Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 677 Table 1. Continued Author Study design Setting N of professionals Years in N of Type of Pre-determined Key findings/themes Type of Type of non- (year) [Age (M, SD)] practice patients, age cognitive list of strategies pharmacological pharmacological [M, SD] impairment recommendation recommendation Qualitative Physician-reported management strategies Ambigga Case report Ma- 1 primary care N/A N/A Case vignette— No preset list Preferences for treatment: Vascular medi- Physical activity et al. (45) laysia physician MCI Provided recom- promote independence in cation Social activity mendations based communication and activ- Diet on evidence ities of daily living Cognitive stimu- Mental exercise (e.g. puz- lation zles) Sleep Healthy lifestyle including Limit alcohol physical activity and diet Getting enough sleep Limit alcohol intake Control vascular risk fac- tors (e.g. hypertension) Regular follow up 3–6 months 16.7 years N/A Case vignette— No preset list Participants provided range Blood pressure Physical activity Hochhalter Focus USA 28 primary care (SD = n/a) MCI of pharmacological and (vascular man- Social activity et al. (44) groups, physicians 26.2 years non-pharmacological strat- agement) Diet semi- <44 years old (SD = n/a) egies. However, key findings Cholesterol Cognitive stimu- structured (53.6%) from study also outlined (vascular man- lation interviews 45–64 years old that some participants felt agement) (39.3%) that the management op- Reassessment of >65 years old tions are too generic and diabetes man- (7.1%) that in some cases dementia agement (not 79% male is not preventable. specified) 21 advanced prac- tice provider <44 years old (9.5%) 45–64 years old (90.5%) >65 years old (0%) 19% male M, mean; SD, standard deviation; N/A, not applicable (not provided). 1993 results. 2001 results. Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 678 Family Practice, 2021, Vol. 38, No. 5 with memory concerns using a preset list of pharmacological and non-pharmacological strategies. Across both studies, the top two recommendations were increasing physical activity and increasing cognitive stimulation (41,42). For physicians surveyed in Day et al., the third most common recommendation was for the patient to improve their diet. However, in Friedman et  al., physicians’ third highest recommendation to patients was to increase social stimula- tion. A small proportion, 40 physicians (4%) from Friedman et al., indicated that they would provide no advice for any treatment or strategies in preventing cognitive decline. Day et al. did not report if any physicians would not provide advice to patients with subjective memory concerns (please see Table 2). Mild cognitive impairment For patients presenting with MCI, three studies investigated primary care physicians’ intentions to provide non-pharmacological man- agement strategies. Werner et  al. (40) used a survey-based measure with 11 preset pharmacological and non-pharmacological strategies, which largely overlapped with DocStyles, but had some different strategies listed. Ambigga et al. (45) provide a case report and vignette on how primary care physicians should manage a patient with MCI. The final study, Hochhalter et al. ( 44), conducted a qualitative study using case vignettes in focus groups and semi-structured interviews. Across all three studies (40,44,45), four recommendations were high- lighted: physical activity, cognitive stimulation, social stimulation and diet. PCPs who participated in semi-structured interviews outlined the importance of recommending physical activity for a key reason ‘Vigorous daily exercise… because it improves, basically, all the vas- cular risks which people in this age group face’ [(44), p. 3]. The min- imum requirement of what is deemed enough physical activity, or for any of the other recommendations, is not outlined across any of the studies. Hochhalter et al. (44) also identified a small number of PCPs who did not provide any sort of management strategies because they felt that cognitive impairment, specifically Alzheimer’s disease, is not preventable ‘Stuff like Alzheimer’s, we can’t do anything about. Either you get it, or you don’t. You can’t prevent it’ [(44), p. 3]. Pharmacological management Pharmacological management for people presenting with either memory problems (SMC or MCI) was investigated by all nine studies, which has been outlined above in the Characteristics and quality of included studies section (please see Table 1 for study characteristics). Subjective memory concern For patients presenting with SMC, three studies investigated PCPs’ in- tentions for pharmacological management strategies (41–43) (please see Table 3). Both Day et  al. (41) and Friedman et  al. (42) used the Docstyles measure. Banjo et  al. (43) utilized a different method by using a case vignette of a patient with memory concerns and then asking how comfortable PCPs would be prescribing cognitive enhan- cers. Banjo et al. averaged the PCPs response to how comfortable they felt prescribing a cognitive enhancer (a preset list of sildenafil, methyl- phenidate and modafinil) on a Likert scale with 1 being ‘Less comfort- able’ and 7 being ‘More comfortable’. The PCPs felt most comfortable prescribing sildenafil. The only management response that appeared across all three studies was to provide no pharmacological response (41–43). Banjo et al. did not report the specific number of physicians providing advice but did report that some physicians did not provide any pharmacological response. A minimum of 1 in 5 physicians within Table 2. Physician behaviour of managing MCI and SMCs using a non-pharmacological response Non-pharmacological treatments - % of family practitioners who would provide treatment advice Study Meditation Sleep Reduce BMI Physical activity Limit alcohol Diet Social stimulation Cognitive No treatment stimulation MCI studies Ambigga et al. (45) N/R N/R N/R N/R N/R N/R – Hochhalter et al. (44) – – – N/R – N/R N/R N/R N/R Werner et al. (40) 57/168 – – 148/168 (88.0%) – 74/168 (44.0%) 148/168 (88.0%) 148/168 (88.0%) – (34.0%) SMC studies Day et al. (41) – – 550/972 (56.6%) 892/972 (91.8%) 626/972 (64.4%) 809/972 (83.2%) 775/972 (79.7%) 829/972 (85.3%) – Friedman et al. (42) – – 457/1000(45.7%) 861/1000 (86.1%) 591/1000 (59.1%) 609/1000 (60.9%) 667/1000 (66.7%) 802/1000 (80.2%) 40/1000 (4.0%) N/R = Numbers not reported but treatment advice still provided;  = treatment advice provided. BMI, body mass index. Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 679 Table 3. Physician behaviour of managing MCI and SMCs using a pharmacological response Pharmacological treatments—% of family practitioners who would provide treatment advice Study Anti-dementia Disease Natural remedies Vitamins/ New drugs Reduce Nootropics Vascular No drug treatment drugs management supplements (not specified) polypharmacy management MCI studies Ambigga et al. (45) – – – – – – – N/R – Hochhalter et al. – N/R – – – – – N/R N/R (44) a a a a a a Maeck et al. (38) 0/159 (0%) – 54/159 (34.0%) – 71/159 (44.7%) – 62/159 (39.0%) 56/159 (35.2%) 47/159 (29.6%) b b b b b b 25/122 (20.5%) 28/122 (23.0%) 1/122 (0.8%) 4/122 (3.3%) 3/122 (2.5%) 69/122 (56.6%) Suribhatla et al. (39) – – – – – – – 16/61 (26.0%) 45/61 (74.0%) Werner et al. (40) – – 17/168 (10.1%) 57/168 (33.9%) 22/168 (13.1%) – – – 72/168 (42.9%) Argimon-Pallas et al. – – – – 53.0% – 24.0% 10.0% 47.0% (37) SMC studies Banjo et al. (43) – – – – – – N/R – N/R Day et al. (41) – – – 332/972 (34.2%) 164/972 (16.9%) 435/972 (44.7%) – – 41/972 (4.2%) Friedman et al. (42) – – – 293/1000 (29.3%) 116/1000 (11.6%) 411/1000(41.1%) – – 180/1000 (18.0%) N/R = numbers not reported but treatment advice still provided;  = treatment advice provided. 1993 survey data. 2001 survey data. No figures given within study but is minimum amount of people who did not receive treatment. Figure based on adding all treatments options up; then taking that total away from 100. Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 680 Family Practice, 2021, Vol. 38, No. 5 Friedmann et al. and 1 in 20 physicians within Day et al. reported that type II diabetes) (44), natural remedies (such as Gingko Biloba) they would not provide any pharmacological response at all. These (38,40) and even anti-dementia drugs (38). In 2001, 122 PCPs in are minimum estimates as these figures are based on adding all treat- Germany (38) were given a case vignette of a patient with MCI who ment options up, then taking that total away from the study popu- has an increased risk of developing dementia. At that time, 12% of lation. However, within two studies, pharmacological response was PCPs (n = 15) would prescribe memantine and 8% (n = 10) would more frequent among physicians than no treatment at all. Reducing prescribe cholinesterase inhibitors to improve cognitive symptoms in polypharmacy was a management response to SMC being reported people with MCI (38). that just under half of physicians highlighted across two studies (41,42). Additionally, approximately a third of physicians in two Discussion studies also reported that they recommended the initiation of supple- ments and vitamins (41,42). However, the specific type of vitamins The review-highlighted PCPs were reporting that they were more and supplements were not specified. likely to provide non-pharmacological strategies than pharmaco- logical treatments. The three most common non-pharmacological Mild cognitive impairment strategies reported as being used to reduce cognitive decline and Five studies investigated PCPs’ intentions and one study investigated dementia risk in people with memory problems were (i) physical PCPs’ observed behaviour for pharmacological management strat- activity, (ii) cognitive stimulation and (iii) social stimulation (40– egies for patients presenting with MCI (please see Table 3). Across 42,44,45). Particular types of physical activity or cognitive and four of the five studies investigating reported management strategies, social stimulation were not specified. However, current evidence sug- physicians would not provide any pharmacological treatment in re- gests that not all types of physical activity are equally effective. For sponse to managing a patient with MCI. Maeck et al. surveyed phys- example, in a recent review, 4–6 months of aerobic exercise twice a icians in 1993 and 2001. In 1993, just under one in three physicians week or one to three times a week combining cognitive and motor reported that they would not typically provide any pharmacological challenges (Tai Chi, dance or dumbbell training) works to improve treatment. In comparison to 2001, just over one in two physicians memory and global cognitive functioning, but short-term resistance would not provide any pharmacological treatment. In a more recent training for less than 4 months did not improve memory or cognitive survey, Werner et al. also indicated that just under one in two phys- functioning (46–48). While there is less evidence in the arenas of cog- icians reported that they would not provide any pharmacological nitive and social activities, it appears that, in these domains too, not treatment. For physicians surveyed over the last 20  years, 43% to all activity types are equally effective (17,46–48). Other key strat- 74% would not prescribe any form of medication (38–40,44). If egies that physicians reported that they used included improving diet physicians were to advise on the use of pharmacological treatment, (40–42,44,45) and reducing alcohol intake (41,42,45). However, it is vascular management appeared the most common, being highlighted important to consider that all studies on non-pharmacological man- across four of the five studies (38,39,44,45). Vascular management agement evaluated self-reported (hypothetical) behaviours and none included any treatments aimed at lowering cholesterol, blood pres- observed actual behaviours. Additionally, three of the five studies sure and blood glucose in order to improve blood flow. One in four investigating non-pharmacological strategies used preset survey lists. physicians in Suribhatla et  al. reported that they would prescribe Therefore, these studies did not provide opportunity for physicians statins to manage vascular-related MCI. This was supported by a to outline other strategies they may implement. similar response rate of using vascular treatment management for For pharmacological treatment offered by PCPs for people with MCI by physicians surveyed in 1993 within the Maeck et al. study. memory problems, the most common across eight of the nine studies However, by 2001, this treatment strategy was reported by only was to provide no drug treatment. This appears to be in line with 3 physicians out of 122 surveyed. Two studies did not report the guidance for MCI management (49), which does not recommend number of physicians as one was a case report and the other was any drug treatments. Additionally, treatment for memory problems a qualitative study (44,45). Physicians within the focus groups out- is typically assessed and initiated by specialists in memory clinics or lined the importance of managing vascular risk factors not just for other secondary care services, which is common practice in coun- risk of conversion to dementia but also other health conditions that tries in North America, Europe and Oceania (50–52). However, it is could occur as a result of vascular disease (44). Only one study in important to consider that, within two studies investigating SMCs, the review (37) investigated observed natural behaviour rather than physicians were more frequently providing some pharmacological physicians’ reported management strategies. Argimon-Pallas et  al. responses, the most common responses being vascular risk man- (37) conducted a 12-month naturalistic descriptive study of 105 agement and vitamins. As for non-pharmacological approaches, the primary care centres across Spain and 202 patients who presented studies did not report the specific vascular management strategies with cognitive impairment. Of these patients, one in four were pre- used, and not all are equally effective. For example, insulin therapy scribed nootropics, which are drugs aimed at enhancing cognition has been associated with an increased risk of developing dementia, and can include piracetam (38), methylphenidate (43) and modafinil whereas thiazolidinedione exposure is associated with protective ef- (43). However, the type of nootropics prescribed in Argimon-Pallas fects and reduces the risk of dementia (14). Some evidence has indi- were not specified. One in 10 patients was prescribed calcium ant- cated that all classes of antihypertensives may have protective effects agonists, which are primarily used for treating hypertension but can for dementia with minimal difference in effect between classes (53). also be used for heart arrhythmia and headaches. This is a similar For vitamin or supplement management, low levels of vitamin D rate to the patients diagnosed with dementia within this study, but (54) or B vitamins (55) (B6, folate and B12) are typically associated Argimon-Pallas et  al. (37) did not provide analysis of any other with increased risk of dementia and are specific vitamin deficiencies comparator groups. that PCPs could address with minimal adverse effects. Other pharmacological strategies that PCPs reported they would Despite mixed evidence, the World Health Organization (48) use included prescription of vitamins (40), new drugs (type not spe- has set out a list of strategies for managing people at high risk of cified) (38,40), review of disease management medication (such as developing dementia that are appropriate for PCPs across the world Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 681 to deliver. This review has demonstrated that most PCPs’ reported Future research management strategies are adhering to most of the generic recom- Though self-report measures may provide some correspondence to mendations outlined in the WHO report. However, within the observed behaviour, there are still large discrepancies between self- included studies, there were some important omissions of manage- reported attitudes and actual observed behaviours (58,59). To gain a ment strategies that PCPs did not report as offering to people with more accurate reflection of primary care current management strat- memory problems. Depression, smoking and hearing loss are asso- egies for people with MCI or SMC, high-quality longitudinal observa- ciated with an increased risk of developing dementia, yet no study tional studies are needed. Observational studies can provide an insight or PCPs acknowledged this as an important strategy. Additionally, it into if people with memory problems are actively being managed is important to note that most of the included studies are reported differently than people who are cognitively healthy. Future research strategies from PCPs and, therefore, may not accurately portray be- should monitor both pharmacological and non-pharmacological de- haviours in observed practice. The only study to use a descriptive mentia prevention strategies offered by primary care. Research should naturalistic design, which was conducted in 2007, demonstrated that also capture the specific types of management strategies offered, such neurotropics (cognitive enhancers) were being prescribed more than as aerobic exercise or weight training for physical activity. is being recommended (37). This is perhaps surprising given the lack of evidence to suggest the effectiveness of neurotropics or acetyl- Conclusion cholinesterase inhibitors in people with MCI and SMC (56,57). In particular, acetylcholinesterase inhibitor prescription in MCI should The current review highlighted that when people are presenting with not be recommended due to many safety issues and minimal im- memory problems, primary care physicians will suggest that the pa- provement in cognition (57). tient can mitigate cognitive decline by improving physical activity, Primary care is in an optimal position to not only first identify cognitive stimulation, social stimulation and diet. Addressing hearing people with memory concerns and problems but also to coordinate loss, smoking and depression were not mentioned as strategies. For the management of risk after the patient is screened as having SMC MCI, most physicians report that they will not intend to prescribe or MCI. Therefore, it is important that PCPs advise people with any pharmacological treatments; but if they did, it would most likely memory problems on the modifiable health and lifestyle factors as- be to manage vascular risk factors. For SMC, there were physicians sociated with dementia, such as hypertension, depression, hearing across all three studies that provided no pharmacological treatment loss and the other nine factors identified in the Lancet commission at all. However, in two studies, physicians were more likely to reduce (4). By informing patients of these strategies, people with memory polypharmacy and increase vitamins than to provide no treatment at problems could reduce the risk of further cognitive decline or delay all. Most studies were surveys of subjective self-reported behaviours the onset of dementia. and there is a lack of strong evidence to accurately answer what are the current treatment responses for people with memory problems Limitations provided by PCPs. Future research using observational study designs is needed to obtain a more accurate reflection of actual current prac- There are some limitations to consider when interpreting the find- tice rather than reported practice. By understanding current prac- ings of this study. Due to heterogeneity in location, population and tices, research can optimize the management of cognitive decline and methods across different studies, we did not pool data across the dementia prevention in primary care. studies for a meta-analysis. We employed inclusive eligibility cri- teria in terms of study design, which allowed survey-based studies, qualitative interviews and observational studies to be synthesized to- Supplementary material gether. The included studies were conducted across a range of coun- Supplementary material is available at Family Practice online. tries, with different guidelines for practice, which may have impacted on the strategies reported by the PCPs. A  major limitation of all studies was that control groups were not used to compare how treat- Acknowledgements ment for an older patient at high risk of developing dementia might The authors are extremely grateful for the support provided in the development differ from an older patient with no memory problems. Therefore, of the search strategy by Jenni Ford, Evidence Services Librarian at Royal Free the percentage of people with memory problems who receive non- Library. The authors would also like to thank the numerous authors who were pharmacological recommendations, such as diet, physical activity contacted and kindly provided additional information regarding their studies. and social stimulation, may be the same percentage of older people who would anyway receive non-pharmacological recommendations Declaration as part of general health promotion advice or to treat other condi- Funding: Brendan Hallam is a PhD student funded by the Economic & Social tions. The lack of description, especially for pharmacological treat- Research Council’s London (UBEL) Doctoral Training Partnership, embed- ments, made it difficult to know the specific types of drugs used. For ded within the APPLE-TREE programme (Project reference: ES/S010408/1). example, Argimon-Pallas et al. (37) used the term nootropics, which Ethics approval: none. is a generic term for substances that aim to improve cognition, and Conflicts of interest: none declared. can range from caffeine to Ritalin. Other limitations in relation to the methodology of the current References review are only selecting English language studies. The current re- 1. World Health Organization. Dementia Factsheet: World Health Organ- view did not have the capacity or resources to translate non-English ization. 2020. Updated 21 September 2020. https://www.who.int/news- articles, which could introduce bias if potential key data from room/fact-sheets/detail/dementia (accessed on 14 January 2021). non-English articles are missed. Additionally, due to limited re- 2. World Health Organization. The Top 10 Causes of Death: World Health sources, the review also prioritized peer-reviewed articles to main- Organization. 2020. Updated 9 December 2020. https://www.who.int/ tain the scientific standard of the literature included in the review news-room/fact-sheets/detail/the-top-10-causes-of-death (accessed on 14 and excluded grey literature. January 2021). Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 682 Family Practice, 2021, Vol. 38, No. 5 3. Office for National Statistics. Deaths Registered in England and Wales: 25. National Institute for Health and Care Excellence. Dementia, Disability 2018. Office for National Statistics. 2019. https://www.ons.gov.uk/ and Frailty in Later Life-Mid-Life Approaches to Delay or Prevent Onset peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bul- (NG16). NICE Guideline. 2015. https://www.nice.org.uk/guidance/ng16 letins/deathsregistrationsummarytables/2018 (accessed on 23 June 2020). (accessed on 23 June 2020). 4. Livingston G, Huntley J, Sommerlad A et al. Dementia prevention, inter- 26. Petersen RC, Lopez O, Armstrong MJ et  al. Practice guideline update vention, and care: 2020 report of the Lancet Commission. Lancet 2020; summary: Mild cognitive impairment: report of the Guideline Develop- 396(10248): 413–46. ment, Dissemination, and Implementation Subcommittee of the American 5. Ngandu T, Lehtisalo J, Solomon A et al. A 2 year multidomain intervention Academy of Neurology. Neurology 2018; 90(3): 126–35. of diet, exercise, cognitive training, and vascular risk monitoring versus 27. Royal Australian College of General Practitioners. Guidelines for Pre- control to prevent cognitive decline in at-risk elderly people (FINGER): a ventive Activities in General Practice. 9th edn, updated. East Melbourne, randomised controlled trial. Lancet 2015; 385(9984): 2255–63. Australia: RACGP; 2018. 6. Viticchi G, Falsetti L, Buratti L et al. Framingham risk score and the risk 28. Wells  CE, Smith  SJ. Diagnostic care pathways in dementia: a review of of progression from mild cognitive impairment to dementia. J Alzheimers the involvement of primary care in practice and innovation. J Prim Care Dis 2017; 59(1): 67–75. Community Health. 2017; 8(2): 103–11. 7. Newson RS, Kemps EB. The nature of subjective cognitive complaints of 29. Ahmad S, Orrell M, Iliffe S, Gracie A. GPs’ attitudes, awareness, and prac- older adults. Int J Aging Hum Dev 2006; 63(2): 139–51. tice regarding early diagnosis of dementia. Br J Gen Pract 2010; 60(578): 8. Steinberg SI, Negash S, Sammel MD et al. Subjective memory complaints, e360–5. cognitive performance, and psychological factors in healthy older adults. 30. Godbee K, Gunn J, Lautenschlager NT, Curran E, Palmer VJ. Implementing Am J Alzheimers Dis Other Demen 2013; 28(8): 776–83. dementia risk reduction in primary care: a preliminary conceptual model 9. Gauthier S, Reisberg B, Zaudig M et al.; International Psychogeriatric As- based on a scoping review of practitioners’ views. Prim Health Care Res sociation Expert Conference on Mild Cognitive Impairment. Mild cogni- Dev 2019; 20: e140. tive impairment. Lancet 2006; 367(9518): 1262–70. 31. Moher  D, Liberati  A, Tetzlaff  J, Altman  DG; PRISMA Group. Preferred 10. Lopez OL, Kuller LH, Becker JT et al. Incidence of dementia in mild cog- reporting items for systematic reviews and meta-analyses: the PRISMA nitive impairment in the cardiovascular health study cognition study. Arch statement. PLoS Med 2009; 6(7): e1000097. Neurol 2007; 64(3): 416–20. 32. Vaportzis E, Gow AJ. People’s beliefs and expectations about how cogni- 11. Stewart R. Subjective cognitive impairment. Curr Opin Psychiatry 2012; tive skills change with age: evidence from a UK-wide aging survey. Am J 25(6): 445–50. Geriatr Psychiatry 2018; 26(7): 797–805. 12. Mitchell AJ, Beaumont H, Ferguson D, Yadegarfar M, Stubbs B. Risk of de- 33. Noyes  J, Booth  A, Moore  G, Flemming  K, Tunçalp  Ö, Shakibazadeh  E. mentia and mild cognitive impairment in older people with subjective memory Synthesising quantitative and qualitative evidence to inform guidelines on complaints: meta-analysis. Acta Psychiatr Scand 2014; 130(6): 439–51. complex interventions: clarifying the purposes, designs and outlining some 13. Tully  PJ, Hanon  O, Cosh  S, Tzourio  C. Diuretic antihypertensive drugs methods. BMJ Glob Health 2019; 4(Suppl 1): e000893. and incident dementia risk: a systematic review, meta-analysis and meta- 34. Hong QN, Fàbregues S, Bartlett G et  al. The Mixed Methods Appraisal regression of prospective studies. J Hypertens 2016; 34(6): 1027–35. Tool (MMAT) version 2018 for information professionals and researchers. 14. McMillan JM, Mele BS, Hogan DB, Leung AA. Impact of pharmacological Educ Inf 2018; 34(4): 285–91. treatment of diabetes mellitus on dementia risk: systematic review and 35. Pace  R, Pluye  P, Bartlett  G et  al. Testing the reliability and efficiency of meta-analysis. BMJ Open Diabetes Res Care 2018; 6(1): e000563. the pilot Mixed Methods Appraisal Tool (MMAT) for systematic mixed 15. Dupré C, Bongue B, Helmer C et al. Physical activity types and risk of de- studies review. Int J Nurs Stud 2012; 49(1): 47–53. mentia in community-dwelling older people: the Three-City cohort. BMC 36. Souto RQ, Khanassov V, Hong QN, Bush PL, Vedel I, Pluye P. Systematic Geriatr 2020; 20(1): 132. mixed studies reviews: updating results on the reliability and efficiency 16. Choi D, Choi S, Park SM. Effect of smoking cessation on the risk of de- of the Mixed Methods Appraisal Tool. Int J Nurs Stud 2015; 52(1): mentia: a longitudinal study. Ann Clin Transl Neurol 2018; 5(10): 1192–9. 500–1. 17. Fratiglioni  L, Marseglia  A, Dekhtyar  S. Ageing without dementia: can 37. Argimón  Pallàs  JM, Riu  Subirana  S, Lizán  Tudela  L, Badia  Llach  X, stimulating psychosocial and lifestyle experiences make a difference? Martínez Lage JM. [Initial treatment of patients with subjective memory Lancet Neurol 2020; 19(6): 533–43. complaints and/or cognitive impairment in primary care: ISSEA Study]. 18. Amieva H, Ouvrard C. Does treating hearing loss in older adults improve Aten Primaria 2007; 39(4): 171–7. cognitive outcomes? A review. J Clin Med 2020; 9(3): 805. 38. Maeck L, Haak S, Knoblauch A, Stoppe G. Primary care physicians’ at- 19. Danat  IM, Clifford  A, Partridge  M et  al. Impacts of overweight and titudes related to cognition enhancers in early dementia: a representa- obesity in older age on the risk of dementia: a systematic literature review tive eight-year follow-up study in Lower Saxony, Germany. Int J Geriatr and a meta-analysis. J Alzheimers Dis 2019; 70(s1): 87–99. Psychiatry 2008; 23(4): 415–21. 20. Chan JY, Yiu KK, Kwok TC, et al. Depression and antidepressants as po- 39. Suribhatla S, Dennis MS, Potter JF. A study of statin use in the prevention tential risk factors in dementia: a systematic review and meta-analysis of of cognitive impairment of vascular origin in the UK. J Neurol Sci 2005; 18 longitudinal studies. J Am Med Dir Assoc 2019; 20(3): 279–86. e1. 229–230: 147–50. 21. Sabia S, Fayosse A, Dumurgier J et  al. Alcohol consumption and risk of 40. Werner  P, Heinik  J, Kitai  E. Familiarity, knowledge, and preferences of dementia: 23 year follow-up of Whitehall II cohort study. BMJ 2018; 362: family physicians regarding mild cognitive impairment. Int Psychogeriatr k2927. 2013; 25(5): 805–13. 22. Andrieu  S, Guyonnet  S, Coley  N et  al.; MAPT Study Group. Effect of 41. Day KL, Friedman DB, Laditka JN, et al. Prevention of cognitive impair- long-term omega 3 polyunsaturated fatty acid supplementation with or ment: physician perceptions and practices. J Appl Gerontol. 2012; 31(6): without multidomain intervention on cognitive function in elderly adults 743–54. with memory complaints (MAPT): a randomised, placebo-controlled trial. 42. Friedman DB, Rose ID, Anderson LA et  al. Beliefs and communication Lancet Neurol 2017; 16(5): 377–89. practices regarding cognitive functioning among consumers and primary 23. Cooper C, Aguirre E, Barber JA, et al. APPLE‐Tree (Active Prevention in care providers in the United States, 2009. Prev Chronic Dis 2013; 10: E58; People at risk of dementia: Lifestyle, bEhaviour change and Technology quiz 8–13. to REducE cognitive and functional decline) programme: protocol. Int J 43. Banjo OC, Nadler R, Reiner PB. Physician attitudes towards pharmaco- Geriatr Psychiatry 2019. logical cognitive enhancement: safety concerns are paramount. PLoS One 24. Yaffe  K, Barnes  DE, Rosenberg  D et  al. Systematic multi-domain 2010; 5(12): e14322. Alzheimer’s risk reduction trial (SMARRT): study protocol. J Alzheimers 44. Hochhalter AK, Bryant LL, Hunter R et al. Multisite qualitative study of Dis 2019; 70(s1): 207–20. primary care physicians’ and midlevel providers’ self-reported practices Downloaded from https://academic.oup.com/fampra/article/38/5/669/6256039 by DeepDyve user on 20 July 2022 Primary care management of memory concerns 683 and perceptions about maintaining cognitive health. Prev Chronic Dis 52. OECD. Care Needed: Improving the Lives of People With Dementia. 2012; 9: E169. Paris: OECD Publishing; 2018. 45. Ambigga D, Suthahar A, Ramli A, Ng K, Radziah A, Marymol K. Diagnosis 53. Murray  MD, Hendrie  HC, Lane  KA et  al. Antihypertensive medication and management of mild cognitive impairment in the community: what is the and dementia risk in older adult African Americans with hypertension: a role of primary care physician? Malays Fam Physician 2011; 6(2–3): 74–8. prospective cohort study. J Gen Intern Med 2018; 33(4): 455–62. 46. Whitty E, Mansour H, Aguirre E et  al. Efficacy of lifestyle and psycho- 54. Sommer I, Griebler U, Kien C et al. Vitamin D deficiency as a risk factor social interventions in reducing cognitive decline in older people: system- for dementia: a systematic review and meta-analysis. BMC Geriatr 2017; atic review. Ageing Res Rev 2020; 62: 101113. 17(1): 16. 47. Alty  J, Farrow  M, Lawler  K. Exercise and dementia prevention. Pract 55. Lefèvre-Arbogast S, Féart C, Dartigues J-F et al. Dietary B vitamins and a Neurol 2020; 20(3): 234–40. 10-year risk of dementia in older persons. Nutrients 2016; 8(12): 761. 48. World Health Organization. Risk Reduction of Cognitive Decline and De- 56. Tricco AC, Soobiah C, Berliner S et al. Efficacy and safety of cognitive en- mentia: WHO Guidelines. France: World Health Organization; 2019. https:// hancers for patients with mild cognitive impairment: a systematic review apps.who.int/iris/bitstream/handle/10665/312180/9789241550543-eng. and meta-analysis. CMAJ 2013; 185(16): 1393–401. pdf?ua=1 (accessed on 17 July 2020). 57. Matsunaga S, Fujishiro H, Takechi H. Efficacy and safety of cholinesterase 49. NICE CKS. Dementia, Scenario: Suspected Dementia. 2019. [2 April 2020]. inhibitors for mild cognitive impairment: a systematic review and meta- https://cks.nice.org.uk/dementia#!scenario (accessed on 25 June 2020). analysis. J Alzheimers Dis 2019; 71(2): 513–23. 50. Jolley D, Benbow SM, Grizzell M. Memory clinics. Postgrad Med J 2006; 58. Eccles MP, Hrisos S, Francis J et al. Do self-reported intentions predict 82(965): 199–206. clinicians’ behaviour: a systematic review. Implement Sci 2006; 1(1): 51. Gruters  AAA, Ramakers  IHGB, Kessels  RPC et  al. Development of 28. memory clinics in the Netherlands over the last 20  years. Int J Geriatr 59. Sheeran P, Webb TL. The intention–behavior gap. Soc Pers Psychol Com- Psychiatry 2019; 34(8): 1267–74. pass. 2016; 10(9): 503–18.

Journal

Family Practice – Oxford University Press

Published: Sep 25, 2021

Keywords: dementia; primary health care; memory; minimal cognitive impairment; physical activity; memory impairment; cognitive impairment; pharmacology; cognitive stimulation; risk reduction; pharmacotherapy

There are no references for this article.