Skip to main content

Table 1 Study characteristics

From: A systematic review and meta-analysis of studies on screening for mild cognitive impairment in primary healthcare

ID Sample Size Country/income level Screening tool Age Mode of delivery Sensitivity and specificity
Arabi 2013 160 Malaysia/ upper-middle EDQ and MMSE 65 >  Face-to-face or telephone interview with the patient and an informed person. The highest score was entered in the analysis A score above 8 for EDQ and a score of 21 or lower for MMSE were considered as the criteria for diagnosing cognitive impairment
Arabi 2016 200 Malaysia/ upper-middle EDQ and MMSE 60 Face-to-face and telephone interviews with patients and knowledgeable people around A score of 6 was considered for EDQ for 95.5% sensitivity and 84.2% for specificity and a score of 21 for MMSE was considered for dementia diagnosis
Brodaty 2016 [20] 1717 Australia/ high GPCOG, MMSE, CAMCOG 75 Nurses examined patients with MMSE and GPCOG, and then specialists used CAMCOG and GDS to evaluate patients at a later stage A score below 5 indicates cognitive impairment and a score above 8 indicates a low probability of developing cognitive impairment. A score of 5 to 8 indicates an uncertain situation that requires assessment with GPCOF. A score of 11/10 out of 15 meant cognitive impairment in males. A score of 23 out of 30 is also considered cognitive impairment for MMSE
Chan 2016 [31] 309 Singapore/ high AD8, MMSE, MOCA 60 Screening was done in two stages. In the first stage, psychiatrists evaluated the patient and in the second stage, a panel of specialists AD8 with a cut-off point of 4.8 with a mean area below the curve of 0.97 and a sensitivity of 0.91 and a characteristic of 0.91 showed the best detection accuracy
Eichler 2015 [26] 4046 Germany/ high DemTec vs MMSE 70 Demtec was first used by a GP for screening, then MMSE was used The MMSE score was used to classify the cognitive impairment
27–30%: No disorder, 20–26%: Mild disorder, 10–19% Moderate disorder, 0–19%: Severe cognitive impairment
Grober 2014 [38] 112 US/ high MMSE, screening 65 Screening was done in two stages. First, the individual's cognitive status was assessed using MMSE. If the individual's cognitive impairment was not diagnosed, a complete assessment using pFCSRT + IR was used An MMSE score of 23 or higher was considered
Grober 2017 [39] 563 US/ high IQCODE- 65 In the first stage, knowledgeable people completed the IQCODE short questionnaire. If a person was diagnosed with severe dementia, a complete evaluation was performed with pFCSRT-IR and DSM-IV The diagnosis of cognitive impairment and dementia was made based on the opinion of a psychiatrist and a geriatrician
Grober 2019 [37] 257 US/ high IQCOD and pFCSRT   Screening is done in two ways based on the patient and based on the informed caregiver and also in two stages. Initially, IQCOD was used if the person was accompanied and PBS was used if they were not. If the person is positive in the first stage, he / she enters the second stage and is evaluated through pFCSRT. If the person is negative, the screening will be repeated one year later. If the result of the second stage is positive, treatment is started and follow-up is done. If it is negative, the screening is repeated one year later The decision criterion based on IQCODE: greater than equal to 3.5. The decision criterion in BPS is: MIS < 5 or AF < 9. In the second stage, if FR < 25 or TR < 46, the person is recognized as positive
Koc Okudur 2019 [36] 357 Turkey/ upper-middle MMSE and RCS-T 60 First the evaluation was done based on MMSE and then the complete evaluation was done with RCS-T. The evaluation was performed by a general practitioner The RCS score is from 0 to 10
A score of less than 4 was considered for the diagnosis of Alzheimer's and less than 6 for the diagnosis of mild cognitive impairment
Larner 2018 [23] 676 England/high MMSE, MOCA and DSM-IV 65 The evaluation was based on MMSE and MOCA and in the second stage, based on the opinion of experts, the cognitive status of individuals was determined through DSM-V Intersection points 24 and 26 were considered for MMSE
Latraki 2017 [28] 319 Greece/ high TYM GPCog, MMSE 60–89 First TYM then GPCOG was used to diagnose cognitive impairment TYM: The highest score is equal to 50, with two cutting points of 39.38 or 36.35. GPCOG: The maximum score is 9, with a cut-off point of 7
MMSE: 23 out of 30 as cutting point
Pandhita 2019 212 Indonesia/ upper-middle CERAD 60 Used a decision tree model to identify cognitive impairment. The information entered in the model is based on CERAD, OLB and fast cognition assessment such as clock drawing, verbal test Based on standard scores of WAHYU, VFT, SMC tests
Petrazzuoli 2014 [30] 121 Italy/ high AQT and MMSE 45–90 Screening during routine referral to primary care was performed by a general practitioner AQT sets different range for diagnosis (in seconds) for different age groups. No cutting point was reported for MMSE
Salami 2019 [29] 114 Iran MMSE and TYM 80 Participants passed a physical examination and completed forms of the MMSE and TYM tests The MMSE test had AUC = 0.991, sensitivity = 0.90 and specificity = 0.96,
Shaaban 2013 [35] 49 Malaysia/ upper-middle M_RUDAS, MMSE, ECAQ 65 Screening by a family physician and a trained expert using
M_RUDAS, M_MMSE and M_ECAQ. Clinical interview was conducted by a psychiatrist using DSM IV
MMSE: Cutting point 17 and less
ECAQ: Cutting point 5/10 and less
M-RUDAS: > 23
Stein 2015
6619 Germany/ high SMSE, MMSE 75 MMSE with 30 questions and SMSE with 6 questions have been used by general practitioners. Additional evaluation by psychiatrists using DSM – III – R,
ROC and AUC have been used as measurement accuracy criteria. Cutting point as follows
MMSE < 24
Teixeira 2017 [32] 436 Portuguese/ high MMSE and Global -GDS +  65 Patients are evaluated by a general practitioner or nurse and then caregivers of dementia patients are evaluated Patients were divided into six groups based on the GDS result:
Very mild cognitive impairment / Mild disorder / Moderate disorder / Moderate to severe disorder / Very severe disorder
Thyrian 2016 [25] 1167 Germany/upper-middle MMSE vs Neuro psychiatric Inventory (NPI) 70 Evaluation of patients using MMSE by general practitioners and evaluation of their psychological status by psychiatrists by NPI Classification of patients based on MMSE:
20–30%: Irrelevant or mild
10–19%: Medium
0–9%: Severe cognitive impairment
An NPI score above 5 indicates a diagnosis of cognitive impairment
Xue 2017 [22] 2731 China/ upper-middle SIS, (MMSE) 60 Screening was performed by trained health workers MMSE: For education 0 to 5:17, Education 6 to 10 years: 20, And higher education than 10:24 as the cutting point, The cut-off point for SIS was set to 4
Yang 2014 [21] 733 China/ upper-middle MMSE, MOCA versus SE + MOCA 60 The initial assessment was performed by nurses and face to face. MMSE and MoCA were used in combination to diagnose cognitive impairment Cut points for MMSE (85.2 sensitivity and 92.75 specification, Illiterate: 17 out of 18, Up to 6 classes: 20 out of 21, More than 6 classes: 24 out of 25, Cut points for MoCA, Illiterate: 13 out of 14, Up to 6 classes: 19 out of 20, More than 20 classes: 24 out of 25
Zaganas 2020
314 Greece/ high MMSE 60–100 Interviews were conducted by a trained nurse. Psychological assessment was performed by a trained psychiatrist who assessed the cognitive status of patients for more than 2.5 h The cut-off point for the Greek version of MMSE was 24.23