TABLE 1.
Citation characteristics
| First author, year (n) | Sample, setting and study type | Baseline age (years [SD]) a | Total follow‐up (years) | Mean (SD) or median (IQR) follow‐up in years a | Psychiatric disorder or clinically significant symptoms b , c | Dementia diagnosis (N cases) c , d | Quality e | Summary of findings on association between psychiatric disorder and dementia f |
|---|---|---|---|---|---|---|---|---|
| Anxiety (n = 6) | ||||||||
| de Bruijn 2014 | Rotterdam Study, The Netherlands, prospective | Sample 1: 68.6 (8.5), sample 2: 75.5 (6.2) | 18 | Sample 1: 11.8 (5)Sample 2: 5.8 (1.9) | Sample 1: HADS ≥8 | Dementia (DSM‐III‐R, n = 358); AD (NINCDS‐ADRDA, n = 291) | Good | No evidence of association between anxiety disorder and dementia (HR: 0.81, 95% CI: 0.50–1.30) |
| Sample 1 n = 2708 | Sample 2: DSM‐IV | |||||||
| Sample 2 n = 3069 | ||||||||
| Gallacher 2009 n = 1160 | Caerphilly Prospective Study, Wales, prospective | 56.1 (4.4) | 20 | 17.3 (1.3) | STAI≥31 | Dementia (DSM‐IV, n = 69) | Good | Anxiety associated with increased odds of dementia (OR: 2.89, 95% CI: 1.27–6.54) |
| Mortamais 2018 n = 5234 | Three‐City (3C) Study, France, prospective | 73.4 (5.2) | 10 | / | STAI≥44 | Dementia (DSM‐IV, n = 378)AD (NINCDS‐ADRDA, n = 259), VaD (NINDS‐AIREN, n = 23) | Good | High trait anxiety associated with a higher rate of dementia (HR: 1.26, 95% CI: 1.01–1.57) |
| Petkus 2016 n = 1082 | Swedish adoption Twin Study of Ageing, prospective | 60.86 (11.15) | 28 | Years to dementia: 14.65 (6.7) | 1 + standard deviation above mean STPI | Dementia (DSM‐III‐R/IV, n = 172) | Good | Anxiety associated with increased risk of dementia after adjustment for confounders, including depression (HR: 1.48; 95% CI: 1.01–2.18) |
| Santabárbara 2019 n = 4057 | Zaragoza Dementia and Depression study (ZARADEMP), Spain, prospective | No AD: 72.83 (9.03) | 4.5 | Median: 4.4 (IQR: 3.0–4.9) | GMS AGECAT ≥3 | AD (DSM‐IV, n = 87) | Poor | AD associated with previous anxiety (SHR: 3.9, 95% CI: 1.59–9.6), but not anxiety ‘subcases’ (SHR: 1.19, 95% CI: 0.75–1.88). |
| AD: 83.72 (7.13) | ||||||||
| Santabárbara 2020 n = 4057 | ZARADEMP, Spain, prospective | VaD (men: 80.3 (8.2), women: 79.8 (7.9)); No VaD (men: 71.7 (9.1), women: 72.3 (9.1)) | 4.5 | Median: 4.4 (IQR: 3.0–4.9) | GMS AGECAT ≥3 | VaD (DSM‐IV, n = 14) | Poor | Anxiety associated with VaD in men (IRR: 3.24, 95% CI: 1.13–9.35), but not women (IRR: 0.68, 95% CI: 0.19–2.23) |
| Post‐traumatic stress disorder (n = 3) | ||||||||
| Flatt 2018 n = 499,844 | Kaiser permanente Northern California health system, United States, register‐based | 71.1 (7.9) | 13 | 8 (4.6) | ICD‐9: 309.81 | Dementia (ICD‐9, n = 59,127) | Good | PTSD associated with higher rate of dementia (female HR: 1.59, 95% CI: 1.3–1.95; male HR: 1.96, 95% CI: 1.51–2.55) |
| Gradus 2019 n = 279,188 | Danish population register | 51 | 17 | Stress cohort median: 6.1; Comparison: 6.8 | Stress disorders: ICD‐10: F43.x | Dementia (ICD‐10, stress cohort n = 1364) | Good | Those with PTSD had 2 times the rate of dementia compared to those without PTSD (95% CI: 1.3–3.2) |
| Wang 2016 n = 8750 | Nationwide Health Insurance Research Database (NHIRD), Taiwan, register‐based | PTSD: 55.44 (9.20)No PTSD: 55.42 (9.22) | 11 | Years to dementia: PTSD: 4.68 (2.31); No PTSD: 6.96 (2.66) | ICD‐9‐CM: 309.81 | Dementia (ICD‐9‐CM, n = 135) | Good | PTSD (HR: 4.37, 95% CI: 2.53–7.55) and depression (HR: 2.16, 95% CI: 1.28–3.66) associated with subsequent dementia, with a dose‐response relationship by PTSD severity |
| Non‐affective psychotic disorder (n = 7) | ||||||||
| Almeida 2018 n = 38,173 | Health in Men Study (HIMS), Western Australia, prospective | 72.5 (4.6) | 17.7 | / | Non‐affective psychotic disorders: ICD‐8/9: 295, 297; ICD‐10: F20, F22, F23, F25, F28, F29 | Dementia (ICD‐8/9/10, n = 8068) | Good | Psychotic disorders associated with subsequent dementia (SHR: 2.67, 95% CI: 2.3–3.09). Stronger association for shorter duration of psychosis. No variation by age‐at‐onset |
| Kodesh 2020 n = 94,120 | Health maintenance organisation data, Israel, register‐based | 68.9 (7.1) | 4.83 | / | VLOSLP: ICD‐9: 295–299; ICD‐10: F20–F29 | Dementia (ICD‐9/10, n = 6026) | Good | Very late‐onset schizophrenia associated with subsequent dementia (HR: 2.67, 95% CI: 1.82–3.91). |
| Kørner 2009 n = 12,616 | Danish population register | Median LOS: 53.56, comparison: 65.13, VLOSLP and comparison: 71.19 | 4.58 | Median (IQR) LOS: 3.15 (1.56–3.50)VLOSLP: 3 (1.25–4.78) | LOS and VLOSLP: ICD‐10: F20‐F20.9 | Dementia (ICD‐10, LOS n = 20, comparison n = 160; VLOSLP n = 18, comparison n = 157). | Good | LOS and VLOSLP had higher dementia rates than osteoarthritis patients (LOS rate ratio (RR): 3.47, 95% CI: 2.19–5.5; VLOSLP RR: 3.15, 95% CI: 1.93–5.14) and general population (LOS RR: 2.36, 95% CI: 1.54–3.62; VLOSLP RR: 2.21, 95% CI: 1.39–3.5). |
| Lin 2018 n = 30,200 | NHIRD, Taiwan, register‐based | / | 14 | Schizophrenia: 9.4 (11.8), control: 9.5 (10.1) | Schizophrenia: ICD‐9‐CM: 295 | Dementia (ICD‐9‐CM, n = 1237), AD (ICD‐9‐CM, n = 147), VaD (ICD‐9‐CM, n = 162) | Good | Patients with schizophrenia had higher rates of dementia (HR: 2.01, 95% CI: 1.42–2.59), AD (HR: 2.27, 95% CI: 1.99–3.51), and VaD (HR: 2.01, 95% CI: 1.36–2.22) than controls |
| Ribe 2015 n = 2,845,440 | Danish population register | 58.7 (11.2) | 18 | 11 (6) | Schizophrenia: ICD‐8/9: 295 (except 295.79); ICD‐10: F20; Schizoaffective disorder: ICD‐8/9: 295.79, 296.8; ICD‐10: F25 | Dementia (ICD‐8/9/10, n = 136,012) | Good | Schizophrenia associated with incident dementia (IRR: 1.71, 95% CI: 1.6–1.82) |
| Stafford 2021 n = 169,499 | Swedish population register | 70.31 (7.2) | 30 | 8.6 (6.68) | VLOSLP: ICD‐10: F20‐F29 (or ICD‐8/9 equivalent) | Dementia (ICD‐8/9/10, n = 13,610) | Good | VLOSLP associated with higher rate of dementia (HR: 4.22, 95% CI: 4.05–4.41). Association attenuated over time but was present for up to 20 years |
| Stroup 2021 n = 8,011,773 | Medicare beneficiaries, US, register‐based | 74 (8.2) | 11 | / | Schizophrenia: ICD‐8/9/10: F25 | Dementia (ICD‐9/10, n = 1,129,646) | Good | Schizophrenia associated with elevated rate of dementia (age 66 dementia rate – schizophrenia: 52.5 (95% CI: 50.1–54.9), controls: 4.5 (95% CI: 4.4–4.6) per 1000 person‐years at‐risk |
| Bipolar disorder (BPD) (n = 4) | ||||||||
| Almeida 2018 n = 38,173 | Western Australian Data Linkage System, register‐based | 72.5 (4.6) | 17.6 | 12.8 (5.3) | ICD‐8: 296.1, 296.3; ICD‐9: 296.0, 296.1, 296.4, 296.5, 296.6, 296.7, 296.80, 296.81; ICD‐10: F30, F31 | Dementia (ICD‐8/9/10, n = 423) | Good | Late‐onset (≥60 years HR: 2.99, 95% CI: 2.17–4.12) and younger onset BPD associated with dementia (<60 years HR: 2.31, 95% CI: 1.77–3.01) |
| Almeida 2016 n = 37,768 | HIMS, Western Australia, prospective | 72.5 (4.6) | 13 | / | ICD‐8: 296.1 and 296.3; ICD‐9: 296.0, 296.1, 296.4, 296.5, 296.6, 296.7, 296.80, 296.81 | Dementia (ICD‐8/9, n = 4925) | Good | BPD associated with higher dementia rates (HR: 2.3, 95% CI: 1.80–2.94). Stronger association for shorter duration of BPD, or illness onset after 70 years old |
| Lin 2020 n = 102,675 | NHIRD, Taiwan, register‐based | BPD: 55.31 (8.48), comparison: 55.25 (8.55) | 9 | BPD: 6.56 (2.97); comparison: 6.86 (2.85) | ICD‐9‐CM: 296 except 296.2x, 296.3x, 296.9x, and 296.82 | Dementia (ICD‐9‐CM, n = 2122), AD (ICD‐9‐CM, n = 1353), VaD (ICD‐9‐CM, n = 447) | Good | BPD associated with incident dementia (HR: 7.52, 95% CI: 6.86–8.25), AD (HR: 13.16, 95% CI: 11.58–14.96) and VaD (HR: 5.5, 95% CI: 4.53–6.69) |
| Wu 2013 n = 64,804 | NHIRD, Taiwan, register‐based | 74.1 (8.6) | 9 | / | ICD‐9‐ CM: 296.0x, 296.1x, 296.4x, 296.5x, 296.6x, 296.7x, 296.80, 296.81, 296.89 | Dementia (ICD‐9‐CM, n = 9304) | Good | BPD associated with increased odds of dementia (OR: 4.32, 95% CI: 3.21–5.82) |
| Multiple psychiatric diagnoses (n = 4) | ||||||||
| Chen 2015 n = 4582 | NHIRD, Taiwan, register‐based | Control: 65.34 (7.47); major depressive disorder: 65.45 (7.53); BPD: 64.72 (7.08) | 13 | Years to dementia: controls: 5.62 (3.23), MDD: 4.01 (3.02); BPD: 4.38 (3.60) | MDD: ICD‐9‐CM: 296.2x, 296.3x; BPD: ICD‐9‐CM: 296.0x, 296.1x, 296.4x, 296.5x, 296.6x, 296.7x, 296.80, 296.81, 296.89 | Dementia (ICD‐9‐CM, n = 547) | Good | BPD (HR: 5.58, 95% CI: 4.26–7.32) and MDD (HR: 3.02, 95% CI: 2.46–3.7) associated with higher rates of dementia |
| Kessing 1999 n = 13,852 | Danish population register | Neurosis: 41.2 (13.8), depression: 51.3 (16.0), schizophrenia: 31.7 (15.4) | 24.7 | Median: 21.6 | Depressive episode: ICD‐8: 296.09, 296.29, Schizophrenia: ICD‐8: 295, Neurosis: ICD‐8: 300 | Dementia (ICD‐8/9/10, n = 424) | Poor | Rates of dementia were increased 14.7‐fold (95% CI: 9.1–22.4) in those with schizophrenia, 13.7‐fold (95% CI: 12.1–15.4) for affective disorder, and 11.2‐fold (95% CI: 9.6–12.9) for neurosis |
| Tapiainen 2017 n = 55,896 | Medication use and Alzheimer's disease study, Finland, register‐based nested case‐control | 79.7 (6.8) | 33 | Time to AD: AD cases: 18.1 years (8.9), controls: 19.4 (9.0) | Psychotic disorders: ICD‐10: F20–F29; mood disorders: ICD‐10: F32–F39; neurotic disorders: ICD‐10: F40–F48 | AD (NINCDS‐ADRDA and DSM‐IV, n = 27,948) | Good | Depression associated with increased odds of AD with 5‐year (OR: 1.17, 95% CI: 1.05–1.3), but not 10‐year interval between diagnoses (OR: 1.08, 95% CI: 0.96–1.23). Anxiety and psychotic disorders were not associated with AD regardless of interval |
| Zotcheva 2018 n = 28,916 | The Nord‐Trøndelag Health Study, Norway, prospective | Midlife moderate‐to‐vigorous physical activity (MVPA): 52.3,No MVPA: 54.6 | 16.3 | 15.2 | Psychological distress: ADI‐4 ≥88th percentile | Dementia (ICD‐10, n = 920) | Good | Psychological distress associated with higher dementia rates (HR: 1.34, 95% CI: 1.03–1.74) |
| Depression (n = 33) | ||||||||
| Almeida 2017 n = 4922 | HIMS, Western Australia, prospective | 77.2 (3.7) | 14.3 | 8.9 | Short version GDS‐15 score ≥7 | Dementia (ICD‐8/9/10, n = 903) | Good | Depression associated with dementia (past depression SHR: 1.3, 95% CI: 1.0–1.6, baseline SHR: 1.5, 95% CI: 1.2–2.0) |
| Becker 2009 n = 288 | Cardiovascular Health Study‐Cognition Study, US, prospective | 77.52 (3.65) | 9 | 7.1 | CES‐D score ≥10 | Dementia (DSM‐IV, n = 48) | Poor | No evidence of association between persistent depression and dementia (HR: 1.33, 95% CI: 0.49–3.65) |
| Berger 1999 n = 222 | The Kungsholmen Project, Sweden, prospective | Incident AD: 85.53 (4.95); no dementia: 83.18 (4.77) | 3 | 3.08 (0.58) | CPRS | AD (DSM‐III‐R, n = 34) | Poor | Depression associated with subsequent AD, particularly for motivation‐related symptoms (OR: 1.4, 95% CI: 1.04–1.89) |
| Buntinx 1996 n = 19,103 | Family‐practice‐based registration network, The Netherlands, register‐based | 50+ | 10 | / | ICPC‐classification: P76 | Dementia (ICPC: n = 137) | Poor | Depression among older people associated with increased risk of dementia (HR: 2.55, 95% CI: 1.19–5.47) |
| Chan 2020 n = 16,725 | NHIRD, Taiwan, register‐based | 41.5 (15.9) | 12 | Time to dementia: control: 10.71 (1.16); MDD: 10.45 (1.86) | ICD‐9‐CM: 296.2x and 296.3x | Dementia (ICD‐9‐CM, n = 508), AD (ICD‐9‐CM, n = 458) | Good | Patients with MDD had increased risk of dementia and AD. Risk highest among difficult‐to‐treat patients |
| Chen 2008UK n = 3,341,China n = 1254 | MRC–Ageing in Liverpool Project Health Aspects (MRC‐ALPHA), UK; Hefei cohort, China | 65+ | MRC‐alpha: 2‐4Hefei‐ 1 | / | GMS‐AGECAT score ≥3 | Dementia (GMS‐AGECAT: Hefei n = 75, MRC‐ALPHA, n = 382 | Poor | Incident dementia associated with severe, but not milder, depression (Hefei cohort HR: 5.44, 95% CI: 1.67–17.8; MRC‐ALPHA, 4‐year follow‐up HR: 2.62, 95% CI: 1.18–5.8). |
| Deckers 2018 n = 278 | The Cambridge City over‐75s cohort (CC75 C), UK, prospective | 87.8 (3.1) | 18 | / | CAMDEX Depressive Symptoms Scale score ≥6 | Dementia (CAMDEX, post‐mortem, cause of death, n = 76) | Good | Depression not associated with incident dementia (OR: 0.78, 95% CI: 0.34–1.78). |
| Ezzati 2019 n = 1219 | The Einstein Ageing Study, US, prospective | 78.3 (5.3) | 17.2 | 4.5 (3.5) | GDS score ≥6 | Dementia (DSM‐IV, n = 132) | Good | Depression associated with dementia in longer‐term (>3 years HR: 1.13, 95% CI: 1.01–1.26), but not short‐term follow‐up (<3 years HR: 1.09, 95% CI: 0.99–1.2). |
| Gatz 2005 n = 766 | Manitoba Study of Health and Ageing, Canada, prospective | 74.5 (6) | 5 | / | CES‐D score ≥16 | Dementia (DSM‐III‐R, n = 56), AD (NINCDS‐ADRDA, n = 36) | Poor | Depression associated with dementia (CES‐D ≥16 HR: 2.37, 95% CI: 1.02–5.54) and AD (OR: 2.75, 95% CI: 1.04–7.24). |
| Geerlings 2000 n = 3147 | Amsterdam Longitudinal Study of the Elderly, The Netherlands, prospective | Depression: 73.6 (5.7), no depression: 73.7 (5.7) | 4 | 3.2 | GMS‐AGECAT score ≥3 | AD (DSM‐III‐R, n = 53) | Good | Depression associated with increased risk of AD (OR: 1.67, 95% CI: 0.76–3.63). |
| Gracia‐Garcia 2015 n = 3864 | ZARADEMP, Spain, prospective | Depression: 73.6 (9.3),No depression: 71.5 (8.9) | 4.5 | / | GMS‐AGECAT score ≥3 | AD (DSM‐IV, n = 70) | Poor | Severe depression associated with subsequent AD (SHR: 4.3, 95% CI: 1.39–13.33). |
| Heser 2013 n = 2663 | German Study on Ageing, Cognition, and Dementia in Primary Care Patients, prospective | 81.2 | 8 | / | CIDI | Dementia (DSM‐IV and ICD‐10, n = 308), AD (DSM‐IV, n = 152) | Good | Very late‐onset depression and current depressive symptoms both predicted all‐cause dementia |
| Heser 2020 n = 97,110 | German health insurance provider, Allgemeine Ortskrankenkasse | 74.7 (6.6) | 9 | 5.82 | ICD‐10: F32 | Dementia (ICD‐10, n = 20,779) | Good | Depression associated with dementia (IRR: 1.58, 95% CI: 1.51–1.64). Stronger association found for shortest interval, men, and younger participants |
| Holmquist 2020 n = 3,341,010 | Swedish population register | Matched cohort: 63.79 (11.89); Sibling cohort‐ depression: 59.10 (8.85), no depression: 59.97 (8.91) | 35 | 10.41 (6.89) | ICD‐10: F32, F33ICD‐8/9: 311, 296B | Dementia (ICD‐8/9/10, n = 9802), AD (ICD‐8/9/10, n = 4201), VaD (ICD‐8/9/10, n = 2329) | Good | Depression associated with dementia (OR: 2.47, 95% CI: 2.35–2.58), VaD (OR: 2.68, 95% CI: 2.44–2.95), and AD (OR: 1.79, 95% CI: 1.68–1.92). Association strongest in first 6 months after depression diagnosis but persisted for more than 20 years |
| Irie 2008 n = 1932 | Honolulu‐Asia Ageing Study, Japan, prospective | 76.3 (3.6) | 6 | / | CES‐D score ≥9 | Dementia (DSM‐III‐R, n = 98) | Good | Depression associated with higher rate of dementia (HR: 2.2, 95% CI: 1.3–3.7) and AD (HR: 2.9, 95% CI: 1.4–5.9), but not VaD (HR: 1.3, 95% CI: 0.3–5.8) |
| Karlsson 2015 n = 2404 | The Swedish Twin Registry | Dementia: 80.1 (6.6), no dementia: 78.9 (6.6) | >10 years | / | ICD‐7: 314.99; ICD‐8: 296.00, 298.00, 300.40–41, 790.20; ICD‐9: 296 C/D/W, 298A, 300E, 309 A/B, 311X; ICD‐10: F32, F33, F34.1, F41.2; Self‐report or anti‐depressant use; CES‐D score ≥20 | Dementia (DSM‐III‐R/IV, n = 804), AD (NINCDS‐ADRDA, n = 469) | Good | Depression associated with increased risk of dementia (HR: 3.41, 95% CI: 2.72–4.27). Strongest association in the 10 years before dementia diagnosis and for late‐onset depression |
| Köhler 2015 n = 35,791 | The Dutch Registration of Family Practices, Limburg, The Netherlands, register‐based | 65+ | 13 | / | ICPC code: P76 | Dementia (ICPC, n = 1680) | Good | Depression associated with increased risk of dementia (HR: 2.03, 95% CI: 1.56–2.64) |
| Köhler 2011 n = 771 | The Maastricht Ageing Study, The Netherlands, prospective | 67.1 (7.3) | 9 | / | Scores in upper quartile of SCL‐90 (revised version). | Dementia (DSM‐III‐R/IV, n = 37), AD (NINCDS‐ADRDA, n = 26), VaD (NINDS‐AIREN, n = 11) | Fair | Depression associated with all‐cause dementia (OR: 2.06, 95% CI: 1.01–4.22), but not AD (OR: 1.81, 95% CI: 0.78–4.23) or VaD (OR: 3.03, 95% CI: 0.86–10.64) |
| Kontari 2019 n = 4589 | English Longitudinal Study of Ageing, prospective | 50+ | 10 | / | CES‐D score ≥4 | Dementia (participant or informant reported physician diagnosed dementia, and/or IQCODE score ≥3.5, n = 216) | Poor | Depression associated with higher rate of dementia during follow‐up (HR: 1.82, 95% CI: 1.13–2.95), with attenuation after adjusting for baseline cognitive function (HR: 1.28, 95% CI: 0.78–2.08) |
| Lenoir 2011 n = 7989 | The 3C Study, France, prospective | 74 (5.4) | 4 | / | Self‐reported lifetime treated depression. MDD (DSM‐IV). CES‐D baseline score ≥22 (women), ≥16 (men) | Dementia (DSM‐IV, n = 276), AD (NINCDS‐ADRDA, n = 180), VaD (NINDS‐AIREN, n = 24) | Poor | Dementia associated with baseline depressive symptoms (HR: 1.5, 95% CI: 1.2–2.2), but not self‐reported treated depression or diagnosed MDD |
| Li 2011 n = 3410 | Adult Changes in Thought Study, US, prospective | Significant depressive symptoms: 75.8 (6.2),No depression: 74.8 (6.2) | 15 | 7.1 | CES‐D score ≥16; self‐reported history of depression | Dementia (DSM‐IV, n = 658) | Good | Baseline depression associated with dementia (HR: 1.71, 95% CI: 1.37–2.13), for late‐life (age‐at‐onset ≥50 years) (HR: 1.46, 95% CI: 1.16–1.84), but not early‐life depression (age‐at‐onset <50 years) HR: 1.1, 95% CI: 0.83–1.47) |
| Lin 2017 n = 49,955 | NHIRD, Taiwan, register‐based | Median: 39, IQR: 29–51 | 10 | Depression median: 7.19 (IQR: 5.95–8.48), control: 7.22 (IQR: 6.01–8.51) | ICD‐9‐CM: 296.2x–296.3x, 300.4, 311.x | VaD (ICD‐9‐CM, n = 117) | Good | Depression associated with higher rate of VaD (HR: 3.1, 95% CI: 2.13–4.52) |
| Luppa 2013 n = 1265 | Leipzig Longitudinal Study of the Aged, Germany, prospective | 81.5 | 8 | 4.3 (2.4) | CES‐D score ≥23; MDD diagnosis in DSM‐III‐R | Dementia (DSM‐III‐R/‐IV and ICD‐10, n = 183) | Good | Major depression diagnosis associated with higher rate of dementia (HR: 2.75, 95% CI: 1.01–7.5) |
| Mirza 2014 n = 4393 | Rotterdam Study, The Netherlands, prospective | 72.7 (7.3) | 13.7 | 8.7 (3.5) | CES‐D score ≥16 | Dementia (DSM‐III‐R, n = 582), AD (NINCDS‐ADRDA, n = 489) | Good | Depression associated with dementia in short (HR: 1.08, 95%CI: 1.00–1.17) (0–5 years HR: 1.13, 95%CI: 1.01–1.27) and intermediate (5–10 years HR: 1.14, 95%CI: 1.01–1.29), but not long‐term follow‐up (>10 years HR: 0.83, 95%CI: 0.66–1.04) |
| Richard 2013 n = 1483 | Washington Heights–Inwood Columbia Ageing Project, US, prospective | Depression: 77.7, (7.2), No depression: 76.7 (7.0) | 10.1 | 5.4 | CES‐D score ≥4 | Dementia (DSM‐III‐R, n = 207), AD (NINCDS‐ADRDA, n = 164), VaD (NINDS‐AIREN, n = 33) | Good | Baseline depression associated with higher rate of dementia (HR: 1.8, 95%CI: 1.2–2.7), and AD (HR: 1.9, 95%CI: 1.2–2.9), but not VaD (HR: 1.7, 95% CI: 0.5–5.6) |
| Rolandi 2020 n = 1100 | Brain ageing in Abbiategrasso (InveCe.Ab), Italy, prospective | 70–74 | 8 | 7 | DSM‐IV‐TR MDD or dysthymiaOR 3+: (i) depression history, (ii) depression treatment, (iii) GDS score ≥8, (iv) depressed mood in last week. | Dementia (DSM‐IV‐TR, n = 111) | Poor | Depression not associated with subsequent dementia (SHR: 0.99, 95% CI: 0.4–2.45) |
| Saczynski 2010 n = 949 | The Framingham Heart Study, US prospective | 79 (5) | 17 | 8 | CES‐D score ≥16 | Dementia (DSM‐IV, n = 164)AD (DSM‐IV, n = 136) | Good | Depression associated with increased risk of dementia (HR: 1.72, 95% CI: 1.04–2.84) and AD (HR: 1.76, 95% CI: 1.03–3.01) |
| Singh‐Manoux 2017 n = 10,308 | Whitehall II, UK, prospective | 1985‐ Depression: 44.5 (6), no depression: 45.1 (6); 2003‐ Depression: 60 (6), no depression: 61.5 (6) | 28 | 1985: 26.6 (4.5)1991: 21.7 (3.6)1997: 16.3 (2.7),2003: 11.1 (1.8) | General Health Questionnaire (GHQ‐30) ≥5CES‐D score ≥16 | Dementia (ICD‐10, n = 322) | Good | Dementia associated with depression in late (mean follow‐up 11 years) (HR: 1.67, 95% CI: 1.11–2.49, but not early study phase (mean follow‐up 22 years) (HR: 1.02, 95% CI: 0.72–1.44) |
| Spira 2012 n = 302 | Study of Osteoporotic Fractures, US, prospective | 86.9 (2.1) | 5 | / | GDS score ≥6 | Dementia (DSM‐IV‐R, n = 84) | Poor | Depression associated with increased risk of dementia (OR: 3.15, 95% CI: 1.03–9.65) |
| Tao 2019 n = 8880 | NHIRD, Taiwan, register‐based | 71.55 (5.47) | 7 | 6.94 (0.5) | ICD‐9‐CM: 296.2x at least 2x in 6 months and prescribed antidepressant medication for 90 days and 6 months + after initial diagnosis. | AD (ICD‐9‐CM, n = 155) | Good | Depression associated with increased risk of AD (HR: 2.21, 95% CI: 1.57–3.31) |
| Vilalta 2013 n = 451 | Estudio de Verona, Spain, prospective | 76.9 (5.5) | 5 | / | DSM‐IV diagnosis of major depression | Dementia (DSM‐IV, n = 52), AD (DSM‐IV, n = 30) | Poor | Late‐onset depression associated with increased risk of dementia (HR: 2.64, 95% CI: 1.15–6.0) |
| Wallin 2013 n = 212 | The Umeå 85+/GERDA (GErontologisk Regional DAtabas), Sweden, prospective | Dementia: 88.54 (3.7), no dementia 88.92 (4.35) | 7 | Dementia: 3.82 (1.22) no dementia: 3.24 (1.71) | DSM‐IV diagnosis of depression | Dementia (DSM‐IV, n = 71) | Good | Baseline (OR: 2.91, 95% CI: 1.37–6.16) and follow‐up depression (OR: 1.61, 95% CI: 1.26–2.05) associated with dementia |
| Wu 2020 n = 16,210 | Survey of Health, Ageing and Retirement in Europe (SHARE), 14 countries in Europe | 70.13 | 10 | 7.98 (2.61) | Europe‐depression (EURO‐D) scale score ≥4 | Dementia (participant‐or proxy‐reported physician diagnosis, n = 1030) | Good | Late‐life depression associated with dementia (SHR: 1.52, 95% CI: 1.32–1.75), although only in those below age 80 years. Dose‐response relationship found between depression severity and dementia risk (p for trend <0.001) |
a
Standard deviation (SD), interquartile range (IQR).
b
Anxiety and Depression Index (ADI‐4), Center for Epidemiologic Studies‐Depression (CES‐D), Composite International Diagnostic Interview (CIDI), Comprehensive Psychopathological Rating Scale (CPRS), Geriatric Depression Scale (GDS‐15), Hospital Anxiety and Depression Scale (HADS), Spielberger State Trait Anxiety Inventory (STAI), State‐Trait Personality Inventory (STPI), Symptom Checklist‐90‐Revised (SCL‐90).
c
International Classification of Health Problems in Primary Care (ICPC), The Geriatric Mental State ‐ Automated Geriatric Examination for Computer Assisted Taxonomy (GMS‐AGECAT), Cambridge Diagnostic Examination for the Elderly (CAMDEX).
d
Alzheimer's disease (AD), Vascular dementia (VaD).
e
Newcastle‐Ottawa Scale for cohort studies.
f
Hazard ratio (HR), incidence rate ratio (IRR), odds ratio (OR), subdistribution hazard ratio (SHR).