Dementia ascertainment using existing data in UK longitudinal and cohort studies: a systematic review of methodology

Sibbett, Ruth A.; Russ, Tom C.; Deary, Ian J.; Starr, John M.

doi:10.1186/s12888-017-1401-4

Table 2 Articles included in final review

From: Dementia ascertainment using existing data in UK longitudinal and cohort studies: a systematic review of methodology

Author	Study topic	Dementia outcome/ cohort^a	Existing data use	Data sources	Dementia criteria	Validation/ comparison^b	Quality score
Baker et al. [56]	Hip fracture risk and mortality in AD	Cohort	Full	THIN^e	Diagnostic code for AD, OR Prescription for an AD defining drug.	No	9
Brayne et al. [31]	Dementia at death and prevention	Outcome	Part	Death certificates	ICD-10 codes for dementia.	Yes	15
Brayne et al. [30]	Neuropathological correlates of dementia	Outcome	Part	Death certificates	DSM-IV criteria using data from all sources including death certificates	Yes	15
Chen et al. [32]	Health care resource utilisation in AD	Cohort	Full	CPRD^f	Diagnosis of AD.	No	8
Clarke et al. [4]	Dementia incidence	Outcome	Part	Death certificates, hospital case notes	‘Diagnostic information’ from both sources.	Yes	14
Cook et al. [37]	Incidence of stroke and seizure in AD	Cohort	Full	THIN^e	Read codes for AD.	No	9
Crugel et al. [7]	Antipsychotic use in dementia	Cohort	Part	Electronic clinical records for trust (Rio)	Confirmed ICD-10 diagnosis of dementia.	No	7
Doll et al. [42]	Smoking and dementia	Outcome	Full	Death certificates	Dementia ‘mentioned’ on death certificate.	Yes	11
Dregan et al. [57]	Inflammation, related therapy and dementia	Outcome	Full	CPRD^f	Diagnostic codes for dementia.	No	13
Goh et al. [13]	Angiotensin receptor blockers and dementia	Outcome	Full	CPRD^f	Read codes for dementia.	No	14
Grant et al. [58]	Incontinence and dementia	Cohort	Full	THIN^e	One or more codes dementia codes, OR Two or more prescriptions for drugs to treat dementia.	No	14
Guthrie et al. [16]	Psychotropic drug prescribing in dementia	Cohort	Full	SPICE-PC^h	Read codes for dementia, OR Prescription of acetylcholinesterase inhibitor.	Yes	14
Heath et al. [9]	Vascular comorbidities in early onset dementia	Outcome	Full	General practice research dataset (unnamed)	Read codes for dementia, OR Prescription of acetylcholinesterase inhibitor.	Yes	16
Houttekier et al. [59]	Place of death in dementia	Both	Full	Death certificates	ICD-10 codes for dementia.	No	13
Imfeld et al. [11]	Epidemiology, comorbidities and medication in AD & VD	Outcome	Full	GPRD^g	Stage I: read code for dementia, OR, prescription for acetylcholinesterase inhibitor. Stage II: algorithm for AD or VD (based on DSM-IV, NINCDS-ADRDA, NINDS-AIREN, NICE & SIGN).	Yes	9
Karlinsky et al. [60]	AD in twins	Outcome	Full	Hospital twin register, hospital case notes	Stage I: discharge diagnosis suggestive of dementia on register, Stage II: case note diagnosis of AD according to DSM-III-R criteria.	No	10
Keenan et al. [29]	Glaucoma and AD/ VD	Both	Full	Hospital Episode Statistics	ICD-10 diagnostic code for AD or VD.	No	15
Kehoe et al. [35]	Angiotensin targeting anti-hypertensives, mortality and hospitalisation in dementia	Cohort	Full	GPRD^g	Read codes for dementia.	No	9
Lu et al. [61]	Gout and risk of AD	Outcome	Full	THIN^e	Diagnostic codes for AD.	No	14
Martinez et al. [62]	Trends in antipsychotic drug use in dementia	Cohort	Full	CPRD^f	Read codes for dementia.	No	9
McCarthy et al. [33]	Experience of dying with dementia	Both	Full	Death certificates	Dementia diagnosis coded.	No	8
McGonigal et al. [22]	Epidemiology of pre-senile AD	Outcome	Full	^dISD Scotland data for psychiatric hospitals, hospital records	Stage I: diagnostic codes for dementia in SMR, Stage II: NINCDS-ADRDA criteria and Hachinski score applied to records.	Yes	13
Morgan et al. [63]	Physical activity in midlife and dementia	Outcome	Part	Death certificates	Mention of dementia in either part one or part two of death certificate (used in conjunction with new data at consensus)	No	15
Newens et al. [26]	Ascertainment, incidence, prevalence & survival in pre-senile AD	Outcome	Part	Electronic hospital information systems (^cHAA, MHE, Körner), neuroradiology records, case notes	Stage I: potential cases identified by ICD-9 codes from information systems, referrals for CT with dementing process, (and contact with services). Stage II: case notes for all examined for DSM-III-R criteria for dementia, then algorithm for pre-senile AD.	Yes	13
Palmer et al. [8]	Dementia and occupational exposure to solvents	Cohort	Full	CT records, hospital case notes	Exact criteria not specified.	No	7
Pendlebury et al. [5]	Risk of dementia after TIA and stroke	Outcome	Part	Clinical records (hospital & GP), death certificates	Recorded diagnosis of dementia in primary care record, DSM-IV criteria met after searching the primary care record or dementia listed on death certificate.	No	14
Perera et al. [34]	Response to acetylcholinesterase inhibition in dementia	Cohort	Full	Electronic medical records, hospital trust case register	Patients receiving Acetylcholinesterase Inhibitor.	No	8
Qizilbash et al. [14]	BMI and risk of dementia	Outcome	Full	CPRD^f, death certificates	Dementia term recorded in CPRD*, OR Dementia diagnosis on death certificate.	No	13
Rait et al. [17]	Survival in dementia	Both	Full	THIN^e	Read codes for dementia.	Yes	10
Renvoize et al. [23]	Prevalence of young onset dementia	Outcome	Part	Local computerised information systems, medical and social care records	Stage I: discharge diagnosis of dementia in computer system, Stage II: “criteria for dementia” from notes (criteria not specified)	Yes	13
Reyniers et al. [18]	International variation in place of death in dementia	Cohort	Full	Death certificates	ICD-10 dementia diagnosis.	No	10
Russ et al. [10]	Geographical variation in dementia (Scottish sub-study only)	Outcome	Full	^dISD Scotland data, death certificates, records for a nursing home medical practice	ICD-9 & 10 codes for dementia from ^dSMR data and death certificates, and dementia status reported by a medical practice.	Yes	16
Ryan [27]	Hospital incidence rates in dementia	Outcome	Full	^dISD Scotland data.	ICD-8 & 9 codes for dementia.	Yes	14
Sampson et al. [24]	Differences in care for those with and without dementia	Cohort	Full	Medical notes	Formal diagnosis of dementia recorded.	No	13
Seshadri et al. [12]	Oestrogen replacement therapy and the risk of AD	Both	Full	GPRD*, GP records	Stage I: computer diagnosis of dementia, Stage II: records for each reviewed to confirm diagnosis based on NINCDS-ADRDA criteria.	Yes	14
Shah et al. [25]	Dementia and influenza vaccination uptake	Outcome	Full	THIN^e	Read codes for dementia.	Yes	12
Sleeman et al. [19]	Place of death and dementia	Cohort	Full	Death certificates	ICD-10 codes for dementia.	No	8
Sorahan et al. [20]	AD, MND and PD in magnetic field exposure	Outcome	Full	Death records	ICD 9 & 10 codes for dementia.	No	12
Stephens et al. [28]	Prescribing of antipsychotics in dementia on the acute general hospital	Cohort	Full	Hospital Episode Statistics, linked to pharmacy records	ICD-10 code diagnosis of dementia.	No	9
Su et al. [64]	MMSE as a predictor of mortality	Cohort	Full	Hospital trust case register	ICD-10 codes for dementia.	No	14
De Vries et al. [65]	Dementia deaths in a hospice	Outcome	Full	Hospice case notes/ referrals	Formal diagnosis of dementia in the notes, other evidence in the notes (no exact criteria given).	Yes	12
Whalley et al. [66]	Effects of anticholinergic drugs on cognition	Outcome	Part	Clinical records, imaging results	Diagnoses based on ICD-10 criteria, using all source data.	No	15
Whalley et al. [67]	Childhood mental ability and dementia (Late onset sub-study only)	Outcome	Part	Psychiatric case register, hospital records	Stage I: diagnosis of dementia on register, Stage II: ICD-10 diagnosis of dementia in case notes.	No	14
White et al. [68]	Dementia, walking outdoors and getting lost	Outcome	Full	Missing persons reports	Presence of a key dementia term.	No	11
Wilcock et al. [69]	Educational intervention to improve the management of dementia	Outcome	Full	Electronic GP records	Search for dementia in records- exact criteria not specified.	No	10
Woodburn et al. [70]	Features of early onset dementia in Scotland	Cohort	Part	Psychiatric case register, neurology service records, case notes	Stage I: ICD-9 codes for dementia in register, Stage II: Feighner criteria for Organic Brain Syndrome applied to notes (criteria for neurology records not specified)	No	12
Wotton et al. [71]	Obesity and subsequent dementia	Outcome	Full	Hospital Episode Statistics linked with death records	ICD-9 & 10 codes for dementia.	No	14

^a: Was dementia ascertainment performed as the study outcome, or was dementia ascertainment performed in order to form a dementia cohort for which another outcome was determined
^bDid the paper perform a validation study for dementia ascertainment or compare incidence/ prevalence rate to expected rate
^c HAA: Hospital Activity Analysis; MHE: Mental Health Enquiry system; Körner: Korner Episode Statstics (hospital information systems)
^d ISD Scotland: Information Services Division Scotland (holds national datasets (SMR: Scottish Morbidity Records) of health outcomes/ diagnoses)
^e THIN: The Health Improvement Network; ^f CPRD: Clinical Practice Research Datalink; ^g GPRD: General Practice Research Database; ^h SPICE-PC: Scottish Programme for Improving Clinical Effectiveness- Primary Care

Back to article page

ISSN: 1471-244X

Contact us

Submission enquiries: bmcpsychiatry@biomedcentral.com
General enquiries: ORSupport@springernature.com

BMC Psychiatry

Contact us