Introduction

This concept defines dementia and then describes the methods that have been used in MCHP research to identify dementia from administrative data in the MCHP Data Repository. Links are provided to report specific research findings.

Dementia is a term used to define the loss of cognitive function of the brain. This usually affects decision-making and problem solving, memory and verbal communications, and in some instances also results in behaviour changes (National Institute of Neurological Disorders and Stroke (NINDS), 2006). See NINDS Web Site - Dementia Information Page at: http://www.ninds.nih.gov/disorders/dementias/dementia.htm for more information.

See MedlinePlus® - Health Topics - Dementia for more information.

Data Sources and General Conditions for Identifying Dementia

The data sources used to identify dementia from the MCHP Data Repository include:


• Hospital Abstracts data; and
  
• Medical Services (Physician Claims) data.

Typically, analysis of the general population is limited to age 55 and over in most studies, but analysis of PCH residents is limited to those individuals aged 75 and over. Prevalence rates are usually calculated using five years of data.

MCHP Research Methods for Identifying Dementia

Over time, different methods have been used to identify dementia in MCHP research. The following describes the methods that have been used and the specific research projects these methods are used in.

1. De Coster C. & Kozyrskyj A. (2000), Menec et al. (2002), Martens et al. (2004), and Doupe et al. (2006, 2008)

In a number of early reports from MCHP, De Coster C. & Kozyrskyj A. (2000), Menec et al. (2002), Martens et al. (2004), and Doupe et al. (2006, 2008) the presence of any of the following broad range of ICD-9-CM codes in either hospital discharge abstracts or medical services/physician claims data were used to identify dementia:


• 290 (senile and pre-senile organic psychotic conditions),
  
• 291-292 (alcoholic and drug psychoses),
  
• 294 (other organic psychotic conditions),
  
• 331 (cerebral degenerations), or
  
• 797 (senility)

This definition was based on input from geriatricians and epidemiologists specializing in research of the older adult.

In The Health and Health Care Use of Manitoba's Seniors: Have They Changed Over Time? deliverable by Menec et al. (2002) they reported the proportion of the seniors population (age 65+) with a dementia diagnosis for 1999. For more information, see section 4.5.2 Demntia in this report.

In the Patterns of Regional Mental Illness Disorder Diagnoses and Service Use in Manitoba: A Population-Based Study deliverable by Martens et al. (2004) they investigated the treatment prevalence of dementia and reported their findings by RHA, RHA District and by income quintiles by age and by age and sex for 1997/98-2001/02. For more information, see section 2.10 Treatment Prevalence of Dementia in the report. Additional information is available in Table 1: Five-year period prevalence (1997/98-2001/02) of Dementia.

In the Using Administrative Data to Develop Indicators of Quality Care in Personal Care Homes (2006) and An Initial Analysis of Emergency Departments and Urgent Care in Winnipeg (2008) deliverables by Doupe et al. they investigated the prevalence of dementia using the definition listed above. For more information see:


• Figure 5.8: Percent of Residents Diagnosed with Dementia, for Manitoba and by RHA, Annual Average from 1999/2000 – 2003/04 in the 2006 (PCH Quality Indicators) report; and
  
• Table 8.4: Distribution of Emergency Department Users by Previous Diagnosis of Select Mental Illnesses in the 2008 (Emergency Department) report.

2. Kozyrskyj et al. (2003)

In the Discharge Outcomes for Long-Stay Patients in Winnipeg Acute Care Hospitals deliverable by Kozyrskyj et al. (2003) they investigated the likelihood of three hospital discharge destinations in comparison to going home for a variety of sociodemographic, health status and hospital factors, including dementia and other cognitive impairment. This involved hospital discharges between 1993/94 to 1999/2000. They defined dementia and other cognitive impairment using the following ICD-9-CM diagnosis codes in any of the 16 diagnosis fields:


• 290.0-290.9,
  
• 291.1, 291.2,
  
• 292.82, 292.83,
  
• 294.0, 294.1, 294.8, 294.9,
  
• 331.0, 331.1, 331.3, 331.7, 331.9,
  
• 797, and
  
• 799.3

For more information on their findings, see Tables B1 - B4 in Appendix B of this report.

3. Fransoo et al. (2009) and Martens et al. (2010)

In the Manitoba RHA Indicators Atlas 2009 deliverable by Fransoo et al. (2009) and the Profile of Metis Health Status and Healthcare Utilization in Manitoba: A Population-Based Study deliverable by Martens et al. (2010) they investigated the prevalence of dementia by regions in Manitoba. Prevalence rates were calculated using five years of data. Both reports defined dementia as any resident age 55 or older with:


• at least one hospitalization with any of the following ICD-9-CM codes: 290, 291, 292, 294, 331, or 797 or any of the following ICD-10-CA codes: F00, F01, F02, F03, F04, F05.1, F06.5, F06.6, F06.8, F06.9, F09, F10-F19 (but excluding: F10.0, F10.1, F10.2, F10.3, F10.4, F10.8, F10.9, F11.1, F11.2, F12.1, F12.2, F13.1, F13.2, F14.1, F14.2, F15.1, F15.2, F16.1, F16.2, F17.1, F17.2, F18.1, F18.2, F19.1, F19.2), G30, G31.0, G31.1, G31.9, G32.8, G91, G93.7, G94, or R54; OR
  
• at least one physician visit with any of the following ICD-9-CM codes: 290, 291, 292, 294, 331, or 797.

In the "2009 RHA Atlas", prevalence rates were calculated for two different, 5-year time periods: 1996/97–2000/01 and 2001/02–2005/06. For more information see section 5.7 - Dementia that contains figures for prevalence by RHA, RHA District and Winnipeg Neighbourhood Clusters.

In the "Profile of Metis" report, prevalence rates were calculated using data from 2002/03-2006/07. For more information see section 6.7: Dementia that contains figures for prevalence by RHA, Metis Region and Winnipeg Community Area, for the Metis cohort compared to All Other Manitobans.

4. Martens et al. (2010) and Chartier et al. (2012)

In the Health Inequities in Manitoba: Is the Socioeconomic Gap in Health Widening or Narrowing Over Time? deliverable by Martens et al. (2010); and the Health and Healthcare Utilization of Francophones in Manitoba deliverable by Chartier et al. (2012) they investigated the prevalence of dementia in the population. The ICD codes used to define dementia in this research included:


• one or more hospitalizations in five years with a diagnosis for dementia, including organic psychotic conditions, cerebral degenerations and senility, using the ICD-9-CM diagnosis codes: 290, 291.1, 291.2, 292.82, 294, 331, 797; OR the ICD-10-CA diagnosis codes: F00, F01, F02, F03, F04, F05.1, F06.5, F06.6, F06.8, F06.9, F09, F10.7, F11.7, F12.7, F13.7, F14.7, F15.7, F16.7, F18.7, F19.7, G30, G31.0, G31.1, G31.9, G32.8, G91, G93.7, G94, R54; OR
  
• one or more physician visits in five years with a diagnosis for dementia, using ICD-9-CM diagnosis codes: 290, 294, 331, 797.

In the "Health Inequities" deliverable (2010), they reported prevalence of dementia over five, 5-year time periods from 1984/85 to 2008/09. For more information, please see the section on Dementia that contains information and figures on prevalence over time by (rural and urban) income quintiles, adjusted Lorenz curves for dementia in rural and urban areas over two different time periods, and disparity rate ratios by urban and rural income.

In the "Francophone" deliverable (2012) they compared a cohort of Francophone Manitobans to all other Manitobans. For more information, please see section 6.7 Dementia that presents rate ratios of dementia for Francophones versus other Manitobans by RHAs and Winnipeg Community Areas.

5. Fransoo et al. (2013) and Chartier et al. (2015)

In The 2013 RHA Indicators Atlas deliverable by Fransoo et al. (2013) and the Care of Manitobans Living with Chronic Kidney Disease deliverable by Chartier et al. (2015) they investigated the prevalence of dementia. The ICD codes used to define dementia in this research included:


• one or more hospitalizations with a diagnosis for dementia, including organic psychotic conditions, cerebral degenerations, and senility -- ICD-9-CM: 290, 291.1, 292.2, 292.82, 294, 331, 797; OR ICD-10-CM: F00, F01, F02, F03, F04, F05.1, F06.5, F06.6, F06.8, F06.9, F09, F10.7, F11.7, F12.7, F13.7, F14.7, F15.7, F16.7, F17.7, F18.7, F19.7, G30, G31.0, G31.1, G31.9, G32.8, G91, G93.7, G94, R54; OR
  
• one or more physician visits with a diagnosis for dementia; ICD-9-CM: 290, 294, 331, 797

In the "2013 RHA Atlas" deliverable (2013) they reported dementia prevalence by RHA, for both the current 5-RHA structure and the former 11-RHA, and also by RHA Districts and by Winnipeg Neighbourhood Clusters. Prevalence rates were compared over two 5-year time periods - 2002/03–2006/07 and 2007/08–2011/12. For more information, please see Section 5.3 Dementia Prevalence in the 2013 RHA Indicators Atlas.

In the "Chronic Kidney Disease" deliverable (2015) they reported dementia prevalence and relative risk for end stage kidney disease (ESKD) and chronic kidney disease (CKD). For more information, please see


• Table 4.2 Mental Disorders in Adults in End Stage Kidney Disease; and
  
• Table 4.6 Mental Disorders in Adults in Chronic Kidney Disease;

6. Chartier et al. (2018)

In the Mental Illness Among Adult Manitobans deliverable by Chartier et al. (2018) they investigated the prevalence of dementia. The following ICD codes were used to define dementia in this research:


• one or more hospitalizations with a diagnosis for dementia, including organic psychotic conditions, cerebral degeneration, and senility: ICD-9-CM codes: 290, 291.1, 291.2, 292.82 294,331,797; OR ICD-10-CA codes: F00, F01, F02, F03, F04, F05.1, F10.7, F11.7, F12.7, F13.7, F14.7, F15.7, F16.7, F18.7, F19.7, G30, G31.0, G31.1, R54; OR
  
• one or more physician visits with a diagnosis for dementia; ICD-9-CM codes: 290, 294, 331, 797

Cautions / Limitations

• Dementia is difficult to code using administrative claims data. There is the potential for substantial under counting. The definition often used at MCHP was created in consultation with geriatricians and epidemiologists specializing in research of the older adult. This definition is meant to be more inclusive (higher sensitivity) rather than exclusive (specificity). Codes 291 and 292 have sub-codes (291.1, 291.2, 292.82) that are clearly dementia. These 4- and 5-digit codes cannot be captured in the Medical Services (physician claims) data because only 3-digit ICD-9-CM codes are recorded, so a more inclusive definition was used. NOTE: In December 2018, a 5-digit diagnosis code variable was added to the Medical Services data. If recorded, the first 3-digits of both the 3-digit and 5-digit diagnosis code variables will match exactly. The 5-digit diagnosis code variable applies to records from 2015/16 forward.
  
  
• The codes 291 and 292 contribute roughly 5% of the total dementia cases to the cohort ( Martens et. al. 2007 - see on-line article at http://publications.cpa-apc.org/media.php?mid=515 ).
  
  
• As of January, 2010, a decision was made at MCHP to include very specific limits on the use of the 291 and 292 ICD-9-CM codes to define dementia. For hospital discharge abstracts data, use the ICD-9-CM codes: 291.1x, 291.2x and 292.82 to define dementia. For medical services/physician claims data, the 3-digit codes 291 and 292 should not be used to define dementia.
  
  
• In discussion at the 2013 RHA Day, it was suggested that the drug Aricept - generic name Donepezil - ATC code N06DA02 - be included in the list of medications for determining dementia (Alzheimer's disease) prevalence. NOTE: As of November 2013, MCHP has not included this medication in its definition of dementia. This is something that future research should investigate and consider.