Introduction

This concept contains information on the methods used in MCHP research to identify the conditions of Alcohol Use Disorder (AUD) and high-risk alcohol use. In Manitoba, there is a concern of the harms associated with misuse of alcohol, which have far-reaching impacts across the health, social and justice sectors. Information for this concept is taken directly from the Deliverable Health and Social Outcomes Associated with High-Risk Alcohol Use by Nickel et al.(2018)

This research used a matched cohort approach by developing an AUD cohort based on alcohol-related diagnoses found in the hospital abstracts and physician services data, and prescriptions related to alcohol treatment found in the Manitoba prescription drug data. A matching process was used to develop a "non-exposed" cohort for comparison. Additionally, survey data was used to identify self-reported alcohol consumption levels that exceeded Health Canada's recommended low-risk daily and weekly guidelines. These methods are described below.

For more information related to the patterns of healthcare use, health outcomes, and social services use outcomes associated with having an AUD and those who exceeded the low-risk alcohol consumption guidelines, please read the on-line Health and Social Outcomes Associated with High-Risk Alcohol Use Deliverable.

Method of Identifying Alcohol Use Disorder (AUD)

Individuals with an indication for an AUD were identified if they met at least one of the following conditions:


• diagnosed with an alcohol-related mental disorder;
  
• diagnosed with a disease caused by excessive alcohol consumption; OR
  
• received a prescription for a medication used to treat alcohol use disorder.

Diagnoses and Prescription Drug Codes Used to Define AUD

The following mental health diagnoses were used to identify individuals with an AUD, using the Hospital Abstracts data and the Medical Services (Physician Claims) data:



ICD-9-CM Diagnosis Codes	ICD-10-CA Diagnosis Codes
291 Alcohol-induced mental disorders	F10 Mental and behavioural disorders due to use of alcohol
303 Alcohol dependence syndrome	.
305.0 Alcohol abuse	.

The following physical health diagnoses were used to identify individuals with an AUD, using the Hospital Abstracts data:



ICD-9-CM Diagnosis Codes	ICD-10-CA Diagnosis Codes
571.0 Alcoholic fatty liver	K70 Alcoholic Liver Disease
571.1 Acute alcoholic hepatitis	.
571.2 Alcoholic cirrhosis	.
571.3 Alcoholic liver damage unspecified	.
425.5 Alcoholic Cardiomyopathy	I42.6 Alcoholic Cardiomyopathy
535.5 Alcoholic Gastritis	K29.2 Alcoholic Gastritis
357.5 Alcoholic Polyneuropathy	G62.1 Alcoholic Polyneuropathy
760.71 FASD (for mothers of children with diagnosis)	Q86.0 FASD (for mothers of children with diagnosis)
977.3 Poisoning by Alcohol Deterrents	X45 Accidental poisoning by and exposure to alcohol
E86.00 Accidental poisoning by Alcoholic Beverages	X65 Intentional self-poisoning by and exposure to alcohol
.	Y15 Poisoning by and exposure to alcohol, undetermined intent
980.0 Toxic effect of alcohol (ethanol)	T51.0 Toxic effect of alcohol (ethanol)
V11.3 Alcoholism	Y90 Evidence of alcohol involvement determined by blood alcohol level
.	Y91 Evidence of alcohol involvement determined by level of intoxication
.	Z50.2 Alcohol rehabilitation
.	Z71.4 Alcohol abuse counselling and surveillance
.	Z72.1 Problems related to lifestyle, Alcohol use
.	E24.4 Alcohol-induced pseudo-Cushing’s syndrome
.	G31.2 Degeneration of nervous system due to alcohol
.	G72.1 Alcoholic Myopathy
.	K85.2 Alcohol-induced acute pancreatitis
.	K86.0 Alcohol-induced chronic pancreatitis
.	O35.4 Maternal care for (suspected) damage to fetus from alcohol

The following medications (ATC codes) were used to identify individuals with an AUD, using the Drug Program Information Network (DPIN) data:



ATC Codes	Generic Drug Names
N07BB01	Disulfiram
N07BB02	Calcium Carbimide
N07BB03	Acamprosate Calcium
N07BB04	Naltrexone

NOTE: Researchers may choose to include or exclude Calcium Carbimide (N07BB02), depending on the purpose of their research. Calcium Carbimide was prescribed as a treatment for AUD, however, it is no longer recommended to be prescribed due to severe reaction when people consume alcohol. The drug was removed from the Canadian market in April, 1999. Researchers should consider the objectives and timing of their study before including Calcium Carbimide in the list of medications to treat alcohol use disorder (AUD). For more information on Calcium Carbimide, see the Government of Canada Health Products web page at:
https://health-products.canada.ca/dpd-bdpp/dispatch-repartition.do;jsessionid=BC562A564254838F8B5AFF93F64EE0C1

The lists of diagnoses and medications documented above are available in Appendix 1: Technical Definitions of the on-line Deliverable.

Methods of Identifying High-Risk Alcohol Use

This research also examined associations between self-reported drinking behaviours, healthcare use and social service use outcomes. Information on self-reported drinking behaviours is available in the Canadian Community Health Survey (CCHS) data.

High-risk alcohol use was identified by whether the response in the CCHS survey indicated they had exceeded Health Canada's recommended low-risk drinking guidelines, which are:


• for women, consuming no more than 10 drinks per week and no more than 2 drinks per day most days; and
  
• for men, consuming no more than 15 drinks per week and no more than 3 drinks per day most days.

A list of the alcohol-related CCHS survey variables used in this research are located in Appendix 1: Technical Definitions - Alcohol Use Disorder - Matches of the Deliverable.

Cautions / Limitations

There are several limitations to the administrative data used in the study. These included:


• many individuals with an AUD go undetected by the medical system;
  
• The Medical Services data we analyzed only has three-digit ICD diagnosis codes, so we could not identify physical conditions from outpatient visits, only from hospitalizations;
  
• those who present to the medical system and receive an indication for having an AUD may be sicker and have more advanced progression of AUD compared with those who remain in the community without seeking medical attention;
  
• if the control group does contain individuals with an AUD, this would serve to attenuate any differences in outcomes between our cases and matches;
  
• we are only able to detect individuals once they present with an AUD; and
  
• there is the potential for coding errors, leading to misclassification of someone as having or not having an AUD.

These limitations result in the fact that the estimated population-level impact of AUD on health and social services presented in the report is likely an underestimate.

There are also two main limitations to using CCHS survey data:

• CCHS does not ask questions about life-long alcohol consumption; and
  
• survey respondents in recovery may have appeared to be non-drinkers based on their self-reported alcohol consumption.

Both scenarios would result in misclassification and attenuate any differences toward the null.