Introduction

This concept contains information on how asthma been defined in MCHP research using administrative data, including a brief definition, a broad review of existing literature published prior to 2006, and specific algorithms since then that have defined asthma using the Manitoba Population Research Data Repository housed at MCHP. These algorithms have typically used a combination of ICD diagnoses codes from hospital abstract and medical services data, and/or relevant ATC codes found in prescription drug data. Future research in this area should investigate specific physician tariff codes that may detect additional asthma cases that are well controlled.

Definition and Literature Review

Definition of Asthma

Asthma is an inflammatory disorder of the airways, characterized by periodic attacks of wheezing, shortness of breath, chest tightness, and coughing. See MedlinePlus® - Health Topics - Asthma for more information.

Literature Review

In Lix et al. (2006), they performed a literature review and found that cases of asthma have been ascertained in administrative data using diagnosis codes alone ( Erzen et al. 1995, Huzel et al. 2002, Wilchesky et al. 2004 ) and diagnosis codes along with prescription drug claims ( Morrison et al. 2001, Kozyrskyj et al. 2004, Macy et al. 2005, Schatz et al. 2005, Lix et al. 2006 ). Table 1 summarizes eight studies, published prior to 2006, that use administrative data to ascertain cases of asthma.

Three of the studies used Manitoba data; among these studies, two used only medical services/physician claims to identify asthma cases, while the third ( Kozyrskyj et al., 2004 ) used a combination of hospital separations, medical services/physician claims, and prescription drug records to identify asthma cases. For the studies that reported validation results, estimates of sensitivity, specificity, and predictive values varied substantially.

Manitoba Asthma Algorithms

The following studies describe the asthma algorithms developed in different research projects over time, using data from the Manitoba Population Research Data Repository.

1. Kozyrskyj et al. (2004)

In Kozyrskyj et al. (2004) children with persistent asthma were identified as having one or more physician claims or hospitalizations for asthma-like diagnoses (ICD-9-CM 493) in one year, or, in the absence of these diagnoses, one or more prescriptions for an inhaled corticosteroid or chromone, or Ketotifen concomitant with an inhaled or oral beta-agonist, or two or more prescriptions for an inhaled or oral beta-agonist. This study focused on children aged 5 to 15 years.

2. Lix et al. (2006)

In Lix et al. (2006) the following ICD-9-CM diagnosis code was used in combination with prescription drugs to identify asthma cases:

• 493 - Asthma

Prescription Drug(s)

Drugs used for treatment of Asthma are complex and often overlap with treatments for other conditions. The definition used in this study followed work done by Kozyrskyj et al. (2004) and was updated by Lix et al. (2006) with input from asthma researchers and pharmacists working with MCHP. A complex set of drugs was defined primarily on Drug Identification Number (DIN) , route, and active ingredient (see SAS code below). Initial selection of drugs was based on Anatomical Therapeutic Chemical (ATC) code: R03A (Adrenergics, inhalants), R03B (other drugs for obstructive airway diseases, inhalants), R03C (other drugs for obstructive airway diseases, inhalants), and R03D (other systemic drugs for obstructive airway diseases). Based on DIN, active ingredients and route a number of drugs were excluded from this list and one additional drug (Ketotifen, R06AX17) was included.

Comparisons were made between different sources of data. DINs were identified using only ATC codes (above), those identified through the use of the CPS with a primary indication for asthma, and the definition developed by Kozyrskyj and others for use by Lix et al. (2006) - Excel File of Drugs - 2006 (internal access only)).

See Table of Prescriptions for more information.

SAS Code for Identification of Asthma Drugs

See SAS code for identification of asthma drugs (internal access only) below.

Calculating Population-based Rates

Based on work done by Lix et al. (2006) , multiple administrative definitions of chronic diseases were compared with survey data. The definitions looked at multiple age groups (12-18, 19-49, 50+), multiple years of administrative data (1, 2, 3, 5), and different data sources of administrative data (medical services/physician claims, prescription drugs, and hospital discharge abstracts). The 'best' definition for identification of individuals with asthma was 5 years of data using all of the data sources. It should be noted that using prescription drugs alone had high sensitivity (81.5%) at 5 years.

Table 2, 3 and 4 report estimates of agreement, sensitivity, specificity and predictive values for the algorithms investigated for each of the three different age groups.

SAS Code for Calculation of Asthma Rates by Age Group and Year

See below for SAS code for calculation of asthma rates by age group and year (internal access only).

Validated work done by Kozyrskyj et al. (2004) for children has a slightly different requirement for prescriptions over time. The additional criteria in the drug definition for children is used to help remove individuals with childhood wheezing. The definition is at least one prescription for an inhaled corticosteroid or Chromone or Ketotifen concomitant with an inhaled or oral beta-agonist, or two or more prescriptions for an inhaled or oral beta-agonist. Each of these groups of drugs is identified in the provided code.

3. Lix et al. (2008)

Lix et al. (2008) provided an update to the 2006 study with a report titled Defining and Validating Chronic Disease: An Administrative Data Approach. An Update with ICD-10-CA. The purpose of the 2008 report is to examine the validity of administrative data for monitoring the prevalence of chronic disease in Manitoba. Specific objectives are:


• Report relevant ICD-10-CA codes for ascertaining cases of chronic disease in administrative health data;
  
• Evaluate the validity of multiple algorithms for identifying disease cases from Manitoba administrative data.

The 2008 report uses the same methods and algorithms as described in the 2006 report, with the following modifications:


• ICD-10-CA codes were used to define specific chronic diseases from hospital separation data, beginning April 1, 2004. This is due to a change in coding systems used in Manitoba hospitals. The same ICD-9-CM codes identified in the 2006 report were used to identify hospital cases prior to April 1, 2004.
  
• data from the Canadian Community Health Survey (CCHS), cycle 3.1, collected from January 2005 to January 2006 were used to evaluate the validity of the administrative data. The cohort consisted of 5,099 adults 19+ years of age and 701 youth 12 to 18 years of age.

The following ICD-10-CA codes were used to define asthma in administrative hospital separation data from April 1, 2004 to March 31, 2006:


• J45 - asthma
  
• J46 - status asthmaticus

Table 5 reports the estimates of agreement kappa(κ), sensitivity, specificity, Youden's Index, PPV (positive predictive value) and NPV (negative predictive value) for each of the 28 asthma algorithms that were investigated for the combined age group of 12 years and older.

Discussion of the validation results for asthma can be found in the full report Chapter 4: Asthma

Calculating Population-based Prevalence Rates

Crude provincial prevalence estimates for the 28 asthma algorithms are reported in Table 6 for the following four age groups: all ages (12 years and older), 12 to 18 years, 19 to 49 years, and 50 or more years.

Discussion of the prevalence rates for asthma can be found in the full report: Chapter 4: Asthma

4. Finlayson et al. (2010)

Finlayson et al. (2010) defined asthma as one or more hospitalizations OR one or more physician visits OR one or more prescriptions over a five-year time period for those aged 24+ where the events are coded with an ICD code representing asthma/COPD or a prescription is dispensed for an asthma/COPD medication. In this study, asthma and chronic obstructive pulmonary disease (COPD) are referred to as the same condition as it is not usually clinically possible to definitively distinguish between the two conditions.

5. Raymond et al. (2011)

Raymond et al. (2011) use a broader scope in their definition for asthma, defining it as one physician claim OR one hospital claim with a corresponding diagnosis of: ICD-9-CM: 464, 466, 490, 491, 493 or ICD-10-CA: J04, J05, J20, J21, J40, J41, J42, J45, J441, J448 OR one prescription for an asthma medication in a three-year period (see medications for asthma and chronic obstructive pulmonary disease (COPD) listed in Table A1.1 in the appendix of this report).

6. Chartier et al. (2012, 2016), Brownell et al. (2012) and Fransoo et al. (2019)

In Chartier et al. (2012, 2016), Brownell et al. (2012), and Fransoo et al. (2019), asthma is defined as one of the following conditions within a two-year period:


• one or more physician visits with a corresponding ICD-9-CM diagnosis code of 493; OR
  
• one or more hospitalization with a corresponding ICD-9-CM diagnosis code of 493 or ICD-10-CA diagnosis code of J45; OR
  
• one or more prescriptions for medication to treat asthma. For the lists of medications used in these reports, see:
  • medication list in Chartier et al. (2012) - NOTE: This list is based on the same ATC codes that are used in the Raymond et al. (2011) report.
    
  • medication list in Brownell et al. (2012) - NOTE: This list is based on the latest ATC codes and Drug Identification Numbers (DINs) available for asthma.
    
  • medication list in Chartier et al. (2016)
    
  • medication list in Fransoo et al. (2019)
  
  NOTE: In Chartier et al. (2016), they focused on asthma for children 6 - 19 years of age. In Fransoo et al. (2019), they focused on asthma for children 5 - 19 years of age.

Asthma Detection Using Physician Billing / Tariff Codes

Upon review of the Medical Services / Physician Claims data (Manitoba Physician's Manual - April 1, 2016), a number of tariff codes related to asthma were identified and could be used to augment existing definitions of asthma. The first tariff code may detect additional cases of asthma that are well controlled and not reported in other current administrative data.


• 8432 - annual management of Asthma - this tariff code is applicable for patients with a confirmed diagnosis of asthma and can only be claimed once per year.

Other relevant tariff codes used for diagnostic testing related to asthma include:


• 8862, 8860, 8861 – Exercise administration for asthma detection;
  
• 8865, 8863, 8864 – Histamine, methacholine, cold air administration for measurement of non specific reactivity for asthma; and
  
• 8868, 8866, 8867 – Antigen administration for detection of specific reactivity for asthma

NOTE: Tariff codes should not be used alone to detect cases of asthma. Rather, they should be used in conjunction with the existing definition using hospital, medical services and pharmaceutical data.

Cautions

• Because the definition used at MCHP includes a number of specific DINs and drug products, the medication list should be reviewed and updated prior to starting any new work using asthma or other diagnoses involving medications.