Concept: Calculating Costs/Expenditures for Pharmaceuticals

Introduction

The addition of pharmaceutical data to the Manitoba Population Research Data Repository in 1995 has given MCHP the ability to study population-based patterns of pharmaceutical use, costs and health outcomes for users of specific types of pharmaceuticals. The addition enhances the Centre's ability to analyze the population's use of health services and the relationship between health care expenditures and health.

This concept describes the methodology for calculating costs/expenditures for pharmaceutical/drug information. NOTE: For an overview of general costing methods (sources of data, types of costs, and approaches) and the methods of costing specific health services (hospitals, physicians, home care, personal care homes and pharmaceutical/prescription drugs) that have been used in MCHP research over time, please see the Costing Methods: An Overview of Costing Health Services in Manitoba concept.

The Drug Program Information Network (DPIN) collects data on prescriptions issued to Manitobans from different sources/carriers, including Pharmacare (C1), Personal Care Homes (PCH) (C2), Employment/Income Assistance (C3), Palliative care (C4), and non-adjudicated sources (third party/client pay). The system does not collect data from hospital pharmacies, nursing stations, ward stock and outpatient visits at CancerCare Manitoba.

A brief synopsis of the data collected includes the type of drug, dosage, prescription date, ingredient cost and professional (dispensing) fee. For more detailed information regarding the pharmaceutical/drug data, see the Drug Program Information Network (DPIN) Data Description.

This concept also provides links to relevant SAS code. Please see the SAS code and formats section containing the following links to: Drug Claims SAS Code, Non-Adjudicated Professional Fees Paid - SAS Code, and PCH Professional Fees Paid SAS Code - (internal access only).

Costs

Total prescription costs include the drug ingredient cost plus the professional (dispensing) fees.


• Drug ingredient cost - Most drugs, including their generic equivalents are included on the Manitoba Drug Benefits and Interchangeability Formulary available from Manitoba Health, available on-line at http://www.gov.mb.ca/health/mdbif/index.html. This formulary identifies identical or very similar drugs produced by at least two manufacturers. The government reimburses drugs based on this formulary at the lower manufacturer's price among the group of interchangeable products. Manitoba Health updates this list every two to four months.
  
  The drug ingredient cost is available in the Pharmacare, Personal Care Home, and Employment and Income Assistance data. They are imputed for the non-adjudicated data set (see Imputation Methodology ).
  
  
• Professional (dispensing) fees - The professional fee is available in the Pharmacare and Employment and Income Assistance data. There is no minimum or maximum fee; retail pharmacies set their own fees. The average fee is based on information gathered by the DPIN program based on actual charges of retail pharmacies.
  
  The professional fees are imputed (see Imputation Methodology) for non-adjudicated data and the Personal Care Home data, the latter because nursing homes pay a set fee to the pharmacy per month based on the number of beds in the home. In Winnipeg, the rate is $26.45 per bed per month and outside of Winnipeg the rate is $26.95 per bed per month.
  
  Note: The professional fee is zero in the following situations:
  
  
  1. When the prescription is filled in drugstores where no professional fee is charged.
     
  2. In claims from the Personal Care Home data. These claims to do not contain a professional fee because nursing homes pay a set fee to the pharmacy per month based on the number of beds in the nursing home, as described above.
  
  
• Total prescription cost - the sum of drug ingredient cost plus the professional (dispensing) fees (using values in the data set, or the imputed values where applicable).
  
  For more information, see the links below to Costs in the DPIN Data - Internal File Documentation (internal access only) and MCHP Data Repository - Drug (DPIN) Database - Drug Database Guide .

Expenditures

Pharmaceutical expenditures can be broken into two categories: those paid by the Pharmacare program (government), and those paid privately.


• Total paid by Pharmacare (government) - total cost for the Personal Care Home Services claims, and the Employment and Income Assistance Drug Program Drug claims. The Pharmacare data contains a variable with the amount paid by the government for the given claim.
  
  
• Total paid for privately (either out-of-pocket or by private insurance) - the total cost of all non-adjudicated claims, and the difference between total cost and total paid by government for Pharmacare only. For the other carriers this amount is zero.

Measures of Pharmaceutical Costs

Several measures of pharmaceutical costs can be derived from data found in the Manitoba Population Research Data Repository:


• Average cost per prescription (and by therapeutic class) - total paid for all drug products (Drug Identification Numbers (DIN)) within each Anatomical Therapeutic Chemical (ATC) class divided by total volume of prescriptions for all DINs.
  
  
• Average cost per Defined Daily Dose (DDD) (and by therapeutic class) - total paid for all drug products (DINs) within each ATC therapeutic class divided by total DDDs dispensed for all DINs.
  
  
• Total drug expenditure by population group and therapeutic class - total paid for all drug products (DINs) within each ATC therapeutic class divided by the number of users or residents eligible to receive health services or by age, sex, region of residence, and other demographic characteristics of interest including income quintile.

Imputation Methodology

The imputation methodology as described in Appendix I of Metge et al. 1999 is:

"For those individuals either receiving social service from Winnipeg city or designated as Status Indians and having prescriptions dispensed, data extracted did not include any expenditure data. Expenditure data was imputed for these claims. Both ingredient costs and professional fees for dispensing were imputed. To impute a professional fee per prescription for these claims, we used the professional fee mode (a probabilistic value) from the Manitoba Family Services plan. Ingredient cost for these claims were imputed from the Manitoba Family Services plan using the product of two fields: metric quantity dispensed [MQTY] x unit price mode per metric quantity. However, there was a consistency problem with MQTY. Many drug products can be described using multiple metric quantity descriptions. For example, the metric quantity in the DPIN database for a 200 mL bottle of 250 mg/5mL amoxicillin could be described as: 1 (bottle), 200 mL, 100 g or 10,000 mg. We used the following rule to impute metric quantity: if MQTY (of the claim) was 90% percentile and > 20 times the MQTY mode, then MQTY mode (from Manitoba Family Services) was imputed.

Days supply and a professional fee for dispensing are not included in the personal care home (PCH) or nursing home data; a professional fee was imputed for PCH claims and rates requiring days supply in the calculation (e.g., DDDs) do not include the residents of PCHs. There were 300 699 prescriptions dispensed (or 4.5% of total 1996 claims) for personal care home residents. A professional dispensing fee had to be imputed for all these claims because of the current method of reimbursement to pharmacies providing these services. Reimbursement to pharmacists for dispensing to PCHs is based on a capitated rate per bed exclusive of ingredient cost. This rate is different for pharmacies dispensing pharmaceuticals inside the city of Winnipeg or rurally ($ 26.45 vs. $ 29.95 per bed per month). A fee for Winnipeg and non-Winnipeg PCH claims was calculated on a monthly basis (total beds x capitation rate ÷ number of prescriptions claims/month) and then imputed for all PCH claims."

Significance

The data on pharmaceuticals includes virtually all prescription drug use in Manitoba. The addition of these data to the Manitoba Population Research Data Repository has given MCHP the ability to:


• Study population-based patterns of pharmaceutical use ( Metge, 1999 )
  
  
• Develop indicators of pharmaceutical appropriateness ( Metge, 2003 )
  
  
• Describe the extent of use of the most influential new pharmaceuticals ( Metge, 2003 )
  
  
• Characterize high-cost prescription users ( Kozyrskyj, 2005 )
  
  
• Determine whether differences in disease prevalence explain the higher prescription costs of high-cost users ( Kozyrskyj, 2005 )

Cautions/Limitations

Although the DPIN database has been validated and is considered to be reliable ( Metge, 1999 ), several cautions are to be noted. First, DPIN does not capture in-hospital use of prescription drugs ( Metge, 2003 ). Secondly, pharmaceuticals dispensed to Status Indians through nursing stations in remote Northern locations are not included in the database ( Metge, 1999 ).

It should be noted that only approximately 75% of Personal Care Homes (PCH) fill prescriptions at community pharmacies (and are therefore included in the DPIN system). Medication use for PCH where prescriptions are filled through hospitals is not known (Doupe et al., 2006).

Also, imputations are required to determine professional fee and days supply for pharmaceuticals dispensed in personal care or nursing homes as well as expenditures for individuals receiving social service from Winnipeg and Status Indians (see Imputation Methodology ).