Skip to main content

Proton Pump Inhibitor Initiative

 

In July 2010, 1,288 general practitioners in BC were sent EQIP’s educational materials and personalized prescribing portraits on evidence-based use of proton pump inhibitors (PPIs) and histamine2 receptor antagonists (H2RAs).

In December 2011, the remainder of eligible general practitioners in BC were sent these materials (2,000 GPs).

Rationale

PPIs and H2RAs are drugs that are prescribed to relieve acid reflux and related conditions. H2RAs are relatively inexpensive, fast-acting, and have fewer side-effects.  

Evidence-based guidelines recommend starting patients on H2RAs before proceeding to a PPI, but often patients are started on a PPI and remain on the drug for much longer than the recommended period of 4 weeks. Attempting to stop taking a PPI can result in painful rebound symptoms unless the dose is properly tapered.

EQIP aims to remind physicians to re-evaluate PPI therapy at 4 weeks and to taper patients off of the drug if symptoms are well-controlled.

Materials

PPI Portrait – July 2010

PPI Portrait – December 2011

PPI Cost Speedometer – July 2010

PPI Cost Speedometer – December 2011

 

How Do You Prescribe PPIs? Portrait Methods and Assumptions - July 2010 Mailout 

Purpose of This Document

The purpose of this document is to describe the detailed methods, processes and assumptions used to generate the personalized data in the EQIP PPI prescribing portraits.
(Download the July 2010 document as a PDF)

(Download the December 2011 document as a PDF)

Table of Contents

1 Portrait Inclusion Criteria

1.1     Physician Inclusion Criteria
1.2     Patient Inclusion and Assignment Criteria

2 Page-2 Top Bar Charts: Duration that Your Patients Are on a PPI

2.1     Overview
2.2     Data
2.3     Methods
2.4     Issues

3 Page-2 Bottom Bar Charts: Your Patients Who Reduced Their PPI Use

3.1     Overview
3.2     Data
3.3     Methods
3.4     Issues

4 Page-3 Top Cost Tables: The Cost of Your Prescribing 2004-2008

4.1     Overview
4.2     Data
4.3     Methods
4.4     Issues

5 Page-3 Bottom Line Graph: Your PPI and H2-Antagonist Prescription Trends

5.1     Overview
5.2     Data
5.3     Methods

Comments? You will need to register and log in


1.1 Physician Inclusion Criteria

1.1     Physician Inclusion Criteria

Physicians must meet all of the following requirements to be eligible to receive an individualized PPI prescribing portrait in either the Early Intervention or Delayed Control Group:

  1. Prescribed a PPI to 3 or more patients during the study period
  2. Defined as a ‘General Practitioner’ by the BC Medical Services Plan
  3. Have a License Status equal to ‘Private Practice’ according to the BC Medical Service Plan
  4. Not retired by 2009 (Fee for Service payments > $0 in the first quarter of 2009)
  5. The physician must have a valid encrypted identifier in both the Medical Services Plan and Pharmaceutical Services Division databases at the Ministry of Health Services

top

1.2 Patient Inclusion and Assignment Criteria

  1. Patients are assigned to the portrait of the physician who initiates their PPI therapy (writes their first PPI prescription that gets filled at a community pharmacy).
  2. Patients must be residents of B.C. and registered with the BC Medical Services Plan.
  3. Patients must possess a valid Personal Health Number in the BC Ministry of Health administrative databases.

  top

2 Page-2 Top Bar Charts: Duration that Your Patients Are on a PPI

2.1 Overview

The side-by-side vertical bar charts show the duration patients are on PPI therapy or PPI+NSAID therapy from 2004 to 2008.  Within these charts, the horizontal axis show the number of weeks of therapy, and the vertical axis show the number of patients.

The study period is from 2004 to 2008; patients treated entirely outside of this period are not included in the portrait data.

Page 2 Top Bar Charts

  top

2.2 Data

PPI and NSAID claims are extracted from the B.C. Ministry of Health’s administrative claims database.  All PPI prescriptions filled at a community pharmacy for patients not federally insured (RCMP, Veterans, federal inmates, etc) are included in the portrait.

The data range for PPI claims selection is Jan 01, 2003 to March 31, 2008. One year of data prior to the study period is used to improve the duration of therapy calculations.

The data range for NSAID claims selection is Jan 01, 2002 to March 31, 2008. The additional data prior to the study period is needed to determine overlapping therapy with a PPI.

Marketed combination products intended for treatment of H. pylori have been excluded from the data.

  top

2.3 Methods

 

Episodes of Drug Therapy

A patient may have more than one course of PPI therapy during the study period and, therefore, may be counted in these charts more than once.  These cases are handled by calculating Episodes of Drug Therapy (EDT).  EDTs are defined according to the following rules:

  1. An EDT may contain one, or a series of, prescription claim records for a patient, prescribed by the same physician, for a defined group of drugs, where the lag between the dispensed date of a new claim and the ending date of the preceding claim is less than a defined ‘buffer’ period.
  2. The EDT will use a buffer period of 60 days to allow for gaps between a prescription start date and the end date of the preceding prescription.
  3. The end date is calculated as the prescription service date plus the number of days supply dispensed.  This inherently assumes full compliance.
  4. EDT time will be the sum of Day Supply Dispensed; it will not be the calendar duration of the series of prescription claims.

In rare cases, when a patient has multiple prescriptions for the same drug on the same day, we will assume full compliance with each prescription as sequential usage (e.g. treat it as vacation supply).

PPI + NSAID Definition:  If NSAID therapy occurs within 60 days before the start of a PPI EDT, or if NSAID therapy occurs within 60 days after the end of PPI EDT, then that PPI EDT is defined as PPI+NSAID.

Where EDT’s for a PPI and an NSAID overlap, the duration is based on the PPI claims ONLY.  Example: if a 30 day supply of a PPI is overlapped by a 90 day supply of an NSAID, the duration of therapy is 30 days.

  top

Calculating Duration of Therapy

In cases where a patient’s first PPI prescription is in the first year of the study period, claims in the 1-year period prior are used to determine duration of therapy.

The PPI only (without NSAID) chart and the PPI+NSAID chart are mutually exclusive.

No data beyond Dec 31, 2008 is used to determine duration of therapy.  Example: If a patient receives their first Rx for a PPI on Dec 30, 2008 for a 15 day supply they will be classified as having a duration of therapy of “1-4 weeks” (refills are not looked for beyond Dec 31, 2008)

Duration of therapy is calculated using the “Days Supply Dispensed” field. Summed days are converted to time categories according to the following:

Chart Duration

Sum of Days Supply Dispensed

<1 week

0 - 6

1-4 weeks

7-31

5-10 weeks

32-73

11-20 weeks

74-143

21-52 weeks

144-365

52+ weeks

366

  top

2.4 Issues

  1. Patients are assigned to the portrait of the physician who initiates their PPI therapy (writes their first PPI prescription that gets filled at a community pharmacy).  Overlapping NSAID therapy may have been prescribed by a different physician.
  2. Out-of-province prescriptions or medication dispensed in a hospital is not included in this portrait.
  3. Drug wastage or other non-compliance may alter a patient’s actual duration of therapy.
  4. Individualized prescribing data is produced using encrypted administrative data provided by the BC Ministry of Health.  Potential inaccuracies in the source data may lead to incomplete or inaccurate individual prescribing portrait results.

  top

3 Page-2 Bottom Bar Charts: Your Patients Who Reduced Their PPI Use

3.1 Overview

This vertical bar chart shows the percentage of patients who reduced their PPI therapy or switched to an H2RA or a less costly PPI.  The individualized physician’s patients are represented by the blue columns; the example evidence based practice is represented by the white columns.

Page 2 Bottom Bar Charts

3.2 Data

PPI claims data and Episodes of Drug Therapy (EDT), as described in Section 1.3, are used to generate the data in this portrait.

 top

3.3 Methods

Maintained Dose/Frequency

 

Because of the variation between PPIs in dosing and strength, a Defined Daily Dose[1] (DDD) is applied to each Drug Identification Number (DIN).   Dose is then calculated by dividing the DIN’s strength by the associated DDD.

A patient has “Maintained Dose” if the dose of the patient’s first PPI prescription in the EDT is equal to the dose of the last PPI prescription in the EDT.

Frequency is calculated by looking at the total days (time) between the patient’s 1st and 2nd PPI prescription in an EDT divided by the number of DDDs dispensed.  The same calculation is done for the patient’s 2nd last and last PPI prescription in an EDT.  If the frequency calculated for the last two prescriptions is within 25% of the frequency calculated for the first two prescriptions then the patient has “Maintained Frequency”.

If a Patient’s EDT satisfied both “Maintained Dose” and “Maintained Frequency” they are counted in the numerator.  The denominator is the total number of PPI EDT’s initiated by the physician with 2 or more PPI prescriptions.

 

 

Reduced Dose/Frequency

 

Dose and Frequency are calculated as described above.

If a patient’s EDT has a lower dose on their last prescription versus their first prescription and the frequency decreases by more than 25%, they are counted in the numerator.  The denominator is the total number of PPI EDT’s initiated by the physician with 2 or more PPI prescriptions.

 top

 

 

Switched to Rx H2RA

The process for determining if a patient’s PPI EDT ended with a switch to an H2RA is outlined in the following steps:

  1. Generate all H2RA EDT’s for each PPI patient using the same EDT algorithm used to create the PPI EDT’s (outlined in Section 1.3 above).
  2. Identify PPI EDT’s where the patient has stopped their therapy (adding an H2RA without stopping their PPI therapy is not counted).  Stopping is defined as not having a subsequent PPI prescription within 60 days after the patient’s supply has ended. The ‘Stop Date’ is defined as the last PPI prescription dispensed date plus the number of days supply dispensed.
  3. Cross reference, for each patient, all H2RA EDTs with PPI EDTs where the patient has stopped therapy.  If the start date of the H2RA EDT is between the dispensing date of the last PPI in the PPI EDT and the PPI stop date plus 60 days then set a potential stop flag to ‘Y’.
  4. If a PPI EDT has a stop flag equal to Y and the PPI physician is the same physician who initiated the H2RA EDT then the PPI EDT is counted as “Switch to Rx H2RA”
  5. The denominator is the sum of all PPI EDTs, for each physician, where the date of the last prescription plus the number of days supply dispensed is not within 60 days of the end of data (if so we are unable to determine stopping).

 top

Switched to Less Costly PPI

 

Switching to a less costly PPI is determined by comparing the first prescription in the PPI EDT to the last prescription.  The ranking of PPIs from highest cost is[2]:

  1. Esomeprazole (Nexium) – $2.25
  2. Pantoprazole (Pantoloc) – $2.16
  3. Lansoprazole (Prevacid) -  $2.14
  4. Pantoprazole generic – $1.47
  5. Rabeprazole (Pariet) – $1.39
  6. Omeprazole generic – $1.26
  7. Omeprazole (Losec) – $1.18
  8. Rabeprazole generic – $0.97

The denominator is the total number of PPI EDT’s initiated by the physician with 2 or more PPI prescriptions

top

 

 

Stopped PPI Therapy

 

For each PPI EDT, stopping therapy is defined as having no further PPI prescriptions for 60 days after the end of the days supply dispensed from the last prescription date.  Where the last prescription date plus the number of days supply dispensed is within 60 days of the end of data we do not define the PPI EDT as stopping.

Evidence Based Targets

 

Within each of the five categories the prevalence of each possible value [1-100] is calculated.  Each of the five categories is analyzed separately.  The physician value at the 75th percentile is used as the evidence based target value.  The exception is the “Maintained Dose/Frequency” category where the 25th percentile value is used.

top

3.4 Issues

  1. Stopping PPI therapy is likely over-estimated as patients who move or switch physicians are counted as stopping PPI therapy.
  2. The five categories on the horizontal axis do not have consistent denominators.  This may lead to confusion and physicians may find it difficult to recognize their own prescribing patterns.
  3. The evidence-based practice figures are based on prevalence data; this may not be entirely consistent with evidence-based guidelines or other literature.
  4. The horizontal axis categories are not mutually exclusive and therefore will not necessarily add to 100%.

4 Page-3 Top Cost Tables: The Cost of Your Prescribing 2004-2008

4.1 Overview

The top two tables on Page-3 show PPI and H2RA utilization and costs over the study period split by chemical and brand vs generic products.

Each row contains the distinct number of patients receiving the product, the total cost of prescriptions in the study period, and the typical daily treatment cost for the product.

The last row in each table provides a distinct patient count - number of patients who took any PPI (H2RA) – and the total cost of all PPI (H2RA) prescriptions.

Page 3 Top Cost Tables

top

4.2 Data

PPI and H2RA claims for each physician are extracted from the B.C. Ministry of Health’s administrative claims database.  All PPI prescriptions filled at a community pharmacy for patients not federally insured (RCMP, Veterans, federal inmates, etc) are included in the cost tables.

 

4.3 Methods

Drug costs are calculated based on the dispensing date of the prescription.  Only the product/ingredient cost portion of the prescription is included in the cost tables; dispensing fees and special service fees are excluded.  The full amount of the product cost for claims dispensed between January 01st, 2004 and Dec 31, 2008 are included.

Patients who are prescribed more than one type of PPI/H2RA are counted more than once.  Patient totals are distinct patient counts.

 

4.4 Issues

  1. The cost tables show counts of patients whereas the graphs on page 2 show Episodes of Drug Therapy (EDTs).  This may cause a perceived discrepancy between page-2 versus page-3 figures.
  2. The column “Total Cost of your Prescriptions” does not account for patient duration.  Physicians may be confused by the discrepancy in their prescribing costs in relation to the difference in typical daily treatment cost.  This confusion most likely arises because of a difference in treatment duration between the PPI products.   The difference in product availability will make this a common occurrence.

top

5 Page-3 Bottom Line Graph: Your PPI and H2-Antagonist Prescription Trends

5.1 Overview

The line chart shows the annual prescribing trends by PPI chemical and the average of all H2RA chemical prescribing.  Calendar year is shown on the horizontal axis and the number of prescriptions is shown on the vertical axis.

Page 3 Bottom Line Graph

5.2 Data

PPI and H2RA claims for each physician are extracted from the B.C. Ministry of Health’s administrative claims database.  All PPI prescriptions filled at a community pharmacy for patients not federally insured (RCMP, Veterans, federal inmates, etc) are included in the cost tables.

top

5.3 Methods

The data for the utilization line chart is created for each physician according to the following steps:

  1. Count the number of prescriptions prescribed for each PPI chemical by calendar year
  2. Count the number of H2RA prescriptions prescribed by calendar year
  3. Divide the number of H2RA prescriptions (in Step 2 above) by 4 to create an average (cimetidine, ranitidine, famotidine, nizatidine). 

 


[1] Defined Daily Dose is based on the World Health Organizations DDD Index at http://www.whocc.no/atc_ddd_index/

[2] Prices are based on PharmaNet allowable cost per tablet for the standard dose as recommended by the Compendium of Pharmaceuticals and Specialties as of the time of portrait production.


Author: Greg Carney | Date Created: July 21, 2009 | Last Updated: June 17, 2010 | Version: 2.2


 

 top

 

 Comment