On this page: Aim | Overview of model and the interpretation of its results | Key results | Structure of this annex | Sources of complexity - interdependency and data paucity | Discussion of the model design | Assumptions and levers | Analysis tables | Projections: First year and 2017 - 26 present value | Results | Sensitivity analyses | Discussion | Output
Scenarios 2, 3, and 4 are expected to have an impact on the purchase and use of low dose codeine and, in the case of Scenario 4, on the decisions to seek treatment for both dependence and the conditions such as chronic pain that consumers are currently treating with OTC low dose codeine combination medicines.
The aim of the economic model is to estimate the projected costs and benefits of each scenario in relation to both of these broad consequences.
Overview of model and the interpretation of its results
Scenarios 2, 3, and 4 are intended to achieve public health outcomes relating to codeine misuse and abuse; however, these gains could come at a cost to the majority of consumers, who do not misuse low dose combination codeine and who will be required to attend a GP to obtain a script or substitute with other OTC analgesics. This comparison of the expected consequences, both costs and benefits is routinely assessed as part of economic and social models that support regulatory impact assessments.
However, part of the expected health benefits (improved quality of life) are not the result of the Scenario 4 per se. Instead, they are the result of additional treatment options being pursued by patients if OTC low dose codeine combination medicines were not available. Patient behaviour has to change and other costs have to be incurred to achieve these intended and expected health consequences. The benefits are available to a broader range of consumers other than the small group who might misuse or abuse OTC low dose codeine combination medicines, including those whose migraines or chronic pain could be treated more effectively with prescription only medicines or by specialised pain centres. This assessment of the costs of additional treatment compared to the health gains is conventionally addressed through health economic models such as those used routinely by the Pharmaceutical Benefits Advisory Committee (PBAC) and the Medical Services Advisory Committee (MSAC). The economic model has a health economic component in addition to the conventional economic and social impact component to ensure that it accommodates both the costs and benefits of the expected health gains.
The economic and social net benefit of the each scenario is reported as a net benefit. This net benefit represents the sum of additional costs (negative figure) and benefits (positive figure) derived from the economic model, where these additional costs and benefits are an aggregation of the consequences for:
- the 'improved treatment' group;
- the same treatment group who continue to use prescription low dose codeine combination drugs; and
- those consumers who switch to other OTC medications including paracetamol and/or ibuprofen.
The net economic benefits of each option and scenario are presented in Table E1. The economic benefits include the gain in quality of life for additional people who improve treatment, deaths prevented and net financial savings due to reduction in expenditure on low dose codeine medicines. The economic costs include the additional costs to consumers, MBS and PBS of additional medications, GP consultations and specialist consultations. The gains in quality of life (measured as QALYs) and deaths prevented are monetised at the rates recommended by OBPR. Gains in quality of life and deaths prevented are experienced only in Scenario 4. The net economic benefit remain positive for Scenario 4 under a full range of sensitivity analyses.
|Element||Scenario 2||Scenario 3||Scenario 4|
|Option 2||Option 5||Option 3||Option 5||Option 4||Option 6|
|Economic costs (PV 2017-26 at 7% $M)||($20.70)||($409.87)||($14.49)||($409.87)||($56.03)||($209.87)|
|Economic benefits (PV 2017-26 at 7% $M)||0||0||0||0||$243.95||$5,353.17|
|Net benefit (option basis) ($M)||($20.70)||($409.87)||($14.49)||($409.87)||$187.92||$5,143.30|
|Net benefit (scenario basis) ($M)||($430.57)||($424.36)||$5,331.22|
Structure of this annex
This annex first sets out the reasoning behind the economic model, with clear illustrations of the potential sources of bias in the analysis and how these were resolved.
Next, the document steps through the economic model, summarising each of the key components: the model's levers and assumptions; the first year analysis; the projections, and the results. Model levers are the model inputs that can be changed in order to explore the results under a range of scenarios. These inputs relate to baseline use of low dose codeine medicines, changes in behaviour in response to each option, and the consequences of these changes. Levers are essential in this model where there is limited evidence and data as well as limited agreement across the range of stakeholders in relation to their actual value. These levers ensure that the base case can easily be respecified in response to different stakeholders understanding of the base case and the impact on the net benefit re-assessed.
Sources of complexity - interdependency and data paucity
The first source of complexity in developing this model relates to the interdependency between the responses by the three main stakeholder groups (consumers, GPs and pharmacists) to Scenario 4. In summary, potential actions by these groups are:
- Consumers. Will need to decide whether they go to their GP, consider the treatment options provided by the GP, and/or use OTC non-codeine analgesic such as paracetamol and/or ibuprofen.
- GPs. Will have a new patient group for whom they can suggest a range of additional options (appropriate for the patient), including referring patients to specialist pain units or allied health providers; providing scripts of low or high dose codeine or other prescription only medicines; and/or suggesting OTC pain relief medications.
- Pharmacists. Will continue to interact with consumers who are seeking pain relief, such as those who require prescription medications, some of which will be new to the consumer, and consumers continuing to attend the pharmacy seeking pain relief without any prescriptions for medications.
The second source of complexity was the paucity of data that could inform the estimate of the key benefit of Scenario 4; that being the proportion of current users of OTC low dose codeine will experience a health gain, including preventing deaths, as a consequence of this change. KPMG was unable to obtain any data to inform the following estimates:
- the number of people who are currently dependent on low dose codeine medicines (note high dose medicines specifically excluded);
- the number of adverse events attributable to low dose codeine medicines(excluding high dose codeine medicines); and
- the number of people who use low dose codeine medicines chronically and, while currently not dependent, are at risk of dependence.
The critical issue in developing the economic model is the combined effect of these two complexities (interdependency between responses and paucity of estimate data). Unless there are changes in treatment (the cumulative result of changes in activity and/or decisions made by the three groups identified above) there will be very limited changes in health outcomes. Scenario 4 in itself will not produce the expected benefits; rather, it will be the changes in people's activity and behaviour that realises the benefits.
Discussion of the model design
Complexity of this model
The complexity of the model is primarily driven by the use of five separate consumer groups, rather than relying on the concept of an 'average' consumer. This segmentation of the population allows the plausibility of responses to the proposed rescheduling to be assessed more meaningfully and also reduces sources of unintentional systematic error that result from working with a model based on the characteristics of an 'average' consumer. However, this approach does increase the number of required assumptions, an important consideration in this data-poor area. A possible consequence of these additional assumptions is that, when these assumptions are all varied between base, worst and best cases, a very wide range for any resultant metric, such as a net benefit, emerges. The model incorporates a set of analyses that identify the most plausible combinations of these parameters, before conducting the sensitivity analyses, hence addressing this risk.
The third feature of this model is its conservative approach to estimating the size of the projected benefits of Scenario 4. (See discussion in Section 2.3 on input uncertainty.) This approach entails ensuring that the base case parameters for estimating these benefits is conservative, and that the key mechanisms by which this gain is achieved, as articulated by stakeholders, is clearly explained. Potential offsets to these gains, also identified by stakeholders, are also incorporated into the model.
The five main considerations
Listed below are five key considerations that the model addresses and which, collectively, reduce the systematic bias that may arise from a simpler model. These considerations are:
- Why is the concept of an 'average' consumer potentially misleading?
- At what point does using an 'average' response model for Scenario 4 produce a biased estimate of the increase in GP visits?
- What are the five consumer groups and what are their differences?
- What are the drivers of the model's benefits?
- How are the model's key parameters determined?
A pre-modelling exercise was performed to explore these issues, using hypothetical data. In the following sections, the results of this pre-modelling exercise are presented and the implications of these results for the model developed for this regulatory change are discussed.
The average consumer
Table E2 provides a hypothetical example of a market with 14 million sales and 1.1 million consumers. It is straightforward to calculate the average use per consumer; 12.7 packs per year. In this case, if 9% of current consumers stop using low dose codeine medicines post up-scheduling, and the remainder continue their use, then there is a reduction in volume of sales of 9%. However, if it assumed that there are two types of consumers: those with an average of two packs per year, and those with an average of 120 packs per year (two to three packs a week). The latter group is 9% of the market, but represent 86% of the sales.
|Low use consumers||High use consumers||Total use|
|Number of consumers||1,000,000||100,000||1,100,000|
|Packs per consumer per year||2||120||12.7|
|Packs per year||2,000,000||12,000,000||14,000,000|
|% of consumers||91%||9%||100%|
|% of packs||14%||86%||100%|
The implications of the above calculations (using hypothetical data) are as follows:
- While there are only a low percentage of consumers (9%) with potential misuse or abuse concerns, they account for the majority of the overall use (86%).
- For consumers using 2 packs per year, one would expect a different response to the shift from S3 to S4 when compared to the second type of consumer that use 120 packs per year.
- The demand for codeine containing medicines by low use consumers is more likely to be price elastic; they are likely to respond to the introduction of an additional cost to obtain the product by reducing their demand for that product, and substituting it with less expensive alternatives. They may reduce use if required to pay to see a GP in addition to the cost of the medicines.
- The demand for codeine containing medicines by high use consumers is more likely to be price inelastic; they will be less likely to reduce their demand in response to a price increase. However, it would also be infeasible for them to visit a GP every week (or multiple visits per week) if they had to obtain a script for each pack used. Therefore, their principal constraint is their time; although they may be willing to visit a GP 12 times a year, they are far less likely to make 50 visits per year. (This pre-modelling exercise initially assumes that scripts have no repeats and hence there is one GP consultation per script.)
- The health benefits are most likely to accrue in the smaller group of high use consumers.
- Problem users might use several pharmacies for supply and hence (in the absence of a tracking system) no single pharmacy has an accurate picture of their overall use.
The average response calculation
A pre-modelling exercise was undertaken to assess the potential limitations of using an 'average patient' approach as a basis for the model's structure. This approach was used in the Cadence report concerning the financial implications of Scenario 4. Noting the absence of additional data to inform the calculation of the number of additional GP visits arising from the Scenario 4, if it is assumed, for the purpose of this pre-modelling exercise, that: 1) annual sales are 14 million packs, and 2) a consumer will only go to their GP to obtain a script for codeine 40% of the time, then this would lead to an additional 5.6 million trips to the GP and the total volume of sales would be 40% of the original value (see Table E3).
|Packs per year||14,000,000|
|% of occasions where the consumer go to the GP||40%|
|Additional trips to the GP||5,600,000|
Given this result, some reasonable assumptions about Scenario 4 are:
- the GP will be able to provide up to five repeats (based on the level of repeats associated with current S4 codeine containing medicines for private scripts); the 'no repeats' scenario is unlikely to apply;
- not all trips to the GP will be solely for the purpose of obtaining a prescription for codeine as some patients will request a script as part of a consultation that would otherwise have occurred;
- different consumers will respond differently to the change, depending upon whether they have acute or chronic pain; and
- GPs will also respond to the actions of their patients; if their patients come every two weeks, for example, for an additional supply then they may refer them to an alternative service such as a pain management clinic.
Table E4 shows that, by incorporating these reasonable assumptions the volume of additional GP visits is reduced to 540,000 from the original estimate in Table E3 (5,600,00). It should be noted that the same base numbers are used as for the previous table. It should also be noted that the assumption that five repeats will be available (rather than zero repeats) is not the only driver of the lower number of GP visits. If only one repeat (or zero repeats) were available, this approach would identify that the high use group would require an implausible number of annual trips to the GP and hence this would act as a constraint on both their volume of scripts and the number of additional GP visits as a response to Scenario 4. The average response approach does not identify this.
|Category/Input||Hypothetical Value||Assumptions/Comments - illustrative only|
|Low use consumer|
|Pre changes to scheduling|
|Number of people||1,000,000|
|Packs per person per year||2|
|Post changes to scheduling|
|% of consumers who get a script||20%||Most people will use OTC paracetamol and/or ibuprofen instead of low dose codeine medicines|
|Consumers who get a scripts||200,000|
|Packs in year per person||2||For those who continue, the same number of packs a year are used|
|Additional GP visits per person||0||Most of the population has at least 2 visits to a GP a year and this would be absorbed into existing visits|
|High use consumer|
|Pre changes to scheduling|
|Number of consumers||100,000|
|Packs per year||120|
|Post changes to scheduling|
|% of consumers who get a script||90%||Most of these consumers will continue to use low dose codeine medicines. The continued use of low dose codeine as the only option other than OTC pharmacotherapy is consistent with the Cadence modelling. (In the actual economic model, it is assumed that some of these consumers will have high dose codeine medicines and others will have other pharmaco-therapy and specialist pain care. This is consistent with the advice from stakeholders who supported the up-scheduling.)|
|No. of consumers who get a script||90,000|
|Maximum additional visits|
|Additional GP visits (0 repeats)||120||If there were no repeats, then this would result in 120 visits to a GP a year, which is implausible.|
|Additional GP visits (5 repeats)||20||With 5 repeats, the patients could go 20 times a year, but that would still mean they were a very frequent user of GP services.|
|Additional GP visits (max with 5 repeats)||6||It is more reasonable to assume they would limit themselves to 6 additional visits a year.|
|Plausible additional visits|
|Max per year||6|
|Plausible packs per year|
|Total GP visits||12||Assuming that they would otherwise have had 6 visits, they now have 12 visits a year. This means that half of their visits when they obtain a script are visits that would otherwise have occurred.|
|Packs - with 5 repeats||72||Consumers could have 6 additional visits a year, 12 in total, and then a script with 5 repeats at each occasion which is a total of 72 packs (12 visits x 6 packs (5 repeats + 1 pack – original dispensing))|
|Total population - the combined effect of the two|
|% of consumers who get a script||26%||The combined effect is that 26% of consumers continue as patients (290,000/1,100,000)|
|No. of consumers who get a script||290,000||(200,000 (low use consumer) + 90,000 (high use consumer))|
|Packs per year per consumer||23.7||The average packs per patient is now 23.7 (6880000/290000), which is higher compared to the average preceding the scheduling change. This is a consequence of two factors. First the identification of two different groups of consumers allows group to change their use at a different rate. Second, the denominator (the number of people who continue to consume) is reduced.|
|Total packs per year||6,880,000|
|As % of previous packs||49%||The volume is now 49% of the previous volume|
|Additional GP visits total||540,000||There are 540,000 additional GP visits|
The average response approach and consistency check against the results of the Cadence Economics Report - Fiscal Impact of Codeine Changes
KPMG reviewed the methods and results of the Cadence Economics report, which predicts an additional 8.7 million GP consultations each year as a consequence of the proposed up-scheduling of codeine to S4. There are two key differences between the KPMG and Cadence economic modelling approaches: one methodological the other relating to a key assumption. The key methodological difference is the Cadence model uses an 'average' consumer concept, while the KPMG model used a segregated population (refer to Figure E1). The differing key assumption was that the Cadence model assumed no repeats while KPMG has assumed 5 repeats. These two key differences were assessed to be the primary drivers of the different results produced by the two models.
The five groups of consumers
The five groups of consumers are set out in Figure E1. Codeine use can be either therapeutic or non-therapeutic. Therapeutic use can be for acute pain and consumers might purchase only one or two OTC packs a year and use it in a way consistent with the product advice: i.e. for no more than three days in a row without seeing their GP and no more than eight tablets a day.Therapeutic use for chronic conditions such as chronic pain can lead to dependence. In the model, dependent consumers use more than the recommended maximum dose each day, whereas non-dependent consumers use the maximum recommended dose each day. All chronic users use the low dose codeine medicines for the majority of days in a year, which in the base case we assume to be 250 days. They are also assumed to be more likely to use the largest pack size (40 tablets) compared to acute users. Consequently, a regulatory change that restricts pack size to three days' supply will impact on this group of consumers more so than acute users.
Chronic therapeutic use is more likely to result in adverse events such as gastrointestinal bleeds as a consequence of the use of ibuprofen for longer periods and at higher daily doses than recommended. They are also more likely to have an accidental death due to an overdose of codeine, although this is considered to be a very rare event. Consumers who use therapeutic low dose codeine medicines for chronic pain are at greater risk of becoming dependent in comparison to users that self-treat acute conditions.
While non-therapeutic use is often referred to in the media, it is likely to be only a very small proportion of consumers in this group. KPMG has taken a conservative approach and included this group to account for the total volume of sales. However, we do not account for any benefits that might result from the proposed rescheduling for this group. KPMG does assume that their use of low dose codeine medicines will be limited substantially if they are required to go to a GP to obtain a script. However, we have assumed that they are likely to substitute this with other forms of recreational drug use.
Figure E1: Patient/consumer groups used in the health economic modelling
The drivers of gains in health outcomes
If low dose codeine medicine is only available by prescription, then the health gains compared to the existing situation (low dose codeine OTC medicines) are driven by changes in treatment and therapy. This change could arise because they discuss alternative treatment options with their pharmacist or, alternatively, they visit their GP, who then either maintains their current use of low dose codeine medicines or changes their therapy (Figure E2). This change may include an alternative analgesic medicine that is more suitable, or a diagnosis of their condition resulting in treatment and ultimately reducing the need for analgesic use. There are also other treatment options that include non-pharmacological options, referrals to allied health professionals or specialist pain clinics. If patients are referred to these services by their GPs, then in many situations, patients will be able to obtain a Medicare rebate for the service. However, there are currently waiting lists for pain clinic services, so capacity constraints must also be considered.
The critical point here is that the same treatment options are available to the consumers before and after Scenario 4 is enacted. The impact of the up-scheduling to Schedule 4 does not introduce the other treatment options but does make the existing treatment options more likely to be explored. The potential health impact is to increase the chance that a patient, who would have better treatment option in comparison to low dose codeine medicines, explores these options with their pharmacist or their GP as a consequence of Scenario 4 being enacted.
Figure E2: One option is excluded as a consequence of Scenario 4
Determining key modelling parameters
Four steps are taken to determine the key modelling parameters given available data (sales data from IMS), reasonable assumptions about the number of tablets consumed per day, days of use for each of the five types of consumers, and the share of total consumers in each group. This method is illustrated in Figure E3. These four steps are outlined below, with the inputs to steps 1), 2) and 3) having been informed by discussions with stakeholders:
- The total number of sales by pack size (dollar value and packs sold) was calculated from the IMS data. The total retail sales value by pack size was calculated by assuming a 44% retail mark up.
- The number of tablets per consumer, from 6-20 tablets per day over 12-365 days per year depending on the type of consumer:
- Therapeutic, acute pain, non-dependent – 8 tablets/day over 12 days/year.
- Therapeutic, chronic pain, non-dependent – 8 tablets/day over 250 days/year.
- Therapeutic, chronic pain, dependent – 12 tablets/day over 250 days/year.
- Non-therapeutic, occasional use, non-dependent –6 tablets/day over 20 days/year.
- Non-therapeutic, regular use, dependent –20 tablets/day over 365 days/year.
- The proportion of consumers in each group:
- Therapeutic – 99%:
- Therapeutic, acute pain, non-dependent – 80% of all users
- Therapeutic, chronic pain – 19% of all users of whom:
- 40% are Therapeutic, chronic pain, non-dependent.
- 60% are Therapeutic, chronic pain, dependent.
- Non-therapeutic– 1%, of whom:
- 10% are Non-therapeutic, regular use, dependent.
- 90% are Non-therapeutic, occasional use, non-dependent.
- Therapeutic – 99%:
- The pack sales were then allocated by the model across consumers, starting with the smaller packs allocated to the lower use consumers, and with the maximum pack sizes allocated last to the most frequent users.
This approach allowed the calculation of the average pack size and expenditure per consumer group. In turn, this enabled the assessment of the plausibility of changes to current behaviour based on a more accurate picture of the current pattern of use.
Figure E3: Determining key modelling parameters
Assumptions and levers
The model includes two types of inputs:
- Administrative inputs, such as the value of a statistical life year ($182,000), which was used to value a Quality Adjusted Life Year (QALY). These inputs are set via a range of protocols detailed in the economic model.
- A small number of inputs that are set in Analysis Tables 1 to 3, which need to be considered in the context of other information and hence are most appropriately entered in a table.
The input panel at the start of the model provides a range of inputs that can be tested. The details of these are set out in the model.
The model assumes very small gains in QALYs from treatment and, in the base case, only some of the consumers who use low dose codeine medicines for chronic pain are assumed to have active treatment as a consequence of Scenario 4 being enacted.
Consumers who are using low dose codeine medicines for non-therapeutic reasons are assumed, conservatively, to not have any health gains, and are not treated under Scenario 4. As these consumers are assumed to be only 1% of the total users of low dose codeine, this assumption is conservative.
It is reasonable to expect that the medications that are PBS listed for pain are cost-effective, assuming that most of these will have been part of the Pharmaceutical Benefits Advisory Committee (PBAC) process. It is also reasonable to expect that the MBS items for pain clinics also represent cost-effective care. The costs of these additional services to both the consumer and the Commonwealth are included.
The model comprises one output table and 5 analyses tables, two of which (Analysis Tables 1 and 2) also include inputs relating to current use of low dose codeine medicines per consumer, the proportion that will take each therapeutic pathway, and the resource use and QALY outcome of each pathway. Analysis Table 3 performs projections from 2017 to 2026.
The last two model tables set out the IMS data on sales for S2 and S3 medicines, the projected sales for 2017, the impact of the changes in pack sizes. These tables inform all Options. Option 6 is additionally informed by the assumptions set out in Analysis Tables 1 to 3.
Projections: First year and 2017 - 26 present value
Projections were performed for ten years. QALY and death results were projected for ten years and then the QALYs, and monetised QALYS and deaths were discounted using 7% (in the base case), to ensure consistency with the OBPR guidelines.
The base case assumes a constant rate of reduction in the total costs, and benefits, of treatment due to a decreasing rate of participation in treatment. Some stakeholders noted that the initial increase in the number of people who are additionally treated and experience a health benefit is likely to decline in a few years until the system recalibrates. That is, patients who are currently using low dose codeine medicines and who, following up-scheduling will pursue therapeutic pathways that improve health outcomes, are part of a cohort. They will receive additional treatment and care for the following year, but this additional treatment, compared to what they would otherwise have received, will reduce each year. The model captures this factor by assuming, in the base case, a 30% annual reduction on previous year's treatment and health gains.
Deaths were assumed to be prevented at the same number each year. A conservative assumption was made regarding the changing prevalence of dependence due to accessibility of low dose codeine medicines; it is assumed that there would be no increase in either, the overall use of these medicines, or in the prevalence of codeine dependence under Scenario 1, the current situation.
The overall results are presented in the RIS table in the model (Table ES2 of the Executive Summary). The main result is the net benefit, which is presented using 2 subset results for the first year and tenth year net benefit. These yearly results are further subdivided into: (1) first year benefit, cost and net benefit; and (2) the ten year NPV benefit, cost and net benefit. Option 6 is the only option with a net benefit hence all other options have a net economic cost, driven by additional out-of-pocket costs to consumers as well as the cost attributed to the administrative burden for businesses and individuals.
The following tables (Tables E5 and E6) presents the 2017 costs and benefits included in the economic model disaggregated by broad categories. (More detailed tables are presented in the Executive Summary)
|2017||Scenario 2||Scenario 3||Scenario 4|
|Economic costs||Option 2||Option 5||Option 3||Option 5||Option 4||Option 6|
|Additional Out-of-pocket costs to consumers||($2.76)||($54.63)||($1.93)||($54.63)||0||0|
|Additional costs to MBS||0||0||0||0||($7.47)||($51.97)|
|Additional treatment related GP consultations ($M)||0||0||0||0||0||($42.86)|
|Additional Low Dose codeine prescription GP consultations ($M)||0||0||0||0||($7.47)||($1.89)|
|Additional specialist consultations ($M)||0||0||0||0||0||($7.21)|
|Additional costs to PBS - for medicines already demonstrated to be cost effective||0||0||0||0||0||($23.06)|
|Total economic costs||($2.76)||($54.63)||($1.93)||($54.63)||($7.47)||($75.03)|
|2017||Scenario 2||Scenario 3||Scenario 4|
|Economic benefits||Option 2||Option 5||Option 3||Option 5||Option 4||Option 6|
|Improved quality of life (ex. Deaths prevented)|
|Total QALY gains||0||0||0||0||0||9,074|
|Monetised QALY gain||0||0||0||0||0||$1,651.52|
|Estimated deaths prevented||0||0||0||0||0||5|
|Monetised deaths prevented ($M)||0||0||0||0||0||$21.00|
|Net financial savings to consumers||0||0||0||0||$32.51||$40.21|
|Total economic benefits||0||0||0||0||$32.51||$1,712.73|
The following tables (Tables E7 and E8) present the disaggregated costs and benefits as present values over the period 2017 to 2026 discounted at 7%. The ten-year results are not simply the one year results multiplied by ten. The following factors drive the differences in the summary statistics in the following (2017 to 2026) and previous (2017) tables:
- Clinical stakeholders indicated that there would be an initial 'hump' in additional treatment for patients who would otherwise be chronic users (dependent or risk of dependency) but that this would taper off over time. Hence the additional benefits and costs of Option 6 are substantially less than ten times the single year (2017) results; that is, the additional benefits and costs are incurred each year at a reducing rate.
- A 7% discount rate was applied to the monetary values, which is consistent with OBPR guidance.
- The population of ongoing low-dose codeine medicines users is assumed to increase by the same rate as the ABS projected annual population growth (1.66%)
|2017-26 present value (7% discount rate)||Scenario 2||Scenario 3||Scenario 4|
|Economic costs ($M)||Option 2||Option 5||Option 3||Option 5||Option 4||Option 6|
|Additional Out-of-pocket costs to consumers ($M)||($20.70)||($409.87)||($14.49)||($409.87)||0||0|
|Additional costs to MBS ($M)||0||0||0||0||($56.03)||($148.43)|
|Additional treatment related GP consultations ($M)||0||0||0||0||0||($112.59)|
|Additional Low Dose codeine prescription GP consultations ($M)||0||0||0||0||($56.03)||($16.56)|
|Additional specialist consultations ($M)||0||0||0||0||0||($19.27)|
|Additional costs to PBS- for medicines already demonstrated to be cost effective ($M)||0||0||0||0||0||($61.43)|
|Total economic costs ($M)||($20.70)||($409.87)||($14.49)||($409.87)||($56.03)||($209.87)|
|2017-26 present value (7% discount rate)||Scenario 2||Scenario 3||Scenario 4|
|Economic benefits ($M)||Option 2||Option 5||Option 3||Option 5||Option 4||Option 6|
|Improved quality of life (ex. Deaths prevented)|
|Total QALY gains (PV at 7%)||0||0||0||0||0||24,173|
|Monetised QALY gain ($M)||0||0||0||0||0||$4,399.47|
|Estimated deaths prevented (not discounted)||0||0||0||0||0||50|
|Monetised deaths prevented (PV at 7%) ($M)||0||0||0||0||0||$147.50|
|Net financial savings to consumers (PV at 7%) ($M)||0||0||0||0||$243.95||$806.21|
|Total economic benefits ($M)||0||0||0||0||$243.95||$5,353.17|
Table E9 presents the economic costs and benefits for 2017 and for 2017-26 summarised as net benefits.
|Net economic benefit||Scenario 2||Scenario 3||Scenario 4|
|Option 2||Option 5||Option 3||Option 5||Option 4||Option 6|
|Economic cost ($M)||($2.76)||($54.63)||($1.93)||($54.63)||($7.47)||($75.03)|
|Economic benefit ($M)||0||0||0||0||$32.51||$1,712.73|
|Net economic benefit ($M)||($2.76)||($54.63)||($1.93)||($54.63)||$25.05||$1,637.70|
|For Scenarios ($M)||($57.38)||($56.56)||$1,662.74|
|2017-26 (PV, 7% discount rate)|
|Economic cost ($M)||($20.70)||($409.87)||($14.49)||($409.87)||($56.03)||($209.87)|
|Economic benefit ($M)||0||0||0||0||$243.95||$5,353.17|
|Net economic benefit ($M)||($20.70)||($409.87)||($14.49)||($409.87)||$187.92||$5,143.30|
|For Scenarios ($M)||($430.57)||($424.36)||$5,331.22|
Sensitivity analyses can be performed in two ways within this model. The first is a structured approach using the Scenario Manager tool. Seven sensitivity analyses have been programmed and they vary a total of nine key inputs, as set out in Table E10. An alternative approach is to use the input panels at the start of the RIS sheet. Any number of these inputs can be changed and the impact on the net benefit can be tested. Regardless of the variations made via the input panel, the Scenario Manager will always use the base case programmed into the tool, not the prevailing values of the variables.
The input panel was used to test the model as a quality assurance process and also to test the overall robustness of the results. The base case value of assumptions that were varied as part of testing the model are presented in Table E10. The values that were varied in the seven standard sensitivity analyses included in 'Scenario manager' to test the robustness of the final results are also included in Table E10. Other inputs were not tested in the sensitivity analysis as they were assessed to be administrative inputs, for example, the monetary value of a statistical year of life. The results indicate that only in extreme cases does the model predict a net loss with Option 6. An example of an extreme scenario is when the QALY gains from treatment are reduced by 30%, the treatment costs are increased by 30%, the loss in QALYs to a small proportion of consumers is assumed to be high and the treatment activity in the first year is expected to continue at a high rate for the next ten years.
|Selected Inputs||Base case value||Value in sensitivity analysis||Present value of net benefit (2017 to 2026)||Change compared to base case||Interpretation|
|Base case value: $ 5,143,299,606 (This is the present value of the annual net benefit over ten years)|
|Univariate sensitivity analysis|
|Deaths prevented per year||5||20||$5,585,785,243||9%||A study indicate that the deaths potentially attributable to OTC low dose codeine medicines are around 15-30 a year. However given these consumers have multiple complexities, only some of these deaths are assumed in the base case to have been prevented. The net benefit increases as the number of deaths prevented increases.|
|Discount rate||7%||10%||$4,705,920,607||-9%||OBPR guidelines require that the base case discount rate is 7% and is varied to 10% and 3% in univariate sensitivity analyses. The net benefit varies in a way that is consistent with expectations.|
|Repeats for a private script of low dose codeine medicines||5||0||$5,072,341,663||-1%||The less repeats in a private script of LD codeine, the more visits to a GP required for a patient to maintain current usage. This only applies to acute users (current S2 and S3) who continue to use these medicines. Some of these visits (75%) are assumed to have occurred without the need for the prescription. The less repeats the higher the cost to the consumer for the co-payment and the higher the cost to MBS. The impact on the net benefits is small because the main driver of the additional MBS costs is the S2 users, who are predicted to only require one pack a year and hence not require a repeat. The Medicare cost for the current S3 users increases by approximately 5 fold, however this is from a low base. The MBS costs for this group are driven by treatment related consultations. And finally, the net benefit is driven by the very high economic benefit relative to the cost.|
|% of all acute users of LD codeine who continue to use LD Codeine after the shift from S3→S4 at the same rate as they do currently: 5 packs per year||20%||25%||$5,137,897,522||0.1%||These are consumers who continue for the next ten years to use LD codeine for acute pain. The less who continue, the less GP visits, the lower the cost to Medicare and to consumers and the greater the savings from less low dose codeine prescriptions.|
|Of the consultations for the above group, what% are additional consultations. Assume all patients get maximum repeats, currently 5 packs per year||25%||50%||$5,125,560,120||0.3%||Only some of the consultations for the above group are additional. The greater the proportion that are additional, the higher the additional cost to consumers and Medicare.|
|% of consumers whose use is for acute pain||80%||90%||$4,829,883,483||-6%||This is the group who are least likely to participate in treatment hence the overall QALY gains are reduced, and the net savings from reduced expenditure are also minimised when this proportion is increased.|
|% of consumers whose use is chronic, who are also dependent||20%||50%||$5,169,648,377||1%|| |
Chronic therapeutic users represent 19% of all users in the model's base case.
At 20% of this group, dependent consumers represent 3.8% of all current users of low dose codeine medicines.
Chronic dependent users have the greatest potential for a health benefit (have a QALY gain) and will make the highest savings in expenditure on LD codeine. However their treatment costs are also expected to be the highest as they are more likely to require referrals to pain clinics. Hence the net effect on the net benefit of increasing the share of consumers in this group is small.
|% of consumers whose use is non-therapeutic||1%||2%||$5,061,830,607||-2%||No therapeutic users are assumed not to participate in the ongoing pathways and so incur neither additional costs nor additional benefits. Hence the larger the share of current consumers in this group, the lower the net benefit because less consumers experience the average net benefit per consumer at baseline, which is maintained.|
|% of therapeutic Chronic dependent and non-dependent users who are referred to pain clinics||10% and 5%||1% and 0% (instead have intensive GP management with smaller health gains and lower cost )||$4,960,883,222||-3.5%|| |
Chronic dependent therapeutic users represent 3.8% of all consumers in the model's base case and chronic non-dependent therapeutic users represent 15.2%. Hence, under the base case. 1.4% of all current consumers are expected to be referred to pain clinics. However, pain clinics are likely to have waiting lists. This sensitivity analysis reflects this situation.
Instead of going to a pain clinic, these customers are assumed to have multiple GP visits and use a range of prescription pain medications. Hence, there is still an additional health gain, and MBS and PBS cost for these patients. The impact on the overall net benefit is a small reduction.
|Annual change in additional treatments (projections)||-30%||0%||$1,423,923,121||-74%||With these benefit minimising and cost maximising assumptions, a net benefit is expected but is substantially reduced.|
|Change in the average QALY per treatment||0%||-80%|
|Change in the average cost per treatment (all costs - consumer, MBS and PBS)||0%||80%|
|Deaths prevented per year||5||0|
|Repeats for a private script of low dose codeine medicines||5||0|
The additional health gains are achieved at an additional cost to the system. The comparison of the health gains with the additional costs of achieving these gains is analogous to the PBAC assessment of the additional costs and benefits of a new medicine.
It is noted that:
- Typically, new medicines listed on the PBS come at an additional cost to the health system; however, if it can be demonstrated that the additional costs are justified by the additional benefits then the new medicine will be usually be listed on the PBS.
- Similarly, this proposed change in regulation comes with additional costs (additional visits to the GP, additional out-of pocket costs to the consumer, etc.); however, the additional costs are justified by the additional benefits (i.e. the net benefit is positive)
The economic model indicates that Scenario 4 is broadly 'cost effective'. One way of summarising and assessing the health and financial consequences of Scenario 4 is as a net benefit, where the QALYs and deaths are summarised as monetised values and compared to the costs. This approach is consistent with OPBR guidelines and presented in this report. Another approach, which is used by the Pharmaceutical Benefits Advisory Committee (PBAC) and the Medical Services Advisory Committee (MSAC), is cost effectiveness analysis: a comparison of the additional costs with the additional effects (QALYs) expressed as an incremental cost effectiveness ratio (ICER).
A formal cost effectiveness analysis was outside the scope of this project. It is possible to use the results generated by the economic model to make a broad estimate of the ICER of Scenario 4 compared to Scenario 1 (current practice). Using the results report in previous tables, the additional costs to MBS and PBS from Scenario 4 relative to current practice were compared to the additional QALYs, for both the first year and the present value over ten years and estimates $9,092 per QALY and $11,097 per QALY respectively are reported. (See Table ES16). Although the decision threshold for PBAC is not publicly reported, it is reasonable to conclude that at around $10,000 per QALY that these preliminary results indicate that Scenario 4 is projected to achieve additional QALYs at a cost per QALY that would be considered to be 'cost effective'.
|Cost and effect (QALYs) - all sourced form previous tables||2017||PV of 2017-26|
|Additional MBS low dose codeine prescription consultation costs ($M)||$9.36||$72.60|
|Additional MBS treatment consultation costs ($M)||$50.08||$134.22|
|Additional PBS costs ($M)||$23.06||$61.43|
|Total MBS and PBS costs ($M)||$82.50||$268.25|
|Additional QALYs (ex. Deaths) (QALYs) (Number)||9,074||24,173|
|ICER = additional cost/additional QALYS ($)||$9,092||$11,097|
The following drivers contribute to the 'cost effectiveness' of Scenario 4.
- Each ongoing pack of low dose codeine combination medicines purchased does not require an additional GP consultation. Specifically:
- up to 5 repeats are possible for consumers should they continue to use low dose codeine combination medicines resulting in up to 6 scripts per consultation; and
- for many consumers, this additional script is unlikely to represent an additional GP consultation and will instead form part of a consultation that would otherwise have occurred in the absence of Scenario 4 being enacted.
- For some consumers (namely ongoing low dose codeine medicine users) there is an additional cost but no change in health outcomes. However, the number of additional GP consultations for these consumers are likely to be substantially less than the number of packs they purchase (due to being able to obtain up to 6 scripts per consultation).
- The additional cost for pain clinics and some GP consultations are for treatment that would otherwise not have occurred in the absence of Scenario 4 being enacted. Treatment costs are accepted as a cost effective part of the health system.
- The additional treatment costs to PBS are largely for medicines already shown to be cost effective.
- There is a financial saving without loss of health benefits for most consumers who shift from low dose codeine combination medicines to paracetamol and/or ibuprofen OTC medicines. There is no loss in health outcomes because for most of these consumers, as the evidence indicates, there is no incremental health effect of the use of low dose codeine combination products compared to using these analgesics without codeine.
In many scenarios KPMG tested, there was a net reduction in out-of-pocket costs to the consumer. When this occurs, the model adjusts the resultant saving to be included as a benefit, not a negative cost. This net savings to consumers in Option 6 is the result of a combination of factors, including:
- the reduction in use of low dose codeine medicines;
- the substitution of low dose codeine medicines to cheaper supermarket products such as paracetamol and/or ibuprofen;
- patients who continue with prescription medicines (whether containing codeine or not) in most cases will pay the same or less than their current expenditure on low dose OTC codeine medicines;
- if patients substitute low dose codeine medicine with high dose prescription codeine medicine via script, they can be provided up to five repeats by their GP reducing the need for less visits to their GP;
- the high bulk-billing rate for GP consultations, which reduces the net additional cost to consumers, and;
- the rate at which pain-related GP consultations can be accommodated within visits that would otherwise have occurred in the absence of Scenario 4 being enacted.
One issue raised by stakeholders is that the predicted demand for additional consultations at pain clinics is unlikely to be accommodated within existing capacity. A sensitivity analysis that assumes only 0.38% rather than 1.4% of all current consumers will be referred to a pain clinic indicates that there would only be a small reduction in the net benefit if less patients were referred. This result occurs because these patients would instead have intensive pain management provided by their GP and still have health gains at an additional cost to the MBS for treatment (compared to the current situation). However, these gains are assumed to be slightly smaller and the costs slightly lower. (See the results of the sensitivity analysis reported in Table E10.)
Most stakeholders indicated that additional face-to-face education for prescribers and pharmacists was unlikely to be necessary. With an estimated one million people using at least one S3 low dose codeine medicine product a year, the need to invest in an education and awareness campaign, particularly for consumers is apparent. This issue was raised with stakeholders. The question of how an education campaign for consumers would be funded, and what form it might take is still to be determined and is dependent on the regulatory process changes, if any. The cost of this campaign was not included in these costings.
For some consumers, Scenario 4 will result in a reduction in out of pocket costs with no change in health status. Many consumers who currently use low dose codeine combination medicines occasionally (only three days at a time for acute pain) and are expected to substitute this medicine with codeine free alternative that do not require a prescription. Generally these will be at a lower cost to the consumer compared to the expenditure that would otherwise have occurred. However, on average, it is expected that there will be no loss in pain relief for these many of these consumer, as indicated by systematic reviews of clinical trials. Hence there will be financial savings to some consumers as they switch to less expensive medications, without a loss in pain relief.
The additional cost to the PBS (relating to Option 6 (Scenario 4))
Additional PBS items are an expected consequence of Scenario 4 being enacted. Some additional pharmaco-therapies used by patients as prescribed by their clinician as a consequence of their attendance at a GP for acute of chronic pain will incur a PBS subsidy. A key clinical stakeholder contact indicated this additional prescribing would occur and provided examples of the medicines prescribed but did not provide estimated numbers of patients who be prescribed these medicines. Another key stakeholder indicated that there would be additional PBS scripts now dispensed by pharmacies, including for medicines other than low dose codeine, and that pharmacists would need to provide information on these new (to the patient) pharmacotherapies.
The assumptions incorporated into the economic model include:
- that the current S2 and S3 codeine medicines would not be listed on the PBS;
- that a proportion of current consumers of S3 codeine medicines (some acute, mainly chronic) will be provided with alternative pharmacotherapy for chronic pain and acute pain such as migraines;
- the ongoing use of these PBS medicines would taper off);
- that these PBS medicines would be cost-effective for patients and hence the additional costs, while recorded, would be accompanied by an additional health benefit for patients (on average); and
- as there was no data on which to inform the exact mix of these additional medicines an average additional PBS cost was assumed.
These assumptions were generated as a result of discussions with key stakeholders and reasonable base case assumptions were then tested in the sensitivity analysis and the net benefit remained positive.
The additional cost to the MBS (Scenario 4)
As discussed above, the additional treatment options that will be pursued by some acute and most chronic users will require additional consultations.
It was assumed that only some patients who are currently chronic users would go to pain clinics, and the constraint on current resources was identified by a key clinical stakeholder).
In addition, some consumers will continue to use low dose codeine combination medicines. Although no stakeholders provided an estimate of this number, all stakeholders who commented on this factor indicated that there would be a substantive reduction in volume of low dose codeine combination medicines used. This ongoing use of low dose codeine combination medicines would require additional GP visits and the model's estimate takes into account the number of repeats (up to 5) and the fact that these patients would likely be attending a GP for some of these consultations (in the absence of Scenario 4 being enacted).
The cost per additional MBS visit (specialist and non-referred) was derived using the data provided by Medicare specifically for this project. This took into account a range of assumptions made by Medicare about the share of these patients who would be bulkbilled and the mix of level A, B and C consultations. This actual mix was not provided to KPMG, only the total cost to Medicare for a range of potential volumes of consultations each year.
The structure of the worksheet that presents the economic model is detailed in Figure E4.
Figure E4: Overview of structure of the Economic Model worksheet