You are here

Assessed listed medicines evidence guidelines

Version 1.1, August 2018

17 August 2018

Book pagination

5. Evidence requirements and standards

This guidance provides information on the general types and standards of evidence required to support an application for an assessed listed medicine. Specific data requirements and dossier formats for individual application categories are provided in Application dossier requirements.

5.1 Methods of establishing efficacy

5.1.1 Overview of methods of establishing efficacy for different application categories

Each application category has different methods for establishing efficacy of the product. These methods are generally summarised in Table 4 below. The methods of establishing efficacy for L(A)3 applications are described in more detail in Table 5.

Table 4: Overview of application categories and efficacy requirements
Application type Product type Method of establishing efficacy
L(A)1 Identical to an existing assessed listed medicine, other than a permitted difference as specified in Table 3
  • Assessment of label
  • Access to reference medicine dossier required
L(A)2 Generic medicine of a fully evaluated assessed listed medicine
Identical to a medicine evaluated by a Comparable Overseas Regulator (COR)
  • Full un-redacted COR evaluation report


  • Products not covered by L(A)1 or L(A)2
  • Includes new products or variation to an existing assessed listed medicine
Any type of product
  • Method 1
Isolated chemical substances (i.e. single chemicals, well-defined chemical complexes, prodrugs, amino acids, vitamins and minerals)
  • Method 2A
Products that meet the requirements for a compliant biowaiver and medicines that do not require biopharmaceutic studies or clinical efficacy studies (e.g. some aqueous oral solutions or some products containing substances that are not systemically or locally absorbed)
  • Method 2B

5.1.2 Methods of establishing efficacy for L(A)3 applications

All indications for assessed listed medicines must be supported by scientific evidence of efficacy of the product. Efficacy refers to the potential of a medicine to produce a beneficial therapeutic effect in tightly controlled circumstances relative to a placebo[6] or other interventions. Efficacy studies focus on demonstrating statistically significant differences between intervention groups in clinical settings.

It is important to note that efficacy is not the same as effectiveness:

  • Effectiveness refers to the extent of perceived or reported a beneficial effect under "real world" settings, and may be different than efficacy as a consequence of factors that are controlled or limited in clinical settings but not in real world use (e.g. different population groups, diets, etc.).
  • Efficacy studies focus on establishing a causal relationship between a treatment and an effect. While medicines eligible for the assessed listed medicines pathway may operate through different therapeutic modalities to conventional medicines, there is no difference in the nature of either the cause-effect relationships being assessed or in the outcomes being studied. Similar principles and standards of efficacy evidence apply to these products. As a result, assessed listed medicines cannot solely be supported by evidence of effectiveness (e.g. historical use), and must be supported by evidence of the efficacy of the finished product, rather than simply the efficacy of separate ingredients.

For L(A)3 applications, there are three methods via which applicants may provide evidence of the efficacy of the proposed product (see Table 5). These three methods are designed to ensure that there is a sufficiently high standard of evidence to support consumer confidence in the indications, while being sufficiently minimalist to enable access to the pathway and support innovation in the sector. In brief:

  • Method 1 utilises the common standard approach of clinical trials on the product, and is suitable for all product types, including traditional medicines, herbal medicines, probiotics and conventional medicines.
  • Method 2A uses a combination of efficacy data and bioavailability/bioequivalence data to support plausible efficacy of the product. It is appropriate for systemically acting isolated chemical substances (i.e. single chemicals, well-defined chemical complexes, prodrugs, amino acids, vitamins and minerals).
  • Method 2B uses ingredient efficacy data in combination with product dissolution/ release data or in vivo pharmacokinetic studies to support plausible efficacy of the product. It can only be used for products that meet the requirements for a compliant biowaiver[7]and certain medicines that do not require biopharmaceutic studies or clinical efficacy studies (e.g. some aqueous oral solutions or some products containing substances that are not systemically or locally absorbed)[8]. See Biopharmaceutic and pharmacokinetic studies for further information.

Table 5 specifies the minimum requirements and may vary depending on the product[9].

Table 5: Methods for establishing the efficacy of assessed listed medicines
Data type Method 1 Method 2A Method 2B
Suitable product types All types Systemically acting isolated chemical substances. Supported by a biowaiver, or not requiring biopharmaceutic or clinical efficacy studies.
Body of scientific information Full literature search report on the finished product[10], or all active ingredients and formulation. Full literature search report on all active ingredients and formulation. Full literature search report on all active ingredients and formulation.
Published studies or clinical study reports (see Table 7) Efficacy evidence on the finished product[11]. Efficacy evidence for each active ingredient. Efficacy evidence for each active ingredient.
Biopharmaceutic and pharmacokinetic evidence Not normally required

Evidence for efficacy of the product formulation, established through:

  1. bioequivalence data to existing products (consisting of evidence of release via dissolution data and absorption of the active ingredient via bioavailability data); or
  2. in some instances, comparative dissolution (against established data) demonstrating release of the active ingredient with appropriate scientific justification regarding bioequivalence.

In vitro dissolution/ release tests or pharmacokinetic studies demonstrating in vivo drug release and availability of the active ingredients at the site of action.

Appropriate scientific justification of the approach and validation of the approach where appropriate.

Formulation All methods must provide a justification of the use of the particular combination of ingredients, including potential interactions between the ingredients.

Note that Methods 2A and 2B are generally not appropriate for herbs, herbal extracts, substances of biological origin, or complex mixtures of chemicals. This is due to the fact that the variable chemical composition and, in many cases, lack of known active component makes it difficult to accurately demonstrate appropriate biopharmaceutical properties of the medicine. A sub-set of chemical markers is not a suitable proxy for establishing the bioavailability or bioequivalence of all active constituents of a complex substance.

Applicants may submit a detailed scientific justification if the data package convincingly demonstrates that bioavailability/bioequivalence data is not required (see Justifications).

Products that do not meet the evidence requirements for Methods 2A and 2B may either be assessed via Method 1, or must be listed via the listed medicines pathway and use only permitted indications.

5.2 Types of applications

Some of the above evidence requirements for assessed listed medicines can be met through:

  • Conventional applications - primarily contain full study reports of company sponsored studies that support the efficacy of the product. These studies can be supported with bibliographical references.
  • Literature-based submissions - rely on bibliographic data or overseas reports to support the efficacy of the product (See Literature-based submissions for complementary medicines). The supporting literature must be relevant to the application. For example, the information should generally relate closely to the formulation, dosage regimen and indications of the proposed product. Unlike for prescription medicines, you do not need to gain approval of literature search strategies prior to submitting your application, and there is no formal pre-submission phase.
  • Mixed applications - consist of a combination of full study reports of limited clinical studies carried out by the applicant supplemented with bibliographical references to support the efficacy of the product.

Regardless of the approach used, a certified translation should be provided for relevant evidence reported in a language other than English. All evidence will be subject to minimum requirements for relevance, quality and consistency.

If applicants utilise a literature based submission or mixed submission, all scientific publications should be peer-reviewed and be published in a reputable journal.

5.3 Literature search report

A literature search report is a description of a logical, transparent and reproducible approach to identifying and retrieving all authoritative published material which contains evidence (both positive and negative) related to the proposed product and/or its components. It is intended to provide a comprehensive and unbiased review of the available literature in relation to the application, and is a key requirement of all evidence-based medicine. It is not the same as a systematic review.

All applicants must include a report of the methodology used for the systematic literature search with the application in Module 1.5.1. The report should include, as a minimum, a well-conducted systematic search of Medline or Embase with descriptions of any additional non-systematic or manual searching. The report must outline:

  • the search strategy, rationale, platform and date;
  • references retrieved and period covered;
  • selection or filter criteria applied to identify relevant reports;
  • list of reports which have been excluded;
  • appraisal of the evidence identified;
  • pivotal studies and the rationale for their selection; and
  • details of how any additional references were retrieved - for example, from in-house databases, lists of references, or hand searching.

For a search strategy to be considered robust it should be reproducible. Applicants should not substitute 'in-house' databases for Medline and/or Embase searches, or use internet search engines as a primary search platform. However, applicants may include other appropriate public databases in addition to Medline/Embase, and must include all relevant studies regardless of whether the findings are adverse to the proposed product or not. Relevant reports include all studies that reference (amongst others) the product ingredients, formulation, dose, health benefit, and context of use.

No single literature search strategy will fit all cases and requirements will vary according to the specific nature of the application.

In planning and conducting systematic literature searches, you may find it useful for an information retrieval expert to be involved in the process.

For more information on documenting literature searches, please refer to guidance on Systematic literature searches on the TGA website.

5.4 Standards of evidence

The evidence provided to support an application for an assessed listed medicine should cover aspects of the pharmacology, clinical safety (e.g. relating to the dose, use in vulnerable populations, specific formulation and dose form), and efficacy of a medicine, and serve to establish the balance of benefits and risks of the medicine in relation to its intended use. It should also provide the scientific evidence to support the claims and directions for use made on product labels and other product literature.

In assessing the evidence, the TGA must be assured that outcomes observed are due to the therapeutic action of the product and are not simply due to chance or sources of experimental bias introduced by the design, execution, or reporting of the study. The outcomes should also be clinically meaningful, and plausibly applicable to the wider population.

Overall, the standard and weight of evidence submitted for an assessed listed medicine should support a plausible cause-effect relationship between the treatment or intervention and the presumed therapeutic outcome. To ensure that such inferences can be made from studies, the standard of evidence submitted by applicants is reviewed on the basis of:

  • the level of evidence (type/ design, and quantity of consistent evidence)
  • evidence quality
  • statistical validity
  • external validity (generalisability)
  • extent of evidence consistency
  • relevance of the evidence to the product and indications

Information on these considerations is outlined below.

5.4.1 Level of evidence

The level of evidence refers to the type and quantity of studies used to demonstrate the claimed efficacy of a product.

Certain sources of evidence provide higher quality information in this regard than others due to their design, methodology or level of review, and consequently, the degree to which sources of bias have been limited. Additionally, certain types of studies are appropriate as support for both intermediate level indications and low level indications, whereas others are only appropriate for low level scientific or low level traditional use indications. The study type and the body of knowledge should therefore be carefully considered by sponsors in evaluating their evidence in supporting the efficacy of the product.

A brief overview of the main relevant study types and their utility is given below, but applicants may consult the National Health and Medical Research Council (NHMRC) publication: How to use the evidence: assessment and application of scientific evidence for a more detailed discussion.

Study types
Clinical trials

These are studies in which there is an intervention given by an investigator and the subject is followed prospectively through time. Clinical trials vary widely in the strength of evidence they provide, depending on the design and measures taken to reduce sources of error. Certain types of trials offer specific advantages and may mitigate some of the challenges of sampling and generalisation. Examples include:

  • Parallel group designs - subjects are assigned to one or more treatment groups and followed through the study. The advantage of this approach is that it is very straightforward and very few assumptions are made.
  • Crossover designs - each subject is assigned to a sequence of two or more treatments and acts as their own control for treatment comparisons. This reduces the number of subjects required to achieve an appropriate statistical power.
  • Multi-centre studies - the intervention is delivered across multiple sites. This improves the chances of recruiting sufficient subjects for good statistical power, and may provide better generalisation of the results to the broader clinical setting. However, they are also subject to greater potential variability.
  • Factorial designs - two or more treatments are studied simultaneously through the use of various combinations of treatments. The key advantage is that more than one intervention question can be examined within the one trial.

Well-conducted clinical trials can be used as evidence to support both intermediate level and low level indications. Clinical trials submitted as evidence to support an indication should be published in a high quality, peer-reviewed journal. Original unpublished clinical trials should meet all the applicable TGA adopted guidelines.

Only human studies are considered appropriate to support indications for listed medicines. The scientific uncertainties involved in extrapolating non-human data from animal and in vitro studies limit their usefulness. However, non-human and in vitro studies may be used to support any discussion on biological plausibility, and in vivo and in vitro studies may be used when providing biopharmaceutic and pharmacokinetic data.

Systematic reviews

Systematic reviews are characterised by an unambiguous research question; explicit and reproducible criteria for the identification and inclusion/exclusion of studies; a rigorous appraisal of the quality of individual studies; and a systematic synthesis of the results of included studies[12]. Statistical methods (meta-analysis) may or may not be used to analyse and summarise the results of the included studies.

A well-conducted systematic review provides a useful summary of clinical information, and may help resolve seemingly conflicting information. Systematic reviews can also potentially overcome issues related to small sample sizes, and be more generalisable to the wider population (see External validity). Reviews by the Cochrane Collaboration are generally well regarded.

While systematic reviews are a recommended source of evidence for assessed listed medicines, they can be subject to a number of weaknesses. The studies they cover need to be appropriately powered and conducted and reviewers have to consider inclusion criteria with prior knowledge of the results. This may lead to bias; and due to differences in study designs, endpoints, populations, the validity of combining data must be carefully considered. Applicants should critically appraise the methodology and conclusions of any systematic review to determine whether it supports the efficacy of their product.

In general, robust systematic reviews can be used as evidence to support both intermediate level and low-level indications.

Observational studies (cohort and case-control)

Cohort studies are prospective observational studies that involve comparison of two groups of subjects that are followed over time from exposure or non-exposure to a treatment to the subsequent outcome. The result is usually reported as a relative ratio of the outcomes in the two cohorts. Cohort studies are potentially cheaper and easier to conduct than randomised clinical trials, and are useful for assessing the relationship between intervention and the outcome of interest.

Case-control studies are retrospective observational studies. They commence from the outcome, and attempt to work backwards to the exposure. This requires identification of a group with a particular characteristic (cases), and a comparison group (control) that is as similar as possible to the case group but lack the trait in question. In these studies there is no direct intervention - information is obtained from direct questioning of the subject or consultation of records. Case-control studies are advantageous in offering a relatively inexpensive approach and short analysis timeframes. They also require a small number of subjects and do not suffer from the same subject attrition as cohort studies.

Case-control and cohort studies suffer from potentially significant sources of selection bias; potentially compromising attrition of subjects; and a lack of reliability for modest treatment effects. An additional major drawback in case-studies is the potential influence of confounding variables[13], which must be mitigated via matching cases with controls with similar age, gender, socioeconomic status and so forth. As a result, cohort and case control studies and are limited in their ability to provide unbiased and unambiguous data regarding the true efficacy of an intervention, and therefore may not provide acceptable evidence for some indications.

Comparative studies (non-control)

Single-arm (interrupted time series and case series) studies are prospective studies in which a group of individuals is exposed to an intervention and the response to the intervention measured at multiple time points or after the intervention. There is no control group. Single-arm studies are useful when the available subject pool is limited and it is therefore not optimal to randomise many participants to a control arm, or when it is unethical to employ a placebo control.

However, these studies are often limited by an inability to distinguish between the effect of the treatment, a placebo effect, and the effect of natural history. Additionally, they may fail to identify positive effects in situations where a negative outcome would have resulted in the absence of the intervention. They are, therefore, not appropriate as a sole source of evidence for the efficacy of an assessed listed medicine, but can be used as evidence in combination with other sources of evidence.

Non-systematic reviews

Non-systematic reviews tend to be mainly descriptive and often focus on a subset of studies selected by availability or author preference. They typically include an outline of the major findings of a collection of studies, and a broad appraisal of the overall body of literature. They do not employ an organised method of identifying, compiling, and evaluating studies using a set of specific criteria. Non-systematic reviews may or may not include a quantitative pooling of data (meta-analysis). Although informative, non-systematic reviews are limited by selection and author bias. They may also fail to provide clear conclusions, particularly if the studies included have conflicting results. Due to this, non-systematic reviews cannot be used as a sole source of evidence for efficacy, but can provide support for other studies.

Comparable Overseas Regulator (COR) reports for L(A)2 applications

Other international regulatory agencies evaluate and approve low risk medicines. Although they may have vastly different regulatory frameworks compared to Australia, the core scientific and technical methodologies employed by many of these agencies during scientific evaluation are largely similar. As a consequence, where the methodologies and standards employed by an agency are comparable to those used by the TGA, it is possible for applicants to submit technical data from CORs to support the efficacy of their proposed product.

However, a set of criteria relating to the international agency and the scientific report must be met. These criteria are currently being finalised. Once finalised, they will serve to ensure that best practice regulatory approaches have been employed. The requirements for the report will include the following:

  • Reports must be complete, un-redacted, and written in English (or include a certified translation).
  • The reports must collectively address all aspects of the evidence requirements for Module 5 (clinical).
  • The report must have been fully written by an overseas regulator, and present an independent assessment of data provided to that regulator – i.e. self-affirmed/ self-assessed reports or 'expert opinion' reports are not acceptable.
  • Reports should be prepared using internationally accepted guidelines and standards consistent with those used by the TGA.
  • The characteristics of the medicine or ingredient and its intended use described in the report should be identical to that described in the sponsor's application. The formulation, dose, route of administration and/or indications described in the COR report(s) must be identical to that being applied for. Minor formulation changes for fragrance, flavour and colouring variants are permitted, where the combined total difference is not more than 2% of the total formulation.
  • There must be no new indications proposed beyond what the report(s) considered.
  • There must be no new contradictory clinical data or regulatory evaluation reports available beyond what the evaluation report has considered.
  • An application for the proposed medicine must not have been withdrawn in response to technical questions from a regulator or rejected in any other jurisdiction.
  • The conditions under which the reports are provided should not preclude general information about the ingredient or medicine from being published (noting that commercial in confidence information would not be disclosed).

All COR reports should be included in Module 1 under 'Foreign evaluation reports' (1.11.4).

The TGA will not re-evaluate the efficacy data previously evaluated and approved by CORs. The TGA will review the COR assessment report and Australian label to ensure that the evidence meets all requirements and supports the indications and claims of the proposed product.

Reports may be used in conjunction, or in combination with independent scientific studies, provided that the medicines to which they refer are sufficiently similar, and the minimum data requirements are met. Reports from CORs that may not meet all criteria may be considered in certain instances, provided the applicant can provide adequate justification or additional data as required.

The use of international reports enables the TGA to conduct abridged evaluations that focus on issues that are specific to the Australian regulatory context, such as the product label. However, where overseas regulatory reports are used, the TGA will continue to make the final regulatory decisions, ensuring that safety, quality and efficacy are not compromised and that the Australian context is taken into account.

Scientific reference texts and monographs

Several internationally recognised monographs and reference texts are available and may be used to support secondary (low level) indications. Only sources that include scientific/ clinical information are appropriate to support secondary scientific indications. Some examples include:

The monographs produced by a particular organisation may not all meet the specific requirements for assessed listed medicines. Applicants should ensure that the particular monograph provides appropriate scientific evidence to support their product's indications and claims.

Traditional reference texts

If your applications has indications that refer to use in a traditional context (these can only be low level indications), then it may use traditional evidence to support the context of traditional use (see Traditional indications).

Many traditional ingredients have a well-established period of widespread traditional use which is recorded in materia medica, monographs, pharmacopoeias and publications from various international regulatory authorities. In some cases, traditional medicine practices have also been recorded in ethnobotanical or sociological papers. This provides an accumulated repository of information that reflects the refinement of dosage and formulation.

Traditional reference texts can be used to support traditional use indications provided that the dosage, formulation, preparation and use of the medicine are the same as that described in the evidence. Evidence of traditional use for an indication must demonstrate that the medicine or the relevant ingredients in the medicine has been used for at least three generations (at least 75 years) in the tradition it belongs to. In some instances, different sources of evidence of traditional use may be required to support a particular indication for an ingredient or formulation. Together these sources should form a combined collective of evidence that will be relevant and of high quality.

An official national pharmacopoeial monograph may be accepted as evidence of use during the years the monograph has been valid. However, applicants should note that most pharmacopoeias do not contain information on therapeutic indications, posology, or safety of a medicine, since their purpose is to detail the quality aspects.

For many traditional medicines there has been little quantifiable scientific research, scientific assessment or scrutiny undertaken on the medicine's mode of action or effect. It is therefore inappropriate to use evidence of traditional use to support a mechanism of action or an underlying physiological process, as these are required to be supported by scientific evidence (refer to Indications).

Some monographs refer to clinical studies or pharmacology of a particular ingredient using citations and reporting study outcomes of auxiliary scientific papers. Such information is considered a secondary source of scientific evidence and cannot be used to support traditional indications.

Some examples of sources of evidence for traditional use include:

Non-reference textbooks, web searches, and publication abstracts are not appropriate sources of evidence to support an application for an assessed listed medicine.

Evidence hierarchy and quantity

Double blinded randomised controlled trials and systematic reviews of multiple randomised clinical trials are considered to be the gold standard in epidemiological and clinical research, as they are most likely to achieve low bias and high precision when studying treatment effects. However, they are not always available or feasible. Acknowledging this, the TGA allows other study types and a range of other sources of evidence to be submitted as potential support for the claimed efficacy of a product. The limitations of these other sources need to be considered. For example, case-control studies and cohort studies may not be a practical means of providing evidence for some indications and are limited in their ability to produce unbiased and unambiguous data regarding the true efficacy of an intervention. They can, however, provide valuable supportive data relating to the likely effectiveness of an intervention for the general population. Case studies and epidemiological surveys do not have sufficient strength in their own right to justify a scientific indication.

The TGA takes a 'weight of evidence' approach - the less robust the studies, the greater the quantity of consistent evidence required. To assist sponsors to ensure that they have at least the required minimum of appropriate level of evidence to support their application, the TGA has developed an evidence hierarchy and minimum evidence framework. These are provided in Tables 6 and 7. The definitions for intermediate and low level indications have been provided previously (see Risk categorisation).

Note that the requirements in Table 7 are generally minimum evidence requirements and the options presented may not be suitable for every situation. This represents the lower threshold below which the efficacy of the medicine cannot be reasonably assessed. Supplying the data in Table 7 is not sufficient for an application to be approved. The information must be of high quality and address the other requirements that address the efficacy of the product as set out in this document (see Evidence requirements and standards and Alignment of indications and evidence). Information on specific areas of concern may need to be addressed through more or better quality studies.

Additional studies serve to strengthen the evidence of efficacy, particularly if any of the pivotal studies are limited in some way, and may improve the likelihood of an application being approved.

Regardless of what studies are used, all sources of evidence must meet the evidence standards and include a full description of the study design, population, treatment(s) and protocols employed. Abstracts, web searches or incomplete references will not be accepted as suitable evidence. An overview of some of these considerations is provided in the following sections.

Table 6: Evidence hierarchy for assessed listed medicines
Category A Category B Category C Category D
Double blind randomised controlled trials (including cross-over trials) Observational studies e.g. cohort and case control studies Non-systematic, generalised reviews - including databases Traditional reference text
Systematic reviews Comparative studies (non-control) Publicised international regulatory authority articles Herbal monograph
Evidence based reference text - scientific Herbal pharmacopoeia
Scientific monographs Materia medica
Publicised international regulatory authority articles – Traditional only
Table 7: Minimum evidence requirements for assessed listed medicines
Indication[14] Primary (intermediate) Secondary (low level)
Indication type Scientific Scientific Traditional[15]
Required evidence

Minimum of one from Category A


Minimum of two from Category B, AND one from Category C

Non-specific indications:

Minimum of two from Category B or Category C

Non-specific indications:

Minimum of two from Category D to support the tradition of use

Specific indications:

Minimum of one from Category A


Minimum of one from Category B, AND two from Category C

Specific indications:

Minimum of two from Category D to support the tradition of use


Additional evidence from Category C or Category D to support the specificity of the traditional indication

To meet the above minimum evidence requirements, the evidence should contain independent sources of information e.g. two publications referencing the same clinical trial or information are not considered to be two independent sources of information.

5.4.2 Study scope


All studies should be conducted according to Good Clinical Practice (GCP) principles and have appropriate ethical certification. They should also be compliant with International Council for Harmonisation (ICH), and European Medicines Agency (EMA) guidelines adopted by the TGA at the time of application. These can be accessed on the TGA website.

In particular, it is important for applicants to ensure that any studies:

  • are conducted in accordance with sound ethical principles and have received prior approval by an independent ethics review committee.
  • involve only participants who have freely given informed consent prior to the start of the study.
  • are conducted by individuals that are appropriately qualified by education, training, and experience to perform their respective tasks.
  • include accurate recording, handling, and storage of information in a way that allows its accurate reporting, interpretation and verification.
  • are scientifically sound, and described in a clear, detailed protocol.
  • report any conflicts of interest.

The Consolidated Standards of Reporting Trials (CONSORT) guidelines provide a useful checklist against which the trial procedures can be measured, and helps standardise the ways in which the findings are reported. A related document, the Quality of Reporting of Meta-analyses (QUOROM) statement, provides specific guidelines for the reporting of meta-analyses, as well as a checklist to promote standardisation and the inclusion of critical components. Applicants may find these useful ways to self-assess the quality of the studies prior to submitting an application.

Population selection

The study population used should be appropriate for the outcomes being tested (e.g. healthy individuals, women, elderly adults etc.). Data obtained from studies using participants with serious health concerns is generally not appropriate to support an indication for assessed listed medicines; unless the indication relates directly to a population with that condition (e.g. a restricted representation). In some circumstances it is also possible to use studies of participants with serious health concerns where positive outcomes were noted, to provide secondary (non-pivotal) sources of evidence.

The study population should be clearly identified in the study protocol, and all the inclusion and exclusion criteria adequately outlined. These criteria are particularly important. If the criteria are too lax, the validity of the inclusion may be questionable, while if the criteria are too tight, the results may not be applicable to the wider population (refer to External validity).

Baseline characteristics of treatment and control groups should be documented to ensure equivalence in key areas such as age, weight, diet and other factors that may contribute to non-treatment differences in health benefit between groups.

Sample size

Studies submitted in support of the efficacy of an assessed listed medicine should involve a sufficient number of participants to enable the reliable detection of clinically significant treatment effects.

The number of participants required to be reasonably certain of a reliable result is described as the 'power' of a study. In formal terms, the power of a study is the likelihood of the study finding a true difference between treatments if one exists. The larger the sample size, the greater the statistical power of the study.

The number of study participants required for a study to demonstrate clinical significance depends on the study aim and design, how the data will be analysed, and the attrition/ drop-out rates. It may also be affected by:

  • The size of the difference to be detected. The required sample size to achieve a particular power is inversely proportional to the square of the difference to be detected.
  • The relative size of the samples. Greatest power is achieved with an equal allocation of subjects to treatment and control groups.
  • The proportion of subjects who experience the treatment outcome.
  • The extent of variation in the outcome. The larger the variation in the outcome, the greater the required sample size.
  • The extent of protocol compliance. The worse the compliance, the greater the sample size needed to retain adequate power.

Many studies are undertaken with an unrealistic expectation of the likely size of the treatment effect, and the impact of variation and non-compliance. This may dramatically affect the acceptability of the conclusions. Applicants should therefore carefully consider any limitations of the statistical calculations that the study authors have reported, including the number of drop outs and the impact this may have on the reported study outcomes.

To ensure that a study is capable of supporting the indications for a product, the power should be at least 80%. The study should also provide a description of how the power/ sample size was determined. Underpowered studies may be submitted as supplementary evidence, however cannot be used as primary evidence due to the statistical uncertainty of the effect.

Studies submitted to support indications must have a statistical power of at least 80%.

Study outcome(s)

The primary variable of a study is the measurement that provides an estimate of the effect or outcome of a treatment. Ideally, studies should only have one primary variable that:

  • is the most clinically relevant to the proposed indication, and consequently, the one capable of demonstrating the efficacy of the medicine to the greatest extent.
  • provides a reliable and validated measure consistent with the standards and norms of the relevant field.
  • is clearly defined in the protocol before the start of the study. This is to avoid artificial result selection via post-hoc definition of the outcome.

If possible, multiple measures should be used.

Secondary endpoints assess other effects of the medicine that may or may not be related to the primary endpoint (e.g. questionnaires to assess subjective pain). If a study includes secondary efficacy outcomes, they should provide supportive evidence for the primary outcome, or provide supportive evidence for secondary outcomes specified and defined in the initial protocol. It is important that these methods are accurately validated, to ensure the results can be reproduced.

In some instances, the direct measurement of a clinically relevant benefit will be neither feasible nor practical (e.g. for the demonstration of a long-term health benefit). In such circumstances a surrogate variable, which relates to a clinically important outcome but does not in itself measure a clinical benefit, may be used. The surrogate variable must be a demonstrably valid and reliable predictor of clinical benefit.

In all cases, applicants should provide a justification addressing the clinical relevance of the outcome and the rationale for its selection. Additionally, they should demonstrate that the methods used to measure the variables that contribute to the study endpoints should be validated and meet appropriate standards for accuracy, precision, reliability, reproducibility, and responsiveness. Results for every measured outcome must be reported, regardless of whether they are positive, negative or non-significant. The report should also report and discuss any side effects or adverse events observed.

5.4.3 Study quality (internal validity)

In assessing the suitability of the evidence to support indications for assessed listed medicines, TGA evaluators review the quality of the studies submitted. In this context, quality refers to the extent to which the design of the study eliminates sources of bias, and therefore, provides confidence in interpreting the results.

Some of the important considerations for assessing the internal validity of evidence submitted are outlined in the sections below. They are, however, dealt with in greater detail in TGA adopted guidelines. The NHMRC publication: How to use the evidence: assessment and application of scientific evidence also provides further useful information on many of these points.

While these principles are mostly applicable to studies used to support intermediate level indications (i.e. studies in Category A and B in Table 6), it is worthwhile applying them to studies submitted in support of low level indications, as this will improve the likelihood of an application being approved.

Sources of bias

The ideal experimental design should compare two groups that do not differ in any significant respect apart from the treatment or intervention of interest. Significant differences between the groups may introduce bias into the comparison. Bias is the tendency of any factors associated with the design, conduct, analysis and interpretation of the results of a study to cause the treatment effect deviate from its true value. Bias can be introduced through deviations in conduct (operational bias), or may be inherent in the design and analysis of the study (statistical bias).

The following examples of factors introducing bias can be considered when assessing the strength of a study:

  • selection bias - study subjects are chosen in a way that results in a non-random sample that is not representative of the population.
  • measurement bias - investigators may subconsciously favour, or more diligently pursue observations, in one treatment arm over the other and this can be reflected in the measurements obtained.
  • observer bias - the observer makes a subjective appraisal of the outcome.
  • recall bias - patients know whether they were allocated to the treatment group or the control group, and this affects their reporting of their symptoms.
  • regression to the mean - a phenomenon occurring when groups have been selected on the basis of extreme scores. If the first measurement is an extreme score, the second measurement will tend to be closer to the average.
  • treatment selection bias - the effects of a treatment are determined by confounders (e.g. other co-interventions) rather than the treatment itself.
Control, randomisation and blinding

Many sources of bias can be addressed through the use of appropriate controls, randomisation, outcome selection, and blinding of both researchers and patients.

  • A control is an observation or treatment designed to provide a comparator in order to enable researchers to identify the effects of variables other than the treatment or intervention under study. Controls may be negative/placebo (to ensure that there is no effect of a placebo when no effect is expected), positive/known active (to ensure that an effect can be measured when an effect is expected), a different dosage regimen, or no treatment. The choice of the control group is be determined by such factors as the availability of standard treatments, the severity of the condition, and ethical considerations. The choice of control should be justified in the applicant's report.
  • Randomisation is a method in which subjects are allocated by a chance mechanism to treatment or control groups. Each subject in the study should have the same chance of being allocated to any of the treatment groups. This eliminates selection bias; helps ensure that treatment groups will tend to be comparable; and ensures the validity of the statistical analysis. Ideally the randomisation process should be centralised and computer-based, and also incorporate concealment of allocation.
  • Blinding is a statistical bias reduction technique in which the study participants have no knowledge of which treatment (placebo or test product) they received. Double blinding is where neither the study participants nor the investigators carrying out the study are aware of the treatment to which the subjects have been allocated. Double-blinding eliminates the potential for observation bias, and reduces the potential for the "placebo-effect". Blinding of outcome measurements becomes more crucial as the measure becomes more subjective and more open to observer bias. This is particularly important for symptoms and other patient self-report measures.

Applicants should take into account the controls, randomisation, blinding, and allocation concealment (where appropriate) when assessing the reliability of their evidence as support for indications. Well-designed and conducted studies should include a detailed description of the subject eligibility criteria, method of randomisation, and blinding technique employed so as to enable assessment of the potential for error or unblinding. Studies that do not employ or report these bias reduction techniques are less likely to be accepted by evaluators as convincing evidence of the veracity of claims.

Changes to the protocol

Departures from the planned conduct of a study may introduce operational bias. It is therefore important that any evaluation of a clinical study report consider the number and severity of protocol violations and deviations and the completeness of patient follow-up. These may include changes to the inclusion or exclusion criteria, outcomes measured, group sizes, treatment protocol, blinding, and/or duration. Any statistical consequences of these changes and revision of analytical approaches should also be addressed.

Dealing with missing data

One of the problems with clinical studies, particularly longer-term studies, is that data may be lost due to patient dropout, treatment failures, non-compliance, adverse events, and missed measurements, among other factors.

Missing data stemming from high attrition rates can lead to non-comparability of treatment and control groups, a reduction in the statistical power of the study, and the introduction of significant bias. This may make the results of the study difficult to interpret, and diminish the ability of the study to support an indication or claim.

Consequently, the design and conduct of a clinical trial should seek to minimise the amount and/or impact of missing data and where relevant, should contain:

  • the subject withdrawal criteria
  • whether and how subjects were replaced
  • procedures for monitoring subject compliance
  • procedures for accounting for missing, unused and spurious data
  • subject accountability, including details for reasons of withdrawal
  • a discussion of the number, time, pattern and possible implications of missing values.

The TGA will consider the reasons for the missing data, whether appropriate replacement methods were used, and the likely effect of any missing data on the study results.

There is no clear definition of what proportion of missing data would mean that a trial was not suitable to support any specific indication or claim, but in broad terms, 5%-10% would generally not be of great concern, while >20% would mean that missing values would need to be one of the key issues addressed in the evaluation.

5.4.4 Statistical validity

Statistical validity refers to the extent to which a measurement is well-founded and accurately reflects reality. The validity of the conclusions of the study is critically dependent on the nature of the metric used to measure an effect, the size of effect, and the type of statistical transformations and analyses performed.

Studies submitted as part of an application for listing of an assessed listed medicine must use valid statistical methods to assess the outcomes, and must account for any potential confounders. These statistical methods should be fully described in the protocol, and should be appropriate for the efficacy outcomes measured. Unplanned analyses undertaken after the completion of a trial (post-hoc analyses) are to be avoided as they are unlikely to have been considered in power calculations and study design.

Analysis set

In general, the main analysis should be in the intention-to-treat (ITT) population. This involves analysis of subjects according to the allocated treatment regimen, rather than the actual treatment experienced (i.e. subjects are analysed according to their allocation, regardless of their compliance to the treatment). This is the most conservative approach as it biases the analysis towards the null hypothesis. However, it serves to minimise bias in instances where the dropout rates are high.

When an ITT is performed, all efforts should be made to obtain outcome measurements from all original participants at the end of the study. In cases where this is not possible, baseline measurements of study parameters should be carried forward (for example, for a study outcome related to weight loss, body weight recorded at the beginning of treatment would be the same at the end of the study). A treatment effect demonstrated in an ITT analysis may underestimate the efficacy of the treatment, but may be a good reflection of effectiveness under real world conditions.

Applicants should be cautious of studies employing a per-protocol analysis. This is an analysis on the subset of study subjects that actually complied with the protocol sufficiently. While it is less conservative than ITT analysis, it maximises the chances of an effect being observed and is heavily prone to bias as adherence to the protocol may be related to the treatment outcome.

Measure of the effect

There are a large number of ways of expressing the effect of a treatment. These include averaged differences, standardised differences, weighted means, relative risk/ risk ratio, and odds ratio, among others. The choice of measure largely depends on the study design, and whether the outcome measured is a continuous number (e.g. blood pressure) or discrete (e.g. improved/ not improved).

For assessing efficacy, the risk difference and the number-needed-to-treat are the most important measures:

  • The risk difference (also known as the absolute risk reduction or absolute effect) is the difference between the proportion or rate of events in the treatment group and the control group i.e. Risk Difference = Proportion treatment - Proportion control. It is the inverse of the number-needed-to-treat.
  • The number-needed-to-treat is the number of patients that need to be treated with the test product before one patient experiences a clinical benefit from treatment. It is the inverse of the risk difference i.e. Number-needed-to-treat = 1/risk difference.

Only expressing the effect of a treatment as relative measures (e.g. relative risk/risk ratio, odds ratio) has the potential to mislead, as relative measures tend to exaggerate estimates of efficacy.

Statistical significance

Even if the study is well-conducted and sources of bias are limited, there is a possibility that the results arose purely by chance. It is therefore essential that studies submitted as evidence for assessed listed medicines use appropriate statistical methods to minimise Type I errors. A Type I error (false positive) is a conclusion that there is a difference between two treatments when no difference exists in reality.

Well-conducted studies will usually report the degree of statistical significance (p-value) associated with the observed difference between treatments. The p-value provides an indication of the probability of claiming that there is a treatment effect when in fact there is no real effect (i.e. the probability of making a Type I error). The p-value provides an indication of whether the treatment effect that has been observed can be explained by chance alone. Although there is no definitive p-value threshold, the lower the p-value the greater the likelihood that the effect observed is real. In practice, a 'p' value of less than 0.05 indicates with acceptable certainty that an observed effect or health benefit is unlikely to be due to chance.

In considering the strength of their evidence, applicants should ensure that:

  • the statistical test used to derive the p-value is appropriate and reliable;
  • the p-value obtained for the primary outcome is less than 0.05; and
  • all the actual p-values (not just p < 0.05) are reported.

It is important to bear in mind that statistical significance does not provide information about the degree of benefit produced or whether it is likely to be clinically meaningful.


The confidence interval (CI) is the range of values within which there is a certain likelihood that the true value can be found. The confidence level is the probability that the CI contains the true difference. Well-conducted studies should usually report the 95% CI. This means that there is a 95% chance that repeated experiments would have outcomes that fall within the specified range.

The precision (or width) of the CI is also an important consideration. A narrow 95% CI is much more desirable than a wide 95% CI. A wide CI indicates a low level of confidence in what the true population effect is. The effect of different is illustrated in part in Figure 1.

When 95% confidence intervals are generated for primary study outcome measures, the 95% CI should include only clinically important treatment effects. The 95% CIs of the intervention and exposed groups must not overlap (refer to Figure 1).

Clinical significance

A statistically significant outcome indicates only that there is likely to be a relationship between intervention and outcome. For a study to provide adequate support for an indication, the observed differences should not only be statistically significant, but also be clinically meaningful.

Clinical significance is the degree of benefit that is worthwhile to justify a particular treatment or intervention. It can be regarded as a measure of how meaningful a particular study outcome might be to the consumer. For example, a study might demonstrate a statistically significant weight loss, but in practice, that effect would not generally be experienced or noticed by members of the wider population, and would be unlikely to significantly advance health goals.

Judgements about clinical significance are often made by experienced clinicians within a context of ongoing monitoring and supervised care. As assessed listed medicines are self-selected and often used without healthcare practitioner intervention or supervision, it is challenging to evaluate how studies result to practical health outcomes. This is particularly true given that the meaningfulness of a predetermined "significant clinical benefit" may vary between patients depending on a number of factors such as state of disease, comorbidities, personal circumstances, and alternative options for treatment.

Nevertheless, applicants should give consideration to the likely meaningfulness of an observed health outcome to the intended target population. In making such judgements, it is useful to consider the significance, confidence interval, confidence level, and the magnitude of the treatment effect. For the evidence to suggest that an intervention is useful, the 95% CI at p <0.05 should include only clinically important treatment effects. This is illustrated in Figure 1.

Figure 1: Illustration of the impact of the size of the effect and the precision of the 95% CI on the interpretation of the outcomes of a clinical study.

Figure 1: Illustration of the impact of the size of the effect and the precision of the 95% CI on the interpretation of the outcomes of a clinical study.

For a study to be considered as reliable support for an indication, the p-value for an observed difference should be <0.05, and the 95% confidence interval should include only clinically meaningful results.

5.4.5 External validity

External validity and extrapolation are extremely important considerations in determining the scope of an indication.

Extrapolation is the application of results from a study to a different population from the one used in the study (e.g. results from a study on 20-25 year old women being applied to 30-35 year old women). Generalisability (or external validity) is a term often used to conceptualise the extent to which study results can be broadly generalised beyond the setting of the study and the particular sample groups used.

The validity of such inferences depends on the representativeness, size and variability of the study sample. The greater the extent of these characteristics, the more generalisable the results. Additional factors to consider when determining if the results of a particular study can be extrapolated or generalised are:

  • The effect of gender, age, or ethnicity. Are physiological differences likely to impact on the efficacy of the treatment?
  • The timing of the treatment. The stage of the condition /illness may impact on treatment outcomes.
  • Variations in the condition being treated. There may be distinct underlying aetiologies despite similar presentations.

It is also worthwhile considering whether the results of a particular study are applicable to individuals as well as groups.

5.4.6 Balance of evidence and conflicting results

The strength of evidence provided by a specific study is greatly enhanced if the effect is reproducible, and if the cause-effect relationship proposed is consistent with existing knowledge.

A well-conducted literature search will identify all related studies, and these should be assessed in relation to the findings of the primary evidence. Positive, null and negative results should be examined. If there are conflicts in the outcomes of different studies, the applicant must provide a plausible explanation for the conflicts in a scientific justification.

In some instances, the conflicts can be readily accounted for by differences in design or methodology (e.g. dose form, population, timing etc.). If suitable explanations for the discrepancies cannot be found, the highest quality studies will receive higher weight in the evaluation.

5.4.7 Summary of considerations for studies

Table 8 summarises many of the considerations outlined in the preceding sections. Applicants may find it helpful in appraising the quality of their evidence prior to submission of an application.

It is important to bear in mind that these principles reflect issues that are taken into account in an ideal study. If there are valid scientific reasons for a particular omission or deviation, applicants can submit a scientific justification which addresses the matter (refer to Justifications).

Applicants may find the following guidance useful:

How to access a pdf document

Table 8: Principles for clinical trials, including trials included in systematic reviews.
Topic Consideration Principle
Scope Conduct
  • The study should be conducted using Good Clinical Practice (GCP).
  • The intervention must be described at a level that allows replication.
  • Conflicts of interest should be reported.
  • The report must specify the sample and how it was obtained, including inclusion/ exclusion criteria.
  • The study population must be appropriate for the outcomes tested.
  • The sample size should provide sufficient statistical power.
  • The stated outcome(s) of the intervention must be measured.
  • The outcomes should be appropriate and relevant.
  • Valid measures of the targeted effect must be used.
  • Results for every measured outcome must be reported, regardless of whether they are positive, negative or non-significant.
  • Potential side-effects or adverse events should be measured.
Quality Controlling bias
  • The methodology should include appropriate blinding or masking.
  • The design must have at least one control condition that does not receive the tested intervention (where relevant).
  • Where possible, assignment to conditions needs to minimise statistical bias through randomisation.
  • Measures of internal consistency must be included.
Changes and missing data
  • Protocol violations and deviations and the completeness of patient follow-up should be reported.
  • Compliance / attrition should be reported and accounted for.
  • Missing data must be reported and handled appropriately.
Statistics Basis
  • The intention-to-treat (ITT) population should be used.
  • The absolute risk difference and the number-needed-to treat should be reported where possible.
  • Statistical methods must be relevant to the experimental design, and aim to produce an unbiased estimate of relative effects along with a statistical measure of confidence.
  • Pre-test differences should be accounted for.
  • The p-value or interval must reasonably exclude chance.
  • Outcomes should have clinical, rather than merely statistical, significance.
  • Efficacy can only be claimed for a consistent pattern of statistically significant effects.
  • The extrapolation and generalisability of the results should be considered.
Balance of evidence  
  • Evidence from multiple sources is desirable.
  • Plausible explanations for contradictory results should be provided.
  • Where two or more studies are available, the results of the highest quality studies will have higher weight.

5.5 Biopharmaceutic and pharmacokinetic studies

Biopharmaceutic and pharmacokinetic studies are a critical part of establishing the efficacy of medicines. These types of studies demonstrate that medicines dissolve and release active ingredients appropriately; and that the active ingredients are absorbed, distributed and metabolised in a manner that allows efficacious quantities to reach the intended site of action. They also serve to ensure that undesirable effects such as dose-dumping, dose retention or in vivo interactions do not either reduce the efficacy of the product or pose a risk to the consumer.

For L(A)3 applications (using Method 2A or 2B) and L(A)2 generic applications, different biopharmaceutic and pharmacokinetic studies are required to ensure that the products are likely to be efficacious. The data requirements are summarised below depending on the type of product. For further guidance see TGA Guidance 15: Biopharmaceutic studies[16].

Pharmacokinetic data is not explicitly required when a clinical study is used as the main evidence for an intermediate indication (Method 1), however it is expected that such studies will address relevant aspects of the medicine's pharmacokinetic properties.

5.5.1 New products (systemically acting)

For L(A)3 applications, in order to establish the efficacy of a new systemically acting product, applicants must provide bioavailability studies. This is particularly important in the case where the evidence is derived from efficacy studies on the individual ingredients (Method 2A), however as described in Table 5, is not usually necessary for Method 1.

Bioavailability is the proportion of the administered dose of an active ingredient that reaches systemic circulation as an intact drug. It may differ between individuals and depends on a large number of factors that cannot usually be reliably inferred from the formulation. For systemically acting oral products these include the:

  • rate and extent of the disintegration of the product, and the rate and extent of dissolution of the active ingredient(s).
  • rate and extent of passage of the active through the gut membranes - a process determined by factors such as the physiochemical characteristics of the active, including its lipid solubility, diffusivity and propensity towards interactions with active transporters in the gut wall, as well as the excipients used in the formulation, the drug coatings, and the gut lumen pH;
  • rate of gastric emptying/ intestinal transit; and
  • extent of first-pass metabolism in the liver - if first-pass metabolism occurs, some proportion of the substance will be removed before the remainder reaches systemic circulation.

For these reasons, applications for new assessed listed products must address several issues that impact on the efficacy of the medicine. These requirements also differ depending on whether the product is intended to be an immediate release product or a delayed/ modified release product.

Immediate release oral dosage forms

The following studies (or a robust scientific justification for not including such studies) are required:

  1. study to establish that the proposed formulation is optimal (e.g. a comparative bioavailability study versus an oral solution of the drug);
  2. bioequivalence studies between the proposed formulation and pivotal clinical trial formulations;
  3. bioequivalence studies amongst the various strengths proposed in the application (if applicable); and
  4. a food effect study.
Modified release oral dosage forms

Modified release products (including delayed, sustained, and combination release products) must be determined to meet the modified release claims; should provide consistent pharmacokinetic performance between dosage units; and should produce plasma concentrations that lie within the therapeutic range.

The following studies (or a robust scientific justification for not including such studies) must be submitted, in addition to the studies required for immediate release oral dosage forms:

  1. steady state versus an appropriate immediate release reference product; and
  2. in vitro and in vivo correlation studies.
  3. in vitro studies confirming the absence of dose-dumping effects in the presence of alcohol.

5.5.2 Generic products

For L(A)2 and in some cases for L(A)3 applications, the proposed assessed listed medicine may be similar or identical to an existing product for which bioavailability data exists. This includes products currently approved as an assessed listed medicine in Australia, evaluated by comparable regulatory authorities, or extensively studied in clinical trials, and excludes 'grandfathered' products. In such cases, the efficacy of the proposed product can be inferred if it can be demonstrated that it has the same / similar pharmacokinetic properties to the reference product.

A generic assessed listed product is a medicine that, in comparison to another fully TGA evaluated assessed listed medicine (the "reference product") included in the ARTG:

  1. has the same quantitative composition of therapeutically active substances, being substances of similar quality to those used in the comparison medicine; and
  2. has the same pharmaceutical form; and
  3. is bioequivalent; and
  4. has the same safety and efficacy properties.

All applications for generic systemically acting products must establish bioequivalence between the reference and the proposed products. Bioequivalence refers to the comparability of medicines that are pharmaceutically equivalent and which have no significant difference in the rate and extent to which the active ingredient becomes available at the site of drug action when administered at the same molar dose under similar conditions. Bioequivalent drugs are similar to such a degree that their effects, with respect to both efficacy and safety can be expected to be essentially the same (see Guidance 15: Biopharmaceutic studies).

Bioequivalence studies should be performed on the innovator product (i.e. a product that has had a full efficacy data package evaluated by the TGA), not a generic of the innovator. This is important to reduce the likelihood of pharmacokinetic drift, whereby generics that refer to other generics no longer resemble the originally evaluated product due to variability in confidence intervals in each bioequivalence study.

The most reliable means to demonstrate that one formulation will be as effective as another is to conduct a randomised, single-dose crossover bioequivalence study in healthy volunteers. In these studies, subjects receive the different formulations on two separate occasions separated by a wash-out period. A minimum of 12 volunteers, sufficiently long wash-out period (5-7 times the half-life of the drug) to prevent carryover effects, and adequate plasma sampling frequencies should be used. Studies on subjects in the fasted state are usually preferred, as this is the most sensitive condition for detecting differences between formulations. In general, the 90% confidence interval of the ratio of the geometric means of the area under the plasma concentration vs. time curve (AUC) and maximum plasma concentration (Cmax) are required to be between 0.8 and 1.25 for bioequivalence to have been demonstrated.

There are two ways to demonstrate bioequivalence:

  1. Where the reference product has previously not been evaluated for bioavailability, bioavailability studies of both the reference and the proposed product are required.
  2. Where the reference product has been evaluated for bioavailability by the TGA, similar dissolution profiles between the reference and the proposed products across the physiological pH are taken into consideration to establish bioequivalence.

If the formulations differ significantly and a different release rate has been designed into the formulation, then a non inferiority study against the reference product may be appropriate.

For further information, refer to the TGA adopted Guideline on the Investigation of Bioequivalence (CPMP/EWP/QWP/1401/98 Rev. 1).

Generic products that have the same excipients as the reference product

If a product is a generic of an existing product for which bioavailability data exists, and has the same excipient formulation, applicants are able to demonstrate bioavailability by providing:

  1. evidence of identical formulations including excipients;
  2. demonstration of similar dissolution profiles between reference and proposed products across physiological pH; and
  3. scientific justification explaining why points a. and b. above sufficiently demonstrate the bioavailability of the specific product.

Applicants may rely on bioavailability data for an existing product, or data that is obtained through literature, with the appropriate justification. The reference product cannot be a 'grandfathered' medicine.

For further information, please refer to Guideline on the investigation of bioequivalence (pdf,233kb) and Guidance 15: Biopharmaceutic studies.

Generic products meeting the requirements for a BCS-based biowaiver

In certain circumstances, despite the product being of a type that would normally require biopharmaceutic studies, it is possible to provide a robust scientific rationale for why bioavailability and/or bioequivalence data might be considered unnecessary for listing of the proposed product. This is generally referred to as a 'biowaiver'.

Biowaivers allow dissolution tests to be used as the surrogate basis for the decision as to whether two products may be considered to be equivalent. In this context, the dissolution and absorption of the medicine is regarded as the critical aspect in determining the equivalence of two products. Consequently, biowaivers are only appropriate for certain classes of products.

The Biopharmaceutics Classification Scheme (BCS) is generally used to determine whether or not a biowaiver may be appropriate. The BCS classifies active substances into four classes based on solubility and permeability, as follows:

High solubility Low solubility
High permeability BCS class I BCS class II
Low permeability BCS class III BCS class IV

Products containing active(s) that are highly soluble, highly permeable (i.e. BCS class I substances) and rapidly dissolving may be considered for a biowaiver. Highly soluble substances are soluble at the highest dose strength in <250 ml water over a pH range of 1 -7.5. Highly permeable substances are those for which the extent of absorption is > 90% of an administered dose based on mass balance or relative to an intravenous reference dose. Rapidly dissolving products are defined as those where no less than 85% of the product dissolves within 30 mins in standard conditions. Excipients that might affect bioavailability should be qualitatively and quantitatively the same in the two products. In general, the use of the same excipients in similar amounts is preferred.

Biowaivers are not usually supported for BCS II–IV substances, or products with more complex formulations such as prolonged release tablets. In some instances, a BCS-based biowaiver may be considered for BCS III products (high solubility, low permeability) where the substance has high solubility and limited absorption; very rapid in vitro dissolution; excipients that might affect bioavailability are qualitatively and quantitatively the same; and other excipients are qualitatively the same and quantitatively very similar.

For additional information about the requirements for demonstrating that a substance can be considered to be BCS class I or III for the purposes of a biowaiver, refer to Appendix III of the TGA-adopted TGA-adopted Guideline on the Investigation of Bioequivalence. The FDA guidance on the BCS may also be helpful.

In all cases, the reference product to which the proposed assessed listed medicine is being compared must have established bioavailability data (e.g. assessed listed or registered medicines, apart from 'grandfathered' products, or products extensively characterised by overseas regulatory authorities). The applicant’s justification should address all the aspects of the products outlined in Guidance 15: Biopharmaceutic studies. Additionally, dissolution profiles across physiological pH showing appropriate release of the active(s) must be supplied (this is a standard quality requirement for all medicines).

5.5.3 Products not requiring biopharmaceutic studies

For some L(A)2 and L(A)3 applications, there is a limited number of products that do not require biopharmaceutic studies, even in the absence of a reference product. These include:

  • Aqueous oral solutions that contain the same active substances in the same concentration as a current evaluated product, and that do not contain any excipients that might affect the in vivo solubility, in vivo stability, gastric passage or absorption of the active ingredient(s). Refer to the Guideline on the investigation of bioequivalence for more information.
  • Oral medicines that are not systemically or locally absorbed (e.g. probiotics, non-digestible polymers, oral suspensions etc.).
  • Locally applied products where the active(s) are not systemically or locally absorbed.
  • Products with only minor formulation changes - i.e. up to 2% of the total formulation can change compared to the reference formulation, if the change is only to a flavour, fragrance and/or colour (including printing inks). However, you may need to provide dissolution profiles across physiological pH showing appropriate release of the active(s).
  • For variations to formulation; medicines with an acceptable correlation between the rate and extent of in vivo absorption and the in vitro dissolution rate, and where the in vitro dissolution rate of the reformulated medicine is equivalent (under the same test conditions used to establish the correlation) to the approved AUST L(A) medicine.

For products that are not systemically or locally absorbed, you may need to provide evidence of non-absorption, and efficacy may need to be demonstrated via clinical studies or other data. Please refer to the relevant guidelines e.g. the Note for Guidance on the Clinical Requirements for Locally Applied, Locally Acting Products Containing Known Constituents (CPMP/EWP/239/95 final).

Further information

For further information on pharmacokinetic and biopharmaceutic studies, refer to the Guideline on the investigation of bioequivalence.

5.6 Justifications

In some cases it might be unfeasible or scientifically unrealistic to supply some required evidence or to meet some of the guidelines mentioned in Sections 4 - 5 above. In such instances, applicants are able to submit a scientific justification (see 'Justification for not complying with technical data requirements or not adhering to guidelines' in the Mandatory requirements for an effective assessed listed medicine application).


  1. A placebo is a substance or treatment with no active therapeutic effect.
  2. A biowaiver is an acknowledgement that in vivo bioavailability and/or bioequivalence studies may be considered unnecessary for product approval.
  3. A study or justification may be required if there is doubt as to whether absorption occurs.
  4. For example, additional biopharmaceutic and pharmacokinetic studies may be required for non-conventional dosage forms such as modified release products.
  5. The finished product is the final dosage form with all active and excipient ingredients.
  6. Up to 2% of the total formulation can change compared to the formulation in the evidence, if the change is only to a flavour, fragrance and/or colour (including printing inks).
  7. Cochrane Collaboration - Glossary
  8. A confounding variable is a factor that may contribute to an apparent association between an exposure and outcome by independently affecting both.
  9. See definition in Risk categorisation
  10. All indications for assessed listed medicines must be supported by scientific evidence of efficacy, however traditional evidence is required to support use in a traditional context (see Low level (secondary) indications – Traditional indications)
  11. Although this guidance refers to prescription medicines, the same principles apply for providing biopharmaceutic studies for the relevant assessed listed medicines application types

Book pagination