Module 8: Science and Society

NSWInvestigating ScienceSyllabus dot point

Inquiry Question 2: Are there limits or boundaries to scientific research, and how are these determined?

Investigate the role of conflicts of interest in scientific research, including industry funding and the responsibilities of scientists to disclose

A focused answer to the HSC Investigating Science Module 8 dot point on conflicts of interest. Covers tobacco industry funding, pharma trials, climate denial, AusVaxSafety, and worked HSC past exam questions on industry-funded research.

Generated by Claude OpusReviewed by Better Tuition Academy5 min answer

Have a quick question? Jump to the Q&A page

What this dot point is asking

NESA wants you to identify conflicts of interest in research, explain how they can shape results, and evaluate the measures used to manage them. This dot point is examined in 4 to 6 mark questions.

The answer

A conflict of interest exists when a researcher has personal, financial, professional or political interests that could shape research outcomes. Such interests are pervasive in modern science: pharmaceutical research depends on industry funding, climate research can attract political pressure, and academic careers depend on publication.

Types of conflict of interest

Financial
The researcher or their institution receives payment, stock, royalty or grant funding from a stakeholder.
Professional
Career advancement depends on positive results, attention or alignment with a research group.
Political
Findings touch on contested public policy, exposing researchers to advocacy pressures.
Personal
Researcher has family, friends or strong personal beliefs affecting the topic.
Patent or intellectual property
Researcher holds or expects to hold patents related to the work.

How conflicts shape research

Conflicts can shape every stage:

  1. Hypothesis formation. Which questions get asked. Industry favours questions whose answers are commercially useful.
  2. Methodology. Industry-funded trials may use favourable comparators (a competitor's older drug at a lower dose) or non-standard primary endpoints.
  3. Data interpretation. Ambiguous data can be presented to favour the funder's preferred conclusion.
  4. Selective publication. Negative results are filed away (publication bias).
  5. Communication. Press releases over-claim findings to attract media attention.

The 2017 Cochrane review by Lundh and colleagues found that industry-sponsored studies are more likely to report results favourable to the sponsor than independent studies, even after adjusting for methodological quality.

Documented examples

Tobacco industry doubt-mongering (1950s onwards)
Tobacco companies internally accepted by the late 1950s that smoking caused cancer. Externally, they funded research, recruited scientists and amplified doubt to delay regulation. The 1969 Brown and Williamson memo stated: "Doubt is our product." Naomi Oreskes and Erik Conway document the strategies in Merchants of Doubt (2010).
Pharma industry trials
Multiple Cochrane reviews show industry-funded drug trials are more likely to report results favourable to the sponsor. Vioxx (rofecoxib) was withdrawn by Merck in 2004 after researchers showed it caused heart attacks; subsequent analyses revealed Merck-funded studies had downplayed the cardiac risk.
Climate denial campaigns
ExxonMobil funded climate-sceptical research and think-tanks for decades despite internal scientific reports confirming anthropogenic warming. Documented in the 2015 Inside Climate News investigation.
Sugar industry and dietary fat
A 2016 JAMA Internal Medicine article revealed that the Sugar Research Foundation (SRF) paid Harvard nutrition researchers in the 1960s to publish a literature review concluding fat (not sugar) was the main dietary cause of heart disease. This shaped US and global dietary policy for decades.

Wakefield: the canonical undisclosed conflict case

Andrew Wakefield's 1998 Lancet paper linking the MMR vaccine to autism failed to disclose:

  • Payments over 400,000 GBP from lawyers preparing litigation against MMR manufacturers.
  • A patent on a competing single-virus measles vaccine.
  • Some procedures performed without ethics approval.

These were major contributors to the eventual 2010 retraction and the GMC striking him off the register. The case is studied in research ethics courses worldwide.

Management measures

1. Mandatory disclosure
Major journals (NEJM, JAMA, Lancet, Nature, Science) require authors to declare all relevant financial relationships, including funding, stock holdings, consulting fees, patents and travel.
2. Pre-registration
Hypotheses, methods and primary outcomes are publicly registered before data collection. Sites like ClinicalTrials.gov and the Australian and New Zealand Clinical Trials Registry (ANZCTR) host these.
3. Open data
Some journals require raw data to be made publicly available for independent re-analysis.
4. Registered reports
Journals accept papers based on methodology, before results are known. This prevents publication bias toward positive findings.
5. Independent replication
Cochrane reviews and other meta-analyses combine multiple independent studies. Effects observed only in funded studies but not in independent ones are flagged as suspicious.
6. Editorial independence
Major journals separate editorial decisions from advertising and commercial relationships.
7. Institutional integrity offices
Universities and research institutes investigate alleged misconduct. The Australian Research Integrity Committee oversees national investigations.

Australian disclosure framework

The Australian Code for the Responsible Conduct of Research (2018), jointly issued by NHMRC, ARC and Universities Australia, requires:

  • Disclosure of all funding sources and conflicts of interest.
  • Annual declarations by researchers.
  • Institutional registers of conflicts.
  • Specific protocols for industry-funded research.

NHMRC and ARC funding requires disclosure as a condition of grant agreement.

What disclosure does and does not do

It does:

  • Make conflicts visible to readers, editors and reviewers.
  • Allow informed weighting of evidence.
  • Create accountability for undisclosed conflicts.

It does not:

  • Eliminate the effect of conflicts on research design or interpretation.
  • Prevent researchers from being shaped by their interests.
  • Guarantee independent or unbiased work.

Disclosure is necessary but not sufficient. Independent replication, pre-registration and open data are needed to verify findings.

Australian success: AusVaxSafety

AusVaxSafety is an Australian vaccine safety surveillance program that monitors adverse events following immunisation using independent observational data. It is funded by the Department of Health but operates with clear protocols and transparent reporting, including conflict-of-interest declarations from all participating researchers. The program has been cited internationally as a model for managing the inherent conflict in vaccine safety research.

Difficulties

Conflicts of interest cannot always be eliminated. Researchers in some fields have specialist expertise hard to find outside industry-aligned roles. The challenge is to manage rather than ban industry involvement.

The pharmaceutical industry funds most drug discovery. Without it, drug development would stall. The challenge is to design trials and disclosure systems that preserve the benefits of industry involvement while minimising bias.

Citizens and conflicts

When evaluating scientific claims, citizens can ask:

  • Who funded the study?
  • Were results pre-registered?
  • Has it been independently replicated?
  • Are the authors connected to relevant industries?
  • Is the conflict declared?

These questions are part of scientific literacy.

Past exam questions, worked

Real questions from past NESA papers on this dot point, with our answer explainer.

2024 HSC6 marksUsing examples, discuss how conflicts of interest can shape scientific research and what measures are taken to manage them.
Show worked answer →

A 6-mark answer needs the mechanism of bias, two examples, the management measures and a balanced judgement.

Mechanism
When researchers or institutions have financial, professional or personal interests in the outcome, those interests can shape which hypotheses are tested, how data is analysed, whether results are published, and how findings are communicated. The bias may be conscious or unconscious and is documented in meta-analyses comparing industry-funded with independent research.
Example 1: Tobacco industry
From the 1950s, tobacco companies funded research arguing the smoking-cancer link was uncertain. Internal documents (Master Settlement Agreement 1998) showed companies concluded internally that smoking caused cancer decades before public admission. The "Doubt is our product" memo (Brown and Williamson 1969) stated the strategy: manufacture doubt to delay regulation.
Example 2: Pharmaceutical trials
Industry-funded drug trials are more likely to report positive results for the funder's product than independent trials (Lundh et al. Cochrane review). Mechanisms include selective publication, favourable comparator choices and selective subgroup analysis.

Management measures.

  • Mandatory disclosure. Major journals require explicit conflict-of-interest declarations.
  • Pre-registration. Hypotheses and analyses locked before data collection.
  • Independent replication. Cochrane reviews give weight to study design rather than funder.
  • Funding transparency. NHMRC and ARC require open declaration.
  • Editorial independence. Journals separate editorial decisions from advertising.

Judgement. Conflicts are unavoidable; disclosure, pre-registration and independent replication mitigate but do not eliminate the effect. Markers reward two examples, multiple measures and clear evaluation.

2021 HSC4 marksWhy must scientists disclose conflicts of interest?
Show worked answer →

A 4-mark answer needs the principle, the practical consequence, an example and the broader benefit.

The principle. Trust in science depends on the integrity of the research process. When researchers have financial or personal interests in a particular outcome, those interests can shape (consciously or unconsciously) the questions asked, methods used and conclusions drawn. Disclosure allows readers, editors and the public to weigh the research accordingly.

Practical consequence. Undisclosed conflicts can lead to:

  • Retraction of published papers (Wakefield's 1998 MMR paper).
  • Loss of professional standing (struck off medical register).
  • Loss of funding from federal and international bodies.
  • Reduced credibility for the institution and the field.

Example. Andrew Wakefield's 1998 Lancet paper failed to disclose that he was being paid by lawyers preparing litigation against vaccine manufacturers and that he held a patent on a competing single-virus measles vaccine. These undisclosed conflicts contributed to the eventual retraction of the paper in 2010.

Broader benefit. Disclosure does not prevent funded research from being published. It allows readers to evaluate the research with full information. The system depends on honest disclosure backed by institutional and peer-review checks.

Markers reward the principle, the consequence, a named example and the broader function.

Related dot points