The limits of user-choice in specialised health and human services

Thu-Trang Tran, University of Melbourne

Thu-Trang is a PhD candidate in the Melbourne School of Government at the University of Melbourne, researching wisdom in public administration. She is a senior strategy advisor in the Victorian Government and a sessional lecturer at the University of Melbourne and La Trobe University, specialising in governance, social policy and issues in international development. Thu-Trang holds a Master of Laws and Bachelor degrees in Law and Commerce from UNSW and a Graduate Diploma in Legal Practice from the College of Law, Sydney. The views in this article are her own.

Credence attributes of specialised services demand a shift in the Productivity Commission’s reform proposals to focus on supply characteristics, rather than user characteristics such as informed user choice.

Specialised health and human services have ‘credence attributes’ that make it difficult for users to discern the ‘true’ technical qualities of the care they receive, even after they have received the service and all the associated information. Recent behavioural economics research confirms that patients were inaccurate in assessing the technical quality of care provided by physicians.

The implications of this are significant. 

The strategy of more information, informed user choice, and increased competition, rests on flawed assumptions and a misconstruction of the nature of the problems (and thus misdirects reform efforts). 

To account for the credence nature of health services, the Productivity Commission needs to expressly address supply-side reforms and integrated regulatory frameworks to ensure the requisite technical quality of care in the first instance. This then enables user-choice, which can only occur on the non-technical qualities of care (such as a physician’s attentiveness, cleanliness of the clinic), service attributes that users can be reasonably expected to inform themselves of.

Recent behavioural economic research and theories on ‘credence’ attributes of goods and services fundamentally challenges the assumptions underpinning the Productivity Commission’s recommendations and adds another layer of important considerations that must be factored into any design of market and regulatory reforms.

Behavioural research in health economics suggests user-choice is not premised on technical qualities of care, but on non-technical qualities: [1]

…patients have been shown to accurately report information on whether a physician was respectful, attentive, clear in explaining clinical issues and operated a clean and efficient office. These same patients were found to be inaccurate in reporting of the technical quality of care. That is, they were not accurate in judging whether a physician supplied appropriate evidence based treatment.[2]

This can be explained using the Search, Experience, Credence (SEC) attributes framework. These categories describe the how and at what point, “if ever, consumers can accurately assess a product's performance.”[3] Product here is used to mean goods and services.

Search attributes are qualities of good or services that consumers can assess by inspection before making a purchase, such as a colour of a car or cleanliness of a clinic. Experience attributes are qualities that cannot be determined prior to purchase, but can reasonably be discerned from experience or consumption, such as the taste of foods or whether a car can be driven.[4]

Credence goods or services have certain qualities that the average consumer cannot assess their true performance due to a combination of product and user factors.

Consumers may not possess sufficient technical expertise to assess the product's true performance, to diagnose his/her own need for the product or service, or because diagnosing a need separately from filling the need at the same time is uneconomical or difficult.[5]

Examples of credence services include complex automobile repairs or medical services such as root canal surgery. “In both cases the consumer is unlikely to know with absolute certainty whether the diagnosis was correct and is unlikely to have the technical expertise to assess the quality of the service provided.”[6] In effect, it would take another expert in the same professional field to make a proper assessment of the true quality of the service or medical treatment.

Credence characteristics of health services can be explained in terms of information asymmetry or ‘user characteristics’[7] as the Productivity Commission has done. However, such framing inevitably leads to problematic proposals situated in the domain of users, such as ameliorating their decision-making processes with more information.[8] While the Productivity Commission has recognised that users lack the technical expertise required to make an informed choice, the Productivity Commission’s cursory note of user-driven workarounds and user-initiated remedies for poor quality of services are wanting:

Level of expertise required to compare alternatives — Even if a user (or their decision maker) has access to information about alternative services, they may not have the technical expertise to make an informed choice. This can be the case where individuals are highly reliant on their doctor to make decisions for them about complex forms of healthcare. A lack of technical expertise may not be an insurmountable barrier if users can obtain a second opinion from another provider or engage an independent expert to advise on alternatives (noting that seeking advice from other sources will tend to raise the cost to the recipient of the service).

Indeed, obtaining a second opinion necessarily brings about the exact credence attribute problems in first instance, and adds to the vulnerability (and arguably confusion and trauma) experienced by service users in situations of immense power imbalance. These other forms of ‘consumer detriment’[9] (in addition to cost) need to be duly recognised and accounted for to ensure a holistic framework of government interventions to make service users better off.

The Productivity Commission goes on to note that people can learn from their family and friends. Behavioural research suggests that people do rely on family and friends when choosing a physician (rather than systematic information).

The implication here is that patients will commonly develop a prior about a primary care physician by relying on reports from family and friends that will be based on observations about some dimensions of medical care and perhaps not on the dimensions that most directly affect their health outcomes.[10]

This can be problematic with respect to health services as family and friends similarly are not in a position to distil the salient technical features of the quality of care provided. Thus this heuristic may create unintended distortions in the market that redirects focus on service dimensions that do not directly affect user health outcomes.

Consumer Affairs Victoria similarly sounds a bell of warning in its discussion paper on credence attributes: [11]

Services can also be credence attributes and this can create incentives for fraudulent behaviour by sellers. Sometimes sellers not only provide the service; they act as experts determining the requirements of consumers. […]

Consumers may never discover whether the advice they acted on was optimal, or even effective. Fraud and over-servicing are more likely when:

  • diagnosis and follow-up occurs jointly, and
  • verifying quality of the end result is difficult or costly, because suppliers think the probability of detection is low. [12]

Consequently, there is a need to revisit the rationale behind the identification of public hospital services, public dental care, and specialist palliative care as a suitable service sectors for reforms, and reassess the efficacy of ‘user informed choice’ as a key driver of service improvements.

Using the SEC framework, it appears that the Productivity Commission’s rationale in the Preliminary Findings Report is based on conceptualising health services as ‘experience’ rather than ‘credence’ services. The proposals for improving user choice[13] thus do not delve into the necessary analytical and evidentiary depth about the saliency of information for service users when they assess the care they received.

The starting point is that irrespective of the full provision of information about the services and service providers, users are not in a position (not through lack of capacity to process information) to properly assess the technical quality of care.

The better assumption to make is that users can only make informed choices about non-technical aspects of health services. While non-technical qualities of care may also act as proxies for assessing technical qualities of care and may also be pertinent considerations in the eyes of services users, they should not be the primary basis for driving improved health outcomes for service users.

The stewardship role of government thus requires careful integration of government interventions and calibration of policy levers to assure the technical quality of care in the first instance. This means focusing on supply characteristics, rather than user or demand characteristics. That is, improvements in the quality of care flow not from the indirect mechanism of informed user-choice, but directly from supply-side measures such as well-trained and qualified service providers. An integrated framework also entails looking not just at point-of-purchase, but also the rest of regulatory and service regime, including post-purchase when the quality of care is not delivered or sustained.

One missing part of the puzzle in the Productivity Commission’s discourse to date is the interaction of the proposed reforms with the Australian Health Practitioner Regulatory Agency’s (AHPRA) framework to ensure a robust monitoring, compliance and enforcement regime[14] that ensures ease of access to justice and meaningful remedies for consumers, especially those experiencing vulnerability and disadvantage[15]. This author notes that APRHA’s complaints model can result in outcomes that penalises health practitioners that have been found to be in breach of the required standard of care. This model however does not directly provide redress for the detriment suffered by services users.

This may mean health regulatory agencies need to be empowered with broader powers, and to play a greater role in supporting service users get meaningful redress. Tools from consumer policy can be examined for appropriate application to healthcare services that simultaneously address the misconduct of service providers, provide remedies for aggrieved service users, and ensure affordable and available access to justice for users.

In summary, the ‘devil is in the detail’ of the market regulatory framework. The Productivity Commission’s recommendations premised on ‘informed user choice’ rests on assumptions that do not marry up to emerging evidence about how people interact with and choose health services. The Productivity Commission’s proposals need to clearly lay out aspects of their market-based interventions to deliver real impacts for consumers that are fully informed by the evidence specific to the healthcare services undergoing reform.


[1] See Richard G Frank (2004), Behavioral Economics and Health Economics”, National Bureau Of Economic Research Working Paper 10881, Cambridge, at, accessed 20 October 2016, p14.

[2] Citing Edgman-Levitan, S., and P.D. Cleary (1996), “What Information Do Consumers Want and Need?” Health Affairs 15(4): 42-56.

[3] See Gary T. Ford, Darlene B. Smith, and John L. Swasy (1988) ,"An Empirical Test of the Search, Experience and Credence Attributes Framework", in NA - Advances in Consumer Research Volume 15, eds. Micheal J. Houston, Provo, UT : Association for Consumer Research, Pages: 239-244.

[4] Frank (2004), see note 1.

[5] Citing Darby, M. and Karni, E. (1973), Free competition and the optimal amount of fraud, Journal of Law and Economics, 16, pp. 67-88 in Ford (1988), see note 4.

[6] Ford et al (1988), see note 3.

[7] Productivity Commission (2016), Human Services: Identifying sectors for reform, Issues Paper, Canberra, p14.

[8] Productivity Commission (2016), see note 7, p14.

[9] Consumer Affairs Victoria (2006), Consumer detriment in Victoria: a survey of its nature, costs and implications, Research Paper No. 10, Melbourne.

[10] Frank (2004), see note 1, p14.

[11] See Consumer Affairs Victoria (2010), Credence Attributes: Making honesty the best policy, Discussion paper, at, accessed 20 October 2016, p6-7.

[12] Citing Darby and Karni (1973) in Consumer Affairs Victoria (2010), see above note 11.[13] Productivity Commission (2016), Introducing Competition and Informed User Choice into Human Services: Identifying Sectors for Reform, Preliminary Findings Report, Canberra, pp17, 20-23.

[14] Consumer Affairs Victoria (2010), see note 11, p15.

[15] Consumer Affairs Victoria (2004), What do we mean by 'vulnerable' and 'disadvantaged' consumers?, Discussion Paper, Melbourne.