The Vipeholm Dental Caries Studies and the capacity for informed consent
Philip M. Gaughwin PhD* |

*Medical Student, The Australian National University

The Vipeholm studies illustrate two well-known sayings: (1) the end sometimes justifies the means, and (2) it is easy to be wise after the event.
–Bo Krasse, senior researcher Vipeholm Dental Caries Experiments 1945-1955 University of Goteborg, 2001 (2)



n Sweden at the beginning of the 20th century, tooth decay had become increasingly prevalent as its population industrialised and shifted from a rural to an urban habitat. By the 1930s, dental caries were prevalent in 83% of children’s deciduous teeth, and in over 99% of army conscripts, a figure comparable to the general population (1, 2). While several animal model and observational studies suggested a link between dietary consumption and caries formation, the cause of this highly prevalent, chronic disease was not known at the time (3, 4). 

In order to establish an effective preventative public health strategy, it was necessary to identify the cause of dental caries.  To this end, the government commissioned the Folktandvåren (Public Dental Service) and Medical Board to establish an interdisciplinary team of clinicians and scientists at Vipeholm, a state hospital for the “uneducable mentally deficient” in Lund (a small university town in southern Sweden).  The resulting ten year study (1945-1955) was one of the first taxpayer-funded scientific investigations ever performed (5).  The Vipeholm Dental Caries Study has since been described as “…one of the most important contributions in the entire dental literature” as it:
“…definitively established that the more frequently sugar is consumed, the greater the risk, and that sugar consumed between meals has a much greater caries potential than when consumed during a meal” (6).

Vipeholm’s ethical transgressions have been a consistent focus for retrospective analysis in Sweden, but are not well known to the international medical community.  Perhaps the most egregious is that all patients involved in the study had not given their informed consent, regardless of their capacity to do so (7). Here, we will briefly describe the Vipeholm experiments, and discuss aspects of the study regarding consent in contemporary medical research.

The Vipeholm studies: rationale, design, and outcomes

Why Vipeholm? 

The following summary of the Vipeholm experiments summarises extensive work that is presented in greater detail elsewhere (1, 2, 5, 8, 9).  Vipeholm was the only state hospital in Sweden for individuals who were intellectually disabled[$], and housed 800 patients from across Sweden.  The Vipeholm patient population was selected by the Folktandvåren for what was considered a highly prestigious, long-term and expensive investment of taxpayer funds for purely scientific reasons. In order to establish a link between dietary consumption and caries formation, it was necessary to perform a controlled investigation in a large, permanently

housed, adult patient population for whom exposure (dietary consumption) could be standardised and outcomes (caries formation, saliva composition, etc) quantified in minute detail, over a period of years. Vipeholm was the only state-owned medical facility in Sweden that met these requirements (2).

The Vipeholm sugar experiments: methods and findings

Initial experiments exposed patients to supplements (Vitamin A, B, C, D, and Fluoride) in conjunction with regular meals in order to see if it reduced caries formation.  By 1947, the researchers had failed to demonstrate any clear protective effect.  The researchers next investigated another line of evidence suggesting an association between sugar and dental caries formation.  In this second series of experiments, sugar was added to the basic (standard) diet of 436 patients in various controlled formats: dissolved in beverages, baked into bread, in plain chocolate, or varying numbers of toffees or caramels over the course of a day.  Control patients were given fats instead of fermentable carbohydrates to compensate for calorie consumption.  Saliva samples were taken every 15 minutes through the day and analysed for sugar and acid levels (5, 10).

By 1949, the study found that, relative to control patients, those who consumed 24 toffees developed 4.02 new carious tooth surfaces per person per year, relative to control patients (0.30 caries per person per year).  In particular, the study established that sticky sweets consumed between meals (rather than within meals) significantly increased carious tooth surface formation (2).  Ethical issues were not discussed formally either before or during the study.  Prior to 1950, patients received no dental care.  By the end of the study, 2,125 new dental caries had been induced in the study period (8).

The Vipeholm experiments: reactions and consequences

When the results emerged in 1953 (11), there was an outcry. Other scientists claimed the evidence had been ‘bought’ by the confectionary and sugar industries.  The media argued the studies were inhumane, and provoked national outrage (5). The parliament was stunned, and introduced a bill to halt funding for further experiments, and to prevent the use of patients at Vipeholm as research subjects from July 1955 onwards (10). 

Because the experiments had employed a rigorous scientific methodology and controlled for the introduction of bias, the researchers were able to successfully defend their experiments to the government and to the public at large.  The results of the Vipeholm experiments were incorporated into a major public health strategy in 1957, and the Folktandvåren became engaged in a successful ongoing campaign of preventative oral health (2).

Capacity and Consent in Medical Research

Ethical and legal considerations of the Vipeholm experiments

Vipeholm was initiated (1944) prior to the Nuremberg Code (1947) and the Declaration of Helsinki (1964), which codify ethical considerations for human experimentation. The Nuremberg Code emphasised that research participants provide voluntary and informed consent free from coercion (article 1) and that the research should not be conducted where there is a priori reason for inducing disabling injury (article 5) (2). In order for an individual to consent, they must (1) be given accurate and balanced information, (2) have the capacity to consent, and (3) make an autonomous and voluntary choice to participate (article 1). The 1975 revisions of the Declaration of Helsinki (adopted by the World Medical Association)(2) later introduced the requirement for an independent, institutional review board (IRB) oversight (Article 4) and formal, ethical consideration of the research protocol (Article 5) (7). In the United States, the Willowbrook study (where children with intellectual disabilities were injected with hepatitis in order to observe its progression in situ) resulted in federal commissions that outlawed inhumane and exploitative research (12,13).  In the context of Australian law, the National Statement on Ethical Conduct of Research Involving Humans (1999, revised 2007) (4) codifies four principles of ethical research: (1) Research Merit and Integrity, (2) Justice, (3) Beneficence and (4) Respect for persons. The ethical ramifications of the Vipeholm studies are described in detail elsewhere (7).

Consent and contemporary medical research for the intellectually disabled: ethical, legal and human rights considerations
The Vipeholm and Willowbrook studies have not discouraged interest in contemporary medical research into intellectually disabled populations. For example, psychiatric disorders are more frequent in the intellectually disabled population.  Consequently, psychiatric treatments hold great promise for improving the quality of life for intellectually disabled persons (14).  Of increasing importance is the extent to which multiple centre, randomised controlled trials (necessary for the therapeutic evaluation of psychiatric drugs) can be conducted for intellectually disabled persons when legal and ethical requirements, particularly concerning informed consent, differ between countries (15).

Edwards (2000) summarised three arguments for the absolute exclusion of individuals with intellectual disabilities from participation in medical research (16).  Firstly, there is the special vulnerability of the research subject. In an institutional setting where a patient is absolutely dependent on others for their wellbeing and social interaction, consent may be given out of desire to be seen to cooperate or ‘help’ their doctors and carers.  Secondly, it can be argued that all research is fundamentally non-therapeutic, in that it is not primarily intended to directly benefit the potential participant.  Thirdly, while one can choose to be used as means to another’s ends, it is morally illegitimate to use others in this way in the absence of their consent. Alternatively, McDonald and Keys (2009) have argued that blanket prohibitions imposed by IRBs may actually limit the autonomy and self-determination of intellectually disabled individuals and thereby increase their marginalisation in society (17).

The current legal and international human rights framework emphasises the capacity of the intellectually disabled for social, physical, emotional and intellectual development.  Autonomy and independence are central features of the United Nations Convention on the Rights of Persons with Disabilities (5).  In Australia, the Disability Act (2006 (Vic)) upholds the rights of those with a disability to human worth, to live free from abuse, neglect or exploitation, and the right to exercise control over their own lives (6).  Under these principles, an individual who assents to participation in medical research cannot be legally obstructed from doing so.  Conversely, any indication by the individual that they refuse to participate must be respected, and participation in the study stopped. 

Guardianship and consent by proxy for medical research
Often, it is difficult to ascertain patient preferences in the absence of decisional capacity. It has been common practice for medical researchers to seek the consent of a proxy or guardian in order to enable participation of intellectually disabled individuals in a research project. Here, the common law position is less clear. Guidelines for decisional competency, including the involvement of proxies or guardian decision makers, were established for medical treatment, not research.  Consequently, there is a tendency for research participation to be assessed on a per-case basis.  A guardian makes the decision to participate based on two principles: patient preferences and wishes, and what is in the best interests of the patient (14).  Importantly, the consideration of ‘best interests’ often is widened to include the interests of similarly placed persons (i.e., intellectually disabled individuals similar to the person being asked to participate).  Regardless, it remains the responsibility of the researchers, not the guardians or patients, to ensure that the rights of the individual are respected at all stages of the research process and to meet the requirements of consent under the law (13).


The Vipeholm studies have had a substantial influence on public health policy and fundamental scientific understanding.  The study identified the cause and means of prevention of a disease that had been long present in human history, and had markedly increased as a consequence of urbanisation and industrialisation. In other ways, Vipeholm serves as a marker against which progress in ethical research conduct can be measured, from paternalistic to reflective and positive engagement.. 


The author is indebted to Elin Bommenel (Lund University) for helpful discussions. 

Conflicts of Interest Statement

The author declares no conflicts of interest.

[$]The term intellectual disability will be used to refer to research participants of the Vipeholm study, although this population likely also housed individuals with severe epilepsy and other disorders. For the purposes of this study, we adopt the World Health Organization (WHO 1992) definition: “a condition of arrested or incomplete development of the mind, which is especially characterised by impairment of skills manifested during the developmental period, which contribute to the overall level of intelligence, i.e. cognitive, language, motor and social abilities.” ICD-10 Classification of Mental and Behavioral Disorders: Clinical descriptions and diagnostic guidelines (World Health Organization, Geneva, 1992), pp. 227-230



  1. Krasse B. Why was the Vipeholm Study done and why have this symposium? Scand J Dent Res. 1989 Apr 1;97(2):99-102.
  2. Krasse B. The Vipeholm Dental Caries Study: Recollections and Reflections 50 Years Later. J Dent Res. 2001 Sep 1;80(9):1785-8.
  3.  [Editorial]. Sweet Tooth And Decay. Brit Med J 1958 Aug 2;2(5091):293.
  4.  [Editorial]. Diet and Prevention Of Dental Disease. Brit Med J. 1965 Nov 20;2(5472):1197-8.
  5. Bommenel E. {Introduction}. “Sockerförsöket: kariesexperimenten 1943-1960 på Vipeholms sjukhus för sinnesslöa” [PhD Dissertation, Linköping University (Swedish)]. Linköping Studies in Arts and Science 2006; 348: 1-9.
  6. Zero DT. Sugars: The Arch Criminal? Caries Res. 2004 Jan 1;38(3):277-85.
  7. Petersson B. The mentally retarded as research subjects. A research ethics study of the Vipeholm investigations of 1945-1955. Studies in Research Ethics. 1993;3:1-32.
  8. Bommenel E. {Process and Analysis of Results.} “Sockerförsöket: kariesexperimenten 1943-1960 på Vipeholms sjukhus för sinnesslöa” [PhD Dissertation, Linköping University (Swedish)]. Linköping Studies in Arts and Science 2006; 348: 55-66.
  9. Bommenel E. {Vipeholm Experiments - a summary.} “Sockerförsöket: kariesexperimenten 1943-1960 på Vipeholms sjukhus för sinnesslöa” [PhD Dissertation, Linköping University (Swedish)]. Linköping Studies in Arts and Science 2006; 348: 18-54.
  10. Fiske J. Research governance. Brit Dent J. 2007; 203(3): 117.
  11. Gustafsson B, Quensel C-E, Lanker L, Lundqvist C, Grahnen H, Bonow B, et al. The Vipeholm caries study.  Acta Odontologica Scandinavia. 1954;11(3-4):232-64.
  12. Diamond EF. The Willowbrook experiments. The Linacre quarterly. 1973 May; 40(2):133-7.
  13. Iacono T, Murray V Issues of informed consent in conducting medical research involving people with intellectual disability. J Appl Res Intellect 2003; 16: 41-51.
  14. Fisher CB, Cea CD, Davidson PW, Fried AL. Capacity of persons with mental retardation to consent to participate in randomized clinical trials. Am J Psychiatry. 2006 Oct 1;163(10):1813-20.
  15. Veenstra MY, Walsh PN, van Schrojenstein Lantman-de Valk HMJ, Haveman MJ, Linehan C, Kerr MP, et al. Sampling and ethical issues in a multicenter study on health of people with intellectual disabilities. J Clin Epidemiol. 2010 Oct 1;63(10):1091-100.
  16. Edwards SD. An argument against research on people with intellectual disabilities. Med Health Care Philos. 2000 Jan 1;3(1):69-73.
  17. McDonald KE, Keys CB. How the powerful decide: access to research participation by those at the margins. Am J Community Psychol. 2008 Sep 1;42(1-2):79-93.