Primary Prevention and the “Polypill”

Samantha Gale, School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, 21 Sassoon Road, Pokfulam, Hong Kong

The use of medication in secondary and tertiary prevention is well established, for example, antihypertensives for asymptomatic hypertension to prevent ischaemic heart disease (IHD), and warfarin for ischaemic stroke survivors to prevent further strokes. Similarly, combination medicines containing more than one active ingredient, such as co-amilofruse, co-amoxiclav and co-codamol, which can simplify treatment regimens and improve patient adherence [1] have long been in regular use in both secondary and tertiary prevention.

However, in 2003, Wald and Law proposed and patented the use of a combination medicine they termed the “polypill” in the primary prevention of cardiovascular disease [2]. Based on an analysis of published meta-analyses, they suggested a theoretical formulation for a tablet that would reduce four risk factors for IHD and stroke: hypertension, hypercholesterolaemia, serum homocysteine and platelet activation. This tablet would contain 75mg aspirin, half dose of three antihypertensives, a statin and folic acid. Controversially, they proposed this tablet be given, not only to those with increased risk, but to everyone over the age of 55 years, with no baseline screening of risk factors or monitoring of effectiveness. Used in this way, they predicted the “polypill” could reduce IHD events by 88% and strokes by 80%. They calculated one third of people would benefit, experiencing an additional 11-12 years of life, whilst 15% would be expected to experience side-effects, including gastro-intestinal bleeding, haemorrhagic stroke and hypotension.

These impressive figures have been criticised due to the theoretical nature of the study, which extrapolates data from individual medications [3]. Two randomised controlled trials have since confirmed the effectiveness of similar combination pills, although again using risk factor reduction as a surrogate outcome for cardiovascular risk reduction [4,5]. The first, the Indian Polycap study (TIPS), assessed the efficacy and safety of a combination pill, termed the Polycap, versus its components in multiple formulations [4]. The Polycap contained five of the six components proposed by Wald and Law, excluding only folate. This study involved over 2,000 participants aged 45 to 80 years, each with one risk factor for cardiovascular disease. Whilst the results showed less benefit than Wald and Law’s theoretical predictions, they were still considerable with an estimated proportional risk reduction of 62% for IHD and 48% for stroke. The second study, by the PILL Collaborative Group, compared the efficacy and tolerability of a combination pill, termed the Red Heart Pill, with placebo [5]. The Red Heart Pill contained four of the six components proposed by Wald and Law (75mg aspirin, 10mg lisinopril, 12.5mg hydrochlorothiazide and 20mg simvastatin). This study involved almost 400 individuals, each with an estimated 5 year cardiovascular disease risk in excess of 7.5%. The results were similar to the TIPS study with an estimated proportional risk reduction of 60% for IHD and 56% for stroke. The risk of side effects was similar to that predicted by Wald and Law at 1 in 6.

One argument Wald and Law give for treating healthy people over the age of 55 years, regardless of other risk factors, is that age itself is the most predictive risk factor for cardiovascular disease [6]. They claim, in another theoretical study, on a simulated population, that age alone has a similar screening performance to Framingham screening [6]. However, in the Framingham Heart Study, the factor with the greatest regression coefficient and hazard ratio for women is systolic blood pressure and not age, suggesting that age is not the most predictive risk factor in women [7].

Treating on the basis of age could be viewed as a form of secondary prevention, similar to treating on the basis of hypertension or any other risk factor. However, if people are to be treated on the basis of risk, why exclude statistically significant risk factors from the risk assessment? The simplicity and lowered costs of using age alone to screen over multiple risk factors are arguably outweighed by overtreatment of those who have a low cardiovascular risk despite being over 55 years.

The key remaining question is therefore whether the “polypill” is appropriate in primary prevention. The current approach is one of secondary prevention. Individuals whose 10-year cardiovascular risk exceeds a set value, currently 20% in the UK [8], are labelled high risk and offered preventative treatment including lifestyle advice and medications in order to significantly reduce their individual risk. The key benefit of this approach is that the individual benefits from a meaningful reduction in their cardiovascular risk. Furthermore, people at low risk are not harmed from risk reduction interventions. However, the threshold for labelling individuals as high risk is somewhat arbitrary as risk is a continuous scale. A person with a 10 year cardiovascular risk of 19% is not at a ‘safe’ level of risk; yet with the high risk approach and its inherent arbitrary thresholds, they would not receive intervention. Only the relatively small numbers of people above the thresholds benefit from reduced risk. Thus to protect greater numbers of people using the high risk approach, such thresholds would have to be lowered and increasing numbers labelled with disease labels such as ‘hypertension’ and ‘hyperlipidaemia’.

A successful primary prevention strategy could be expected to have a greater impact than secondary prevention strategies. This is because there are large numbers of individuals at small risk, and only low numbers at high risk. Since the Population Attributable Risk is influenced by both the relative risk and the prevalence, the low risk group could account for more cases of disease. This is described by Geoffrey Rose as the ‘prevention paradox’. 9 Rose also demonstrated that using the ‘population strategy’, only a small improvement is required at the individual level for a significant reduction in population disease cases [9].

The simplicity of using a medication for primary prevention of cardiovascular disease is appealing. The alternative population approach involves modulating both the individual and social determinants that influence healthy behaviour through education, incentives and legislation. A recent King’s Fund report found that whilst current approaches have had a positive effect in reducing population risk, they have widened health inequalities [10]. In addition, it could be expected that education- or legislation-based strategies would take considerable time and finance to achieve risk reduction, in comparison to a tablet which reduces risk in only 12 weeks 2 and has cheap, ‘off-patent’ active ingredients.

However, the non-maleficence principle of medical ethics needs to be considered, as well as beneficence [11]. Medications have the potential for harm. Antihypertensives can cause hypotension and renal failure; myopathy is a side-effect of statins; and aspirin can cause gastro-intestinal bleeding and haemorrhagic stroke. In secondary prevention, an individual’s risk of treatment side effects is weighed against the risk of no treatment before any intervention is started. This intervention would aim for significant risk reduction for the individual. In contrast, the use of the “polypill” in primary prevention would involve exposing healthy individuals, at low risk of harm from cardiovascular disease, to a greater risk of harm from medication side-effects for modest individual benefit.

There are additional, less predictable, possible harms from using the “polypill” in primary prevention. Although individuals would no longer receive disease-labels, such as hypertension, following screening, the prescription of medicine may still act to ‘medicalise’ the population and could negatively influence self-perceptions of health and self-efficacy. Social perceptions of the older population, including their health and societal contributions, could also change. For example, treating on the basis of age could promote a concept of older age as an illness. The Health Belief Model suggests that individuals who perceive their health to be under the control of others, such as physicians, will be less likely to adopt healthy behaviours [12]. Physician prescription of the “polypill” may therefore lead to individuals taking less responsibility in leading a healthy lifestyle. Similarly, the impetus for governments to focus on improving the social determinants of cardiovascular disease may be reduced.

A further consideration is how to communicate information about the “polypill” and its benefits and risks when used in primary prevention, to the population. Individuals need to be fully-informed before deciding whether to take a pill which can cause side effects and offers only modest benefits at the individual level, in order to reduce the population risk. The lack of individual cardiovascular risk assessment makes this particularly challenging. In addition, previous statistics provided to the population through the media have been incomplete [13] and arguably misleading.

In conclusion, the use of the “polypill” in the primary prevention of cardiovascular disease could deliver impressive benefits to the population and reduce a significant health burden but at an unquantified, arguably unethical, cost to individuals. Questions that need to be answered before implementation of this intervention is further considered include: its potential effect on equity, individual health behaviour and policy priorities; the value of the personalised risk assessment in patient treatment concordance; how to fully inform the population of the benefits and risks; and its acceptability to a fully informed population.

References: 

1. Bangalore S, Kamalakkannan G, Parkar S, Messerli FH. Fixed-dose combinations improve medication compliance: a meta-analysis. Am J Med. 2007;120(8):713–719. doi:10.1016/j.amjmed.2006.08.033

2. Wald NJ, Law MR, A strategy to reduce cardiovascular disease by more than 80%. BMJ 2003; 326:1419. doi:10.1136/bmj.326.7404.1419

3. Mayor S. Polypill will not change prevention of heart disease. BMJ 2004; 329:589.2. doi:10.1136/bmj.329.7466.589-a

4. The Indian Polycap Study. Effects of a polypill (Polycap) on risk factors in middle-aged individuals without cardiovascular disease (TIPS): a phase II, double-blind, randomised trial. The Lancet 2009; 373(9672):1341-1351 doi:10.1016/S0140-6736(09)60611-5

5. PILL Collaborative Group (2011) An International Randomised Placebo-Controlled Trial of a Four-Component Combination Pill (“Polypill”) in People with Raised Cardiovascular Risk. PLoS ONE 6(5):e19857. doi:10.1371/journal.pone.0019857

6. Wald NJ, Simmonds M, Morris JK. Screening for Future Cardiovascular Disease Using Age Alone Compared with Multiple Risk Factors and Age. PLoS ONE 2011; 6(5): e18742. doi:10.1371/journal.pone.0018742

7. D’Agnostino RB, Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, Kannel WB. General Cardiovascular Risk Profile for Use in Primary Care. The Framingham Heart Study. Circulation 2008; 117:743-753 doi:10.1161/CIRCULATIONAHA.107.699579

8. National Institute for Health and Clinical Excellence. Lipid modification: cardiovascular risk assessment and the modification of blood lipids for the primary and secondary prevention of cardiovascular disease. 2008. Available online at: http://guidance.nice.org.uk/CG67 (Accessed 31 March 2013)

9. Rose G. Strategy of Preventive Medicine. New York: Oxford University Press 1992

10. Buck D, Frosini F Clustering of unhealthy behaviours over time: Implications for policy and practice. The King’s Fund 2012

11. Beauchamp TL, Childress JF. Principles of Biomedical Ethics. 6th Edition. USA: Oxford University Press 2008

12. Rosenstock IM, Strecher VJ, Becker MH. Social learning theory and the Health Belief Model. Health Educ Behav 1988; 15:175-183 doi:10.1177/109019818801500203

13. McCartney M. The press release, relative risks, and the polypill. BMJ 2011;343:d4720. doi:10.1136/bmj.d4720

Story image from Wikimedia Commons