Fixed-dose combination therapy for the prevention of cardiovascular disease

  • Review
  • Intervention

Authors


Abstract

Background

Cardiovascular disease (CVD) is the leading cause of death and disability worldwide, yet CVD risk factor control and secondary prevention rates remain low. A fixed-dose combination of blood pressure and cholesterol lowering and antiplatelet treatments into a single pill, or polypill, has been proposed as one strategy to reduce the global burden of CVD by up to 80% given its potential for better adherence and lower costs.

Objectives

To determine the effectiveness of fixed-dose combination therapy on reducing fatal and non-fatal CVD events and on improving blood pressure and lipid CVD risk factors for both primary and secondary prevention of CVD. We also aimed to determine discontinuation rates, adverse events, health-related quality of life, and costs of fixed-dose combination therapy.

Search methods

We searched the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library (2013, Issue 6), MEDLINE Ovid (1946 to week 2 July 2013), EMBASE Ovid (1980 to Week 28 2013), ISI Web of Science (1970 to 19 July 2013), and the Database of Abstracts of Reviews of Effects (DARE), Health Technology Assessment Database (HTA), and Health Economics Evaluations Database (HEED) (2011, Issue 4) in The Cochrane Library. We used no language restrictions.

Selection criteria

We included randomised controlled trials of a fixed-dose combination therapy including at least one blood pressure lowering and one lipid lowering component versus usual care, placebo, or a single drug active component for any treatment duration in adults ≥ 18 years old with no restrictions on presence or absence of pre-existing cardiovascular disease.

Data collection and analysis

Three review authors independently selected studies for inclusion and extracted the data. We evaluated risk of bias using the Cochrane risk of bias assessment tool. We sought to include outcome data on all-cause mortality, fatal and non-fatal CVD events, adverse events, changes in systolic and diastolic blood pressure, total and low density lipoprotein (LDL) cholesterol concentrations, discontinuation rates, quality of life, and costs. We calculated risk ratios (RR) for dichotomous data and weighted mean differences (MD) for continuous data with 95% confidence intervals (CI) using fixed-effect models when heterogeneity was low (I2 < 50%) and random-effects models when heterogeneity was high (I2 > 50%).

Main results

We found nine randomised controlled trials with a total of 7047 participants. Seven of the nine trials evaluated the effects of fixed-dose combination therapy on primary CVD prevention, and the trial length ranged from six weeks to 15 months. We found a moderate to high risk of bias in the domains of selection, performance, detection, attrition, and other types of bias in five of the nine trials. Compared with the comparator groups, the effects of the fixed-dose combination treatment on mortality (1.2% versus 1.0%, RR 1.26, 95% CI 0.67 to 2.38,  N = 3465) and cardiovascular events (4.0% versus 2.9%, RR 1.38, 95% CI 0.91 to 2.10, N = 2479) were uncertain (low quality evidence). The low event rates for these outcomes, limited availability of data as only two out of nine trials reported on these outcomes, and a high risk of bias in at least one domain suggest that these results should not be viewed with confidence. Adverse events were common in both the intervention (30%) and comparator (24%) groups, with participants randomised to fixed-dose combination therapy being 20% (95% CI 9% to 30%) more likely to report an adverse event. Notably, no serious adverse events were reported. Compared with placebo, the rate of discontinuation among participants randomised to fixed-dose combination was higher (14% versus 11%, RR 1.26 95% CI 1.02 to 1.55). The weighted mean differences in systolic and diastolic blood pressure between the intervention and control arms were -7.05 mmHg (95% CI -10.18 to -3.87) and -3.65 mmHg (95% CI -5.44 to -1.85), respectively. The weighted mean differences (95% CI) in total and LDL cholesterol between the intervention and control arms were -0.75 mmol/L (95% CI -1.05 to -0.46) and -0.81 mmol/L (95% CI -1.09 to -0.53), respectively. There was a high degree of statistical heterogeneity in comparisons of blood pressure and lipids (I2 ≥ 70% for all) that could not be explained, so these results should be viewed with caution. Fixed-dose combination therapy improved adherence to a multi-drug strategy by 33% (26% to 41%) compared with usual care, but this comparison was reported in only one study. The effects of fixed-dose combination therapy on quality of life are uncertain, though these results were reported in only one trial. No trials reported costs.

Authors' conclusions

Compared with placebo, single drug active component, or usual care, the effects of fixed-dose combination therapy on all-cause mortality or CVD events are uncertain; only few trials report these outcomes and the included trials were primarily designed to observe changes in CVD risk factor levels rather than clinical events. Reductions in blood pressure and lipid parameters are generally lower than those previously projected, though substantial heterogeneity of results exists. Fixed-dose combination therapy is associated with modest increases in adverse events compared with placebo, single drug active component, or usual care but may be associated with improved adherence to a multidrug regimen. Ongoing trials of fixed-dose combination therapy will likely inform key outcomes.

Résumé scientifique

Traitement combiné à dose fixe pour la prévention des maladies cardiovasculaires

Contexte

La maladie cardio-vasculaire (MCV) est la principale cause de décès et d'invalidité dans le monde. Toutefois, le contrôle des facteurs de risque des MCV et les taux de prévention secondaire restent faibles. La combinaison à dose fixe en une seule pilule, ou poly pilule, de traitements antiplaquettaires et diminuant la pression artérielle et le cholestérol, a été proposée comme stratégie pour réduire la charge mondiale des MCV jusqu'à 80 %, étant donné son potentiel pour une meilleure adhérence et une réduction des coûts.

Objectifs

Déterminer l'efficacité du traitement combiné à dose fixe sur la réduction des effets mortels et non mortels des MCV et sur l'amélioration de la pression artérielle et des facteurs de risque lipidiques des MCV pour la prévention primaire et secondaire des MCV. Nous avons également cherché à déterminer les taux d'abandon, les effets indésirables, la qualité de vie liée à la santé et les coûts du traitement combiné à dose fixe.

Stratégie de recherche documentaire

Nous avons effectué des recherches des essais contrôlés du registre Cochrane (CENTRAL) dans La Bibliothèque Cochrane ( 2013, numéro 6), MEDLINE Ovid (de 1946 à la 2ème semaine de juillet 2013), EMBASE Ovid (de 1980 à la semaine 28 de 2013), ISI Web of Science (de 1970 au 19 juillet 2013) et Database of Abstracts of Reviews of Effects (DARE), Health Technology Assessment Database (HTA), Health Economics Evaluations Database (HEED) (2011, numéro 4) dans la Bibliothèque Cochrane . Nous n'avons utilisé aucune restriction concernant la langue.

Critères de sélection

Nous avons inclus des essais contrôlés randomisés portant sur un traitement combiné à dose fixe et incluant au moins un composant pour abaisser la pression artérielle et un hypolipidémiant par rapport aux soins habituels, à un placebo ou à une composante médicamenteuse unique pour n'importe quelle durée de traitement chez les adultes ? de 18 ans, sans aucune restriction concernant la présence ou l'absence de maladie cardio-vasculaire préexistante.

Recueil et analyse des données

Trois auteurs de la revue ont indépendamment sélectionné les études à inclure et extrait les données. Nous avons évalué le risque de biais en utilisant l'outil d'évaluation Cochrane de risque de biais. Nous avons cherché à inclure les données des résultats sur la mortalité toutes causes confondues, les effets mortels et non mortels des MCV, les effets indésirables, les changements de la pression artérielle systolique et diastolique, les taux de cholestérol total et les taux de cholestérol à lipoprotéines à basse densité (LDL), les taux d'abandon, la qualité de vie et les coûts. Nous avons calculé les risques relatifs (RR) pour les données dichotomiques et les différences moyennes pondérées (DM) pour les données continues avec des intervalles de confiance (IC) à 95 % en utilisant des modèles à effets fixes lorsque l'hétérogénéité était faible (I 2 < 50 %) et des modèles à effets aléatoires lorsque l'hétérogénéité était élevée (I 2 > 50 %).

Résultats principaux

Nous avons trouvé neuf essais contrôlés randomisés avec un total de 7 047 participants. Sept des neuf essais évaluaient les effets du traitement combiné à dose fixe sur la prévention primaire des MCV et la durée des essais allait de 6 semaines à 15 mois. Nous avons trouvé un risque de biais variant de modéré à élevé dans les domaines de sélection, de performance, de détection, d'attrition et d'autres types de biais dans cinq des neuf essais, Comparés aux groupes de comparaison, les effets du traitement combiné à dose fixe sur la mortalité (1,2 % par rapport à 1,0%, RR de 1,26, IC à 95 % de 0,67 à 2,38, N = 3 465) et sur les effets cardio-vasculaires (4,0 % par rapport à 2,9%, RR de 1,38, IC à 95 % de 0,91 à 2,10, N = 2 479) étaient incertains (preuves de faible qualité). Le faible taux d'effets pour ces critères de jugement, la disponibilité limitée des données, car seuls deux des neuf essais rendaient compte de ces critères de jugement, et le risque élevé de biais dans au moins un domaine, suggèrent d'interpréter ces résultats avec prudence. Les effets indésirables étaient courants dans les groupes d'intervention (30 %) et dans les groupes de comparaison (24 %), avec 20 % des participants randomisés au traitement combiné à dose fixe (IC à 95 % de 9 % à 30 %) plus susceptibles de rapporter un effet indésirable. Surtout, aucun effet indésirable grave n'a été rapporté. Par rapport à un placebo, le taux d'abandon chez les participants randomisés au traitement combiné à dose fixe était plus élevée (14 % par rapport à 11%, RR de 1,26, IC à 95 % de 1,02 à 1,55). Les différences moyennes pondérées de la pression artérielle systolique et diastolique entre le groupe d'intervention et le groupe témoin étaient de -7,05 mmHg (IC à 95 % de -10,18 à -3,87) et de -3,65 mmHg (IC à 95 % de -5,44 à -1,85), respectivement. Les différences moyennes pondérées (IC à 95 %) pour le cholestérol total et à LDL entre le groupe d'intervention et le groupe témoin étaient de -0,75 mmol / l (IC à 95 % de -1,05 à -0,46) et de -0,81 mmol / l (IC à 95 % de -1,09 à -0,53), respectivement. Un degré élevé d'hétérogénéité statistique dans les comparaisons de la pression artérielle et des lipides (I 2 ≥ 70 % pour toutes) ne pouvait pas être expliqué, ces résultats doivent donc être interprétés avec prudence. Le traitement combiné à dose fixe améliorait de 33 % (26 % à 41 %) l'adhérence à une stratégie à plusieurs médicaments par rapport aux soins habituels, mais cette comparaison a été rapportée dans une seule étude. Les effets du traitement combiné à dose fixe sur la qualité de vie sont incertains, bien que ces résultats aient été rapportés dans un seul essai. Aucun essai n'a rapporté les coûts.

Conclusions des auteurs

Par rapport à un placebo, à une composante médicamenteuse active unique ou aux soins habituels, les effets du traitement combiné à dose fixe sur la mortalité toutes causes confondues ou sur les effets des MCV étaient incertains; seuls quelques essais rendaient compte de ces critères de jugement et les essais inclus étaient principalement conçus pour observer des changements sur les niveaux de facteur de risque cardio-vasculaire plutôt que sur les effets cliniques. Les baisses de la pression artérielle et des paramètres lipidiques sont généralement plus faibles que celles précédemment prévues, bien que les résultats soient substantiellement hétérogènes. Le traitement combiné à dose fixe est associé à une augmentation modeste des effets indésirables par rapport à un placebo, à une composante médicamenteuse active unique ou aux soins habituels, mais peut être associé à une meilleure adhérence à un schéma thérapeutique à médicaments multiples. Les essais en cours sur le traitement combiné à dose fixe rendront probablement compte des principaux critères de jugement.

Plain language summary

Fixed-dose combination therapy for the prevention of cardiovascular disease

Cardiovascular diseases (CVD), including heart attacks and strokes, are the leading cause of death and disability worldwide. Drug therapy with blood pressure and cholesterol lowering medications, particularly statins, have been proven to reduce the likelihood that individuals will experience a fatal or non-fatal cardiovascular event. Aspirin has also been proven to prevent heart attacks, certain types of strokes, and death in people with prior cardiovascular disease. The concept of fixed-dose combination therapy is to combine mulitple medications in a single pill as this has been shown to improve adherence in patients with high blood pressure and human immunodeficiency virus (HIV). There have been recent randomised controlled clinical trials to evaluate the effect of fixed-dose combination therapy for CVD prevention. The aim of this systematic review was to determine the effects of fixed-dose combination therapy on all-cause mortality, fatal and non-fatal CVD events, adverse events, blood pressure, lipids, discontinuation rates, quality of life, and costs for CVD prevention.

We searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, and EMBASE until 2013. We found nine randomised controlled trials of two-drug through to five-drug fixed-dose combination therapy with placebo, single drug active component, or usual care in 7047 patients, dating from 2009 to 2013. Trials were generally short-term, ranging from six weeks to 15 months, and included middle-age adults with and without prior CVD.

Compared with placebo, single drug active component, or usual care, the effects of fixed-dose combination therapy on all-cause mortality or CVD events were uncertain. However, the event rates for these outcomes were relatively uncommon, only two out of nine trials reported these outcomes, these trials were primarily designed to observed changes in CVD risk factor levels rather than clinical events, and the trials had a high risk of bias in at least one domain, suggesting that these results should not viewed with confidence. Of 1000 people treated with fixed-dose combination therapy during the study period, 297 (range 264 to 315) would experience a side effect compared with 242 people treated with placebo. Fixed-dose combination therapy was associated with lower systolic blood pressure (-7.05 mmHg, range -10.18 to -3.87) and total cholesterol (-0.75 mmol/L, range -1.05 to -0.46). However, there was a high degree of statistical heterogeneity in these comparisons so these results should be viewed with caution. Of 1000 patients treated with fixed-dose combination therapy during the study period, 140 (range 122 to 186) would discontinue the therapy compared with 115 patients treated with placebo. The effects on quality of life were uncertain, and no cost data were reported. Ongoing trials of fixed-dose combination therapy will likely inform these important endpoints.

Résumé simplifié

Traitement combiné à dose fixe pour la prévention des maladies cardiovasculaires

Les maladies cardiovasculaires (MCV), y compris les crises cardiaques et les AVC, sont la principale cause de décès et d'invalidité dans le monde. Les traitements utilisant des médicaments pour abaisser la pression artérielle et le cholestérol, en particulier les statines, se sont révélés efficaces pour réduire la probabilité d'événements cardiovasculaires mortels ou non mortels. L'aspirine a également démontré prévenir les crises cardiaques, certains types d'AVC et les décès chez les patients ayant des antécédents de maladie cardio-vasculaire. Le concept du traitement combiné à dose fixe est de combiner plusieurs médicaments en une seule pilule car il a démontré améliorer l'adhérence au traitement chez les patients souffrant d'une pression artérielle élevée et du virus de l'immunodéficience humaine (VIH). De récents essais cliniques contrôlés randomisés ont évalué le traitement combiné à dose fixe pour la prévention des MCV. L'objectif de cette revue systématique était de déterminer les effets du traitement combiné à dose fixe sur la mortalité toutes causes confondues, les effets mortels et non mortels des MCV, les effets indésirables, la pression artérielle, les lipides, les taux d'abandon, la qualité de vie et les coûts pour la prévention des MCV.

Nous avons effectué des recherches dans le registre Cochrane des essais contrôlés (CENTRAL), MEDLINE et EMBASE jusqu'en 2013. Nous avons trouvé neuf essais contrôlés randomisés comparant un traitement combiné à dose fixe incluant deux à cinq médicaments par rapport à un placebo, une composante médicamenteuse unique ou aux soins habituels chez 7 047 patients, datant de 2009 à 2013. Les essais étaient généralement de courte durée, allant de 6 semaines à 15 mois et comprenaient des adultes d'âge moyen, avec et sans antécédents de MCV.

Par rapport à un placebo, à une composante médicamenteuse unique ou aux soins habituels, les effets du traitement combiné à dose fixe sur la mortalité toutes causes confondues ou sur les effets des MCV étaient incertains. Cependant, les taux des effets pour ces critères de jugement étaient rarement rapportés, seulement deux des neuf essais rendaient compte de ces critères de jugement. Ces essais étaient principalement conçus pour observer des changements sur les niveaux de facteur de risque cardio-vasculaire plutôt que sur les effets cliniques. De plus, les essais présentaient un risque élevé de biais dans au moins un domaine, ce qui suggère que ces résultats devraient être interprétés avec prudence. Sur 1 000 patients traités avec un traitement combiné à dose fixe pendant la période d'étude, 297 (plage de 264 à 315) souffraient d'un effet secondaire par rapport à 242 patients traités sous placebo. Le traitement combiné à dose fixe était associé à une plus faible pression artérielle systolique (de -7,05 mmHg, allant de -10,18 à -3,87) et à un plus faible taux de cholestérol total (-0,75 mmol / l, allant de -1,05 à -0,46). Cependant, ces comparaisons comprenaient un degré élevé d'hétérogénéité statistique, de sorte que ces résultats doivent être interprétés avec prudence. Sur 1 000 patients traités avec un traitement combiné à dose fixe pendant la période d'étude, 140 (plage de 122 à 186) ont arrêté le traitement par rapport à 115 patients traités sous placebo. Les effets sur la qualité de vie n'étaient pas clairs et aucune donnée sur les coûts n'a été rapportée. Les essais en cours du traitement combiné à dose fixe devront probablement rendre compte de ces importants critères de jugement.

Notes de traduction

Traduit par: French Cochrane Centre 6th August, 2014
Traduction financée par: Financeurs pour le Canada : Instituts de Recherche en Santé du Canada, Ministère de la Santé et des Services Sociaux du Québec, Fonds de recherche du Québec-Santé et Institut National d'Excellence en Santé et en Services Sociaux; pour la France : Ministère en charge de la Santé

Laienverständliche Zusammenfassung

Kombinationstherapie mit festgelegter Dosierung zur Vorbeugung von kardiovaskulären Erkrankungen

Kardiovaskuläre Erkrankungen (CVD), einschließlich Herzinfarkt und Schlaganfall sind weltweit die häufigste Ursache für Tod und Behinderung. Es wurde gezeigt, dass medikamentöse Therapie mit Blutdruck- und Cholesterinsenkenden Mitteln, vor allem mit Statinen, die Wahrscheinlichkeit für ein tödliches oder nicht-tödliches kardiovaskuläres Ereignis verringert. Es wurde auch gezeigt, dass Aspirin Herzinfarkten, bestimmten Schlaganfällen und Tod bei Menschen mit bestehender CVD vorbeugen kann. Das Konzept der Kombinationstherapie mit festgelegter Dosierung bedeutet die Kombination verschiedener Medikamente in eine einzige Tablette, da dies nachweislich die Einhaltung der Therapie von Menschen mit hohem Blutdruck und HIV verbessert. Kürzlich gab es randomisierte kontrollierte klinische Studien, die die Wirkung der Kombinationstherapie mit festgelegter Dosierung zur Vorbeugung von CVD untersuchten. Das Ziel dieses systematischen Reviews war es, die Wirkungen der Kombinationstherapie mit festgelegter Dosierung auf die Gesamtmortalität, die tödlichen und nicht-tödlichen CVD Ereignisse, Nebenwirkungen, Blutdruck, Lipide, Abbruchraten, Lebensqualität und Kosten für eine CVD Prävention zu bestimmen.

Wir durchsuchten das Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE und EMBASE bis 2013. Wir fanden neun randomisierte kontrollierte Studien über Zwei-Komponenten- bis Fünf-Komponenten-Kombinationstherapie mit festgelegter Dosierung verglichen mit Placebo, Medikamente mit einzelnen aktiven Komponenten oder Standardtherapie. Diese Studien schlossen insgesamt 7047 Patienten ein und stammten von 2009 bis 2013. Studien waren in der Regel von kurzer Dauer, reichten von sechs Wochen bis 15 Monate, und schlossen Erwachsene mittleren Alters mit und ohne bestehende CVD ein.

Im Vergleich zu Placebo, Medikamenten mit einer einzelnen aktiven Komponente oder Regelversorgung, waren die Auswirkungen der Kombinationstherapie mit festgelegter Dosierung auf die Gesamtmortalität oder CVD-Ereignisse unsicher. Allerdings waren die Ereignisraten für diese Endpunkte relativ selten, nur zwei von neun Studien berichtet diese Endpunkte. Diese Studien wurden in erster Linie durchgeführt, um Veränderungen im Vorkommen der CVD-Risikofaktoren statt klinischer Ereignisse zu beobachteten, und die Studien hatten ein hohes Risiko für Bias in mindestens einer Domäne, was darauf hindeutet, dass diesen Ergebnissen nicht vollständig vertraut werden sollte. Von 1000 Menschen, die mit Kombinationstherapie mit festgelegter Dosierung während der Studiendauer behandelt wurden, würden 297 (Bereich 264-315) eine Nebenwirkung erfahren, verglichen mit 242 Personen, die mit Placebo behandelt wurden. Die Kombinationstherapie mit festgelegter Dosierung wurde mit einem niedrigeren systolischen Blutdruck (-7,05 mmHg, Bereich-10,18 bis -3,87) und Gesamtcholesterin (-0,75 mmol/L, Bereich -1,05 bis -0,46) assoziiert. Jedoch gab es einen hohen Grad an statistischer Heterogenität bei diesen Vergleichen, so dass diese Ergebnisse mit Vorsicht betrachtet werden sollten. Von 1000 Patienten, die während der Studiendauer mit Kombinationstherapie mit festgelegter Dosierung behandelt wurden, brachen 140 (Bereich 122 bis 186) die Therapie ab. Das gleiche gilt für 115 Patienten der Placebogruppe. Die Wirkungen auf Lebensqualität waren unklar und Daten zu Kosten wurden nicht berichtet. Laufende Studien zu Kombinationstherapie mit festgelegter Dosierung werden wahrscheinlich über diese wichtigen Endpunkte berichten.

Anmerkungen zur Übersetzung

K. Kunzweiler und I. Töws, Koordination durch Cochrane Schweiz.

Summary of findings(Explanation)

Summary of findings for the main comparison. 
Fixed-dose combination therapy for the prevention of cardiovascular disease (CVD)

Patient or population: Adults older than 18 years, with no restriction regarding presence of CVD; participants generally had elevated risk of CVD (as estimated by the presence of at least one abnormal cardiovascular risk factor) without prevalent CVD (two studies included > 10% of participants with prior CVD)

Settings: Outpatient

Intervention: Fixed-dose combination therapy of varying drug combinations ranging from two to five drugs

Comparison: Usual care, placebo, or single drug therapy from alternate drug class

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of Participants
(studies)
Quality of the evidence
(GRADE)
Comments
Assumed riskCorresponding risk
Standard practice or placeboFixed-dose combination therapy
All-cause mortality Total

RR = 1.26

[0.67, 2.38]

3465

(2 studies)

⊕⊕⊝⊝
low
Downgraded due to study limitations (risk of bias) in 1 of 2 included studies and imprecision of effect.
10 per 1000

12 per 1000

(7 to 24)

CVD event TotalRR = 1.38 [0.91, 2.10]

2479

(2 studies)

⊕⊕⊝⊝
low
Downgraded due to study limitations (risk of bias) in 1 of 2 included studies and imprecision of effect.
29 per 1000

40 per 1000

(26 to 61)

Any adverse event

6 weeks to 15 months

Total

RR = 1.19

[1.09, 1.30]

4864

(7 studies)

⊕⊕⊝⊝
low
Downgraded due to study limitations (risk of bias) and difficulty in assessing indirectness of evidence.
242 per 1000

297 per 1000

(264 to 315)

Systolic blood pressure

mmHg

The mean change in systolic blood pressure ranged across control groups from -17.9 mmHg to -2 mmHg.The mean change in systolic blood pressure in the intervention groups was on average a -7.05 mmHg (95% CI -10.18 to -3.87) greater reduction compared with control. 

5787

(9 studies)

⊕⊕⊕⊝
moderate
Downgraded due to study limitations (risk of bias) and unexplained heterogeneity.

Diastolic blood pressure

mmHg

The mean change in diastolic blood pressure ranged across control groups from -9.8 mmHg to -0.5 mmHg.The mean change in diastolic blood pressure in the intervention groups was on average a -3.65 mmHg (95% CI -5.44 to -1.85) greater reduction compared with control. 

5787

(9 studies)

⊕⊕⊕⊝
moderate
Downgraded due to study limitations (risk of bias) and unexplained heterogeneity.

Total cholesterol

mmol/L

The mean change in total cholesterol ranged across control groups from -1.6 mmol/L to 0 mmol/L.The mean change in total cholesterol in the intervention groups was on average a -0.75 mmol/L (-1.05 to -0.46) greater reduction compared with control. 

5569

(9 studies)

⊕⊕⊝⊝
low
Downgraded due to study limitations (risk of bias), unexplained heterogeneity, and funnel plot asymmetry.

LDL cholesterol

mmol/L

The mean change in LDL cholesterol ranged across control groups ranged from
-1.4 mmol/L to -0.04 mmol/L.
The mean change in LDL cholesterol in the intervention groups was on average a
-0.81 mmol/L (95% CI -1.09 to -0.53) greater reduction compared with control.
 

5365

(8 studies)

⊕⊕⊕⊝
moderate
Downgraded due to study limitations (risk of bias) and unexplained heterogeneity.

Discontinuation for any reason

6 weeks to 15 months

TotalRR = 1.26 [1.02, 1.55]

2423

(6 studies)

⊕⊕⊝⊝
low
Downgraded due to study limitations (risk of bias) and difficulty in assessing indirectness of evidence.
120 per 1000 140 per 1000
(122 to 186)
*The basis for the assumed risk (e.g. the median control group risk across studies) is the outcomes of the study control arms. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CVD = cardiovascular disease;
CI: confidence interval; RR: risk ratio.
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Description of the condition

Cardiovascular disease (CVD) is a principal cause of death worldwide. In 2010, more than 17 million deaths globally were attributed to CVD, over 80% of which occurred in low and middle-income countries (WHO 2010 (a)). Furthermore, the situation is not expected to improve, with global CVD mortality estimated to increase by six million over the next 20 years (WHO 2010 (a)). Ischaemic heart disease and cerebrovascular diseases, the major CVDs, are also major causes of disability resulting in 130 million disability-adjusted life years (DALYs) lost in 2010 (WHO 2011). Therefore, preventing deaths and disease due to CVD is a priority for global public health.

Optimising modifiable risk factors reduces CVD mortality and morbidity (Cowie 2005). Individuals with both hypertension and dyslipidaemia have a greater risk of CVD than those with either hypertension or dyslipidaemia alone (Neaton 1992; Thomas 2002), highlighting the importance of considering overall CVD risk as opposed to individual risk factors (Perk 2012). Therefore, adopting a multi-factorial approach to CVD risk management, where multiple risk factors are modified simultaneously, is a more effective way of reducing CVD events than focusing on single risk factors in isolation (Gaede 2003).

Current national and international approaches to CVD prevention incorporate both primary and secondary prevention (Perk 2012; NICE 2010). Primary prevention aims to prevent CVD events in those who have no clinical evidence of CVD. To achieve this, guidelines recommend intervening when the 10 year risk for any CVD event when the 10 year risk exceeds recommended thresholds or when the risk of a fatal CVD event is estimated to be at 5% using validated risk scores (NICE 2008; NICE 2010; Perk 2012; Stone 2013). CVD incidence and mortality are reduced by antihypertensives (Collins 1990) and statins, which improve the lipid profile (Taylor 2013). Secondary prevention requires blood pressure control, cholesterol lowering, and use of antiplatelet drugs to prevent further CVD events, which is known to be effective (ATT-Collaboration 2002; Baigent 2005; Rashid 2003).

The same CVD risk factors operate globally (Yusuf 2004) making multi-factorial prevention strategies relevant, but conventional approaches targeting high risk individuals, conducting investigations, prescribing various medications, regular monitoring, and drug dose titration to optimise CVD risk factors are difficult to implement. Three major issues arise for global CVD prevention.

(i) Reducing risk factors in a selected high risk group does not yield as much benefit to a population as reducing risk factors in the whole population (Cooney 2009; NICE 2010).

(ii) Lipid-lowering with statins reduces CVD events at pre-treatment lipid levels that are considered normal (Colhoun 2004; HPSCG 2002; O'Keefe 2004; Sever 2003) making blood testing for lipid levels less relevant and potentially increasing the number of people who would benefit from statins.

(iii) Implementing conventional CVD prevention would be a challenge for the healthcare systems in most low and middle-income countries due to financial and time costs, human resource availability, laboratory capacity, drug acquisition, and adminstration.

Therefore, alternative and complementary population-wide strategies are required.

Description of the intervention

A fixed-dose combination pill was proposed in 2001 by a World Health Organization (WHO) and Wellcome Trust expert group (WHO 2002) and was subsequently specified as a combination of four drugs (beta-blocker, angiotensin converting enzyme (ACE)-inhibitor, aspirin, and statin), which was estimated to reduce CVD events by 75% in people with clinical evidence of CVD (Yusuf 2002). This concept was followed in 2003 by a proposed Polypill® (a combination of folic acid, aspirin, three low-dose antihypertensives, and a low-dose statin), which was intended for both secondary prevention and primary prevention in all people aged 55 years and over and was estimated to reduce CVD events by about 80% (Wald 2003). Recent evidence has indicated that the effects of fixed-dose combination treatment may be less than was initially proposed, but that this strategy may improve the blood pressure and lipid profile to near expected levels (PILL-collaborative 2011; TIPS 2009). The controversial aspect of the Polypill® was that it was intended to be used at a population level without screening of blood cholesterol or blood pressure (Wald 2011) because an age threshold of 55 years and above would be used to determine eligibility for treatment (Lonn 2010; Wald 2003).

While aspirin is indicated for secondary prevention of CVD, the use of aspirin for primary prevention of CVD is generally indicated when the absolute risk of cardiovascular disease outweighs the risk of severe bleeding (Baigent 2009). Also, doubt exists regarding folic acid since recent large randomised trials have indicated no CVD benefit (Armitage 2010; Holmes 2011). On the other hand, statins and antihypertensives as single treatments are known to be relatively safe and individually beneficial in terms of reducing CVD risk and thereby cardiovascular events for both secondary prevention and primary prevention (ALLHAT-investigators 2002; Colhoun 2004; CTT 2012; HPSCG 2002; Julius 2004; Kearney 2008; LaRosa 2005; Ostergren 2008; Papademetriou 2003; Sever 2003; Taylor 2013; Turnbull 2003). Therefore, although uncertainty exists regarding possible components, the consensus is that the ideal fixed-dose combination therapy for primary and secondary CVD prevention should include at least one antihypertensive and one statin.

There is some recent evidence regarding the efficacy and safety of antihypertensives and statins when administered concomitantly (Messerli 2006; Preston 2007), and of multiple antihypertensives when administered as a single tablet (Gupta 2010; Bangalore 2007). Clinicians may be wary of combination therapy due to the potential restrictions on individualised management (Viera 2011); that is, the ability to amend standard therapy because of medical history or adverse events, such as avoiding a beta-blocker in an asthmatic or changing from an ACE-inhibitor due to cough, and because of the inability to titrate each drug prescribed according to clinical response (Lonn 2010). It is also unclear if there are unique adverse events associated with fixed-dose combination therapy beyond the individual components.

How the intervention might work

The effectiveness of the drugs comprising a fixed-dose combination are generally well understood, and the principles behind using pharmacotherapy at a population level are that the drugs themselves are inexpensive, simple to administer for easier clinical decision making, might not require a medically trained practitioner, and may provide a more effective option that the promotion of lifestyle changes for multiple risk factor control. Yet convincing evidence of the benefits of such interventions has not been achieved. (Beaglehole 2011; Ebrahim 2011; Lonn 2010). Although modifying national health policy has been successful in some high-income countries, such as in Scandinavia (Vartiainen 2010), population-level pharmacotherapy can be politically challenging in both high and low to middle-income countries (Lonn 2010; Yusuf 2011) and may not meet with patient approval. However, patient adherence to the fixed-dose combination therapy is expected to be better than with multiple tablets, but it has been argued that they will likely have a greater potential for adverse effects than behavioural or lifestyle changes and that a purely biological approach is too narrow to allow the social, economic, and behavioural complexities of CVD prevention to be appreciated and confronted (Franco 2004).

Recent global epidemiological data from the Prospective Urban Rural Epidemiology (PURE) Study investigators indicate that the overall use of secondary prevention medication was less than 30% and that levels of use are particularly poor in low and middle-income countries and in rural regions (Yusuf 2011). The likely result is inadequate prevention of further CVD events. Prescribing fixed-dose combination therapy to individuals who are above an accepted absolute risk threshold for initiation of pharmacotherapy for primary CVD prevention may help to resolve these challenges.

However, fixed-dose combination therapy still has many unknowns. These include (i) the best constituents, whether two or three or four or five drugs are required; and (ii) evidence of safety, effectiveness, and cost-effectiveness, and whether increasing the number of constituents will produce a favourable risk-benefit profile and be worth the increased cost. In particular, the evidence is sparse concerning benefits and risks of fixed-dose combination therapy for primary prevention in those people with low CVD risk. Several authors have questioned whether a fixed-dose combination strategy may have unforeseen negative effects on other aspects of CVD risk reduction, for example, individuals neglecting to exercise because of a sense of CVD security with fixed-dose combination therapy (Lonn 2010). As yet there are limited long-term follow-up outcome and safety data, which is of particular importance beacuse the Polypill® concept was designed with long-term use of fixed-dose combination therapy in mind.

Why it is important to do this review

Various fixed-dose combination pills are now being manufactured, and there is evidence that physicians are aware of this option and are potentially willing to prescribe it, though perhaps not without some reservations (Viera 2011). There is an emerging literature of randomised controlled trials comparing fixed-dose combination therapy with placebo or standard practice in both the primary and secondary prevention of CVD, as well as in assessing safety and tolerability (Elley 2012). Since the publication of this review (Elley 2012), additional fixed-dose combination trial data have been published.

Objectives

To determine the effectiveness of fixed-dose combination therapy on reducing fatal and non-fatal CVD events and on improving CVD risk factors for both primary and secondary prevention of CVD. We also aimed to determine discontinuation rates, adverse events, health-related quality of life, and costs of fixed-dose combination therapy.

Methods

Criteria for considering studies for this review

Types of studies

Randomised controlled trials (RCT).

Types of participants

Adults 18 years and older with no restriction regarding presence of CVD.

Types of interventions

A fixed-dose combination therapy, a combination of several active components into a single pill with the aim being to optimise CVD risk and reduce CVD fatal and non-fatal events. At least one statin and one antihypertensive agent should be included. We examined different combinations and doses in stratified analyses, where possible.

Trials were considered where the comparison group was usual care, placebo, or a single drug comparator.

Types of outcome measures

Primary outcomes
  • Clinical outcomes including mortality (cardiovascular and all-cause); non-fatal CVD endpoints such as myocardial infarction, coronary artery bypass grafting (CABG), percutaneous transluminal coronary angioplasty (PTCA), angina or angiographically defined ischaemic heart disease, stroke, transient ischaemic attack (TIA), carotid endarterectomy, or peripheral arterial disease (PAD).

  • Adverse events including overall rates of discontinuation, proportion of participants experiencing specific symptoms or results and rates of discontinuation by specific symptoms. These included but were not limited to: myalgias, cough, elevated liver enzymes, gastric irritation or dyspepsia.

Secondary outcomes
  • Systolic and diastolic blood pressure

  • Total and LDL cholesterol

  • Adherence

  • Health-related quality of life, measured according to any well validated and adjusted scale concerning quality of life

  • Costs of fixed-dose combination therapy

Search methods for identification of studies

Electronic searches

The following electronic databases were searched:

  • Cochrane Central Register of Controlled Trials (CENTRAL, Issue 6, 2013) on The Cochrane Library;

  • MEDLINE (Ovid) (1946 to week 2 July 2013);

  • EMBASE (Ovid) (1980 to Week 28 2013);

  • ISI Web of Science (1970 to 19 July 2013);

  • Database of Abstracts of Reviews of Effects (DARE), Health Technology Assessment Database (HTA), and Health Economics Evaluations Database (HEED) in The Cochrane Library (2011, Issue 4).

The searches were limited to records published since 2000. The fixed-dose combination therapy was conceptualised in 2001, so relevant trials will only appear after this date. The searches were initially run in January 2012 (Appendix 1) and updated in July 2013 (Appendix 2). The latest searches utilised limits to core clinical journals in MEDLINE and priority journals in EMBASE. The Cochrane sensitive-maximising RCT filter (Lefebvre 2011) was used for MEDLINE and adaptations of it were used for EMBASE and Web of Science.

Searching other resources

We searched the metaRegister of controlled trials (mRCT) (www.controlled-trials.com/mrct), clinicaltrials.gov (www.clinicaltrials.gov), and the World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (http://apps.who.int/trialsearch/) for ongoing trials on 11 July 2011. This search was updated on 24 December 2011 to review existing ongoing studies that had been identified and identify any recent registrations. In addition, reference lists of reviews and retrieved articles were checked for additional studies and citation searches performed on key articles. Experts in the field were contacted for unpublished and ongoing trials. Authors were contacted where necessary for additional information.

Data collection and analysis

Selection of studies

From the searches, the title and abstract of each paper were reviewed by three authors (AdeC, MF, NW) and potentially relevant references retrieved. Following this initial screening, the full text reports of potentially relevant studies were obtained, and three authors (AdeC, MF, NW) independently selected studies to be included in the review using predetermined inclusion criteria. The rapid review search was completed by one author (MH). In all cases disagreements about any study inclusions were resolved by consensus, and a fourth author (KR) was consulted if disagreement persisted.

Data extraction and management

Data were extracted independently by two authors (AdeC, MF) using a proforma, and principal investigators were contacted to provide additional relevant information where necessary. Data extraction from the rapid review was performed by one author (MH). Details of the study design, participant characteristics, study setting, intervention and comparator, and outcome data including details of outcome assessment, adverse effects, and methodological quality (randomisation, blinding, attrition) were extracted from each of the included studies. Disagreements about extracted data were resolved by consensus, and a third author was consulted if disagreement persisted (KR).

Assessment of risk of bias in included studies

Risk of bias was assessed according to the Cochrane risk of bias assessment tool, including examining the quality of the random sequence generation and allocation concealment, description of dropouts and withdrawals (including intention-to-treat analysis), blinding (participants, personnel, and outcome assessment), and selective outcome reporting (Higgins 2011). For cluster randomised trials, we have followed the Cochrane Handbook for Systematic Reviews of Interventions recommendations for assessing risk of bias, with particular attention across the domains of: recruitment; baseline imbalances; loss of clusters; incorrect analyses; and comparability with individually randomised trials (Higgins 2011). The risk of bias in the included studies was assessed independently by three authors (AdeC, MF, MH).

Measures of treatment effect

Data were processed in accordance with the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Dichotomous outcomes were expressed as relative risks, and 95% confidence intervals (CI) were calculated for each study. For continuous variables, net changes were compared (that is intervention group minus control group differences) and a weighted mean difference (MD) and 95% CI were calculated for each study. For TIPS 2009, we compared the effects of fixed-dose combination therapy on mean (SD) levels of blood pressure and cholesterol against the study arms without active components as reported by the study authors. Where SDs were not reported in the outcomes of interest (TIPS 2009), we used baseline SDs per Elley 2012 and Furukawa 2006.

Assessment of heterogeneity

For each outcome, tests of heterogeneity were carried out using the Chi2 test of heterogeneity and the I2 statistic. Where no or minimal heterogeneity was present, we performed fixed-effect model meta-analyses. Where substantial heterogeneity was detected (I2 > 50%), we evaluated the results for possible explanations (for example participants and interventions) and performed random-effect model meta-analyses with cautious intepretation.

Subgroup analysis and investigation of heterogeneity

If there were sufficient studies, we aimed to conduct the following subgroup analyses.

  • Age.

  • Sex.

  • Primary prevention (populations where 10% or less had pre-existing CVD) versus secondary prevention (population where > 10% had pre-existing CVD).

  • Two-drug versus three-drug or more fixed-dose combination therapies.

  • Comparator group as usual care versus placebo or inactive control.

The first four of these analyses were pre-specified in our protocol, and the last subgroup analysis was performed post hoc. Data were available to perform subgroup analyses on the latter three analyses.

Sensitivity analysis

Sensitivity analyses were performed by excluding studies at high risk of bias. We created funnel plots and performed tests of asymmetry (Egger 1997) according to the available outcomes of systolic blood pressure and total cholesterol to assess possible publication bias through funnel plot asymmetry.

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies; Characteristics of ongoing studies.

Results of the search

The searches in 2012 generated 9731 hits and 5067 papers after de-duplication. Screening the titles and abstracts identified 41 papers for formal inclusion or exclusion. Of these, eight RCTs (eight papers, three abstracts) met the inclusion criteria. We identified five ongoing trials (FOCUS 2011; IMPACT 2011; Kanyini-GAP 2010; Merat 2010; PolyIran 2010). The search in 2013 generated 287 hits and 278 papers after de-duplication. Screening titles and abstracts identified 16 papers for formal inclusion or exclusion. None of these studies met the inclusion criteria. We further identified one paper published after the latest search (UMPIRE 2013) through communication with the study authors. The study flow diagram is presented in Figure 1. One study was a cluster randomised trial (CRUCIAL 2011), one study was a randomised, cross-over design clinical trial (Wald 2012), and the remaining seven were individual-level randomised trials.

Figure 1.

Study flow.

Included studies

Details of the methods, participants, intervention, comparison group and outcome measures for each of the studies included in the review are shown in the Characteristics of included studies table. Nine trials were included with 7047 participants randomised. The three largest trials (CRUCIAL 2011; TIPS 2009; UMPIRE 2013) randomized 5518 (78%) of all participants. The duration of the intervention and follow-up periods was generally short-term (six weeks in one study (TOGETHER 2010), eight weeks in one study (CUSP 2009), 12 weeks in four studies (PILL 2011; Soliman 2009; TIPS 2009; Wald 2012)), but three studies had median follow-up periods of 12 to 15 months (CRUCIAL 2011; Malekzadeh 2010; UMPIRE 2013). All trials reported changes in blood pressure and cholesterol, whereas mortality was only reported in two trials (CRUCIAL 2011; UMPIRE 2013). Three trials (CRUCIAL 2011; Soliman 2009; UMPIRE 2013) compared fixed-dose combination therapy against usual care, whereas the other six trials compared combination therapy against either active control or placebo. One trial (TIPS 2009) included nine arms with different drug combinations, which led to restricting our analyses to comparisons between fixed-dose combination therapy and groups without either blood pressure or cholesterol lowering drugs (depending upon the analysis) and lowered the sample sizes in these analyses.

The included studies frequently had complex inclusion and exclusion criteria that were generally based upon freedom from prior cardiovascular disease, an age threshold ranging from > 21 years to > 55 years in women, a composite measure of short-term (10 year) risk (five year predicted Framingham CVD risk ≥ 7.5% in PILL 2011), or one to three elevated cardiovascular disease risk factors. UMPIRE 2013 specifically enrolled participants with established CVD or a five year risk of CVD ≥ 15%, while CRUCIAL 2011 included > 18% of particpants with peripheral artery disease (PAD) and > 14% with prior transient ischaemic attack (TIA) or stroke. The participants were generally middle-aged with a mean (SD) age ranging from 52.6 (9.6) years (CUSP 2009) to 62.1 (10.4) years (UMPIRE 2013). The majority of trials enrolled predominantly men with two trial randomising more then 80% men (PILL 2011; UMPIRE 2013) compared with one trial that enrolled only 27% men (Soliman 2009). Baseline systolic blood pressure ranged from 125 mmHg to 166 mmHg, and baseline total cholesterol ranged from 4.2 to 6.1 mmol/L.

The drugs included in the various fixed-dose combination pills varied (Table 1) with three studies including two drugs (CRUCIAL 2011; CUSP 2009; TOGETHER 2010), five studies including four drugs (PILL 2011; Soliman 2009; Malekzadeh 2010, Wald 2012; UMPIRE 2013), and one study including five drugs (TIPS 2009). Aspirin was included in five studies (Malekzadeh 2010; PILL 2011; Soliman 2009; TIPS 2009; UMPIRE 2013), and blood pressure and cholesterol lowering drugs were included, by definition, in all nine studies. The blood pressure components included either a calcium channel blocker, thiazide diuretic, beta-blocker, ACE-inhibitor, or angiotensin receptor blocker (ARB), or a combination thereof. In terms of lipid lowering drugs, simvastatin was used in five trials (PILL 2011; Soliman 2009; TIPS 2009; Wald 2012; UMPIRE 2013), and atorvastatin was used in four trials (CRUCIAL 2011; CUSP 2009; Malekzadeh 2010; TOGETHER 2010).

Table 1. Polypill content by trial
  1. 1 Site investigators could request dosages of amlodipine and atorvastatin 5/20 mg and 10/20 mg

StudyPolypill contents (dose)Comparator
CRUCIAL 2011

Amlodipine 5 to 10 mg

Atorvastatin 10mg1

Usual care
CUSP 2009

Amlodipine 5 mg

Atorvastatin 20 mg

Placebo
Malekzadeh 2010

Aspirin 81 mg

Atorvastatin 20 mg

Enalapril 2.5 mg

Hydrochlorothiazide 12.5 mg

Placebo
PILL 2011

Aspirin 75 mg

Hydrochlorothiazide 12.5 mg

Lisinopril 10 mg

Simvastatin 20 mg

Placebo
Soliman 2009

Aspirin 75 mg

Hydrochlorothiazide 12.5 mg

Lisinopril 10 mg

Simvastatin 20 mg

Usual care
TIPS 2009

Aspirin 100 mg

Atenolol 50 mg

Hydrochlorothiazide 12.5 mg

Ramipril 5 mg

Simvastatin 20 mg

8 other drug/drug combination groups:

1) Aspirin 100mg

2) Aspirin 100mg, hydrochlorothiazide 12.5mg, atenolol 50mg, ramipril 5mg

3) Hydrochlorothiazide 12.5mg

4) Hydrochlorothiazide 12.5mg, atenolol 50mg

5) Hydrochlorothiazide 12.5mg, ramipril 5mg

6) Hydrochlorothiazide 12.5mg, atenolol 50mg, ramipril 5mg

7) Ramipril 5mg, atenolol 50mg

8) Simvastatin 20mg

TOGETHER 2010

Amlodipine 5 to 10 mg

Atorvastatin 10mg

Amlodipine 5 to 10 mg
UMPIRE 2013

Aspirin 75mg

Atenolol 50mg

Lisinopril 40mg

Simvastatin 40mg

or

Aspirin 75mg

Hydrochlorothiazide 12.5mg

Lisinopril 40mg

Simvastatin 40mg

Usual care
Wald 2012

Amlodipine 2.5 mg

Hydrochlorothiazide 12.5 mg

Losartan 25 mg

Simvastatin 40mg

Placebo

Excluded studies

Details and reasons for exclusion for the studies that most closely missed the inclusion criteria are presented in the Characteristics of excluded studies table. The majority of excluded studies were not RCTs.

Risk of bias in included studies

Details are provided for each of the included studies in the risk of bias tables in Characteristics of included studies and in Figure 2 and Figure 3.

Figure 2.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Figure 3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Allocation

The methods of random sequence generation and allocation concealment were unclear in three of the included studies (CRUCIAL 2011; CUSP 2009; Soliman 2009). In the six studies where randomisation and allocation concealment were clear, the methods used were judged to have a low risk of bias (Malekzadeh 2010; PILL 2011; TIPS 2009; TOGETHER 2010; Wald 2012; UMPIRE 2013).

Blinding

Three of the nine included studies had a high risk for performance bias because the comparator group was usual care (CRUCIAL 2011; Soliman 2009; UMPIRE 2013). However, one of these studies included blinded outcome assessment (UMPIRE 2013) and had low risk of detection bias. The remaining six trials stated that they were double blinded (participants and study personnel, including outcome assessors, were blinded to treatment allocation) and were regarded as having low risk of bias in this domain.

Incomplete outcome data

Most studies reported losses to follow-up but there were differences in the proportion of losses to follow-up between the interevention and control arms. Two studies had a high risk of attrition bias (CRUCIAL 2011; TOGETHER 2010), including use of last observation carried forward for missing continuous variables. Four studies had an unclear risk of attrition bias (CUSP 2009; Malekzadeh 2010; Soliman 2009; TIPS 2009), and three studies had low risk of attrition bias (Malekzadeh 2010; Wald 2012; UMPIRE 2013).

Selective reporting

The risk of bias associated with selective reporting was low in five studies (CUSP 2009; Malekzadeh 2010; Soliman 2009; TIPS 2009; UMPIRE 2013), unclear in three studies (CRUCIAL 2011; TOGETHER 2010; Wald 2012), and high in one study (PILL 2011).

Other potential sources of bias

In CRUCIAL 2011, different doses of fixed-dose combination therapy were used among participants randomised to the intervention arm, which was associated with an uncertain risk of bias because the option for drug titration could attenuate the effect size if investigators did not titrate the dose of fixed-dose combination therapy; conversely, the differential dosing could accentuate the effect size because of higher drug doses. In Malekzadeh 2010, a run-in period was used to exclude potential participants who had adherence rates < 70%. In Soliman 2009, participants had varying degrees of background blood pressure and lipid lowering therapies between groups. In other cases there was insufficient information to judge the risk of bias in other sources of bias not covered above, and all were categorised as unclear. In UMPIRE 2013, participants randomised to the intervention arm received fixed-dose combination therapy at no cost compared with participants randomised to usual care who were responsible for their drug costs, which may have led to increased adherence in the intervention arm.

Effects of interventions

See: Summary of findings for the main comparison

Primary outcomes

All-cause mortality

Two studies (CRUCIAL 2011; UMPIRE 2013) reported death rates at the end of the study period with follow-up at 12 and 15 months, respectively. Mortality rates were low in both groups (1.2% in the intervention group compared with 1.0% in the comparator group), and participants randomised to the intervention had no evidence of increased mortality compared with the comparator group (RR 1.26, 95% CI 0.67 to 2.38) (Analysis 1.1) in the context of relatively few events. Both studies included > 10% of participants with prevalent CVD and both studies included usual care as the comparator group, so no subgroup analyses could be performed in these domains. The results were similar when restricting this analysis to UMPIRE 2013, which included a fixed-dose combination intervention with four components (compared with two in CRUCIAL 2011) (data not shown).

Major CVD events

Only two out of nine studies (Malekzadeh 2010; UMPIRE 2013) reported rates of cardiovascular events. Cardiovascular events were uncommon in both groups (4.0% rate in the intervention group compared with 2.9% in the comparator group), and participants randomised to the intervention had no evidence of increased event rates compared with the comparator group (RR 1.38, 95% CI 0.91 to 2.10) (Analysis 1.2). However, these results were imprecise, and there was only one event reported in both arms of Malekzadeh 2010. Participants in Malekzadeh 2010 did not have prevalent CVD, and the comparator group received placebo, compared with the participants in UMPIRE 2013. Both trials included four-drug fixed-dose combination therapy.

Adverse events

Seven trials including 4864 participants reporting aggregated rates of adverse events in both groups were included in the meta-analysis. The risk for adverse events was higher in participants in the intervention arm compared with participants in the control arm (30% versus 24%, RR 1.19, 95% CI 1.09 to 1.30) (Analysis 1.3). Specific side effects that were evaluated included myalgias (five studies, 12% versus 11%, RR 1.14, 95% CI 0.81 to 1.60), increased liver enzymes (three studies, 8% versus 7%, RR 1.01, 95% CI 0.72 to 1.43), cough (four studies, 6% versus 3%, RR 2.34, 95% CI 0.77 to 7.08), gastric irritation and dyspepsia (four studies, 3% versus 2%, RR 1.33, 95% CI 0.66 to 2.74), and bleeding (one study, 2% versus 0.5%, RR 4.00, 95% CI 0.45 to 35.46). Results were similar after excluding trials with >10% of participants with prevalent CVD or usual care as the comparator group (CRUCIAL 2011; UMPIRE 2013 for both) (RR 1.36, 95% CI 1.16 to 1.60) and after excluding trials with less than three drug combinations (CRUCIAL 2011; CUSP 2009; TOGETHER 2010) (RR 1.31, 95% CI 1.13 to 1.51). Rates of discontinuation were reported in both groups in the six trials with active control or placebo as the comparator and were higher in participants randomised to fixed-dose combination therapy (14% versus 11.5%, RR 1.26, 95% CI 1.02 to 1.55) (Analysis 4.1).

Secondary outcomes

Blood pressure

All nine trials reported changes in systolic and diastolic blood pressure in 5787 participants. There was a large degree of heterogeneity among the trials for both systolic blood pressure (I2 = 92%) and diastolic blood pressure (I2 = 91%). No single trial explained this heterogeneity, nor was it explained by primary versus secondary prevention trials nor two-drug versus three or more drug combinations. Using a random-effects model, the WMD in systolic blood pressure between the intervention and control arms was -7.02 mmHg (95% CI -10.18 to -3.87) (Analysis 2.1), and the WMD in diastolic blood pressure between the intervention and control arms was -3.65 mmHg (95% CI -5.44 to -1.85) (Analysis 2.2). Trials that included usual care in the comparator group (CRUCIAL 2011; CUSP 2009; UMPIRE 2013) did not have as large reductions in systolic blood pressure (MD -4.76 mmHg, 95% CI -11.24 to -1.71) compared with other trials (Analysis 2.5). These results should be interpreted with caution given the degree of heterogeneity. There was no evidence of funnel plot asymmetry for systolic blood pressure.

Lipids

All nine trials reported changes in total cholesterol in 5569 participants, and eight trials reported changes in LDL cholesterol in 5365 participants. There was a large degree of heterogeneity among the trials for both total cholesterol (I2 = 97%) and LDL cholesterol (I2 = 97%). No single trial explained this heterogeneity, nor was it explained by primary versus secondary prevention trials, nor two-drug versus three or more drug combinations. Using a random effects model, the weighted mean difference (WMD) in total cholesterol between the intervention and control arm was -0.75 mmol/L (95% CI -1.05 to -0.46) (Analysis 3.1). Using a random-effects model, WMD in LDL cholesterol between the intervention and control arms was -0.81 mmol/L (95% CI -1.09 to -0.53) (Analysis 3.2). Trials that included usual care in the comparator group (CRUCIAL 2011; CUSP 2009; UMPIRE 2013) did not have as large reductions in total cholesterol (MD -0.28 mmol/L, 95% CI -0.66 to 0.10) compared with other trials (Analysis 3.5). These results should be interpreted with caution given the degree of heterogeneity. There was evidence of funnel plot asymmetry for total cholesterol (Figure 4).

Figure 4.

Funnel plot of comparison: 3 Cholesterol, outcome: 3.1 Total cholesterol.

Adherence

In trials with usual care comparisons, assessing adherence or discontinuation was problematic. In fact, only one (UMPIRE 2013) out of three trials that included a usual care arm reported adherence, which was defined as taking aspirin, statin, and two or more blood pressure lowering drugs at least 4 days per week. In UMPIRE, adherence at 15 months was 86% in the intervention group compared with 65% in the comparator group (RR = 1.33 [95% CI: 1.26, 1.41]). However, the discontinuation rate among individuals randomized to fixed-dose combination was 22%.

Health-related quality of life

One trial (UMPIRE 2013) reported health-related quality of life measures at the end of the study period using the EQ-5D instrument. Mean (SD) summary index scores were similar between the intervention and comparator groups (0.82 (0.01) versus 0.81 (0.1), P = 0.43).

Costs

No studies have reported costs or cost-effectiveness associated with fixed-dose combination therapy to date.

Discussion

Summary of main results

The trials included in this systematic review demonstrated no differences in mortality and cardiovascular events between participants randomised to the fixed-dose combination group compared with comparator groups. However, the event rates for these outcomes were low so they were relatively uncommon, only two out of nine trials reported these outcomes, and these trials included at least one domain that had a high risk of bias, suggesting that these results should not viewed with confidence. Adverse events were common in both the intervention (30%) and comparator (24%) groups, with participants randomised to fixed-dose combination therapy being 20% (95% CI 9% to 30%) more likely to report an adverse event. Notably, no serious adverse events were reported. The trials reported weighted mean reductions in systolic (-7.02 mmHg, 95% CI -10.18 to -3.87) and diastolic blood pressure (-3.65 mmHg, 95% CI -5.44 to -1.85) and total (-0.75 mmol/L, 95% CI -1.05 to -0.46) and LDL cholesterol (-0.81 mmol/L, 95% CI -1.09 to -0.53). However, there was substantial heterogeneity in these estimates, which should be interpreted with caution. The trials demonstrated a 26% (95% CI 2% to 55%) increased risk of discontinuing the study medication (discontinuation rate range 10% to 23%) compared with either usual care, placebo, or a single drug (aspirin, statin, or thiazide in the case of TIPS 2009). We were unable to explain the heterogeneity of effects on blood pressure or lipids in terms of primary versus secondary prevention trials, the number of drugs in the fixed-dose combination pills, or the comparator group being active control, placebo or usual care. It is possible that the heterogeneity is due to the characteristics of the patients studied, differences in the potency of the antihypertensives and statins used, and the differences in treatments used in the comparison groups.

Overall completeness and applicability of evidence

The included trials used five different polypills: three of the studies (CRUCIAL 2011; CUSP 2009; TOGETHER 2010) included polypills with only two drugs (one blood pressure lowering drug (amlodipine) and one statin (atorvastatin)); three studies (PILL 2011; Soliman 2009; UMPIRE 2013) used the Dr Reddy's Lab Red Heart Pill that includes four drugs (aspirin, lisinopril, simvastatin, and hydrochlorothiazide), and the remaining studies included different four-drug (Malekzadeh 2010; Wald 2012) or five-drug combinations (TIPS 2009). The decision to combine the estimates of these different drug combinations and different comparators was made and meta-analysis for this review was performed to evaluate the estimated effect size of fixed-dose combination therapy. A rationale for fixed-dose combination therapy is that it is more likely to be taken than multiple dose regimens. However, we found a higher likelihood of discontinuation for fixed-dose treatment than for placebo. Comparisons of adherence across trials are hampered by differing definitions, which should be standardised in future reporting of these trials. Trials using 'usual care' comparison groups have reported reasonably high levels of adherence and low levels of discontinuation, but these may be misleading as there is no relevant comparison.

There are six ongoing trials (FOCUS 2011; IMPACT 2011; Kanyini-GAP 2010; Merat 2010; PolyIran 2010; TIPS-3 2012), including three that are part of the Single Pill Against Cardiovascular Events (SPACE) collaboration. These results are likely to have an important impact on our confidence in the estimate of effect and may change the estimate.

Quality of the evidence

Our review was limited by the presence of a moderate to high risk of bias in the domains of selection, performance, detection, attrition, and other types of bias in five of the nine trials that were included, which limits the confidence with which we have in these results. Using other GRADE domains, the quality of evidence was also limited by the imprecision of results for the effects on all-cause and CVD mortality, the heterogeneity of effects on blood pressure and cholesterol, and funnel plot asymmetry suggestive of publication bias (total cholesterol only). Indirectness of evidence was further limited in evaluating the effects of fixed-dose combination on adverse events and discontinuation rates, particularly because these comparisons excluded participants receiving 'usual care' (because these groups cannot 'discontinue' usual care). However, these comparisons are likely very relevant in assessing the overall effect of fixed-dose combination therapy.

Potential biases in the review process

For the TIPS 2009 and Wald 2012 studies, we relied upon the point estimates and standard deviations extracted by Elley 2012, since these data points were not specifically provided in the text of the manuscripts or by the study authors. Elley and colleagues estimated the outcome standard deviations using baseline standard deviations as reported by Furukawa and colleagues (Furukawa 2006).

Agreements and disagreements with other studies or reviews

Our results demonstrated modestly lower reductions in systolic (-7.02 mmHg versus -9.20 mmHg) and diastolic blood pressure (-3.65 mmHg versus -5.00 mmHg) and lower total (-0.75 mmol/L versus -1.22 mmol/L) and LDL cholesterol (-0.81 mmol/L versus -1.02 mmol/L) compared with an earlier systematic review (Elley 2012). The absolute and relative adverse event rates were similar to those reported by Elley 2012, but the absolute and relative discontinuation rates were lower in our review. These differences are accounted for by our inclusion of three additional studies (CRUCIAL 2011; Soliman 2009; UMPIRE 2013).

The changes in blood pressure were lower than those predicted by Wald and Law (diastolic blood pressure: -3.65 mmHg versus -11 mmHg), which may be due to the use of one blood pressure lowering drug in three of the studies (CRUCIAL 2011; CUSP 2009; TOGETHER 2010), all of which used a calcium channel blocker (amlodipine) rather than a combination of thiazide, ACE-inhibitor, or beta-blocker as previously proposed (Wald 2003). In addition, the baseline blood pressure from which Wald and Law were operating was 150/90 mmHg (Wald 2003, Lonn 2010), compared with a range of blood pressures of 125 to 165 mmHg/78 to 91 mmHg, with seven of nine studies having a baseline systolic blood pressure less than 150 mmHg. The changes in LDL cholesterol were also lower than those predicted by Wald and Law (-0.75 mmol/L versus 1.8 mmol/L), likely due to differences in the dose and type of statin used in these trials (Wald 2003). The baseline LDL cholesterol proposed by Wald and Law was 4.8 mmol/L (Lonn 2010), compared with a range of LDL cholesterol from 2.3 to 3.7 mmol/L with all nine studies having a mean baseline LDL cholesterol less than 4.8 mmol/L. Three trials used simvastatin 20 mg (Soliman 2009; PILL 2011; TIPS 2009) compared with simvastatin 40 mg proposed by Wald and Law (Wald 2003).

Bangalore and colleagues have previously performed a systematic review and meta-analysis of the effect of fixed-dose combination therapy on adherence for chronic conditions including hypertension, diabetes, and HIV (Bangalore 2007) and reported a 24% (95% CI 19% to 29%) lower rate of discontinuation compared with control. These results were similar to those reported by Gupta and colleagues, who reported an increased odds of adherence with fixed-dose combination therapy for blood pressure compared with usual care (OR 1.21, 95% CI 1.03 to 1.43) (Gupta 2010). Gupta and colleagues demonstrated trends toward improved blood pressure control and side effects (Gupta 2010). The differences in discontinuation rates and adherence between these studies and our study may be due to the fact that patients in the Bangalore and Gupta meta-analyses received active drug in either arm compared with our meta-analysis where comparator group participants received either usual care (and possibly no drugs), placebo, or alternative drugs with potentially lower rates of side effects (TIPS 2009).

Virdee and colleagues interviewed 11 primary care physicians and five practice nurses in nine Birmingham, UK practices about their knowledge and attitudes toward fixed-dose combination therapy (Virdee 2013). The majority of respondents were uncertain about how they would incorporate fixed-dose combination therapy in their practice and whether it was designed for primary or secondary CVD prevention. Most felt reluctant about using a specific age cut-off to initiate therapy, despite acknowledging potential advantages to this approach. Most respondents felt unease at the concept of minimial or no monitoring of patients taking a fixed-dose combination therapy, despite the proposal by Wald and Law (Wald 2003). In March 2010, Viera and colleagues surveyed US physicians about their willingness to prescribe fixed-dose combination therapy. Nearly two out of every three physicians reported that they would prescribe fixed-dose combination therapy for patients at moderate risk for CVD and more than four out of every five physicians reported that they would prescribe fixed-dose combination therapy for patients at high risk for CVD. These disparate data using different methods of data collection suggest varying potential for uptake among physicians.

Authors' conclusions

Implications for practice

Compared with usual care, active control, or placebo for CVD prevention, the effects of fixed-dose combination therapy on all-cause mortality or CVD events are uncertain due to low event rates, imprecision, and risk of bias. Participants randomised to fixed-dose combination therapy had moderately higher rates of adverse events (RR 1.18, 95% CI 1.09 to 1.30) and discontinuation (RR 1.26, 95% CI 1.02 to 1.55). Fixed-dose combination therapy is associated with a -7.05 mmHg (95% CI -10.18 to -3.87) and -3.65 mmHg (95% CI -5.44 to -1.85) greater reduction in systolic and diastolic blood pressure and a -0.75 mmol/L (95% CI -1.05 to -0.46) and -0.81 mmol/L (95% CI -1.09 to -0.53) greater reduction in total and LDL cholesterol, but there is substantial heterogeneity in these results. The heterogeneity may reflect differences in primary compared with secondary prevention studies, the composition of fixed-dose combinations, comparator groups, or all of the above. Fixed-dose combination therapy improved adherence to a multi-drug strategy by 33% (26% to 41%) compared with usual care, but this comparison was reported in only one study. Fixed-dose combination therapy may be an alternative therapy for risk factor control in patients for CVD prevention but future studies will likely have an important effect on these estimates.

Implications for research

High-quality randomised controlled trials are needed to evaluate if the effect of fixed-dose combination therapies on risk factor levels translates into improvements in fatal and non-fatal events in both primary and secondary CVD prevention settings. Studies evaluating the effects of fixed-dose combination therapy compared with usual multiple variable dose therapies should also be performed to compare adherence rates more directly, since discontinuation rates are generally lower among participants receiving placebos. Larger studies are also needed to evaluate the risk of serious adverse events.

Acknowledgements

We are grateful for the assistance from Dr Curt Furberg who provided additional information on data reported in Soliman 2009, for the assistance from Dr Tom Marshall and colleagues for extra data for Malekzadeh 2010, for the assistance of Henry Lishi Li for translation of one excluded manuscript, and for the assistance of Ms Amy Rogers for her editorial assistance.

Data and analyses

Download statistical data

Comparison 1. Mortality, cardiovascular events, and adverse events
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 All-cause mortality23465Risk Ratio (M-H, Fixed, 95% CI)1.26 [0.67, 2.38]
2 Cardiovascular events22479Risk Ratio (M-H, Fixed, 95% CI)1.38 [0.91, 2.10]
3 Any adverse event74864Risk Ratio (M-H, Fixed, 95% CI)1.19 [1.09, 1.30]
Analysis 1.1.

Comparison 1 Mortality, cardiovascular events, and adverse events, Outcome 1 All-cause mortality.

Analysis 1.2.

Comparison 1 Mortality, cardiovascular events, and adverse events, Outcome 2 Cardiovascular events.

Analysis 1.3.

Comparison 1 Mortality, cardiovascular events, and adverse events, Outcome 3 Any adverse event.

Comparison 2. Blood pressure
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Systolic blood pressure95787Mean Difference (IV, Random, 95% CI)-7.02 [-10.18, -3.87]
2 Diastolic blood pressure95787Mean Difference (IV, Random, 95% CI)-3.65 [-5.44, -1.85]
3 Systolic blood pressure: primary prevention trials72366Mean Difference (IV, Random, 95% CI)-7.45 [-11.05, -3.84]
4 Systolic blood pressure: 3+ drugs only64014Mean Difference (IV, Random, 95% CI)-7.00 [-11.40, -2.60]
5 Systolic blood pressure: comparator as usual care33624Mean Difference (IV, Random, 95% CI)-4.76 [-11.24, 1.71]
Analysis 2.1.

Comparison 2 Blood pressure, Outcome 1 Systolic blood pressure.

Analysis 2.2.

Comparison 2 Blood pressure, Outcome 2 Diastolic blood pressure.

Analysis 2.3.

Comparison 2 Blood pressure, Outcome 3 Systolic blood pressure: primary prevention trials.

Analysis 2.4.

Comparison 2 Blood pressure, Outcome 4 Systolic blood pressure: 3+ drugs only.

Analysis 2.5.

Comparison 2 Blood pressure, Outcome 5 Systolic blood pressure: comparator as usual care.

Comparison 3. Cholesterol
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Total cholesterol95569Mean Difference (IV, Random, 95% CI)-0.75 [-1.05, -0.46]
2 LDL cholesterol85365Mean Difference (IV, Random, 95% CI)-0.81 [-1.09, -0.53]
3 Total cholesterol: primary prevention trials72147Mean Difference (IV, Random, 95% CI)-0.92 [-1.18, -0.65]
4 Total cholesterol: 3+ drugs only63796Mean Difference (IV, Random, 95% CI)-0.65 [-1.10, -0.21]
5 Total cholesterol: comparator as usual care33624Mean Difference (IV, Random, 95% CI)-0.28 [-0.66, 0.10]
Analysis 3.1.

Comparison 3 Cholesterol, Outcome 1 Total cholesterol.

Analysis 3.2.

Comparison 3 Cholesterol, Outcome 2 LDL cholesterol.

Analysis 3.3.

Comparison 3 Cholesterol, Outcome 3 Total cholesterol: primary prevention trials.

Analysis 3.4.

Comparison 3 Cholesterol, Outcome 4 Total cholesterol: 3+ drugs only.

Analysis 3.5.

Comparison 3 Cholesterol, Outcome 5 Total cholesterol: comparator as usual care.

Comparison 4. Adverse events
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Discontinuation of study drug62423Risk Ratio (M-H, Fixed, 95% CI)1.26 [1.02, 1.55]
2 Myalgias53014Risk Ratio (M-H, Fixed, 95% CI)1.14 [0.81, 1.60]
3 Increased liver chemistries31427Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.72, 1.43]
4 Cough42093Risk Ratio (M-H, Random, 95% CI)2.34 [0.77, 7.08]
5 Dyspepsia/gastrointestinal irritation43417Risk Ratio (M-H, Random, 95% CI)1.33 [0.64, 2.74]
6 Bleeding1378Risk Ratio (M-H, Fixed, 95% CI)4.0 [0.45, 35.46]
Analysis 4.1.

Comparison 4 Adverse events, Outcome 1 Discontinuation of study drug.

Analysis 4.2.

Comparison 4 Adverse events, Outcome 2 Myalgias.

Analysis 4.3.

Comparison 4 Adverse events, Outcome 3 Increased liver chemistries.

Analysis 4.4.

Comparison 4 Adverse events, Outcome 4 Cough.

Analysis 4.5.

Comparison 4 Adverse events, Outcome 5 Dyspepsia/gastrointestinal irritation.

Analysis 4.6.

Comparison 4 Adverse events, Outcome 6 Bleeding.

Appendices

Appendix 1. Search strategies 2012

The Cochrane Library

#1 MeSH descriptor Cardiovascular Diseases explode all trees
#2 cardio*
#3 cardia*
#4 heart*
#5 coronary*
#6 angina*
#7 ventric*
#8 myocard*
#9 pericard*
#10 isch?em*
#11 emboli*
#12 arrhythmi*
#13 thrombo*
#14 atrial fibrillat*
#15 tachycardi*
#16 endocardi*
#17 (sick next sinus)
#18 MeSH descriptor Stroke explode all trees
#19 (stroke or stokes)
#20 cerebrovasc*
#21 cerebral vascular
#22 apoplexy
#23 (brain near/2 accident)
#24 ((brain* or cerebral or lacunar) near/2 infarct*)
#25 MeSH descriptor Hypertension explode all trees
#26 hypertensi*
#27 peripheral next arter* next disease*
#28 ((high or increased or elevated) near/2 (blood next pressure))
#29 MeSH descriptor Hyperlipidemias explode all trees
#30 hyperlipid*
#31 hyperlip?emia*
#32 hypercholesterol*
#33 hypercholester?emia*
#34 hyperlipoprotein?emia*
#35 hypertriglycerid?emia*
#36 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35)
#37 MeSH descriptor Drug Combinations, this term only
#38 polypill*
#39 (drug near/2 combin*)
#40 ((multi* or several) near/2 (ingredient* or component))
#41 policap
#42 quintapill
#43 (single near/2 pill* near/2 comb*)
#44 single-pill
#45 Red Heart pill*
#46 (#37 OR #38 OR #39 OR #40 OR #41 OR #42 OR #43 OR #44 OR #45)
#47 36 and 46, from 2000 to 2012

MEDLINE Ovid

1 exp Cardiovascular Diseases/
2 cardio*.tw.
3 cardia*.tw.
4 heart*.tw.
5 coronary*.tw.
6 angina*.tw.
7 ventric*.tw.
8 myocard*.tw.
9 pericard*.tw.
10 isch?em*.tw.
11 emboli*.tw.
12 arrhythmi*.tw.
13 thrombo*.tw.
14 atrial fibrillat*.tw.
15 tachycardi*.tw.
16 endocardi*.tw.
17 (sick adj sinus).tw.
18 exp Stroke/
19 (stroke or stokes).tw.
20 cerebrovasc*.tw.
21 cerebral vascular.tw.
22 apoplexy.tw.
23 (brain adj2 accident*).tw.
24 ((brain* or cerebral or lacunar) adj2 infarct*).tw.
25 exp Hypertension/
26 hypertensi*.tw.
27 peripheral arter* disease*.tw.
28 ((high or increased or elevated) adj2 blood pressure).tw.
29 exp Hyperlipidemias/
30 hyperlipid*.tw.
31 hyperlip?emia*.tw.
32 hypercholesterol*.tw.
33 hypercholester?emia*.tw.
34 hyperlipoprotein?emia*.tw.
35 hypertriglycerid?emia*.tw.
36 or/1-35
37 Drug Combinations/
38 polypill*.tw.
39 (drug adj2 combin*).tw.
40 ((multi* or several) adj2 (ingredient* or component*)).tw.
41 policap.tw.
42 quintapill.tw.
43 (single adj2 pill* adj2 comb*).tw.
44 single-pill.tw.
45 Red Heart pill*.tw.
46 or/37-45
47 randomised controlled trial.pt.
48 controlled clinical trial.pt.
49 randomised.ab.
50 placebo.ab.
51 drug therapy.fs.
52 randomly.ab.
53 trial.ab.
54 groups.ab.
55 47 or 48 or 49 or 50 or 51 or 52 or 53 or 54
56 exp animals/ not humans.sh.
57 55 not 56
58 36 and 46
59 58 and 57
60 limit 59 to yr="2000 -Current"

EMBASE Ovid

1 exp Cardiovascular Diseases/
2 cardio*.tw.
3 cardia*.tw.
4 heart*.tw.
5 coronary*.tw.
6 angina*.tw.
7 ventric*.tw.
8 myocard*.tw.
9 pericard*.tw.
10 isch?em*.tw.
11 emboli*.tw.
12 arrhythmi*.tw.
13 thrombo*.tw.
14 atrial fibrillat*.tw.
15 tachycardi*.tw.
16 endocardi*.tw.
17 (sick adj sinus).tw.
18 exp cerebrovascular disease/
19 (stroke or stokes).tw.
20 cerebrovasc*.tw.
21 cerebral vascular.tw.
22 apoplexy.tw.
23 (brain adj2 accident*).tw.
24 ((brain* or cerebral or lacunar) adj2 infarct*).tw.
25 exp Hypertension/
26 hypertensi*.tw.
27 peripheral arter* disease*.tw.
28 ((high or increased or elevated) adj2 blood pressure).tw.
29 exp Hyperlipidemias/
30 hyperlipid*.tw.
31 hyperlip?emia*.tw.
32 hypercholesterol*.tw.
33 hypercholester?emia*.tw.
34 hyperlipoprotein?emia*.tw.
35 hypertriglycerid?emia*.tw.
36 or/1-35
37 Drug Combinations/
38 polypill*.tw.
39 (drug adj2 combin*).tw.
40 ((multi* or several) adj2 (ingredient* or component*)).tw.
41 policap.tw.
42 quintapill.tw.
43 (single adj2 pill* adj2 comb*).tw.
44 single-pill.tw.
45 Red Heart pill*.tw.
46 or/37-45
47 36 and 46
48 random$.tw.
49 factorial$.tw.
50 crossover$.tw.
51 cross over$.tw.
52 cross-over$.tw.
53 placebo$.tw.
54 (doubl$ adj blind$).tw.
55 (singl$ adj blind$).tw.
56 assign$.tw.
57 allocat$.tw.
58 volunteer$.tw.
59 crossover procedure/
60 double blind procedure/
61 randomised controlled trial/
62 single blind procedure/
63 48 or 49 or 50 or 51 or 52 or 53 or 54 or 55 or 56 or 57 or 58 or 59 or 60 or 61 or 62
64 (animal/ or nonhuman/) not human/
65 63 not 64
66 47 and 65
67 limit 66 to yr="2000 -Current"

ISI Web of Science

25 #24 AND #23
24 TS=(random* or blind* or allocat* or assign* or trial* or placebo* or crossover* or cross-over*)
23 #22 AND #14
22 #21 OR #20 OR #19 OR #18 OR #17 OR #16 OR #15
21 TS=(single-pill or "red heart pill")
20 TS=(single near/2 pill* near/2 comb*)
19 TS=(policap or quintapill)
18 TS=(several near/2 ingredient* or several near/2 component)
17 TS=(multi* near/2 ingredient* or multi* near/2 component)
16 TS=(drug near/2 combin*)
15 TS=polypill*
14 #13 OR #12 OR #11 OR #10 OR #9 OR #8 OR #7 OR #6 OR #5 OR #4 OR #3 OR #2 OR #1
13 TS=(hyperlipid* or hyperlip?emia* or hyperchlosterol* or hypercholester?emia* or hyperlipoprotein?emia* or hypertriglycerid?emia*)
12 TS=(high near/2 "blood pressure" or increased near/2 "blood pressure" or elevated near/2 "blood pressure")
11 TS=(hypertensi* or "peripheral arter* disease*")
10 TS=(brain* near/2 infarct* OR cerebral near/2 infarct* OR lacunar near/2 infarct*)
9 TS=(brain near/2 accident)
8 TS=apoplexy
7 TS=(stroke or strokes or cerebrovasc* or "cerebral vascular")
6 TS=("sick sinus")
5 TS=(tachycardi* or endocardi*)
4 TS="atrial fibrillat*"
3 TS=(pericard* or isch?em* or emboli* or arrhythmi* or thromo*)
2 TS=(cardia* or heart* or coronary* or angina* or ventric* or myocard*)
1 TS=(cardio)

Appendix 2. Search strategies 2013

The Cochrane Library

#1 MeSH descriptor Cardiovascular Diseases explode all trees
#2 cardio*
#3 cardia*
#4 heart*
#5 coronary*
#6 angina*
#7 ventric*
#8 myocard*
#9 pericard*
#10 isch?em*
#11 emboli*
#12 arrhythmi*
#13 thrombo*
#14 atrial fibrillat*
#15 tachycardi*
#16 endocardi*
#17 (sick next sinus)
#18 MeSH descriptor Stroke explode all trees
#19 (stroke or stokes)
#20 cerebrovasc*
#21 cerebral vascular
#22 apoplexy
#23 (brain near/2 accident)
#24 ((brain* or cerebral or lacunar) near/2 infarct*)
#25 MeSH descriptor Hypertension explode all trees
#26 hypertensi*
#27 peripheral next arter* next disease*
#28 ((high or increased or elevated) near/2 (blood next pressure))
#29 MeSH descriptor Hyperlipidemias explode all trees
#30 hyperlipid*
#31 hyperlip?emia*
#32 hypercholesterol*
#33 hypercholester?emia*
#34 hyperlipoprotein?emia*
#35 hypertriglycerid?emia*
#36 (#1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14 OR #15 OR #16 OR #17 OR #18 OR #19 OR #20 OR #21 OR #22 OR #23 OR #24 OR #25 OR #26 OR #27 OR #28 OR #29 OR #30 OR #31 OR #32 OR #33 OR #34 OR #35)
#37 MeSH descriptor Drug Combinations, this term only
#38 polypill*
#39 (drug near/2 combin*)
#40 ((multi* or several) near/2 (ingredient* or component))
#41 policap
#42 quintapill
#43 (single near/2 pill* near/2 comb*)
#44 single-pill
#45 Red Heart pill*
#46 (#37 OR #38 OR #39 OR #40 OR #41 OR #42 OR #43 OR #44 OR #45)
#47 36 and 46, from 2000 to 2013

MEDLINE Ovid

1 exp Cardiovascular Diseases/
2 cardio*.tw.
3 cardia*.tw.
4 heart*.tw.
5 coronary*.tw.
6 angina*.tw.
7 ventric*.tw.
8 myocard*.tw.
9 pericard*.tw.
10 isch?em*.tw.
11 emboli*.tw.
12 arrhythmi*.tw.
13 thrombo*.tw.
14 atrial fibrillat*.tw.
15 tachycardi*.tw.
16 endocardi*.tw.
17 (sick adj sinus).tw.
18 exp Stroke/
19 (stroke or stokes).tw.
20 cerebrovasc*.tw.
21 cerebral vascular.tw.
22 apoplexy.tw.
23 (brain adj2 accident*).tw.
24 ((brain* or cerebral or lacunar) adj2 infarct*).tw.
25 exp Hypertension/
26 hypertensi*.tw.
27 peripheral arter* disease*.tw.
28 ((high or increased or elevated) adj2 blood pressure).tw.
29 exp Hyperlipidemias/
30 hyperlipid*.tw.
31 hyperlip?emia*.tw.
32 hypercholesterol*.tw.
33 hypercholester?emia*.tw.
34 hyperlipoprotein?emia*.tw.
35 hypertriglycerid?emia*.tw.
36 or/1-35
37 Drug Combinations/
38 polypill*.tw.
39 (drug adj2 combin*).tw.
40 ((multi* or several) adj2 (ingredient* or component*)).tw.
41 policap.tw.
42 quintapill.tw.
43 (single adj2 pill* adj2 comb*).tw.
44 single-pill.tw.
45 Red Heart pill*.tw.
46 or/37-45
47 randomized controlled trial.pt.
48 controlled clinical trial.pt.
49 randomized.ab.
50 placebo.ab.
51 drug therapy.fs.
52 randomly.ab.
53 trial.ab.
54 groups.ab.
55 47 or 48 or 49 or 50 or 51 or 52 or 53 or 54
56 exp animals/ not humans.sh.
57 55 not 56
58 36 and 46
59 58 and 57
60 limit 59 to yr="2000 -Current"
61 (2012* or 2013*).ed.
62 60 and 61
63 limit 62 to “core clinical journals (aim)”

EMBASE Ovid

1 exp Cardiovascular Diseases/
2 cardio*.tw.
3 cardia*.tw.
4 heart*.tw.
5 coronary*.tw.
6 angina*.tw.
7 ventric*.tw.
8 myocard*.tw.
9 pericard*.tw.
10 isch?em*.tw.
11 emboli*.tw.
12 arrhythmi*.tw.
13 thrombo*.tw.
14 atrial fibrillat*.tw.
15 tachycardi*.tw.
16 endocardi*.tw.
17 (sick adj sinus).tw.
18 exp cerebrovascular disease/
19 (stroke or stokes).tw.
20 cerebrovasc*.tw.
21 cerebral vascular.tw.
22 apoplexy.tw.
23 (brain adj2 accident*).tw.
24 ((brain* or cerebral or lacunar) adj2 infarct*).tw.
25 exp Hypertension/
26 hypertensi*.tw.
27 peripheral arter* disease*.tw.
28 ((high or increased or elevated) adj2 blood pressure).tw.
29 exp Hyperlipidemias/
30 hyperlipid*.tw.
31 hyperlip?emia*.tw.
32 hypercholesterol*.tw.
33 hypercholester?emia*.tw.
34 hyperlipoprotein?emia*.tw.
35 hypertriglycerid?emia*.tw.
36 or/1-35
37 Drug Combinations/
38 polypill*.tw.
39 (drug adj2 combin*).tw.
40 ((multi* or several) adj2 (ingredient* or component*)).tw.
41 policap.tw.
42 quintapill.tw.
43 (single adj2 pill* adj2 comb*).tw.
44 single-pill.tw.
45 Red Heart pill*.tw.
46 or/37-45
47 36 and 46
48 random$.tw.
49 factorial$.tw.
50 crossover$.tw.
51 cross over$.tw.
52 cross-over$.tw.
53 placebo$.tw.
54 (doubl$ adj blind$).tw.
55 (singl$ adj blind$).tw.
56 assign$.tw.
57 allocat$.tw.
58 volunteer$.tw.
59 crossover procedure/
60 double blind procedure/
61 randomized controlled trial/
62 single blind procedure/
63 48 or 49 or 50 or 51 or 52 or 53 or 54 or 55 or 56 or 57 or 58 or 59 or 60 or 61 or 62
64 (animal/ or nonhuman/) not human/
65 63 not 64
66 47 and 65
67 limit 66 to yr="2000 -Current"
68 (2012* or 2013*).em.
69 67 and 68
70 limit 69 to priority journals

ISI Web of Science

25 #24 AND #23
24 TS=(random* or blind* or allocat* or assign* or trial* or placebo* or crossover* or cross-over*)
23 #22 AND #14
22 #21 OR #20 OR #19 OR #18 OR #17 OR #16 OR #15
21 TS=(single-pill or "red heart pill")
20 TS=(single near/2 pill* near/2 comb*)
19 TS=(policap or quintapill)
18 TS=(several near/2 ingredient* or several near/2 component)
17 TS=(multi* near/2 ingredient* or multi* near/2 component)
16 TS=(drug near/2 combin*)
15 TS=polypill*
14 #13 OR #12 OR #11 OR #10 OR #9 OR #8 OR #7 OR #6 OR #5 OR #4 OR #3 OR #2 OR #1
13 TS=(hyperlipid* or hyperlip?emia* or hyperchlosterol* or hypercholester?emia* or hyperlipoprotein?emia* or hypertriglycerid?emia*)
12 TS=(high near/2 "blood pressure" or increased near/2 "blood pressure" or elevated near/2 "blood pressure")
11 TS=(hypertensi* or "peripheral arter* disease*")
10 TS=(brain* near/2 infarct* OR cerebral near/2 infarct* OR lacunar near/2 infarct*)
9 TS=(brain near/2 accident)
8 TS=apoplexy
7 TS=(stroke or strokes or cerebrovasc* or "cerebral vascular")
6 TS=("sick sinus")
5 TS=(tachycardi* or endocardi*)
4 TS="atrial fibrillat*"
3 TS=(pericard* or isch?em* or emboli* or arrhythmi* or thromo*)
2 TS=(cardia* or heart* or coronary* or angina* or ventric* or myocard*)
1 TS=(cardio)

Contributions of authors

All authors contributed to the development of the protocol. Angharad de Cates screened titles and abstracts, assessed studies for inclusion and exclusion, extracted data, contacted authors, and drafted the review. Matthew Farr screened titles and abstracts, assessed studies for inclusion and exclusion, and extracted data. Nicola Wright screened titles and abstracts and assessed studies for inclusion and exclusion. Karen Rees supervised the first three authors and contributed to writing the review. Mark Huffman contacted authors, screened titles and abstracts for the rapid review, extracted data, contributed to the analyses, and contributed to writing the review. Shah Ebrahim interpreted findings and contributed to writing the review. Mark Huffman and Shah Ebrahim performed the analyses.

Declarations of interest

Mark Huffman has received grant support (modest) from Scientific Therapeutic Initiative (subsidiary of Astra Zeneca) for a project unrelated to this paper. He has also received travel support from the World Heart Federation through a grant (significant) from Astra Zeneca on research training in implementation science, health system strengthening, and health policy.

Sources of support

Internal sources

  • Warwick Medical School, University of Warwick, UK.

  • Department of Non-Communicable Disease Epidemiology, London School of Hygiene and Tropical Medicine, UK.

External sources

  • NIHR Cochrane Programme Grant, UK.

  • Karen Rees is also funded by the National Institute for Health Research (NIHR) Collaboration for Leadership in Applied Health Research and Care West Midlands at University Hospitals Birmingham NHS Foundation Trust, UK.

Differences between protocol and review

The background section has been shortened. Previous inclusion of HDL cholesterol and triglycerides as outcomes were excluded, and subgroup analysis evaluating the comparator group as usual care versus placebo or inactive control added.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

CRUCIAL 2011

MethodsOpen label cluster randomised trial
Participants136 clusters; 1461 total participants (779 intervention; 682 control participants), men and women aged 35-79 years with hypertension and total cholesterol <250 mg/dl plus three or more risk factors (current smoker, peripheral artery disease, type 2 diabetes, family history of early CHD before aged 55 years in first dgree relative; left ventricular hypertrophy on electrocardiogram [ECG]; history of transient ischemic attack or stroke three or more months prior to screening; ECG abnormalities; age >55 years [men] or >65 years [women], total cholesterol >250mg/dl, or HDL <40mg/dl).
Interventions

Intervention: Single pill amlodipine/atorvastatin (5mg/10mg to 10mg/10mg; site investigators could request dosages of 5/20 mg and 10/20 mg) in addition to other hypertensive / lipid lowering therapy as required, as well as therapeutic lifestyle counselling change

Control: Usual care, including therapeutic lifestyle counselling change

OutcomesSBP, DBP, LDL-C, total cholesterol; all-cause mortality reported
NotesControl: inactive/usual care
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear risk"Investigators - randomly assigned", "randomisation was stratified", "investigator as unit of randomisation"
Allocation concealment (selection bias)Unclear riskDue to cluster randomisation
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label
Blinding of outcome assessment (detection bias)
All outcomes
High riskOpen label
Incomplete outcome data (attrition bias)
All outcomes
High risk93/779 (11.9%) discontinued intervention; 44/682 (6.5%) discontinued in usual care arm
Selective reporting (reporting bias)Unclear riskNot all outcomes available for meta-analysis
Other biasUnclear riskSignificant differences between two arms in terms of baseline blood pressure and ECG abnormalities/PVD; underpowered according to authors' own calculations; different doses of intervention were available upon request to investigators

CUSP 2009

MethodsIndividual-level randomised controlled trial
Participants130 participants (66 intervention; 64 control) with coexisting, untreated hypertension (SBP=140-169 mmHg or DBP=90-105 mmHg) and dyslipidemia (LDL-C=110-160 mg/dl) but without a history of cardiovascular disease; age >21 years
Interventions

Intervention: Single pill amlodipine/atorvastatin (5mg/20mg) + therapeutic lifestyle changes

Control: Therapeutic lifestyle changes

OutcomesTarget for BP <140/90 mm Hg and LDL-C <100 mg/dL [2.59 mmol ⁄L] at week 4 and week 8: the percentage of patients in whom the single LDL-C goal was reached at weeks 4 and 8; mean changes from baseline in SBP and DBP at weeks 4 and 8; mean changes from baseline in LDL-C at weeks 4 and 8; 10-year Framingham risk of CHD at weeks 4 and 8
NotesControl: inactive/usual care
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot specifically stated: "Patients were randomised in a double-blind manner"
Allocation concealment (selection bias)Unclear riskNot specifically stated
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskNot specifically stated
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskNot specifically stated
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskUnclear how data from particpants lost to follow-up were handled
Selective reporting (reporting bias)Low riskPrimary outcomes reported (week 4 blood pressure and LDL targets)

Malekzadeh 2010

MethodsBlock randomisation
Participants475 participants (241 polypill; 234 control) without cardiovascular disease, hypertension, or hyperlipidaemia aged 50 to 79 years (men) and 55 to 79 years (women)
Interventions

Intervention: Polypill (aspirin 81 mg, enalapril 2.5 mg, atorvastatin 20 mg and hydrochlorothiazide 12.5 mg)

Control: placebo

OutcomesHospital admissions / major cardiovascular events / seated and standing BP, LDL-C, total cholesterol, triglycerides, HDL-C and fasting glucose
NotesControl: inactive/placebo
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer generated block randomisation
Allocation concealment (selection bias)Low riskComputer generation allocation to numbered list of blister packs manufactured by Alborz Darou
Blinding of participants and personnel (performance bias)
All outcomes
Low riskIdentical blister packs used for participant blinding
Blinding of outcome assessment (detection bias)
All outcomes
Low riskOutcome assessors (clinicians) blinded to allocation
Incomplete outcome data (attrition bias)
All outcomes
High riskHigh rate of loss to follow-up at 12 months (experimental 32%; control 22%)
Selective reporting (reporting bias)Low riskPrimary outcome reported (changes in blood pressure and LDL cholesterol)
Other biasHigh riskRun-in period excluded participants with low (<70%) adherence; large differences in baseline characteristics between intervention and control groups

PILL 2011

MethodsIndividual-level randomised controlled trial
Participants378 participants (189 intervention; 189 control) with 5-year Framingham coronary heart disease risk ≥7.5% or if Framingham risk was between 5% and 7.5%, two or more additional untreated risk factors were needed (body mass index >30kg/m2, waist circumference >102cm in men or >88cm in women; heart rate > 80 bpm; fasting glucose 5.6-7 mmol/L, triglycerides >1.7 mmol/L; family history of first degree relative with premature ischemic heart disease or stroke (men < 55 years; women: <65 years), or glomerular filtration rate <60ml/min
Interventions

Intervention: Red heart pill (aspirin 75 mg, lisinopril 10mg, hydrochlorothiazide 12.5mg and simvastatin 20mg)

Control: placebo

OutcomesChange in SBP; change in LDL-C; tolerability; secondary outcomes included adherence, DBP, total cholesterol, HDL-C, total cholesterol:HDL cholesterol ratio, non-HDL cholesterol, triglycerides, frequency of switching/adding open-label treatment, estimated effects on CVD risk
NotesControl: inactive/placebo
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskCentral computer based randomisation
Allocation concealment (selection bias)Low riskCentral computer based randomisation
Blinding of participants and personnel (performance bias)
All outcomes
Low riskSpecifically reported and use of placebo control
Blinding of outcome assessment (detection bias)
All outcomes
Low riskOutcome assessors and study staff all blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskLow rates of loss to follow-up (experimental 2%; control 1%)
Selective reporting (reporting bias)High riskLast observation carried forward for missing data at week 12.

Soliman 2009

MethodsOpen label, parallel group randomised clinical trial
Participants216 (105 Polypill; 111 control); ≥40 years for men and ≥50 years for women; estimated 10 year World Health Organization total cardiovascular risk score ≥20% without established cardiovascular disease
Interventions

Intervention: Red Heart pill 2b (75 mg aspirin, 20 mg simvastatin, 10 mg lisinopril and 12.5 mg hydrochlorothiazide)

Control: Standard practice defined by the study investigators

OutcomesSBP, total cholesterol, 10-year cardiovascular disease risk, adherence, fasting glucose, creatinine, potassium, and liver enzymes
NotesControl: inactive/usual care
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNo method of randomisation stated
Allocation concealment (selection bias)High riskOpen label
Blinding of participants and personnel (performance bias)
All outcomes
High riskOpen label
Blinding of outcome assessment (detection bias)
All outcomes
High riskOpen label
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskUnclear how missing data were handled
Selective reporting (reporting bias)Low riskPrimary outcomes (blood pressure, cholesterol, ten year CVD risk) all reported
Other biasHigh riskUse of non-study antihypertensives and statins very different between centres

TIPS 2009

MethodsIndividual-level randomised controlled trial
Participants2053 participants (205 aspirin; 205 thiazide; 209 thiazide + ramipril; 207 thiazide + atenolol; 205 ramipril + atenolol; 204 thiaizde + ramipril + atenolol; 204 thiaizide + ramipril + atenolol + aspirin; 202 simvastatin; 412 Polycap [thiazide + ramipril + atenolol + simvastatin + aspirin); 45 to 80 years old without prior cardiovascular disease but with at least one risk factor: type 2 diabetes; blood pressure >140/90 mmHg but <160/100 mmHg; smoker within the past five years; waist-to-hip ratio >0.85 for women and 0.90 for men; LDL cholesterol >3.1 mmol/L but less 4.5 mmol/L or HDL cholesterol <1.04 mmol/L.
Interventions

Intervention: Polycap (thiazide 12·5 mg, atenolol 50 mg, ramipril 5 mg, simvastatin 20 mg, aspirin 100 mg)

Control: 8 other drug/drug combination groups listed above

OutcomesLDL for the effect of lipid-lowering drugs, BP for antihypertensive drugs, heart rate for the effects of atenolol, urinary 11-dehydrothromboxane B2 for the antiplatelet effects of aspirin, rates of discontinuation of drugs for safety
NotesControl: active
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskCentral computer randomisation
Allocation concealment (selection bias)Low riskCentral computer randomisation
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo control using identical capsule
Blinding of outcome assessment (detection bias)
All outcomes
Low riskDouble-blinding reported; probably occurred given research team's prior studies
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskUnclear how missing SBP and LDL-C data at week 12 follow-up were handled
Selective reporting (reporting bias)Low riskPrimary outcomes reported

TOGETHER 2010

MethodsIndividual-level randomised, double dummy controlled trial
Participants244 participants (122 intervention; 122 control) with history of hypertension but no history of CVD or diabetes with ≥2 risk factors: age ≥45 years for men; ≥55 years for women; current smoker; family history of premature coronary heart disease in first degree relative; HDL cholesterol <40 mg/dl; waist circumference >102 cm in men and >88 cm in women.
Interventions

Intervention: single pill amlodipine (5/10mg) plus atorvastatin 20mg + therapeutic lifestyle changes

Control: amlodipine (5/10mg) + therapeutic lifestyle changes

OutcomesProportion achieving a BP goal <140/90 mmHg and LDL-C<100 mg/dl at week 6; BP and LDL-C goal at week 4; BP goal at weeks 4 and 6; change in SBP, DBP, LDL-C, total cholesterol, HDL-C, triglycerides at weeks 4 and 6; predicted 10 year Framingham coronary heart disease risk score, adverse events
NotesControl: active
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskCentral, computer based telerandomisation
Allocation concealment (selection bias)Low riskCentral, computer based telerandomisation
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskDouble blind labeled bottles but unclear if pills were identical
Blinding of outcome assessment (detection bias)
All outcomes
Low riskReportedly double blinded
Incomplete outcome data (attrition bias)
All outcomes
High riskLast observation carried forward used for non-completers for final analysis
Selective reporting (reporting bias)Unclear riskPrimary outcomes reported

UMPIRE 2013

MethodsRandomised, open label, blinded endpoint clinical trial of an FDC-based treatment strategy compared with usual care
Participants≥18 years old and established CVD or an estimated 5 year CVD risk of 15% or greater in India and 3 European countries (England, Ireland, and the Netherlands)
Interventions

Intervention: one of two versions of the fixed-dose combination ((1) aspirin 75mg, simvastatin 40mg, lisinopril 10mg, atenolol 50mg or (2) aspirin 75mg, simvastatin 40mg, lisinopril 10mg, hydrochlorothiazide 12.5mg)

Control: usual care

Outcomes

Primary: adherence to indicated medications (self-reported current use of antiplatelet, statin, and ≥2 BP-lowering therapies, defined as taking the medication for at least 4 days during the week preceding the visit) at baseline and at the end of the trial and changes in SBP and LDL-C from baseline to the end of the trial.

Secondary: adherence at 12 months, reasons for stopping cardiovascular medications, quality of life, serious adverse events, and changes in total cholesterol, HDL-C, triglycerides, and creatinine from baseline to 12 months and end of study and cardiovascular events (including coronary heart disease, heart failure leading to death or hospital admission, and cerebrovascular or peripheral arterial disease events)

Notes

Control: inactive/usual care

Trial is part of "Single Pill Against Cardiovascular Events (SPACE)" collaboration, which encompasses the "Improving Adherence using Combination Therapy (IMPACT)" and "Kaniyini Guidelines Adherence with the Polypill (Kanyini-GAP)" trials.

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation occurred through web-based clinical data management system
Allocation concealment (selection bias)Low riskRandomisation occurred through web-based clinical data management system
Blinding of participants and personnel (performance bias)
All outcomes
High riskParticipants and personnel were unblinded
Blinding of outcome assessment (detection bias)
All outcomes
Low riskOutcome assessors were blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskAt the end of the study, data on self-reported adherence, systolic BP, and LDL-C were available for 1921 (96%), 1849 (92%), and 1807 (90%) randomized participants, respectively
Selective reporting (reporting bias)Low riskAll primary outcomes reported; quality of life outcomes were not reported in this initial report
Other biasUnclear riskParticipants randomized to the intervention arm received fixed-dose combination therapy at no cost compared with participants randomized to usual care who were responsible for their drug costs

Wald 2012

MethodsIndividual-level randomised double-blind placebo-controlled cross-over trial
Participants86 individuals (43 Polypill then placebo; 43 placebo then Polypill) aged 50 years or over without history of cardiovascular disease who were previously taking simvastatin and blood pressure lowering drugs; limited to participants living in London or could travel easily to London
Interventions

Intervention: fixed-dose combination (amlodipine 2.5mg, losartan 25mg, hydrochlorothiazide 12.5mg, simvastatin 40mg) daily for 12 weeks

Control: placebo

OutcomesSBP, DBP, total cholesterol, LDL-C, HDL-C, triglycerides, apoB, adherence
NotesControl: inactive/placebo
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer generated block randomisation
Allocation concealment (selection bias)Low riskComputer generated block randomisation with sequential identical blister packs
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo controlled
Blinding of outcome assessment (detection bias)
All outcomes
Low riskOutcome assessors reported as being blinded
Incomplete outcome data (attrition bias)
All outcomes
Low riskPrimary outcomes reported
Selective reporting (reporting bias)Unclear riskAdverse event data not clearly described; only proportion of individuals with "symptom", which was assumed to be an adverse event
Other biasLow riskNo need for intention-to-treat analysis as cross-over design. Any losses to follow-up clear

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Avenell 2012Not related to cardiovascular outcomes
Bakris 2012Did not include statin fixed-dose combination intervention
Blank 2005Intervention fixed-dose combination therapy but not RCT
Boger-Megiddo 2010Case-control study, not RCT
CAPABLE 2009aIntervention fixed-dose combination therapy but not RCT
CAPABLE 2009bIntervention fixed-dose combination therapy but not RCT
Chapman 2010Intervention polypill but not RCT
Chyrsant 2011Review article, not RCT
Derosa 2013Did not include statin fixed-dose combination intervention
Gemini-AALA 2009Intervention fixed-dose combination therapy but not RCT
JEWEL 2006Intervention fixed-dose combination therapy but not RCT
JEWEL 2006aSee above
Li 2011Non-randomized, non-comparator study
Liew 2009Intervention fixed-dose combination therapy but not RCT (cost-effectiveness analysis)
Neldam 2012Not RCT
Neutel 2012Not RCT
Nitsch 2013NSAID coprescription, not RCT
Patel 2010Intervention fixed-dose combination therapy but not RCT
Sun 2012Comparator: fixed-dose combination
TIPS-2 2012Comparator group includes fixed-dose combination therapy
Wald review 2012Review article, not RCT
Weber 2012Retrospective analysis, not RCT
Weber 2013Secondary analysis of ACCOMPLISH trial
Zhu 2012Did not include statin fixed-dose combination intervention

Characteristics of ongoing studies [ordered by study ID]

FOCUS 2011

Trial name or titleFOCUS
MethodsRandomised controlled trial
Participants1340 post-MI patients followed up for 9 months
InterventionsFixed-dose combination (aspirin 100mg, ramipril 2.5mg/5mg/10mg, simvastatin 40mg) or three drugs separately
Outcomes

Primary outcomes: adherence to treatment

Secondary outcomes: changes in SBP/DBP/LDL-C, adverse events, economic data

Starting dateProtocol published November 2011
Contact information 
Notes 

IMPACT 2011

Trial name or titleIMProving Adherence using Combination Therapy (IMPACT)
MethodsOpen-label randomised controlled trial
Participants600 participants who have had CVD events or are at high risk of CVD followed up for 12 months
InterventionsFixed-dose combination (aspirin 75mg, simvastatin 40mg, lisinopril 10mg, hydrochlorothiazide 12.5mg/atenolol 50mg) or current medications
Outcomes

Primary outcomes: adherence to prescribed medication, changes in SBP/LDL-C

Secondary outcomes: other serum lipids, medication dispensing, barriers to adherence, CVD events, other serious adverse events, quality of life, prescriber acceptability

Starting dateProtocol published July 2011
Contact information 
NotesLinked to Kanyini-GAP and UMPIRE

Kanyini-GAP 2010

Trial name or titleKanyini-Guidelines Adherence with the Polypill (Kanyini-GAP)
MethodsOpen randomised controlled trial
Participants1000 participants at high risk of cardiovascular events recruited from mainstream GP practices and Aboriginal health services followed up for an average of 18 months
InterventionsOne of two versions of fixed-dose combination (chosen by treating clinician: (1) aspirin 75mg, simvastatin 40mg, lisinopril 10mg, atenolol 50mg or (2) aspirin 75mg, simvastatin 40mg, lisinopril 10mg, hydrochlorothiazide 12.5mg) or to usual care
Outcomes

Primary outcomes: change in cholesterol / SBP, self-reported use of aspirin / simvastatin / at least two antihypertensives

Secondary outcomes: cardiovascular events, renal outcomes, self-reported barriers to indicated therapy, prescription of indicated therapy, serious adverse events, changes in quality of life

Starting dateProtocol published August 2010
Contact information 
NotesLinked closely to IMPACT and UMPIRE

Merat 2010

Trial name or titlePolypill and Nonalcoholic Steatohepatitis (PolyIran-L)
MethodsUnclear in trial registration
ParticipantsUnclear in trial registration, followed up for 5 years
InterventionsFixed-dose combination (unspecified)
OutcomesCardiovascular events (also non-alcoholic steatohepatitis-specific outcomes)
Starting dateRegistered 19/11/2010 (NCT01245608)
Contact information 
NotesNo publication of protocol, design or any data as yet

PolyIran 2010

Trial name or titlePolyIran
MethodsThree-armed open randomised controlled trial
Participants30000 participants over 50 years in Iran followed up between 2 and 5 years
InterventionsFixed-combination therapy (aspirin, statin, antihypertensives - not detailed) + minimal care; minimal care alone; usual care alone
OutcomesCardiovascular events, cardiovascular-specific mortality
Starting dateRegistered 14/12/2010 (NCT01271985)
Contact information 
NotesNo publication of protocol, design or any data as yet

TIPS-3 2012

Trial name or titleThe International Polycap Study-3
Methods2 x 2 x 2 randomised controlled trial, factorial design (3 arms: Polycap D, aspirin, vitamin D)
Participants5500 participants (women 60 years or older and men 55 years or older) without known heart disease or prior stroke and without a clear indication or contraindication to any of the study medications and INTERHEART risk score of 10 or greater
InterventionsPolycap DS vs. placebo; embedded in trial comparing enteric coated aspirin vs. placebo and vitamin D vs. placebo
OutcomesComposite of major CVD (CV death, non-fatal stroke, non-fatal MI), plus heart failure, resuscitated cardiac arrest, or revascularisation with evidence of ischemia in participants taking Polycap versus placebo
Starting dateProtocl updated on clinicaltrials.gov on October 2012 (ClinicalTrials.gov Identifier:NCT01646437)
Contact informationDr. Salim Yusuf, Population Health Research Institute
Notes 

Ancillary