Postgame Wrap of the Ultimate Blood Pressure Megatrial
Did It Score an ALLHAT Trick or Was It “Three Strikes and You’re Out?”
I never heard a crowd boo a homer, but I’ve heard plenty of boos after a strikeout.
— —Babe Ruth
It is a long base path that links the development and execution of a groundbreaking clinical trial with the assessment of its effect on clinical practice and ultimately health outcomes. The bases include the following: (1) publishing and disseminating the results of the study; (2) incorporating the findings into clinical practice recommendations (developed either informally or formally, as per the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure Guidelines process); (3) impacting on clinical practice of healthcare professionals; and (4) impacting on patient outcomes and/or healthcare resource use.
In hypertension there has been a succession of landmark clinical trials that have dramatically affected the way we manage this disorder, eg, the initial Veterans Affairs trials, which proved that hypertension was not really “essential,” and the Systolic Hypertension in the Elderly Program Study,1 which validated the benefits of treating systolic hypertension (remember the bad old days when it was taught that a normal blood pressure was “100 mm Hg plus their age”). These were truly “home runs” in the management of hypertension.
In this context, it is very appropriate to begin to examine the box score for the Antihypertensive and Lipid Lowering treatment to prevent Heart Attack Trial (ALLHAT), certainly the largest and most robust primary prevention hypertension trial ever.2 This 3-way comparison among a diuretic, angiotensin-converting enzyme inhibitor, and calcium channel blocker finished as a “dead heat” (the fourth comparator drug, an α-adrenergic antagonist, having been stopped early because of lack of any benefit for the primary outcome and a signal suggesting potential harm in some of the secondary outcomes, eg heart failure). For the primary outcome there were no significant differences among any of the 3 competing drugs, although differences in secondary outcomes were debated both verbally and in print for years after. However, the major message “marketed” by the steering committee was that the results of the ALLHAT Study endorsed the more widespread use of thiazide diuretics as preferred first-line therapy in the treatment of hypertension. The rationale for this message was that the comparable efficacy of diuretics (versus the newer agents) in the primary outcome and superior efficacy in some of the secondary outcomes, coupled with much lower cost of diuretics, should translate into their preferential use.
How well has ALLHAT done in rounding the bases toward influencing patient outcomes? The current article by Muntner et al in this issue of Hypertension3 examines an interim step along the way, ie, determining whether ALLHAT influenced prescribing rates for diuretics versus other antihypertensive classes. The study does provide a quantitative answer regarding the change in diuretic prescription rates over time. However, the role of ALLHAT in this phenomenon is not definitively proven (and cannot be from an observational study, such as the present one). To definitively prove that a particular piece of evidence influenced practice would require an experimental study in which clinicians (and their patients) were randomly allocated to be exposed to the evidence or not, with standardization of all other influences on clinician practice (including not only other sources of evidence but also pharmaceutical and direct-to-consumer marketing, drug samples, etc). Clearly, such a study is impossible to conduct when exploring the impact of a large clinical trial published in the Journal of the American Medical Association and reported in various secondary journals.
There was and is widespread appreciation of the superb design and execution of the ALLHAT Study. In addition, the primary outcome (fatal coronary heart disease or nonfatal myocardial infarction combined) was unambiguous: all 3 of the drug regimens performed comparably. There was widespread dissemination of the results via publication in the Journal of the American Medical Association for the primary outcome and in several other high-impact journals for subgroup analyses. Thus, ALLHAT reached first base.
In regard to getting to second base, incorporation into practice guidelines, ALLHAT seems to have had significant impact as well, at least within the United States. The major message from the study, ie, that “thiazide-type diuretics should be used in drug treatment for most patients with uncomplicated hypertension . . .” was prominently incorporated into the 2003 Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure Guidelines recommendations.4 However, ALLHAT had less impact on hypertension guidelines formulated beyond the borders of the United States (such as the International Society of Hypertension/European Society of Hypertension recommendations and those of the Canadian Hypertension Education Program).5
Although we will not know for a number of years whether ALLHAT positively impacted patient outcomes in nontrial participants, Muntner et al3 have provided us with data demonstrating a potential impact on prescribing practices. Over a 3-year window before and after release of the study results, the use of thiazides increased from 31% of all antihypertensive prescriptions to 37%. However, renin-angiotensin system inhibitors remained the most commonly prescribed antihypertensives in all of the years studied. Further, there was a comparable rise in the proportional use of renin-angiotensin system inhibitors before and after ALLHAT (from 41% to 45%). This primarily related to a rise in the use of angiotensin receptor blockers. Indeed, even after ALLHAT, only 21% of initial antihypertensive prescriptions were for thiazides.
Whether this cup is half full or half empty is in the eye of the beholder. There was an increase in diuretic use, which represented a reversal of the progressive decline seen in diuretic use in the years preceding ALLHAT’s publication. However, over the same time frame, there was also a near doubling in the use of angiotensin receptor blockers, a class of agents that was not part of the ALLHAT “field.” Indeed, the authors of the far smaller Losartan Intervention for Endpoint Reduction in Hypertension Study (LIFE),6 a trial comparing angiotensin receptor blocker with β-blocker released in 2002, could point to a much more dramatic rise in angiotensin receptor blocker use than in diuretic use over the years studied by Muntner et al.3 In that context, one could argue that LIFE had much more influence on prescribing practices than ALLHAT.
Of course, inferring the impact of ALLHAT versus LIFE on the basis of prescribing patterns alone is unfair, because physician prescribing is influenced by more than just the quality of the evidence.7 Although some may argue that any trial marketed by the pharmaceutical industry will have more impact than those without industry championship. However, there is evidence that even public sector/volunteer initiatives can effectively market evidence with substantial impact on both clinical practice and outcomes.
One only has to look north of the border to the efforts of the Canadian Hypertension Education Program (CHEP), an all-volunteer initiative charged with developing, marketing, and implementing the Canadian national hypertension guidelines. Using similar observational approaches as those used by Muntner et al,3 several studies have documented significant and substantial improvements in antihypertensive management within Canada since initiation of CHEP. For example, after the introduction of CHEP, the use of diuretics for hypertension in Canada increased by >10% per year, even after adjusting for the pre-CHEP prescribing rates.8 In addition, since the introduction of CHEP there has been a marked improvement in hypertension control rates (≥65%, as reported in the recent Ontario Blood Pressure Study)9 with concomitant reductions in hypertension-related cardiovascular complication rates in Canada.10
The barriers that impede the translation of randomized clinical trial evidence into clinical practice are legion and well documented.11 In addition to these more general barriers, the fact that the ALLHAT message was not universally accepted by many key hypertension opinion leaders in the United States undoubtedly negatively impacted its ability to influence practice. These dissenting voices, amplified by the divergent goals of the pharmaceutical companies representing non–ALLHAT-recommended drugs, could have significantly attenuated the impact of the study.
What can we learn from the ALLHAT experience? First, we can learn that a single study, no matter how large and how well done, may not be seen as the “last word” in any field. Second, the incorporation of clinical trial findings into clinical practice guidelines and, ultimately, clinical practice requires a concerted effort. Strategies proven to enhance the implementation of evidence into practice include critical pathways, real-time clinical reminders, disease management strategies, local opinion leaders, academic detailing, and (sometimes) audit and feedback. It is incumbent on guideline developers to foster active implementation processes that incorporate such proven strategies to ensure that the outstanding contribution of trialists such as the ALLHAT team move beyond the printed page and into clinical practice.
R.D.F. is the past chair of the Canadian Hypertension Education Program. F.A.M. reports no conflicts.
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
The ALLHAT Officers and Coordinators for the ALLHAT Collaborative Research Group. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs. diuretic: the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), JAMA. 2002; 288: 2981–2997.
Muntner P, Krousel-Wood M, Hyre AD, Stanley E, Cushman WC, Cutler JA, Piller LB, Goforth GA, Whelton PK. Antihypertensive prescriptions for newly treated patients before and after the main Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial results and seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure guidelines. Hypertension. 2009; 53: 617–623.
Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL, Jones DW, Materson BJ, Oparil S, Wright JT, Roccella EJ, and the National High Blood Pressure Education Program Coordinating Committee. The seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. 2003; 42: 1206–1252.
Dahlof B, Devereux RB, Kjeldsen SE, Julius S, Beevers G, de Faire U, Fyhrquist F, Ibsen H, Kristiansson K, Lederballe-Pedersen O, Lindholm LH, Nieminen MS, Omvik P, Oparil S, Wedel H, the LIFE Study Group. Cardiovascular morbidity and mortality in the Losartan Intervention for Endpoint Reduction in Hypertension Study (LIFE): a randomised trial against atenolol. Lancet. 2002; 359: 995–1003.
Campbell NRC, Tu K, Duong-Hua M, Brant R, McAlister FA. The impact of the Canadian Hypertension Education Program on antihypertensive prescribing trends. Hypertension. 2006; 47: 22–28.
Leenen FH, Dumais J, McInnis NH, Turton P, Stratychuk L, Nemeth K, Lum-Kwong MM, Fodor G. Results of the Ontario survey on the prevalence and control of hypertension. CMAJ. 2008; 178: 1441–1449.
Campbell NR, Brant R, Johansen H, Walker RL, Wielgosz A, Onysko J, Gao RN, Sambell C, Phillips S, McAlister FA; for the Canadian Hypertension Education Program Outcomes Research Task Force. Increases in antihypertensive prescriptions and reductions in cardiovascular events in Canada. Hypertension. 2009; 53: 128–134.