INTRODUCTION


The epidemic of heart failure (HF) is perhaps one of the most important and challenging public health issues in the United States today. Recent decades have seen a dramatic rise in the number of persons who carry the HF diagnosis and the number of HF-related hospitalizations, as well as the resulting economic impact on the health-care system. Current estimates by the American Heart Association report that 5.2 million Americans have HF.1 In addition, 600,000 incident cases and more than 1 million hospitalizations occur annually at a cost of >$33 billion.1 If current trends in incidence and survival remain constant, studies project marked future increases in both HF prevalence and cost.2 This review will focus on recent epidemiologic trends and will concentrate on the latest scientific contributions to the field of HF epidemiology.


INCIDENCE


As background to any discussion of current trends in HF incidence, it is important to understand the limitations of the data available to researchers. No national HF surveillance system exists, and current data on HF incidence are derived primarily from cohort studies and administrative databases. Cohort studies, such as the Framingham Heart Study and Rochester Epidemiology Project, provide the most reliable information, as they include data from both inpatients and outpatients diagnosed with HF, and have the ability to apply standard validated criteria for HF diagnosis consistently over decades of follow-up. However, these cohort studies analyze relatively small homogeneous patient populations in limited geographic regions, and their findings may not reflect the experience of diverse populations across the United States. Although large administrative database studies generally overcome these limitations by evaluating populations of patients across broad geographic regions, they have several major shortcomings as well. First, they are typically limited to patients hospitalized with HF, and do not include those diagnosed in the outpatient setting. Second, they frequently rely on administrative billing codes for HF diagnosis, which are less accurate than clinical diagnostic criteria and may substantially underestimate the number of HF-related hospitalizations.3 Finally, administrative databases usually do not provide detailed clinical information on important clinical variables such as HF etiology and left ventricular systolic function. Nevertheless, despite their limitations, recent studies of HF incidence offer important epidemiological insights.


Data from the Framingham Heart Study indicate that HF incidence in the United States is quite high. Lloyd-Jones and colleagues demonstrated that the lifetime risk of developing HF at the age of 40 years is close to 20% in both men and women.4 In another Framingham study of age-adjusted HF incidence trends between the 1950s and 1990s, Levy and colleagues showed an overall trend toward lower age-adjusted HF incidence during this time period that was statistically significant in women. (Table 2-1).5 Specifically, the age-adjusted incidence decreased from 627 to 564/100,000 person-years in men (7% relative risk reduction) and from 420 to 327/100,000 person-years in women (31% relative reduction) between the periods of 1950–1969 and 1990–1999.5 However, close analysis of the data indicates that the reduction in HF incidence actually took place between the periods of 1950–1969 and 1970–1979. In fact, it appears that no decrease in HF incidence has occurred since the 1970s.


Other studies have reported a trend of stable or even increasing age-adjusted incidence more recently. Data from the Resource Utilization Among Congestive Heart Failure (REACH) study of the Henry Ford Health System indicate that age-adjusted HF incidence cases rose slightly among women from 3.7 to 4.2 cases/1000 patients, while there was a trivial decline among men from 4.0 to 3.7 cases/1000 patients between 1989 and 1999 (neither change was statistically significant).6 Findings of the Rochester Epidemiology Project from Olmsted County, Minnesota, show that there has been a nonsignificant increase in age-adjusted HF incidence of 4% in men and 11% in women between the periods of 1979–1984 and 1996–2000.7


It is somewhat surprising that the age-adjusted HF incidence has not declined in the past 20–30 years. Recent data suggest that the control of hypertension has improved in recent years.8 Hypertension remains one of the main risk factors for HF, precedes HF in >90% of patients, and has a 39% population-attributable risk of HF in men and a 59% population-attributable risk in women.9 According to the Framingham Heart Study, the lifetime risk of HF doubles in patients with blood pressure ≥160/100 mm Hg compared with those who have blood pressure <140/90 mm Hg.4 Data from the National Health And Nutrition Examination Survey (NHANES) showed that there were small improvements in the proportion of patients who were treated and whose hypertension was controlled (6% and 6.4%, respectively) between 1988 and 1991 and 1999 and 2000.8


Furthermore, recent trends in the epidemiology of myocardial infarction (MI) and coronary artery disease also suggest that age-adjusted incidence of HF should be declining. Several studies have documented better survival in patients with acute MI, such as the community-based study of patients hospitalized in Worcester, Massachusetts, that showed a decline in in-hospital mortality from 17.8% to 11.7% between the mid-1970s and 1990s.10 However, the incidence of MI has remained unchanged during the same time period. According to data from the Rochester Epidemiology Project, the age and sex-adjusted incidence of MI experienced a nonsignificant 6% decline between 1979 and 1998.11 The incidence of all coronary artery disease has also remained relatively stable, with a nonsignificant 9% decline between 1988 and 1998.11 Although it has been hypothesized that better survival after MI without concomitant decline in its incidence may result in higher incidence of post-MI HF, this notion is not supported by evidence. In fact, better management of patients following MI has resulted in declining post-MI HF incidence rates (28% relative risk reduction during 1979–1994, mean follow-up 7.6 years) according to the data from Olmsted County, Minnesota.12


One of the key reasons for the lack of decline in age-adjusted HF incidence is the increasing prevalence of several important HF risk factors. Recent data from NHANES demonstrated that the prevalence of hypertension increased by 3.7% between 1988 and 1991 and 1999 and 2000.8 Further, although treatment and control of hypertension improved by 6% and 6.4%, respectively, 69% of patients still did not have adequate hypertension control.8 Similar findings were observed in the Cardiovascular Health Study, which showed that although treatment of blood pressure improved among elderly patients, it was still suboptimal, with more than half of the patients not having adequate hypertension control.13 Rising obesity rates may be contributing to the current lack of improvement in HF incidence as well. Data from the Framingham Heart Study indicate that the risk of developing HF in obese patients (body mass index >30) is double that of patients with normal body mass index, and after adjustment for established risk factors there is a 5–7% rise in the relative risk of HF with every 1 point increase in body mass index.14 According to data from NHANES and the Behavioral Risk Factor Surveillance System (BRFSS), the prevalence of obesity has been rising dramatically over the last decade.15–19 Possibly related to increased obesity rates, the prevalence of diabetes is on the rise as well.20 Diabetes has been shown to be a significant risk factor for HF,21 and the data from NHANES show that the control of glucose, hypertension, and hyperlipidemia remains poor among diabetic patients, and did not improve from 1988 to 2000.22 The rising prevalence of obesity and diabetes are important contributors to the lack of improvement in age-adjusted HF incidence rates, and will likely become even more important in the near future.


Although data on the temporal trends in overall (not age-adjusted) HF incidence are lacking, incidence is likely to increase in the near future because of the aging of the American population. Data from the REACH study clearly show a dramatic increase in HF incidence with age (Fig. 2-1), with a rate as high as 40/1000 in women aged >85 years, and even higher in similarly aged men.6 A European study from Rotterdam confirms these findings and shows that incidence rate increases with age from 1.4/1000 person-years in those aged 55–59 to 47.4/1000 person-years in those aged ≥90.23 Thus, the rising number of elderly persons in the United States will invariably affect the <65 66–74 75–84 > 85 Age group number of new HF diagnoses, even if age-adjusted incidence remains stable.


PREVALENCE


There has been a marked rise in the prevalence of HF over the past 2–3 decades. The American Heart Association estimates that in 2004, 5.2 million patients had the diagnosis of HF in the United States—an increase from the 1 to 2 million estimated during the period of 1971–1975.1 Similar to incidence, the prevalence of HF rises markedly with age. According to data from the Cardiovascular Health Study, the prevalence of HF rises from 4% in women aged 70–74 years to 14% in women aged >85 years, with a comparable marked increase among similarly aged men.24 Comparable observations were made in the Rotterdam study from Europe, with prevalence increasing from 0.9% in subjects aged 55–64 years to 17.4% in those aged ≥85 years.23


Analysis of temporal trends in HF prevalence from NHANES shows dramatic increases among all age groups between the late 1970s and early 1990s, although the most rapid increase took place among the elderly (Fig. 2-2). Experience from the individual health-care systems also shows that substantial increases in HF prevalence took place during the 1990s. Data from REACH demonstrate that age-adjusted HF prevalence has increased from 3.7% to 14.3% in women and from 4.0 to 14.5/1000 patients in men during 1989–1999.6


Again, the aging of the American population is likely one of the important factors behind the rising prevalence rates (Fig. 2-3). The U.S. Bureau of the Census estimates that the number of persons aged >65 years has increased by nearly 4 million and the number of very elderly (those aged >85 years) increased by 1.2 million between 1990 and 2000.25 More importantly, future projections suggest that the proportion of patients aged >60 years will increase dramatically from 16.5% in 1997 to 24.6% in 2025, with the actual number of older Americans nearly doubling from >44 million to >82.5 million during the same time period.26 Given the higher HF prevalence in the elderly, the continuous rise in the overall numbers as well as proportion of patients with the diagnosis is likely to continue. In fact, a recent study from Scotland projects that even if the overall HF prevalence rates in the population do not increase, the aging of the population itself will cause a 17–31% increase in the number of patients with HF between 2000 and 2020.2,27


The other two likely contributors to the rising prevalence rates are current trends in HF Millions incidence and mortality. As mentioned above, there is no convincing evidence that HF incidence is declining. Furthermore, the most recent data from the Framingham Heart Study and Rochester Epidemiology Project suggest that the long-term survival of HF patients has been improving over the past 2–4 decades.5,7 With no decrease in the number of patients diagnosed each year, and better long-term survival, it is inevitable that prevalence rates will continue to climb.


HOSPITALIZATIONS AND ECONOMIC BURDEN


The increase in HF hospitalizations has been even more dramatic than the rise in the numbers of patients with HF during the past two to three decades (Fig. 2-4). According to data from the National Hospital Discharge Survey, there was a 189% increase in the number of patients hospitalized with HF as the primary discharge diagnosis between 1979 and 1999, specifically, from 377,000 to 1,088,349.1,28 Most of the increase, once again, has been among the elderly (Fig. 2-5). This is directly translated to considerable economic impact on society; in 2005, the American Heart Association estimated that the direct cost of HF will be >25 billion dollars, with hospital charges accounting for nearly 60% of this cost.1


A substantial proportion of HF-related hospitalizations is accounted for by readmissions of patients previously hospitalized with HF. In one study, 44% of patients hospitalized with a primary discharge diagnosis of HF were readmitted within 6 months.29 Economic analysis of HF-related admissions showed that the average cost of repeat hospitalization was in excess of $7000 per patient.30 There is also evidence that HF-related hospital readmissions are on the rise, possibly due to shorter hospital length of stay. The studies of Medicare beneficiaries with HF show a 9% increase in the odds of hospital readmission at 30 days during the period of 1993–1999.31


Although European countries have experienced a similar epidemic increase in HF-related hospitalizations during the late 1980s and early 1990s, recent data from Scotland suggest that the number of hospitalizations peaked in 1993–1994 and leveled off during 1995–1996 among both women and men (Fig. 2-6).32 Similar observations were made in Canada, where the number of HF hospitalizations decreased by 7% between 1994 and 1995 and 1999 and 2000.33 There is evidence that a similar plateau may be occurring in the United States. The National Hospital Discharge Survey indicates that the number of hospitalizations peaked in the late 1990s, and decreased to 970,000 in 2002.1 Although the cause of this plateau is not well understood, it is possible that recent advances in HF management, including pharmacologic and disease management interventions, have decreased the number of preventable HF hospitalizations.


Nevertheless, projections for trends in HF hospitalizations and associated cost are bleak. Studies from Scotland predict a 12–34% increase in HF hospitalizations by the year 2020 and in Canada the rate is expected to double by the year 2025.2,34 Data from Canada suggest that to keep the number of new HF hospitalizations at the current level, the incidence of HF will need to decrease by 2.6% yearly.34


OUTCOMES


Multiple studies have documented that despite recent advances in HF management, short- and long-term mortality rates remain alarmingly high. Data from the Framingham Heart Study show that age-adjusted 30-day, 1-year, and 5-year mortality in HF patients from 1990 to 1999 were 10–13%, 24–28%, and 45–59% respectively.5 A study from the Rochester Epidemiology Project reported lower 30-day (4–6%) and 6-month (17–21%), but similar 5-year (46–50%) mortality during 1996–2000.7 The mortality estimates from administrative database studies are even more staggering. A recent study of Medicare beneficiaries hospitalized with HF demonstrated 1-year HF mortality of nearly 32%,31 while a report from the National Health Service in Scotland showed that the overall case-fatality rate among both inpatients and outpatients with HF during 1986–1995 was as high as 44% at 1 year and 76% at 5 years.35


Several recent reports have suggested that longer-term HF mortality has been improving over the past several decades. Levy and others demonstrated that 5-year mortality has declined from 70% to 59% in men, and from 57% to 45% in women between the periods of 1950–1969 and 1990–1999, an overall relative risk decrease of 32% (Table 2-2).5 Considerable improvement in long-term HF survival was also seen in the Rochester Epidemiology Project, with a 28–52% decrease in the relative risk of long-term mortality among men between the periods of 1979–1984 and 1996–2000.7 The relative risk reduction among women was more modest (6–33%) and not statistically significant in women aged >80 years.7 Finally, a study of the entire HF patient population in Scotland showed that median survival improved from 1.23 to 1.64 years from 1986 to 1995.35


Interestingly, this improvement in long-term survival is likely contributing to the rising HF prevalence and hospitalization rates. As patients with HF live longer, and HF incidence does not decline, the total number of HF patients will rise and utilization of health-care services by these patients will be increasing. Scientific simulations and forecasts clearly show that increasing health-care utilization in the face of decreasing mortality is not a paradox, but in fact an anticipated tradeoff of lower mortality for higher morbidity, which is further compounded by the aging of the American population.27

All values were adjusted for age (<55, 55–64, 65–74, 75–84, and >85 years). Source: Reprinted with permission from Levy et al. N Engl J Med. 2002;347:1397–402, Massachusetts Medical Society.


Whether long-term HF mortality has continued to improve during the 1990s is less clear. Although one study of Medicare beneficiaries in Northeast Ohio showed a 14.6% relative risk decline in 1-year mortality during 1991–1997, other studies have been less convincing.36 Although analysis of administrative data from Ontario, Canada, showed an overall trend towards lower 1-year mortality during 1992–2000, examination of the crude data from that study suggests that most of the improvement occurred between 1992 and 1993, with minimal change thereafter.37 The study of temporal mortality trends among Medicare beneficiaries during the 1990s similarly showed a decline in 1-year mortality between 1993 and 1994, but suggested no subsequent improvement after 1994.31 No statistically significant difference in 1-year case fatality rates could be found in the study of the Henry Ford Health System patients with HF from 1989 to 1999.6


Data on temporal trends in short-term HF mortality are also mixed. The above mentioned study of HF patients in Scotland showed a 17–26% relative risk reduction in 30-day HF mortality from 1986 to 1995, and the study of Medicare patients in Northeast Ohio demonstrated a 12% relative risk reduction in 30-day mortality from 1991–1997.35,36 However, recent studies of very large patient populations in Canada and the United States did not show similar 30-day mortality improvements. The study of 77,421 patients from Ontario, Canada, did not show any trend towards better 30-day mortality either among all patients or hospitalization survivors during 1992–2000.37 Recent analysis of 3,957,520 Medicare beneficiaries hospitalized with the primary discharge diagnosis of HF showed that there was no significant change in risk-adjusted 30-day mortality between 1993 and 1999.31 Although most studies show a clear and significant decrease in in-hospital mortality during the 1990s (nearly 50% relative risk reduction between 1991 and 1997 in one study), this appears to be mostly explained by a dramatic decrease in the mean length of stay among the hospitalized HF patients during the same time period.36 In fact, it appears that mortality immediately following hospital discharge (postdischarge mortality) increased from 1991 to 1997, suggesting that although fewer patients die during hospitalizations, more HF patients die out of hospital with no overall change in short-term mortality (Fig. 2-7).36


Overall, it appears that although the outcomes of HF patients improved over the past several decades, recent progress has been modest at best. This is somewhat paradoxical, as some of the most dramatic advances in HF management took place during the 1990s. There are several likely explanations. The vast and rising majority of HF patients are elderly. Less than 20% of elderly HF patients fit enrollment criteria for angiotensin-converting enzyme (ACE) inhibitor and b-blocker clinical trials.38 Due to their high burden of comorbid illness, many elderly HF patients have contraindications to these therapies. In addition, more than half have HF with preserved left ventricular systolic function, a condition in which the efficacy of ACE inhibitors and b-blockers has not been established.39


Recent studies of Medicare beneficiaries hospitalized with HF demonstrate that there is underuse and misuse of evidence-based therapies. Among those patients who were “ideal candidates” for ACE inhibitor and spironolactone therapies, <70% and 25%, respectively, were treated.40 Data from the Cardiovascular Health Study showed that <20% of elderly patients with HF were being treated with both an ACE inhibitor and b-blocker during the 1990s.41 Further, there is evidence that some therapies may be prescribed to patients at high risk for medication side effects. For example, nearly 31% of spironolactone prescriptions in the Medicare population were given to patients who had a major contraindication to that drug.42


Another possible explanation for the apparent lack of major improvement in outcomes over the past decade is that recent advances in HF management are preventing HF hospitalizations, thus resulting in an overall “sicker” population of HF patients being hospitalized over time. Since studies based on administrative data typically focus on hospitalized HF patients, apparent lack of improvement in HF outcomes could be partially due to this selection bias.


CONCLUSION


Given the aging of the American population, no clear change in HF incidence, and improving long-term survival of patients with established HF over the last 20–30 years, the overall number of patients with HF will likely continue to increase substantially in the near future. This may have a profound effect on both the number of HF-related hospitalizations and the economic impact on the health-care system. Although further innovations in HF management and cost-saving interventions such as disease management are important, better primary HF prevention appears to be the critical factor in stemming the current HF epidemic. This will undoubtedly involve better detection and treatment of key HF risk factors, including hypertension, coronary artery disease, obesity, diabetes, and valvular disease.


REFERENCES


1. American Heart Association. Heart Disease and Stroke Statistics––2007 Update. Available at: www.americanheart.org/presenter.jhtml?identi fier=3000090. Accessed April 6, 2007.

1. Stewart S, MacIntyre K, Capewell S, et al. Heart failure and the aging population: an increasing burden in the 21st century? Heart (British Cardiac Society). 2003;89:49–53.

2. Khand AU, Shaw M, Gemmel I, et al. Do discharge codes underestimate hospitalisation due to heart failure: validation study of hospital discharge coding for heart failure. Eur J Heart Fail. 2005;7:792–797.

3. Lloyd-Jones DM, Larson MG, Leip EP, et al. Lifetime risk for developing congestive heart failure: the Framingham Heart Study. Circulation. 2002;106:3068–3072.

4. Levy D, Kenchaiah S, Larson MG, et al. Long-term trends in the incidence of and survival with heart failure. N Engl J Med. 2002;347:1397–1402.

5. McCullough PA, Philbin EF, Spertus JA, et al. Confirmation of a heart failure epidemic: findings from the Resource Utilization Among Congestive Heart Failure (REACH) study. J Am Coll Cardiol. 2002;39:60–69.

6. Roger VL, Weston SA, Redfield MM, et al. Trends in heart failure incidence and survival in a community-based population. JAMA. 2004;292:344–350.

7. Hajjar I, Kotchen TA. Trends in prevalence, awareness, treatment, and control of hypertension in the United States, 1988–2000. JAMA. 2003;290:199–206.

8. Levy D, Larson MG, Vasan RS, et al. The progression from hypertension to congestive heart failure. JAMA. 1996;275:1557–1562.

9. Goldberg RJ, Yarzebski J, Lessard D, et al. A two-decades (1975 to 1995) long experience in the incidence, in-hospital and long-term case-fatality rates of acute myocardial infarction: a community-wide perspective. J Am Coll Cardiol. 1999;33: 1533–1539.

10. Arciero TJ, Jacobsen SJ, Reeder GS, et al. Temporal trends in the incidence of coronary disease. Am J Med. 2004;117:228–233.

11. Hellermann JP, Goraya TY, Jacobsen SJ, et al. Incidence of heart failure after myocardial infarction: is it changing over time? Am J Epidemiol. 2003;157:1101–1107.

12. Psaty BM, Manolio TA, Smith NL, et al. Time trends in high blood pressure control and the use of antihypertensive medications in older adults: the Cardiovascular Health Study. Arch Intern Med. 202;162:2325–2332.

1. Kenchaiah S, Evans JC, Levy D, et al. Obesity and the risk of heart failure. N Engl J Med. 2002; 347:305–313.

2. Mokdad AH, Bowman BA, Ford ES, et al. The continuing epidemics of obesity and diabetes in the United States. JAMA. 2001;286:1195–1200.

3. Mokdad AH, Ford ES, Bowman BA, et al. Prevalence of obesity, diabetes, and obesity-related health risk factors, 2001. JAMA. 2003;289: 76–79.

4. Mokdad AH, Serdula MK, Dietz WH, et al. The spread of the obesity epidemic in the United States, 1991–1998. JAMA. 1999;282:1519–1522.

5. Mokdad AH, Serdula MK, Dietz WH, et al. The continuing epidemic of obesity in the United States. JAMA. 2000;284:1650–1651.

6. Ford ES, Mokdad AH, Giles WH. Trends in waist circumference among U.S. adults. Obes Res. 2003; 11:1223–1231.

7. Mokdad AH, Ford ES, Bowman BA, et al. Diabetes trends in the U.S.: 1990–1998. Diabetes Care. 2000;23:1278–1283.

8. He J, Ogden LG, Bazzano LA, et al. Risk factors for congestive heart failure in US men and women: NHANES I epidemiologic follow-up study. Arch Intern Med. 2001;161:996–1002.

9. Saydah SH, Fradkin J, Cowie CC. Poor control of risk factors for vascular disease among adults with previously diagnosed diabetes. JAMA. 2004;291:335–342.

10. Bleumink GS, Knetsch AM, et al. Quantifying the heart failure epidemic: prevalence, incidence rate, lifetime risk and prognosis of heart failure: the Rotterdam Study. Eur Heart J. 2004;25: 1614–1619.

11. Kitzman DW, Gardin JM, Gottdiener JS, CHS Research Group. Cardiovascular Health Study. Importance of heart failure with preserved systolic function in patients > or = 65 years of age. Am J Cardiol. 2001;87:413–419.

12. United States Department of Commerce, Economics and Statistics Administration. Bureau of the Census. The 65 years and over population: 2000. Census 2000 brief. Available at: http://www.census.gov/prod/2001pubs/c2kbr0 1-10.pdf. Accessed October 14, 2005.

13. United States Department of Commerce, Economics and Statistics Administration. Bureau of the Census. Aging of the Americas into the XXI century. Available at: http://www.census.gov/ ipc/prod/ ageame.pdf. Accessed October 14, 2005.