[Home] [Headlines] [Latest Articles] [Latest Comments] [Post] [Mail] [Sign-in] [Setup] [Help] [Register]
|
Status: Not Logged In; Sign In
United States News Title: What Changes In Survival Rates Tell Us About US Health Care Many advocates of US health reform point to the nation’s relatively low life-expectancy rankings as evidence that the health care system is performing poorly. Others say that poor US health outcomes are largely due not to health care but to high rates of smoking, obesity, traffic fatalities, and homicides. We used cross-national data on the fifteen-year survival of men and women over three decades to examine the validity of these arguments. We found that the risk profiles of Americans generally improved relative to those for citizens of many other nations, but Americans’ relative fifteen-year survival has nevertheless been declining. For example, by 2005, fifteen-year survival rates for forty-five-year-old US white women were lower than in twelve comparison countries with populations of at least seven million and per capita gross domestic product (GDP) of at least 60 percent of US per capita GDP in 1975. The findings undercut critics who might argue that the US health care system is not in need of major changes. In 1950, the United States was fifth among the leading industrialized nations with respect to female life expectancy at birth, surpassed only by Sweden, Norway, Australia, and the Netherlands.1 The last available measure of female life expectancy had the United States ranked at forty-sixth in the world.2 As of September 23, 2010, the United States ranked forty-ninth for both male and female life expectancy combined.3,4 The United States does little better in international comparisons of mortality. Americans live 5.7 fewer years of "perfect health"—a measure adjusted for time spent ill—than the Japanese.5 Meanwhile, per capita health spending in the United States increased at nearly twice the rate in other wealthy nations between 1970 and 2002.6 As a result, the United States now spends well over twice the median expenditure of industrialized nations on health care, and far more than any other country as a percentage of its gross domestic product (GDP).7 The observation that Americans are spending relatively more on health but living relatively shorter, less healthy lives has led some critics to allege that the US health care system is "uniquely inefficient."8 Although these statistics appear to suggest that the US system performs poorly, some policy experts claimed that the system was not to blame for this apparently dismal picture. They also quibbled with the data, arguing that life expectancy figures could be biased by variation in the way births are coded or by differences in reproductive health policy.9,10 Whether a fetus or newborn infant who dies is considered a case of infant mortality or a fetal demise can very from nation to nation. Thus, the United States might count some cases as an infant death, and other countries might not. Critics who were skeptical about health reform also questioned the validity of international comparisons of existing health outcomes data. They argued that variations in international outcomes might result from differences in demographics or population health risk factors.11,12 And given Americans’ expanding waistlines, poor health habits, car culture, and homicide rates, the critics argued, the United States might rank worse without its costly health system.10 Is the US health system at least partly to blame for this deterioration in international rankings for life expectancy and medical costs? Or can the declines be better explained by statistical, demographic, behavioral, and social factors? In this paper we explore changes in fifteen-year survival at middle and older ages, alongside per capita health care spending, in the United States and twelve other wealthy nations. We then examine the extent to which the survival and cost variations over time among these nations can be explained by demographics, obesity, smoking, or mortality events that are not closely related to health care, such as traffic accidents and homicide. By comparing health system costs and mortality rates over time, it is possible to assess whether trends in risk factors for health or causes of death can explain the observed relative decline in broad health outcomes among American men and women over the past thirty years. Comparison Countries We compared the performance of the United States to that of twelve nations that have populations of at least seven million and per capita GDP of at least 60 percent of the US per capita GDP since 1975. These nations are Australia, Austria, Belgium, Canada, France, Germany, Italy, Japan, the Netherlands, Sweden, Switzerland, and the United Kingdom. These comparison countries all provide universal health insurance coverage. However, they have a fairly diverse set of health care systems, ranging from care that is organized and paid for by the government to regulated private insurance markets.13 Note that data on all indicators are not available for all comparison countries at all points in time. Outcomes We examined relative changes in two measures of system performance—cost and fifteen- year survival—from 1975 to 2005, a period for which good comparative data are available. We focused on forty-five- and sixty-five-year-old men and women and measured costs as health care spending per capita, adjusted to constant US dollars using purchasing power parity. We chose to study adult survival, rather than measures at birth, because variations in family planning policy and coding at birth may affect life expectancy at birth. Most of the important chronic diseases that can be successfully prevented or treated with health interventions—such as cardiovascular disease, stroke, and diabetes— arise in midlife, and deaths due to these diseases cluster in the second half of life.14 We measured fifteen-year survival rather than life expectancy because the latter can be biased by the survival experiences of a small number of elderly people, among whom coding errors are common.15 Focusing on survival also allowed us to distinguish between the experiences of specific cohorts. We explored fifteen-year survival for men and women separately because risk-factor profiles differ greatly by sex and country.16 Data Sources We obtained cross-national data on obesity and health system costs from the Organisation for Economic Co-operation and Development (OECD),17 which in turn collected the data from each of the countries. In many cases, the survey year varied slightly by country. In other cases, there were changes in the national survey data reported to the OECD over time. For instance, in Australia, earlier data came from the Risk Factor Prevalence Survey of the National Heart Foundation, and later data from the National Nutrition Survey of the Australian Bureau of Statistics. Overall survival data were supplied by member countries to the World Health Organization (WHO), from which we obtained the information. Data for US non-Hispanic whites came from the Centers for Disease Control and Prevention (CDC), which is the original source of the US data reported to the WHO. The CDC database contains race-specific data, while the WHO information does not.18 In our analysis of deaths by cause, we also used data from the CDC.19 Finally, we used the International Mortality and Smoking Statistics (IMASS) software, version 4.04, to examine smoking data by period and birth cohort. The software database is populated with international data from the WHO. In these data, we constructed birth cohorts by combining period and age data. Even though the sample differs from period to period, it is nationally representative; thus, different samples may be treated as if they are a single cohort followed over time. We assumed that subjects who were age five in the 1950 sample and age ten in the 1955 sample were drawn from the same longitudinal cohort, when, in fact, they were different nationally representative samples. Although the birth cohort data are not as complete as cross-sectional data across all countries and age groups, they offer the distinct advantage of providing information on the life course of each of our comparison groups by age, sex, and time period. Life-Expectancy Calculations We used Wolfram Alpha—an online "data engine," which allows various types of calculations using rigorous, scientific tools—to calculate survival probabilities for people between ages 45–60 and ages 65–80 in all of the countries under study.20 This engine uses compiled mortality data from the WHO. The methods used for estimating survival probabilities are described in detail elsewhere.21 Briefly, we employed standard life table methods, in which age-specific mortality probabilities were applied to a hypothetical cohort of 100,000 people. Using life-table data, we calculated the fifteen-year survival probability for forty-five-year-olds and sixty-five-year-olds. To hold race constant, we also examined US data for non-Hispanic whites. Because race-specific data were not available from Wolfram Alpha, we used life tables from 1975, 1985, 1995, and 2004—the latest year for which data were available—to generate sex- and age-specific mortality ratios for whites relative to all Americans in the same age groups.18 Below we present temporal trends by sex for the outcomes of interest. Our graphical results present outcomes for 1975 and 2005 only. Limitations Our study used longitudinal data, which helped eliminate idiosyncratic findings and allowed us to observe the temporal relationship between changes in behavioral risk factors, costs, and life expectancy. In addition, we focused on survival for people ages forty-five and sixty-five, which improved the comparability of the data across countries. Nevertheless, our study was subject to a number of important limitations. First, it necessarily relied on diverse sources of data compiled by the OECD. Although it is likely that both random and nonrandom errors were introduced by the use of disparate sources of data, we minimized the range of error by including only larger industrialized nations in our analysis. Second, we did not use comparative data on specific causes of death. Some researchers have argued, for instance, that the United States does much better than other nations in terms of spending on some types of preventive care and cancer care, leading to increased survival.11 Comparative data on specific causes of death over time could provide additional insights into health system performance. However, comparisons over time and across nations are muddied by differences in coding and changes in disease coding over time in some countries but not others, and by the availability of prevalence data only. For instance, a high prevalence of heart disease could reflect either a systematic failure of efforts to prevent heart disease or a highly successful effort to prevent mortality from heart disease. Finally, we examined only relative changes in smoking and obesity over time. There may be other important, but unidentified, sociocultural factors that differentiate the United States from other industrialized nations. One example is social capital, which is hypothesized to contribute to population health through both health-inducing social ties and the greater likelihood of voter support for social programs, such as mass transit and education.22,23 Unfortunately, there are no good international measures of social capital over time, so we could not include this variable in our analysis.
Post Comment Private Reply Ignore Thread Top • Page Up • Full Thread • Page Down • Bottom/Latest
#1. To: go65 (#0)
In the 1950's few women worked. There were few fast food restaurants needed.
-----------------------------------------------------------
There were few fast food restaurants needed. I guess men made up the largest percentage of nurses, maids, laundry workers, seamtresses and school teachers then according to you. Plus seasonal farm workers. White Sands... always showing her foreign ignorance whenever she posts.
Seventy percent of Swedish women work and they rank 6th in life expectancy to our 38th.
yeah, but they are socialists.
yeah, but they are socialists. And eat lots of fish.
LoL - at least they don't eat Lutefisk like their neighbors to the west.
And eat lots of fish. Well they can't have anything to offer.
|
|
[Home] [Headlines] [Latest Articles] [Latest Comments] [Post] [Mail] [Sign-in] [Setup] [Help] [Register]
|