Library Index :: Death and Dying Reference :: The End of Life: Medical Considerations - Causes Of Death, The Study To Understand Prognoses And Preferences For Outcomes And Risks Of Treatments (support)

The End of Life: Medical Considerations - The Study To Understand Prognoses And Preferences For Outcomes And Risks Of Treatments (support)

During the twentieth century in the United States, the process of dying shifted from the familiar surroundings of home to the hospital. While hospitalization ensures that the benefits of modern medicine are readily available, many patients dread leaving the comfort of their homes and losing, to some extent, control over their end-of-life decisions.

Between 1989 and 1994, in an effort to "improve end-of-life decision making and reduce the frequency of a mechanically supported, painful, and prolonged process of dying," a group of investigators from various disciplines undertook the largest study of death and dying ever conducted in the United States. The project, known as the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT), included more than 9,000 patients who suffered from life-threatening illnesses. Patients enrolled in the study had about a 50 percent chance of dying within six months.

The researchers published the results of their study in "A Controlled Trial to Improve Care for Seriously Ill Hospitalized Patients" (Journal of the American Medical Association,

TABLE 4.1
Death rates for the 15 leading causes of death, 2001, and percent change, 2000–2001
[Death rates on an annual basis per 100,000 population; age-adjusted rates per 100,000 U.S. standard population]

Age-adjusted death rate
Percent change Ratio
Rank* Cause of death (Based on the Tenth Revision International Classification of Diseases, 1992) Number Percent of total deaths 2001 crude death rate 2001 2000 to 2001 Male to female Black to white Hispanic to white non-Hispanic
All causes 2,416,425 100.0 848.5 854.5 −1.7 1.4 1.3 0.8
1 Diseases of heart 700,142 29.0 245.8 247.8 −3.8 1.5 1.3 0.8
2 Malignant neoplasms 553,768 22.9 194.4 196.0 −1.8 1.5 1.3 0.7
3 Cerebrovascular diseases 163,538 6.8 57.4 57.9 −4.9 1.0 1.4 0.8
4 Chronic lower respiratory diseases 123,013 5.1 43.2 43.7 −1.1 1.4 0.7 0.4
5 Accidents (unintentional injuries) 101,537 4.2 35.7 35.7 2.3 2.2 1.0 0.8
6 Diabetes mellitus 71,372 3.0 25.1 25.3 1.2 1.2 2.1 1.7
7 Influenza and pneumonia 62,034 2.6 21.8 22.0 −7.2 1.4 1.1 0.9
8 Alzheimer's disease 53,852 2.2 18.9 19.1 5.5 0.8 0.7 0.5
9 Nephritis, nephrotic syndrome and nephrosis 39,480 1.6 13.9 14.0 3.7 1.5 2.4 1.0
10 Septicemia 32,238 1.3 11.3 11.4 0.9 1.2 2.3 0.8
11 Intentional self-harm (suicide) 30,622 1.3 10.8 10.7 2.9 4.6 0.5 0.5
12 Chronic liver disease and cirrhosis 27,035 1.1 9.5 9.5 0 2.1 1.0 1.8
13 Assault (homicide) 20,308 0.8 7.1 7.1 20.3 3.3 4.3 2.1
14 Essential (primary) hypertension and hypertensive disease 19,250 0.8 6.8 6.8 4.6 1.0 2.9 1.1
15 Pneumonitis due to solids and liquids 17,301 0.7 6.1 6.1 0 1.8 1.1 0.7
All other causes 400,935 16.6 140.8
… Category not applicable.
*Rank based on number of deaths.
SOURCE: Elizabeth Arias, et al., "Table C. Percent of Total Deaths, Death Rates, Age-Adjusted Death Rates for 2001, Percent Change in Age-Adjusted Death Rates from 2000 to 2001, and Ratio of Age-Adjusted Death Rates by Race and Sex for the 15 Leading Causes of Death for the Total Population in 2001: United States," in Deaths: Final Data for 2001, National Vital Statistics Reports, National Center for Health Statistics, Hyattsville, MD, vol. 52, no. 3, September 18, 2003

vol. 274, no. 20, November 22/29, 1995). The SUPPORT investigators hypothesized that increased communication between patients and physicians, better understanding of patients' wishes, and the use of computer-based projections of patient survival would result in "earlier treatment decisions, reductions in time spent in undesirable states before death, and reduced resource use."

Phase I of the study was observational. The researchers reviewed patients' medical records and interviewed patients, surrogates (people who make decisions if patients became incompetent), and patients' physicians. Discussions and decisions about life-sustaining measures were observed.

The researchers interviewed patients, families, and surrogates about the patients' thoughts on cardiopulmonary resuscitation (CPR), their perceptions of their quality of life, the frequency and severity of their pain, and their satisfaction with the care provided. The physicians who acknowledged responsibility for the patients' medical decisions were also interviewed to determine their understanding of patients' views on CPR and how patients' wishes influenced their medical care. The surrogates were again interviewed after the patients' deaths.

Problems with End-of-Life Care

Phase I of SUPPORT found a lack of communication between physicians and patients, showed aggressive treatment of dying patients, and revealed a disturbing picture of hospital death. Of the 4,301 patients, 31 percent expressed a desire that CPR be withheld. But only 47 percent of physicians reported knowledge of their patients' wishes. About half (49 percent) of patients who requested not to be resuscitated did not have a Do Not Resuscitate (DNR) order in their medical charts. Of the 79 percent who died with a DNR order, 46 percent of the orders were written within only two days of death.

The patients' final days in the hospital included an average of eight days in "generally undesirable states"—in an intensive care unit (ICU), receiving artificial respiration, or in a coma. More than a third (38 percent) stayed 10 days in the ICU, while almost half (46 percent) were mechanically ventilated within three days prior to death. Surrogates reported that 50 percent of conscious patients complained of moderate or severe pain at least half the time in their last three days.

Phase II: Intervention Fails to Improve Care

Phase II of SUPPORT, the intervention phase, was implemented to address the shortcomings documented in Phase I. It lasted another two years, and involved patient participants with characteristics similar to those in Phase I. This time, however, the doctors were given printed reports about the patients and their wishes regarding life-sustaining

TABLE 4.2
Death rates, by age, for the 15 leading causes of death, 1999–2001
[Rates on an annual basis per 100,000 population in specified group; age-adjusted rates per 100,000 U.S. standard population. Rates are based on populations enumerated as of April 1 for 2000 and estimated as of July 1 for all other years.]

Age
Cause of death (Based on the Tenth Revision, International Classification of Diseases, 1992) and year All ages1 Under 1 year2 1–4 years 5–14 years 15–24 years 25–34 years 35–44 years 45–54 years 55–64 years 65–74 years 75–84 years 85 years and over Age adjusted rate
All causes
2001 848.5 683.4 33.3 17.3 80.7 105.2 203.6 428.9 964.6 2,353.3 5,582.4 15,112.8 854.5
2000 854.0 736.7 32.4 18.0 79.9 101.4 198.9 425.6 992.2 2,399.1 5,666.5 15,524.4 869.0
1999 857.0 736.0 34.2 18.6 79.3 102.2 198.0 418.2 1,005.0 2,457.3 5,714.5 15,554.6 875.6
Diseases of heart
2001 245.8 11.9 1.5 0.7 2.5 8.0 29.6 92.9 246.9 635.1 1,725.7 5,664.2 247.8
2000 252.6 13.0 1.2 0.7 2.6 7.4 29.2 94.2 261.2 665.6 1,780.3 5,926.1 257.6
1999 259.9 13.8 1.2 0.7 2.8 7.6 30.2 95.7 269.9 701.7 1,849.9 6,063.0 266.5
Malignant neoplasms
2001 194.4 1.6 2.7 2.5 4.3 10.1 36.8 126.5 356.5 802.8 1,315.8 1,765.6 196.0
2000 196.5 2.4 2.7 2.5 4.4 9.8 36.6 127.5 366.7 816.3 1,335.6 1,819.4 199.6
1999 197.0 1.8 2.7 2.5 4.5 10.0 37.1 127.6 374.6 827.1 1,331.5 1,805.8 200.8
Cerebrovascular diseases
2001 57.4 2.7 0.4 0.2 0.5 1.5 5.5 15.1 38.0 123.4 443.9 1,500.2 57.9
2000 59.6 3.3 0.3 0.2 0.5 1.5 5.8 16.0 41.0 128.6 461.3 1,589.2 60.9
1999 60.0 2.7 0.3 0.2 0.5 1.4 5.7 15.2 40.6 130.8 469.8 1,614.8 61.6
Chronic lower respiratory diseases
2001 43.2 1.0 0.3 0.3 0.4 0.7 2.2 8.5 44.1 167.9 379.8 644.7 43.7
2000 43.4 0.9 0.3 0.3 0.5 0.7 2.1 8.6 44.2 169.4 386.1 648.6 44.2
1999 44.5 0.9 0.4 0.3 0.5 0.8 2.0 8.5 47.5 177.2 397.8 646.0 45.4
Accidents (unintentional injuries)
2001 35.7 24.2 11.2 6.9 36.1 29.9 35.4 34.1 30.3 42.8 100.9 276.4 35.7
2000 34.8 23.1 11.9 7.3 36.0 29.5 34.1 32.6 30.9 41.9 95.1 273.5 34.9
1999 35.1 22.3 12.4 7.6 35.3 29.6 33.8 31.8 30.6 44.6 100.5 282.4 35.3
Diabetes mellitus
2001 25.1 * * 0.1 0.4 1.5 4.3 13.6 37.8 91.4 181.4 321.8 25.3
2000 24.6 * * 0.1 0.4 1.6 4.3 13.1 37.8 90.7 179.5 319.7 25.0
1999 24.5 * * 0.1 0.4 1.4 4.3 12.9 38.3 91.8 178.0 317.2 25.0
Influenza and pneumonia
2001 21.8 7.4 0.7 0.2 0.5 0.9 2.2 4.6 10.7 36.3 148.5 685.6 22.0
2000 23.2 7.6 0.7 0.2 0.5 0.9 2.4 4.7 11.9 39.1 160.3 744.1 23.7
1999 22.8 8.4 0.8 0.2 0.5 0.8 2.4 4.6 11.0 37.2 157.0 751.8 23.5
Alzheimer's disease
2001 18.9 * * * * * * 0.2 2.1 18.7 147.5 710.3 19.1
2000 17.6 * * * * * * 0.2 2.0 18.7 139.6 667.7 18.1
1999 16.0 * * * * * * 0.2 1.9 17.4 129.5 601.3 16.5
Nephritis, nephrotic syndrome and nephrosis
2001 13.9 3.3 * * 0.2 0.6 1.7 4.6 13.0 40.2 104.2 287.7 14.0
2000 13.2 4.3 * 0.1 0.2 0.6 1.6 4.4 12.8 38.0 100.8 277.8 13.5
1999 12.7 4.4 * 0.1 0.2 0.6 1.6 4.0 12.0 37.1 97.6 268.9 13.0
Septicemia
2001 11.3 7.7 0.7 0.2 0.3 0.7 1.8 5.0 12.3 32.8 82.3 205.9 11.4
2000 11.1 7.2 0.6 0.2 0.3 0.7 1.9 4.9 11.9 31.0 80.4 215.7 11.3
1999 11.0 7.5 0.6 0.2 0.3 0.7 1.8 4.6 11.4 31.2 79.4 220.7 11.3

TABLE 4.2
Death rates, by age, for the 15 leading causes of death, 1999–2001
[Rates on an annual basis per 100,000 population in specified group; age-adjusted rates per 100,000 U.S. standard population. Rates are based on populations enumerated as of April 1 for 2000 and estimated as of July 1 for all other years.]

Age
Cause of death (Based on the Tenth Revision, International Classification of Diseases, 1992) and year All ages1 Under 1 year2 1–4 years 5–14 years 15–24 years 25–34 years 35–44 years 45–54 years 55–64 years 65–74 years 75–84 years 85 years and over Age adjusted rate
Intentional self-harm (suicide)
20013 10.8 0.7 9.9 12.8 14.7 15.2 13.1 13.3 17.4 17.5 10.7
2000 10.4 0.7 10.2 12.0 14.5 14.4 12.1 12.5 17.6 19.6 10.4
1999 10.5 0.6 10.1 12.7 14.3 13.9 12.2 13.4 18.1 19.3 10.5
Chronic liver disease and cirrhosis
2001 9.5 * * * 0.1 1.0 7.4 18.5 22.7 30.0 30.2 22.2 9.5
2000 9.4 * * * 0.1 1.0 7.5 17.7 23.8 29.8 31.0 23.1 9.5
1999 9.4 * * * 0.1 1.0 7.3 17.4 23.7 30.6 31.9 23.2 9.6
Assault (homicide)
20013 7.1 8.2 2.7 0.8 13.3 13.1 9.5 6.3 4.0 2.9 2.5 2.4 7.1
2000 6.0 9.2 2.3 0.9 12.6 10.4 7.1 4.7 3.0 2.4 2.4 2.4 5.9
1999 6.1 8.7 2.5 1.1 12.9 10.5 7.1 4.6 3.0 2.6 2.5 2.4 6.0
Essential (primary) hypertension and hypertensive renal disease
2001 6.8 * * * 0.1 0.3 0.7 2.4 5.8 15.5 47.7 171.9 6.8
2000 6.4 * * * * 0.2 0.8 2.3 5.9 15.1 45.5 162.9 6.5
1999 6.1 * * * * 0.2 0.7 2.2 5.5 15.2 43.6 152.1 6.2
Pneumonitis due to solids and liquids
2001 6.1 * * * 0.1 0.2 0.4 1.0 2.6 10.0 45.8 189.4 6.1
2000 5.9 * * * 0.1 0.2 0.4 1.0 2.5 10.3 44.5 187.6 6.1
1999 5.5 * * * 0.1 0.2 0.4 0.8 2.5 9.5 41.1 175.6 5.6
*Figure does not meet standards of reliability or precision.
… Category not applicable.
1Figures for age not stated included in "All ages" but not distributed among age groups.
2Death rates for "Under 1 year" (based on population estimates) differ from infant mortality rates (based on live births).
3Figures include September 11, 2001 related deaths for which death certificates were filed as of 10/24/02.
SOURCE: Elizabeth Arias, et al., "Table 9. Death Rates by Age and Age-Adjusted Death Rates for the 15 Leading Causes of Death in 2001: United States, 1999-2001," in Deaths: Final Data for 2001, National Vital Statistics Reports, National Center for Health Statistics, Hyattsville, MD, vol. 52, no. 3, September 18, 2003

TABLE 4.3
Estimated numbers of deaths of persons with AIDS, by year of death and selected characteristics, 1998–2002

Year of death
1998 1999 2000 2001 2002 Cumulative through 20021
Age at death (years)
< 13 104 102 51 49 33 5,071
13–14 9 19 10 5 10 244
15–24 258 232 206 261 190 9,507
25–34 3,785 3,252 2,765 2,377 1,971 139,977
35–44 7,991 7,679 6,998 7,077 6,401 207,324
45–54 4,784 5,004 5,082 5,202 5,395 97,027
55–64 1,511 1,546 1,584 1,758 1,728 31,179
≥65 562 622 652 673 641 11,340
Race/ethnicity
White, not Hispanic 6,228 5,800 5,331 5,061 4,555 223,623
Black, not Hispanic 9,116 9,097 8,723 8,915 8,566 185,080
Hispanic 3,449 3,353 3,118 3,236 3,056 87,888
Asian/Pacific Islander 125 116 107 111 93 3,350
American Indian/Alaska Native 79 79 61 70 72 1,424
Exposure category
Male adult or adolescent
Male-to-male sexual contact 7,120 6,615 6,098 5,971 5,418 249,198
Injection drug use 4,735 4,501 4,145 4,129 4,038 103,714
Male-to-male sexual contact and injection drug use 1,338 1,319 1,287 1,262 1,082 36,224
Heterosexual contact 1,300 1,358 1,363 1,462 1,384 20,820
Other2 222 203 210 177 160 9,797
Subtotal 14,715 13,997 13,104 13,001 12,083 419,754
Female adult or adolescent
Injection drug use 1,970 2,083 1,936 1,947 1,933 37,758
Heterosexual contact 2,100 2,140 2,134 2,280 2,197 34,661
Other2 97 104 99 101 95 4,087
Subtotal 4,167 4,327 4,169 4,328 4,226 76,507
Child(<13 years)
Perinatal 117 122 71 69 59 4,882
Other3 6 8 4 4 4 526
Subtotal 122 130 75 73 62 5,407
Region of residence
Northeast 5,681 5,814 5,089 5,263 5,419 161,971
Midwest 1,936 1,700 1,659 1,697 1,315 48,415
South 7,515 7,263 7,289 7,201 6,671 168,670
West 3,092 2,967 2,624 2,490 2,259 104,559
U.S. dependencies, possessions, and associated nations 781 710 687 752 707 18,054
Total4 19,005 18,454 17,347 17,402 16,371 501,669
Note: These numbers do not represent actual cases in persons who died with AIDS. Rather, these numbers are point estimates of cases in persons who died with AIDS that have been adjusted for delays in reporting of deaths and for redistribution of cases in persons initially reported without an identified risk. The estimates have not been adjusted for incomplete reporting.
1Includes persons who died with AIDS, from the beginning of the epidemic through 2002.
2Includes hemophilia, blood transfusion, perinatal, and risk not reported or not identified.
3Includes hemophilia, blood transfusion, and risk not reported or not identified.
4Includes persons of unknown or multiple race and of unknown sex. Cumulative total includes 304 persons of unknown or multiple race and 1 person of unknown sex. Because column totals were calculated independently of the values for the subpopulations, the values in each column may not sum to the column total.
SOURCE: "Table 7. Estimated Numbers of Deaths of Persons with AIDS, by Year of Death and Selected Characteristics, 1998–2002—United States," in HIV/AIDS Surveillance Report, Centers for Disease Control and Prevention, Atlanta, GA, vol. 14, October 27, 2003

treatments. SUPPORT nurses facilitated the flow of information among patients, families, and health care personnel, and helped manage patients' pain. The SUPPORT investigators were appalled that no improvement in the quality of hospital death occurred.

What Went Wrong?

Dr. Bernard Lo ("Improving Care Near the End of Life: Why Is It So Hard?," Journal of the American Medical Association, vol. 274, no. 20, November 22/29, 1995) believes that the results reported in the SUPPORT study raise more questions than answers. Among other issues, Dr. Lo claims that while Phase I showed poor doctor-patient communication, Phase II, instead of directly addressing this shortcoming, added a third party, the SUPPORT nurses, to do the physicians' job. The issue of inadequate pain control was not resolved, in part because the physicians were not routinely advised of the patients' pain.

Dr. Lo questioned the physicians' apparent lack of interest in patients' preferences regarding CPR. Did the physicians think they knew what was best for the patients and, therefore, ignored information relayed by the nurses? Furthermore, no comparison was ever made between patients' perceived prognosis of their own condition and that made by their physicians. Patients who overestimated the likelihood of recovery might have requested life-sustaining measures, which their doctors might not have deemed appropriate.

Daniel Callahan, former president of the Hastings Center, a think tank for biomedical ethics, questions why the idea of patient empowerment, which includes issues that Americans love—self-determination, choice, patients' rights, the provision of timely scientific knowledge, a legal domestication of death—doesn't appear to work. Callahan believes Americans continue to be ambivalent in their attitude toward death ("Once Again, Reality: The Lessons of the SUPPORT Study," Special Supplement, Hastings Center Report, vol. 25, no. 6, 1995). Medical professionals and laypersons alike are often unsure whether to accept death as a part of life or see it as some kind of misplaced biological occurrence to be fought. In the SUPPORT study, this ambivalence was depicted in physicians' management of terminal care and even in the actions of some patients and their families.

Callahan points out that SUPPORT illustrated the fact that a number of variables influence end-of-life decisions. Some of these are:

  • Patients change their minds when it comes to life-sustaining treatments. Some do not really know what they want or what may happen in the course of their illness.
  • Doctors may have differing opinions about prognosis, what specific treatments would accomplish, or how to balance their professional judgment and patient preferences.
  • Families do not always know what should be done and can be uncertain in interpreting patients' preferences.

Stella Reiter-Theil noted that the SUPPORT study provided evidence regarding an additional variable that influences end-of-life decisions: ageism ("The Ethics of End-of-Life Decision in the Elderly: Deliberations from the ECOPE Study," Best Practice and Research Clinical Anaesthesiology, vol. 17, no. 2, 2003). Reiter-Theil questions whether age is an appropriate criterion by which to allocate health care resources, and she suggests that decisions to provide resources and treatment to patients should be based on their ability to benefit from treatment and not on their age. Yet physicians in the SUPPORT study were less likely to think that older patients wanted care to extend their lives, and were more likely to think that younger patients wanted such care, when this was not necessarily the case.

User Comments Add a comment…