Townsend Letter The Examiner of Alternative Medicine
Alternative Medicine Conference Calendar
Check recent tables of contents

From the Townsend Letter
August / September 2010

The United States Program on Cancer, 1975–2006: A Failure (Part 1)
by Anthony D. Apostolides, PhD, and Ipatia K. Apostolides, BA

Search this site

 

Full article: Online publication only
Part 2 also online

Introduction
It seems to us that we are living in a cancer epidemic in the US. Both of us have lost parents, relatives, and friends to cancer; and people we know have lost parents and friends to the disease. Based on our own experience, we were driven to carry out research on this matter and examine the cancer program and its effects on people of the US. This article is the result of that research; hopefully, it contributes toward informing American citizens and policy makers on the outcomes of the cancer program and possible changes.

In this study, we assess the effectiveness of the program against cancer as guided by the National Cancer Institute (NCI). The period of analysis ranges from 1975 to 2006. Such a long period enables one: (1) to show the long-term, accurate trends in the disease, and thus (2) to appraise the effectiveness of the program of the US against the disease.

The criteria used in this study to rate the effectiveness of the NCI program are incidence rate, mortality rate, and probability. If the incidence rate increases over time, the NCI program to prevent cancer is shown to be ineffective. If the incidence rate decreases, the prevention program is shown to be effective. If the mortality rate increases over time, the NCI program to treat cancer is shown to be failing. If the mortality rate decreases over time, the program is shown to succeed.

The third measure used in this assessment is the probability facing Americans of getting cancer in their lifetimes. This statistic is tied to the prevention side of the cancer program; so if the probability increases, the NCI program is ineffective and if the probability decreases, the program is effective.  
 
Data
Period of Analysis and Data
Our analysis covers the period 1975–2006. The assessment based on such a long time period provides results that are more comprehensive and, thus, more reliable as compared to a smaller interval. This is particularly relevant for cancer, which can take years to appear after a person's exposure to carcinogens.

The US does not have a national database on cancer incidence rates or mortality rates. The data for the analysis were primarily obtained from the SEER-9 (Surveillance Epidemiology and End Results) database of NCI. SEER-9 provides data that are obtained from nine geographic areas: five states (Connecticut, Hawaii, Iowa, Utah, and New Mexico) and four metropolitan areas (Detroit, Atlanta, Seattle, and San Francisco). This geographic group constitutes approximately 10% of the US population.

 The US cancer mortality data are obtained by NCI from the Centers for Disease Control and Prevention (CDC), and are incorporated into the SEER database. The CDC has developed national cancer mortality rates for the period 1979–2006 and they include actual (referred to as "crude" in official statistics) and age-adjusted rates. Data on the US population were obtained from the US Bureau of the Census, Statistical Abstract of the United States.


SEER provides cancer data in age-adjusted form. Those data are calculated numbers of incidence and mortality by holding constant the age distribution of the US population to that of a given year (presently it is 2000).Thus, an incidence rate of 200 for a particular year means that 200 people per 100,000 population were expected to have been diagnosed with the disease if the population of the US was the same as in year 2000.

Shortcomings of Existing Cancer Data
There are a number of shortcomings with the existing official cancer data on the incidence and mortality rates, which are discussed in Appendix A. The following issues are covered there: (1) SEER-9 reports only age-adjusted data and not actual/crude data. However, crude mortality rates from CDC show a different picture than age-adjusted rates. (2) The age-adjusted mortality rates of SEER-9 are lower than the age-adjusted rates of CDC. (3) Mortality statistics are not adjusted by CDC and SEER-9 for possible mistakes/updates. (4) Implications of age-adjusted cancer rates; and (5) Delays in the reporting of cancer incidence counts.


Cancer Incidence

Cancer Incidence Rates
Table 1 presents data and tabulations on the annual incidence rate of cancer, for all invasive cancers, in the US for 1975 to 2006. The alarming observation is that the incidence rate of cancer in 2006 was substantially higher than in 1975. In 1975, 400 people per 100,000 of population were afflicted with cancer; and in 2006 – 31 years later – the cancer incidence had worsened significantly to 456 people being afflicted per 100,000 population.

From 1975 to 1992, the cancer incidence rate jumped from 400 to 510 per 100,000. After 1992, the incidence rate declined somewhat; however, in 2006, that rate was still significantly higher than in 1975.

Table 1: Cancer Incidence in the US, 1975–2006

Year

Cancer Incidence Rate (age-adjusted)

Population – US (July 1)

Number of People Afflicted by Cancer

(1)

(2)

(3)

1975

400.4

215,973,000

864,756

1976

407.4

218,035,000

888,275

1977

407.7

220,239,000

897,914

1978

407.3

222,585,000

906,589

1979

412.5

225,055,000

928,352

1980

417.9

227,726,000

951,667

1981

425.2

229,966,000

977,815

1982

424.5

232,188,000

985,522

1983

431.1

234,307,000

1,010,097

1984

439.9

236,348,000

1,039,695

1985

448.7

238,466,000

1,069,997

1986

451.3

240,651,000

1,086,058

1987

467.9

242,804,000

1,136,080

1988

463.7

245,021,000

1,136,162

1989

467.5

247,342,000

1,156,324

1990

481.7

250,132,000

1,204,761

1991

502.7

253,493,000

1,274,309

1992

510.2

256,894,000

1,310,673

1993

493.0

260,255,000

1,283,057

1994

482.9

263,436,000

1,272,132

1995

476.1

266,557,000

1,269,078

1996

478.3

269,667,000

1,289,817

1997

485.5

272,912,000

1,324,988

1998

486.9

276,115,000

1,344,404

1999

489.2

279,295,000

1,366,171

2000

484.0

282,385,000

1,366,743

2001

486.5

285,267,000

1,387,824

2002

483.0

288,028,000

1,391,060

2003

469.3

290,704,000

1,364,274

2004

467.7

293,310,000

1,371,811

2005

459.9

295,994,000

1,361,276

2006

456.2

298,766,000

1,362,970

Sources:  Data on incidence rates were obtained from SEER "Cancer Statistics Review 1975–2006." They relate to the SEER-9 areas.  Rates are per 100,000 of population and are age-adjusted to the US 2000 Standard Population. The US population was obtained from the Statistical Abstract of the United States.

Number of Total Cancer Cases in the US
Next, we estimated the total number of people diagnosed with cancer during 1975–2006. This was done by multiplying the annual cancer incidence rate (age-adjusted) by the annual population of the US (per 100,000). For this calculation, one needs to assume that the SEER incidence rate is representative enough for the nation. The results are shown in Table 1, column 3. In 1975, 864,756 Americans were diagnosed with cancer. This number continued to climb rapidly and reached over a million (1,010,097) by 1983 – less than 10 years after 1975.The total number of people getting cancer continued to increase; by 2002, the number reached a formidable 1,391,060. The numbers in 2001 and 2002 were the highest annual numbers during the period of analysis.

Over the period 1975 to 2006, the total number of Americans who were afflicted with cancer reached close to 38 million. In the last ten-year period from 1997 to 2006, the total number of people who were afflicted reached close to 14 million. This most recent period has the largest annual numbers of American contracting cancer over the analysis period.  

Graphical Illustration of the Cancer Incidence
The increases of cancer incidence over time are observed graphically in Chart 1 (with data from Table 1). The height of each bar represents the (age-adjusted) incidence rate for that year. The heights of the bars rise annually from 1975, indicating a steady increase in the cancer incidence rate. The height of the bar in 2006 is significantly higher than in 1975. If the prevention side of the cancer program had stayed at the same level of efficiency (or inefficiency) as in 1975, that would be shown by the height of the bar in 1975 continuing to be the same in all the other years.

Chart 2 displays the total number of Americans who contracted cancer in each year during 1975–2006 (with data from Table 1). This number is represented by the height of the bar. There are steady increases in the heights of the bars over 1975–2006, indicating the steadily increasing number of people who were afflicted with cancer. It is again observed that the last 10 years (since 1997) have the largest numbers of people being afflicted by cancer.

Increases of the Cancer Incidence Rate
Over the period of analysis (1975–2006), the incidence rate of cancer increased at an annual rate of 0.4%. The cancer incidence rate is the ratio of the number of people who are diagnosed with cancer to the population of the US (per 100,000 persons). Consequently, a positive growth rate means that the number of people diagnosed with cancer every year (the numerator) grew faster than the population of the country. This is a crucial point. If the cancer program had been effective, there would have been a drop in the growth rate – that is, it would have been negative. Instead, the opposite happened.

In addition, the total number of people in the US who were afflicted with cancer also rose annually at 1.5%. During this time period, the population of the country grew at an annual rate of 1.1%. Thus, the total number of Americans afflicted by cancer, over 1975–2006, climbed faster than the US population.

Cancer Mortality
Mortality Rate of Cancer

Of those who are diagnosed with cancer, a significant number, unfortunately, succumb to the disease and die – often with considerable suffering. To show the fatal results of cancer, Table 2 presents data on (age-adjusted) mortality rates (per 100,000), for the 1975–2006 period. The cancer mortality rates over time reveal a similar pattern as observed for the cancer incidence. The SEER-9 reported mortality rate in 1975 was 199 (per 100,000) and steadily increased up to 1991 when it reached 215, the highest rate over the period.1

Table 2: Cancer Mortality in the US, 1975–2006

Year

Cancer Mortality (age-adjusted)

Population – US (July 1)

Number of People Who  Died from Cancer

(1)

    (2)

(3)

1975

199.1

215,973,000

430,002

1976

202.3

218,035,000

441,085

1977

203.0

220,239,000

447,085

1978

204.4

222,585,000

454,964

1979

204.5

225,055,000

460,237

1980

207.0

227,726,000

471,393

1981

206.4

229,966,000

474,650

1982

208.3

232,188,000

483,648

1983

209.2

234,307,000

490,170

1984

210.9

236,348,000

498,458

1985

211.3

238,466,000

503,879

1986

211.8

240,651,000

509,699

1987

211.9

242,804,000

514,502

1988

212.6

245,021,000

520,915

1989

214.3

247,342,000

530,054

1990

214.9

250,132,000

537,534

1991

215.1

253,493,000

545,263

1992

213.7

256,894,000

548,982

1993

213.4

260,255,000

555,384

1994

211.7

263,436,000

557,694

1995

209.9

266,557,000

559,503

1996

207.0

269,667,000

558,211

1997

203.6

272,912,000

555,649

1998

200.8

276,115,000

554,439

1999

200.7

279,295,000

560,545

2000

198.7

282,385,000

561,099

2001

195.9

285,267,000

558,838

2002

193.7

288,028,000

557,910

2003

190.1

290,704,000

552,628

2004

185.9

293,310,000

545,263

2005

184.2

295,994,000

545,221

2006

181.1

298,766,000

541,065

Sources:  The data in column 1 were obtained from SEER "Cancer Statistics Review 1975-2006."  Rates are per 100,000 of population and are age-adjusted to the US 2000 Standard Population. Population data came from the same source as Table 1.   

For every year following 1975, the number of Americans dying from cancer rose annually until 2000. If the cancer treatment program of the US had been successful, there would have been declines in the cancer mortality rate after 1975 instead of increases. This indicates an inefficiency in the treatment of cancer patients.

From 1992, the mortality rate began to decline somewhat. However, it was still higher than in 1975. This continued until 2000, when the mortality rate reached 199 (rate rounded). It was only after 2000 that the age-adjusted mortality rate fell below what it was in 1975. If one compares the rate in 1975 with that in 2006, the result is a decrease of 9% for the entire 32-year period. This translates into a miniscule 0.28% annual rate of decrease.

Number of Americans who Died from Cancer
In order to calculate the total number of Americans who lost their lives to cancer annually, the SEER mortality rates of Table 2 were multiplied by the population (in 100,000s). The results are depicted in column 3 of Table 2. The number of Americans who died from cancer in 1975 was 430,002. By 1985, the number had surpassed 500,000 people per year and by 2000 had reached the highest level at over 561,099. The number in 2006 is estimated at 541,065. These numbers also tell a story of failure.2

A disturbing finding is that a much higher number of Americans lost their lives to cancer in 2006 – 541,065 – than in 1975 – 430,002 people. The total number of Americans who died from cancer, over 1975–2006, was close to 17 million; that is 6.5% of the average population of the US over the period.3

In putting the mortality numbers in a different perspective, the number of Americans who have been losing their lives to cancer, during the 1990s and 2000s, reached over 1,500 people a day – the count of a World Trade Center building of the 9/11/2001 tragic event. It is as though the US is stricken by such a catastrophe every day. That number translates into 63 Americans dying from cancer every hour of the day – or 1 American dying from cancer every minute. These are indeed grim statistics showing the terrible outcome from the failure of the US cancer program.

Graphical Illustration of Cancer Mortality
The mortality rates of cancer in the US are also shown graphically in Chart 3, by the heights of the bars. From 1975 the bars continue to rise over time, reaching their highest level in 1991. That is, for 16 consecutive years after 1975, the number of Americans dying from cancer (per 100,000 population) continued to increase. After 1991, there is a decline in the bar heights; that, however, translates into a very small annual decrease of the mortality rate over the period of analysis.

It is observed in Chart 4 that from 1975 the bar heights rise steadily over time – indicating an ever-increasing number of people dying from the disease. The bar heights continued to rise, reaching a peak in 2000.

With regard to the relationship between cancer incidence and mortality, over the period of analysis (1975–2006), a total of 38.6 million Americans were afflicted by cancer. During that period, a total of 16.6 million Americans lost their lives to the disease. The ratio between these two numbers provides an approximation of those who die from cancer in relation to those afflicted by the disease – at 44%. This is a grim statistic for all Americans.

Notes
We use SEER data for the analysis over 1975–2006. CDC provides data on cancer mortality for the entire US, but these data are only available from 1979. The advantages of the CDC data are that they cover the entire US and they are provided in both the raw (unadjusted) form and in age-adjusted form. This enables one to compare the cancer outcomes in both forms, which makes for a more comprehensive picture of cancer. The CDC data on cancer mortality are discussed in Appendix A, and compared to the SEER data on cancer mortality.  Appendix Table A-1 presents cancer mortality rates from SEER and CDC. In contrast to the SEER age-adjusted rates which show a decrease over time, the CDC crude rates increase over time. The average population was obtained by summing the population of the first and last year, and dividing by 2.

References
Apostolides AD. Alternative treatments of cancer: trends over time and implications for the future. J Health Soc Policy. 1999;10(3).
Bailar JC III, Smith EM. Progress against cancer?
N Engl J Med. May 8 1986;314(19).
Bailar JC III, Gornick HL. Cancer undefeated.
N Engl J Med. May 29 1997;336(22).
Clegg LX, Feuer EJ, Midthune DN, Hankey BF. Impact of reporting delay and reporting error on cancer incidence rates and trends.
J Natl Cancer Inst. Oct. 16 2002;94(20).
Engel LW, Strauchen JA, Chiazze L Jr, Heid M. Accuracy of death certification in an autopsied population with specific attention to malignant neoplasms and vascular diseases.
Am J Epidemiol. 1980;111(1).
Epstein SS.
Cancer-Gate: How to Win the Losing Cancer War. Amityville, NY: Baywood Publishing Company; 2005.
Faquet GB.
The War on Cancer: An Anatomy of Failure, A Blueprint for the Future. Dordrecht, Netherlands: Springer; 2005.
Horner MJ, Ries LAG, Krapcho M, et al., eds. SEER cancer statistics review, 1975–2006. National Cancer Institute. http://seer.cancer.gov/csr/1975_2006; based on November 2008 SEER data submission, posted 2009.
Kolata G. In long drive to cure cancer, advances have been elusive. New York Times. April 24, 2009.
Landrigan PJ. Commentary: Environmental Disease – A Preventable Epidemic.
Am J Public Health. July 1992;82(7).
Leaf C. Why we're losing the war on cancer.
Fortune. March 22, 2004.
Percy C, Stanek E III, Gloeckler L. Accuracy of cancer death certificates and its effect on cancer mortality statistics.
Am J Public Health. March 1981;71(3).
Spector R. The war on cancer – a progress report for skeptics.
Skeptical Inquirer. January/February 2010.
United States Bureau of the Census. Statistical Abstract of the United States; section on Population.

Appendix A – Shortcomings of Existing Cancer Data
The reporting by SEER of only age-adjusted data instead of crude and age-adjusted data; actual cancer mortality rates show a different picture than age-adjusted rates.

The SEER data are made available in age-adjusted form but not in their actual form – referred to as "crude" in official NCI statistics. One needs to do special tabulations on the NCI web site in order to translate the SEER age-adjusted data into the crude data.

Data on cancer mortality rates are available from CDC as both crude and age-adjusted rates. These are presented in Appendix Table A-1 (along with age-adjusted mortality rates from SEER). From 1979 to 2006, the crude rates show an annual increase (0.2%), while the age-adjusted rates show an annual decrease (−0.4%). There is a substantial difference in the information provided by the age-adjusted data and crude data. Consequently, an incomplete picture of cancer is given when only the age-adjusted data are provided by NCI to the public. 

Table A-1: Cancer Mortality in the US, 1975–2006: SEER and CDC Data

SEER

CDC

Year

Age-Adjusted Rate Per 100,000 Population

Crude Rate Per 100,000 Population

Age-Adjusted Rate Per 100,000 Population

(1)

      (2)

        (3)

1975

199.1

1976

202.3

1977

203.0

1978

204.4

1979

204.5

182.2

207.0

1980

207.0

186.5

211.0

1981

206.4

186.6

209.5

1982

208.3

190.0

211.5

1983

209.2

192.3

212.3

1984

210.9

194.9

213.9

1985

211.3

196.7

214.4

1986

211.8

198.2

214.6

1987

211.9

199.5

214.7

1988

212.6

201.0

215.5

1989

214.3

203.6

217.1

1990

214.9

205.7

219.0

1991

215.1

206.1

218.2

1992

213.7

205.6

216.5

1993

213.4

206.7

216.5

1994

211.7

205.8

214.7

1995

209.9

205.0

212.9

1996

207.0

203.0

209.7

1997

203.6

200.6

206.3

1998

200.8

199.1

203.7

1999

200.7

201.8

205.6

2000

198.7

201.3

204.5

2001

195.9

199.2

200.9

2002

193.7

197.9

198.1

2003

190.1

196.2

194.4

2004

185.9

193.2

190.4

2005

184.2

193.3

188.3

2006

181.1

191.7

185.3

Note:  The age-adjusted rates use the population of year 2000 as the standard population.
Sources:  The data of SEER are from Table 1. The data of CDC were obtained at CDC WONDER (http://wonder.cdc.gov/mortSQL.html)

The age-adjusted mortality rates of SEER are lower than those of CDC.

The age-adjusted cancer mortality rates from CDC are higher than the SEER rates.

Mortality statistics are not adjusted by CDC or SEER for possible mistakes.

Cancer mortality statistics are obtained by NCI from CDC and are incorporated into the SEER database. However, the cancer mortality statistics are not revised by CDC. Once these statistics are collected for any year, they stay at that number and are not updated. This can create problems with measurement since any past mistakes that may have been made in the count stay unchanged. Since large data series typically need to be revised and updated over time, we wonder about the quality and reliability of the mortality statistics of CDC used by SEER, that undergo no updating. A number of studies have revealed errors in the accuracy of cancer death certificates and the effect on cancer mortality statistics (Percy et al. 1981, Engel et al. 1980).

Implications of age-adjusted cancer rates. 
Age-adjusted cancer rates have been adjusted – that is, changed – for the impact of age in cancer incidence or mortality. In doing this adjustment, a particular year is chosen whose population becomes the standard population. This population (by age brackets) is used to multiply the incidence rates, by age brackets, of other years. This holds the population of the county the same in every year of a period of years. Consequently, the age-adjusted incidence (or mortality rate) is calculated as to what it would have been if the population had not changed over time.

The rationale behind the age adjustment is that the cancer incidence will increase in a country if the population is aging (that is, when the older age brackets grow faster than the young age brackets) because older people have higher incidence rates. However, an increased incidence of people in the older age brackets can be the result of two factors. One would be that age results in increased disease – other things, like the environment, being constant. The second factor is the environment which contains carcinogens (which tend to increase over time) and older Americans would have been exposed to these carcinogens over longer periods of time than younger people. In time, they succumb to increased attacks on their immune system by the carcinogens, and cancer occurs.

The age-adjustment holds the age of the population constant but does not hold constant the effect of the environment on people over time. This makes it more relevant for NCI/SEER to provide both crude cancer data and age-adjusted data. This would provide a more complete and accurate situation of cancer to the public.

Delays in the reporting of cancer incidence counts.
In 2002, research by staff of NCI reported that there were delays in the reporting of cancer cases. These delays resulted in significant undercounting of the cancer cases for a particular year. Consequently, the cancer incidence rates, reported by SEER, were underestimates of the actual cancer incidence rates. The study analyzed data for six important types of cancer: female breast, colorectal, lung/bronchus, prostate, and melanoma. According to the study, although the SEER program allows about 2 years for the collection of newly diagnosed cancer cases, it would actually take 4–17 years for 99% or more of the cancer cases to be reported (given the incidence counts initially reported). The incidence counts reported, at the 2-year delay, accounted for just 88–97% of the estimated final incidence counts (Clegg et al. 2002).

The SEER data for cancer incidence available for 1975–2006 have been adjusted for reporting delays. Thus, it is possible to compare incidence rates reported for 1975–2003, which were not adjusted for reporting delays, with more recent incidence rates for 1975–2006, which had been adjusted for reporting delays. These data are presented in Appendix Table A-2. In examining the two sets of numbers, one does not observe substantial differences between them.

Table A-2: Incidence Rates – SEER – 2003 and 2006 Adjustment for Reporting Delays

Year

Cancer Incidence Rate (age-adjusted)

Cancer Incidence Rate (age-adjusted) –Adjusted for reporting delays

(1)

(2)

1975

400.3

400.4

1976

407.2

407.4

1977

407.6

407.7

1978

407.2

407.3

1979

412.4

412.5

1980

417.7

417.9

1981

425.1

425.2

1982

424.3

424.5

1983

430.9

431.1

1984

439.6

439.9

1985

448.4

448.7

1986

451.0

451.3

1987

467.8

467.9

1988

463.5

463.7

1989

467.3

467.5

1990

481.2

481.7

1991

502.3

502.7

1992

509.8

510.2

1993

492.6

493.0

1994

482.5

482.9

1995

475.5

476.1

1996

477.3

478.3

1997

484.3

485.5

1998

485.7

486.9

1999

487.6

489.2

2000

481.9

484.0

2001

483.7

486.5

2002

478.9

483.0

2003

459.6

469.3

2004

467.7

2005

459.9

2006

456.2

Note:  The data in column 1 are not adjusted for reporting delays, by SEER.
Sources:  Data in column 1 were obtained from http://seer.cancer.gov/csr/1975_2003/results_merged/ topics_annualrates.pdf
For data in column 2, same source as in Table 1.

Using data provided in the NCI study, we estimated that 7.5% of cancer cases would not be reported, on the average, at the 2-year period. The calculation is shown below. With that magnitude of cases not reported (at the 2-year period), one would have expected the incidence rates after adjustment for reporting delays to be noticeably higher than the ones shown. Nevertheless, these are the numbers provided by SEER, unless there is appraisal in the future of the technique used for the adjustment.

Calculation: (1) 97–88 = 9% of cases are not reported at the 2-year period; (2) 9%/2 = 4.5% average percentage of cases not reported; (3) 88+4.5 = 92.5% of the cases would be reported, on the average, at the 2-year period; (4) 100-92.5 = 7.5% of cancer cases would be missed on the average at the 2-year period.

 

Consult your doctor before using any of the treatments found within this site.

Subscriptions are available for Townsend Letter, the Examiner of Alternative Medicine magazine, which is published 10 times each year.

Search our pre-2001 archives for further information. Older issues of the printed magazine are also indexed for your convenience.
1983-2001 indices ; recent indices

Once you find the magazines you'd like to order, please use our convenient form, e-mail subscriptions@townsendletter.com, or call 360.385.6021 (PST).

Who are we? | New articles | Featured topics |
Tables of contents
| Subscriptions | Contact us | Links | Classifieds | Advertise | Alternative Medicine Conference Calendar | Search site | Archives |
EDTA Chelation Therapy | Home

© 1983-2010 Townsend Letter for Doctors & Patients
All rights reserved.
Website by Sandy Hershelman Designs
September 25, 2010

Order back issues
Advertise with TLDP!

Visit our pre-2001 archives