Skip to main content
NCBI home page
Public Health Reports logoLink to Public Health Reports
. 2019 Dec 5;135(1):150–160. doi: 10.1177/0033354919893029

Male Mortality Trends in the United States, 1900-2010: Progress, Challenges, and Opportunities

Wanda K Jones 1,✉, Robert A Hahn 2, R Gibson Parrish 3, Steven M Teutsch 4,5, Man-Huei Chang 6
PMCID: PMC7119244  PMID: 31804898

Abstract

Objectives:

Male mortality fell substantially during the past century, and major causes of death changed. Building on our recent analysis of female mortality trends in the United States, we examined all-cause and cause-specific mortality trends at each decade from 1900 to 2010 among US males.

Methods:

We conducted a descriptive study of age-adjusted death rates (AADRs) for 11 categories of disease and injury stratified by race (white, nonwhite, and, when available, black), the excess of male mortality over female mortality ([male AADR − female AADR]/female AADR), and potential causes of persistent excess of male mortality. We used national mortality data for each decade.

Results:

From 1900 to 2010, the all-cause AADR declined 66.4% among white males and 74.5% among nonwhite males. Five major causes of death in 1900 were pneumonia and influenza, heart disease, stroke, tuberculosis, and unintentional nonmotor vehicle injuries; in 2010, infectious conditions were replaced by cancers and chronic lower respiratory diseases. The all-cause excess of male mortality rose from 9.1% in 1900 to 65.5% in 1980 among white males and a peak of 63.7% in 1990 among nonwhite males, subsequently falling among all groups.

Conclusion:

During the last century, AADRs among males declined more slowly than among females. Although the gap diminished in recent decades, exploration of social and behavioral factors may inform interventions that could further reduce death rates among males.

Keywords: male health, mortality trends, chronic disease, sex and gender disparities

From 1900 to 2010, age-adjusted death rates (AADRs) per 100 000 for the US population declined 70%, from 2518 to 747; life expectancy at birth increased from 47.3 to 78.7 years; and major causes of death (CODs) changed.1,2 Female mortality rates declined substantially during this period,3 likely as a result of broad progress in public health, health behaviors, social and environmental factors, and, more recently, clinical care. During this period, male mortality rates persistently exceeded female mortality rates.1 Here, we explored trends in male mortality from 1900 to 2010, trends in male excess mortality, and possible explanations.

We extended previous analyses, recognizing changes in death reporting and disease classification. Changes include an increase since the early 20th century in the number of states that report death data, improvements in clinical knowledge that have led to better disease classification, and 9 revisions (approximately decennially) of the original list of International Causes of Death (ICD).3 Racial/ethnic data were limited to the categories of white and nonwhite until 1970, when data on black persons were first reported separately. Consideration of differences in death rates by cause, sex, race, and other factors may help elucidate biological, behavioral, social, and environmental contributions to excess male mortality, suggesting public health and medical interventions.

Methods

We tabulated male death rates: all-cause age-specific death rates, trends in AADRs for specific CODs, and AADRs for major CODs by race. We examined trends in male excess AADRs ([male AADR − female AADR]/female AADR) for major CODs by race. We obtained data for the analysis of decadal trends in death rates (1900-2010) from tabulations of CODs from published compilations and from public-use computer data files, as previously described.4 Because all data were from previously published US government sources, no institutional review board review was required.

Analyses

We reported death rates and AADRs for the underlying COD for males and females by race (white and nonwhite) previously,4 and we used those rates in these analyses. The CODs examined were the 5 major causes either in 1900 or in 2010—ICD groupings that constitute the largest proportions of deaths at those times. Two physician epidemiologists (S.M.T. and R.G.P.) linked multiple ICD codes. From 1900 to 1940, leukemia was not classified with cancers in ICDs, and rates for leukemia are available for these years only for the whole population (ie, without sex- or race-specific rates). Therefore, to show rates for cancers for the entire period of 1900-2010, leukemia was excluded from the “all cancer” category before 1950.

All analyses reported here examined male mortality, unless otherwise noted. Because age- and cause-specific rates by white race and nonwhite race were not available until 1920, we tabulated cause-specific AADRs by race only for 1920-2010. Because age- and cause-specific data for black persons were not available until 1970, we tabulated data for black males only for 1970-2010. All rates were tabulated per 100 000 US 2000 standard population.

We calculated the following:

  1. Age-specific all-cause death rates for white males and nonwhite males from 1900-2010 in decadal years.

  2. Major CODs in 1900 and in 2010, based on AADRs for each major cause in 1900 and 2010. Direct comparison of ranked causes between the 2 years was not possible, because data for chronic lower respiratory diseases were not available in 1900.

  3. AADR trends for selected major CODs for the whole male population from 1900-2010, for white males and nonwhite males from 1920-2010, and for black males from 1970-2010.

  4. Cause-specific AADRs from 1900-2010 for selected conditions combined: chronic (ie, heart disease, stroke, cancers combined), infectious (ie, influenza and pneumonia, tuberculosis, and enteritis and diarrhea combined), and unintentional injury (ie, motor vehicle injury [MVI)] and unintentional nonmotor vehicle injuries [UNMVIs], such as poisoning [including drug overdose], falls, and drowning).

  5. Relative differences between male and female AADRs ([male AADR − female AADR]/female AADR) for all causes and for selected specific causes for the whole population from 1900-2010, for white males and nonwhite males from 1920-2010, and for black males from 1970-2010. Relative differences >0 were defined as “excess.”

We used Microsoft Excel software for all analyses.

Results

Age-Specific All-Cause Death Rates, by Race, 1900-2010

Age-specific death rates declined from 1900 to 2010 for all age groups among both white and nonwhite males, with greater declines among younger age groups among both white and nonwhite males (Table 1). Age-specific death rates among nonwhite males exceeded those among white males in all decades and all age groups except for males aged ≥75, but declines in death rates were greater among nonwhite males than among white males in all age groups. The greatest declines were among males aged 1-4 years—a decline of 98.6% among white males and 99.2% among nonwhite males.

Table 1.

Age-specific death rates (per 100 000 US 2000 standard population) among males, by race (white and nonwhite), death registration states, 1900-1930, and United States, 1940-2010a

Age, y, and Raceb Death Registration States United States % Decline 1900 to 2010
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
<1
 White 17 587 14 297 9810 7154 5670 3401 2694 2113 1230 896 668 584 97
 Nonwhite 36 926 25 762 16 774 12 226 10 122 5992 5189 4020 2350 1805 1305 974 97
1-4
 White 2016 1424 984 554 281 136 105 84 66 46 33 27 99
 Nonwhite 4343 3013 1498 996 527 271 207 145 103 79 48 37 99
5-14
 White 378 296 271 181 112 67 53 48 35 26 20 14 96
 Nonwhite 777 543 365 266 164 97 75 65 45 37 25 17 98
15-24
 White 580 469 425 298 197 152 144 171 167 131 106 92 84
 Nonwhite 1183 1001 994 781 499 290 214 305 202 217 150 117 90
25-34
 White 813 672 594 407 281 185 163 177 171 176 124 136 83
 Nonwhite 1249 1264 1215 1214 853 496 386 504 358 350 199 163 87
35-44
 White 1059 969 765 651 507 381 333 344 257 268 234 207 81
 Nonwhite 1419 1752 1445 1700 1318 861 729 874 594 559 359 235 83
45-54
 White 1550 1495 1198 1231 1139 985 932 883 699 549 497 492 68
 Nonwhite 2466 2483 2009 2633 2453 1857 1551 1646 1309 1018 797 568 77
55-64
 White 2852 2841 2421 2552 2522 2304 2225 2203 1729 1467 1163 1033 64
 Nonwhite 4213 3874 3114 4039 3711 3481 3152 3047 2606 2219 1661 1304 69
65-74
 White 5913 5858 5419 5512 5400 4865 4848 4810 4036 3398 2906 2232 62
 Nonwhite 7164 6632 6016 6727 6283 5795 5664 5474 4747 4349 3545 2547 65
75-84
 White 12 823 12 764 12 250 11 921 12 202 10 526 10 300 10 099 8830 7845 6933 5704 56
 Nonwhite 13 139 11 379 11 599 11 763 10 877 9030 8663 8981 8689 8316 7352 5615 57
≥85
 White 26 918 25 794 25 358 23 762 25 135 22 116 21 750 18 552 19 097 18 268 17 716 15 640 42
 Nonwhite 24 929 17 407 24 709 22 845 19 972 16 022 15 239 11 405 15 775 16 081 15 488 13 293 47
Open in a new tab

a The states providing mortality data are known as “death registration states”; they included 10 states and the District of Columbia in 1900 (40.5% of the US population) and gradually expanded to include all 48 states and the District of Columbia by 1933. Data source: tabulations of causes of death from published compilations and from public-use computer data files, as previously described in Hahn et al.4

b Disaggregation of race beyond white and nonwhite did not occur until 1970.

Declines in death rates among white and nonwhite males from birth through age 14 years occurred every decade from 1900 to 2010. For white and nonwhite males aged 15-44, death rates increased from 1960 to 1970 and were irregular thereafter. For several white and nonwhite age groups aged ≥35, death rate declines were irregular in decades after 1910, increased in 1930, and declined irregularly thereafter.

Major CODs in 1900 and 2010

Five major CODs for males comprised 44.3% of the all-cause AADRs in 1900: pneumonia and influenza (10.9%), heart disease (10.9%), stroke (9.4%), tuberculosis (8.3%), and UNMVIs (4.8%, minimally greater than diarrhea and enteritis; Table 2). In 2010, 5 major CODs comprised 62.0% of all-cause AADRs: heart disease (25.4%), all cancers (23.7%), chronic lower respiratory diseases (5.4%), stroke (4.4%), and UNMVIs (4.0%).

Table 2.

Age-adjusted death rates (per 100 000 US 2000 standard population) among males for all causes and selected causes, by race (white, nonwhite, and black) and available years, death registration states, 1900-1930, and United States, 1940-2010a

Cause/Raceb Death Registration States United States
1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
All causes
 All 2631 2460 2213 2089 1976 1674 1609 1542 1348 1203 1054 887
 White 2613 2442 2167 2014 1925 1643 1586 1514 1318 1166 1029 879
 Nonwhite 3577 3088 2748 2846 2459 1950 1778 1755 1569 1443 1187 911
 Black — — — — — — — 1874 1698 1645 1404 1104
Selected infectious diseasesc
 All 631 520 458 285 200 96 83 60 44 49 29 19
 White — — 436 261 182 86 77 57 42 48 29 19
 Nonwhite — — 718 514 374 191 135 89 54 57 33 22
 Black — — — — — — — 93 56 61 35 23
Tuberculosis
 All 218 181 130 86 61 35 12 5 2 1 0 0
 White — — 117 73 51 29 10 4 1 1 0 0
 Nonwhite — — 275 210 154 91 29 14 5 5 2 1
 Black — — — — — — — 15 6 5 2 1
Influenza and pneumonia
 All 287 239 281 172 128 55 66 54 42 48 29 18
 White — — 272 162 120 51 62 52 41 47 28 18
 Nonwhite — — 382 263 201 90 99 72 48 52 31 21
 Black — — — — — — — 77 50 56 34 21
Diarrhea and enteritis
 All 126 100 48 27 12 5 5 1 0 0 0 1
 White — — 46 25 11 5 5 1 0 0 0 1
 Nonwhite — — 61 41 20 10 7 2 0 0 0 1
 Black — — — — — — — 2 0 0 0 1
Selected chronic diseasesd
 All 624 730 753 874 999 1094 1099 1039 913 762 631 474
 White — — 759 873 1006 1094 1101 1039 904 747 620 469
 Nonwhite — — 648 861 894 1065 1056 1033 977 869 700 497
 Black — — — — — — — 1106 1057 986 828 607
All cancers except leukemia
 All 91 118 140 162 187 208 225 248 271 280 249 210
 White — — 145 168 192 210 225 245 265 272 244 208
 Nonwhite — — 61 81 118 180 225 273 324 347 286 219
 Black — — — — — — — 292 353 398 346 269
Cancers of respiratory system
 All — — — — — 29 48 72 89 95 80 63
 White — — — — — 30 48 72 88 93 78 62
 Nonwhite — — — — — 21 47 76 103 113 88 63
 Black — — — — — — — 81 114 132 107 78
Major cardiovascular diseases
 All 579 716 720 792 877 965 950 850 683 511 407 282
 White — — 724 786 880 963 952 853 680 505 401 277
 Nonwhite — — 663 846 832 961 912 812 692 552 444 302
 Black — — — — — — — 870 745 623 518 367
Heart disease
 All 286 371 383 508 632 699 688 634 539 412 320 225
 White — — 383 505 638 701 695 640 540 409 317 223
 Nonwhite — — 381 530 552 653 601 568 521 429 334 229
 Black — — — — — — — 607 561 485 393 281
Stroke
 All 248 241 230 204 180 186 186 157 102 69 62 39
 White — — 231 200 176 182 182 154 99 66 60 38
 Nonwhite — — 206 250 225 231 230 192 131 92 80 49
 Black — — — — — — — 206 142 103 90 57
Chronic lower respiratory diseases
 All — — — — — — — 39 50 56 56 49
 White — — — — — — — 41 52 57 57 51
 Nonwhite — — — — — — — 26 32 44 43 34
 Black — — — — — — — 27 34 48 48 40
Unintentional injuriese
 All — 156 125 147 130 102 86 90 71 54 49 52
 White — — 125 146 129 100 83 86 68 53 49 54
 Nonwhite — — 124 154 137 119 111 123 89 65 50 40
 Black — — — — — — — 127 94 71 58 46
Unintentional motor vehicle injuries
 All — 3 17 47 45 39 35 42 34 27 22 16
 White — — 18 48 45 38 35 40 34 26 22 17
 Nonwhite — — 9 43 45 44 41 51 34 28 22 14
 Black — — — — — — — 51 34 30 24 17
Unintentional nonmotor vehicle injuries
 All 127 152 108 100 85 64 51 49 37 28 28 35
 White — — 107 98 84 62 49 46 35 26 27 37
 Nonwhite — — 115 111 91 75 70 73 55 37 29 26
 Black — — — — — — — 76 59 42 34 29
Ill-defined conditions
 All — — — — — 30 20 20 18 12 13 12
 White — — — — — 23 15 16 15 11 13 12
 Nonwhite — — — — — 109 66 58 44 22 16 13
 Black — — — — — — — 64 49 26 20 17
Senility
 All — — — — — — — 1 1 1 1 2
 White — — — — — — — 1 1 1 1 2
 Nonwhite — — — — — — — 4 2 1 1 1
 Black — — — — — — — 4 2 1 1 2
Other ill-defined conditions
 All — — — — — — — 19 17 12 12 10
 White — — — — — — — 15 14 10 11 10
 Nonwhite — — — — — — — 54 42 12 15 12
 Black — — — — — — — 60 47 25 19 15
Open in a new tab

Abbreviation: —, data not available.

a The states providing mortality data are known as “death registration states”; they included 10 states and the District of Columbia in 1900 (40.5% of the US population) and gradually expanded to include all 48 states and the District of Columbia by 1933. Data source: tabulations of causes of death from published compilations and from public-use computer data files, as previously described.4

b Disaggregation of race beyond white and nonwhite did not occur until 1970.

c Selected infectious disease totals are the sum of the rates for influenza and pneumonia, tuberculosis, and enteritis and diarrhea.

d Selected chronic disease totals are the sum of the rates for heart disease, stroke, and all cancers. They do not include chronic lower respiratory diseases.

e Unintentional injury totals are the sum of the rates for unintentional motor vehicle injuries and unintentional nonmotor vehicle injuries.

Cause-Specific AADRs

From 1900 to 2010, the greatest decreases in AADRs were for selected infectious diseases: pneumonia and influenza declined by 93.7% (from 287 to 18 per 100 000 population), tuberculosis by 99.9% (from 218 to 0.2 per 100 000 population), and diarrhea and enteritis by 99.4% (from 126 to 1 per 100 000 population; Table 2; Figure 1). By 1950, these infectious conditions combined already had declined by 84.9% (from 631 to 95 per 100 000 population). AADRs for major cardiovascular diseases (dominated by heart disease) increased by 66.7% from 1900 (579 per 100 000 population) to 1950 (965 per 100 000 population) and then declined 70.8% (to 282 per 100 000 population) by 2010. AADRs for UNMVIs peaked in 1910 (152 per 100 000 population) and then declined 76.8% by 2010 (to 35 per 100 000 population).

Figure 1.

Figure 1.

Open in a new tab

Age-adjusted death rates by cause of death among US males, 1900-2010. Rates are per 100 000 US 2000 standard population. Data for 1900-1930 are from death registration states; data for 1940-2010 are from the United States. The states providing mortality data are known as “death registration states”; they included 10 states and the District of Columbia in 1900 (40.5% of the US population) and gradually expanded to include all 48 states and the District of Columbia by 1933. Data source: tabulations of causes of death from published compilations and from public-use computer data files, as previously described.4 Abbreviations: MVI, motor vehicle injuries; UNMVI, unintentional nonmotor vehicle injuries.

The AADR trends among males for 3 additional major CODs varied during the study period (Table 1). For cancers, the AADR almost tripled from 1900 to 1990 and then declined by 25.1% thereafter. Chronic lower respiratory conditions, first reported in 1970 at 39 per 100 000 population, increased by 41.6% in 2000 and then declined by 12.7% in 2010. The AADR for MVIs, first recorded in 1910 at 3 per 100 000 population, increased rapidly to a peak of 47 per 100 000 population in 1930 and then declined by 65.7% to 16 per 100 000 population in 2010, after 4 decades of fluctuation.

All-Cause and Cause-Specific AADRs Among White and Nonwhite Males (1920-2010) and Black Males (1970-2010)

From 1900 to 2010, the decline in all-cause AADR was greater among nonwhite males (74.5%) than among white males (66.4%), but the AADR among white males was lower than the AADR among nonwhite males throughout the study period (Table 1). The AADRs among black males exceeded the rate among white and nonwhite males from 1970 to 2010, but they declined 41% overall during this period.

The AADRs for heart disease among white and nonwhite males rose 83.1% and 71.5%, respectively, from 1920 to a peak in 1950 and subsequently fell by 68.2% and 64.9% in 2010 (Table 2). The AADRs for stroke started at their peak of 231 per 100 000 population in 1920 among white males; they peaked at 250 per 100 000 population in 1930 among nonwhite males. Stroke rates then declined by 83.7% and 80.5% among white and nonwhite males, respectively, by 2010. In 2010, the AADR for stroke among nonwhite males (49 per 100 000 population) was 30.1% higher than among white males (38 per 100 000 population).

The AADRs for cancer among both white and nonwhite males rose from 145 and 61 per 100 000 population, respectively, to peaks in 1990—272 and 347 per 100 000 population, respectively—and declined thereafter. The AADRs for cancer were substantially greater among white males than among nonwhite males until 1960, when the AADRs for cancer among nonwhite males surpassed the AADRs for cancer among white males (Table 2).

The AADRs declined similarly among white and nonwhite males for enteritis and diarrhea, tuberculosis, and pneumonia and influenza; the AADRs among nonwhite males exceeded those among white males in all decades for tuberculosis and for pneumonia and influenza but only until 1980 for enteritis and diarrhea (Table 2).

The AADRs for UNMVIs generally declined among nonwhite males from 1920 to 2010; among white males, the AADRs for UNMVIs declined until 1990 and increased sharply until 2010. The AADRs for UNMVIs among nonwhite males exceeded those among white males in all decades except 2010. The AADRs for MVIs peaked at 48 per 100 000 population among white males in 1930 and generally declined thereafter; the AADRs for MVIs among nonwhite males reached a peak of 51 per 100 000 population in 1970 and then declined (Table 2).

In 1920, the AADRs for major CODs among nonwhite males exceeded those among white males only for infectious diseases and for unintentional UNMVIs. By 2010, the AADRs among nonwhite males exceeded those among white males for major cardiovascular diseases, cancers, and infectious diseases. From 1970 through 2010, the AADRs among black males generally exceeded those among nonwhite males for all conditions and followed similar temporal trends (Table 2).

Trends Among Males for Selected Combinations of Chronic, Infectious, and Unintentional Injury Conditions Combined (1900-2010)

The selected chronic diseases, infectious diseases, and unintentional injuries assessed in this analysis accounted for 54.3% of the AADRs among males in 1900 and 62.4% in 2010 (Table 1; Figure 2). Trends in the AADRs for these combined conditions varied markedly. Rates per 100 000 population for the selected chronic conditions rose from 670 in 1900 to a peak of 1165 in 1960, an increase of 75.4%, and then declined by 56.8% to 483 in 2010. Rates per 100 000 population for infectious conditions decreased from 631 in 1900 to 96 in 1950, a decrease of 84.9%, and then declined another 80.8% thereafter to 19 in 2010. Unintentional injury rates remained relatively steady through 1930, declining from 147 to 52 per 100 000 population from 1930 to 2010.

Figure 2.

Figure 2.

Open in a new tab

Age-adjusted death rates for selected chronic diseases, selected infectious diseases, and unintentional injury, US males, 1900-2010. Selected infectious disease totals are the sum of the rates for influenza and pneumonia, tuberculosis, and enteritis and diarrhea. Selected chronic disease totals are the sum of the rates for heart disease, stroke, and all cancers; they do not include chronic lower respiratory diseases. Unintentional injury totals are the sum of the rates for unintentional motor vehicle injuries and unintentional nonmotor vehicle injuries. Rates are per 100 000 US 2000 standard population. Data for 1900-1930 are from death registration states; data for 1940-2010 are from the United States. The states providing mortality data are known as “death registration states”; they included 10 states and the District of Columbia in 1900 (40.5% of the US population) and gradually expanded to include all 48 states and the District of Columbia by 1933. Data source: tabulations of causes of death from published compilations and from public-use computer data files, as previously described.4

Trends in Relative Differences in AADRs Between Males and Females for All Causes (1900-2010) and Specific Causes by Race (1920/1970-2010)

The relative differences between male and female all-cause AADRs began a steady increase after 1920, with greater excesses among white males than among nonwhite males until 1990 (Figure 3). Excess all-cause death rates peaked in 1980 among white males and in 1990 among nonwhite and black males and declined thereafter. Despite recent declines, the excess in male AADRs was greater in 2010 (39.7%) than in 1900 (9.1%; Figure 3).

Figure 3.

Figure 3.

Open in a new tab

Relative difference between male and female age-adjusted death rates (AADRs) for all causes and selected causes, by race (white, black, and nonwhite), calculated as ([male AADR − female AADR]/female AADR), 1900-2010, United States. Rates are per 100 000 US 2000 standard population. Data for 1900-1930 are from death registration states; data for 1940-2010 are from United States. The states providing mortality data are known as “death registration states”; they included 10 states and the District of Columbia in 1900 (40.5% of the US population) and gradually expanded to include all 48 states and the District of Columbia by 1933. Not all data were available in all years. Disaggregation of race beyond white and nonwhite did not occur until 1970. Data source: tabulations of causes of death from published compilations and from public-use computer data files, as previously described.4

The relative differences in AADRs between males and females varied markedly by cause (Figure 3). Trends for heart disease, cancers, and pneumonia and influenza followed all-cause trends. The AADRs for stroke differed little by sex. Of all categories of disease and injuries studied, excess male death rates were greatest for respiratory system cancers (peaking at an excess of 5.3-fold in 1960), chronic lower respiratory disease (3.9-fold excess for the first year of data in 1970), and MVIs (3.0-fold excess in 1910). AADRs for MVIs showed the greatest excess in male mortality during the entire period, with large excesses more recently for UNMVIs, respiratory system cancer, chronic lower respiratory disease, and tuberculosis.

Discussion

Our analysis of AADRs and major CODs by sex and race across 110 years in the United States demonstrates substantial progress in increasing longevity. However, among several age groups and racial populations, progress was interrupted early in the 20th century and again in 1970. Trends also suggest a need to further explore the factors contributing to sustained (and even increased) excess death rates among males compared with females. Changes in diagnostic and classification acumen may have influenced trends, particularly for chronic diseases,4 and improvement in the representation of the US population in vital records by the mid-20th century may also have affected trends. These changes, however, were unlikely to have substantially affected the major trends or trend differences observed in our analysis.

For several age groups of white and nonwhite males, death rate declines between 1900 and 2010 were interrupted by increases. One cluster of increases occurred among men aged ≥35 beginning in 1910 and continuing until 1940. Another cluster occurred among white and nonwhite males aged 15-44, beginning in 1970 and continuing into the next 2 decades for a few subgroups. This cluster, along with a similar cluster among females at the same time,4 precedes by more than a decade the onset of the US opioid overdose epidemic. The increase in UNMVI deaths among white men after 1990 may be associated with the opioid overdose epidemic.5

Although the AADRs for most major conditions declined during the study period, we found 3 principal exceptions—heart disease, all cancers, and MVIs, which rose substantially before declining. During the study period, male AADRs exceeded female AADRs in every decade. The greatest increase in the relative difference occurred during 5 decades: from 16% in 1930 to almost 65% in 1980—about 4 times the 1930 excess. Although the excess then dropped to about 40% in 2010, it was still higher than the excess of 9% in 1900. Death rates for respiratory system cancer had the largest excess, followed by MVIs, then, to a lesser extent, UNMVIs, and, more recently, chronic lower respiratory diseases.

Although data are scarce for the first half of the 20th century, we have potential explanations for differences in AADRs between males and females, including changes in social determinants, biological markers, and public health and medical interventions. Women have made relative gains in the 20th century in education, employment, and income—all associated with long-term health benefits.4 From 1900 to 1950, high school graduation rates among females increased from <30% to approximately 80%.6 Postsecondary education for women also increased.7 Between 1940 and 1970, the median years of schooling increased for nonwhite and white females.8 Increases in proportions of women in the workforce, married women working outside the home, and women in professions—and a narrowing of the gap between earnings among men and women7—all may have contributed to improved economic well-being for women.9 These factors may partially explain the increased mortality gap between males and females. The health of women has improved more rapidly overall than the health of men, in part because of women’s gains in socioeconomic position.

Another major factor contributing to excess male deaths throughout the 20th century is smoking. Waldron estimates that approximately half of the male–female difference in mortality in the United States is attributable to smoking.10 Men took up cigarette smoking earlier in the 20th century than women did; more men smoked, and they smoked more cigarettes. Men also began to quit smoking before women did.11 These differences may largely account for the differences in death rates for cardiovascular disease, cancers (respiratory system cancers in particular), and lower respiratory disease—the initial rise and subsequent decline as well as the earlier peak for heart disease than for cancers. Despite reductions in smoking prevalence among males since the 1960s, smoking prevalence is higher among males than among females.11

During the 20th century, rates of hypertension declined more rapidly among females than among males.12 Men generally have had higher rates of hypertension than women since 1960.13 Recent data14 suggest that rates of high total cholesterol among males and females are nearly equal but that significantly more males have low levels of high-density lipoprotein. Combined, these excesses among males are consistent with higher death rates for heart disease among males. In contrast, females have had higher rates of obesity since 1960,15 a risk factor also associated with heart disease mortality. Males and females report similar rates of physical activity.16

The apparent lack of association of trends in smoking and hypertension with stroke mortality is puzzling. However, a study from Britain indicates that 2 distinct forms of stroke have different risk factors and that trends in cerebral infarction mortality follow trends in coronary heart disease, while trends in hemorrhagic stroke do not.17

Male engagement with the health care system differs from that of females and may also contribute to excess male mortality. Among persons aged >18, females are more likely than males to report having a usual place of health care18 and more likely than males to visit physician offices and hospital outpatient departments.19 Males without a usual source of care may not receive preventive services available for 9 of the 10 major CODs.20 The use of 8 recommended preventive services is greater among females than among males.21

Remarkable among the comparisons of male and female mortality trends is the male excess in MVI deaths. Although all-cause excess trends do not surpass 69%, MVI trends generally range from 122% to 300% (ie, a male maximum rate that is 4 times that of females). In 2017, men drove approximately 1.09 times as many miles as women,22 and male drivers were associated with 2.43 times as many motor vehicle fatalities.23

Plausible explanations for trends in male excess MVI fatalities are available. In a US sample in 2006 and 2008, men reported 3.1 times as many episodes of alcohol-impaired driving as did women.24 Men are also approximately 50% more likely than women to drive above the speed limit,25 and 9%-13% of men are less likely than women to use seat belts.26 Moreover, male and female drivers involved in speeding-related motor vehicle fatalities are more likely to have elevated blood-alcohol levels and to not use seat belts than nonspeeding drivers.27 Diverse programs and policies are effective in reducing motor vehicle fatalities.28,29

Although male mortality declined markedly from 1900 to 2010, substantial room for improvement exists. As recognized in Healthy People 2020, much can be achieved by the equitable distribution of social determinants of health.30 Further health gains can be achieved by reductions in behaviors such as smoking, physical inactivity, and the consumption of foods that promote obesity, hypertension, and high cholesterol. Finally, the full use of multiple preventive and curative health interventions would also contribute to increased longevity.

Limitations

This study had several limitations. First, mortality is not an optimal measure of the burden of disease and suffering and, thus, is not ideal for planning and monitoring interventions to improve public health. However, mortality is a measure available over the time span examined. Second, available mortality data are imperfect for many reasons. Because data on all states in the United States became available only in 1933, early rates may not be representative of the US population, and early trends are assessed in different populations over time. Third, analysis of mortality rates is limited by the consistency and completeness of data available. Completeness of population and mortality reporting is likely to have increased substantially since 1900. Incomplete reporting leads to underestimates of death rates.

Fourth, the comparability of ICD categories between revisions is commonly assessed by the comparability ratio, which compares the number of deaths attributed to a given cause using the previous ICD revision to the number of deaths in the same year attributed to the comparable cause for the subsequent ICD revision. For example, comparability ratios among 56 causes for the sixth and fifth revisions range from 0.57 (for “sepsis of pregnancy, childbirth, and puerperium”) to 1.79 (for “acute nephritis and nephritis with edema, including nephrosis”).31 There is also evidence that coronary heart disease is often overdiagnosed,32 which may inflate coronary heart disease death rates and, if overdiagnosis has changed over time, make apparent trends inaccurate. Finally, racial data were not specific for nonwhite groups (eg, black, Asian, American Indian) until recent decades, and racial classification in health statistics is problematic (eg, inconsistency of coding in numerator and denominator data).33 The specification of black populations among nonwhite populations in mortality data became available only in 1970.

Conclusion

The mortality data presented in our study demonstrate that both males and females are likely to have been harmed by exposures such as smoking and motor vehicles early in the 20th century but are likely also to have benefited from the environmental, behavioral, social, technological, and medical advances of the century—especially in the second half. Both sexes gained >30 years of life expectancy during the century. Nonetheless, male death rates continue to exceed those of females at every life stage, and the relative difference in 2010 exceeded that in 1900. Some excess may be attributable to risk-taking behavior among men, which may include lower rates of use of health care services and greater participation in harmful behaviors. Males also have higher rates of some biological markers than women. All of these features provide corresponding opportunities for health improvement, and evidence-based interventions are available to do so.34

Acknowledgments

The authors recognize the contributions of Dorit Carmelli, PhD, and Li Shu-Chuan Cheng, PhD, from SRI International for compiling the mortality data prior to 1968 and participating in the design and preliminary analysis of these data. We are grateful for the guidance of Daniel Friedman for methodological advice and the CDC WONDER team for their technical support and assistance.

Footnotes

Declaration of Conflicting Interests: The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The work of Dr Parrish for this publication was supported by cooperative agreement grant number U38OT000216 from the Centers for Disease Control and Prevention (CDC). The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the US Department of Health and Human Services or CDC.

ORCID iD: Wanda K. Jones, DrPH Inline graphic https://orcid.org/0000-0002-2903-711X

References

  • 1. Centers for Disease Control and Prevention. Death rates and life expectancy at birth. 2018. https://data.cdc.gov/NCHS/NCHS-Death-rates-and-life-expectancy-at-birth/w9j2-ggv5. Accessed June 26, 2019.
  • 2. Jones DS, Podolsky SH, Greene JA. The burden of disease and the changing task of medicine. N Engl J Med. 2012;366(25):2333–2338. doi:10.1056/NEJMp1113569 [DOI] [PubMed] [Google Scholar]
  • 3. Wilbur CL, King WA. Manual of International Classification of Causes of Death. Washington, DC: US Government Printing Office; 1902. [Google Scholar]
  • 4. Hahn R, Chang MH, Parrish RG, Teutsch SM, Jones WK. Trends in mortality among females in the United States, 1900-2010: progress and challenges. Prev Chronic Dis. 2018;15:E30 doi:10.5888/pcd15.170284 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Dowell D, Arias E, Kochanek K, et al. Contribution of opioid-involved poisoning to the change in life expectancy in the United States, 2000-2015. JAMA. 2017;318(11):1065–1067. doi:10.1001/jama.2017.9308 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Heckman JJ, LaFontaine PA. The American high school graduation rate: trends and levels. Rev Econ Stat. 2010;92(2):244–262. doi:10.1162/rest.2010.12366 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Caplow T, Hicks L, Wattenberg B. The First Measured Century: An Illustrated Guide to Trends in America, 1900-2000. Washington, DC: American Enterprise Institute; 2001. [Google Scholar]
  • 8. US Census Bureau. Historical Statistics of the United States, Colonial Times to 1970. Washington, DC: US Department of Commerce, US Census Bureau; 1975. [Google Scholar]
  • 9. Katz MB, Stern MJ, Fader JJ. Women and the paradox of economic inequality in the twentieth century. https://repository.upenn.edu/cgi/viewcontent.cgi?article=1047&context=spp_papers. Accessed June 26, 2019.
  • 10. Waldron IJ. Patterns and causes of gender differences in smoking. Soc Sci Med. 1991;32(9):989–1005. doi:10.1016/0277-9536(91)90157-8 [DOI] [PubMed] [Google Scholar]
  • 11. Jamal A, Phillips E, Gentzke AS, et al. Current cigarette smoking among adults—United States, 2016. MMWR Morb Mortal Wkly Rep. 2018;67(2):53–59. doi:10.15585/mmwr.mm6702a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Goff DC, Howard G, Russell GB, Labarthe DR. Birth cohort evidence of population influences on blood pressure in the United States, 1887-1994. Ann Epidemiol. 2001;11(4):271–279. doi:10.1016/S1047-2797(00)00224-6 [DOI] [PubMed] [Google Scholar]
  • 13. Burt VL, Cutler JA, Higgins M, et al. Trends in the prevalence, awareness, treatment, and control of hypertension in the adult US population data from the health examination surveys, 1960 to 1991. Hypertension. 1995;26(1):60–69. doi:10.1161/01.hyp.26.1.60 [DOI] [PubMed] [Google Scholar]
  • 14. Carroll MD, Fryar CD, Nguyen DT. Total and high-density lipoprotein cholesterol in adults: United States, 2015-2016. NCHS Data Brief. 2017;(290):1–8. [PubMed] [Google Scholar]
  • 15. Flegal KM, Carroll MD, Kuczmarski RJ, Johnson CL. Overweight and obesity in the United States: prevalence and trends, 1960-1994. Int J Obes Relat Metab Disord. 1998;22(1):39–47. [DOI] [PubMed] [Google Scholar]
  • 16. Brownson RC, Boehmer TK, Luke DA. Declining rates of physical activity in the United States: what are the contributors? Annu Rev Public Health. 2005;26:421–443. doi:10.1146/annurev.publhealth.26.021304.144437 [DOI] [PubMed] [Google Scholar]
  • 17. Lawlor DA, Smith GD, Leon DA, Sterne JA, Ebrahim S. Secular trends in mortality by stroke subtype in the 20th century: a retrospective analysis. Lancet. 2002;360(9348):1818–1823. doi:10.1016/S0140-6736(02)11769-7 [DOI] [PubMed] [Google Scholar]
  • 18. Blackwell D, Villarroel M, Clarke T. Tables of Summary Health Statistics for US Adults: 2014 National Health Interview Survey. Hyattsville, MD: National Center for Health Statistics; 2015. [Google Scholar]
  • 19. National Center for Health Statistics. Health, United States, 2016: With Chartbook on Long-Term Trends in Health. Report no 2017-1232 Hyattsville, MD: National Center for Health Statistics; 2017. [PubMed] [Google Scholar]
  • 20. Fox JB, Shaw FE. Receipt of selected clinical preventive services by adults—United States, 2011-2012. MMWR Morb Mortal Wkly Rep. 2015;64(27):738–742. [PMC free article] [PubMed] [Google Scholar]
  • 21. Centers for Disease Control and Prevention. Use of selected clinical preventive services among adults—United States, 2007-2010. MMWR Morb Mortal Wkly Rep. 2012;61(suppl):1–2. [Google Scholar]
  • 22. US Department of Transportation, Federal Highway Administration. National Household Travel Survey. 2017 https://nhts.ornl.gov. Accessed June 26, 2019.
  • 23. Insurance Institute for Highway Safety. Fatality facts 2017: gender. https://www.iihs.org/topics/fatality-statistics/detail/gender . Accessed October 24, 2019.
  • 24. Bergen G, Shults RA, Beck LF, Qayad M. Self-reported alcohol-impaired driving in the US, 2006 and 2008. Am J Prev Med. 2012;42(2):142–149. doi:10.1016/j.amepre.2011.10.015 [DOI] [PubMed] [Google Scholar]
  • 25. National Highway Traffic Safety Administration. National Survey of Speeding and Unsafe Driving, Attitudes and Behaviors. Volume II—findings report. http://www.trb.org/Main/Blurbs/153203.aspx. Published 2003. Accessed June 26, 2019.
  • 26. Centers for Disease Control and Prevention. Vital signs: nonfatal, motor vehicle-occupant injuries (2009) and seat belt use (2008) among adults—United States. MMWR Morb Mortal Wkly Rep. 2011;59(51):1681–1686. [PubMed] [Google Scholar]
  • 27. US Department of Transportation, National Highway Traffic Safety Administration. Traffic safety facts: 2014 data. Speeding. https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812265. Accessed June 26, 2019.
  • 28. The Guide to Community Preventive Services. Findings for motor vehicle injury. https://www.thecommunityguide.org/content/task-force-findings-motor-vehicle-injury. Published 2019. Accessed June 26, 2019.
  • 29. Goodwin AH, Thomas L, Kirley B, Hall W, O’Brien NP, Hill K. Countermeasures That Work: A Highway Safety Countermeasure Guide for State Highway Safety Offices: 2015. Washington, DC: US Department of Transportation, National Highway Traffic Safety Administration; 2015. [Google Scholar]
  • 30. US Department of Health and Human Services, Office of Disease Prevention and Health Promotion Healthy People 2020. Washington, DC: Office of Disease Prevention and Health Promotion; 2010. [Google Scholar]
  • 31. Faust MM, Dolman AB. Comparability Ratios Based on Mortality Statistics for the Fifth and Sixth Revisions, United States, 1950. Washington, DC: US Department of Health, Education, and Welfare, Public Health Service, National Center for Health Statistics; 1964. [Google Scholar]
  • 32. Agarwal R, Norton JM, Konty K, et al. Overreporting of deaths from coronary heart disease in New York City hospitals, 2003. Prev Chronic Dis. 2010;7(3):A47. [PMC free article] [PubMed] [Google Scholar]
  • 33. Rosenberg HM, Maurer JD, MacDorman MF, Hoyert DL, Scott C. Quality of death rates by race and Hispanic-origin; a summary of current research, 1999. Vital Health Stat 2. 1999;(128):1–13. [PubMed] [Google Scholar]
  • 34. Community Preventive Services Task Force. The community guide. https://www.thecommunityguide.org. Published 2019. Accessed June 26, 2019. [DOI] [PubMed]

Articles from Public Health Reports are provided here courtesy of SAGE Publications

RESOURCES