Current Costs of Families' Payments Compared to Estimates of Health Plans

Flora Williams, Purdue University
[ to cite ]:
Flora Williams (1975) ,"Current Costs of Families' Payments Compared to Estimates of Health Plans", in NA - Advances in Consumer Research Volume 02, eds. Mary Jane Schlinger, Ann Abor, MI : Association for Consumer Research, Pages: 891-904.

Advances in Consumer Research Volume 2, 1975      Pages 891-904

CURRENT COSTS OF FAMILIES' PAYMENTS COMPARED TO ESTIMATES OF HEALTH PLANS

Flora Williams, Purdue University

[This study contributes to inter-regional Agricultural Experiment Station research project NC-90, "Factors Affecting Patterns of Living of Disadvantaged Families." Cooperating states: California, Hawaii, Illinois, Indiana, Iowa, Kansas, Missouri, Nebraska, Nevada, Ohio, Texas, Vermont, Wisconsin. Journal Paper No. 5643, Purdue Agricultural Experiment Station. West Lafayette. Indiana 47907.]

[Flora Williams is Assistant Professor, Home Management and Family Economics Department, School of Home Economics, Purdue University.]

The current costs of families' fixed payments for medical care were compared to estimated fixed payments under proposed health plans. Families selected randomly from six metropolitan areas were paying an average of $149 per year for total fixed medical costs compared to the estimated $84 insurance cost under the Health Security Plan, $150 under the Employee Health Insurance Plan, and $600 under the Assisted Health Insurance Plan. Total fixed medical costs were composed of expenses for health insurance which averaged $35, payroll deductions which averaged $27, and fixed or regular health costs which averaged $87. Families at every income level had fixed expenses greater than they would have under the Health Security Plan. Families under $7500 income were paying more than they would under proposed EHIP costs but families over $7500 were paying less.

Various plans of paying health expenses are being studied to assure good health care for all Americans. Senator Kennedy (Congressional Record, 1973), when discussing the crisis in rising health care costs, said the answer is "not to cut back on benefits, to raise insurance premiums even more, or to simply offer more insurance to more Americans. The answer is to reform our health care system and bring these costs under control."

The proposals currently before Congress outline in detail the structure, benefits, reimbursement policy, regulation and administration, financing, and special provisions. Not reported is the immediate impact on the consumer's budget allocations. Consumer behavior would probably be adjusted as fixed commitments are established by whichever plan is adopted. Flexibility in expenditures for health care might be decreased even if the coverage is better and the cost pressures of irregular medical expenses are reduced. The immediate impact of any increased fixed commitments for health maintenance might be felt before most consumers realize the improved health benefits.

One question to be asked when studying various health plans for predicting the immediate impact on consumers' financial behavior is: Would one plan more than another more closely match what families themselves pay currently in terms of fixed costs? This question looks only at the impact of changes in fixed commitments regardless of additional benefits or additional expenses for deductibles or expenses not covered by the plans.

What are the fixed expenses families have had for their medical costs? Which of several selected variables influenced the amount of insurance payments by families? What would be the possible changes in amount of money allocated by families for fixed medical costs when the new plan would be adopted nationwide? Answers to these questions might influence the initial reaction of families to any change.

Fixed medical costs in this study were defined as the total money spent by families for health insurance, payroll deductions for health insurance, and health expenses reported by the families as fixed. These costs could be compared to aggregated estimates of fixed costs under Senator Kennedy's "Health Security Plan" and the Administration's "Comprehensive Health Insurance Plan." Not examined were extent of coverage, employer's contributions to insurance coverage, deductibles, or health expenses that would be irregular or not fixed. Therefore, such provisions in the health plans are not discussed in depth or included in the cost comparisons for this report.

The proposed Health Security -Act of 1973 would cover every resident with health insurance regardless of residence, work status, income, size of family, or history of medical problems. Payment for the coverage would be based on a person's ability to pay. The Health Security Plan converts

The existing Medicare hospital insurance payroll taxes into Health Security taxes, and raise the rates to 1 percent on employees and 3.5 percent on employers. ...raises the wage base for the employer tax from $12,000 (as present law provides after 1973) to $15,000, with subsequent further increase if wage levels rise, so that the Health Security Wage base will always be 125 percent of the Social Security wage base. ...adds a new 1 percent Health Security tax on unearned income (unless such income is less than $400 a year), subject to the same maximum on taxable income as is applicable to the employee and self-employment taxes. Taxable unearned income is adjusted gross income up to the stated maximum, minus wages and self-employment income already taxed for Health Security purposes (excluding certain items of income specifically excluded from the other taxes and excluding $3,000 in unearned income for persons over age 60) (Congressional Record, January 31, 1973, p. 30).

The two structures of the Comprehensive Health Insurance Plan (The White House, 1974) are the (a) Employee Health Insurance Plan (EHIP) whereby all employers would be required to offer the basic insurance plan or Health Maintenance Organization coverage while the plan would also be available to self-employed and non-working families; and (b) Assisted Health Insurance Plan (AHIP) whereby families would enroll who are not covered under EHIP, are below $5,000 income, are non-working or high risk with income between $5,000 and $7,500, or work in unusually high risk insurance type jobs.

Employers would contribute 65 percent of the premium expenses for covered employees with Federal Government subsidy if an employer's payroll rises by more than 3 percent due to required contributions to coverage. Eventually the employer would pay 75 percent of premium costs and employees the remaining 25 percent. "Premiums for employer groups of 51 or more employees and other families and groups being offered EHIP would be negotiated between employer and other groups and the insurance carrier" (The White House, 1974, p.4). The average family cost for health insurance premiums would be about $600 per year with the employee paying approximately $150 per year with the employer paying approximately $450 for each employee (Congressional Record, February 6, 1974, S1341).

The premiums, deductibles, coinsurance, and maximum liability would be related to income under the AHIP. Working families with income below $5,000 would pay no premiums at all.

METHODOLOGY

An interdisciplinary, regional research project, NC-90, entitled "Factors Affecting Patterns of Living in Disadvantaged Families" provided the data used for this study. The disciplines participating in the project were: rural and family sociology, child and educational psychology, economics, family economics, home management and child development. Thirteen state experiment stations were represented in the project. Members of the NC-90 committee developed the interview schedule and provided instructions in training interviewers. Interviewers were indigenous to the six urban disadvantaged areas supplying data for this part of the study. Data were collected during 1970 and subsequently analyzed.

Source of the Data

Data collected randomly from families in six metropolitan areas were analyzed for purposes of this study. The homemaker served as the respondent. Area samples were drawn by the Survey Section of the Iowa State University Statistical Laboratory for the studies in Indiana, Ohio, and Nevada. East Chicago, Indiana and Toledo, Ohio families were selected from areas designated as poverty tracts by the Bureau of the Census (Maps of Major Concentrations of Poverty, 1966). The population from which the Nevada sample was drawn consisted of families living in designated "low cost housing" areas as determined through the compilation of information by the City Planning Department of Las Vegas. Hawaii data were collected within eight census tracts in metropolitan Honolulu on the island of Oahu. In these tracts during the three year period of 1964-1967, 40 percent or more of the families had an annual income of $3,000 or less. The Statistical Laboratory at the University of Illinois sampled a low income area based upon the assessed valuation of property in Champaign-Urbana, Illinois. In Superior, Wisconsin the families were selected from an economically depressed area.

To be eligible for the personal interview, a family had to have at least one child under age 18 and a homemaker, either gainfully employed or not, between 18 and 64 years of age. Families varied widely in the distribution of income, education, number in family, ethnicity and amount of payments for insurance. Although areas were considered disadvantaged, the majority of the families would not be classified as low income as defined by the Social Security Administration (Orshansky, 1969).

Characteristics of the Sample

Characteristics of the families in this sample from metropolitan areas varied widely in their demographic characteristics. By ethnicity, 49 percent were white, 30 percent were black, 5 percent were oriental, 12 percent were Spanish-Americans, and 4 percent were of other ethnic origins. The mean total disposable income was $8,434.51.

Characteristics of the sample families in 1970 may be compared to national figures in 1970 in order to judge application of findings to families other than those in the sample. The percentage of the families in the sample who had over $10,000 income was 22, whereas in the nation as a whole it was 49 (Statistical Abstract, 1971, p. 322). As defined by the Social Security Administration, 15 percent of the families were classified as below the poverty level (below 100 income indeX) and about ten percent were classified in the near-poverty level (between 100 and 125 income index). The national figure for families below the poverty level was also 15 percent (Statistical Abstract, 1971, p. 323). The mean number in the family was-five. The mean age of the husband was 37. Median years of schooling completed for husband and wives was 10.8. The national median was 12.2 (Statistical Abstract, 1971, p. 111).

Data Analysis

Data from these six metropolitan areas were combined for analysis purposes. The number of families supplying useable data was 202 from Hawaii, 287 from Illinois, 193 from Indiana, 225 from Nevada, 170 from Ohio, and 208 from Wisconsin. In some analyses the number of cases was less when information was incomplete.

The one-way analysis of variance technique was used to determine differences of means between fixed health expenditures by selected variables. The Pearson's product-moment correlation was used to determine the relationship of selected variables with fixed medical expenditures. Multiple regression analysis was used to measure the simultaneous effects of different variables and to indicate the most significance variables related to payment for health insurance. Unless indicated otherwise, the .05 level of significance was used to determine statistically significant differences and relationships.

Selected Variable Definitions

Health insurance expenditures were reported by the families in answer to the questions, "Could you list the bills or expenses you are supposed to meet regularly? That is, the sort of things you feel are rather fixed, or that you are obliged to pay, or that you have promised to pay each week or every month or every so often, or that are taken out of a paycheck?" The amount for a given period was reported. The number of payments due the past 12 months, the dollar amount per year, and dollar amount that was a payroll deduction were recorded.

Although health insurance premiums buy various types and amounts of coverage, this study looked at only the annual premium payment by the family for that coverage. Employers may have paid all or part of health insurance premiums but this study looked only at the payments or contributions made by the family.

Health costs were similarly reported. The dollar amount considered as the medical bills or expenses the family was supposed to meet regularly, felt obligated to pay, or promised to pay were recorded as fixed expenditures. Examples of these were regular purchase of drugs such as insulin or periodic payments on a hospital or dental bill. The total amount per year was calculated after recording the number of payments due during the past 12 months.

Life insurance payments were reported in the same manner as health insurance payments and calculated on an annual basis. This variable was included as an indicator of insurance orientation.

Total disposable income was the flow of purchasing power received from investments, social security benefits, job-related benefits such as disability or unemployment insurance, armed services benefits, welfare benefits, legal arrangements, and earned income. Earned income did not include FICA or income taxes. Income was grouped to correspond to annual income groups in the schedule of cost estimates for the "Comprehensive Health Insurance Act of 1974."

Income index was computed by dividing the total family disposable income by its poverty threshold and multiplying by 100. The poverty threshold as measured by Orshansky (1969) attempts to take into account the composition and the number in the family and is based on the assumption that no more than a third of the family income is used for food. An income index of less than 100 indicates that the family's income is below its poverty threshold when this threshold is based on the economy food plan of the USDA, a plan designed for short term emergency situations.

In addition to the objective measure of income adequacy, the perceived adequacy of income was examined. Perceived adequacy was reported from the question, "To what extent do you think your income is enough to live on?" Responses were: (l) not at all adequate, (2) can meet necessities only, (3) (3) can afford some of the things we want, but not all we want, (4) can afford about everything we want, and (5) can afford about everything we want and still save money.

A financial problem of money for medical costs was identified when the respondent was asked, "Do you have the problem of not enough money for dentist, doctor, or medicine?" Responses were: (1) never, (2) seldom, (3) sometimes, and (4) often.

Ethnicity was observed by the interviewer and respondents were classified as white, black, oriental, Spanish-American or other. "Other" included Polynesians, Indians, mixture, and unknown.

Education referred to the number of years completed in school. Education was reported for the respondent and for her husband.

Occupation was determined by asking the respondent to describe the kind of job held and what business or product was involved. The job contributing the most to the respondent's or her husband's total income was used to determine the occupation held. The scale with 29 levels was based on North-Hatt and Duncan's rankings (Reiss, 1961).

Disability was derived from the question, "Is anyone in this family sick all the time or disabled in any way?" Responses were: (1) not disabled, (2) not limited in any of the following ways, (3) able to work, keep house, go to school, or play, but limited in other activities, (4) able to work, etc., but limited in kind or amount of work, etc., and (5) not able to work, etc., at all.

FINDINGS

Fixed Expenses Families Have Had for Medical Costs

Health Insurance Excluding Payroll Deductions. Eighty-three percent of all families did not report having fixed expenses for health insurance excluding payroll deductions, although the percent covered by insurance is not known for this sample. This large number of families with no insurance expenses brought the average amount for health insurance payments to a low of $35.

The average for just those families who had any expenses for health insurance was $212 (Table 1). The means did not differ significantly among families in various income groups. The maximum spent for insurance by any family was over $600 for families in the three lowest income levels and just under $500 for the two highest income levels. Families classified as low income, below $5000, were 30.3 percent of the sample. About 90 percent of these low-income families were not paying any health insurance premiums themselves.

TABLE 1

HEALTH INSURANCE, FIXED HEALTH COSTS AND PAYROLL DEDUCTIONS BY INCOME LEVEL

Families who had no fixed payments for health insurance were compared with families who had $1 to $199, those who had $200 to $299, and those who had $300 or more expenses by the analysis of variance test. Significant differences (p = < .01) among groups occurred for mean education of husband, total disposable income and income index. The average years of schooling for the no-expense group was 10.49 compared to 11.14, 12.43 and 11.42 for other groups, respectively. One might argue the explanation that the low educated husbands would be employed in jobs providing some health insurance coverage. The argument probably would not hold, however, because the mean occupational level did not differ significantly among the expense groups and the Pearson-product moment correlation value (r = -.09) for husband's occupational level and expenses for health insurance was negative (p < .05).

The average total disposable income for the no-payment group was $6,934 compared to $8,779, $9,384, $8,040 for the other groups, respectively, by level of health insurance payments. The average income index for the no-expense group was 161 compared to 218, 262 and 247 for the other groups, respectively.

Payroll Deductions. Payroll deductions occurred for only 15 percent of the families. For all families, the average was $28.24. For all those families who had payroll deductions the average amount was $185 (Table 1). Means of payroll deductions differed by income level. The lowest mean was for families in the $2,500 through $4,999 income group.

Payroll deductions for just the husbands averaged $191 (Table 2). Two thirds of the husbands had deducted this amount plus or minus $114. The means for various occupational groups did not differ significantly. However, the percentage having any deductions for health insurance did vary considerably among occupational groups. The highest percentage of men who hat deductions was in the highest income group: professionals, managers, officials, and proprietors.

Health Costs. The average fixed health costs excluding health insurance were $102.73 when all families were analyzed. Of the 28 percent who had any health costs considered fixed or regular, the average cost was $315 (Table 1).

Families in different income levels having fixed health costs were as follows:

Under $2500 15.1 percent

$2,500 through $4,999 21.8 percent

$5,000 through $7,4993 4.0 percent

$7,500 through $9,9993 0.3 percent

$10,000 and over 27.5 percent

Fixed health costs, when they did occur, varied significantly by income level but the relationship was not linear. The highest income families, with $10,000 and over, averaged $417 in fixed health costs followed by the lowest income families, with under $2500, who averaged $404 in fixed health costs. This leads to some interesting speculation as to what is cause and what is effect.

Total Medical Costs. Total fixed medical costs composed of payments for health insurance, payroll deductions for health insurance, and other fixed health costs were analyzed and shown in Table 3. The average for the sample families was $149. The average costs for all families ranged from $77 for families with income under $2500 to $219 for families with $10,000 or more income.

TABLE 2

PAYROLL DEDUCTION BY HUSBAND'S OCCUPATIONAL GROUP

Average medical costs for various income levels were compared. Significant differences were found when all families were included in the analysis but not when just those families having medical costs were included. The over-all mean for these families was $302. The percentage of families at different income levels who had any costs was as follows:

Under $2500 21.0 percent

$2,500 through $4,999 31.1 percent

$5,000 through $7,499 55.6 percent

$7,500 through $9,999 56.0 percent

$10,000 and over 64.5 percent

Among the low-income families, 105 or about 28 percent had some fixed medical costs (Table 3). These families with income below $2500 averaged $377 for medical costs and families with income from $2500 to $5000 averaged $305.

Influence of Selected Variables Upon Amount of Insurance Payments

The multiple regression step-wise technique was applied to find the association of selected variables with the fixed expenses for health insurance. Fifteen variables were selected because they were available in the data collected. Through stepwise regression, seven variables remained as significant contributors at the .10 level and yielded a multiple R of .31.

The standardized multiple regression equation was Y = .33X1 -.25X2 + .15X3 + .12X4 + .09X5 + .06X6 + .04X7 in which

Y = expenses for health insurance

X1 = income index

X2 = total disposable income

X3 = husband's age

X4 = life insurance

X5 = wife's years of schooling

X6 = fixed health costs

X7 = wife's disability index

TABLE 3

TOTAL FIXED MEDICAL COSTS BY INCOME LEVEL

In terms of standardized partial regression coefficients (betas), income index (an objective measure of income adequacy) was the most important in explaining expenses for health insurance. As income index increased, families allocated more of their income to life insurance. Total disposable income, which does not take into consideration the cost of food for the given composition of the family, was negatively associated. Variables eliminated from the equation were number in the family, expenses for car insurance, husband's disability index, income dependability, not enough money for doctor or dentist, husband's occupational level, perceptions of income adequacy, and husband's years of schooling.

The finding that both fixed health costs and woman's disability were positively associated with expenses for health insurance has relevance to the proposals for health insurance premiums regardless of the consumer's medical history or risk. An insurance orientation for some families may have been indicated by the finding that life insurance was positively associated with expenses for health insurance (r = .14).

Health insurance expenses by families themselves were examined in relation to payroll deductions for health insurance. The Pearson-product moment correlation coefficient (r = -.10) was significant beyond the .01 level. As payroll deductions increased, non-payroll expenses for health insurance decreased; although little of the variance in health insurance expenditures by families themselves could be explained by the payroll deductions.

Possible changes for families who were not paying out of their own finances for health insurance would be increases for all under the Health Security Plan and increases for high income families under the Comprehensive Health Insurance Plan (Figure 1).

FIGURE 1

MODEL OF POSSIBLE CHANGES FAMILIES WOULD MAKE UNDER PROPOSED HEALTH PLANS

Some of the families currently not paying for health insurance are covered by insurance through work-related benefits. Other families are not currently covered but would be covered under a proposal when adopted. If these other families are low-income, below $5,000, they would continue to not pay under the "Comprehensive Health Insurance Plan" but would pay one percent of $15,000 under the "Health Security Plan" unless the type of income was specifically excluded from the other taxes (Congressional Record, January 31, 1973, p. 30). Families with high income, regardless of their past coverage, would have to allocate money for insurance premiums that heretofore they had not done. Under the "Comprehensive Health Insurance Plan," their payroll deductions would average $150 with the employer contributing an average of $450. Under the "Health Security Plan," one percent of their income in addition to current social security contributions would be paid on 125 percent of the social security wage base.

Comparison of Current Medical Costs Compared to Estimated Costs. Which proposed plan most closely matches what families were paying for fixed medical costs? As shown in Table 4, at every income level, families as a whole were paying more than the cost estimates for the Health Security Plan. The overall average paid was nearly the same as the cost estimate for the EHIP, but families with income under $7,500 were paying more and families with $7,500 and over were paying less. Under AHIP families with income under $5,000 would pay nothing whereas families with $5,000 or over would pay considerably more than they did pay. The projected average would be $600 compared to the overall average of $149 paid by the sample families.

Of families in the lowest income group, the majority (80 percent) did not have any medical expenses considered fixed. Under the Health Security Plan many of these would pay one percent of their income. For those families who had had fixed expenses which averaged $377 this reduced fixed cost would be extremely helpful. Under EHIP and AHIP these lowest income families would pay nothing. A similar change would occur for families in the $2,500 through $4,999 level.

Of families in the $5,000 and $7,500 levels, the minority having no fixed expenses would begin paying one percent under the Health Security Plan and $105 and $210, respectively, under the EHIP. These amounts would be less than the average of fixed medical expenses ($294 and $249) for families having such expenses. Under AHIP they would have to pay more than their average fixed expenses. Of families in the $7,500 through $9,999 level, they would pay an average of $600 compared to an average $249 for those who had any expenses and none for the others.

Of families with $10,000 and over income, the 22 percent that heretofore did not have fixed expenses would pay $100 to $150 under Health Security Plan, $315 under EHIP and $900 under AHIP. For families with fixed expenses averaging $340 and ranging from $20 to $3400 some would pay considerably less under any of the three estimates. Many of those working in high risk occupations or who were health risks would be under AHIP and would pas more than previously.

SUMMARY AND CONCLUSIONS

The current cost of families' fixed payments for medical care was compared to estimated fixed payments under proposed health plans. Families selected randomly from six metropolitan disadvantaged areas were paying an average of $149 per year for total fixed medical costs. The average for just those families who had any fixed medical costs was $305. In contrast, estimates for insurance costs under the Health Security Plan averaged $84 and under the Comprehensive Health Insurance Plan averaged $150 for those qualifying for the Employee Health Insurance Plan and $600 for those qualifying for the Assisted Health Insurance Plan.

Compared to cost estimates for the proposed Health Security Plan, families at every income level, on the average, were paying more for fixed medical expenses. Compared to the estimates for EHIP in the proposed Comprehensive Health Insurance Plan, families on the average were paying the same amount. All families under $7,500 income were paying more than they would under proposed EHIP costs but families over $7,500 were paying less. However, the families that had fixed medical costs were paying considerably more than estimated costs under EHIP. Compared to AHIP of the Comprehensive Health Insurance Plan, families over $7,500 income were paying considerably less.

TABLE 4

CURRENT AND ESTIMATED ANNUAL COSTS (FIXED) FOR FAMILIES BELOW AGE 65 BY INCOME GROUP

In this sample, 51 percent of the families did not consider any expenses, fixed or regular, for medical costs. Forty-three percent of these families were low income and, therefore, would not pay under proposed AEIP and EHIP but all would pay under the Health Security Plan unless income was specifically excluded as a determinant.

Health insurance, other than payroll expenses, was paid by 17 percent of the families themselves. The average spent on health insurance premiums was $212 for those that did pay. Variables associated with amount spent on health insurance, in order of importance, were income index, total disposable income (negatively), husband's age, life insurance, wife's years of schooling, fixed health costs, and wife's disability index but not number in family, expenses for car insurance, husband's disability, income dependability, perceptions of income adequacy and husband's years of schooling. Income index took into account the composition and number in the family as well as total disposable income.

Payroll deductions for health insurance were not commonly found among families. Fifteen percent had such deductions. Under the EHIP the employee would deduct 35 percent of the insurance payment and the employer would contribute 65 percent. Heretofore, for these sample families most employers contributed all or none. Adjustments would be made in this realm, especially for the 85 percent of the families not having had deductions. For those having deductions, the $185 average is higher than the proposed $150.

Fixed or regular health costs other than health insurance occurred for 28 percent of the families. Their average cost was $315.

Total fixed medical costs which were composed of expenses for health insurance, health costs, and payroll deductions that were fixed or regular for families varied with income levels. Families below $2,500 income averaged total costs of $77; those between $2,500 and $5,000 averaged $96; those between $5,000 and $7,499 averaged $164; those between $7,500 and $9,999 averaged $139; and those with $10,000 and above averaged $273. The low averages occurred because only about half of the families had such expenses.

The immediate impact upon families as consumers will probably be the change in their budgets for fixed allocations for medical expenses. The increased benefits or coverage will be realized as programs are implemented and utilized by families. Although the sample families at every income level, on the average, were paying more for fixed medical expenses than the estimates for the proposed Health Security Plan, over half had no expenses as fixed. Those families would lose their flexibility in health expense allocations under this plan. Many of the other families would pay considerably less.

Families under $5,000 income would pay nothing for the Comprehensive Health Plan which would be considerably less for 44 percent of the families and no change for the others. Families with $5,000 to $7,500 incomes, on the average, would pay about the same under AHIP but less under EHIP than they were paying for fixed health care. Of course the 29 percent not having fixed expenses in this income group would have considerable increases in fixed expenses. The greatest impact upon budget allocations would be for families with incomes over $7,500 who would qualify for AHIP. These cost estimates are much higher than they paid previously.

REFERENCES

Congressional Record. 93d Congress, First Session, January 31, 1973, 119(17), 30.

Congressional Record 93d Congress, Second Session, February 6, 1974, 120 (12), S1341.

Orshansky, M. How poverty is measured Monthly Labor Review, February, 1969, 92 (6), 32-62.

Reiss, A. G. Occupation and social status. New York: Free Press of Glencoe, 1961.

United States Bureau of the Census. Maps of major concentrations of poverty in standard metropolitan statistical areas of 250,000 or more populations Vol. 1. Prepared for the Office of Economic Opportunity. Washington, D.C.: U.S. Government Printing Office, 1966, iv.

United States Bureau of the Census. Statistical abstract of the United States: 1971. (92d edition) Washington, D.C., 1971.

The White House. Fact sheet: The comprehensive health insurance plan. Office of the White House Press Secretary, February 6, 1974, 6.

----------------------------------------