payments at a monthly rate of $21.8 million. About 103,000 other persons had established entitlement to these payments, but the payments were being withheld at the close of December because these individuals were receiving cash payments under public assistance or because the amount of their government pension was at least as large as the special social security payment.

The number receiving monthly benefits increased by 1.9 million (9 percent) in 1966 and totaled almost 22.8 million at the end of the year. The total amount of monthly benefits in currentpayment status at the end of December-$1,639 million-was $122 million (8 percent) larger than at the end of 1965. Retired workers and their dependents numbered 14.8 million and made up 65 percent of all beneficiaries. Their monthly benefits amounted to $1,116 million and represented 68 percent of the total payable. Almost 5.4 million persons, or 24 percent of the total number, were receiving survivor monthly benefits that amounted to $366 million-22 percent of the total. Persons receiving the special payments under the 1966 amendments accounted for 3 percent of the total number and 1 percent of the monthly amount. Nine percent (2.0 million) of all beneficiaries were disabled workers and their dependents, and the $136 million they were receiving represented 8 percent of the total monthly benefit amount.

At the end of 1966, about 17.0 million or 89 percent of the Nation's 19.2 million persons aged 65 or over were eligible for monthly benefits under the program. Some 15.6 million persons aged 65 or over or 81 percent of the total were actually receiving monthly benefits, about 1.3 million more than a year earlier.

Benefit payments in 1966 totaled $20.0 billion -about $1.7 billion higher than the amount paid in 1965. Retired workers and their dependents and the survivors of deceased workers were paid monthly benefits amounting to almost $18.0 billion, a 9 percent increase from the amount paid. in 1965. Benefits to disabled workers and their dependents rose 13 percent to almost $1.8 billion. Lump-sum death payments increased by $20 million to $237 million, and monthly payments to noninsured persons aged 72 and over-first payable for October 1966-totaled $44 million.

During January-December 1966, 4,300 disabledworker families were awarded benefits that were either partially reduced or completely withheld

as a result of the workmen's compensation reduction provision in the 1965 amendments. At the end of December, the 3,075 disabled-worker beneficiary families currently affected by the reduction provision were entitled to monthly benefits under the social security program totaling $567,000, $371,000 of which was being withheld.

As a result of the revised definition of disability provided by the 1965 amendments, the number of workers for whom a period of disability was established during 1966 set a new high-287,000. Almost 23,000 applicants for child's benefits who were aged 18 or over were found to have a disability that began before they had reached age 18; the number was about 1,000 more than that in 1965.

had also been received. In December, almost 133,000 hospital admission notices and close to 19,000 home health notices were reported.

For the first 6 months of the health insurance program, more than $900 million was drawn from the hospital insurance trust fund by intermediaries for benefit payments to institutions such as hospitals and home health agencies providing care to beneficiaries. During the same period, more than $127 million was drawn from the supplementary medical insurance trust fund to pay for benefits provided under the medical insurance program.

Besides the Government agencies involved in administering the health insurance program for the aged, many other institutions and individuals have given the program its support and cooperation. Program operations in the first half-year have involved the staffs of some 6,700 hospitals, some 200,000 practicing physicians and their staffs, the staffs of 1,400 participating home health agencies, 74 Blue Cross organizations, 33 Blue Shield Plans, 15 insurance companies and several independent health insurers, and more than 75 group-practice prepayment plans. The first data from the medical insurance sample of the Current Medicare Survey, with a description of the survey procedures and findings, are shown on pages 4-9.

TOTAL INCOME-MAINTENANCE PAYMENTS
HIGHER AT YEAREND

Total income-maintenance payments under public programs rose in December to $3.9 billion$565 million or nearly 17 percent more than the amount in December 1965. Most of the growth during the year occurred in OASDHI ($375 million) and in public assistance ($107). The substantial rise in OASDHI benefits results primarily from the payment of health insurance benefits for the aged, beginning July 1966, and the special payments to persons aged 72 and over, beginning October 1966. The new programs of medical assistance under title XIX of the Social Security Act (in 28 States by the end of 1966) accounted for much of public assistance's rise. Unemployment insurance benefit payments during December reached $167 million—the lowest level for that month since 1956. The rate of unemployment for 1966 was 3.9 percent-somewhat lower than the rate for 1965 and the lowest rate since 1953.

RISE IN PA MONEY PAYMENT TOTALS

December changes in public assistance caseloads followed the same pattern as November changes for the money payment programs. Seasonal influences on employment conditions were reflected in the rises in the number receiving aid to families with dependent children (97,000) and in the number of general assistance recipients (51,000). About 37,000 of the increase in the number of recipients of aid to families with dependent children occurred in the unemployedparent segment of the program; California, New York, Pennsylvania, and Washington accounted for most of this increase and for about half of the rise in the rest of the caseload.

There were 5,500 fewer recipients of old-age assistance, and aid to the blind edged down slightly. A rise of 2,800 was reported in aid to the permanently and totally disabled. The special monthly cash payments under the social security program-first payable for October 1966 to noninsured individuals aged 72 or over-continued to be a factor in the discontinuance of money payments to assistance recipients. In October-December, an estimated 20,000 discontinuances were related to the receipt of such payment, and 15,000 of these terminations were in old-age assistance.

In medical assistance for the aged, payments were made in behalf of 800 more persons than in November. In medical assistance under title XIX of the Social Security Act, States reporting on the number in whose behalf payments were made showed a drop of 55,000 recipients.

Total expenditures for assistance were $577.1 million in December-$5.9 million more than in November. Total money payments rose $8.0 million, largely because of the increases of $4.5 million in aid to families with dependent children and $1.7 million in general assistance. Old-age assistance, aid to the blind, and aid to the permanently and totally disabled also showed rises. Total medical vendor payments declined $2.1 million, reflecting mainly decreases of $940,000 in medical assistance and an estimated $1 million in general assistance.

New Mexico initiated a program of medical assistance under title XIX in December, bringing to 28 the number of jurisdictions with these programs.

by JACK SCHARFF*

The January issue of the Social Security Bulletin carried a full description of the statistical program established to record and maintain data on the utilization and charges for medical care services covered under the health insurance program for the aged. The statistical system is based on the receipt by the Social Security Administration of bills presented to and paid by fiscal intermediaries throughout the country. Considerable delays in the statistical reporting of current information are inherent in the billing system. The Social Security Administration has therefore initiated a continuing monthly Current Medicare Survey (CMS) to provide current estimates of the hospital and medical care services used and of the charges incurred by persons covered under the program. This article describes the medical insurance sample of the CMS and presents information for July, August, and September 1966— the first 3 months of the program's operation. A subsequent article will describe the hospital insurance sample of the CMS and report data for the initial months of the program.

CMS HIGHLIGHTS

PERIODIC HOUSEHOLD interviews of persons enrolled in the medical insurance program for the aged form the basis for the medical insurance sample of the CMS. Information on the use of and charges for medical care and related services has been collected since the beginning of the program. Highlights of the first 3 months of the program's operation (without adjustments for seasonal variations) reveal that

-during July 1966, approximately one-third of the medical insurance enrollees-5.6 million personsused medical services covered under the program, and of this total, about 10 percent met the $50 deductible during the month

-charges per person using covered medical services in July averaged $31

* Chief, Current Medicare Survey Staff, Division of Health Insurance Studies, Office of Research and Statistics.

-utilization of medical services increases somewhat with age, is greater for women than men, but does not vary materially by region

-the number of enrollees utilizing medical services and total charges incurred remained relatively unchanged in August and September

-approximately the same number of people met the deductible in July, August, and September

-about 2 million persons or 11 percent of the enrollees met the deductible by the end of September -by the end of the first quarter of operation, charges potentially reimbursable by the program totaled about $227 million.

The health insurance program for the aged, popularly known as "Medicare," went into effect on July 1, 1966. The program comprises two related parts: The hospital insurance program (HI) provides financial assistance in paying part of the cost of hospital care, post-hospital extended care and home health services, and outpatient hospital diagnostic services. The supplementary medical insurance program (SMI) provides for payment of part of the charges for physicians' and other medical services.

The statistics for the medical insurance program depend upon receipt by the Social Security Administration of copies of bills transmitted to and paid by about 50 intermediaries throughout the country. It was anticipated, and actual experience during the first months of the program's operation has demonstrated, that there are time lags and delays of varying duration in the receipt and payment of bills by the intermediaries. Several factors contribute to this delay. Beneficiaries may accumulate their bills until the charges exceed the $50 deductible. This process may present special problems to the aged who have difficulties in fully understanding this part of the program. For the beneficiaries whose bills are not assigned to physicians, delays occur when they must present receipted bills to the intermediaries for reimbursement.

Knowledge of the volume of services and charges incurred for specified periods is helpful for administrative purposes to provide a current picture of the extent of obligations incurred by

the Federal Government and to estimate the future claims against the trust fund. The CMS is intended to produce program information on a statistical basis about 3 months after the reference period. It will provide, considerably in advance of the time when data become available from operating records, national estimates of the number of persons receiving medical care and related services and the charges incurred, including deductibles and coinsurance.

SMI SAMPLE DESIGN

To obtain information on the utilization and charges for medical care and related services under the medical insurance program, the survey design calls for monthly personal interviews of nearly 4,000 persons selected from the primary 5-percent statistical sample of those enrolled in this part of the health insurance program.1 The sample, chosen to be self-weighting within 105 primary sampling units, represents the 17.5 million persons residing in the 50 States and the District of Columbia who were enrolled for medical insurance benefits as of July 1, 1966. Persons selected in July remained in the sample through the end of December 1966. A sample of persons was selected for interviews starting in October 1966. This group will remain in the survey for 15 months. The cycle was determined by the fact that any expenses incurred by an individual in the last 3 months of a calendar year and applied to the deductible for that year may be carried over and applied to the deductible for the next calendar year.

The SMI sample, used to provide estimates in this report, consists of: (1) A basic sample of 3,800 individuals who will normally remain in the sample for 15 months and (2) a small incremental sample drawn to include persons "aging in" to the universe and added to the sample each month.

SMI SURVEY PROCEDURES AND DATA COLLECTION

The basic sampling unit is the individual beneficiary. To obtain current information on medical insurance benefits, the beneficiaries are contacted

1 For a description of the 5-percent sample, see Howard West, "Health Insurance for the Aged: The Statistical Program," Social Security Bulletin, January 1967.

periodically to supply the required information. Personal interviews, utilizing a questionnaire and a diary form, are conducted by the Bureau of the Census acting as a collector of data for the Social Security Administration. Experienced field interviewers obtain information about the use of medical care and related services during the preceding month. A careful editing and screening process identifies those items not covered by the program, and, up to the present, only bills incurred for covered services have been coded and tabulated. Charges are accumulated so that the total covered charges for an individual may be located along a continuum from any point below the deductible to any point above.

The questionnaire and diary form are designed to obtain the following items of information: name and address of respondent, date and place of doctor visits, type of physician, condition treated, other medical services received, including covered medical services received in the hospital and nursing home, as well as X-rays, medical tests, ambulance services, and the like. Also included are questions relating to the total amount of the bill, the portion not covered by the program, and the source of payment.

No attempt is made to ascertain charges or services for hospital-based physicians such as pathologists and radiologists, except where the patient is aware of such treatment and is billed separately for the physician's services.

Information on the dollar amount of the physician's bill is often unknown because no bill had been received by the date of the interview. The interviewer attempts to obtain this information in the following month. Experience has proven that the doctor's bill is frequently available during the follow-up interview.

There are several instances where the bill information is not normally available to the beneficiary, regardless of the elapsed time. Welfare beneficiaries, for example, would not generally know the amount of the doctor's bill. Where no information on charges is available, an estimating procedure was established that is based on the assumption that charges will be the same for similar services rendered in the same area. For example, a doctor's home visit in a specific city is valued at the amount last reported in the sample for this type of service in the same geographic

area.

The nonresponse rate, where no interview was obtained, amounted to 9 percent in July, 8 percent in August, and 8 percent in September. Two techniques were employed to impute the required data: First, for persons known to be hospitalized during the survey month and, therefore, not available for interview, the imputation was based on the hospital experience of reported persons. Second, a random substitution of experience of other persons in the same color, sex, and age group was used.

The monthly survey results were derived by multiplication of the sample data by a single weight obtained from the ratio of an independent estimate of the persons enrolled in the medical insurance program as of the beginning of the month to the sample population. For July, tabulated enrollment data were employed as the independent figure. For August, enrollment data were independently estimated by adjusting the July 1 tabulated enrollment for increments of persons reaching age 65 and for decrements of persons who died or dropped from enrollment. The estimated August 1 enrollment was used in a similar manner as a basis for deriving the estimated September 1 enrollment figures.

Special efforts are made to obtain the data for persons in the sample who died during the survey month because these individuals probably had higher utilization of medical services during the survey month than that of other persons. The results of the first month's attempt to obtain this information pointed to the desirability of postponing the interview with the next of kin or any other proxy respondent until the following month. On this basis, interviews were obtained for 93 percent of the known deaths occurring between July 1, 1966, and September 30, 1966.

CMS FINDINGS ON SMI

During the first month of the program's operation, approximately 5.6 million aged persons, or about one-third of the 17.5 million persons enrolled, used medical services covered under the medical insurance program (table 1). No adjustment was made for seasonal variation. Excluded

2 See "Enrollment in the Health Insurance Program for the Aged," Social Security Bulletin, March 1966, pp. 21–24.