Longitudinal Analyses of Four Years of Experience of a Prepaid Comprehensive Medical Care Plan1

The spectacular contributions of science to medicine and the popularization of the results have greatly stimulated the demand on the part of the public for making modern medical benefits more readily accessible to a larger portion of the population. But the very advances which have stimulated this demand and which have focused attention on the preventive aspects of disease have also contributed greatly to the cost of medical care. On the one hand, this increased cost has led to an enormous growth in the application of the insurance principle to prepayment for medical and hospital care; on the other, it has led to new approaches in the organization of medical practice to provide this care. These changes in the methods of paying for and of providing medical care are comparative newcomers among American social institutions. Provision for prepayment of hospitalization began in 1933 with the establishment of Blue Cross plans. Prepayment of medical care is an even more recent phenomenon. Fewer than six persons per 1,000 held surgical and medical insurance in 1939. Today, according to the Health Information Foundation,2 almost 70 per cent of the population is protected or “covered” by some combination of hospital, surgical, or medical insurance. On the organizational side change has been less rapid, but there has been a steady growth in the number of group practice organizations. It has been estimated3 that while in 1932 there were 239 medical groups with 1,466 physicians in the United States, by 1951 there were 600 such groups with over 5,000 physicians. Growth in voluntary health insurance both in terms of people covered and in benefits offered will undoubtedly continue. New organizational patterns will continue to emerge. Unlike their predecessors, however, these new developments will be able to call upon a body of experience to guide them which was not available to most of the present voluntary prepayment plans for medical care at the time they were founded. The experience of one such plan—the Health Insurance Plan of Greater New York (HIP)—has been analyzed in the belief that it will be of value not only to HIP but also to administrators of other medical care plans, to welfare fund administrators, to public health officials, to community planners, to sociologists, and to the public in general, all of whom are interested in facts that bear on the problem of making the benefits of modern medical care available to the population. The HIP experience will be presented in three parts: Part I, the present paper, deals with enrollment experience and sets forth data on the extent to which enrollees of different classes remain in the Plan and factors influencing continuance of their enrollment; Part II will present information on the number and kinds of physicians' services utilized by enrollees and how this utilization varies with length of coverage in the Plan; Part III will contain data on morbidity among enrollees of the Plan and will seek to develop measures of prevalence and incidence of various diagnoses as reflected in the operating records of HIP. These data represent the results of a “longitudinal” analysis of the first four years of operation in HIP—that is, they show the experience of the same group of individuals followed through four years of calendar time. They supplement material contained in a published report on the findings of a field survey inquiring into the health and medical care of HIP and New York City households4 which was essentially a cross-sectional study depicting the situation at a particular point in time. Since the meaning of the data to be presented cannot be properly assessed without an understanding of the organizational framework in which they have been collected, a brief description of the HIP program is in order. HIP grew out of the late Mayor Fiorello LaGuardia's concern for the health of the employees of the City of New York and of their families. This concern extended not only to providing a means for meeting the costs of medical care but also to the kind of medical care to be provided. Mr. LaGuardia felt that such medical care should be comprehensive in coverage and should emphasize maintenance of health and prevention of illness. With this philosophy as a guiding principle, the Health Insurance Plan of Greater New York began operation as a nonprofit corporation in March, 1947. It is generally classed as a “service” program which means that in return for the premium the enrollee is entitled to receive a variety of medical services. He receives no bills for these services, the premium paying the entire cost.5 In this sense, once the premium is paid there are no additional costs for medical care in HIP. This contrasts with the “indemnity” type of program in which the enrollee is billed by the doctor and is indemnified in cash according to a fixed schedule of fees for those medical services covered in the contract. The indemnity received may or may not equal the amount of the physician's bill.6 The HIP contract makes available to the enrollee the following services in the home, physician's office, or hospital: general medical care; surgical, maternity, pediatric, and other specialist care; periodic health examinations; immunizations; laboratory and x-ray services; physical therapy; administration of blood and plasma; psychiatric consultations; visiting nurse services; and ambulance services.7 There are no waiting periods for service and no limitations on the number of services or duration of medical care. Group enrollment is required, and, with few exceptions the employer pays at least half the premium directly or, in the case of union welfare funds, all of the premium is usually paid as a fringe benefit. A person losing coverage under the group contract is legally entitled to continue as an individual subscriber with a “policy in conversion.” Medical services are provided on a group-practice basis through medical groups distributed throughout New York City. The organization of these medical groups during the period of the present study is described in detail by Baehr.8 The medical groups are paid on a per capita basis for each individual on their rolls, irrespective of the volume of service. The payment of the individual doctor in the medical groups is determined by each medical group. The present paper sets forth the extent to which HIP's enrollees have remained in the Plan during the four-year period 1948–1951 and examines some of the factors influencing retention of coverage in the Plan. From the economic standpoint it is desirable to have a picture of retention of membership in a medical care plan because (a) the plan may be embarrassed financially if its enrollee population tends to stay in the plan only during periods of need for extensive medical services, (b) estimates of personnel and facilities required under certain conditions and considerations of such social questions as that of health insurance of the aged require forecasts of future enrollment in the plan, and such forecasts require knowledge of attrition rates of various components of the population, and (c) a high rate of turnover results in higher operational costs. Knowledge of retention of coverage is also of importance from the standpoint of health maintenance. If, as in HIP, one of the objectives of a plan is to maintain and protect the health of its enrollees, continuity in membership is essential to the achievement of this aim. Moreover, research on the impact of the plan on the health of its enrollees is hampered if survivorship in the plan is low, since follow-up studies become exceedingly difficult if not impossible. In short, facts on retention of coverage by enrollees in a medical care plan are of importance from an administrative, medical, and sociological standpoint. The basic facts needed to study membership retention in a medical care plan are the date of entry to the plan and the date of leaving. These basic data are available for each enrollee in the Registrar Division of HIP. All changes in enrollment status are noted on the enrollee's file card. Also available for each enrollee are date of birth; sex; whether the enrollee is the subscriber in whose name the contract is made out, a spouse, or a dependent; whether the type of contract covers the family or only the subscriber; and the particular class of account to which the subscriber belongs. Thus, it is possible to examine the extent to which enrollees remain in the Plan in relation to each of these factors. Analysis of retention of coverage in this report extends from January 1, 1948, to December 31, 1951. This period covers the first four full calendar years of HIP's existence. During these four years the enrollment advanced from 76,106 to 287,659. For the purpose of this study a ten per cent sample of all enrollees resident in New York City was chosen. Selection was made on the basis of all subscribers active at any time during the study period whose certificate number ended in a specified digit9 and who entered the Plan before January 1, 1951. This closing date for entrants was chosen to provide at least one year of coverage for persons included in the study. Altogether the sample available for analysis consisted of 27,130 enrollees.10 The distribution of these 27,130 persons by the basic variables of the study are shown in detail in Table 1 and Appendix Tables A1 and A2. However, of the 27,130 total persons, 11,928 were subscribers and 15,202 were dependents. In what follows the term “enrollees” refers to the entire group of 27,130 individuals but the term “subscribers” refers only to the 11,928 individuals in whose name the certificate or policy is issued. The procedure for calculating the proportion of persons remaining in the Plan within a given period of time after enrollment is detailed in Appendix Table A3. This proportion is called the retention rate and may be thought of as the number of persons out of a thousand who remain in the Plan one, two, three, or four years following entry. The experience in retaining coverage for the 27,130 enrollees as a whole is shown in Figure 1. The detailed data are presented in Appendix Table A4. The chart shows that, during the study period, the chances were that about 7 out of every 10 persons entering would still be in the Plan four years later. The curve is slightly convex to the horizontal axis suggesting that the greatest loss occurred during the first year in the Plan. This suggestion is supported by examination of the proportions remaining at the end of each successive year among those present at the beginning of the year. These proportions are as follows: Retention of Coverage in HIP of All Enrollees In other words, if a person remains in HIP throughout the first year, the likelihood that he will leave in any one of the three subsequent years becomes progressively less, though the trend is not a very marked one. The retention of coverage of the successive cohorts was not uniform (Figure 2, Appendix Table A4). Enrollees in the 1947 cohort showed the highest retention rates while those in the 1950 cohort had the lowest rates. About as many persons in the 1950 cohort were lost to the Plan by the end of two years as by the end of three years for the 1948 and 1949 cohorts. Retention of Coverage in HIP of All Enrollees, by Year of Entry In considering this variation in the experience of the several cohorts, it is important to bear in mind their composition in regard to class of account. This distribution is shown in Table 1. It can be seen that nearly 90 per cent of the 1947 cohort were employees of the City of New York and the largest single department was the Board of Education. In the 1950 cohort the category “Other Accounts” represented 19 per cent of the new enrollees although City departments still accounted for three fourths of the new enrollees.11 The difference in the retention rates of the 1947 cohort and the 1950 cohort may therefore be a difference in the class of accounts or it may be due to a difference in type of coverage. Enrollees in City accounts are covered under family contracts but a substantial proportion of those in non-City accounts fall in a type known as “employee-only” accounts. The influence of the type of coverage and of the class of account on retention of membership in the Plan will, therefore, be examined next. In general there are two broad classes of coverage provided by HIP contracts—those in which the subscriber covers himself and his family (family coverage) and those in which the subscriber is eligible to cover only himself (employee-only coverage). The HIP retention experience of persons covered under family contracts is very much better than that of those covered under employee-only contracts. (Figure 3 and Appendix Table A5.) This difference is observed in all of the cohorts. The enrollees under family contract are for the most part in City departments, while those under employee-only contract are covered predominantly in Union and Trustee accounts. All City departments have higher retention rates than either the Union and Trustee accounts or the category “Other Accounts,” which is made up of small private employer contracts and housing projects. (Figure 4, Appendix Table A6.) Retention of Coverage in HIP of All Enrollees, by Type of Contract and Year of Entry Retention of Coverage in HIP of All Enrollees, by Class of Account, Type of Contract, and Family Status In considering these differences, these facts must be borne in mind: (1) subscribers under City accounts pay only half the premium, the other half being paid for by the City; (2) subscribers under “Other” accounts in many instances pay the entire premium; (3) subscribers under Union and Trustee accounts for the most part pay none of the premium as individuals, the premium being paid for out of union welfare funds. Subscribers under City and commercial (“Other”) accounts may decide as individuals to drop their coverage, but this individual decision is not open to subscribers of Union and Trustee Accounts since they remain covered as long as the union continues to pay the premium for its members. Because loss of coverage by a subscriber in Union and Trustee accounts is a result either of death or of failure to remain eligible for coverage by the union welfare fund, the low retention rates in this category must be ascribed primarily to instability of employment,12 and to a low rate of conversion to individual policies (see discussion on Experience with Conversions, on page 662). Retention rates for subscribers in “Other” accounts are poor under both family and employee-only type of contract. While instability of employment undoubtedly plays a role here also, an additional factor of importance in these accounts is the fact that many of the enrollees bear the entire cost of the premium as individuals. Study of the curves (Figure 4) for City accounts (all of which are family type of contract) shows that the existence of dependents is another important factor which influences retention of coverage in the Plan. The rate of retention is always higher for subscribers with dependents than for those without dependents. The same relationship exists between subscribers under family contract with and without dependents in the group of “Other” accounts. The factors just discussed probably play a major part in determining retention of subscribers in the Plan. In some instances anticipation of acquiring dependents, or psychological factors related to being associated with enrollees in the same accounts who have dependents, may also exercise an influence on retention of coverage. In addition, the higher retention rates for enrollees of the Board of Education than for employees of other City departments suggest the possibility that education may also be a factor in determining continuance of coverage in the Plan. One would have to examine education explicitly to determine this. There is a slightly but consistently higher retention rate of females as compared with males (Figure 5, Appendix Table A7) in three out of four of the cohorts. Moreover, when the data are broken further by age as well as by sex (Figure 6, Appendix Table A8) the higher retention rate of females is more pronounced and is seen in every age group except 65 and over. In that age group the number of females on which the rates are based is quite small. Retention of Coverage in HIP of All Enrollees, by Sex and Year of Entry Retention of Coverage in HIP of All Enrollees, by Sex and Age at Entry There is an apparently greater variability in the retention rates with age among the men than among the women. This greater variability, however, is largely due to the much rate of males in the age group than of females in this age group. The of the rates in the is about the same in the two the rates being at a higher The retention rates of males in the fact that all of the enrollees in employee-only accounts are males and, as has been this type of account is by poor retention of coverage. attention is focused on age as a factor in retention of coverage, it is seen that under of both have a high retention rate as both men and in the age group These two age groups accounted for nearly half per of all enrollees with coverage who had entered HIP to January 1, 1951. Family are in the age and this undoubtedly is reflected in the high retention rate of this group. Retention of coverage in HIP in the three age groups and 65 and over. this is to some extent due to the of The retention at the end of four for these three age groups in males and these rates are to from the and although slightly changes are in the retention rates for females of these through the of of has no on the retention rates between the of and The HIP survivorship with are shown in Figure Table of the retention rates from age there is still a with age but not as as This trend may represent a combination of several factors. In the age group 65 and persons from covered employment and their may not be to them to pay the full premium even though they have the of a the retention rate is factors may though with less in the age group in the and is Retention of Coverage in HIP of Enrollees by Age and with It may also be that the trend HIP retention rates with age in part a greater likelihood of an person medical to to HIP which he to it may be that in some instances the trend the fact that the a person is the more he is to have a this disease to the point of not being able to continue in his his is he cannot continue his premium and he his coverage in the Plan. It is not known to what extent this is but if the situation with any it is a which should be of concern to the community since such persons are very to become upon the community in one or From the standpoint of the community it is desirable to have more data on the of this The low retention rate of enrollees a Insurance who age are no as dependents. an enrollee is to have and is no covered by the the enrollee 65 or the year may to individual coverage by paying the entire premium but this is an age when rates are low and the to remain in the Plan is not Also some of these persons may be which provide at least some medical services. The retention rate of persons in with those may be accounted for by a higher proportion of single persons and and rates in the group. In the discussion of retention of coverage and family status it was noted that subscribers with dependents have a better retention rate than those without dependents. In Figure Table the retention of subscribers in family accounts is examined further in relation to the number of dependents Retention of Coverage by HIP Subscribers under Family Contract by Age and by of at Entry It is that among the group of the greater the number of dependents the more it is that the subscriber will remain in HIP throughout the four-year among those with three or more dependents, per cent were still in the Plan four years later. The for those with two, one, and dependents covered were and per cent These in per cent retaining membership probably by of family in the need for and services. In to the experience of the of a subscriber years of age or remaining in the Plan over the four year period varies very in relation to the number of covered dependents. The highest retention rates are still observed among those with the greatest number of dependents, but the over that of subscribers with dependents is not nearly as marked as among subscribers years It may be that the of retention of coverage in the Plan of the subscriber is primarily by the of persons in his age group than by the medical care of other of his It has been noted that subscribers without covered dependents in family accounts are more to remain in the Plan than subscribers in employee-only contracts. A of Figure with Figure shows this to be of whether the group of subscribers is under or over years of In other words, this difference between the class of subscribers without covered dependents apparently is not due to any influence of age on the survivorship It is more related to factors such as or to the fact that the subscriber in the employee-only account may be to the or of the of his family who receive their medical care the Plan. the subscriber without dependents under the family contract is not to these The patterns of retention of coverage in relation to the of the covered which have just been discussed are essentially the same in each of the four cohorts classed by year of entry Table that retention is greatest among subscribers with the greatest number of dependents. The variation in retention of coverage among the different cohorts is largely due to the different class of accounts entering in different years which has been the present time enrollment in most medical care plans is largely on a group However, persons an enrollment group because of or are in some instances to their enrollment to an individual policy with more coverage. In HIP, all enrollees have the of conversion to an individual policy with the same coverage as they the premium rate in HIP for such individual policies is only slightly higher than for the group conversion for most enrollees usually means that they have to the entire premium the portion paid by an employer or by a union health and welfare the total persons in the Plan at the end of per cent had to individual contracts. This low may suggest that are of However, the conversion a different when it is that through the conversion the aged and other of the population with health have an to remain the conversion may be of importance to the the medical care plan, and the community at It is for this that the following data are presented on the of and of retaining coverage after The of among those under family contracts as compared with those under employee-only contracts is shown in Table the group are more than the to exercise the conversion if they are for It to that this an need for medical care for dependents on the part of those under the family contract. However, other factors may also play some part in about the observed in conversion enrollment conditions in the two of coverage are quite contracts were in part those of which covered their and paid the entire In the case of people with family coverage, however, individual had to be made to HIP and the subscriber himself usually paid at least half the This in the decision to may account for a part of the difference in conversion rates. part may be due to a economic of enrollees in employee-only accounts. In any as is detailed the of dependents in the family an important influence on the of to individual contracts if the need subscribers are much more to than men subscribers when they are no eligible for group The difference is greatest in the age group although it is still in the age group. Part of the difference the fact that there are more employees in the employee-only type of contract but it is also that the high conversion rate of females in the age group a need for and services in the it must be that a proportion of subscribers are employees of the Board of Education. The of the Board of Education therefore, influence the per cent on or leave must or coverage for the duration of the They may return to group the in the age group 65 and over are small the conversion rate in this group would to be probably a need for medical care. In family the subscriber with two dependents is more to maintain coverage in the Plan through conversion than the subscriber in or larger (Figure Appendix Table This to be irrespective of the length of coverage. One would to examine the data further by age as well as of family but the material is to this. to by Subscribers under Family Contract by of and of Coverage of Subscribers in which the subscriber has two dependents the of a subscriber with the length of his coverage. A trend to be present among subscribers in larger though there is difference between the of in the year of entry and the year. The picture for subscribers with no dependents and those with one is not quite as but the conversion rates in the calendar year after entry to be higher than in It would that in each family there is a for the conversion rate to the the coverage and that this is there are two and three or more dependents than in families. The subscriber who covers only himself is more to if he is covered under a family contract than under an employee-only contract. For the group the is per cent (Figure and for the not more than per cent This difference a combination of factors among which are the of a individual decision to HIP in the first and the under the family contract of eligible dependents if and when they are the several classes of subscribers in the Board of Education had a higher rate of conversion when for than any other of the Board of Education are largely It is of that the study of Health and Medical in New York City