Among all women who made a medical visit for family planning from 1987 to 1988, a majority (64%) went to a private physician, while 36% attended a family planning clinic.1 Although there has been extensive research on publicly funded family planning clinics and the services that they provide,2 few studies of family planning care have focused on private physicians.
A survey of private office-based physicians in 1983 found that 100% of obstetrician-gynecologists and 84% of general and family practitioners provided reversible contraceptives to their patients. In addition, 94% of obstetrician-gynecologists, 18% of general and family practitioners and 89% of urologists performed sterilization procedures.3
In the early 1980s, obstetrician-gynecologists accounted for one-third of the private physicians offering reversible contraceptive services. However, these providers saw, on average, four times as many contraceptive patients as did general and family practitioners; thus, they served more than two-thirds of all patients visiting private physicians for reversible contraception.4
Surveys of low-income women show that although most would prefer to go to private physicians for their contraceptive care, financial constraints typically compel them to obtain such care at publicly supported family planning clinics.5 In fact, during the 1980s, the percentage of low-income women (those below 150% of poverty) who made their most recent family planning visit to a private physician declined from 52% in 1982 to 40% in 1988, while the percentage of women with an income at or above 150% of poverty who visited a private physician for family planning remained fairly stable (at approximately 75%) over the same period.6
Medicaid, the federal-state entitlement program that covers medical care for welfare recipients and low-income pregnant women, theoretically affords patients access to both public and private family planning providers. However, Medicaid enrollees may not be able to find a private physician who will serve them. Among obstetrician-gynecologists and general and family practitioners in 1983 who offered contraceptive services, only 56% accepted Medicaid.7 Similarly, in 1987, only 54% of obstetrician-gynecologists served patients on Medicaid,8 and in 1990-1993, only two-thirds of all physicians in independent or small group practices participated in Medicaid.9
Moreover, private physicians who do accept Medicaid often see very few of these patients, as low reimbursement levels, the burden of paperwork for filing claims and the delay for reimbursements make serving such patients unattractive.10 In 1988-1990, only 8% of all visits to obstetrician-gynecologists were paid for by Medicaid.11 In 1987-1988, 71% of women who had made a medical family planning visit paid for by Medicaid had gone to a family planning clinic and only 29% had received care from a private provider.12
However, the context of health care delivery and the role of private physicians is changing dramatically, and these changes may increase interest among private physicians in providing services to low-income patients. Moreover, insured individuals are increasingly covered by managed care plans rather than traditional indemnity insurance. By 1996, more than four in 10 privately insured Americans were in managed care plans, as were one in four Medicaid recipients.13
Managed care plans have relieved physicians of the paperwork burden, and competition between plans may have narrowed the gap in reimbursement rates between Medicaid and private insurance. The growth of managed care plans has also increased competition for patients in general. These changes may be especially relevant for low-income women seeking contraceptive services, since eligibility criteria for Medicaid have been expanded to include coverage during pregnancy and up to 60 days postpartum (a time when contraceptive use is usually initiated) for women not previously eligible.14
This article presents information on the characteristics of patients and physicians involved in office-based contraceptive services at an early point in the health care transition. We use the patient visit information available from the National Ambulatory Medical Care Survey (NAMCS), a nationally representative survey of physicians, to identify characteristics of patients making private office visits for contraceptive services and to explore the extent to which the services they receive are affected by their personal characteristics, in particular their expected source of payment. This cross-sectional information provides a foundation for investigation into how the provision of contraceptive services by private physicians may change in the coming years, and identifies some important gaps in the available data.
Data and Methodology
The data used in this study come from the 1990, 1991 and 1992 NAMCS, which is conducted annually by the National Center for Health Statistics (NCHS). The survey contains information from a nationally representative sample of physicians about patient visits to private office practices and the services provided during those visits. Most of the data reflect judgments made or services provided by the physician: the patient's diagnoses (up to three are permitted); whether the physician had seen the patient before; the services provided (including diagnostic and screening services, counseling and advice, specific medications ordered or provided, and any therapy not involving medication); and the amount of time the physician actually spent with the patient.
Patient information reported by the physician includes the patient's date of birth, sex, race, ethnicity, expected source or sources of payment, and the patient's primary complaint, as well as other complaints and symptoms or other reasons for the visit (up to three). The physician is instructed to record the reasons for the visit "in [the] patient's own words" (to minimize any potential bias in the physician's interpretation of the patient's stated reason), and these verbatim statements are coded into categories by the NCHS.
Physicians are asked to code the expected source or sources of payment for the visit using the following categories: private or commercial insurance; health maintenance organization (HMO) or other prepaid plan; Medicare; Medicaid; other government; no charge; patient paid; and other. As defined by the NAMCS, the term "HMO or other prepaid" refers to a wide range of managed care plans, including independent practice associations and preferred provider organizations. We have therefore chosen to use the broader term "managed care" to represent this category. In our analyses, we have combined "no charge" and "other" into one category ("other").
The actual source or sources of payment made directly to the physician or indirectly through reimbursement to the patient may not always be the expected payment at the time of the visit. However, the expected source of payment is most relevant to our analysis, since it is what the patient and physician considered as the source of payment at the time of the visit.
Physicians responding to the NAMCS may code more than one expected source of payment. However, for only 5% of all contraceptive visits were more than one of the five expected payment categories coded (for example, among those in which one of the expected sources of payment was Medicaid or another government source, 3% were also coded as private insurance and 1% as managed care). In such cases, we assumed only one to be the main payment source and assigned the visit to that one source according to the following order of precedence: private insurance; managed care; Medicaid or other government assistance; other payment; and self-pay.
Physicians principally engaged in office practice are selected from master files maintained by the American Medical Association and the American Osteopathic Association. The sample includes physicians working in HMOs and in private, non-hospital-based clinics, but those employed by the federal government are excluded. All participating physicians provide information on each patient visit to their office during a specified time period, using a standard data collection form. The data include only office-based visits and exclude telephone contacts with patients and other types of patient encounters, such as those in hospitals or institutional settings.
We identified as contraceptive visits those visits reported in the NAMCS for patients 10 years of age and older that met one of the following criteria: Any of the three patient-identified complaints, symptoms or other reasons for the visit were recorded as "family planning" (not including "an abortion to be performed at this visit" or "artificial insemination"); any of the three physician diagnoses were classified under the International Classification of Diseases, ninth edition, as contraceptive management, family planning or contraception; any of up to five medications or devices ordered or provided were contraceptives; "family planning" as a therapeutic service was checked (an option only in 1991 and 1992); or either of the two possible ambulatory surgical procedures reported (also an option only in 1991 and 1992) involved contraception (such as insertion or removal of contraceptive devices or male or female sterilization). Each visit was counted only once, even if it met more than one of the criteria.
We recorded the patients' principal complaint (the first reason for visit) into six categories: specific contraception (includes insertion, removal, exams and advice regarding specific contraceptive methods); general family planning (includes counseling, examinations and general advice regarding unwanted pregnancy, sterilization, infertility and genetics and birth control (not pertaining to a specific method); gynecological exam or Pap test; prenatal or postpartum exam; general medical exam; and other. All main reasons for the visit were mutually exclusive. We measured the actual provision of contraceptives by computing if any of the five medications (or devices) that were ordered or provided during the visit was a contraceptive.
To obtain more reliable estimates, we combined the 1990, 1991 and 1992 NAMCS data sets. While survey design and procedures were identical in each year, the samples were independently drawn by NCHS. The response rates of physicians participating in the 1990, 1991 and 1992 surveys were 74%, 72% and 73%, respectively. A total of 1,772 patient visits to private physicians participating in the NAMCS during 1990-1992 were related to contraception. Weights provided by NCHS that adjust for the probability that a physician was selected and for nonparticipation of sampled physicians were used to calculate national estimates from the visit sample.15
Weighted percentage distributions reflect the average distributions over the three-year period. Subgroup differences noted in the text are all significant at p<.05, based on t-tests. (Standard errors were estimated following procedures outlined by Shappert.16) Logistic regressions were performed using SPSS. In order to perform tests of statistical significance, we reweighted the data in the regressions so the total number of cases in the analysis equaled the unweighted total. The new weight equaled the weight NCHS assigned to each visit divided by the ratio of the total number of relevant weighted visits to the total unweighted number. Thus, while the tests of statistical significance in the multivariate analysis are appropriately based on the unweighted number of cases, the odds ratio estimates reflect the weighted distributions of respondents.
Visits for Contraceptive Services
According to the NAMCS, an annual average of 13.5 million contraceptive visits were made to private physicians in 1990-1992. Some 12.6 million of these visits (94%) were made by women, accounting for 7% of all visits made to private physicians by women aged 15-44.
Only 39% of women making private office visits for family planning were aged 30 or older, while this age group accounted for 79% of office visits by men (Table 1). Women making contraceptive visits were roughly twice as likely as men to be black or Hispanic (18% compared with 9%), but the difference is not large enough to be statistically significant.
The majority of office visits made by women obtaining contraceptive care from a private physician were for a reversible method; a contraceptive method was ordered or prescribed at 66% of these visits. In 80% of these cases, the method was reported to be an oral contraceptive; in another 18%, it was identified only as a "contraceptive agent"; this could include devices such as a condom or diaphragm (data not shown). In contrast, 38% of family planning visits made by men were for a vasectomy performed at that visit; it is likely that most other recorded visits were related to consideration of, or follow-up for, the surgical procedure.
According to physicians' reports, only a minority of women and men whose visits were related to contraception actually stated this as their main reason for seeing the doctor. Physicians indicated that 81% of women and 64% of men gave a reason other than contraceptive care for making the visit. The most commonly stated reason for visits made by women were for a general medical exam (15%), a gynecologic exam or Pap test (13%) and a prenatal or postpartum visit (12%).
Most contraceptive visits by women (69%) were to obstetrician-gynecologists, while urologists were the most common type of physician seen by men (43%). Approximately 22% of all visits made by both men and women were to a family practitioner. Forty-two percent of visits by men were first visits to that physician, compared with only 17% of visits by women.
Expected Sources of Payment
Table 1 also indicates that private insurance was the most common expected source of payment for all visits (38%). Patients covered by managed care plans accounted for 22% of all contraceptive visits, and another 22% of visits were made by patients who paid for the visit themselves. Medicaid and other government programs were the expected payment sources for about 12% of all private office visits.
Visits by women expecting to pay with private insurance or through managed care plans were made largely by those aged 25 and older; 43% were by women aged 30 and older, and another 26-30% were by those aged 25-29 (Table 2). In contrast, more than half (55%) of women paying for their visits through Medicaid or through other government assistance were younger than 25; only 31% of such visits involved women 30 or older. Fifty-three percent of the visits by self-paying women were made by 20-29-year-olds.
Fifty-six percent of Medicaid visits by women were made by whites, whereas black and Hispanic women accounted for 31% and 13% of these visits, respectively. Women covered for their family planning visit through Medicaid or other government assistance were 3-5 times more likely to be black than were women covered by private insurance or managed care.
Only 10% of family planning visits to obstetrician-gynecologists were to be paid by Medicaid or another government source, compared with 17% of visits to general and family practitioners and 19% of visits to other physicians (not shown). Women's visits paid for by Medicaid or other government sources were less likely than those covered by private insurance or managed care to be made to an obstetrician-gynecologist (54% versus 73%). There was little difference by expected payment source as to whether the woman had visited that physician before or as to the amount of time she spent with the physician.
Visits paid for by Medicaid or another government source were less likely to involve a Pap test (32%) than were visits in which the costs were expected to be covered by private insurance (58%), by managed care (50%) or by the women themselves (57%). Publicly funded visits were also less likely than self-paid visits to result in the provision of a contraceptive method or a prescription (56% vs. 79%). Thirty-two percent of family planning visits covered with public funds were for general family planning services or for care related to a specific contraceptive method, compared with only 14-15% of visits expected to be covered by private insurance or by the patient herself.
For the multivariate analysis, we examined factors associated with whether a woman's main reason for a visit to a private physician was for general family planning advice or an examination or for care regarding a specific contraceptive method. In addition, three characteristics of each visit were examined: whether a woman was given a contraceptive method or prescription; whether a Pap test was performed or ordered (a rough indicator of the extent to which the woman received broader general or reproductive health care at the visit), and whether the woman spent less than 15 minutes with the physician (a rough measure of the complexity or thoroughness of the visit).
Reason for Visit
As shown in Table 3, a woman's age and race had no significant effect on whether the main reason for her office visit was for general family planning advice or services or was for care regarding a specific contraceptive method. However, Hispanic women were less likely than black women to report coming about a specific contraceptive method (odds ratio of 0.14).
Women seeing a physician who was not an obstetrician-gynecologist were less likely than those who visited an obstetrician-gynecologist to have stated general family planning needs as their main reason for seeking care (odds ratios of 0.37 for general and family practitioners and 0.24 for other physicians). However, the odds that visits to general and family practitioners were about a specific contraceptive method were twice those for visits to obstetrician-gynecologists. Women making their first visit to a responding physician were more likely than ongoing patients to have come for general family planning advice or examination (odds ratio of 3.47), but they were no more or less likely than ongoing patients to have made a visit regarding a specific contraceptive method.
Expected source of payment had a significant impact on the stated reason for a woman's visit to a private physician. Those visits covered by managed care and by Medicaid or other government sources were more likely than visits paid for through private insurance to have general family planning advice or examination recorded as the woman's main reason for the visit (odds ratios of 2.14 and 2.88, respectively). Only those women who were receiving free services or whose source of payment was unknown or unreported were more likely than those covered by private insurance to have reported that their main reason for making the visit was for a specific contraceptive method (2.59).
Table 3 also indicates the impact of patient characteristics and visit characteristics on three selected visit variables. If the main recorded reason for the patient's visit was for a gynecologic examination or a Pap test, the odds that a contraceptive method was actually provided or prescribed were greater than if the stated reason for the visit was general family planning advice or an examination (odds ratio 6.31). Office visits initiated in order to get a specific method or for advice or examination concerning a specific contraceptive, for a general medical examination or for something other than reproductive or general health care were more than three times as likely to result in the provision of a contraceptive method as were visits made for general family planning advice or examination.
Visits made by women aged 30 or older and by women for whom this was their first visit to the respondent physician were less likely than others to involve provision of a method. General and family practice physicians were more likely than obstetrician-gynecologists to have provided a contraceptive method. The odds that a contraceptive method was provided at the visit were lower if Medicaid or another public source was the expected source of payment than if private insurance was expected to cover the cost (odds ratio of 0.65). These odds were higher, though, for visits in which the woman was expected to pay the costs herself (1.61).
The strongest predictor of a woman's receiving a Pap test during a contraceptive visit was whether or not she received a contraceptive method at that visit: The odds that a woman received such a test during a visit in which a contraceptive was provided were 3.43 times higher than during one in which she did not obtain a method. (This relationship is not surprising, since the overwhelming majority of women receiving a method were obtaining oral contraceptives, a method for which a pelvic examination and Pap test are commonly performed when women begin taking pills and at annual intervals during pill use.)
Visits made by women aged 20-29 were less likely than those made by women younger than 20 to involve a Pap test. Visits to a physician other than an obstetrician-gynecologist were less likely to involve a Pap test, and a first visit was more likely to include a Pap test than was a visit made by a continuing patient (odds ratio of 1.36). The odds that a visit included a Pap test were 11 times higher for visits initiated for a gynecologic examination or Pap test, and seven times greater for those sought for a general medical exam, than for visits sought for family planning advice or examination.
The likelihood that a Pap test was provided during an office visit was lower when the woman's expected source of payment was anything other than private insurance. For example, the odds that visits included a Pap test were only 39% as high for visits covered by Medicaid as for those covered by private insurance, and only 47% as high when no payment was expected or when the expected method of payment was unknown.
Few of the available variables were significant predictors of visit length. The odds that a visit lasted less than 15 minutes were lower for first visits and for visits sought for gynecologic exams or for Pap tests (odds ratios of 0.35 and 0.56) while the odds that a visit was less than 15 minutes long were greater when a contraceptive method was provided (1.29). These relationships appear to reflect the longer time that physicians take to get to know a new patient and the added length of a full gynecologic exam compared with a visit for family planning advice. Visits covered by private insurance did not differ significantly from those of any other expected payment source in the likelihood of lasting less than 15 minutes.
This study documents an area in which current knowledge is limited: the family planning care rendered by the most commonly used providers—office-based private physicians. There is substantial disagreement between major national data sources over how many visits, and ultimately how many women, actually receive contraceptive care through private physicians. The NAMCS indicates that some 12.6 million contraceptive visits are made by women to private physicians annually. However, another NCHS data set more commonly used to measure family planning visits, the National Survey of Family Growth (NSFG), produces considerably higher estimates of the total number of family planning visits women made to private medical offices. As measured by the NSFG, there were a total of 24.5 million family planning visits to private medical providers in 1988.17
Markedly different survey designs and definitions appear to be the cause of the large discrepancy between the NAMCS and the NSFG estimates. In the NAMCS, a physician records information about office visits at the time they occur, while in the NSFG, women respond from memory to interviewers about visits made in the past 12 months. Only visits in which the reporting physicians actually saw a patient are included in the NAMCS, while women responding to the NSFG are categorized as having visited a private provider if they obtained any care from the provider, whether or not they actually received services directly from a physician. Finally, the criteria that determine whether family planning was a purpose of the visit may be evaluated differently by a woman responding to a survey about contraceptive and fertility behavior than by a physician coding multiple visit characteristics. In addition, in the NSFG, the definition of a family planning visit is broader than the definition constructed for use with the NAMCS.
Data reported by women in the context of the NSFG may result in an overestimate of the use of private physicians during a given time period, even though the NSFG may be accurate in identifying those women who consider doctors to be their source of contraceptive care. In contrast, the NAMCS may underestimate the number of private physician visits related to contraception. As we have shown, most of the visits categorized as family planning visits were identified not because the woman had given contraceptive care as her main reason for making the visit but because a contraceptive method was provided or prescribed or some other specific family planning service was provided. It is possible that the NAMCS missed visits, initiated for other reasons, at which contraception may have been addressed but was not actually provided.
Despite the fact the NSFG and the NAMCS produce widely different estimates of the number of contraceptive visits, the surveys are in close agreement in estimating the characteristics of patients who seek contraceptive care in private physician settings by such variables as age, race, and source of payment.18 The NAMCS data are therefore useful for examining in greater detail than is possible with the NSFG the characteristics of private physicians' family planning care.
Further research will be needed to better understand the strengths and limitations of the different approaches to obtaining data on the use of health care services. Avenues that appear useful to pursue in the design and development of future national surveys include use of more comparable definitions of visits, reasons for visits and providers seen, aids to help women accurately identify when specific visits occurred and the ability to collect information on the same visit from both the patient and the physician.
Our results confirm that family planning services for men are not well established.19 Men make only 6% of family planning visits to private physicians. The only contraceptive method used by men that requires a family planning visit is a vasectomy. However, men can certainly profit from contraceptive counseling and services before they reach the stage where they are seeking a vasectomy. Indeed, much of the work to develop new male contraceptives is focused on systemic approaches that are likely to require men to visit a clinician for screening and method provision.20
One important contrast between family planning visits made by women to private physicians and those to clinics is the extent to which the woman or the medical provider perceives family planning to be a central focus of the visit. For the most part, women obtaining contraceptive services at family planning clinics are going specifically to get a contraceptive method or for follow-up while they are using a method. Clinics typically provide contraceptive patients with other general medical and social services as well, such as basic medical history-taking and physical exams, risk assessments and testing for sexually transmitted diseases, and contraceptive education and counseling.21
Most private physician visits in which contraceptive care was provided, however, were made for other reasons. Contraceptive care may be seen more as an incidental service among both private physicians and their patients. If this is the case, women obtaining family planning care through private physicians may not be giving or receiving adequate attention to their choice and use of methods and could benefit from specialized contraceptive care such as that typically given in family planning clinics.
Alternatively, contraceptive care may be more integrated into general medical or gynecologic care for women using the private sector. In this case, women utilizing private physicians would have opportunities to raise questions about contraception in a more natural, ongoing way and may be served by someone with more complete knowledge of their general health than would be possible through infrequent visits to a provider who specializes in contraceptive care.
The fact that many women making contraceptive visits say that getting a Pap test or having a general or gynecologic examination is their main reason for seeing a doctor also suggests that for many physicians and their female patients, general or gynecologic care may be synonymous with contraceptive provision. Rather than being incidental to women's overall health care, the need for contraception may in fact propel women to seek a visit for a gynecologic or general health exam. Thus, even though the physical examination typically performed prior to the dispensing of oral contraceptives seldom provides useful information about the appropriateness of this method,22 it is nonetheless common practice—and common expectation—for the woman to receive physical, pelvic and breast exams and a Pap test at these visits.
In surveys such as the NAMCS, and in actual practice, a woman's insurance coverage may influence how her medical care is categorized. A recent survey by The Alan Guttmacher Institute found that both managed care organizations and private indemnity insurance are more likely to pay for general gynecologic services than they are to cover contraceptive services or supplies. However, managed care plans are more likely than private insurance to cover services related to contraception and general reproductive health.23 Medicaid is distinguished from private indemnity insurance and private managed care plans in its mandatory coverage of contraceptive services and supplies.
These differences in coverage within and between plans for services that are often provided within the same visit probably lead many health care providers to "creatively" code visits in order to maximize third-party payment. A habit of coding a frequently covered service (such as an illness treated at the visit or a gynecologic exam and Pap test) instead of a less covered service (such as contraceptive care) in order to receive payment may lead both women and their providers to think of contraception as a less central part of their care.
Our data indicate that women whose visits will be covered by Medicaid are more likely than those covered by private insurance to give general family planning and specific contraceptive care as the main reasons for their visit, although the relationship is only significant for general family planning. Women enrolled in managed care plans are more likely than those relying on private insurance to be making a visit primarily for general family planning. These differences, moreover, are not explained by other characteristics of the visit or of the patient herself.
Perhaps women covered by Medicaid or managed care may be more conscious of their need for contraceptive services. Alternatively, since women and their providers know that Medicaid, and to a large extent managed care, will cover a visit that is clearly a contraceptive visit, this may influence how both a woman and her provider think of and report on the care being provided.
However, the odds of getting a contraceptive method at a particular private office visit are 35% lower and the odds of getting a Pap test are 61% lower for women covered by Medicaid or another government source than for those with private insurance, regardless of the women's characteristics or those of the visit. This finding may suggest that women depending on public sources, primarily Medicaid, are not getting the same quality of care as are those relying on private insurance, who tend to be better off financially. Alternatively, it may simply reflect the typically greater reliance of low-income women on tubal sterilization,24 a method that would be linked to more intense advance counseling and to follow-up after surgery.
Finally, this result may indicate that care for women on Medicaid is likely to extend over several visits, since these women have more complex health care needs. More visits may also be involved in the provision of family planning services to Medicaid enrollees because the reimbursement rates are flat rates covering any and all services provided at a given visit, rather than charges added separately for the specific procedures and services performed. The physician's ability to charge for individual procedures may also explain why women whose visits are covered by private insurance are the most likely to receive a Pap test. That women who pay for their care themselves are more likely to have a contraceptive provided at a private office visit suggests that both the woman and her provider are conscious of the financial pressure to get her needs met in a single visit.
It should be noted that the available data do not allow us to distinguish which of the services rendered at a particular visit were covered by the expected source of payment, or whether the cost of contraceptive supplies was covered in addition to the visit itself. Thus, data about these sources of payment must be viewed with some caution. Indeed, the specific restrictions or provisions of third-party payers may be affecting even basic information about the number and types of visits at which contraceptive care is provided. However, our findings do suggest that the woman's expected source of payment affects the care she receives.
The growth of capitated managed care plans may make it even more difficult in the future to track components of specific visits other than through surveys such as the NAMCS. Even though new forms of health care financing and delivery are emerging throughout the United States, private physician practices will undoubtedly be an important source of contraceptive care for some time to come. Continued research is needed to assure that under all payment plans women have access to the medical care and supplies they need to select and use contraceptives.