Eastwood, Katherine L. MD1; Kacmar, Jennifer E. MD1; Steinauer, Jody MD, MAS2; Weitzen, Sherry PhD1; Boardman, Lori A. MD, ScM1
Nearly one half of all unintended pregnancies in the United States end in elective abortion.1 Although the total number of elective terminations has declined over the past decade, the proportion of early first-trimester abortions has continued to increase. By 2002, 88% of all abortions were performed before 13 weeks of gestation, and the vast majority were accomplished surgically.2 Ongoing provision of abortion services continues to be of concern for several reasons. In a 1995 survey of practicing obstetricians, 59% of physicians aged 65 years and older performed terminations, while only 28% of those aged 50 years and younger did the same.3 The rising age of abortion providers, compounded by the fact that 85% of counties in the United States are without any provider of such services, has led to increasing apprehension that the upcoming generation of obstetricians and gynecologists will either be inadequately trained or unwilling to provide this essential service to their patients.3,4
In response to the changing demographics of abortion provision, the Accreditation Council for Graduate Medical Education established guidelines in January 1996 mandating that residency education must include experience with induced abortion. Although programs with religious or moral objections are not required to provide such experience, provision of training in the complications of abortion is requisite.5 Before 1996, surveys of residency program directors revealed a steadily decreasing number of obstetrics and gynecology residency programs offering training in abortion. Between 1978 and 1995, the percentage of residency programs requiring training in first-trimester terminations decreased from 26% to 12%.6–8 In 1998, after the Accreditation Council for Graduate Medical Education policy change, a similar survey conducted of residency program directors indicated an increase in both first- and second-trimester abortion training. Routine training (defined as required training unless residents express moral objections) in first-trimester terminations was provided in 46% of programs, while another 34% offered elective training (residents choose to receive training). Despite the increases in availability and access to training, however, only 26% of programs reported all of their residents were trained.9
As current abortion provision in the United States is disproportionately distributed among older physicians, continuing patient access to safe abortion services represents a major public health concern. The current study was designed to assess the availability and type of abortion training presently available to obstetrics and gynecology residents and to estimate the proportion of residents who actually received such training. Program characteristics as well as estimation of the incidence and type of resistance to abortion training were also assessed.
PARTICIPANTS AND METHODS
To assess the availability and type of abortion training, we mailed questionnaires to the directors of all 252 accredited obstetrics and gynecology residency programs in the United States. Directors were identified using information provided by the Council on Resident Education in Obstetrics and Gynecology. The first mailing occurred in April 2004. Return envelopes were provided, along with a stamped and numbered postcard to allow for subsequent mailings. Second and third mailings, with instructions to use data from the 2004 graduating class, were sent in November 2004 and May 2005 to program directors who had not returned a numbered postcard indicating prior completion of the questionnaire. Approval from the Institutional Review Board of Women and Infants’ Hospital of Rhode Island was obtained in March 2004.
Questionnaires consisted of 27 closed-ended, multiple-choice questions. The investigator-designed written questionnaire incorporated elements from prior surveys,9,10 with care taken to clarify the distinction between medically indicated and elective abortion. Data collected from program directors included characteristics of the training programs, such as geographic location, program size, program type (university, community or military) and the presence and nature of any religious affiliation. Program directors were asked to classify the type of training available as 1) all residents are trained to perform abortions unless they have a religious or moral objection (termed routine or opt-out training); 2) residents elect to receive training (referred to as optional or opt-in training); 3) not available; or 4) other. Methods of abortion training taught (medical, suction dilation and curettage, manual vacuum aspiration, dilation and extraction, and induction) were ascertained, as were estimates of the number of residents who participated in both first and second trimester elective abortion training (answers were quantified as none, less than 50%, 50–75%, more than 75% to 99%, or 100%). Directors also were asked to provide estimates of the average number of terminations residents trained in elective abortion performed at various gestational ages (less than 13 weeks, 13–17 weeks, and more than 17 weeks).
The final questions of the survey required program directors to indicate whether they believed residents should be trained in abortion and to enumerate the presence and sources of any resistance to such training they had encountered. Other characteristics of the training programs, including the number of attending physicians providing elective abortion, as well as the presence of a family planning expert in the department concluded the questionnaire. Respondents were then invited to provide any additional written comments in the space below the final question.
Differences in types of abortion training on program and regional levels were determined using χ2 or Fisher exact tests where appropriate. The site of abortion training was also measured to identify possible tendencies in the location of abortion training. The results were compared with prior national surveys to determine whether changing trends in abortion training affected the numbers of residents trained. All analyses were performed using STATA 8.0 (StataCorp LP, College Station, TX).
Of the 252 program directors identified, 121 (48%) initially completed and returned the questionnaire. A second mailing resulted in 42 (163 or 65% of total) additional returned surveys. Following the third and final mailing, 185 (73%) program directors returned completed questionnaires. Response rates varied based on program size (46/84 or 55% of small programs [3 or fewer residents/year] compared with 35/43 or 81% of large programs [7 or more residents/year]) and on program location (50% among programs in West North Central (6/12) and East South Central (6/12) regions compared with more than 80% in programs in New England (13/16 or 81%) and Mountain (7/8 or 88%) regions). Programs in regions with response rates more than 70% (New England, Mid-Atlantic, Mountain, and Pacific) were most likely to provide abortion training (Table 1).
In general, large residency training programs, programs located in the Mid-Atlantic, New England or the West Coast of the United States, and programs without a religious affiliation (as compared with those programs identified as Catholic or Protestant) were significantly more likely to provide routine abortion training (P<.01). Both university-affiliated and community-based programs were more likely than military programs to provide routine training in elective termination (Table 1).
Differences also emerged in both the nature and extent of abortion training. In programs with routine training, 79 (85%, 95% confidence interval [CI] 78–92%) trained greater than 50% of their residents in elective first-trimester surgical abortion compared with only 15 (21%, 95% CI12–30%) programs with optional training (P<.001). Similar differences were seen for medical first-trimester abortions and second-trimester dilation and extractions and inductions. For example, although more than 50% of residents received instruction in dilation and extraction in nearly 40% (95% CI 26–46%) of programs with routine training, the same could be said of only 14% (95% CI 6–22%) of programs with optional training (Table 2). Similarly, the number of procedures performed by residents receiving training also varied based on whether the training was routine or optional. Independent of gestational age at the time of the termination, residents in programs with routine training, as compared with optional training, were significantly more likely to obtain adequate experience (P<.01). “Adequate experience” is defined here as at least 10 procedures per gestational age category; Table 3).
In terms of support, both on the individual program director as well as departmental level, differences again emerged depending on the type of training offered. At programs with routine or opt-out training compared with programs with optional or opt-in training, program directors were significantly more likely to agree with the statement “residents should be trained in elective abortion” (82/92 or 89% (95% CI 83–95%) compared with 35 of 70 or 50% (95% CI 38–62%), P<.01). Of the programs with optional training, 12 program directors (17%, 95% CI 8–26%) did not agree that residents should receive training in elective termination practices. Resistance to the teaching of abortion techniques came from a variety of sources. Nearly 50% of the 185 program directors surveyed indicated resistance from nursing (47%, 95% CI 40–54%) and residents themselves (46%, 95% CI 39–53%), followed by the community (34% 95% CI 27–41%), their physician colleagues (30%, 95% CI 23–37%), and the hospital administration (28%, 95% CI 22–34%). Significant differences in the sources of resistance were not found based on the type of training offered. What did seem to differ when considering the type of training was the percentage of attending physicians available to educate residents in abortion techniques. In programs with routine training, 25% (95% CI 16–34%) of staff physicians performed first-trimester abortions, and 8% (95% CI 2–14%) performed second-trimester terminations. These percentages decreased significantly for programs with optional training (5%, 95% CI 0–10%) of physicians performed first-trimester abortions, and 1% (95% CI 0–3%) second-trimester, P < .01 for both comparisons). Finally, family planning experts were on the faculties of 50% (88/185) (95% CI 40–55%) of the programs responding. Such experts were as likely to be on the faculty in programs with routine (54%, 95% CI 44–64%) as optional (50%, 95% CI 38–62%) training (full data not presented).
Abortion continues to be one of the most common procedures in the United States, with over 1,290,000 legal terminations reported in 2002. The annual number of legal induced abortions increased gradually from 1973, peaked in 1990, and thereafter began a general decline until 1998. Abortion rates since that year have stabilized at 20.9 per 1,000.10 Women seeking such care, the majority of whom are poor, require access to treatment by well-qualified providers, and if complications arise, need appropriate and compassionate care. Continued abortion provision necessitates ongoing training of new providers, a task that can largely be accomplished by educating residents in obstetrics and gynecology in various techniques of elective abortion and the management of complications associated with the various procedures used.
Between 1985 and 1998, the number of obstetrics and gynecology training programs in which abortion methods were not included in the curriculum rose from 28% to 44%. Based on our study, only 10% of all programs currently fail to offer training in abortion to their residents (Table 4). Even if we assumed that the 67 nonresponding program directors came from programs that did not offer training in terminations, our estimate would rise from 10% to 34%, still indicating an increase from 1998 in the training of resident physicians. Furthermore, the reversal in dominance from optional to routine training seen in 1998 has been carried forward to 2004. This shift in training is significant, because routine status clearly leads to training in a greater variety of abortion techniques for the management of both first- and second-trimester pregnancies.
In a survey of recent graduates of California residency programs, two factors were clearly associated with the likelihood of future provision of abortion services, training within the hospital setting and performing more than 25 abortions in residency.11 Not only does training in elective abortion need to be routine, it also needs to be of sufficient quantity. In programs with routine training, the majority of residents trained performed more than 10 first-trimester abortions, in contrast with the residents in programs with optional training, where almost one half performed less than 10 first-trimester abortions. It can be inferred that residents with more experience will feel most comfortable with the provision of abortion after residency. Sufficient training, then, seems best accomplished with routine training integrated into the residency program.
The type of training offered clearly varied in our study by multiple program characteristics, including program size and location. One explanation for the regional trends in abortion training may in part stem from recent state-initiated legislative actions. For example, both New York and California enacted initiatives that mandate public medical schools and hospitals to provide abortion training in residency programs.12 Both states also contain a significant proportion of the residency programs (many of which are large programs) in the Middle Atlantic and Pacific regions.
As with any national survey, the results generated are affected by response rates. Although the overall response rate was comparable to that demonstrated in prior national surveys (Table 4), participation rates, as with prior surveys, did clearly vary by region and program size. Programs responding to later mailings were more likely to be smaller and less likely to provide comprehensive abortion training. Other limitations we encountered included the finding that not all respondents answered every question on the survey, which created variable total numbers for each category and may represent biased results.
Although access to elective abortion services continues to be subject to increasing restrictions, the results of our survey indicate that availability of training in abortion techniques has increased over the past decade. Programs with routine or opt-out training are clearly more likely to train residents and provide greater exposure to a variety of termination practices for both first- and second-trimester pregnancies. Although opposition to abortion training was rare among program directors in programs with routine training, resistance to training in these programs did arise, most commonly from nursing. In establishing clinics where abortion services are provided, then, one possible way to reduce such resistance would be to assemble a staff of individuals openly committed to supporting women through the process of termination.
As the majority of the current providers reach retirement, younger graduates will need to be involved in both abortion provision and training of future physicians. Although the individual’s decision to provide elective abortion services after residency will ultimately drive the nature and range of such services for American women, adequate training during residency is fundamental to protecting the patients’ rights to determine the outcome of an unwanted pregnancy. Ongoing training in abortion techniques during residency helps guarantee access to safe and legal abortion.
1. Henshaw SK. Abortion incidence and services in the United States, 1995–1996. Fam Plann Perspect 1998;30:263–70, 287.
2. Strauss LT, Herndon J, Chang J, Parker WY, Bowens SV, Berg CJ. Abortion surveillance—United States, 2002. MMWR Surveill Summ 2005;54:1–31.
3. Kaiser Family Foundation (KFF). National survey of obstetricians/gynecologists on contraception and unplanned pregnancy: attitudes and practices with regard to aborations. Menlo Park (CA): KFF; 1995.
4. Grimes DA. Clinicians who provide abortions: the thinning ranks. Obstet Gynecol 1992;80:719–23.
6. Lindheim BL, Cotterill MA. Training in induced abortion by obstetrics and gynecology residency programs. Fam Plann Perspect 1978;10:24–8.
7. Darney PD, Landy U, MacPherson S, Sweet RL. Abortion training in U.S. obstetrics and gynecology residency programs. Fam Plann Perspect 1987;19:158–62.
8. MacKay HT, MacKay AP. Abortion training in obstetrics and gynecology residency programs in the United States, 1991-1992. Fam Plann Perspect 1995;27:112–5.
9. Almeling R, Tews L, Dudley S. Abortion training in U.S. obstetrics and gynecology residency programs, 1998. Fam Plann Perspect 2000;32:268–71, 320.
11. Steinauer JE, Landy U, Jackson RA, Darney PD. The effect of training on the provision of elective abortion: a survey of five residency programs. Am J Obstet Gynecol 2003;188:1161–3.
12. Foster AM, van Dis J, Steinauer J. MSJAMA. Educational and legislative initiatives affecting residency training in abortion. JAMA 2003;290:1777–8.