In the United States 14% of the gross national product is spent on health care.1 In response to the rising cost of medical care, managed care—defined as organizations that deliver comprehensive health care at a decreased cost by controlling the provision of services2—was conceived. Although some attempts to control medical costs through prepaid contracts began as early as the 1930s,3 membership in managed care organizations has only recently become common. Sixty million people, approximately 30% of the insured population of the United States, are served by managed care.4 Weiner5 estimates that by the year 2000, up to 65% of Americans will participate in some form of managed care.
Managed care organizations provide medical care at a 20–40% lower cost than traditional fee-for-service plans3 because they control the provision of care. Concern exists that the quantity and quality of patient care may be compromised by such cost containment. Although some studies have found few differences between managed care and fee-for-service regarding access or quality,6 other studies found that the elderly and chronically ill patients at the poverty line fared worse in the managed care system.7 Furthermore, some patients with potentially life-threatening illnesses have been denied care by their managed care plans,4 and managed care gatekeeping practices have been associated with adverse outcomes in patients presenting to emergency rooms.8 Francis et al9 demonstrated that in patients with colorectal cancer, membership in a health maintenance organization was associated with a longer delay in receiving definitive surgical treatment.
Because of concerns about the effect of managed care plans on the quality of patient care and because changing demographics seem to ensure an increasing proportion of elderly women,10 it is important to examine the effects of managed care in the fields of gynecology and gynecologic oncology. This study addresses the question of whether membership in a managed care organization is associated with a delay in receiving definitive surgical care in patients with benign gynecologic or gynecologic oncologic diseases.
Materials and Methods
A retrospective case-control study was carried out in which we identified 400 women who had definitive surgical treatment for benign gynecologic or gynecologic oncologic diseases between 1994 and 1997 at our institution (Figure 1). Patients were identified by reviewing all billing records for the specified time period and were included if any one of the procedures of interest was coded and the patient had one of the specified preoperative diagnoses. Procedures accepted as definitive for this study included hysterectomy (abdominal, vaginal, laparoscopic-assisted vaginal, or radical abdominal hysterectomy), endometrial ablation, anterior or posterior colporrhaphy, myomectomy, oophorectomy, retropubic urethropexy, or sacrospinous ligament fixation. Additional inclusion criteria included gynecologic diagnoses of severe cervical dysplasia, fibroids, dysfunctional uterine bleeding, stress urinary incontinence, and uterovaginal or vaginal prolapse, or gynecologic oncologic diagnoses of ovarian cancer, cervical cancer, endometrial carcinoma, and pelvic or adnexal masses. Exclusion criteria included diagnoses of pelvic inflammatory disease, ectopic pregnancies, chronic pelvic pain, or endometriosis. Pelvic inflammatory disease and ectopic pregnancies were excluded because they are acute disease processes for which urgent care is frequently rendered outside the standard preauthorization channels, and chronic pelvic pain and endometriosis were excluded because they are diseases for which definitive surgical treatment is often delayed for other reasons.
Each of these two groups was subdivided into patients with managed care plans and those with fee-for-service plans. Patients were determined to belong to a managed care organization by correlating the Financial Status Classification number to the need for a referral. If a referral was required, the patient was included in the managed care group. Both health maintenance organizations and preferred provider organizations were included in the managed care group. The fee-for-service group included self-pay patients, those with traditional indemnity plans, and those with medicare coverage, which functions as fee-for-service in our community. Each patient's chart was reviewed independently by an investigator who did not participate in the data analysis in conjunction with two billing specialists to ensure proper group assignment.
The primary predictor variable was defined as membership in a managed care organization. The primary outcome variable, delay of receipt of definitive surgical treatment, was defined as the interval between initiation of care and the date of the definitive surgical therapy; initiation of care was considered to be the time at which the diagnosis that eventually led to surgery was established. Delay times for all groups are expressed as mean days ± standard error of the mean (SEM), because they did not follow a Gaussian distribution. Demographic data and dependent variables examined included patient age, primary diagnosis at admission, primary surgical procedure, operative time, estimated blood loss, number of surgical and postsurgical complications, length of hospital stay, number of emergency room visits attributable to the gynecologic diagnosis during the delay period, number of clinic visits excluding the preoperative visit, number of second opinions required, number of alternative treatments attempted, and evidence of evaluations or treatments for the primary diagnosis before presentation to our clinic.
The data analysis was performed by an investigator blinded to group assignment. An unpaired comparison of means test (Student t test) was performed for continuous variables, except for populations with unequal standard deviations, in which case the Mann-Whitney test was substituted. Fisher exact test was applied for dichotomous variables and nominal variables with more than two categories. Two-tailed P testing was specified, and significance was set at .05 for all tests. Statistical analysis was carried out using Instat v2.01 (Graphpad Software, San Diego, CA) and SPSS 6.1 (SPSS, Inc, Chicago, IL).
An interim power analysis was performed with 300 patients. The observed mean delay of treatment was 72.43 ± 7.5 days (mean ± SEM). With a two-tailed P, an α of 0.05, and a β of 0.2, a sample size of 96 patients in each arm was necessary to detect a 40% increase in the mean delay of treatment as a result of membership in a managed care organization.
Of the 400 patients who met inclusion criteria, 207 had benign gynecologic diagnoses and 193 had gynecologic oncologic diagnoses. In the benign gynecologic group, 122 belonged to a managed care organization and 85 had fee-for-service arrangements. In the gynecologic oncologic group, 96 belonged to a managed care organization and 97 had fee-for-service insurance (Figure 1). The distribution of definitive surgical procedures performed is shown in Table 1.
There were no statistically significant differences in the distribution of disease diagnoses between subgroups in either the benign gynecologic or gynecologic oncologic groups, (Table 2) except that fee-for-service patients had a significantly higher incidence of genital prolapse at the time of definitive surgical therapy than their managed care cohorts (P < .01).
The primary outcome of interest in this study was the delay in receipt of definitive surgical care. Managed care patients with benign gynecologic diagnoses had a longer mean delay than did fee-for-service patients (133.7 ± 21 days compared with 84.9 ± 12.8 days, P = .03; data expressed as mean ± SEM). No statistically significant difference was noted in managed care patients with oncologic diagnoses compared with fee-for-service patients (35.7 ± 7.4 days compared with 20.5 ± 2.5 days, P = .29; Figure 2). A subanalysis of the time interval between the date on which surgical preauthorization was submitted (managed care patients) or on which the decision was made to proceed with surgery (fee-for-service patients) and the date the surgery was performed found no difference between the two groups (36.9 ± 28.0 days, managed care, compared with 37.0 ± 35.0 days, fee-for-service; P = .98).
Managed care patients were significantly younger than their fee-for-service counterparts when grouped by gynecologic diagnoses (46.4 ± 9.7 years compared with 56.5 ± 14.9 years, P < .001) and when grouped by gynecologic oncologic diagnoses (47.5 ± 13.2 years compared with 60.9 ± 15.8 years, P < .001; Table 3). The mean age of all managed care patients in this study regardless of diagnosis was lower than that of all fee-for-service patients (46.9 ± 11.4 years compared with 58.9 ± 15.5 years, P < .001).
No significant differences were noted between subgroups for operative time, number of complications, length of stay, number of emergency room visits, number of clinic visits, number of second opinions required, number of alternative treatments required, number of prior evaluations, or number of prior treatments. A statistically significant, but clinically nonrelevant, increase in estimated blood loss was found in managed care patients with gynecologic diagnoses compared with fee-for-service patients (Table 3).
Managed care has evolved rapidly since the early 1980s, with the number and complexity of plans increasing exponentially.11 This rapid growth has had profound implications for physicians and patients alike. Patients' choices of physicians and access to specialists have become restricted, while physician autonomy and remuneration has diminished, and dominance has shifted from nonprofit hospitals to for-profit corporations.12 One unifying characteristic of all managed care plans is the intent to modify the behavior of health care providers to contain costs.3 An obvious concern has been the impact of these cost containment measures on patient care.13 Because there are 147 elderly women for every 100 elderly men in this country,14 the adverse impact of managed care will be particularly important in the fields of gynecology and gynecologic oncology.
In the present study, gynecology patients served by managed care plans had a delay in definitive surgical therapy. Participation in managed care has also been associated with a delay in surgical care for patients with colorectal carcinoma.9 It is interesting to note that all of the gynecologic diagnoses in the patients with benign gynecologic diagnoses (genital prolapse, dysfunctional uterine bleeding, cervical dysplasia) were non-life-threatening conditions with multiple, alternative, non-definitive treatments available. This finding suggests that a patient enrolled in a managed care organization might experience delays in receiving definitive surgical treatment of a nonurgent condition for which there are alternative, less expensive treatments. We were not able to confirm any significant increase in the use of alternative treatments because of the retrospective study design. It is possible that alternative treatments, such as hormone therapy for dysfunctional uterine bleeding, might be as effective as definitive surgery and that a fee-for-service arrangement contributes to earlier performance of a definitive surgery at the expense of effective alternatives. Such questions are difficult to answer with a case-control design; a prospective cohort trial is warranted to better define the factors contributing to the delay of care noted in managed care gynecology patients. Related issues that might be examined in the context of a prospective trial are the appropriateness of patient awareness regarding causes for delay in definitive surgical treatment and the impact of these factors on patient satisfaction.
We did not find a significant delay in surgical care for gynecologic oncology patients, which we interpret as managed care policymakers recognizing the more urgent need for definitive surgical therapy in those cases. Our study was sufficiently powered to detect a 40% delay in care with this group, a level we considered clinically relevant. The 15-day delay we noted did not reach statistical significance, although it is possible that with a sufficiently large study we could have demonstrated a significant delay at a different effect size. A relevant question, though, is whether a 10- to 15-day delay in receiving definitive surgical care for gynecologic malignancies has any meaningful clinical impact. We found, however, one potentially important effect of managed care on this population; shorter length of hospital stay in the managed care population compared with the fee-for-service population almost attained significance (2.9 ± 1.6 days compared with 3.7 ± 2.7 days, P = .06). Although this study did not have the power to detect this dependent variable as a primary outcome, it supports previously reported findings in which managed care patients have up to a 20% shorter hospital stay than fee-for-service patients.15
We defined “delay of surgery” as the interval between the date of initiation of care and the date of surgery, believing this to be the most clinically relevant definition. With this definition, we were able to demonstrate a delay in the receipt of definitive surgical care in managed care patients with benign gynecologic diagnoses. We did not find a difference between groups in the time interval between the date on which surgical preauthorization was submitted (managed care patients) or on which the decision was made to proceed with surgery (fee-for-service patients) and the date the surgery was performed. The delay in treatment occurred before the decision to perform definitive surgical treatment was made, and there was no difference between managed care or fee-for-service plans in approving or scheduling of surgery. This finding might support the hypothesis that that managed care plans promote alternative treatments.
A surprising finding was that genital prolapse was more frequent in our fee-for-service population compared with our managed care population. However, subjects were identified retrospectively on the basis of receipt of definitive surgical treatment. One explanation for the observed disparity might be that fee-for-service patients with genital prolapse were more successful at obtaining definitive surgical treatment than were managed care patients. This type of hypothesis is best tested with a prospective study design to better control sample bias.
One negative finding of interest was the apparent absence of impact of managed care on traditional measures of quality of surgical care, such as operative times, complication rates, length of stay, and emergency room visits. Although these factors were only secondary outcome measures in this study, more sensitive assessments of surgical quality will be necessary in future studies. Alternatively, managed care might exert a greater effect preoperatively rather than intraoperatively or postoperatively.
Our findings suggest that women in managed care plans with nonurgent benign gynecologic diagnoses might experience a delay in receiving definitive surgical care. As managed care organizations evolve, ongoing evaluation of them will be essential in maintaining appropriate quality of patient care.
1. Acker JC. The future of oncology under managed care. N C Med J 1996;57:84–6.
2. Iglehart JK. The American health care system. N Engl J Med 1992;327:742–7.
3. Fairfield G, Hunter DJ, Mechanic D, Rosleff F. Managed care. Origins, principles, and evolution. BMJ 1997;314:1823–6.
4. Zautcke JL, Fraker LD, Hart RG, Stevens JS. Denial of emergency department authorization of potentially high-risk patients by managed care. J Emerg Med 1997;15:605–9.
5. Weiner JP. Forecasting the effects of health reform on US physician workforce requirement. JAMA 1994;272:222–30.
6. Riley GF, Potosky AL, Lubitz JD, Brown ML. Stage of cancer at diagnosis for Medicare HMO and fee-for-service enrollees. Am J Public Health 1994;84:1598–604.
7. Ware JE, Bayliss MS, Rogers WH, Kosinski M, Tarlov AR. Differences in 4-year health outcomes for elderly and poor, chronically ill patients treated in HMO and fee-for-service systems. JAMA 1996; 276:1039–47.
8. Young GP, Lowe RA. Adverse outcomes of managed care gatekeeping. Acad Emerg Med 1997;4:1129–36.
9. Francis AM, Polissar L, Lorenz AB. Care of patients with colorectal cancer—a comparison of a health maintenance organization and fee-for-service practices. Med Care 1984;22:418–26.
10. Taueber CM. America in transition: An aging society. U.S. Department of Commerce, Special Studies, Series P-23. Washington DC, 1983;128:1–18.
11. Miller RH, Luft HS. Managed care plans: Characteristics, growth, and premium performance. Annu Rev Public Health 1994;15:437–59.
12. Fairfield G, Hunter DJ, Mechanic D, Rosleff F. Implications of managed care for health systems, clinicians, and patients. BMJ 1997;314:1895–8.
13. Miller RH, Luft HS. Does managed care lead to better or worse quality of care? Health Aff 1997;16:7–25.
14. Butler RN, Collins KS, Meier DE, Muller CF, Pinn VW. Older women's health: “Taking the pulse” reveals gender gap in medical care. Geriatrics 1995;50:6–8.
15. Miller RH, Luft HS. Managed care plan performance since 1980 JAMA 1994;271:1512–9.