The remainder of the survey consisted of 10 questions asking the participants to supply their primary specialty; professional organization affiliations; year of medical school graduation; training status; sex; primary practice structure; primary practice community; and percentage of patients currently seen who had Medicare, Medicaid, or no insurance.

Statistical Analysis

We used inverse probability weights to correct for disproportionate sampling of particular physician subgroups when presenting data on overall physician attitudes. Because 2 major physician groups [surgeons and those who had graduated from medical school since 1990] were oversampled in our study, we divided respondents into 4 separate weighting groups: 1) nonsurgeons who graduated before 1990, 2) nonsurgeons who graduated during or after 1990, 3) surgeons who graduated before 1990, and 4) surgeons who graduated during or after 1990. We used data from the AMA Physician Masterfile to estimate the total number of U.S. physicians in each of these 4 groups when calculating our sampling weights. The probabilities used to determine these weights were 1 in 590, 1 in 513, 1 in 223, and 1 in 187, respectively. For statistical comparisons, we condensed "strongly support" and "generally support" responses into a single "support" category and "strongly oppose" and "generally oppose" responses into a single "oppose" category. We used logistic regression to assess relationships between demographic and professional characteristics and attitudes toward national health insurance. Multivariate models were adjusted for covariates chosen a priori, including sex; year of graduation; training status; specialty distinction; practice setting; practice location; and self-reported proportions of patients on Medicare, on Medicaid, and without insurance. In adjusted analyses, we dichotomized proportions of patients receiving Medicare, patients receiving Medicaid, and uninsured patients into categories of "low" and "high" using the approximate median values for each of those insurance options. We excluded neutral respondents from adjusted analyses to determine major predictors of support or opposition. Calculations were performed using the Stata statistical package, version 7.0 (Stata Corp., College Station, Texas).

Role of the Funding Sources

The funding sources had no role in the design, conduct, or reporting of the study or in the decision to submit the manuscript for publication.

Results

Top Editors' Notes Methods Results Discussion Author & Article Info References

Sample

Of 3188 mailed surveys, 338 were returned by the post office with no forwarding address and 121 were returned by nonpracticing physicians. We received 1650 surveys from the 2729 eligible participants, yielding an effective response rate of 60%. This response rate varied less than 3% among the large random sample of 2101 physicians and the 2 random oversamples of 310 additional recent medical school graduates and 318 additional surgeons. Because of missing data for individual questions, not all totals equaled 1650. Respondents were similar to nonrespondents in sex and years since medical school graduation. Response rates did not vary substantially by major specialty category (primary care, 60%; surgeons, 62%; subspecialists, 59%). Table 1 shows the major characteristics of respondents from our general random sample of all physicians from the AMA Physician Masterfile.

Attitudes about National Health Insurance Financing

Our survey found that 49% of physicians supported legislation to establish national health insurance ("strongly support," 18%; "generally support," 31%) and 40% opposed it ("strongly oppose," 21%; "generally oppose," 19%) (Table 2). Twenty-six percent of physicians supported a single federal payer system ("strongly support," 9%; "generally support," 17%), and 60% opposed it ("strongly oppose," 33%; "generally oppose," 27%). However, among the 49% of all physicians who indicated support for governmental legislation to establish national health insurance, 61% supported a single federal payer option to achieve that goal. Among major practice specialties, pediatricians and psychiatrists were most supportive of legislation to establish national health insurance, while anesthesiologists and surgical subspecialists (neurosurgeons, orthopedists, urologists, or those completing a fellowship after general surgery training) were least supportive (Figure 2).

View larger version (27K): [in this window] [in a new window]   Figure 2. Physician support for national health insurance (NHI) financing by practice specialty.

In analyses adjusting for potential differences in other personal and practice characteristics, several predictors of support were statistically significant (Figure 3). Physicians in primary care specialties, those reporting that at least 20% of their patients received Medicaid, those not in private practices, and those practicing in the inner city were more likely to support legislation to establish national health insurance than were their respective counterparts. Among those indicating support for legislation to establish national health insurance, physicians in primary care specialties and those not in private practice were statistically significantly more likely to indicate "strong" (rather than "general") support than were their counterparts. In addition, among all physicians who supported the general principle of governmental legislation to establish national health insurance, the single federal payer option was particularly appealing to physicians in primary care specialties. Seventy-one percent of 190 primary care physicians in this group indicated support for the single federal payer option, compared with 58% of non–primary care physicians in this group (P = 0.05).

View larger version (30K): [in this window] [in a new window]   Figure 3. Predictors of physician support for governmental legislation to establish national health insurance (NHI).

Conversely, our survey indicated that opposition to governmental legislation to establish national health insurance was related to a few particular physician characteristics. For example, 53% of physicians who reported that fewer than 20% of their patients had Medicaid and 52% of physicians who reported that fewer than 10% of their patients had no insurance opposed this principle. Similarly, 59% of surgical subspecialists, 59% of anesthesiologists, 73% of physicians in rural practice locations, and 54% of physicians in private practice settings opposed governmental legislation to establish national health insurance. In analyses adjusting for potential differences in other personal and practice characteristics, physicians in private practice (P < 0.001), those practicing outside the inner city (P = 0.002), and those reporting that fewer than 20% of their patients had Medicaid (P = 0.018) were more likely to oppose governmental legislation to establish national health insurance than were their respective counterparts. Among physicians who opposed legislation to establish national health insurance, those who were male (P = 0.003), those who graduated from medical school before 1990 (P = 0.017), those who were in private practice settings (P = 0.017), and those who were not categorized as subspecialists (P = 0.024) were significantly more likely to indicate "strong" (rather than "general") opposition than were their respective counterparts.

Discussion

Top Editors' Notes Methods Results Discussion Author & Article Info References

Although our study indicates that more U.S. physicians support governmental legislation to establish national health insurance than oppose it, physician opinion on this topic remains mixed. In addition, only 26% of all U.S. physicians support a system in which the federal government is the sole payer for health care services. In analyses adjusting for differences in personal and practice characteristics, we found that respondent physicians who were in a primary care specialty, those who reported that at least 20% of their patients had Medicaid, those who practiced in a nonprivate setting, and those who were located in an inner city were statistically significantly more likely to support governmental legislation to establish national health insurance. With the exception of anesthesiologists, family practitioners, and surgical subspecialists, more than half of physician respondents in major practice specialties supported the principle of establishing national health insurance through governmental legislation.

There is arguably some face validity to the finding that physicians with lower patient care continuity and greater exposure to underinsured patient populations are less likely to oppose the principle of governmental action to expand health care financing. However, we were surprised to find that fewer than 50% of family practitioners supported governmental legislation to establish national health insurance. One possible explanation is that some relevant personal or practice characteristics of family practitioners differ from those of other primary health care providers. Although a detailed evaluation of these potential differences was beyond the scope of this study, we found that family practitioners were more likely than other primary health care providers (that is, general internists or general pediatricians) to be located in rural areas (20% vs. 11%; P = 0.007). Since rural physicians were less supportive of governmental legislation to establish national health insurance than were physicians in urban and suburban locations, less support among family practitioners might be explained, in part, by the fact that a greater proportion are located in rural settings.

We were also surprised that the overall level of physician support for governmental action to establish national health insurance was not higher. Although we found a majority of support in several important subgroups, none of these subgroups represented a majority of all U.S. physicians. For example, although 62% of physicians practicing in inner-city locations supported governmental legislation to establish national health insurance, fewer than one fifth of all physicians were located in inner cities. Similarly, of all practicing U.S. physicians, only approximately 24% were women, 27% were in primary care specialties, 42% reported that at least 20% of their patients received Medicaid, and 48% reported being in nonprivate practice settings. Because the strongest support for governmental legislation was concentrated within minority physician subgroups, overall support remained less than 50%.

Although differences in wording, format, and context make comparisons with other physician surveys difficult, we found no previous studies reporting that a plurality of U.S. physicians supports governmental action to guarantee health insurance for all citizens. A recent Harris Interactive poll found that 81% of physicians felt that at least some "fundamental change" is needed to make the current system more workable (13). This was a substantial increase above the 57% who wanted "fundamental change" when the Clinton health care reform proposal was defeated in 1994. This poll also found a similar growing discontent with the current health care system among large, medium, and small employers; health plan managers; and the general public. Our study enhances the findings of this poll by demonstrating that a substantial proportion of U.S. physicians support establishing national health insurance through governmental legislation. In addition, our survey identifies specific physician characteristics that are related to opposition or support of this financing strategy.

Our study has the typical limitations of self-report surveys, such as response bias and an inability to establish a causal relationship between differences in personal and professional characteristics and physician attitudes. However, we had a return rate of 60% from a large physician sample that was similar in demographic characteristics to the entire AMA Physician Masterfile. Furthermore, we found no statistically significant differences in response rates or in major characteristics of responding and nonresponding physicians across any major specialty categories. The survey's simple design, focus on physician attitudes about general strategies for guaranteeing health care financing, and timing outside the context of any major national proposals for health care reform may have avoided some of the potential biases inherent to previous physician surveys about national health insurance. Conversely, it is conceivable that some respondents may have misinterpreted the wording of 1 or both of the major study questions. Because we did not perform more exhaustive pilot testing to assess the cognitive qualities of our major study questions within larger subsets of our sample, we cannot be certain that each question was interpreted consistently across all physician specialties and practice settings. If the questions were not interpreted consistently, the level of physician support for establishing national health insurance through governmental legislation could be misrepresented. It is also possible that support for governmental legislation to establish national health insurance in principle may not predict supportive action toward particular "real world" proposals for health care reform. However, although the future actions of supportive or "neutral" respondents may be less predictable, it is likely that physicians who oppose the principle of governmental legislation to establish national health insurance will indeed oppose specific plans that are based on this financing strategy.

At the very least, this survey "takes the pulse" of the general state of support among U.S. physicians for governmental action to arrange health insurance financing. It also improves our understanding of how the level of this support varies across different personal, professional, and practice characteristics. Because physician attitudes may be relevant to policymakers interested in the timely design of a more widely acceptable national health care reform proposal, the reasons underlying the opposition of major financing principles warrant further research. Through a better understanding of physician attitudes toward the financing of national health insurance, we can better predict the overall acceptability of future reform proposals designed to address the increasing number of uninsured Americans.