Abstract
OBJECTIVE
To describe primary care office policies regarding care of uninsured patients.
DESIGN
Telephone survey of all adult primary care sites advertising in the area telephone directory. Sites were defined by ownership status, number of physicians, use of physician-extenders, and location. Policies assessed were whether the site was accepting new uninsured patients, billing policies, the availability of free or discounted care, and payment plans.
SETTING
Allegheny County, Pennsylvania.
PARTICIPANTS
Of the 359 sites identified, 240 (66.9%) responded, representing 794 physicians. Survey respondents included receptionists (40.4%), office managers (36.2%), and physicians (22.9%).
RESULTS
While the majority of all sites reported accepting new patients without health insurance (87.5%), policies regarding these patients varied significantly by ownership status and the number of physicians. Sites with 3 or fewer physicians were more likely to accept uninsured patients. Self-owned practices were more likely to require payment at the time of service, and provide discounted care, free care, and payment plans compared with hospital/health system practices or multisite group practices.
CONCLUSIONS
Willingness to accept uninsured patients does not always equate to affordable or accessible care. Office policies have the potential to be substantial obstacles to primary care.
Keywords: uninsured patients, access to primary care, practice characteristics
The growing numbers of persons without health insurance has been a central focus of the health care reform debate of recent years.1–3 While the benefits of health care coverage are well-documented, less is known about current policies regarding care of uninsured patients. Several authors have reported different levels and types of care provided to those patients with insurance compared to those without insurance.4–8 While aggregated and per-physician dollar amounts of free and discounted primary and hospital care have been reported,9–11 less is known about factors influencing the availability of this care. This is important given the increased consolidation and integration occurring in health care.12 Physicians working in communities with greater managed care penetration have been shown to provide less free care.13 Ownership status of the practice has also been associated with the level of free care provided, with self-employed/physician-owned practices providing more free care in one study13 and less in another.14
The aim of this study was to identify practice policies regarding uninsured patients among primary care providers in Allegheny County, Pennsylvania, identifying both the proportion of sites accepting new uninsured patients and office policies for the care provided.
METHODS
The sampling unit was primary care sites identified from primary care, general practice, or family care listings in the 1998 Yellow Pages telephone directory. Multiple physicians listed under a common telephone number or practice name were considered as a single site. Federally qualified health centers and federally funded community health clinics were not surveyed. However, academic health center–affiliated or –based clinics were included. The assessment was limited to family practice physicians and internists (allopathic and osteopathic) listed in the telephone directory as providing primary care to adults. Specialists providing primary care were included. Pediatricians providing primary care and subspecialists with limited or focused practice interests were not included.
Sites identified through this initial screening were contacted for a telephone interview by the project coordinator. The interviewer (PMS) introduced himself as calling on behalf of the Allegheny County Health Department to conduct a survey of practice policies regarding uninsured patients. All participants were informed that responses were strictly confidential, that no individual site would be identified by their response, and that they had the right to refuse the interview. The interview lasted less than five minutes and began by asking whether the site accepted uninsured patients. Those sites accepting uninsured patients were then asked their billing and payment policies. Policy questions included whether the site offered patients discounted-fee care, free care, payment plans, and whether they required full payment at the time of service. They were also queried as to whether they only accepted uninsured patients by referral, or for either acute or chronic conditions. Finally, all sites were asked about practice size, ownership structure, and use of housestaff and physician-extenders (to determine whether lower-cost providers affected office policies). Practice location was determined from the telephone listing and classified as urban (within city limits) or suburban. The survey instrument was developed for this study. Questions were field tested with different potential respondents (receptionists, office managers, physicians) prior to actual data collection.
Data from this survey were analyzed using FoxPro (Visual FoxPro; Microsoft Corp., Seattle, Wash) software for data entry and STATA software (STATA Corp., College Station, Tex) for statistical analyses. Categorical data were analyzed using χ2statistics with a two-sided α of 0.05 for significance. The dependent variables were acceptance of uninsured patients and payment options for billing. Independent variables were practice size, use of physician-extenders, ownership structure, and practice location.
RESULTS
Of 359 sites identified in the initial screening, 240 completed the interview (66.9%), representing 794 primary care physicians (86.0% of those practicing in Allegheny County). Respondents completing the surveys were receptionists (40.4%), office managers (36.2%), and physicians or other health providers on site (22.9%). The respondent groups did not differ in uninsured patient acceptance rates or billing practice responses.
Site characteristics are shown in Table 1). Most sites had 3 or fewer physicians (71.1%). Only 15.8% of sites were owned by the on-site physicians, compared with 42.0% of sites owned by a hospital/health system. Overall, 26.6% of sites had either a nurse practitioner or physician's assistant on site and 6.2% had housestaff rotating there. Of nonrespondent sites, 63.0% had 3 or fewer physicians, 63.0% were owned by a hospital or health system, and 53.4% were in urban locations.
Table 1.
Site Characteristics
| Primary care sites identified, N | 359 |
| Respondents, N (%) | 240 (66.9) |
| Physicians represented, N | 794 |
| Proportion of sites with* | |
| 1 Physician, n (%) | 94 (39.1) |
| 2–3 Physicians, n (%) | 77 (32.0) |
| 4–5 Physicians, n (%) | 37 (15.4) |
| >5 Physicians, n (%) | 29 (12.0) |
| Proportion of sites with | |
| Nurse practitioners, n (%) | 14 (5.8) |
| Physician assistants, n (%) | 50 (20.8) |
| Interns/residents, n (%) | 15 (6.2) |
| Ownership structure of practice site | |
| Hospital/integrated health system, n (%) | 101 (42.0) |
| Insurance company/HMO, n (%) | 13 (5.4) |
| Larger group practice, n (%) | 88 (36.6) |
| Solely by providers at that site, n (%) | 38 (15.8) |
| Sites that currently treat uninsured patients, n (%) | 215 (89.5) |
| Sites accepting new patients with no insurance, n (%) | 210 (87.5) |
| Sites accepting uninsured new patients that require | |
| Full payment at time of service, n (%) | 149 (70.9) |
| Will only treat acute problems, n (%) | 61 (29.0) |
| Will only treat chronic conditions, n (%) | 58 (27.6) |
| Sites accepting uninsured new patients that offer | |
| Sliding fee scale, n (%) | 76 (36.2) |
| Free care, n (%) | 82 (39.0) |
| Payment plan, n (%) | 163 (77.6) |
| Mean no. days for someone with no insurance to get an initial appointment | |
| <7 Days (%) | 96 (45.7) |
| 7–14 Days (%) | 78 (37.1) |
| 15–30 Days (%) | 8 (3.8) |
| >30 Days (%) | 12 (5.7) |
| Mean cost for an initial office visit (not including any lab or diagnostic fees), N = 140 | |
| <$25 (%) | 5 (3.5) |
| $26–$50 (%) | 60 (42.9) |
| $51–$100 (%) | 54 (38.6) |
| >$100 (%) | 20 (10.7) |
3 Missing.
Of the 240 sites, 215 (89.5%) reported they were currently treating uninsured patients and 210 sites (87.5%) reported they were accepting uninsured new patients. The most common reasons for not accepting uninsured patients were: 1) limited practice or the practice was full (10 of 30); 2) financially unfeasible (7 of 30); and 3) site regulations/policies prohibited it (4 of 30). Only 12 of the 30 sites not accepting uninsured patients had a policy of referring patients to sites where care would be available.
Of those sites accepting uninsured patients, 45.7% reported being able to see the patient within 7 days. Compared with urban sites, suburban sites were significantly more likely to accept patients without insurance (92.1% vs 84.3%; P < .001). Those sites with 4 or more physicians were significantly less likely to accept uninsured new patients compared with those practices with 3 or fewer physicians (84.8% vs 90.0%; P = .02). The use of nurse practitioners, physicians' assistants or interns/residents was not associated with accepting uninsured patients. There was also no significant difference in rates of accepting uninsured patients among practices with different ownership structures. Of those sites reporting the average cost for an initial “well adult” appointment (N = 140), 42.9% had professional fees between $26 and $50, and 38.6% between $51 and $100.
Overall, 70.9% of sites accepting uninsured patients required full payment at the time of service. In addition, 29.0% of sites limited the care available to only treating acute problems and 27.6% limited the care to only managing chronic conditions. However, 77.6% of sites did report they were willing to provide a payment plan for uninsured patients, and 36.2% reported having a sliding scale for their fees. Similarly, 39.0% reported free care available at their site. Those sites offering either discounted or free care were more likely to be owned by the physicians on site (Table 2) Practices owned by hospitals/health systems or large group practices were less likely to require payment at the time of service compared to physician-owned sites (61.6% vs 67.9% vs 97.2%; P < .001). However, these practices were also significantly less likely to offer discounted fees for their services (26.7 vs 26.1 vs 63.1; P < .001), free care (23.7% vs 26.1% vs 89.4%; P < .001), or payment plans (59.4% vs 62.5% vs 100%; P < .001). Free or discounted-fee care was not associated with practice location (urban vs suburban).
Table 2.
Policies of Primary Care Office Sites Regarding Uninsured Patients
| Hospital/Health System–owned Sites N = 101 | Part of a Larger Group Practice N = 88 | On-site Physician-owned Sites N = 38 | P Value* | |
|---|---|---|---|---|
| Accepts new patients without insurance, n (%) | 86 (85) | 78 (89) | 36 (95) | .12 (H/HS)6.28 (LGP) |
| n = 86†(%) | n = 78†(%) | n = 36†(%) | ||
| Requires payment at the time of service, n (%) | 53 (62) | 53 (68) | 35 (97) | <.001 (H/HS)<.001 (LGP) |
| Only sees uninsured patients with an acute problem, n (%) | 28 (33) | 23 (30) | 6 (17) | .07 (H/HS).14 (LGP) |
| Only sees uninsured patients with a chronic condition, n (%) | 26 (30) | 22 (28) | 7 (19) | .22 (H/HS).32 (LGP) |
| Offers a sliding scale/discount care, n (%) | 27 (31) | 23 (30) | 24 (67) | <.001 (H/HS)<.001 (LGP) |
| Offers a payment plan, n (%) | 60 (70) | 55 (70) | 36 (100) | <.001 (H/HS)<.001 (LGP) |
| Offers free care, n (%) | 24 (28) | 23 (30) | 34 (94) | <.001 (H/HS)<.001 (LGP) |
P value calculations are for matched samples, comparing proportions of hospital/health system–owned (H/HS) with physician–owned proportions and large group practice–owned (LGP) with physician-owned proportions.
The number of sites accepting new patients without insurance is used as the denominator for the percentages shown below.
CONCLUSIONS
This survey of primary care sites represents a novel approach to understanding issues and obstacles associated with accessing care if uninsured. Findings from this survey are notable for the conflicting policies that exist in many sites that potentially exclude patients from primary care. While the vast majority of sites surveyed reported they were accepting patients without health insurance, they also reported billing practices that are likely prohibitive, particularly for lower-income individuals. Over 70% of sites required full payment at the time of service, with professional fees between $51 and $100 dollars at almost 40% of all sites. At the same time, 36% of sites reported offering discounted-fee care, 39% provided free care, and 77.6% provided payment plans for medical bills. It is unclear from our survey how well advertised or consistently applied these policies are, particularly given the overlap in sites reporting both full payment at the time of service and the availability of payment plans. Findings from an earlier consumer survey indicated that most patients were not aware of the payment options of their providers (TPO, unpublished data, 1999). There were also significant limitations on the type of care made available to uninsured patients. Almost one in three sites limited care to only acute problems and a similar proportion to only chronic conditions. This is consistent with findings by Woolhandler et al. that fewer preventive services are available to uninsured persons.15
Site characteristics associated with different policies regarding care of uninsured patients are notable given the current market trends toward more consolidation, mergers, and purchases of smaller practices. Sites with more physicians on staff and those located in urban areas were less likely to accept uninsured patients. This may represent a confounding association of more large sites being located in urban settings. Alternately, the association between larger practices and uninsured care policies may reflect more restrictive business plans at these sites that preclude uninsured care. Practices in urban areas, where there is a greater concentration of uninsured and Medicaid populations, may have either reached a capacity for uninsured patients or be operating at a lower profit margin that precludes this care. Alternately, in urban settings where there are more community health centers for referrals, sites may feel less civic obligation to provide care to uninsured patients. Additional research is needed to clarify these findings.
Compared with physician-owned practices, hospital/health system–owned and larger group practice–owned sites were less likely to offer discounted-fee care, free care, or payment plans. These findings are consistent with a previous study13 and may reflect centralized billing and policy making along with a greater reliance on contracted billing services by health systems and large group practices. At physician-owned sites, where there is greater autonomy and on-site control in policy decisions, a greater proportion accepted patients without insurance and provided discounted-fee or free care and payment plans. That physician-owned practices more commonly require payment at the time of service may reflect smaller operating margins and more strained revenue streams. Since all of the hospitals and health systems in Allegheny County are not-for-profit, this designation does not seem to define practice behavior.
These findings have several policy implications. First, the data suggest that a willingness to accept patients without insurance does not always equal access to affordable care, and office policies have the potential to be a substantial obstacle to accessing primary care. As researchers assess community health access, it is important to take a more in-depth approach that accurately describes true health services availability. Second, uninsured adults need greater availability of free care and discounted-fee care. It is unclear how well-informed patients are regarding office policies. The current trends in health care consolidation make it unlikely that the practice of free and discounted care will expand without an external stimulus. Further work is needed to determine whether this care needs subsidization from a pooled “charity fund” or other incentives for more sites to participate. Finally, the issue of disproportionate share of uninsured care and the need to better coordinate available services among primary care sites is inferred by the urban/suburban distinction in uninsured care policies.
This survey has several limitations. First, it is a survey of only those sites listed in the telephone directory available to the general public and may not be entirely representative of primary care sites in the region. We purposely excluded those care sites specifically chartered to care for uninsured patients, to center our evaluation on the willingness and availability of the broader medical community to care for those without health insurance. The data presented here are self-reported, typically by office staff, to a public health agency–sponsored survey and may not necessarily reflect what actually takes place at that site. We did not confirm our findings with documented office policy or actual patient experience. Finally, the data represent findings from only one region and may not be representative of other parts of the country. However, Pittsburgh, like many metropolitan areas, has recently undergone a period of practice buy-outs and consolidations and has very high penetration of managed care, so we suspect that this environment is not unique.
In summary, barring a public policy breakthrough that expands the availability of affordable insurance to the 43 million Americans currently uninsured, we need to look to our existing systems of care to make it easier for uninsured persons to get primary and preventive care. This responsibility should be collectively shared.
Acknowledgments
This project was funded solely by the Allegheny County Health Department as part of their Health Access Initiative.
REFERENCES
- 1.Kitzhaber J. Uncompensated care—the threat and the challenge. West J Med. 1988;148:711–6. [PMC free article] [PubMed] [Google Scholar]
- 2.Donelan K, Blandon RJ, Hill CA, et al. Whatever happened to the health insurance crisis in the United States? Voices from a national survey. JAMA. 1996;276(16):1346–50. [PubMed] [Google Scholar]
- 3.Rice DP. Health status and national health priorities. West J Med. 1991;154:294–302. [PMC free article] [PubMed] [Google Scholar]
- 4.Mort EA, Edwards JN, Emmons DW, Convery K, Blumenthal D. Physician response to patient insurance status in ambulatory care clinical decision-making. Implications for quality of care. Med Care. 1996;34(8):783–97. doi: 10.1097/00005650-199608000-00006. [DOI] [PubMed] [Google Scholar]
- 5.Hafner-Eaton C. Physician utilization disparities between the uninsured and insured. JAMA. 1993;263(6):787–92. [PubMed] [Google Scholar]
- 6.Hadley J, Steinberg EP, Feder J. Comparison of uninsured and privately insured hospital patients. JAMA. 1991;265:374–9. [PubMed] [Google Scholar]
- 7.Patrick DL, Madden CW, Diehr P, Martin DP, Cheadle A, Skillman SM. Health status and use of services among families with and without health insurance. Med Care. 1992;30(10):941–9. doi: 10.1097/00005650-199210000-00006. [DOI] [PubMed] [Google Scholar]
- 8.Weissman J, Epstein AM. Case mix and resource utilization by uninsured hospital patients in the Boston metropolitan area. JAMA. 1989;261:3572–6. [PubMed] [Google Scholar]
- 9.Dunham NC, Kindig DA, Lastiri-Quiros S, Barham MT, Ramsay P. Uncompensated and discounted Medicaid care provided by physician group practices in Wisconsin. JAMA. 1991;265:2982–6. [PubMed] [Google Scholar]
- 10.Ohsfeldt RL. Uncompensated medical services provided by physicians and hospitals. Med Care. 1985;23(12):1338–44. doi: 10.1097/00005650-198512000-00003. [DOI] [PubMed] [Google Scholar]
- 11.Kilpatrick KE, Miller MK, Dwyer JW, Nissen D. Uncompensated care provided by private physicians in Florida. Health Serv Res. 1991;26(3):277–302. [PMC free article] [PubMed] [Google Scholar]
- 12.Goldberg LG. Health care consolidation and the changing health care marketplace. A review of the literature and issues. EBRI Issue Brief. 1999;214:1–15. [PubMed] [Google Scholar]
- 13.Cunningham PJ, Grossman JM, St Peter RF, Lesser CS. Managed care and physicians' provision of charity care. JAMA. 1999;281:1087–92. doi: 10.1001/jama.281.12.1087. [DOI] [PubMed] [Google Scholar]
- 14.Blumenthal D, Rizzo JA. Who cares for uninsured persons? A study of physicians and their patients who lack health insurance. Med Care. 1991;29(6):502–20. [PubMed] [Google Scholar]
- 15.Woolhandler S, Himmelstein DU. Reverse targeting of preventive care due to lack of health insurance. JAMA. 1988;259:2872–4. [PubMed] [Google Scholar]