Although prices for medical services are known to vary markedly between hospitals, it remains unknown whether variation in hospital prices is explained by differences in hospital quality or reimbursement from major insurers. We obtained “out-of-pocket” price estimates for coronary artery bypass grafting (CABG) from a random sample of US hospitals for a hypothetical patient without medical insurance. We compared hospital CABG price to (1) “fair price” estimate from Healthcare Bluebook data using each hospital’s zip code and (2) Society of Thoracic Surgeons composite CABG quality score and risk-adjusted mortality rate. Of 101 study hospitals, 53 (52.5%) were able to provide a complete price estimate for CABG. The mean price for CABG was $151,271 and ranged from $44,824 to $448,038. Except for geographic census region, which was weakly associated with price, hospital CABG price was not associated with other structural characteristics or CABG volume (p >0.10 for all). Likewise, there was no association between a hospital’s price for CABG with average reimbursement from major insurers within the same zip code (ρ = 0.07, p value = 0.6), Society of Thoracic Surgeoncomposite quality score (ρ = 0.08, p value = 0.71), or risk-adjusted CABG mortality (ρ = −0.03 p value = 0.89). In conclusion, the price of CABG varied more than 10-fold across US hospitals. There was no correlation between price information obtained from hospitals and the average reimbursement from major insurers in the same market. We also found no evidence to suggest that hospitals that charge higher prices provide better quality of care.
Health care in the United States (US) is costly and accounts for nearly 18% of the gross domestic product, As a result, health care expenses are a leading cause of financial stress in US households. Moreover, unlike other businesses where consumers choose services based on prices, the price of health care services in the US is usually not known until the services are received, making it difficult for patients to function as educated consumers. Furthermore, although major insurers negotiate discounted prices for their members, uninsured patients do not enjoy a similar protection. Therefore, increasing transparency in medical prices has gained significant traction in recent years as a measure to reduce costs by engaging patients in comparison shopping and encouraging competition between hospitals. Although previous studies have shown a 10-fold variation in price of health care services across US hospitals, it remains unknown whether hospitals that charge higher prices for uninsured patients also receive high reimbursement for insured patients or provide greater quality of care. Clarifying this relation will be of critical importance to patients who may be willing to pay a higher price for superior quality of care. To address this gap in knowledge, we contacted a sample of US hospitals by telephone to request hospital and physician prices for coronary artery bypass grafting (CABG) for a hypothetical patient without health insurance. We examined the association between a hospital’s price for CABG with (1) average hospital reimbursement from major insurers within the hospital’s zip code and (2) risk-adjusted mortality rate and composite quality score for CABG obtained from the Society of Thoracic Surgeons (STS) database.
Methods
We used 4 main sources of data in our study: (1) Medicare part A data, 2010 to identify CABG hospitals; (2) American Hospital Association (AHA) data, 2010 for hospital structural characteristics; (3) Healthcare Bluebook data for average reimbursement for CABG in the hospital’s zip code of location ( https://www.healthcarebluebook.com ); and (4) STS data, 2013 for hospital risk-adjusted CABG mortality and composite quality score.
We used Medicare data to identify all US hospitals that performed at least 10 CABG surgeries on Medicare patients in 2010 (latest year of Medicare data available at our institution). From that list, we randomly selected 2 hospitals from each of the 50 US states and the District of Columbia. A total of 101 hospitals were identified (Vermont only had one hospital that met our CABG volume criteria) for participation in the study ( Table 1 ). The list of 101 study hospitals is provided in the Supplementary Table 1 .
| Variables | All U.S. hospitals, N = 1141 | Study hospitals, N = 101 | p values |
|---|---|---|---|
| CABG volume | |||
| <50 | 293 (26%) | 19 (19%) | 0.35 |
| 50-99 | 334 (30%) | 33 (33%) | |
| 100-199 | 348 (30%) | 30 (30%) | |
| >200 | 166 (15%) | 19 (19%) | |
| Geographic area | |||
| North Mid-Atlantic | 149 (13%) | 17 (17%) | |
| South Atlantic | 170 (15%) | 18 (18%) | |
| North Central | 305 (27%) | 24 (24%) | 0.43 |
| South Central | 266 (23%) | 16 (16%) | |
| Mountain-Pacific | 234 (21%) | 26 (26%) | |
| Teaching status | |||
| Yes | 229 (20%) | 31 (31%) | 0.014 |
| No | 903 (79%) | 70 (69%) | |
| Location | |||
| Urban | 1036 (91%) | 93 (92%) | 0.9 |
| Rural | 85 (7%) | 8 (8%) | |
| Ownership | |||
| Non-profit | 795 (70%) | 75 (74%) | 0.66 |
| For-profit | 219 (19%) | 16 (16%) | |
| Government | 118 (10%) | 10 (11%) | |
Based on previous work and method developed by our author group, we developed a standardized script to conduct telephone interviews to obtain price for CABG from our study hospitals (included in the Supplementary Material ). We developed a hypothetical scenario in which the 62-year-old father of one of the study investigators (BDG) was advised to undergo CABG, and the caller was trying to obtain a price estimate for that procedure. Briefly, the caller’s father was a previously healthy man who developed angina over the past few months. He had undergone multiple testing and imaging procedures, had failed medical therapy, and had multiple evaluations from physicians, all of whom recommended CABG. The caller’s father did not have health insurance but would be willing to pay for the procedure “out-of-pocket.” Therefore, the caller was “shopping” for the best possible price for the surgery and would be comparing prices from different hospitals before choosing a hospital. Additional details regarding the medical history, results of the cardiac testing, social history, expected length of stay, postdischarge care, the Current Procedural Terminology and International Classification of Diseases–Ninth Clinical Modification codes were also included in the script. The study’s lead investigator (BDG) made all the telephone calls to study hospitals. During the telephone call, every effort was made to use the script as much as possible. Pilot testing of the script was completed on 4 hospitals that were not included in the study sample to assess for content, structure, and clarity; modifications and revisions were made as needed.
All telephone interviews were conducted during January 1, 2014, to February 11, 2014. The invesigator (BDG) called the main hospital telephone number for each hospital and requested to be connected to an office or a department that could provide a “cash” or “out-of-pocket” price estimate for a surgery. If the main hospital operator was unable to find the appropriate contact person or department, the caller requested to speak with either the financial department or the patient billing office. On being transferred to the appropriate department, the caller repeated her request to speak with someone who could provide a price estimate for a surgery. Once connected with someone who stated they could provide that information, the caller transitioned immediately to the script for the remainder of the interview. The caller requested an estimate for the bundled price (hospital and physicians fees). If the hospital was unable to provide a complete bundle price, information regarding affiliated cardiac surgery practices was obtained. If more than one cardiac surgery practice was affiliated with the hospital, each practice was contacted in alphabetical order until a price estimate was obtained. All prices were recorded in addition to information regarding the phone call process. When the caller was unable to speak directly with the appropriate person, a standard message with the reason for the phone call and a callback telephone number was left. If a hospital declined to provide a price estimate, the reason for this was recorded. Every hospital was contacted a maximum of 3 times. If the caller was unable to obtain a price estimate after 3 separate attempts that hospital was deemed as unable to provide an estimate.
Our primary outcome was the complete price for CABG obtained as a bundled price or after summing the individual hospital and physician price. We used the AHA data, 2010 to obtain information regarding each hospital’s structural characteristics—geographic census region (North Mid-Atlantic, South Atlantic, North Central, South Central, and Mountain Pacific), teaching status, location (urban vs rural), and ownership structure (nonprofit, for-profit, and government). In addition, for each hospital that was able to provide a compete bundle price, we also obtained data regarding average reimbursement price for CABG within the zip code of the hospital location using the Healthcare Bluebook website.
Finally, we obtained data regarding hospital quality from the STS website. Briefly, the STS database collects data on 11 quality measures that are endorsed by the National Quality Forum and are reported under the following 4 domains: (1) perioperative medical care, (2) operative care, (3) avoidance of risk-adjusted mortality, and (4) avoidance of risk-adjusted major morbidity. Performance in individual domains is combined to yield a composite STS CABG quality score for each hospital. For this study, we obtained the risk-adjusted mortality and the overall CABG composite score from the STS website for hospitals that provide a price estimate. Additional details regarding the quality measures from the STS database are described in detail elsewhere.
First, we compared characteristics of our study hospitals with all hospitals that met our study inclusion criteria (i.e., performed CABG on at least 10 Medicare patients during the year 2010). Second, we compared characteristics of hospitals that were and were not able to provide a complete price (either bundled or separate) using the chi-square test or unpaired t test when appropriate. Third, we examined hospital variation in price for CABG between hospitals using graphical methods and examined the association between hospital price, structural characteristics (geographic census region, teaching status, urban/rural location, and ownership status), and Medicare 2010 CABG volume. We also examined the correlation between a hospital’s quoted price for CABG with (1) average reimbursement for CABG obtained from the Healthcare Bluebook; (2) hospital’s overall STS CABG quality composite score; and (3) STS risk-adjusted mortality rate using Pearson’s correlation.
All analyses were conducted with SAS 9.4 (Cary, North Carolina) and Microsoft Excel (version 14.4). The Institutional Review Board at the University of Iowa, whom waived the requirement for informed consent, approved the study.
Results
A total of 101 hospitals were included in our study. A comparison between our study hospitals and all US hospitals (n = 1,141) that performed CABG in Medicare patients during 2010 is provided in Table 1 . As can be seen from the table, hospitals included in our study were generally similar to US. CABG hospitals except a higher prevalence of teaching hospitals were in our study population (31% vs 20%).
Of the 101 hospitals in our study, only 9 hospitals (8.9%) were able to provide a complete bundled price (combined physician and hospital price) for CABG. Complete price was obtained from an additional 44 hospitals (43.6%) after calling the hospital and physician offices separately, resulting in a total of 53 hospitals (52.5%) that were able to provide a price for CABG.
Hospitals that were able to provide a price for CABG were more likely to be located in the North Central census region (p value <0.01) and in rural locations compared to hospitals that were not able to provide a complete price ( Table 2 ). However, there was no difference between hospitals that were and were not able to provide a price for CABG with regards to the Medicare CABG volume, teaching status, and ownership status ( Table 2 ). Data regarding STS quality measures for CABG were available from only 23 hospitals (50.9%) that were able to provide a price for CABG, and 16 (33%) that were not able to provide a price for CABG. Overall, the mean STS composite score for these hospitals was 96.4 (SD 1.14, range 94.1 to 98), and mean risk-adjusted CABG mortality rate was 2.1% (SD 0.33, range 1.6 to 2.9), and did not differ between hospitals that were and were not able to provide a CABG price.
| Variable | Hospitals | ∗ p value | ||
|---|---|---|---|---|
| All hospitals, N=101 | Hospitals able to provide a price, N=53 | Hospitals unable to provide a price, N=48 | ||
| CABG volume | ||||
| < 50 | 24 (24%) | 12 (23%) | 12 (25%) | 0.87 |
| 50-99 | 34 (34%) | 16 (30%) | 18 (38%) | |
| 100-199 | 28 (28%) | 17 (32%) | 11 (23%) | |
| >200 | 15 (15%) | 8 (15.%) | 7 (15%) | |
| Geographic Region | ||||
| North Mid-Atlantic | 17 (17%) | 6 (11%) | 11 (23%) | < 0.01 |
| South Atlantic | 18 (18%) | 9 (17%) | 9 (19%) | |
| North Central | 15 (15%) | 14 (26%) | 1 (2%) | |
| South Central | 25 (25%) | 6 (11%) | 19 (40%) | |
| Mountain-Pacific | 26 (26%) | 18 (34%) | 8 (17%) | |
| Teaching Status | ||||
| Yes | 29 (29%) | 14 (26%) | 15 (31%) | 0.6 |
| No | 72 (71%) | 39 (74%) | 33 (69%) | |
| Location | ||||
| Urban | 82 (81%) | 39 (74%) | 43 (90%) | 0.04 |
| Rural | 19 (19%) | 14 (26%) | 5 (10%) | |
| Ownership | ||||
| Non-profit | 73 (72%) | 41 (77%) | 32 (67%) | 0.4 |
| For-profit | 16 (16%) | 6 (11%) | 10 (21%) | |
| Government | 12 (12%) | 6 (11%) | 6 (13%) | |
| STS Scores availability | N = 39 | N = 23 | N = 16 | |
| Mortality scores | 2.1 | 2.1 | 2.2 | 0.53 |
| Composite scores | 96.4 | 96.6 | 96.2 | 0.34 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
