The Reasons That Many Radiology Practices Don't Use Off-Hours Services

Article Outline

• Abstract
• Introduction
• Methods
• Results
  • Demographics
  • Practices With No Off-Hours Services
  • Practices With Internal Off-Hours Services
• Discussion
  • Demographics
  • Practices With No Off-Hours Services
  • Practices With Internal Off-Hours Services
  • Limitations
• Conclusions
• Acknowledgments
• References
• Copyright

Purpose

To compare radiology practices that use external, internal, and no off-hours services.

Methods

From August 2005 to June 2006, 300 nonspecialty hospitals randomly selected from the AHA Guide 2005 Edition were contacted by telephone, e-mail, and mail, with attempts made to speak to the chiefs of radiology. A total of 115 responses were obtained (a 38.3% response rate), with 64 from radiology practices that used external off-hours services, 13 from practices with internal services, and 38 from practices with no services. The demographics of the practices in the 3 categories were compared, and answers to category-specific survey questions were tabulated. Responses were analyzed using descriptive statistics.

Results

Radiology practices using internal off-hours services were significantly larger (mean size, 19.9 full-time radiologists) than those using external off-hours services (mean size, 8.2 full-time radiologists) and those not using any off-hours service (mean size, 10.7 full-time radiologists). A sufficient number of radiologists or residents covering nights had the highest reported importance in the decision not to adopt an external service. Cost and quality concerns were also cited. The consistency of interpreting radiologists known to a practice had the highest importance in the decision to use an internal rather than an external off-hours service. Frequent reasons cited for radiologists to take regular internal off-hours employment were financial incentives provided and a preference for off-hours shifts.

Conclusions

As long as there are sufficient numbers of radiologists and residents to handle the volume of interpretations, many practices will not use external off-hours services. Such services could help increase their adoption by offering lower cost and proven quality.

Key Words: Off-hours services, teleradiology, survey

Back to Article Outline

Introduction 

External off-hours interpretation services are groups of teleradiologists who interpret imaging for other radiology practices, primarily during night, weekend, and holiday hours. They are a new but increasingly prevalent phenomenon in the field of diagnostic radiology [1]. Although several studies have shown these services to be both safe and efficacious in the interpretation of imaging [2, 3, 4, 5], little has been documented regarding radiologists' thoughts and use of these services. Kaye et al [1] examined radiology practices' motivations for using external off-hours services, described how these services were used, and reported on current satisfaction with them. They found that the use of external off-hours services was limited to a small percentage of a practice's total interpretations, and that the primary reason for using these services was convenience, not a shortage of staff members [1].

Despite the convenience of external off-hours services, some practices have organized similar, but internal, services, and many practices use neither external nor internal off-hours services. In this paper, we describe the reasons practices with no off-hours services or with internal services gave for selecting their particular types of arrangements, as well noting some of the characteristics of each. We also present 3-way comparisons among practices that did not use off-hours services at all, those that used external off-hours services, and those that had internal off-hours arrangements. Although there is extensive literature about off-hours services [6, 7, 8, 9, 10, 11, 12, 13, 14], no study has addressed the subject of why practices choose their particular types of arrangements by actually surveying a substantial number of these practices.

Back to Article Outline

Methods 

The subjects for this study were diagnostic radiology practices at 300 US hospitals chosen randomly from the AHA Guide 2005 Edition. The specific inclusion and exclusion criteria were that the hospitals were listed in the AHA Guide 2005 Edition and that they were not specialty, rehabilitation, or psychiatric hospitals. The surveyors attempted to contact by telephone the physician heads of radiology at these hospitals. If the second attempt to reach the chief or head of radiology at a hospital was unsuccessful, the surveyors attempted to obtain an e-mail or regular mail address to which a survey could be sent and returned. Of the 300 hospitals contacted, 115 surveys were completed via telephone, e-mail, and regular mail between August 2005 and June 2006, for a 38.3% response rate. One hundred nine of the 115 surveys (94.8%) were completed by telephone.

We compared the demographic characteristics of radiology practices using the 3 types of arrangements (external, internal, and no off-hours services) and analyzed questions providing detailed information on the reasons for the use of the latter 2 categories of arrangements and on how these arrangements worked. For this analysis, we defined an external off-hours service as a group of radiologists that was a separate legal and physical entity from radiology practices with which it was contracted for off-hours (night, weekend, and holiday) image interpretation. We defined an internal off-hours service as a dedicated group of radiologists within a radiology practice who performed off-hours image interpretation (ie, a practice at which every member rotated call did not have an internal off-hours service). Finally, we defined a radiology practice with no off-hours service as one that did not contract with a separate physical and legal group of radiologists or use its own dedicated group of radiologists to interpret off-hours imaging. This classification generally referred to practices that rotated call among most, if not all, members. It also included practices at which residents were used for off-hours interpretations.

Exemption from requiring consent from each participant was granted by the Human Investigation Committee at Yale University School of Medicine. The surveyors assured each participant of their anonymity, and all identifying information was deleted from the surveys after the assignment of numerical codes for subsequent analysis.

The survey data were entered into an Excel spreadsheet (Microsoft Corporation, Redmond, Washington) and screened for missing values. There were 18 missing values relevant to our analyses; all omitted responses from survey questions. Because this number was quite small in relation to the total number of responses, these omissions did not meaningfully affect any statistical calculations. Depending on the type of data, we report averages and percentages. Statistical significance was evaluated with 2-tailed t tests.

Back to Article Outline

Results 

Demographics 

Of the 115 surveys completed, 64 (55.7%) were from radiology practices that used external off-hours services, 13 (11.3%) were from radiology practices that used internal off-hours service, and 38 (33.0%) were from radiology practices with no external or internal off-hours services. The 38 responding practices that did not use external or internal off-hours service had a median of 8 full-time radiologists, compared with a median of 4 among the 64 practices that used external off-hours services and a median of 19 among the 13 practices that used internal off-hours services (Table 1). The mean sizes of the practices were 10.7, 8.2, and 19.9 full-time radiologists, respectively. Mean practice size was significantly different between radiology practices using external off-hours services and those using internal off-hours services (P < .01), and was marginally significant between radiology practices using internal off-hours services and those using no off-hours services (P < .10). There was, however, no statistically significant difference between practices using no off-hours services and those using external off-hours services.

Table 1. Comparisons of radiology practices using demographic information

Demographic	External Off-Hours Service	Internal Off-Hours Service	No External or Internal Off-Hours Service
No. of practices	64	13	38
Mean (median) no. of full-time radiologists	8.2(4)	19.9(19)	10.7(8)
No. of solo practices	11(17.2%)	1(7.7%)	6(15.8%)
No. of practices with 2 to 9 full-time radiologists	32(50.0%)	4(30.8%)	16(42.1%)
No. of practices with 10 to 19 full-time radiologists	16(25.0%)	2(15.4%)	10(26.3%)
No. of practices with 20 or more full-time radiologists	5(7.8%)	6(46.2%)	6(15.8%)
No. of practices with part-time radiologists	29(45.3%)	4(30.8%)	15(39.5%)
Mean (median) no. of part-time radiologists (for practices with part-time radiologists)	2.6(2)	5.0(4)	2.4(2)
No. of practices with residents	1(1.6%)	1(7.7%)	7(18.4%)
Mean (median) no. of residents (for practices with residents)	15.0(15)	10.0(10)	9.2(4)
Mean (median) no. of hospitals covered	3.0(2)	6.6(6)	3.0(2)
Mean (median) no. of hospitals covered with emergency departments	2.5(1)	6.5(6)	2.8(2)

Sixteen percent of the practices with no off-hours services had 20 or more members, compared with 8% of the practices using external off-hours services and 46% of those using internal off-hours services. This difference was significant between practices using external off-hours services and those using internal off-hours services (P < .01), as well as between practices using internal off-hours services and those using no off-hours services (P < .05). Again, however, there was no statistically significant difference between practices using no off-hours services and those using external off-hours services.

Eighteen percent of the radiology practices with no off-hours service had residents, whereas only 2% of the practices using external off-hours services and 8% of those using internal off-hours services had residents. This difference was significant between practices that used external off-hours services and those that did not use any off-hours services (P < .05), but other differences were not significant.

Practices With No Off-Hours Services 

Having enough radiologists or residents to cover the night had the highest importance in driving the decision not to adopt an external off-hours service, with a mean score of 3.0 on a scale of 1 (“negligible or no importance”), 2 (“little importance”), 3 (“substantial importance”), and 4 (“very important”) (Table 2). Not rated equally important, but with the difference not significant, was an off-hours service being too costly for the benefits it offered. Rated significantly lower in importance was the off-hours service interruption of the involvement of radiologists in the practice of care (P < .01), the quality of an off-hours service being insufficient (P < .01), and having too much volume, which would mean that an off-hours service would cost too much (P < .01).

Table 2. Reasons for not using an external off-hours service

Of the 38 practices that did not use off-hours services, 82% said that the issue of using an off-hours service had been raised. In 80% of these cases, members of the practices had brought it up (Table 3).

Table 3. Who brought up the issue of off-hours services?

Of the 38 practices without off-hours services, only one had previously used an external off-hours service.

Sixty-eight percent of the radiology practices without off-hours services cited increases in their workloads, either through added imaging volume or staffing decreases, as a change that could cause them to adopt off-hours services (Table 4). Twenty-one percent of these practices cited cost decreases through such things as increased reimbursement, increased competition, cheaper technology, and cost sharing as a change that could cause them to adopt off-hours services. Other less often cited changes that could spur adoption included a change in the attitudes of radiologists toward off-hours services, the need to recruit new or desirable radiologists to the practice, and an improvement in the quality of off-hours services' interpretations. Finally, a few practices mentioned that they might adopt off-hours services if it gave final interpretations or if it maintained or improved the continuity of care.

Table 4. The types of changes that could take place leading to the use of an off-hours service

Practices With Internal Off-Hours Services 

The consistency of the radiologists known to a practice interpreting imaging had the highest importance in driving the decision to use an internal off-hours service rather than an external one, with a mean score of 3.8 out of 4 on a scale of 1 (“negligible or no importance”), 2 (“little importance”), 3 (“substantial importance”), and 4 (“very important”) (Table 5). All other reasons given for using an internal off-hours service were significantly (P < .05) lower in importance (Table 5). There were no significant differences among them in importance, however. Seven practices mentioned other reasons as driving their decisions, including 3 practices that stated that maintaining good relationships with clinicians by interpreting their own imaging around the clock (and thus demonstrating a commitment to patient care) was very important. Two other practices cited an increase in the quality of interpretations with an internal off-hours service as very important.

Table 5. Reasons for using an internal off-hours service

With a mean practice size of 19.9 full-time radiologists and a mean of 1.4 radiologists on off-hours shifts, radiology practices with internal off-hours services staffed those services with, on average, one radiologist for every 14 radiologists in the practices.

Of the 13 radiology practices that used internal off-hours services, 7 had permanent crews of radiologists who worked off-hours, whereas 6 used rotation systems. The 7 practices that used permanent crews of radiologists were asked about incentives to work off-hours. Four of these 7 practices said that there were no incentives (ie, the individuals wanted to work off-hours), and 3 said that financial incentives were used. The 6 practices that used rotation systems were asked how long radiologists spent on the rotations on average. Four of the 6 practices said one day, one said 7 days, and one said that the time varied between one and several days.

Of the 13 practices that used internal off-hours services, 11 reported that their services covered weekends in addition to weekdays. Ten of the 13 internal off-hours services were situated locally; the remaining 3 were located within the same states.

Three of the 13 practices that used internal off-hours services also performed interpretations for other radiology practices. Of these 3 practices, 2 reported that 20% of their interpretations were for other radiology practices, while the other practice estimated 5%. Two of these 3 practices stated that they wanted to increase the number of interpretations done for other practices, while the other desired no change. Finally, 7 of the 10 practices with internal off-hours services that did not perform interpretations for other radiology practices stated that they had no desire to do this in the future, whereas 3 reported that they were starting to recruit other practices.

Back to Article Outline

Discussion 

Demographics 

We were surprised that there were no statistically significant differences in size between practices that did not use any off-hours service and those using external off-hours services. We expected that the former would be larger, because larger size generally permits a less onerous call schedule. In any case, the relatively large sizes of practices with internal off-hours arrangements indicated that size facilitates such arrangements.

Our results suggest that radiology practices with residents have a relatively low likelihood of using external off-hours services, because they most likely use their residents for preliminary nighttime interpretations. However, this conclusion may be partially invalidated by the inclusion of practices with fewer than 10 members in our data. The rules of the Radiology Residency Review Committee in essence require a minimum size of 10 to cover subspecialties for a site to be a primary site for a residency. A majority of respondents in the external and no off-hours service categories had fewer than 10 radiologists.

As described below under “Limitations,” we have good evidence that practices using off-hours services were more likely to respond to our survey than practices not using such services. That is, well fewer than half of all practices were using external off-hours services as of the 2005-2006 date of our survey. Similarly, we would judge that fewer than 10% were using internal off-hours services at that time, although somewhat more than 10% of respondents reported using such services.

Practices With No Off-Hours Services 

Our results indicate that the adequacy of staffing is most important in the decision not to use an external off-hours service. A perception that the benefits of off-hours services were not worth the cost was also of high importance in this decision. This is in contrast with practices that do use external off-hours services. Cost did not seem to be an important issue for these practices, although two-thirds of them paid the off-hours services about as much as they collected or more.

Our data on how many practices had discussed the issue of using external off-hours services suggest that although off-hours services are not universally used by radiology practices, they are a very well known and widely discussed option for the practices not using them. Furthermore, our results suggest that off-hours services themselves are not important in bringing the possibility of using such services to the attention of radiology practices.

The fact that there was only one discontinuation of use of an external service among the practices not currently using any off-hours services shows that users were generally satisfied. Indeed, the one practice that did discontinue an off-hours service did so not because of a problem with the service but rather because it adopted its own Web-based picture archiving and communication system that its radiologists could use from home.

Finally, the responses from practices not using off-hours services as to what could make them use such a service corroborate our earlier findings that the adequacy of staffing is paramount in this decision. Our results suggest that a change in workload, either through increased imaging volume or through decreasing staffing, can have a powerful effect on the use of an off-hours service.

Practices With Internal Off-Hours Services 

Our results indicate that a consistency of known radiologists is most important in the decision to use an internal off-hours arrangement rather than an external off-hours service. This is a distinct advantage of using an internal off-hours service. Having internal accountability as well as a known quality of interpretations will continue to make internal off-hours services attractive to some practices. In addition, with radiology practices that use internal off-hours services staffing those arrangements with one radiologist for every 14 radiologists in the practices, on average, call in these arrangements is certainly not an obstacle to their adoption.

Our results also suggest that radiology practices that use permanent crews to staff their internal off-hours services do not have difficulty finding incentives for radiologists to cover these shifts. This underscores the fact that the field of radiology lends itself quite well to individuals who prefer off-hours work or who are seeking extra income.

Interestingly, all of the radiology practices using internal off-hours services situated their radiologists locally or within the same states. This is in contrast to external off-hours services, which often elect to locate their off-hours radiologists overseas to take advantage of the time difference.

Finally, our results indicate that almost half of radiology practices that used internal off-hours service were interested in doing more interpretations for other practices as an additional source of income or in starting such activity.

Limitations 

An important limitation of this study is the limited response rate of 38.3%. Because our survey's subject was off-hours services, we would expect that practices not using off-hours services would have a lower than average response rate, and therefore the percentage of practices using such services would be smaller among practices generally than among respondents. In fact, the external off-hours services industry, which has been growing rapidly, reported a 50% coverage rate only as of mid-2007 (William Bradley, MD, PhD, personal communication, July 2007), well after the 2005-2006 date of our survey. Similarly, an as yet unpublished mid-2007 ACR survey of members analyzed by two of us (MB, JHS) also found that half were in practices using external off-hours services as of that date. In contrast, at most 15% of practices were using such services in 2003 [15]. Despite this bias, there is no particular reason to think that there was response bias within the universe of practices not using off-hours services or within the universe of practices using one or the other type of off-hours service, so findings from respondents within each category, the main subject of this study, should be relatively unbiased. Nonetheless, with only 38 responses from radiology practices that did not use off-hours services and 13 from practices that used internal off-hours services, we had small sample sizes, and our estimates are thus necessarily imprecise.

Possibly, there may be some differences in understanding between respondents who replied by telephone and those who replied in writing. We attempted to minimize this by reading answer options to telephone respondents, and we know of no such differences.

Some of the differences between the findings of this study and the findings of Kaye et al [1] may be due to changes in practices' perspectives once they are using off-hours services. For example, Kaye et al presented an explanation for why the use of external off-hours services may be financially advantageous despite the fact that most practices pay the off-hours services approximately as much (or more) than the practices collect for the images the off-hours services interpret. The explanation is nonobvious and thus may mainly be recognized by practices once they are using off-hours services. That could be a reason why practices not using off-hours services express considerable concern about the cost of off-hours services, whereas practices using external off-hours services do not do so.

Finally, we surveyed only radiology practices. Other constituencies, most obviously hospitals, have a role in the decision to adopt an off-hours service.

Back to Article Outline

Conclusions 

As long as there are sufficient numbers of radiologists and residents to handle the volumes of interpretations at some practices, off-hours radiology services will be well known as an option, but not used, by a large number of practices. Off-hours radiology services could help increase their adoption by offering less expensive service with proven quality. Despite the growth of external off-hours services, the financial incentives and quality assurances afforded by internal off-hours services will continue to make them an attractive option.

Back to Article Outline

Acknowledgments 

We thank the radiologists who took the time to respond to our survey.

Back to Article Outline

References 

1. Kaye AH, Forman HP, Kapoor R, Sunshine JH. A survey of radiology practices' use of after-hours radiology services. J Am Coll Radiol 5:748-58.
   • View In Article
2. Wong WS, Roubal I, Jackson DB, Paik WN, Wong VK. Outsourced teleradiology imaging services: an analysis of discordant interpretation in 124,870 cases. J Am Coll Radiol. 2005;2:478–484
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (91 KB)
   • CrossRef
3. DeCorato DR, Kagetsu NJ, Ablow RC. Off-hours interpretation of radiologic images of patients admitted to the emergency department: efficacy of teleradiology. AJR Am J Roentgenol. 1995;165:1293–1296
   • View In Article
   • MEDLINE
4. Kalyanpur A, Neklesa VP, Pham DT, Forman HP, Stein ST, Brink JA. Implementation of an international teleradiology staffing model. Radiology. 2004;232:415–419
   • View In Article
   • MEDLINE
   • CrossRef
5. Kalyanpur A, Weinberg J, Neklesa V, Brink JA, Forman HP. Emergency radiology coverage: technical and clinical feasibility of an international teleradiology model. Emerg Radiol. 2003;10:115–118
   • View In Article
   • MEDLINE
   • CrossRef
6. Harrigal CL, Erly WK. On-call radiology: community standards and current trends. Semin Ultrasound CT MR. 2007;28:85–93
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (1174 KB)
   • CrossRef
7. Larson PA, Janower ML. The nighthawk: bird of paradise or albatross?. J Am Coll Radiol. 2005;2:967–970
   • View In Article
   • Full Text
   • Full-Text PDF (67 KB)
   • CrossRef
8. Robinson J. Re: “The nighthawk: bird of paradise or albatross?”. J Am Coll Radiol. 2006;3:302
   • View In Article
   • Full Text
   • Full-Text PDF (44 KB)
   • CrossRef
9. Brant-Zawadzki M. The goose and the nighthawk: a bedtime fable for young radiologists (with apologies to the Brothers Grimm). J Am Coll Radiol. 2006;3:231–232
   • View In Article
   • Full Text
   • Full-Text PDF (44 KB)
   • CrossRef
10. Wagner AL. After-hours coverage: problems and solutions. J Am Coll Radiol. 2004;1:351–355
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (90 KB)
    • CrossRef
11. Sunshine JH, Meghea C. How could the radiologist shortage have eased?. AJR Am J Roentgenol. 2006;187:1160–1165
    • View In Article
    • CrossRef
12. Ahmed N, Guo A, Elhassan H, Qaiser R, Chung R. Residents managed trauma adequately using their own radiological interpretation as compared with “nighthawk” radiologists. J Trauma. 2006;61:555–557
    • View In Article
    • MEDLINE
13. McLean TR, Richards EP. Teleradiology: a case study of the economic and legal considerations in international trade in telemedicine. Health Aff (Millwood). 2006;25:1378–1385
    • View In Article
    • CrossRef
14. Mun SK, Tohme WG, Platenberg RC, Choi I. Teleradiology and emerging business models. J Telemed Telecare. 2005;11:271–275
    • View In Article
    • MEDLINE
    • CrossRef
15. Ebbert TL, Meghea C, Iturbe S, Forman HP, Bhargavan M, Sunshine JH. The state of teleradiology in 2003 and changes since 1999. AJR Am J Roentgenol. 2007;188:103–112
    • View In Article