Public or Private Magnetic Resonance Imaging: What Do the Patients Think?

Article Outline

• Abstract
• Résumé
• Introduction
• Materials and Methods
• Results
• Discussion
• Appendix 1. Health opinion survey
• References
• Copyright

Abstract 

Purpose

We described the demographic, clinical, and attitudinal profiles of patients awaiting magnetic resonance imaging (MRI) at a private and at a hospital-based facility, and hypothesized that they would not differ significantly.

Methods

A survey of patients attending a hospital facility and a privately owned venue in an Ontario city. Descriptive, bivariate, and logistic regression analyses were performed.

Results

A total of 314 patients provided data, with a higher response rate at the private clinic than at the hospital-based clinic (97% vs 60%). For the majority of patients (58%), the MRI was scheduled to follow up known disease; 55.8% waited more than 4 weeks; 6.4% waited more than 6 months. One-third of patients expressed a willingness to travel to the United States and pay for the MRI, 41% expressed a willingness to pay within Ontario, and 66% were willing to travel elsewhere in Ontario. They were more likely to be at the hospital-based MRI if they were being followed up for known disease and had a diagnosis of cancer, whereas those patients at the private MRI facility reported significantly more pain; 59% of the hospital-based sample and 72% of the private clinic sample reported significantly reduced quality of life because of their health problem.

Discussion

These data provide interesting insights into the characteristics of patients awaiting an MRI and the attitudes of patients towards public and private MRI clinics. There were significant attitudinal differences between those patients attending the 2 facilities. Pain, coupled with a long wait, may create an incentive for patients to conclude that private clinics should be permitted if the hospital environment is unable to improve access times.

Résumé 

Objectif

Nous avons décrit les profils démographique, clinique et attitudinal des patients en attente d’un examen d’imagerie par résonance magnétique (IRM) dans deux milieux, en clinique privée et en milieu hospitalier, en formulant l’hypothèse qu’il n’y avait pas de différence significative entre les patients des deux établissements.

Méthodes

Une enquête auprès de patients fréquentant un établissement hospitalier public ou un établissement privé situés dans une ville ontarienne. Des analyses descriptives, bivariées et de régression logistique ont été réalisées.

Résultats

Au total, 314 patients ont fourni des données. Le taux de réponse était plus élevé à la clinique privée qu’en milieu hospitalier (97 % contre 60 %). Pour la majorité des patients (58 %), l’IRM était prévue dans le cadre du suivi d’une maladie connue; 55,8 % ont attendu plus de quatre semaines; 6,4 % ont attendu plus de six mois. Le tiers des patients ont affirmé être prêts à se rendre aux États‑Unis pour subir l’examen et à en débourser eux-mêmes les frais, 41 % ont affirmé être prêts à payer l’IRM qu’ils subiraient en Ontario et 66 % ont indiqué être prêts à se rendre ailleurs dans la province pour subir l’examen. Les patients faisant l’objet d’un suivi pour une maladie connue et ayant reçu un diagnostic de cancer étaient plus susceptibles de fréquenter la clinique d’IRM en milieu hospitalier. Le degré de douleur signalé par les patients fréquentant la clinique privée était significativement plus élevé; 59 % des patients en milieu hospitalier et 72 % des patients de la clinique privée ont indiqué que leur problème de santé avait entraîné une diminution significative de leur qualité de vie.

Discussion

Ces données offrent une perspective intéressante des caractéristiques des patients en attente d’un examen d’IRM et des attitudes des patients envers les cliniques d’IRM privées et les établissements publiques. Des différences significatives dans les attitudes des patients ont été observées entre les patients des deux milieux. Conjuguée à une longue période d’attente, la douleur peut pousser les patients à conclure que les cliniques privées devraient être autorisées si le milieu hospitalier est incapable d’améliorer l’accès aux services.

Key Words: MRI, Health care, Funding, Policy, Health services research

Back to Article Outline

Introduction 

Over the past 20 years, there has been an exponential growth in the use of new imaging technologies for the diagnosis of disease. Foremost among these technologies is magnetic resonance imaging (MRI), which has become increasingly important in the diagnosis, staging, treatment planning, and follow-up of known disease. MRI is less available in comparison with other diagnostic equipment, such as computed tomography, ultrasonography, or plain radiographs because of higher costs and a shortage of operational staff, which contributes to lengthy wait times and pent-up demands, although precise data are not available [1]. With an ever-increasing volume of MRI studies being performed [2] and a relatively low number of MRI machines per capita compared with other Organization for Economic Co-operation and Development countries (4.5/million vs an average 7.6/million) [3], the question of how our health care system will meet these demands becomes a concern. Indeed, only 13% of Canadian physicians surveyed in 2005 thought that access to advanced diagnostic services was excellent or very good compared with 54% who viewed it as fair or poor [4].

Previous surveys confirm that, whereas Canadians support a publicly funded system, many would be willing to entertain increased private provision of services if this would enhance access [5]. The ruling by the Supreme Court of Canada in the Chaoulli case [6] affirmed the necessity of providing timely access to health care services and may foreshadow increased availability of private care to achieve this goal. However, little research has been published on the role that private clinics could play in our health care system [7] or patient receptivity to such care venues.

This exploratory study focused on patients’ attitudes towards privately operated MRI clinics. We surveyed outpatients awaiting MRI studies to assess their satisfaction with current MRI waiting times and their opinion regarding trends towards privately owned MRI clinics. By comparing responses from patients receiving their study at a public hospital facility to those at a private freestanding facility, we assessed whether there were any significant differences in attitudes and clinical or demographic characteristics of the 2 patient populations and whether any factors were associated with greater acceptance of a private sector system. It is important to note, however, that at neither facility did the patient incur any direct charge. The privately owned clinic is an example of what has been categorized as “private for profit small business delivery.” This category includes entrepreneurs and small businesses that are privately owned and do not answer to shareholders, and includes almost all physician services other than salaried hospital employees [8]. We hypothesized that, in the absence of direct cost to the patient, there should be no attitudinal differences.

Back to Article Outline

Materials and Methods 

The outpatient clinic (KGHMRI) at the public hospital is located within Kingston General Hospital (KGH), at the south-central end of the city. The hospital serves a population of approximately 300,000 people from 3 counties in southeastern Ontario, with a rural population of approximately 45%. The privately owned freestanding clinic (PMRI) is in a mall at a major intersection at the west end of the city. Although it is a private clinic, the Ontario Health Insurance Program covers patient costs, provided that patients have a valid Ontario Health Insurance Program card. The distance between the 2 clinics is 8.7 kilometres.

The Health Opinion Survey was developed on the basis of questions obtained from a review of the literature and input from the department head, the business manager, and other senior colleagues in the department of radiology. After pilot testing, a number of revisions and eventual consensus, it was considered ready for distribution (Appendix 1). Ethics approval was obtained from Queen’s University Research Ethics Board.

Sample size calculations indicated that a minimum of 50 per group would permit valid comparisons of demographics and responses of the 2 groups (alpha 0.05, power 0.80, for a difference of 25% in responses). However, because it was expected that many people would not respond or simply take the survey with them, 200 surveys were placed in the reception or waiting area of each of the 2 clinics. Patients were given the survey to complete voluntarily while they were waiting. A covering letter indicated the purpose of the study, described confidentiality, and provided contact numbers in the event that there were questions or concerns. Patient contact information was not collected. Questions included demographics, perceptions about current state of waiting for an MRI, and perceptions regarding alternatives (Appendix 1). The survey took approximately 5 minutes to complete and was deposited in a locked box upon completion.

Data were entered into an Excel spreadsheet (Microsoft Inc., Redmond, WA) designed for data capture and imported into SPSS (SPSS Inc., Chicago, IL) for analysis. After descriptive analysis, between-group comparisons used 2-sample t tests (continuous data) and χ² analyses (categorical data). Logistic regression was used to identify predictors of clinic use (KGHMRI vs PMRI). Variables were offered into the model on the basis of the strength of the bivariate associations with the outcomes (P < .20).

Back to Article Outline

Results 

A total of 314 patients responded to the survey, with a higher response rate at the PMRI (97%) than the KGHMRI (60%). The sample characteristics for the hospital-based and private MRI samples are contained in Table 1. Statistically significant differences were found between the 2 groups in age category (P = .013), priority code (P = .028), and reason for MRI (P = .001), with a greater likelihood of being at the hospital-based MRI if they were being followed up for known disease. The PMRI group was more likely to report a significant decrease in quality of life (P = .030) and a longer wait (P < .001). Cancer patients were more likely to be at the hospital-based clinic, whereas those with musculoskeletal problems were more likely to be at the private clinic.

Table 1. Sample characteristics

Sample characteristics	Hospital MRI, n (%) (N = 120)∗	Private MRI, n (%) (N = 194)∗	P value†
Age (y)			.013
≤19	10 (8.3)	2 (1.0)
20–39	21 (17.5)	45 (23.3)
40–59	55 (45.8)	96 (49.7)
60–79	31 (25.8)	48 (24.9)
≥80	3 (2.5)	2 (1.0)
Men	54 (45.0)	86 (44.6)	.94
Education			.64
≤High school	64 (53.3)	96 (50.3)
College/university or higher	56 (46.7)	95 (49.7)
Gross family income ($)			.19
<25,000	24 (21.8)	37 (21.3)
25,000–49,999	35 (31.8)	43 (24.7)
50,000–74,999	33 (30.0)	44 (25.3)
75,000–99,999	8 (7.3)	22 (12.6)
≥100,000	10 (9.1)	28 (16.1)
MRI			.001
For new diagnosis	60 (52.2)	62 (33.0)
For follow-up	55 (47.8)	126 (67.0)
Priority code			.028
1 (highest)	2 (2.7)	5 (3.3)
2	38 (34.5)	52 (34.4)
3	38 (34.5)	62 (41.1)
4	13 (11.8)	25 (16.6)
5	18 (16.4)	7 (4.6)
Quality of life			.030
No/minimal drop	47 (40.9)	50 (28.1)
Moderately/significantly worse	68 (59.1)	128 (71.9)
Length of wait			<.001
<7 d	20 (17.4)	7 (3.8)
1 wk–1 mo	43 (37.4)	55 (30.2)
1–6 mo	43 (37.4)	109 (59.9)
>6 mo	9 (7.8)	11 (6.0)
Health problem			<.001
Musculoskeletal	25 (22.7)	91 (51.7)
Cancer	31 (28.2)	11 (6.2)
Trauma/injury	6 (5.5)	14 (8.0)
Neurologic	32 (29.1)	44 (25.0)
Other	16 (14.5)	16 (9.1)

MRI = magnetic resonance imaging.

The means and standard deviations for the ratings of pain, worry, inability to work, caregiver stress, and other sources of stress, as well as the importance of the MRI and the satisfaction with the wait, for the 2 groups are contained in Table 2. The PMRI group rated pain as significantly more important (mean value of 2.3 on a scale of 1-5, with 5 being the least important) than the KGHMRI group (mean rating, 3.4; P < .001), and the patient in the PMRI group also expressed more dissatisfaction with their wait (P = .008). The patients in the KGHMRI group were more likely to report concerns with worry (1.8 vs 2.1, P = .031) and caregiver stress (2.7 vs 3.3, P = .012).

Table 2. Aspects of the MRI wait

Statistically significant P values are in bold type.

MRI = magnetic resonance imaging.

The perceptions regarding alternatives are shown in Figure 1. The PMRI group was far more likely to indicate that private health clinics are in keeping with the Health Act, that they would alleviate pressure, that they would be willing to pay for a private clinic, would travel elsewhere in Ontario, would travel to the United States to pay for this service, and believe that hospitals should be able to charge for MRIs that are sooner (P < .01 for all). The KGHMRI group was more likely to believe that private MRI alternatives would create a 2-tiered system, although it fell short of statistical significance (P = .158).

• 
  • View Large Image
  • Download to PowerPoint

• Figure 1. 

  Perceptions regarding magnetic resonance imaging alternatives.

Group differences with a significance level of .20 or less were offered into a multivariable logistic regression model to identify the subset of factors associated with being in the KGHMRI group vs the PMRI group. Six variables were identified, with an overall model χ² of 81.7 (P < .001) and accounting for 29.0% of the variation in outcome (Cox & Snell R-square). The most significant predictor was the importance attributed to pain, with an odds ratio (OR) that suggested that those patients who gave it the highest rating were 4.6 times more likely to go to the private clinic (P < .001), whereas those who rated it as the second most important stressor were 4.3 times more likely to go the private clinic compared with those who did not give pain as high a rating. A wait of between 1 and 6 months was also associated with a 7.4-fold increased odds of going to a private clinic (P = .001). The patients in the PMRI group were twice as likely to indicate that they thought that hospitals should allow people to pay for sooner MRIs, and were also 2.6 times more likely to indicate that they would be willing to travel elsewhere in Ontario for the service. However, those patients who reported cancer as their health problem were 3.3 times more likely to be at the hospital-based clinic (obtained by inverting the ORs in Table 3, so 1/0.3 = 3.3). Finally, those patients being seen for a new diagnosis were 1.7 times more likely to be seen at the private MRI. Although this fell short of significance (P = .109), it was considered clinically relevant and, therefore, was retained in the model.

Table 3. Logistic regression model for public vs private MRI (KGH = 0, private = 1)^∗

Variable	Coefficient	Significance	Odds ratio (95% confidence interval)
Constant	–2.9	<.001	n/a
Length of wait (reference <7 d)
1–4 wk	1.2	.056	3.4 (1.0–11.8)
1–6 mo	2.0	.001	7.4 (2.2–24.9)
>6 mo	1.0	.211	2.8 (0.6–13.5)
Hospitals should allow people to pay for sooner MRI (0 = no, 1 = yes)	0.8	.032	2.2 (1.1–4.4)
I would be willing to travel elsewhere in Ontario (0 = no, 1 = yes)	1.0	.006	2.6 (1.3–5.2)
Rating of importance of pain (reference was 3rd through 5th combined)
Second most important stressor	1.5	.011	4.3 (1.4–13.4)
Most important source of stress	1.5	<.001	4.6 (2.3–9.4)
Cancer (0 = no, 1 = yes)	-1.1	.029	0.3 (0.1–0.9)
Reason (0 = follow-up, 1 = new diagnosis)	0.5	.109	1.7 (0.9–3.3)

KGH = Kingston General Hospital; MRI = magnetic resonance imaging; n/a = not applicable.

Back to Article Outline

Discussion 

These data provide important preliminary insights into the attitudes of patients towards public vs private MRI clinics. Regardless of which clinic they visited, 33.8% indicated that they would travel to the United States and pay for a shorter wait, 40.6% indicated a willingness to pay for their MRI, and 65.5% were willing to travel elsewhere in Ontario for an MRI at a private clinic. These findings did not differ across health problems.

Those patients who attended the privately owned MRI facility differed from those at the hospital site in several respects. The patients in the PMRI group had more musculoskeletal cases and reported more concerns with pain, whereas the patients in the KGHMRI group had more cancer cases and had more concerns about worry and caregiver stress. This may reflect referral patterns, because cancer care is located at the KGH. Surprisingly, waiting times at the PMRI were longer, although this may have been because patients had been waiting at the KGHMRI and switched to the PMRI. These data were not collected, so this cannot be confirmed. Patients in the KGHMRI group were more satisfied with the length of their waits, whereas patients in the PMRI group were somewhat more affluent and more willing to pay or travel to receive service.

Recent data from the Ministry of Health and Long-term Care suggest that the median wait time for MRI is 39 days, with the 90th percentile at 104 days [9]. Waiting times within the 2 groups were consistent with this estimate and were quite long in both the KGHMRI and PMRI groups, with 37% of the KGHMRI group waiting between 1 and 6 months, and an additional 8% waiting more than 6 months. The equivalent numbers for the PMRI group were 60% (1 - 6 months) and 6% (> 6 months), which suggests that there is a significant wait regardless of the facility chosen.

The private facility differs from KGH primarily in location, and one, therefore, would expect that there should not be any real difference in the attitudes of the patients who go to the 2 facilities for a new diagnosis or for follow-up, even if their clinical profiles differ. Instead, the choice of which facility would depend on family physician referral habits, provider beliefs about which facility is more expeditious in the context of perceived clinical urgency, and patient preferences, none of which should be associated with different attitudes about health care delivery.

Our regression analyses, however, did suggest that there were both clinical and attitudinal differences between the patients served at the 2 facilities. How might this be explained? Pain, coupled with a long wait, may create an incentive for the patients, independent of income or age, to reflect on the health care system and to conclude that, to achieve timely access for patients with their type of need, hospitals should consider other options, such as payment for services, and private clinics should be permitted as well.

There were several limitations of the study, which may affect generalizability. First, the use of a passive sample of patients awaiting their MRI may have resulted in a self-selection bias, in that those patients who responded may have been different from those who did not. Moreover, Ontario is in a somewhat unique situation in that private MRI facilities receive public funding for routine health services, which is not the case in all provinces, so the attitudinal differences may not be generalizable to other Canadian settings. Finally, patients are likely triaged according to perceived acuity and, as mentioned above, referral patterns of the individual physicians would also have an effect.

In addition, there are a number of questions that could not be addressed with these data, even though they would have been of interest. The primary health problem was self-reported and often vague, which made it difficult to assess the severity of the problem or the accuracy of the reporting. For example, were these patients being seen to rule out the problem that had been identified or to see if treatment of that problem had been effective? With this additional information, it would have been possible to test questions such as whether those who were willing to pay were those patients with serious illness or significant pain or those who were well but worried about their health. We also were unable to determine whether the waiting times at 1site included waiting times from another, for example if the wait at 1 location was too long, and the patient and/or the referring physician then switched to the other facility.

However, because we found an attitudinal difference between patients at a public facility and those attending a “private for profit small business delivery” venue, despite their basic similarity of operation, lends some credence to the slippery slope argument that “creeping privatization” may erode Medicare [10]. That is, based on dissatisfaction with their waiting experience in the public system, patients may first come to support privately owned facilities and then move to support for private payment. Further research is needed to establish not what the general public thinks about private care venues in theory but rather the perspectives of that subset of the public that is actually faced with making a choice about the source of care.

Back to Article Outline

Appendix 1. Health opinion survey 

• 
  • View Large Image
  • Download to PowerPoint
• 
  • View Large Image
  • Download to PowerPoint

Back to Article Outline

References 

1. Tu JV, Pinfold SP, McColgan P, et al. Access to health services in Ontario (Chapter 6: CT and MRI scanning). Toronto: Institute for Clinical Evaluative Sciences (ICES); 2005. Available at: http://www.ices.on.ca/webpage.cfm?site_id=1&org_id=67&morg_id=0&gsec_id=0&item_id=2862&type=atlas. Accessed April 21, 2010.
   • View In Article
2. Canadian Institute for Health Information (CIHI). Medical imaging in Canada. Ottawa: CIHI; 2003;Available at: http://secure.cihi.ca/imaging/index.htmlAccessed June 20, 2007
   • View In Article
3. Organization for Economic Co-operation and Development (OECD) 2005. Health data 2005: How does Canada compare? Available at: http://www.oecd.org/document/24/0,2340,en_2825_497118_2671576_1_1_1_1,00.html#selection. Accessed April 21, 2010.
   • View In Article
4. Canadian Medical Association. MD Pulse 2005: the National Physician Survey and the future of medicine in Canada. Ottawa: CMA; 2005;Available at: http://www.cma.ca/index.cfm/ci_id/43053/la_id/1.htmAccessed April 21, 2010
   • View In Article
5. Vail S. Canadians’ values and attitudes on Canada’s health care system, a synthesis of survey results. Available at: Ottawa: Conference Board of Canada; 2000;http://www.conferenceboard.ca/documents.asp?rnext=215Accessed April 21, 2010
   • View In Article
6. Skolrood RA. Chaoulli v. Quebec (Attorney General) The Supreme Court of Canada sets the stage for fundamental health care reform. Vancouver: Lawson Lundell LLP; 2005;Available at http://www.lawsonlundell.com/Resources/News-and-Publications/Chaoulli-v.-Quebec-Attorney-General-The-Supreme-Court-of-Canada-Sets-the-Stage-for-Fundamental-Health-Care-ReformAccessed October 5, 2010.
   • View In Article
7. Sanmartin C, Shortt S, Barer M, et al. Waiting for medical services in Canada: lots of heat, but little light. CMAJ. 2000;162:1305–1310
   • View In Article
   • MEDLINE
8. Deber RB. Delivering health care services: public, not-for-profit or private? Commission on the Future of Health Care in Canada, Discussion Paper No. 17. Ottawa: Health Canada; 2002;Available at: http://www.hc-sc.gc.ca/english/care/romanow/hcc0381.htmlAccessed April 21, 2010
   • View In Article
9. Ministry of Health and long-Term Care. Wait times in Ontario 2006, Available at: http://www.health.gov.on.ca/transformation/wait_times/wt_data/data_ontario.html. Accessed April 21, 2010.
   • View In Article
10. Chodos H, MacLeod J. Romanow and Kirby on the public/private divide in healthcare: demystifying the debate. Healthc Pap. 2004;4:10–25
    • View In Article