Begin main content
CADTH Document Viewer

The Canadian Medical Imaging Inventory, 2017

March 2018


EXECUTIVE SUMMARY

In 2015, CADTH assumed the task of conducting a biennial survey of medical imaging providers in Canada, following the Canadian Institute for Health Information’s (CIHI’s) data collection until 2012. This report summarizes the results of the second iteration of the CADTH Canadian Medical Imaging Inventory.


Methods

We collected information on six modalities using a Web-based survey, supplemented with information from suppliers of diagnostic imaging equipment, data validators, report reviewers, and literature searches. The modalities were:

  • computed tomography (CT)
  • magnetic resonance imaging (MRI)
  • single-photon emission computed tomography (SPECT)
  • positron emission tomography–computed tomography (PET-CT)
  • positron emission tomography–magnetic resonance imaging (PET-MRI)
  • single-photon emission computed tomography–computed tomography (SPECT-CT).

The survey was launched on April 18, 2017, and data collection formally closed on June 9, 2017. Previous data were entered into a database with a Web interface and respondents were invited to amend the data associated with their sites — or, if there was no pre-existing entry, to complete the survey. High-level data were reviewed by pre-identified regional and provincial validators, who supplied corrections and supplementary data, which were entered into the database. Identified stakeholders and survey respondents were given the opportunity to review the report during the stakeholder feedback process.


Results

A submitted survey was received from 147 sites, and an additional 223 surveys were accessed and modified but not submitted as final. At least minimal data (modalities and unit counts) were available for a total of 505 sites. The majority of sites were publicly funded facilities. Sixty-nine per cent were urban, 29% were rural, and 2% were remote.


Computed Tomography

  • We identified a total of 561 CT units in Canada, up from 419 units in 2007. All provinces or territories had at least one unit. Ontario and Quebec had the most units, followed by British Columbia and Alberta. The three northern territories had one unit each.
  • From examination data reported by validators, we estimated that a total of 5.61 million CT examinations were performed in the most recent fiscal year in Canada, up from 3.38 million in 2007. This is equivalent to 153.0 exams per 1,000 people, up from 103.3 in 2007.
  • CT units operate for an average of 76.5 hours per week and 12.1 hours per day. Most operate on weekends. CT is used across disciplines, with approximately a quarter of the time used in the fields of oncology, followed by neurology, respiratory, and hepatobiliary.
  • When the number of CT units per population for Canada is compared with other countries that report CT units to the Organisation for Economic Co-operation and Development (OECD), Canada appears in the lower half of the reported numbers. For the number of exams per population, Canada appears around the midpoint.
  • Half the units reported on in the survey had 64 cross sectional images or slices, with about an eighth having 128 slices. Three-quarters incorporate image reconstruction techniques for dose reduction, 81% are equipped with dose management controls, and 92% record dose by exams.
  • One-third of CT units were five-years-old or less, 40% were six- to ten-years-old, and 23% were 11- to 15-years-old. None were older than 20 years.


Magnetic Resonance Imaging

  • We identified a total of 366 MRI units in Canada, up from 222 units in 2007. All provinces and one territory had at least one unit. Ontario and Quebec had the most units, followed by British Columbia and Alberta. Yukon has a single unit.
  • From examination data reported by validators , we estimated that a total of 1.86 million MRI examinations were performed in the most recent fiscal year in Canada, up from around 1 million in 2007. This is equivalent to 51.0 exams per 1,000 people, up from 31.2 in 2007.
  • MRI units operated for an average of 78.7 hours per week and 13.1 hours per day. Most also operated on weekends. MRI is used across disciplines, with approximately a quarter of the time used for musculoskeletal exams, followed by neurological, oncological and hepatobiliary exams.
  • When the number of MRI units per population for Canada is compared with other countries that report MRI units to the OECD, Canada appears in the lower half of the reported numbers. For the number of exams per population, Canada appears around the midpoint. This is consistent with the positions for 2015.
  • The majority (83%) of MRIs have field strengths of 1.5 Tesla.
  • Thirty-six per cent of MRI units were five-years-old or less, 33% were six- to ten-years-old, and 26% were aged 11- to 15-year-old. None were older than 20 years.


Positron Emission Tomography–Computed Tomography or Positron Emission Tomography

  • We identified a total of 51 PET-CT units in Canada, up from 21 units in 2007. All were PET-CT, indicating that PET had been replaced as a modality. Nine provinces had at least one unit, up from eight in 2007, with Ontario and Quebec having the most. Newfoundland and Labrador installed a unit between 2015 and 2017.
  • From examination data reported by validators, we estimated that a total of 90,530 PET-CT examinations were performed in the most recent fiscal year in Canada. Exam data for 2007 was not available.
  • PET-CT units operate for an average of 40.4 hours per week and 8.6 hours per day. PET-CT is primarily used for oncology (80%), followed by cardiac and neurological use.
  • Two-thirds of units have sixteen slices. The majority of units (90%) were equipped with dose management controls and 77% recorded patient radiation dose by exam.
  • Thirty-two per cent of PET-CT units are five-years-old or less, 53% are six- to ten-years-old, and 16% are between 11- and 15-years-old.
  • A quarter of sites had access to a cyclotron for generating radioisotopes. Two-thirds of the sites without access to a cyclotron obtained radioisotopes from commercial sources.


Positron Emission Tomography–Magnetic Resonance Imaging

  • The hybrid modality of PET-MRI is the newest specialist imaging modality. We identified three units currently operating for clinical research purposes in Ontario.
  • As the PET-MRI has yet to enter clinical use, we do not have any examination or use data.


Single-Photon Emission Computed Tomography

  • We identified a total of 330 SPECT units in Canada, down from 603 units in 2007, although this figure is approximate because Ontario reported a combined figure for SPECT and SPECT-CT. Nine provinces had at least one unit. Ontario and Quebec had the most units, followed by Alberta and British Columbia.
  • Individual data for SPECT exams were not available for all provinces, so the combined exams are reported under SPECT-CT. Exam data for 2007 was not available.
  • SPECT units operate for an average of 43.5 hours per week and nine hours per day. A minority operate on weekends. SPECT is primarily used for cardiac examinations (40%), followed by oncology and musculoskeletal exams.
  • Three-quarters of units have two detector heads, and one-quarter are dedicated cardiac units.
  • Overall, Canada has some of the oldest SPECT units. Thirteen per cent of these units are less than five-years-old, 29% are between six- and 10-years-old, 36% are 11- to 15-years-old, and 21% are more than 15-years-old.


Single-Photon Emission Computed Tomography–Computed Tomography

  • We identified a total of 261 SPECT-CT units in Canada, up from five units in 2007, although the number is approximate because Ontario reported a combined figure for SPECT and SPECT-CT. All 10 provinces had at least one unit. Ontario and Quebec had the most units, followed by British Columbia and Alberta.
  • From data reported by validators, with a small number of imputed exams, a total of 1.35 million SPECT or SPECT-CT exams were carried out in Canada.
  • SPECT-CT units operate for an average of 45.2 hours per week and nine hours per day. A minority operate on weekends. SPECT-CT is primarily used for cardiac examinations (36%), followed by oncology and musculoskeletal exams.
  • Almost all units have two detector heads. One-third have four slices, and 20% have one and 16 slices.
  • Two-thirds were equipped with dose management controls and 61% recorded patient radiation dose by exam. More than half incorporated image reconstruction techniques for dose reduction.
  • Forty per cent of SPECT-CTs are less than five-years-old, half (48%) are between six- and 10-years-old, and 12% are 11- to 15-years-old.


Picture Archiving and Communications System

  • One-third (28%) of sites had access to a local or institutional picture archiving communications system (PACS) network, 39% had access to a regional network, and one-third (33%) had access to a provincial network.
  • Almost all sites allowed access to PACS images outside the imaging department, and two-thirdsallowed access to other sites within the provincial health care system.
  • Almost all sites with CT, MRI, SPECT, PET-CT, or SPECT-CT stored and accessed images for these modalities on PACS. A minority of sites without each of these modalities could also access images. We do not have information about PET-MRI.


Limitations

  • For feasibility, this iteration of the survey was restricted to six specialist imaging modalities and does not include others that are more common and widespread (e.g., X-ray and ultrasound).
  • As we do not have a definitive list of facilities containing the equipment, and, as the survey was voluntary, we cannot ensure that all facilities or departments containing the modalities were contacted or responded.
  • A limited amount of imputation was used to carry forward data from previous years. Data for hours and types of use was available for only a limited number of sites.


 ABBREVIATIONS

CADTH Canadian Agency for Drugs and Technologies in Health
CIHI Canadian Institute for Health Information
CMII Canadian Medical Imaging Inventory
CT computed tomography
MRI magnetic resonance imaging
OECD Organisation for Economic Co-operation and Development
PACS picture archiving and communication system
PET positron emission tomography
PET-CT positron emission tomography–computed tomography
PET-MRI positron emission tomography–magnetic resonance imaging
SPECT single-photon emission computed tomography
SPECT-CT single-photon emission computed tomography- computed tomography
 


INTRODUCTION

In Canada, medical imaging is a vital service within our health care system, providing the basis for diagnosis, staging, and monitoring in a variety of diseases and conditions. Computed tomography (CT), magnetic resonance imaging (MRI), and nuclear medical imaging, including single-photon emission computed tomography (SPECT) and positron emission tomography (PET), have become commonplace in medical imaging and nuclear medicine departments across Canada. In recent years, hybrid technologies such as single-photon emission computed tomography– computed tomography (SPECT-CT), positron emission tomography–computed tomography (PET-CT), and positron emission tomography–magnetic resonance imaging (PET-MRI) have further expanded the imaging repertoire. Other recent innovations include techniques that provide higher quality imaging, lower radiation doses, and faster examinations.1,2

As imaging modalities advance, decision-makers and clinicians face complex choices about which medical imaging technologies to acquire and use. Each modality offers unique characteristics, advantages, and disadvantages. However, decisions about adoption and implementation are made within the context of a finite health care budget and limited availability of clinical and technical expertise. In addition, the need for appropriate use of imaging studies has been highlighted by Choosing Wisely Canada3,4 and Canada Safe Imaging.5 From the patient’s perspective, overuse is associated with unnecessary exposure to radiation, inconvenience due to travel and appointments, increased wait times for necessary procedures, and risks of further investigation and treatment arising from false-positive or incidental findings.6 From the health care system’s perspective, overuse is associated with increased wait times for necessary procedures, and increased costs.7

Given these considerations, current information regarding the status of medical imaging equipment in Canada is critical. In 2001, CADTH (then the Canadian Coordinating Office for Health Technology Assessment) conducted its first inventory of diagnostic imaging equipment in Canada. From 2003 to 2012, the Canadian Institute for Health Information (CIHI) continued to collect data on the inventory and use of diagnostic imaging equipment.8-10 In 2015, CADTH resumed work on the inventory to meet the ongoing need, producing its first report in 2016.11 This updated report summarizes the findings of the 2017 iteration of the inventory.

For this iteration, data were collected on six modalities: CT, MRI, SPECT, PET-CT (which has almost entirely superseded PET), PET-MRI, and SPECT-CT. (Appendix A describes the modalities and their use.) These six were given priority over several modalities captured in previous iterations of the survey (angiography units, cardiac catheterization units, bone densitometers, and lithotripsy imaging),8-10 and over the widely used modalities of X-ray and ultrasound imaging. We limited the scope to ensure feasibility and were guided by stakeholder prioritization of these six modalities; however, the scope will be re-assessed and expanded in future iterations if feasible. Data were also captured on infrastructure requirements of current concern: use of and access to a picture archiving and communication system (PACS) for each of the six modalities, and the source of isotopes for PET and its hybrid modalities.


OBJECTIVES

The purpose of this pan-Canadian inventory is to document current practices and developments related to the supply, distribution, technical operation, and general clinical use of selected medical imaging modalities at public and private Canadian health care facilities. The specific overall objectives of the project are:

  1. to determine the number of units (medical imaging devices) of selected medical imaging modalities in Canada
  2. to provide accurate and timely data on the supply, distribution, and use of selected medical imaging units in Canada
  3. to improve current understanding of the technical characteristics of medical imaging equipment in Canada
  4. to report on trends and developments in medical imaging equipment use across Canada
  5. to inform medical imaging–related strategic planning on a national, provincial, or territorial basis.


METHODS


Identification of Respondents and Data Collection


Data Sources

Data were primarily collected via a Web-based survey on the CADTH website. (The English language survey is shown in Appendix B.) Both English and French versions were offered. The 2017 survey was based on the 2015 iteration of the survey, with the following changes:

  • Questions Removed
    • breakdown of use between diagnostic, interventional, and other. 2015 responses to the question indicated almost exclusively diagnostic use.
  • Questions Added
    • breakdown of use according to body system for the most relevant body systems for CT, MRI, PET-CT, and SPECT-CT, based on literature searches and expert input
    • whether a site’s ordering system included a screen for appropriate ordering.

Survey respondents who had participated in previous iterations of the inventory were presented with pre-populated forms for updating and completion. Respondents for new sites were presented with blank forms. Data for pre-population were obtained from the following sources:

  • site-level data from the Canadian Medical Imaging Inventory (CMII) 2015 survey: These data included responses from site survey responses, supplemented by data validators, and a supplementary grey literature search conducted for the 2015 report
  • unit-level (technical) data from the CMII 2015 survey, for units installed between 2012 and 2015
  • unit-level data from the CIHI 2012 data set, for units installed before 2012 (site-level data were not carried forward)
  • unit-level availability data provided by three major suppliers of diagnostic imaging equipment (Toshiba, Siemens, GE Healthcare).

The CMII 2015 final data set was restructured and merged with the new data. Differences in site names, unit technical specifications, and dates of installation were reconciled manually across data sources. Site names and first year of operation of imaging equipment, in particular, were inconsistent across sources. If the data source offered a year of installation, but indicated that the first year of operation was not the first year of operation, then the first year of operation was assumed to be the year after installation.

Site-level data consisted of unit availability and counts by modality, and the following measures by modality: average hours of use per day and per week, 24-hour and weekend use, total number of exams in the last fiscal year across all units (some sites had only the last calendar year available, and reported that), and breakdowns of types of use into categories. Site-level data also included the type of facility, the use of PACS, and the source of isotopes (for PET or hybrid PET modalities). Unit-level information consisted of manufacturer, model, year of installation, and modality-specific technical characteristics, such as the number of detectors/slices and availability of dose-management controls for CT, and the field strength for MRI.

Respondents to the 2017 survey were asked to update the available data to reflect the status at the time of survey response. In particular, survey respondents were asked to identify units that had been decommissioned, regardless of the year, and to provide the year of decommissioning. If survey responders did not update the survey, we assumed that there had been no changes from the 2015 survey, and we specifically mentioned this assumption when we sent out the survey.

Data from validators for unit counts and examinations in the last fiscal year were preferentially used over survey data to calculate unit counts and counts per population, exams, and exams per population. If data from validators were incomplete, data from the survey were used. For 2017, validated examination data were available for most jurisdictions, and imputation was used only for a minority of jurisdictions for SPECT and SPECT-CT, in comparison with 2015, when all examination counts for all modalities were estimated using imputation. Site-level unit counts supplied by validators were used to identify surplus units in the database (duplicates or decommissioned units that had not been identified as such, based on the assumption that the oldest unit or units were the ones decommissioned).


Identification of Potential Respondents

Most respondents were pre-identified using a database of previous participants. These contacts were updated to account for changes due to position turnover, retirement, and restructuring. Potential respondents included individuals working in private or public health care settings that operate medical imaging equipment. Occupations included executive (e.g., president of a private facility, hospital administrator) and leadership positions (e.g., chief technologist, manager or director of diagnostic imaging, site coordinator). Some respondents were identified through CADTH liaison officers, external stakeholders, and participant referrals.

Passive methods of recruitment included promotion of the renewed survey on the CADTH website and social media, including Twitter and LinkedIn. It is unclear whether any participants were identified through these channels.

To access the survey, all participants were asked to register a profile on the CADTH website to ensure that their response was linked to a unique registration profile. Registrants were matched to sites if there was a pre-existing record, or, if there was no pre-existing record, registrants were presented with a blank form to create a new record.


Duration of Survey

The survey opened on April 18, 2017, and data collection closed on June 9, 2017.


Validation

Upon survey closure on June 9, 2017, we sent summary statistics of the number of units per modality in each jurisdiction to pre-identified validators. Depending upon the jurisdiction, each validator reviewed data for an entire province or for one or more health regions within a province. Validators assessed the summaries for accuracy and provided corrections and information (unit counts or examinations, either at a site or jurisdictional level) on non-responders. Validators were also asked to encourage non-responders in their regions to participate in the survey by the extended deadline (September 11, 2017).

Validators also provided unit counts for their jurisdiction (province or region) and examination data for a subset of jurisdictions.

Identified stakeholders and survey respondents were given the opportunity to review a draft report during a stakeholder feedback process. The report underwent two rounds of internal review and a formal peer-review process before publication.


Data Analysis

Additional data sources are presented in Table 1, and use of the data sets in the analysis in Table 2.


Table 1: Additional Data Sources
Data Source Application of Data
Industry data List of installed equipment and upgrades from General Electric, Siemens, and Toshiba
Canadian population data Population data for Canada and the provinces from Statistics Canada
International comparison data International comparison data for the number of computed tomography and magnetic resonance imaging units and computed tomography and magnetic resonance imaging exams from the website of the Organisation for Economic Co-operation and Development


Table 2: Use of Data Sets in Analyses
                                                                                                                        Data Sources
Data Summaries CMII 2017 Survey CMII 2015 Surveya,b CIHI 2012 Datac Validation Data Industry Data Statistics Canada International Datad
Summaries of site characteristics          
Summaries of modality availability, number of units      
Summary of planned installations and planned decommissioning            
Summary of units at sites that had responses to the 2015 survey but no responses or validation data for the CMII 2017 surveya            
Maps of machine locations      
Summaries of exams in one fiscal year        
Summaries of average hours per week and hours per day of operation, summaries of proportions of types of use          
Summaries of units per site and units per population    
Comparisons of inventory with international availability for CT and MRI    
Age of units, current and decommissioned      
Technical specifications of current units      

CMII = Canadian Medical Imaging Inventory; CT = computed tomography; MRI = magnetic resonance imaging.

aSites that had a response to the 2015 survey, but not the 2017 survey, identified by the lack of receipt of a submission form for the 2017 survey, or of correspondence indicating no change or describing changes.
b These data included responses from site survey responses, supplemented by data validators, and a supplementary grey literature search conducted for the 2015 report.
c From the data set originally supplied to CADTH by the Canadian Institute for Health Information, comprising data collected between 2003 and 2012, as described in the 2015 CMII report.11
d International data from the Organisation for Economic Co-operation and Development (OECD).12-15


Data Summaries

We present the data using descriptive summaries and graphs of site- and province-level findings. We use counts for discrete data, such as the number of sites with a given modality or the number of units at a site. Continuous values are presented either as summary statistics such as mean (average), or range between minimum and maximum values, or as assigned categories (e.g., hours of use per day as less than eight hours, eight to less than 12 hours, 12 to less than 18 hours, and 18 hours or more). Where we asked respondents to choose between two or more responses (e.g., Yes/No), we report the counts and/or percentages of respondents who selected each response.

Stacked bar charts were used to display number of units and hours and percentage of use as categories. The geographical distribution of modalities was presented using geocoded data presented on maps.

The survey form also included a field to invite respondents to give additional detail from which we extracted information regarding decommissioning activities and sharing of mobile units.


Missing Data and Imputation


Handling of Data From Sites Without Updated Data in 2017

If the 2017 survey was not updated for a site, we assumed there was no change from 2015 data. We carried forward data from previous surveys according to the following rules:

  • Data for site characteristics, PACS use, unit and modality availability, counts, age, technical information, and use collected during the CMII 2015 survey were carried forward unchanged.
  • Data for unit technical specifications collected up to 2012 were used in data summaries for age and technical information, provided the unit had not been identified as surplus to validated counts.
  • Data for site characteristics, PACS use, unit and modality availability and counts, and modality use collected up to 2012 were not incorporated into data summaries unless they were confirmed in 2015 or 2017 (e.g., for unit counts, by comparison with validators’ data).


Imputation of Missing Data

We imputed data for a limited number of missing values. In particular, if the questions regarding planned installations or decommissioning were left blank, we assumed the answer was “no.” If the completed use categories added up to 100%, then any missing values were assumed to be 0%. Out-of-range values for the number of hours of operation per week (>168 hours) or per day (>24 hours) were set to “missing.”


Imputation of Missing Examination Data

By preference, examination data supplied by the validators was reported. If we did not have validators’ data for a given province, then data from the survey were used. When sites with available unit counts were missing data for the total number of examinations for 2017, we imputed the missing data, according to the following rules:

  • Within each province or territory, we calculated the mean number of exams per unit for sites that reported examination data, and used this mean to impute the total number of exams for the remaining units. The total number of exams for each province or territory was the sum of the exams reported and exams imputed.
  • The national total for Canada was the overall sum of reported and imputed examinations for all provinces and territories.
  • Provinces or territories that had no examination data did not contribute to the Canadian total, even if they had units.

 

 


RESULTS


Response Rate for the 2017 Update

A survey response was received (survey was submitted as final) for 147 sites, indicating either that available data had been reviewed and updated for 2017, or that new data had been entered. An additional 223 surveys were accessed and modified by respondents after the opening of the survey on April 18, 2017, but had not been submitted as final by the survey’s close on June 9, 2017. Provincial and territorial validators provided information for non-responding publicly funded health facilities. At least minimal data (modalities and unit counts) were available for a total of 505 sites.

By comparison, 222 completed initial surveys were received for the 2015 CMII (respondents had to enter all data), which were supplemented with data from the 2012 CIHI survey, validators, and the results of a CADTH survey on the use of PET and PET-CT in Canada. At least minimal data (modalities and unit counts) were available for a total of 460 sites.


Characteristics of Facilities Responding to the 2017 Update

Of the 324 sites (of a total of 505 sites) with known facility type, most were identified as hospitals, accounting for 209 (64.5%) of reported sites, whereas 53 (16.4%) were community hospitals, 40 (12.3%) were free-standing facilities, and 22 (6.8%) were tertiary care centres (Appendix C, Table 18 shows summaries by province or territory). A facility was defined as a single hospital or hospital campus site that was part of an amalgamation of hospitals. The definition of free-standing facility was broad and captured privately funded and publicly funded sites, as well as sites that received both types of funding. See Appendix B for the detailed definitions of facility type.

Of the 223 sites that provided setting information, most self-reported that they were urban (154 or 69.1%), while 64 (28.7%) were rural, and 5 (2.2%) were remote (Appendix C, Table 19 shows summaries by province or territory).

Of the 329 sites that provided funding information, most were publicly funded (292 or 88.8%), while 27 (8.2%) were privately funded, and 10 (3%) received both public and private funding (Appendix C, Table 20 shows summaries by province or territory).


Overall Inventory of Medical Imaging Equipment in Canada in 2017

In this section, we briefly describe the overall reported inventory of units and use for the six imaging modalities of interest across all provinces and territories. Subsequent sections present in greater detail the inventory and use for CT, MRI, PET-CT, PET-MRI, SPECT, and SPECT-CT.


Total Unit Counts

Figure 1 and Table 3 show the overall provincial and territorial inventory of all six modalities.

Table 3 shows the total number of units per modality for each jurisdiction, as well as the number of sites with that modality available, based on survey responses, validators’ and stakeholders’ counts, and additional units identified in the CIHI survey as free-standing facilities.


Figure 1: Overall Provincial or Territorial Inventory of CT, MRI, PET-CT, PET-MRI, SPECT, and SPECT-CT in 2017

CT = computed tomography; MRI = magnetic resonance imaging; PET = positron emission tomography; PET-CT = positron emission tomography–computed tomography; PET-MRI = positron emission tomography–magnetic resonance imaging; SPECT = single-photon emission computed tomography; SPECT-CT = single-photon emission computed tomography– computed tomography.


Table 3: Overall Provincial or Territorial Inventory and Availability of CT, MRI, PET-CT, PET-MRI, SPECT, and SPECT-CT in 2017
  Number of Unitsa (Number of Sites With Units)b
Province / Territory CT MRI PET-CT PET-MRI SPECT SPECT-CT
Alberta 56 (41) 41 (29) 4 (3) 0 (0) 42 (29) 32 (19)
British Columbia 66d (47) 46 (42)d 3d (2) 0 (0) 28 (16) 31 (18)
Manitoba 23 (16) 12 (7) 1 (1) 0 (0) 9 (6) 8 (5)
New Brunswick 15 (11) 11 (9) 2 (2) 0 (0) 5 (3) 5 (5)
Newfoundland and Labrador 16 (14) 5 (5) 1 (1) 0 (0) 2 (2) 9 (4)
Northwest Territories 1 (1) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
Nova Scotia 18 (14) 12 (11) 1 (1) 0 (0) 7 (7) 10 (8)
Nunavut 1 (1) 0 (0) 0 (0) 0 (0) 0 (0) 0 (0)
Ontario 184 (105) 120 (74) 17d (14) 3 (3) 151 (74) 78 (48)
Prince Edward Island 2 (2) 1 (1) 0 (0) 0 (0) 0 (0) 2 (1)
Quebec 163d (97) 107d (75) 21d (20) 0 (0) 77c (39) 76c (42)
Saskatchewan 15 (13) 10 (7) 1 (1) 0 (0) 9 (4) 10 (5)
Yukon 1 (1) 1 (1) 0 (0) 0 (0) 0 (0) 0 (0)
Canada 561 (363) 366 (261) 51 (45) 3 (3) 330 (181) 261 (155)

CT = computed tomography; MRI = magnetic resonance imaging;PET = positron emission tomography; PET-CT = positron emission tomography–computed tomography; PET-MRI = positron emission tomography–magnetic resonance imaging; SPECT = single-photon emission computed tomography; SPECT-CT = single-photon emission computed tomography– computed tomography.
aPer-province unit counts according to the validator.
b Per-province unit availability according to the validator if the validator provided lists of sites with availability; otherwise, according to the survey.
c Validator-supplied combined unit counts for SPECT and SPECT-CT. Distribution was assumed to be 50:50, based on the division in Quebec and overall.
d Provincial totals included private units.


Growth in Inventory Over 2015–2017

There was an overall increase in the number of units for all modalities between 2015 and 2017 (see Table 28, Table 29, Table 30, and Table 31). The increase may represent a net increase in installation, but may also represent more complete data collection, due to the addition of industry and validator counts. This particularly applies to SPECT and SPECT-CT.


Number of Units by Provincial or Territorial Population

Table 4 shows the number of units reported per million people for all the provinces or territories and for Canada as a whole.


Table 4: CT, MRI, PET-CT, PET-MRI, SPECT, and SPECT-CT Units by Provincial or Territorial Population in 2017
                                                         Populationa   Number of Unitsb per Million Populationa
Province / Territory   CT MRI PET-CT PET-MRI SPECT SPECT-CT
Alberta 4,291,980 13.05 9.55 0.93 0.00 9.79 7.46
British Columbia 4,789,221 13.78 9.60 0.63 0.00 5.85 6.47
Manitoba 1,332,629 17.26 9.00 0.75 0.00 6.75 6.00
New Brunswick 757,641 19.80 14.52 2.64 0.00 6.60 6.60
Newfoundland and Labrador 528,683 30.26 9.46 1.89 0.00 3.78 17.02
Northwest Territories 44,381 22.53 0.00 0.00 0.00 0.00 0.00
Nova Scotia 953,173 18.88 12.59 1.05 0.00 7.34 10.49
Nunavut 37,462 26.69 0.00 0.00 0.00 0.00 0.00
Ontario 14,135,610 13.02 8.49 1.20 0.21 10.68 5.52
Prince Edward Island 149,790 13.35 6.68 0.00 0.00 0.00 13.35
Quebec 8,371,498 19.47 12.78 2.51 0.00 9.20 9.08
Saskatchewan 1,161,365 12.92 8.61 0.86 0.00 7.75 8.61
Yukon 37,808 26.45 26.45 0.00 0.00 0.00 0.00
Canada 36,591,241 15.33 10.00 1.39 0.08 9.02 7.13

CT = computed tomography; MRI = magnetic resonance imaging;PET = positron emission tomography; PET-CT = positron emission tomography–computed tomography; PET-MRI = positron emission tomography–magnetic resonance imaging; SPECT = single-photon emission computed tomography; SPECT-CT = single-photon emission computed tomography– computed tomography.
a The population (estimated) as of July 1, 2017.16
b Per-province unit counts according to the validator.


Overall Number of Examinations

Table 5 shows the total number of examinations reported for all modalities across Canada for the most recent fiscal (or calendar) year for each site. Data from validators are reported, supplemented with survey data if validator data are unavailable (SPECT and SPECT-CT). Several jurisdictions combined SPECT and SPECT-CT exam data; therefore, an aggregated total is reported for these two modalities. A total of five exams were reported by a single site with PET-MRI. More detailed presentations of each modality appear in following sections, and the data are presented in Appendix D, Table 21.


Table 5: Total Examinations for the Latest Fiscal (or Calendar) Year for All Modalities Across Canada in 2017
Examinations CT MRI PET-CT SPECT and SPECT-CT
Exams reported by validatorsa 5,611,107 1,855,110 90,530 1,125,516
Additional exams estimated from survey datab -- -- -- 228,605
Number of unitsc 561 366 51 591
Exams per 1,000 peopled 153.0 51.0 2.0

37.0

CT = computed tomography; MRI = magnetic resonance imaging;PET-CT = positron emission tomography–computed tomography; SPECT = single-photon emission computed tomography; PET-CT = positron emission tomography–computed tomography.

a Validator reported exams were for public sites only.
b Data derived from survey question: “For all [modality] units, how many examinations on average were conducted in the last fiscal year.” Number of exams for sites that did not report exams were imputed from the average for that province.
c Unit counts were those reported by validators.
d The population (estimated) as of April 1, 2017 (see Table 4).16


Overall Summary of Use: Hours per Day and Hours per Week

Figure 2 and Figure 3 show the pattern of use by hours per day and hours per week, respectively, for all sites with data available. The graph depicts the percentage of sites with units used less than eight hours a day, eight to less than 12 hours per day, 12 to less than 18 hours per day, and greater than 18 hours per day, and the bars are labelled with the number of sites in each category. CT is the most heavily used modality, followed by MRI.


Figure 2: Overall Hours of Use per Day for All Modalities: Percentage of Sites Reporting Specific Hours of Use in 2017

CT = computed tomography; MRI = magnetic resonance imaging; PET = positron emission tomography; PET-CT = positron emission tomography–computed tomography; PET-MRI = positron emission tomography–magnetic resonance imaging; SPECT = single-photon emission computed tomography; SPECT-CT = single-photon emission computed tomography–computed tomography.

Data derived from survey question: “In a regular 24-hour workday, how many hours are the [modality] units in use?”
Where 2017 data were not available, 2015 data were used, where available. Sites without data from 2015 or 2017 were not included in the figures.
Bars are labelled with the number of sites in each category.


Figure 3: Overall Hours of Use per Week for All Modalities: Percentage of Sites Reporting Specific Hours of Use in 2017

CT = computed tomography; MRI = magnetic resonance imaging; PET = positron emission tomography; PET-CT = positron emission tomography–computed tomography; PET-MRI = positron emission tomography–magnetic resonance imaging; SPECT = single-photon emission computed tomography; SPECT-CT = single-photon emission computed tomography–computed tomography.

Data derived from survey question: “In an average 168 hour week, how many hours are the [modality] units in use?”
Where 2017 data were not available, 2015 data were used, where available. Otherwise, no imputation was done, and the site was not included in the totals.
Bars are labelled with the number of sites in each category.


Overall Age of Units and Relationship of Age to Use

The age of imaging equipment was assessed, with the results presented in Table 6. The age of each unit was estimated by calculating the number of years since the first year of operation (i.e., 2017 – first year of operation). The age of imaging equipment that was identified as previously used in the survey could not be estimated. Therefore, these units are not included in the age cohorts, but are presented on their own. Information on a total of 1,615 units was retrieved. The age could be estimated in 83.5% of these units, 2.4% had been installed as previously used, and 14.2% of units did not have age data available. With the exception of SPECT, the majority of imaging equipment has 10 or fewer years of operation: 74.3% of CT units, 69.9% of MRI units, 84.2% of PET-CT units, and 87.6% SPECT-CT units (Table 6). In the case of SPECT, 42.5% of units are less than 10 years of age. SPECT is the only modality for which equipment installed more than 20 years ago is still in use (4.4%).


Table 6: Age of Units for All Modalities Across Canada, Numbers in 2017
Age of Unitsa             Number (and %)
  CT MRI PET-CT SPECT SPECT-CT
0–5 years 167 (34.0) 115 (36.4) 12 (31.6) 41 (12.9) 73 (39.5)
6–10 years 198 (40.3) 106 (33.5) 20 (52.6) 94 (29.2) 89 (48.1)
11–15 years 114 (23.2) 84 (26.6) 6 (15.6) 115 (36.2) 22 (11.9)
16–20 years 12 (2.4) 11 (3.5) 0 54 (17.0) 1 (0.5)
>20 years 0 0 0 14 (4.4) 0
Used Units 15 11 2 8 2
Total 506 327 40 326 197

CT = computed tomography; MRI = magnetic resonance imaging; PET-CT = positron emission tomography–computed tomography; SPECT = single-photon emission computed tomography; SPECT-CT = single-photon emission computed tomography–computed tomography.

a Age for each unit calculated from survey question: “What year did (or will) the [modality] unit become operational?” subtracted from 2017.


RESULTS-Computed Tomography (CT)  


Number and Location of CT Units

Three hundred and sixty-three sites in 13 provinces and territories have one or more CT units. There were up to nine units per site, for a total of 561 units. Ontario, Quebec, and British Columbia had the most CT units, and Nunavut, Northwest Territories, and Yukon the fewest, with one unit each. The number of units in provinces or territories with the modality ranges from 13.02 per million population in Ontario to 30.26 per million population in Newfoundland and Labrador, but this does not necessarily reflect accessibility, particularly in provinces and territories with large remote areas.

Eighty-eight new CT units were installed between 2015 and 2017. Thirty-four were replacement units for decommissioned units, 22 were new units, and 32 were not specified as new or replacement. Forty-six sites decommissioned one or more CT units (most decommissioned one unit) since the last survey in 2015, and 15 sites (of 147 sites with submitted surveys) reported planned installations of one or more CT units in the next two years (Table 7).


Table 7: Summary of Availability and Status of CT Units by Province in 2017
Province / Territory Sites With Unitsa Number of Unitsb Sites Planning to Installc Sites That Decommissioned Units Since 2015d Units per Million Populatione
Alberta 41 56 7 6 13.05
British Columbia 47f 66f 4 7 13.78
Manitoba 16 23 1 3 17.26
New Brunswick 11 15 0 2 19.80
Newfoundland and Labrador 14 16 0 3 30.26
Northwest Territories 1 1 0 1 22.53
Nova Scotia 14 18 0 1 18.88
Nunavut 1 1 0 0 26.69
Ontario 105 184 3 17 13.02
Prince Edward Island 2 2 0 1 13.35
Quebec 97f 163f 0 4 19.47
Saskatchewan 13 15 0 1 12.92
Yukon 1 1 0 0 26.45
Canada 363 561 15 46 15.33

CT = computed tomography. a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned), as provided by survey respondents to CMII and CIHI, and industry sources.
b Data obtained from provincial validators.
c Data derived from survey question: “Do you have plans to install the following in the next two years?” Data available from 147 submitted surveys.
d Data derived from survey question: “Has this unit been decommissioned?” if the year of decommissioning was 2015 or later.
e The population (estimated) as of July 1, 2017 (see Table 4).16
f Provincial totals included private units.


Geographical Distribution of CT Units

Figure 4 shows the present geographical distribution of CT units across Canada mapped to the level of settlement (city or town), with circle diameter proportional to the number of units. Counts for all sites within a city/town were aggregated.


Figure 4: Distribution of CT Units Across Canada in 2017

CT = computed tomography. Availability and unit counts by site were derived from validator data if site-level data were available; otherwise, from survey data. Mobile units appear as one unit at each of the sites served.


Mobile CT

Two CT units were identified as mobile units in urban Quebec. No information was available about whether they were shared between sites or operated as fixed units.


Number of Examinations in a Fiscal Year

Across Canada, an overall total of 5,611,107 CT examinations per year were reported for 561 units. Data quoted was supplied by provincial validators, at a provincial level, for either the latest fiscal or calendar year. The average number of exams per unit per year was 10,002. Table 8 shows the number of exams by province, and the number of exams per 1,000 people.


Table 8: Total Examinations per Fiscal Year for CT in 2017
Province / Territory All Unitsa Total Examsb,c Populationd Exams per 1,000 Population
Alberta 56 405,332e 4,291,980 94.4
British Columbia 66 695,248 4,789,221 145.2
Manitoba 23 186,197 1,332,629 139.7
New Brunswick 15 142,294 757,641 187.8
Newfoundland and Labrador 16 90,985 528,683 172.1
Northwest Territories 1 4,695 44,381 105.8
Nova Scotia 18 155,099 953,173 162.7
Nunavut 1 2,000 37,462 53.4
Ontario 184 2,430,739 14,135,610 172.0
Prince Edward Island 2 15,811 149,790 105.6
Quebec 163 1,350,792 8,371,498 161.4
Saskatchewan 15 128,415 1,161,365 110.6
Yukon 1 3,500 37,808 92.6
Canada 561 5,611,107 36,591,241 153.0

CT = computed tomography.

a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned).
b Validator-supplied data on examinations.
c Validator-supplied data for public sites only.
d The population (estimated) as of July 1, 2017 (see Table 4).16
e Exams from Alberta Health Services facilities only.

The reported examination data are summarized by province or territory in Appendix D, Table 21.


Typical Hours of Operation in a Week and Day, and All-Day and Weekend Use

Two hundred and twenty-five sites out of 363 provided data for the average number of hours per day that CT units were in use. Across all provinces or territories where a modality was available, CT units were used for an average of eight to 13.3 hours per day, depending on the province or territory (Appendix D, Table 22). Thirty-one per cent of units were used for less than eight hours per day, 30.7% were used for eight to less than 12 hours per day, 26.2% of units were used for 12 to less than 18 hours a day, and 12% of units were used for more than 18 hours per day. Sixty-five sites reported that at least one unit at their site was used for 24 hours a day (Appendix D, Table 23). We did not distinguish between scheduled and on-call availability in this question.

Two hundred and twenty-seven sites out of 363 provided data for the average number of hours per week that CT units were in use. Across all provinces or territories where a modality was available, CT units are used for an average of 40 to 88.3 hours per week (Appendix D, Table 24). Twenty-six per cent of units are used for less than 20 hours per week, 22% are used for 40 to less than 60 hours per week, 15.4% are used for 60 to less than 80 hours per week, 22.5% of units are used for 80 to less than 120 hours a week, and 14.5% of units are used for more than 120 hours per week. One hundred and seventy-five sites reported that at least one unit at their site is used on weekends (Appendix D, Table 25). We did not distinguish between scheduled and on-call availability in this question.


Types of CT Use

Survey participants were asked to provide the overall percentage of use for cardiac exams, non-cardiac exams, research, and any other type of use. Use breakdown was available for 159 sites. On average, the highest percentage of use for CT was non-cardiac, at 94.6% (use at individual sites ranged from 0% to 100%) followed by other, at 3.5% (range 0% to 100%), and cardiac, at 1.3% (range 0% to 15%). Details are available in Appendix D, Table 26.

Survey participants were asked to provide an overall percentage of use by discipline for CT. The categories included oncology, respiratory, hepatobiliary, musculoskeletal, inflammatory, neurological, cardiac, trauma, and other. Usage breakdown was available for 63 sites. On average, the highest percentage of use for CT was oncological exams, at 23.1% (use at individual sites ranged from 0% to 100%), followed by neurological, at 18.9% (range 0% to 55%), respiratory, at 12.6% (range 0% to 30%), and hepatobiliary exams, at 11.5% (range 0% to 35%). Details are available in Appendix D, Table 27. Due to the low response rate, data may not be representative of all sites.


Technical Characteristics of CT Units

The number of slices was available for 460 units. CTs with 64 slices (51.5%) are the most common, followed by those with 128 slices (13.2%) and 16 slices (13.2%; Table 9). Six per cent (5.8%) of units have more than 256 slices. Information on dual-energy options was available for 409 units. Almost a quarter (23.5%, 96/409) of CT units have a dual-energy option, which allows for the simultaneous acquisition of images at two different energies, as a means of enhancing images and reducing radiation exposure. Information on dual-target options was available for 209 units. A dual-target option is available in 12.9% (27/209) of CT units.


Table 9: Reported Number of Slices in CT Units in 2017
Slices 1 2 4 6 8 10 16 32
Number (%) 9 (2.0) 3 (0.7) 15 (3.2) 1 (0.2) 9 (2.0) 1 (0.2) 81 (13.2) 2 (0.4)
Slices 40 64 80 128 160 192 256 320
Number (%) 3 (0.7) 237 (51.5) 2 (0.4) 61 (13.2) 6 (1.3) 3 (0.7) 14 (3.0) 13 (2.8)

CT = computed tomography.
Data from question: “How many multidetectors does the CT unit have (how many slices)?”

The survey also asked about features intended to manage radiation safety. Three-quarters (77.2%, 183/237) of CT units incorporate image reconstruction techniques for dose reduction, and 91.6% (185/202) of CT units record patient radiation dose by exam. More than 80% (81.4%, 197/242) of CT units are equipped with dose-management controls, and 93.0% (160/172) of the survey responders reported use of these controls. Data for the others were missing.

 


RESULTS-Magnetic Resonance Imaging (MRI)


Number and Location of MRI Units

Two hundred and sixty-one sites in 11 provinces or territories have one or more MRI units. There were up to eight units per site, for a total of 366 units. Ontario, Quebec, and British Columbia had the most MRI units. Yukon is the only territory to have an MRI unit. The number of units per million population in provinces or territories with the modality ranges from 6.68 in Prince Edward Island to 26.45 in Yukon, but this does not necessarily reflect accessibility, particularly in provinces and territories with large remote areas.

Forty-six new MRI units were installed between 2015 and 2017. Six were replacement units for decommissioned units, 15 were new units, and 25 were not specified as new or replacement. Twelve sites decommissioned one or more MRI units (most decommissioned one unit) since the last survey in 2015, and nine sites (of 147 sites with submitted surveys) reported planned installations of one or more MRI units in the next two years (Table 10).


Table 10: Summary of Availability and Status of MRI Units by Province in 2017
Province / Territory Sites With Unit(s)a Number of Unitsb Sites Planning to Installc Sites That Decommissioned Units Since 2012d Units per Million Populatione
Alberta 29 41 2 3 9.55
British Columbia 42f 46f 1 0 9.60
Manitoba 7 12 1 1 9.00
New Brunswick 9 11 0 2 14.52
Newfoundland and Labrador 5 5 0 0 9.46
Northwest Territories 0 0 0 0 0.00
Nova Scotia 11 12 0 0 12.59
Nunavut 0 0 0 0 0.00
Ontario 74f 120f 5 2 8.49
Prince Edward Island 1 1 0 0 6.68
Quebec 75 107 0 4 12.78
Saskatchewan 7 10 1 0 8.61
Yukon 1 1 0 0 26.45
Canada 261 366 9 12 10.00

MRI = magnetic resonance imaging.
a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned), as provided by survey respondents to CMII and CIHI, and by industry sources.
b Data obtained from provincial validators.
c Data derived from survey question: “Do you have plans to install the following in the next two years?” Data were available from 147 surveys.
d Data derived from survey question: “Has this unit been decommissioned?” if the year of decommissioning was 2015 or later.
e The population (estimated) as of July 1, 2017 (see Table 4).16
f Provincial totals included private units.


Geographical Distribution of MRI

Figure 5 shows the geographical distribution of MRI across Canada mapped to the level of settlement (city or town), with circle diameter proportional to the number of units. Counts for all sites within a city/town were aggregated.


Figure 5: Distribution of MRI Units Across Canada in 2017

MRI = magnetic resonance imaging.

Availability and unit counts by site were derived from validator data if site-level data were available; otherwise, from survey data. Mobile units appear as one unit at each of the sites served.


Mobile MRI

Twenty-two sites indicated that they were served by mobile units, with two units in British Columbia, two in Quebec, and one each in Alberta and New Brunswick. In British Columbia, one mobile unit is shared by three facilities in the Okanagan and Kootenays, and another is shared by two facilities on Vancouver Island. The unit in New Brunswick is shared among five facilities. The unit in Alberta is shared by two facilities outside of Edmonton in Central Alberta. One unit in Quebec is shared by five facilities, all in the administrative region of Gaspésie–Îles-de-la-Madeleine, while the other is shared by three facilities in Abitibi-Témiscamingue.


Number of Examinations in a Fiscal Year

Across Canada, an overall total of 1,855,110 MRI examinations per year were reported for 366 units, with each site reporting its last fiscal (or calendar) year. The average number of exams per unit was 5,082. Table 11 shows the total number of exams by province and the number of exams per 1,000 people.


Table 11: Total Examinations per Fiscal Year for MRI in 2017
Province / Territory All Unitsa Total Examsb,c Population Exams per 1,000 Populationd
Alberta 41 192,375e 4,291,980 44.8
British Columbia 46 173,678 4,789,221 36.3
Manitoba 12 77,735 1,332,629 58.3
New Brunswick 11 44,592 757,641 58.9
Newfoundland and Labrador 5 20,990 528,683 39.7
Northwest Territories 0 NA 44,381 NA
Nova Scotia 12 47,490 953,173 49.8
Nunavut 0 NA 37,462 NA
Ontario 120 866,953 14,135,610 61.3
Prince Edward Island 1 4,279 149,790 28.6
Quebec 107 380,357 8,371,498 45.4
Saskatchewan 10 44,461 1,161,365 38.3
Yukon 1 2,200 37,808 58.2
Canada 366 1,855,110 36,591,241 51.0

NA = not applicable.

a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned).
b Exams supplied by provincial validators.
c Validator-supplied data were for public sites only.
d The population (estimated) as of July 1, 2017 (see Table 4).16
e Exams from Alberta Health Services facilities only.

The reported examination data are summarized by province or territory in Appendix D, Table 21.


Typical Hours of Operation in a Week and Day, and All-Day and Weekend Use

One hundred and forty-five sites (145) out of 261 provided data for the average number of hours per day that MRI units were in use. Across all provinces or territories where a modality was available, MRI units were used for an average of 9 to 16.2 hours per day, depending on the province or territory (Appendix D, Table 22). Seventeen per cent of units were used for less than eight hours per day, 29% were used for eight to less than 12 hours per day, 46.9% of units were used for 12 to less than 18 hours a day, and 6.9% of units were used for more than 18 hours per day. Eleven sites reported that at least one unit at their site was used for 24 hours a day (Appendix D, Table 23). We did not distinguish between scheduled and on-call availability in this question.

One hundred and forty-four sites out of 261 provided data for the average number of hours per week that MRI units were in use. Across all provinces or territories where a modality was available, MRI units were used for an average of 40 to 108.6 hours per week (Appendix D, Table 24). Eighteen per cent of units were used for less than 20 hours per week, 20.8% were used for 40 to less than 60 hours per week, 18.8% were used for 60 to less than 80 hours per week, 32.6% of units were used for 80 to less than 120 hours a week, and 9.7% of units were used for more than 120 hours per week. Eighty-three sites reported that at least one unit at their site was used at weekends (Appendix D, Table 25). We did not distinguish between scheduled and on-call availability in this question.


Types of MRI Use

Survey participants were asked to provide the overall percentage of use for cardiac exams, non-cardiac exams, research, and any other type of use. Use breakdown was available for 86 sites. On average, the highest percentage of use for MRI was non-cardiac exams, at 85.7% (use at individual sites ranged from 0% to 100%) followed by other uses, at 4.3% (range 0% to 100%) and cardiac exams, at 3.5% (range 0% to 50%). Details are available in Appendix D, Table 26.

Survey participants were asked to provide the overall percentage of use by discipline for MRI. The categories included oncology, cardiac, respiratory, hepatobiliary, musculoskeletal, neurological, trauma, and other uses. Use breakdown was available for 34 sites. On average, the highest percentage of use for MRI was musculoskeletal use, at 27.3% (use at individual sites ranged 0% to 70%), followed by neurological use, at 25.4% (range 0% to 45%), oncology use, at 20.8% (range 0% to 100%), and hepatobiliary use, at 11.3% (range 0% to 25%). Details are available in Appendix D, Table 27. Due to the low response rate, data may not be representative of all sites.


Technical Characteristics of MRI Units

For all MRIs with available information on field strength (321 units), the majority (267, 83.2%) operate with a 1.5 Tesla (T) magnetic field strength. The second most common field strength is 3.0 T (13.7%). The remaining MRI units used field strengths of 0.3 T (0.3%), 0.35 T (0.3%), 1.0 T (1.6%), 4.0 T (0.3%), 5.0 T (0.3%), and 9.4 T (0.3%). Data on MRI configuration were available for 170 units. The survey indicates 6.5%, 55.3%, and 38.2% of MRI units use an open bore, closed bore, and wide bore, respectively.


 


RESULTS-Positron Emission Tomography–Computed Tomography (PET-CT) or Positron Emission Tomography (PET)


Number and Location of PET-CT or PET Units

Forty-five sites in nine provinces have one or more PET-CT units. There are up to two units per site, for a total of 51 units. Quebec, Ontario, and Alberta have the most PET-CT units, and Prince Edward Island and the three territories do not have any units. Although we asked about PET-CT or PET, findings of a recent CADTH Environmental Scan17 suggest these units are almost exclusively PET-CT, since stand-alone PET units have not been available for purchase in Canada for the past decade. Therefore, this category is referred to as PET-CT, except when quoting survey questions. The number of units per million population in provinces or territories with the modality ranges from 0.63 in British Columbia to 2.64 in New Brunswick, but this does not necessarily reflect accessibility, particularly in provinces and territories with large remote areas.

One new PET-CT unit was installed between 2015 and 2017. No sites decommissioned a PET-CT unit, and two sites (of 147 submitted surveys) reported planned installations of one or more PET-CT units in the next two years (Table 12).


Table 12: Summary of Availability and Status of PET-CT Units by Province in 2017
Province / Territory Sites With Unitsa Number of Unitsb Sites Planning to Installc Sites That Decommissioned Units Since 2012d Units per Million Populatione
Alberta 3 4 1 0 0.93
British Columbia 2f 3f 1 0 0.63
Manitoba 1 1 0 0 0.75
New Brunswick 2 2 0 0 2.64
Newfoundland and Labrador 1 1 0 0 1.89
Northwest Territories 0 0 0 0 0.00
Nova Scotia 1 1 0 0 1.05
Nunavut 0 0 0 0 0.00
Ontario 14f 17f 0 0 1.20
Prince Edward Island 0 0 0 0 0.00
Quebec 20f 21f 0 0 2.51
Saskatchewan 1 1 0 0 0.86
Yukon 0 0 0 0 0.00
Canada 45 51 2 0 1.39

PET-CT = positron emission tomography–computed tomography.
a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned), as provided by survey respondents to CMII and CIHI, and by industry sources.
b Data obtained from provincial validators.
c Data derived from survey question: “Do you have plans to install the following in the next two years?” Data available from 147 surveys.
d Data derived from survey question: “Has this unit been decommissioned?” if the year of decommissioning was 2015 or later.
e The population (estimated) as of July 1, 2017 (see Table 4).16
f Provincial totals include private units.


Geographical Distribution of PET-CT

Figure 6 shows the present geographical distribution of PET-CT across Canada mapped to the level of settlement (city or town), with circle diameter proportional to the number of units. Counts for all sites within a city/town were aggregated.


Figure 6: Distribution of PET-CT Across Canada in 2017

PET-CT = positron emission tomography–computed tomography.
Availability and unit counts by site were derived from validator data if site-level data were available; otherwise, from survey data.


Number of Examinations in a Fiscal Year

Across Canada, an overall total of 90,530 PET-CT examinations per year were reported for 51 units, with each site reporting its last fiscal year. The average number of exams per unit was 1,775. Table 13 shows the total number of exams by province and the number of exams per 1,000 people.


Table 13: Reported and Imputed Total Examinations per Fiscal Year for PET-CT in 2017
Province / Territory All Unitsa Total Examsb,c Population Exams per 1,000 Populationd
Alberta 4 11,050e 4,291,980 2.6
British Columbia 3 9,280 4,789,221 1.9
Manitoba 1 2,009 1,332,629 1.5
New Brunswick 2 1,808 757,641 2.4
Newfoundland and Labrador 1 0 528,683 0.0
Northwest Territories 0 NA 44,381 NA
Nova Scotia 1 2,512 953,173 2.6
Nunavut 0 NA 37,462 NA
Ontario 17 10,998 14,135,610 0.8
Prince Edward Island 0 NA 149,790 NA
Quebec 21 50,823 8,371,498 6.1
Saskatchewan 1 2,050 1,161,365 1.8
Yukon 0 NA 37,808 NA
Canada 51 90,530 36,591,241 2.0

NA = not applicable; PET-CT = positron emission tomography–computed tomography.
a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned).
b Exams supplied by provincial validators.
cValidator-supplied data were for public sites only.
dThe population (estimated) as of July 1, 2017 (see Table 4).16
e Exams from Alberta Health Services facilities only.

The reported examination data, without imputation, are summarized by province or territory in Appendix D, Table 21.


Typical Hours of Operation in a Week and Day, and All-Day and Weekend Use

Twenty-six of 45 sites provided data for the average number of hours per day that PET-CT units were in use. Across all provinces where a modality was available, PET-CT units were used for an average of six to 14.5 hours per day, depending on the province (Appendix D, Table 22). Fifty-four per cent of units were used for less than eight hours per day, 34.6% were used for eight to less than 12 hours per day, and 11.5% of units were used for 12 to less than 18 hours a day. No sites reported 24 hours-a-day operation (Appendix D, Table 23). We did not distinguish between scheduled and on-call availability in this question.

Twenty-five of 45 sites provided data for the average number of hours per week that PET-CT units were in use. Across all provinces where a modality was available, PET-CT units were used for an average of 18 to 79 hours per week (Appendix D, Table 24). Fifty-two per cent of units were used for less than 20 hours per week, 36% were used for 40 to less than 60 hours per week, and 12% were used for 60 to less than 80 hours per week. Two sites reported that at least one unit at their site was used at weekends (Appendix D, Table 25). We did not distinguish between scheduled and on-call availability in this question.


Type of PET-CT Use

Survey respondents were asked to provide the overall percentage of use for cardiac exams, non-cardiac exams, research, and any other type of use. Use breakdown was available for 19 sites. On average, the highest percentage of use for PET-CT was non-cardiac exams, at 80.7% (use at individual sites ranged from 20% to 100%), followed by cardiac exams, at 12.1% (range 0% to 80%) and research use, at 6.8% (range 0% to 50%). Details are available in Appendix D, Table 26.

Survey respondents were asked to provide the overall percentage of use by discipline for PET-CT. The categories were oncology, cardiac, inflammatory, neurology, and other. The breakdown of use was available for 13 sites. On average, the highest percentage of use for PET-CT was oncology use, at 80.2% (use at individual sites ranged from 0% to 100%) followed by cardiac use, at 10.7% (range 0% to 95%), neurological use, at 7% (range 0% to 50%), and inflammatory use, at 1.2% (range 0% to 5%). Details are available in Appendix D, Table 27. Due to the low response rate, data may not be representative of all sites.


Technical Characteristics of PET-CT Units

The number of detector row slices available in the CT component of these PET-CT machines was available for 37 units. Sixteen slices is the most common number (67.6%), followed by 64 slices (21.6%) and 40 slices (5.4%). No PET-CTs are used exclusively for head imaging. The CT component is used independently (i.e., to provide extra CT capacity) in 40.0% (6/15) of units answering this question.

The survey also asked about features intended to manage radiation safety. Of the 24 units with responses available to that question, 87.5% are equipped with dose-management controls, and, of those, 90.0% (18/20) reported use of these controls. Data for the others were missing. Two-thirds (14/21) of PET-CTs incorporate image reconstruction techniques for dose reduction, and 77.3% (17/22) of PET-CT units record patient radiation dose per exam.


Isotope Supply for PET Hybrid Modalities

We asked sites reporting a PET-CT whether they have access to a cyclotron, and, if not, where they obtain isotopes. Of the 19 sites who responded to the question, eight have access to a local cyclotron. The Canadian Nuclear Safety Commission website identified a total of 10 PET cyclotrons in Canada.18

Nine sites of the 18 sites without a cyclotron reported obtaining isotopes elsewhere, the majority from commercial suppliers (6). The remainder obtained isotopes from other sites with cyclotrons (2) or did not indicate a source (10).

 


RESULTS- Positron Emission Tomography-MRI (PET-MRI)


Number and Location of PET-MRI Units

Three sites in Ontario have one PET-MRI unit each.

One new PET-MRI unit was installed between 2015 and 2017.


Figure 7: Distribution of PET-MRI units across Canada in 2017

PET-MRI = positron emission tomography–magnetic resonance imaging;.


Patterns of PET-MRI use

All units are currently for research use only. However, five examinations were reported for one unit.


Technical Characteristics of PET-MRI

Three PET-MRIs were identified in the survey. No information on the imaging scope or mobility of these units was available.


RESULTS- Single-Photon Emission Computed Tomography (SPECT)


Number and Location of SPECT Units

One hundred and eighty-one sites in nine provinces have one or more SPECT units. There are up to nine units per site, for a total of 330 units. Ontario, Quebec, and Alberta have the most SPECT units, while Prince Edward Island and the three territories have none. The number of units per million population in provinces or territories with SPECT ranges from 3.78 in Newfoundland and Labrador to 10.68 in Ontario, but this does not necessarily reflect accessibility, particularly in provinces and territories with large remote areas.

Thirteen new SPECT units were installed between 2015 and 2017. Three were replacement units for decommissioned units, two were new units, and eight were not specified as new or replacement. Eighteen sites decommissioned one or more SPECT units (most decommissioned one unit) since the last survey in 2015, and three sites (of 147 sites that submitted surveys) reported planned installations of one or more SPECT units in the next two years (Table 14).


Table 14: Summary of Availability and Status of SPECT Units by Province in 2017
Province / Territory Sites With Unitsa Number of Unitsb Sites Planning to Installc Sites That Decommissioned Units Since 2012d Units per Million Populatione
Alberta 29 42 2 1 9.79
British Columbia 16 28 1 3 5.85
Manitoba 6 9 0 1 6.75
New Brunswick 3 5 0 0 6.60
Newfoundland and Labrador 2 2 0 1 3.78
Northwest Territories 0 0 0 0 0.00
Nova Scotia 7 7 0 1 7.34
Nunavut 0 0 0 0 0.00
Ontario 74 151f 0 9 10.68
Prince Edward Island 0 0 0 0 0.00
Quebec 39g 77g 0 2 9.20
Saskatchewan 4 9 0 0 7.75
Yukon 0 0 0 0 0.00
Canada 181 330 3 18 9.02

SPECT = single-photon emission computed tomography.
a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned), as provided by survey respondents to CMII and CIHI, and by industry sources.
b Data obtained from provincial validators.
c Data derived from survey question: “Do you have plans to install the following in the next two years?” Data available from 147 submitted surveys.
d Data derived from survey question: “Has this unit been decommissioned?” if the year of decommissioning was 2015 or later.
e The population (estimated) as of July 1, 2017 (see Table 4).16
f Validator-supplied combined unit counts for SPECT and SPECT-CT. Distribution was assumed to be 50:50, based on the division in Quebec and overall.
g Provincial totals included private units.


Geographical Distribution of SPECT

Figure 8 shows the present geographical distribution of SPECT across Canada mapped to the level of settlement (city or town), with circle diameter proportional to the number of units. Counts for all sites within a city/town were aggregated.


Figure 8: Distribution of SPECT Across Canada in 2017

SPECT = single-photon emission computed tomography.

Availability and unit counts by site were derived from validator data if site-level data were available; otherwise, from survey data. Mobile units appear as one unit at each of the sites served.


Mobile SPECT

Four SPECT units at three sites in Ontario were identified as mobile units. Three were multifunctional, and one was a dedicated cardiac unit. There was no information on whether the units were shared among the identified sites, or were operated as fixed units.


Number of Examinations in a Fiscal Year: SPECT

For several provinces, validators supplied aggregated information for SPECT and SPECT-CT only; therefore, examination data are summarized for both SPECT and SPECT-CT in the section on SPECT-CT (Table 16).


Typical Hours of Operation by Week and Day, and All-Day and Weekend Use

Seventy-five of 181 sites provided data for the average number of hours per day that SPECT units were in use. Across all provinces where a modality was available, SPECT units were used for an average of seven to 9.9 hours per day, depending on the province (Appendix D, Table 22). Sixty-three per cent of units were used for less than eight hours per day, 29.3% were used for eight to less than 12 hours per day, 6.7% of units were used for 12 to less than 18 hours a day, and 1.3% of units were used for more than 18 hours per day. No sites reported that at least one unit at their site was used for 24 hours a day (Appendix D, Table 23). We did not distinguish between scheduled and on-call availability in this question.

Seventy-six of 181 sites provided data for the average number of hours per week that SPECT units were in use. Across all provinces or territories where a modality was available, SPECT units were used for an average of 35 to 49.4 hours per week (Appendix D, Table 24). Sixty-three per cent of units were used for less than 20 hours per week, 27.6% were used for 40 to less than 60 hours per week, 7.9% were used for 60 to less than 80 hours per week, and 1.3% of units were used for 80 to less than 120 hours a week. Nine sites reported that at least one unit at their site was used on weekends (Appendix D, Table 25). We did not distinguish between scheduled and on-call availability in this question.


Types of SPECT Use

Survey respondents were asked to provide the overall percentage of use for cardiac exams, non-cardiac exams, research, and any other type of use for all units at their site. Use breakdown was available for 50 sites. On average, the highest percentage of use for SPECT was non-cardiac exams, at 73.9% (use at individual sites ranged from 0% to 100%), followed by cardiac exams, at 26.3% (range 0% to 100%), and research use, at 0.3% (range 0% to 10%). Details are available in Appendix D, Table 26.

Survey respondents were asked to provide the overall percentage of use by discipline for SPECT for all units at their site. For SPECT, the categories were oncology, cardiac, inflammatory, neurological, and other use. The breakdown of use by discipline was available for 19 sites. On average, the highest percentage of use for SPECT was cardiac exams, at 39.8% (use at individual sites ranged 0% to 100%), followed by oncology use, at 24.7% (range 0% to 100%), musculoskeletal use, at 10.4% (range 0% to 48%), and respiratory use, at 7.2% (range 0% to 65%). Details are available in Appendix D, Table 27. Due to the low response rate, data may not be representative of all sites, as suggested by the difference between cardiac use captured by this question, and cardiac use reported for the breakdown of use by cardiac, non-cardiac, and other use, above.


Technical Characteristics of SPECT Units

The number of detector heads was reported for 83 units, with two detector heads being the most common configuration (73.5%), followed by one (18.1%), three (7.2%), and six (1.2%). CT capability of one- to 64-slice resolution was reported for 37 units. Of the 90 units with the information available, a quarter (24.4%) are dedicated cardiac imaging units, with the remaining 75.6% having multipurpose or non-cardiac use. Note that the section on use reports average use for all units at a site. Field of view was reported for 77 units, with 14.3%, 84.4%, and 1.3% using a dedicated limited, multipurpose, or other view, respectively. SPECT units generally use one of two types of software to generate images: filtered projection or iterative reconstruction. Of the 66 units with available information, 48.5% use filtered projection software, and 51.5% use interactive reconstruction.


RESULTS-Single-Photon Emission Computed Tomography–Computed Tomography (SPECT-CT)


Number and Location of SPECT-CT Units

One hundred and fifty-five sites in 10 provinces or territories have one or more SPECT-CT units. There are up to six units per site, for a total of 261 units. Ontario, Quebec, and Alberta have the most SPECT-CT units. The number of units per million population in provinces or territories with the modality ranges from 5.52 in Ontario to 17.02 to Newfoundland and Labrador, but this does not necessarily reflect accessibility, particularly in provinces and territories with large remote areas.

Thirty-seven new SPECT-CT units were installed between 2015 and 2017. Three were replacement units for decommissioned units, 13 were new units, and 21 were not specified as new or replacement. No sites decommissioned any SPECT-CT units since the last survey in 2015, and five sites (of 147 sites with submitted surveys) reported planned installations of one or more SPECT-CT units in the next two years (Table 15).


Table 15: Summary of Availability and Status of SPECT-CT Units by Province in 2017
Province / Territory Sites With Unitsa Number of Unitsb Sites Planning to Installc Sites That Decommissioned Units Since 2012d Units per Million Populatione
Alberta 19 32 1 0 7.46
British Columbia 18 31 0 0 6.47
Manitoba 5 8 0 0 6.00
New Brunswick 5 5 1 0 6.60
Newfoundland and Labrador 4 9 0 0 17.02
Northwest Territories 0 0 0 0 0.00
Nova Scotia 8 10 0 0 10.49
Nunavut 0 0 0 0 0.00
Ontario 48 78 3 0 5.52
Prince Edward Island 1 2 0 0 13.35
Quebec 42 76 0 0 9.08
Saskatchewan 5 10 0 0 8.61
Yukon 0 0 0 0 0.00
Canada 155 261 5 0 7.13

SPECT-CT = single-photon emission computed tomography–computed tomography.
a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned) as provided by survey respondents to CMII and CIHI, and by industry sources.
b Data obtained from provincial validators.
c Data derived from survey question: “Do you have plans to install the following in the next two years?” Data available from 147 submitted surveys.
dData derived from survey question: “Has this unit been decommissioned?” if the year of decommissioning was 2015 or later.
e The population (estimated) as of July 1, 2017 (see Table 4).16
fValidator-supplied combined unit counts for SPECT and SPECT-CT. Distribution was assumed to be 50:50, based on the division in Quebec and overall.


Geographical Distribution of SPECT-CT

Figure 9 shows the present geographical distribution of SPECT-CT across Canada mapped to the level of settlement (city or town), with circle diameter proportional to the number of units. Counts for all sites within a city/town were aggregated.


Figure 9: Distribution of SPECT-CT across Canada in 2017

SPECT-CT = single-photon emission computed tomography–computed tomography.

Availability and unit counts by site were derived from validator data if site-level data were available; otherwise, from survey data. Mobile units appear as one unit at each of the sites served.


Number of Examinations in a Fiscal Year: SPECT and SPECT-CT

For several provinces, validators supplied aggregated information for SPECT and SPECT-CT only. Data as provided are shown in Appendix D, Table 21. For the purposes of summary, exams for SPECT and SPECT-CT are pooled for all provinces. Across Canada, an overall total of 1,354,121 examinations per year were reported for 591 SPECT and SPECT-CT units, with each site reporting its last fiscal year. The average number of exams per unit was 2,273. Table 16 shows the total number of SPECT and SPECT-CT exams by province, and the number of exams per 1,000 people.


Table 16: Reported and Imputed Total Examinations per Fiscal Year for SPECT and SPECT-CT in 2017
Province / Territory All Unitsa Total Examsb,c Population Exams per 1,000 Populationd
Alberta 74 26,130e 4,291,980 6.1
British Columbia 59 148,578f 4,789,221 31.0
Manitoba 17 22,074 1,332,629 16.6
New Brunswick 10 39,635 757,641 52.3
Newfoundland and Labrador 11 49,835 528,683 94.3
Nova Scotia 17 25,413f 953,173 26.7
Nunavut 0   37,462  
Northwest Territories 0   44,381  
Ontario 229 200,833 14,135,610 14.2
Prince Edward Island 2 2,299 149,790 15.3
Quebec 153 786,594 8,371,498 94.0
Saskatchewan 19 52,730f,g 1,161,365 45.4
Yukon 0   37,808  
Canada 591 1,354,121 36,591,241 37.0

SPECT = single-photon emission computed tomography; SPECT-CT = single-photon emission computed tomography–computed tomography.
a Data derived from number of units with technical details (make, model, first year of operation, and status as decommissioned/non-decommissioned).
b Validator-supplied from provincial validators, unless otherwise noted.
c Validator-supplied data were for public sites only.
d The population (estimated) as of July 1, 2017.16
e Exams from Alberta Health Services facilities only.

f No validator data for the province (British Columbia and Saskatchewan) or a health region (NS). Values imputed from available survey data.
g SPECT-CT data available, only.


Typical Hours of Operation in a Week and Day, and All-Day and Weekend Use

Eighty-seven of 157 sites provided data for the average number of hours per day that SPECT-CT units were in use. Across all provinces or territories where a modality was available, SPECT-CT units were used for an average of seven to 10.8 hours per day, depending on the province (Appendix D, Table 22). Fifty-six per cent of units were used for less than eight hours per day, 34.5% were used for eight to less than 12 hours per day, and 9.2% of units were used for 12 to less than 18 hours a day. No sites have 24-hour use (Appendix D, Table 23). We did not distinguish between scheduled and on-call availability in this question.

Eighty-seven of 157 sites provided data for the average number of hours per week that SPECT-CT units were in use. Across all provinces or territories where a modality was available, SPECT-CT units were used for an average of 35 to 54 hours per week (Appendix D, Table 24). Fifty-six per cent of units were used for less than 20 hours per week, 32.2% were used for 40 to less than 60 hours per week, 9.2% were used for 60 to less than 80 hours per week, and 2.3% of units were used for 80 to less than 120 hours a week. Nine sites reported that at least one unit at their site was used on weekends (Appendix D, Table 25). We did not distinguish between scheduled and on-call availability in this question.


Types of SPECT-CT Use

Survey respondents were asked to provide the overall percentage of use for cardiac exams, non-cardiac exams, research, and any other type of use. Use breakdown was available for 55 sites. On average, the highest percentage of use for SPECT-CT was non-cardiac uses, at 78.1% (use for individual sites ranged from 15% to 100%), followed by cardiac uses, at 21.6% (range 0% to 85%) and other uses, at 0.2% (range 0% to 10%). Details are available in Appendix D, Table 26.

Survey respondents were asked to provide the overall percentage of use by discipline. For SPECT-CT, the categories were oncology, cardiac, hepatobiliary, inflammatory, lymphatic, musculoskeletal, neurological, respiratory, thyroid, and other uses. The breakdown of use was available for 15 sites. On average, the highest percentage of use for SPECT-CT was cardiac use, at 35.7% (use for individual sites ranged from 0% to 100%), followed by oncological use, at 25.3% (range 0% to 100%), musculoskeletal uses, at 18.7% (range 0% to 90%), and respiratory use, at 4.7% (range 0 to 19%). Details are available in Appendix D, Table 27. Due to the low response rate, data may not be representative of all sites, as suggested by the difference in cardiac use captured by this and the breakdown of use by cardiac, non-cardiac, and other uses.


Technical Characteristics of SPECT-CT Units

The number of CT multi-detectors (slices) was reported for 132 units, and the number of detector heads for 116 units. The number of slices ranged from one to 40, with the most common being four (31.8%), one (20.5%), and 16 slices (19.7%). A large majority of machines (94.8%) were equipped with two detector heads, and the remainder had one or three heads. Information on the field of view was gathered for 91 SPECT-CT units. The most common field of view was multi-purpose (94.5%), with the remaining five machines utilizing dedicated limited (1.1%), standard polyvalent (1.1%), and other (3.3%) views.

Eighty-six units included responses for software for image processing. A third (32.6%, 28/86) of SPECT-CT units used dedicated limited software for image processing, with the remaining 67.4% using iterative reconstruction software. Out of 106 units that provided information on whether they were dedicated to cardiac imaging or used for multipurpose or non-cardiac imaging, a small minority  7.5% (8/106)  were dedicated cardiac units, with the remaining 92.5% (98/106) used for multi-purpose or non-cardiac imaging. Finally, 95 units included information on whether the CT component of SPECT-CT units is operated as a stand-alone CT.  Of these, 12.6% (12/95) of SPECT-CT do operate as stand-alone CT units, which allows for extra CT capacity in some facilities.

The survey also asked about features intended to manage radiation safety.  Information on the number of machines equipped with dose management controls was available for 92 units, two-thirds (63.0%, 58/92) of which had these controls. Information on whether dose management controls were used was available for 46 units, of which 84.8% (39/46) use these controls. Data for the others were missing. Information was available for 79 units on whether they incorporated reconstruction techniques for dose reduction. Just more than half (55.7%, 44/79) incorporate reconstruction techniques for dose reduction.  Information was available for 95 units on whether units recorded patient radiation dose per exam, and 61.1% (58/95) reported that they did.


RESULTS-Picture Archiving and Communication Systems (PACS)

Of the 251 sites (of 443 total) in 13 provinces or territories that reported PACS access, 33.2% have provincial/territorial access, 38.5% have regional access, and 28.3% have local access.

PACS images were widely accessible to referring physicians outside the imaging department, with access available at all sites with provincial/territorial or regional access and at 90.1% (64/71) of sites with local/institutional access. Access to PACS images throughout a provincial health care network, without the need to manually push images from any location or modality, was provided by 97.9% (94/96) of sites with provincial/territorial access, 63.9% (53/83) of sites with regional access, and 68.6% (48/70) of sites with local (institutional) access.

Some sites that lack a particular modality have access to images from that modality taken elsewhere. Table 18 shows the relationship between modality availability and PACS status for all sites and modalities with data available.


Table 17: Availability of Modalities and Images on PACS (Percentage of Sites) in 2017
Modality and PACS Status CT MRI PET-CT PET-MRI SPECT SPECT-CT
At site, on PACS 91.3 62.1 12.5 0 33.0 38.9
At site, not on PACS 1.2 4.7 2.3 0 13.8 10.6
Not at site, on PACS 1.2 4.7 3.7 50 5.4 3.0
Not at site, not on PACS 6.3 28.5 81.5 50 47.8 47.4

CT = computed tomography; MRI = magnetic resonance imaging, PACS = picture archiving and communication system; PET = positron emission tomography; PET-CT = positron emission tomography–computed tomography; PET-MRI = positron emission tomography–magnetic resonance imaging; SPECT = single-photon emission computed tomography; SPECT-CT = single-photon emission computed tomography–computed tomography.


RESULTS-Appropriateness of Orders Received

In response to a question about whether sites had a process for determining the appropriateness of received orders, 83.2% of 179 sites responded “yes,” and 16.8% responded “no.”


RESULTS-Canadian Data Versus International Data

The availability of unit count and use data allowed us to compare the Canadian data we collected with international data from the Organisation for Economic Co-operation and Development (OECD).12-15 Last observation carried forward was used for all countries to impute values for comparison with the CMII 2017 data. Years of comparison were 2014 to 2016.


Computed Tomography

Based on the unit counts supplied by provincial validators, Canada appears in the lower third of OECD countries with data on number of CT units per million people collected by OECD,15 a position similar to that in the last survey (Figure 10).Comparator data are the latest reported for each country, up to 2016.


Figure 10: Comparison of Canadian and International Data for CT-Total Units per Million People

BRA = Brazil; AUS = Australia; AUT = Austria; CAN = Canada; CHE = Switzerland; CHL = Chile; CZE = Czech Republic; DEU = Germany; DNK = Denmark; ESP = Spain; EST = Estonia; FIN = Finland; FRA = France; GBR = Great Britain; GRC = Greece; HUN = Hungary; IRL = Ireland; ISL = Iceland; ISR = Israel; ITA = Italy; JPN = Japan; KOR = South Korea; LTU = Lithuania; LUX = Luxembourg; LVA = Latvia; MEX = Mexico; NLD = Netherlands; NZL = New Zealand; POL = Poland; RUS = Russia; SVK = Slovak Republic; SVN = Slovenia; TUR = Turkey.

Canadian data from CMII 2017 survey (red square) are compared with the latest data from each of the comparator countries, up to 2016.

Based on the exam totals for the most recent fiscal year provided by provincial validators, Canada appears in the middle third relative to the countries with data for CT exams collected by OECD,13 a position similar to that in 2015 (Figure 11). Examination data are for publicly funded sites only. Comparator data are the latest reported for each country, up to 2016.


Figure 11: Comparison of Canadian and International Data for CT: Total Exams per Thousand People

BRA = Brazil; AUS = Australia; AUT = Austria; CAN = Canada; CHE = Switzerland; CHL = Chile; CZE = Czech Republic; DEU = Germany; DNK = Denmark; ESP = Spain; EST = Estonia; FIN = Finland; FRA = France; GBR = Great Britain; GRC = Greece; HUN = Hungary; IRL = Ireland; ISL = Iceland; ISR = Israel; ITA = Italy; JPN = Japan; KOR = South Korea; LTU = Lithuania; LUX = Luxembourg; LVA = Latvia; MEX = Mexico; NLD = Netherlands; NZL = New Zealand; POL = Poland; RUS = Russia; SVK = Slovak Republic; SVN = Slovenia; TUR = Turkey.

Canadian data from CMII 2017 survey (red square) are compared with the latest data from each of the comparator countries, up to 2016.


Magnetic Resonance Imaging

Based on the unit counts supplied by provincial validators, Canada appears in the middle third of OECD countries with data for MRI units collected by OECD,14 a similar position to 2015 (Figure 12).Comparator data are the latest reported for each country, up to 2016.


Figure 12: Comparison of Canadian and International Data for MRI: Total units per Million People

BRA = Brazil; AUS = Australia; AUT = Austria; CAN = Canada; CHE = Switzerland; CHL = Chile; CZE = Czech Republic; DEU = Germany; DNK = Denmark; ESP = Spain; EST = Estonia; FIN = Finland; FRA = France; GBR = Great Britain; GRC = Greece; HUN = Hungary; IRL = Ireland; ISL = Iceland; ISR = Israel; ITA = Italy; JPN = Japan; KOR = South Korea; LTU = Lithuania; LUX = Luxembourg; LVA = Latvia; MEX = Mexico; NLD = Netherlands; NZL = New Zealand; POL = Poland; RUS = Russia; SVK = Slovak Republic; SVN = Slovenia; TUR = Turkey.

Canadian data from CMII 2017 survey (red square) are compared with latest data from each of the comparator countries, up to 2016.

Based on the exam totals for the most recent fiscal year provided by provincial validators, Canada is in the lower half for MRI exams per thousand people relative to OECD countries with data for MRI exams collected by OECD,12 a position similar to that in 2015 (Figure 12). Examination data are for publicly funded sites only.


Figure 13: Comparison of Canadian and International Data for MRI: Exams per Fiscal Year per 1,000 People

BRA = Brazil; AUS = Australia; AUT = Austria; CAN = Canada; CHE = Switzerland; CHL = Chile; CZE = Czech Republic; DEU = Germany; DNK = Denmark; ESP = Spain; EST = Estonia; FIN = Finland; FRA = France; GBR = Great Britain; GRC = Greece; HUN = Hungary; IRL = Ireland; ISL = Iceland; ISR = Israel; ITA = Italy; JPN = Japan; KOR = South Korea; LTU = Lithuania; LUX = Luxembourg; LVA = Latvia; MEX = Mexico; NLD = Netherlands; NZL = New Zealand; POL = Poland; RUS = Russia; SVK = Slovak Republic; SVN = Slovenia; TUR = Turkey.

Canadian data from CMII 2017 survey (red square) are compared with the latest data from each of the comparator countries, up to 2016.

 


DISCUSSION


Overall Findings

These results are based on responses from a pan-Canadian survey of health care facilities, combined with historical data from previous surveys and data from provincial validators.


Type of Facility

Most sites that responded were publicly funded hospitals, community hospitals, or tertiary care centres in urban areas. The number of free-standing clinics varied across provinces, depending on policy and funding within the provinces. Some provinces use referral to free-standing clinics to manage wait times. Responding sites were predominantly urban, with just more than a quarter of the sites being rural, and a small number being remote. The setting data are incomplete, as sites that were identifiably remote did not have a response to that particular question. Some closely situated rural sites are served by mobile units (particularly CT or MRI) shared among several sites.


Modalities and Number of Units

Of the modalities surveyed, CT is the most widely distributed, with the highest number of units and highest volume of use overall (based on the number of exams and hours of use), followed by MRI. All provinces and territories have at least one CT unit; all provinces and Yukon have at least one MRI unit; and all provinces have at least one SPECT and SPECT-CT unit. None of the territories have SPECT or SPECT-CT. Nine provinces have PET-CT in clinical use. One province, Ontario, has PET-MRI.


Variation in Number of Exams and Hours of Use

For this iteration of the CMII, we reported validator data at the provincial level for the final summary, although we also collected site-level data on the number of exams and used these for imputation if validator data were unavailable. In some instances, there was a considerable difference between the data collected through the survey (whether imputation was conducted or not) and the data supplied by the validators. Data on hours per week, hours per day, and 24-hour or weekend use were collected by the survey. As in the 2015 survey, there was substantial variation in data collected from validators and from the survey in the number of exams, the hours of use (per week or day), and 24-hour or weekend use across and within jurisdictions. Some of this variation may be due to the availability of units, the availability of trained operators and clinicians, and the age of units. Regional variations in practice, guidelines, adherence to guidelines, and data collection may also play a role.


Variation in Type of Use Across Modalities

As in the previous iteration, we collected data on the distribution of use among cardiac, non-cardiac, research, and other categories of use. In addition, for this iteration, we collected more detailed data on distribution of use by clinical discipline. All modalities were used predominantly for non-cardiac purposes, although SPECT and SPECT-CT have higher cardiac use. Among disciplines, oncology, musculoskeletal imaging, and neurology form the major uses for all modalities, with the exception of PET-CT, which is almost entirely used for oncology with a small proportion of neurology imaging. Cancer, injuries, degenerative musculoskeletal disease, and neurological injury and disease are all common conditions. PET-MRI is at present used for research purposes only.

Again, we focused on reporting imaging for clinical purposes in our selection of survey contacts, so imaging for research purposes may be underreported. In large centres, research facilities for medical imaging may be separate for clinical (diagnostic or interventional) imaging, and the survey and validation may not have captured all units primarily used for research. Other uses may have been captured but not identified under the “other” category, such as animal research studies or veterinary imaging, although the percentage under the “other” category was small.


Age of Diagnostic Imaging Equipment

Using the information presented in Table 6, direct comparisons between the age of equipment in Canada and the age of imaging equipment guidelines published by the Canadian Association of Radiologists (CAR)19 and the European Coordination Committee of the Radiological, Electromedical and Healthcare IT Industry (COCIR)20 can be made.

The CAR guidelines19 propose life expectancies for imaging equipment according to use, which can be classified as either high, medium, or low, based on the number of examinations per year. For the five modalities with age information from the CMII (CT, MRI, PET-CT, SPECT, and SPECT-CT) the CAR guidelines propose life expectancies of eight, 10, and 12 years for high, medium, and low usage machines, respectively. When CMII data are compared with the CAR life expectancies for medium-use machines (as an example), 25.7% of CTs, 30.0% of MRIs, 15.8% of PET-CTs, 57.5% of SPECTs, and 12.4% of SPECT-CTs are 11 years of age or older (as determined by the survey), whereas the CAR guideline recommends a device life expectancy of 10 years for the medium-use range. The CAR guidelines also recommend that the maximum life expectancy and clinical relevance for any imaging equipment should not be expected to exceed 15 years. The age of PET-CT, the newest modality in clinical use, is consistent with this recommendation, suggesting that the transition from PET to PET-CT is complete. Two per cent (2.4%) of CT units, 3.5% of MRI units, and 0.5% SPECT-CT units are older than 15 years, and a substantial percentage (21.4%) of SPECT units are beyond the CAR’s recommended life expectancy. Trends over time suggest that SPECT units are progressively being upgraded to or replaced by SPECT-CT units.8,10,11,21 The maturity and stability of the technology and the lack of incentives for or barriers to upgrading may be factors in prolonging operation of SPECT units.

The COCIR guidelines20 contain three “Golden Rules” created to evaluate medical equipment age and aid procurement decisions, creating an age profile that aims to balance keeping equipment current with the need to maintain efficient health care systems.

  • At least 60% of imaging equipment should be five years old or less. None of the imaging modalities meet this criterion, with 34.0% of CT units, 36.4% of MRI units, 31.6% of PET-CT units, 12.9% of SPECT units, and 39.5% of SPECT-CT units being five years of age or newer.
  • No more than 30% of imaging equipment should be between six and 10 years old. Only SPECT meets this criterion, and, as previously described, SPECT machines are older. Overall, 40.3% of CT units, 33.5% of MRI units, 52.6% of PET-CT units, 29.6% of SPECT units, and 48.1% of SPECT-CT units are between the ages of six and 10.
  • No more than 10% of imaging equipment should be older than 10 years. None of the modalities meet this criterion, with 25.7% of CT units, 30.1% of MRI units, 15.7% of PET-CT units, 57.5% SPECT units, and 12.4% of SPECT-CT units older than 10 years.

The CAR is a professional medical association with industry partnerships based in Canada, so its recommendations are more cognizant of Canadian practice and the health care landscape. The COCIR is a European-based organization representing the manufacturers of diagnostic imaging equipment, and therefore may be less relevant to the Canadian context. Overall, imaging equipment in Canada appears to trend older than the recommended profile, but, as the CAR guidelines indicate, lifetime depends on extent of use. We do not currently have unit-level exam data, having requested average exams for all units at a site. In addition, the number of exams is unavailable for many sites.


PACS Accessibility

Most facilities with units for modalities at a site also stored PACS images for those modalities, while a minority of sites without units available had access to PACS images for those modalities. We do not know what forms of storage are used at sites where PACS is not used, and how the images are shared with referring physicians and consultants.


Jurisdictional Differences


The Influence of Geography

The survey was restricted to the six advanced medical imaging modalities identified by experts as of the most interest, and these modalities are concentrated in major urban centres and in provinces and territories with larger populations. The larger provinces — Alberta, British Columbia, Ontario, and Quebec — have the greatest variety of modalities and number of individual units, followed by Manitoba, New Brunswick, Newfoundland and Labrador, Nova Scotia, and Saskatchewan, which have a relatively modest number of units. Jurisdictions with the fewest units include the territories and Prince Edward Island.

The number of units per million people is more consistent across provinces than the total unit counts per population, but this measure does not account for population distribution in the respective provinces or territories. As an example of the geographic challenges, the Northwest Territories and Nunavut each have a single CT unit serving the entire region (22.53 and 26.69 units per million population, respectively). Quebec, a province of comparable area, has 163 units (19.47 units per million population), predominantly in the south of the province.

Some of the smaller and less populated provinces or territories lack within-jurisdiction access to most modalities, and access may depend on cross-jurisdictional partnerships, patients’ abilities and willingness to travel, and integrated telemedicine services.


Funding Structures

One aspect of medical imaging practice that the survey did not address was potential differences in funding structures across jurisdictions, specifically private–public partnerships22,23 and cost-sharing across jurisdictions (e.g., Ontario/Manitoba).24 Based on comments provided by survey respondents, there may be cross-jurisdictional care of patients, as well as referral of wait-listed patients to private clinics. In both of these scenarios, it is not clear whether private or public funds are used to cover the costs of imaging and care. Regulatory frameworks in place to govern operation of private facilities may also differ across jurisdictions25 and may influence the type of imaging modality and the number and utilization of private clinics.


Trained Personnel

The availability of trained personnel to conduct and interpret imaging exams may also contribute to some of the variation observed across sites. An academic training centre, research facilities, and large health care facilities that provide employment opportunities may be necessary to attract clinicians, technologists, technicians, and other support staff (e.g., radiation safety specialists and nuclear medicine physicists). Remote or rural centres may face challenges in attracting and retaining highly trained professionals, or in providing training and continuing education for existing staff.26 As telemedicine and mobile technologies evolve, there may be opportunities to provide improved access through the combination of these innovations.


Impact of Availability on Wait Times

Wait times for medical imaging are an ongoing concern in Canada.27 While provincial strategies are in place to reduce wait times for these services, Canadian patients are still waiting beyond targeted wait times.28,29 Overall, wait times have increased for CT and MRI (the only modalities for which wait times are recorded and that are relevant to this report) over the last five years. The CAR recommends maximum MRI and CT wait time targets of 24 hours for emergency/life-threatening conditions, seven days for urgent conditions, 30 days for semi-urgent conditions, and 60 days for non-urgent conditions.27 Median wait times for medically necessary elective MRI are longer (10.8 weeks) than for CT (4.1 weeks), and there is variation across jurisdictions, with half of patients receiving an MRI exam within 33 to 84 days and receiving a CT within six to 34 days.28,29 While wait times have increased, so have the number of exams performed by radiology professionals, with a 75% increase in CT exams and a 82% increase in MRI exams over the last decade. Our findings indicate that imaging machines are used for an average of eight to 13.3 hours per day for CT and nine to 16.2 hours per day for MRI, suggesting there is capacity to do more exams on existing equipment, rather than invest in new equipment, which is a common strategy used to reduce wait times.30 This will not, however, improve access in regions where there is no existing imaging equipment, or address restrictions due to funding of imaging and availability of radiologists, technicians, and support staff.



Overall Summary

Each imaging modality, with the exception of SPECT, experienced growth over the last 10 years in Canada, both in numbers of units and in number of units per million people (Figure 14). The 2017 figures are drawn from this report, and the 2007 figures, from a report by CIHI published in 2007.11 The most rapid growth occurred with the hybrid modality SPECT-CT, followed by PET-CT and MRI. The most widely available imaging modality in Canada and the only imaging modality that is available in all provinces and territories (CT) experienced the slowest growth rate of all imaging modalities, at 35% over the last decade. This slow growth rate, compared with other imaging modalities, may be related to that fact that the CT market may already be saturated in Canada. Based on the available data in the 2017 CMII, 53% of CT units installed in the last five years were replacements, compared with 17% of MRI units.

In 2006-2007, SPECT was the most popular imaging modality, with 603 units in 10 jurisdictions.10 In 2016-2017, the number of SPECT units declined to 330 in nine jurisdictions. The decline of 45% may be connected to the rapid adoption of SPECT-CT. For each modality, approximately 60% of all growth can be attributed to two provinces, Quebec and Ontario. PET-MRI had yet to emerge.

Examination data for 2006-2007 were reported only for CT and MRI. The number of exams increased by 75% and 82% for the two modalities, respectively, and the number of exams per thousand population increased by 48% and 63%, respectively.


Figure 14: Change in units per population between 2007 and 2017 for CT, MRI, PET-CT, SPECT, and SPECT-CT


Computed Tomography

CT is the only imaging modality reviewed in this report that is available in every jurisdiction (province or territory). There were 419 CT units in Canada in 2006-2007,10 compared with 561 in 2016-2017, representing a 34% increase over the last 10 years. There was at least one CT unit, to a maximum of 184 CT units, per jurisdiction in 2016-2017, compared with 10 years ago, when 12 jurisdictions had between one and 130 CT units each. For both 200710 and 2017, 60% of all CT units were located in the two most densely populated provinces, Ontario and Quebec.

Over the last 10 years, growth in CT units outstripped population growth, with the greatest increase in Newfoundland and Labrador (21.6 units per million10to 30.2 units per million) and Quebec (15.5 units per million10 to 19.5 units per million). CT units per million people decreased for five jurisdictions, as the number of CT units remained unchanged while the population grew. For all of Canada, CT units per million people expanded from 12.8 per million people in 2006-200710 to 15.4 CT per million people in 2016-2017.

There were 3,380,597 publicly funded CT exams in Canada in 2006-2007,10 compared with 5,611,107 in 2016-2017, representing a 75% increase over a decade. Six jurisdictions (British Columbia, Ontario, Quebec, Prince Edward Island, Yukon, and the Northwest Territories) experienced a growth rate of 60% or more in the number of exams. Nationwide, the overall rate of CT examinations per 1,000 population increased from 103.3 to 153, representing a 48% increase over the last 10 years.


Magnetic Resonance Imaging

Of the imaging modalities reviewed in this report, MRI is the second most numerous. There were 222 MRI units in 10 jurisdictions in 2006-200710 compared with 366 in 11 jurisdictions in 2016-2017, representing a 65% increase over the last 10 years. The two jurisdictions with no existing MRI capacity also have the lowest populations in Canada. Approximately 60% of all MRI units are located in Ontario and Quebec, and this was the same in 2006-2007.

The number of MRI units per million people increased for all jurisdictions with existing MRI capacity over the last 10 years, with the exception of Prince Edward Island, where the number of units remained unchanged although the population experienced 9% growth. The jurisdictions with the greatest growth in MRI units per million — New Brunswick, Nova Scotia, and Saskatchewan — doubled their MRI capacity over the 10-year period. For all of Canada, MRI units increased from 6.8 per million people in 2006-200710 to 10.0 MRIs per million people in 2016-2017.

In 2016-2017, the number of MRI exams conducted in Canada was higher in all jurisdictions compared with 2006-2007. Around 1 million MRI exams were conducted in Canada in 2006-2007,10 compared with 1,855,110 in 2016-2017, representing a 82% increase over 10 years. In 2016-2017, Ontario accounted for 47% of all MRI exams, compared with 44% of all MRI exams in 2006-2007. For Canada as a whole, the rate of MRI examinations per 1,000 population rose from 31.210 to 51.0, an increase of 63%.


Positron Emission Tomography and Positron Emission Tomography–Computed Tomography

There were 21 PET-CT in Canada in 2006-2007,10 compared with 51 in 2016-2017, representing a 65% increase over the last 10 years. There were PET-CT units in nine jurisdictions in 2016-2017, compared with six jurisdictions 10 years ago.10 The jurisdictions with no PET-CT are the jurisdictions with the smallest populations. About 74% of all PET-CT units were in Ontario and Quebec in 2006-2007, and there is no notable change in 2016-2017.

At the national level, PET-CT units increased from 0.9 per million people in 2006-200710 to 1.4 per million people in 2016-2017. Seven jurisdictions experienced a slight growth in the number of PET-CT per million people over the last 10 years. One jurisdiction experienced a slight decline in units per million population.


Single-Photon Emission Computed Tomography

SPECT is the only imaging modality reviewed in this report that decreased in numbers over the last 10 years. In 2006-2007, SPECT was the most commonly available imaging modality in Canada, with 603 units in 10 jurisdictions.10 Ten years later, there were 330 SPECT units in nine jurisdictions, representing an overall decline of 45%. The number of SPECT units may be lower than reported because some jurisdictions include planar imaging in this count.

For all of Canada, SPECT declined over the last 10 years, from 18.4 per million people in 2006-2007,10 to 9.0 per million people in 2016-2017. The decline is SPECT may be attributed to its gradual replacement by SPECT-CT. In 2006-2007, Ontario and Quebec accounted for 67% of all SPECT, compared with 71% in 2016-2017.


Single-Photon Emission Computed Tomography–Computed Tomography

SPECT-CT is unique among the imaging modalities reviewed in this report, in that it experienced significant rapid growth over the last 10 years compared with the other modalities. In 2006-2007, there were 35 SPECT-CT units in five jurisdictions,10compared with 261 SPECT-CT units in 10 jurisdictions in 2016-2017, representing an increase of 646% over the last decade. Ontario and Quebec account for 58% of all SPECT-CT in 2016-2017, compared with 80% in 2006-2007.

Correspondingly, SPECT-CT units increased from 1.0 per million people in 2006-200710 to 7.3 per million in 2016-2017.


Appropriate Imaging and Radiation Safety

There is growing concern about potential health hazards associated with imaging exams that use radiation.31 Most of the emphasis is placed on CT exams because CT accounts for the majority of total radiation received by patients from any imaging modalities,31 although hybrid imaging modalities that use CT are also significant contributors to individual patient radiation dose.32 Our survey indicates that 80% of facilities have a process in place for determining the appropriateness of imaging exams. As well, dose-management controls and unit recording of patient radiation dose per exam are widely used for CT, PET-CT, and SPECT-CT. A CADTH pan-Canadian survey of awareness and implementation of the 10 priorities of the Bonn Call for Action33 on radiation protection indicated a commitment to improving radiation safety, with most respondents reporting at least partial implementation (with intent toward further implementation) at their institutions and in Canadian clinical practice.34


Canadian Data Versus International Data

Compared with other countries of varying levels of development around the world, as recorded by the OECD, Canada currently appears in the lower third of countries reporting the number of CT units per population to the OECD, slightly down in ranking from 2015. Canada ranks around the middle for MRI units. Canada appears in the upper 50% for number of CT exams per population and in the lower 50% for MRI exams.

Our estimated number of exams is based on validator-supplied data, which included only publicly funded facilities, potentially leading to undercounting for the jurisdictions with private facilities.

No OECD data were publicly available on the other modalities of interest, and so Canada’s status compared with other countries is unclear in these cases. In the future, it would be interesting to compare Canada’s adoption of emerging technologies (e.g., PET-MRI) against other countries, especially as adoption and clinical use increases.


Strengths

The data collected for this report are part of an ongoing survey of medical imaging equipment in Canada that was first reported in 2001.11 As such, this survey meets a need that is especially relevant in the current environment of proliferating use of medical imaging and emergent technologies and clinical applications.

The data have been compiled through several iterations of surveys, augmented by information on installations from three of the major suppliers of imaging equipment in Canada. Extensive efforts have been made to obtain high-level data from provincial and territorial validators. Through these efforts, we believe an accurate characterization of the medical imaging landscape in Canada has been achieved. This version of the survey also reports data on technical specifications and age of imaging units in Canada, and trends over time, allowing further insight of associations with patterns of use not only by modality, but also by specific type of modality.


Limitations


Selection of Imaging Modalities

For reasons of feasibility, this iteration of the survey was restricted to six specialist imaging modalities, and omits others that are more common and widespread (for example, X-ray and ultrasound) or that were included in earlier years of the survey (planar gamma cameras, angiography, and bone densitometry). This focus biases the coverage toward urban areas, and does not capture alternative imaging options available outside these regions, especially in remote or rural areas, where patients need to travel or be transferred significant distances for imaging. In addition, these exclusions may limit understanding of the relationship among modalities within the health care system (for example, in pathways that involve multiple modalities), and may limit consideration of funding allocation for diagnostic imaging across all modalities. We will review inclusion of additional modalities for future iterations of the survey as needs and technologies evolve, and we will consider the possibility of conducting a survey limited by geography (i.e., focused on remote areas) rather than by modality.


Private Versus Public Coverage

As participation in the survey was not mandatory, and a definitive up-to-date list of facilities using medical imaging equipment in Canada was lacking, we cannot ensure that all facilities were contacted or represented. In particular, there was a notable difference in the representation of public and privately funded facilities, with more responses from the former. Publicly funded facilities were more readily identified than private facilities, as their data tend to be held at multiple administrative levels. Most provinces lack a publicly available repository of private imaging facilities. This may lead to underestimation of the number of units and of the total number of exams, particularly in jurisdictions where privately run imaging contributes to the overall use.


Variable Instrument Coverage

The quality and completeness of the data collected appear to be relatively high for CT and MRI compared with the other modalities. Both modalities are well-established and have seen longstanding use. For SPECT and SPECT-CT, data are more variable; for instance, one province (Ontario) reported combined SPECT and SPECT-CT counts, and several provinces reported combined SPECT and SPECT-CT exams, or reported a single total for nuclear medicine exams as a whole. It is possible that, in facilities with both a diagnostic imaging and a nuclear medicine department, that we failed to reach the latter. With repeated iterations of the survey, we expect to extend our lists of contacts and obtain more specific information.


Reliability

The accuracy of the data in this report is in part reliant on the personal knowledge of survey participants of their particular health care setting. Level of insight and accuracy of estimates may vary substantially and contribute to variability in the quality and completeness of reporting. Recall bias cannot be avoided, as we were unable to assess whether all information was visually verified and based on real data, or whether questions were answered from memory. Further, respondent fatigue may have affected the responses for difficult questions, such as those regarding the number of examinations or hours of use, particularly if real-time data were not recorded at the facilities.

Industry data came in the form of technical extracts, and contained not only original installations but upgrades, which were on occasion difficult to distinguish. Entries contained abbreviations and industry-specific labelling, which had to be interpreted by the researchers entering the data, with the possibility of misinterpretation, especially misidentifying an upgrade as a new installation. This would not affect overall counts, which were supplied by validators, but might introduce errors into the summaries of technical specifications and age, although these are expected to be minor. We hope to explore the impact of equipment upgrades in future cycles.


Inconsistency in Data Sources

The data uploaded to the survey before opening it for responses were derived from several sources: unit technical specifications from the CMII 2012 data set, facility and technical responses from the 2015 CMII survey, and technical specifications supplied by industry contacts. Several rounds of data reconciliation were required to assemble the data set and remove duplicate entries. Facility names required standardization to identify variations in names and trace name changes and restructuring. Units had to be matched across data sets using available data to avoid duplication, with interpretation of abbreviations and industry-specific terminology. Dates were variously reported as year of installation or first year of operation, and frequently varied across data sets, leading to a one- or two-year uncertainty in the age of individual units, and possibly a corresponding uncertainty in the averages (depending on whether all the errors were in the same direction).

Sites varied in whether they considered certain units to be SPECT or SPECT-CT. Validator reporting of SPECT and SPECT-CT examinations was particularly variable, with combined reporting of SPECT and SPECT-CT exams for four provinces, preventing calculation of the exams for individual modalities, including other nuclear medicine modalities in the totals for two, inflating the results, and reporting only partial information for one province.


Effect of Missing Responses, Assumptions, and Imputations

The technical data for individual units were collected by CIHI for units installed before 2012, and by CADTH for units installed after 2012. This compilation included older units that had not been identified as having been decommissioned. Sites were asked to identify whether individual units had been decommissioned, but, as not all sites reviewed and updated their data, not all decommissioned units were identified. When counts were reconciled with validator data, the oldest surplus units of each modality were assumed to have been decommissioned at each site. If this assumption were incorrect, this would affect the summaries of ages and technical specifications.

Use data (e.g., hours per day and per week) were not updated for all sites. The assumption was made that use data were unlikely to have changed for most sites, and hours per day, hours per week, and types of use were carried forward from 2015 when unavailable for 2017. The data for the new questions around use by discipline were relatively sparse, and we do not know whether the responding sites were representative of all sites.


Variable Interpretation of Questions

Based on comparisons of weekly and daily hours, there was variation in whether sites averaged hours over all calendar days, or only days of operation. If the former, the hours of operation would be underestimated. The question on whether modalities operated on a 24-hour basis did not differentiate between scheduled and on-call availability. Comparison of the data for hours per day with the data for 24-hour operation, and the data for hours per week with the data for weekend operation, suggested that sites varied in their definition of operation (some reported too few hours to have had 24 hours of staffed operation). Therefore, the overall hours of potential availability might be underestimated, and the hours of operation overestimated.


Future Directions and Next Steps


Future Directions for the CMII

Further aspects to explore related to the conduct of the inventory include the following:

  • How should we capture hardware upgrades to existing installations, which several respondents have remarked upon (in comments on the survey) and which have the potential to extend the usable lifespan of equipment substantially?
  • How might we capture the impact of software upgrades on the currency of imaging equipment?
  • How might we capture the eventual impact of artificial intelligence / machine learning / deep learning in terms of use of current equipment and requirements for adjunct equipment?
  • How should we extend capture of availability and use for privately funded sites, which in some jurisdictions contribute significantly to overall imaging use?
  • How should we determine whether the age profile of current units is appropriate to their level of use, as the data were not available to categorize levels of use and compare with age, as described by the CAR?
  • How should we capture the installation of previously used equipment?


Policy, Research, and Clinical Practice Questions

Other questions have been provoked by developments in diagnostic imaging, health technology assessment, and the current medical imaging context in Canada.

  • How might tracking of equipment through the entire life cycle enable planning for the replacement of equipment, through entire life cycle
  • The inventory of equipment might assist in planning the implementation of other therapies that depend on imaging (e.g., proton beam therapy requires a CT, MRI, and PET/CT scanner, and the inventory can identify where these already exist).
  • How can the health care system improve access to imaging for patients in remote and rural areas?
  • How does practice in remote and rural areas adapt to the lack of ready access to specialized diagnostic imaging?
  • What is the cost-effectiveness of medical imaging technologies (taking into account wait times, clinical pathways, and clinical utility)?
  • What is the regulatory framework in place to support private–public partnerships, specifically in terms of eligibility for private imaging (e.g., length of wait-list) and proportion of public funding provided?
  • How does legislation regarding diagnostic imaging differ across jurisdictions, and does that influence the way devices are distributed and used?


Conclusions and Implications of Findings

This report presents data on the number of units, their distribution, and their volume and type of use across Canada for six medical imaging modalities, as informed through a comprehensive survey and data-collection process, building on previous iterations of the survey by CADTH and others.35-37 It discusses changes over time, the age of units, technical characteristics, and Canada’s status compared with other countries.

The survey results provide insight into the current context of medical imaging across Canada. They raise relevant questions related to how medical imaging is monitored and regulated, and how it is optimally used. As well, they raise questions about how funding structures are organized, what the most cost-effective practices are, as well as whether access is equitable, especially in rural and remote areas. Overall, the findings of this report may help decision-makers identify gaps in service; inform medical imaging–related strategic planning on a national, provincial, or territorial basis; and help anticipate future growth and need for replacement. CADTH plans to explore the possibility of investigating some of these issues in future knowledge-synthesis work.

Data for the next iteration of the survey is planned to be collected in the summer and fall of 2019, and the update will be published in 2020.

 


REFERENCES

  1. Xu-Welliver M, Yuh WT, Fielding JR, Macura KJ, Huang Z, Ayan AS, et al. Imaging across the life span: innovations in imaging and therapy for gynecologic cancer. Radiographics [Internet]. 2014 Jul [cited 2016 Mar 10];34(4):1062-81. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4319524
  2. Mouden M, Timmer JR, Ottervanger JP, Reiffers S, Oostdijk AH, Knollema S, et al. Impact of a new ultrafast CZT SPECT camera for myocardial perfusion imaging: fewer equivocal results and lower radiation dose. Eur J Nucl Med Mol Imaging. 2012 Jun;39(6):1048-55.
  3. Choosing Wisely Canada [Internet]. Toronto: Choosing Wisely Canada. 2014 [cited 2016 Feb 5]. Available from: http://www.choosingwiselycanada.org/
  4. Choosing Wisely: an initiative of the ABIM Foundation [Internet]. Philadelphia: American Board of Internal Medicine. 2012 [cited 2016 Feb 5]. Available from: http://www.choosingwisely.org/
  5. Canada Safe Imaging [Internet]. Ottawa: Canadian Association of Radiologists; 2009. [cited 2016 Mar 2]. Available from: http://www.car.ca/en/education/canadasafeimaging.aspx
  6. Rao VM, Levin DC. The overuse of diagnostic imaging and the Choosing Wisely initiative. Ann Intern Med. 2012 Oct 16;157(8):574-6.
  7. Fine B, Dhanoa D. Imaging appropriateness criteria: why Canadian family physicians should care. Can Fam Physician. 2014 Mar;60(3):217-8. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3952750
  8. Medical imaging in Canada 2012 (executive summary) [Internet]. Ottawa: Canadian Institute for Health Information (CIHI); 2013 Feb 12. [cited 2015 Aug 25]. Available from: https://www.cihi.ca/en/mit_summary_2012_en.pdf
  9. QuickStat. Selected medical imaging equipment in Canada [Internet]. Ottawa: Canadian Institute for Health Information (CIHI); 2003 -. What is the number of devices in hospitals for computerized tomography in Alberta, 2012?; 2015 Oct 28 [cited 2016 Jan 20]. Available from: https://www.cihi.ca/en/medical-imaging-equipment-in-canada-2012
  10. Medical imaging in Canada 2007 [Internet]. Ottawa: Canadian Institute for Health Information; 2008. [cited 2016 Feb 5]. Available from: https://secure.cihi.ca/free_products/MIT_2007_e.pdf
  11. Sinclair A, Quay T, Pyke L, Morrison A. The Canadian medical imaging inventory, 2015 [Internet]. Ottawa: CADTH; 2016 Mar. [cited 2017 Dec 12]. (CADTH Optimal Use Report). Available from: https://www.cadth.ca/canadian-medical-imaging-inventory-2015
  12. Magnetic resonance imaging (MRI) exams [Internet]. Paris (FR): Organisation for Economic Co-operation and Development; 2017. [cited 2017 Dec 8]. Available from: https://data.oecd.org/healthcare/magnetic-resonance-imaging-mri-exams.htm
  13. Computed tomography (CT) exams [Internet]. Paris (FR): Organisation for Economic Co-operation and Development; 2017. [cited 2017 Dec 8]. Available from: https://data.oecd.org/healthcare/computed-tomography-ct-exams.htm
  14. Magnetic resonance imaging (MRI) units [Internet]. Paris (FR): Organisation for Economic Co-operation and Development; 2017. [cited 2017 Dec 8]. Available from: https://data.oecd.org/healtheqt/magnetic-resonance-imaging-mri-units.htm
  15. Computed tomography (CT) scanners [Internet]. Paris (FR): Organisation for Economic Co-operation and Development; 2017. [cited 2017 Dec 8]. Available from: https://data.oecd.org/healtheqt/computed-tomography-ct-scanners.htm
  16. CANSIM table [Internet]. Statistics Canada; 2012 -. Population by year, by province and territory; 2015 Sep 29 [cited 2016 Jan 20]. Available from: http://www.statcan.gc.ca/tables-tableaux/sum-som/l01/cst01/demo02a-eng.htm
  17. Publicly funded uses of PET scans in Canada [Internet]. Ottawa: CADTH; 2015 Nov. [cited 2016 Jan 29]. (Environmental scan; issue 53). Available from: https://www.cadth.ca/sites/default/files/pdf/ES0297_publicly_funded_uses_of_PET_scans_in_canada.pdf
  18. Canadian Nuclear Safety Commission [Internet]. Ottawa: Canadian Nuclear Safety Commission. Maps of nuclear facilities; 2014 Mar 2 [cited 2016 Jan 29]. Available from: http://nuclearsafety.gc.ca/eng/resources/maps-of-nuclear-facilities/results.cfm?category=medical-facilities#locations
  19. Lifecycle guidance for medical imaging equipment in Canada; summary report [Internet]. Ottawa: Canadian Association of Radiologists; 2013. [cited 2017 Dec 13]. Available from: https://car.ca/wp-content/uploads/car-lifecycleguidance-summary.pdf
  20. Act now: medical imaging equipment age profile & density [Internet]. 2016 edition. Brussels: European Coordination Committee of the Radiological, Electromedical and Healthcare IT Industry (COCIR); 2015. [cited 2017 Nov 27]. Available from: http://www.cocir.org/uploads/media/16052_COC_AGE_PROFILE_web_01.pdf
  21. MIT 2012 data release - static figures & tables [Spreadsheet on the Internet]. Ottawa (ON): Canadian Institute for Health Information; 2012. [cited 2016 Jan 29]. Available from: https://www.cihi.ca/en/types-of-care/specialized-services/xlsx/internet/STATS_MIT_2012_EN
  22. Barrows D, MacDonald HE, Supapol AB, Dalton-Jez O, Harvey-Rioux S. Public-private partnerships in Canadian health care: A case study of the Brampton Civic Hospital. OECD Journal on Budgeting. 2012 [cited 2018 Feb 14];1. Available from: https://www.oecd.org/gov/budgeting/PPP%20Canadian%20healthcare.pdf
  23. Carson S. The Role of the Private Sector in Canadian Healthcare: Accountability, Strategic Alliances, and Governance [Internet]. Kingston: Queen's University; 2015 Jun 4. [cited 2018 Feb 14]. Available from: https://smith.queensu.ca/insight/whitepapers/the_role_of_the_private_sector_in_canadian_healthcare
  24. Northern Health Travel Grants [Internet]. Toronto: Ministry of Health and Long-Term Care; 2017 Oct 24. [cited 2018 Feb 14]. Available from: http://www.health.gov.on.ca/en/public/publications/ohip/northern.aspx
  25. Reid L. Alberta regulatory body makes important move to address private MRIs [Internet]. Toronto: Healthydebate.ca; 2013 Sep 11. [cited 2018 Feb 14]. Available from: http://healthydebate.ca/opinions/alberta-physician-regulatory-college-makes-the-right-move-to-address-private-medical-imaging
  26. Promed Associates Ltd. Medical imaging in northern Canada: a snapshot in time [Internet].Canadian Association of Radiologists; 2009 Apr 18. [cited 2018 Feb 14]. Available from: https://car.ca/wp-content/uploads/Medical-Imaging-in-Northern-Canada-A-Snapshot-in-Time.pdf
  27. The value of radiology in Canada [Internet]. Ottawa: The Conference Board of Canada; 2017 Jan. [cited 2017 Dec 11]. Available from: http://www.conferenceboard.ca/e-library/abstract.aspx?did=8532
  28. Barua B. Waiting your turn: wait times for health care in Canada, 2017 report [Internet]. Toronto: Fraser Institute; 2017. [cited 2018 Feb 13]. Available from: https://www.fraserinstitute.org/sites/default/files/waiting-your-turn-2017.pdf
  29. Canadian Institute for Health Information (CIHI). Wait Times for Priority Procedures in Canada, 2017 [Internet]. Ottawa: CIHI; 2017 Mar. [cited 2017 Dec 11]. Available from: https://secure.cihi.ca/free_products/wait-times-report-2017_en.pdf
  30. Emery DJ, Forster AJ, Shojania KG, Magnan S, Tubman M, Feasby TE. Management of MRI wait lists in Canada. Healthc Policy [Internet]. 2009 Feb [cited 2017 Dec 11];4(3):76-86. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2653696
  31. Natarajan MK, Paul N, Mercuri M, Waller EJ, Leipsic J, Traboulsi M, et al. Canadian Cardiovascular Society position statement on radiation exposure from cardiac imaging and interventional procedures. Can J Cardiol. 2013 Nov;29(11):1361-8.
  32. Rehani MM. Challenges in radiation protection of patients for the 21st century. AJR Am J Roentgenol. 2013 Apr;200(4):762-4.
  33. Bonn call-for-action: joint position statement by the IAEA and WHO [Internet]. Vienna: The International Atomic Energy Agency; 2012. [cited 2018 Feb 13]. Available from: http://www.who.int/ionizing_radiation/medical_exposure/Bonn_call_action.pdf
  34. Radiation Protection and Safety: Awareness and Implementation of the Bonn Call for Action Priorities in Canada. [Internet]. Ottawa: CADTH. Forthcoming 2018. [cited 2018 Feb 13]. Available from: https://www.cadth.ca/radiation-protection-and-safety-awareness-and-implementation-bonn-call-action-priorities-canada
  35. Limb M. Sharp rise in CT scans prompts call for new safeguards on radiation exposure. BMJ. 2014;349:g5218.
  36. Cressman S, Lam S, Tammemagi MC, Evans WK, Leighl NB, Regier DA, et al. Resource utilization and costs during the initial years of lung cancer screening with computed tomography in Canada. J Thorac Oncol [Internet]. 2014 Oct [cited 2015 Aug 25];9(10):1449-58. Available from: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4165479
  37. Siegel BA. 2014 Cassen Lecture: What have we learned from the National Oncologic PET Registry? J Nucl Med. 2014 Dec;55(12):9N-15N.
  38. Diagnostic imaging [Internet]. Geneva: World Health Organization. Computed tomography; 2016 [cited 2016 Jan 29]. Available from: http://www.who.int/diagnostic_imaging/imaging_modalities/dim_comptomography/en/
  39. Ionizing radiation in pregnant women: a review of the safety and guidelines [Internet]. Ottawa: CADTH; 2015 Jun 9. [cited 2016 Mar 7]. (Rapid response report: summary with critical appraisal). Available from: https://www.cadth.ca/sites/default/files/pdf/htis/june-2015/RC0665-IonizingRadiation-Pregnancy-Final.pdf
  40. Radiation emissions from computed tomography: a review of the risk of cancer and guidelines [Internet]. Ottawa: CADTH; 2014 Jun 4. [cited 2016 Mar 7]. (Rapid response report: summary of critical appraisal). Available from: https://www.cadth.ca/sites/default/files/pdf/htis/jul-2014/RC0558%20CT%20Radiation%20Emission%20Final.pdf
  41. Diagnostic imaging [Internet]. Geneva: World Health Organization. Magnetic resonance imaging; 2016 [cited 2016 Jan 29]. Available from: http://www.who.int/diagnostic_imaging/imaging_modalities/dim_magresimaging/en/
  42. Consiglio N. MRI and patient safety. Can J Med Radiat Technol. 2006;37(2):5-9.
  43. Shellock FG. MRIsafety.com [Internet]. [Playa Del Rey (CA)]: Shellock R & D Services, Inc. Safety information article list; 2015 [cited 2016 Jan 29]. Available from: http://www.mrisafety.com/SafetyInfog.asp
  44. Shellock FG, Crues JV 3rd, editors. MRI: Bioeffects, safety and patient management. Los Angeles: Biomedical Research Publishing Company; 2014.
  45. Diagnostic imaging [Internet]. Geneva: World Health Organization. Nuclear medicine; 2016 [cited 2016 Jan 29]. Available from: http://www.who.int/diagnostic_imaging/imaging_modalities/dim_nuclearmed/en/
  46. Appropriate utilization of advanced diagnostic imaging procedures: CT, MRI, and PET/CT [Internet]. Ottawa: CADTH; 2013 Feb. [cited 2016 Jan 29]. (Environmental scan; issue 39). Available from: https://www.cadth.ca/media/pdf/PFDIESLiteratureScan_e_es.pdf
  47. Breastfeeding in nuclear medicine (NM specific), best practice guidelines [Internet]. Ottawa: Canadian Association of Medical Radiation Technologists; 2013 Apr 18. [cited 2016 Feb 3]. Available from: https://ww2.camrt.ca/bpg/patientsafety/radiationsafety/breastfeedinginnuclearmedicinenmspecific/
  48. Hricak H, Choi BI, Scott AM, Sugimura K, Muellner A, von Schulthess GK, et al. Global trends in hybrid imaging. Radiology [Internet]. 2010 Nov [cited 2015 Aug 25];257(2):498-506. Available from: http://pubs.rsna.org/doi/full/10.1148/radiol.10100579
  49. Drugs and health products [Internet]. Ottawa: Health Canada. Medical isotopes - frequently asked questions; 2009 Aug 31 [cited 2016 Jan 29]. Available from: http://www.hc-sc.gc.ca/dhp-mps/brgtherap/activit/fs-fi/isotopes-med-faq-eng.php
  50. Couillard F. Anticipating a problem, proposing solutions [Internet]. Ottawa: Canadian Association of Medical Radiation Technologists; 2014 Nov 25. [cited 2016 Feb 3]. Available from: http://www.camrt.ca/blog/2014/11/25/anticipating-a-problem-proposing-solutions/
  51. Weber WA. PET/MR imaging: a critical appraisal. J Nucl Med [Internet]. 2014 May 12 [cited 2016 Jan 29];55(Suppl 2):56S-8S. Available from: http://jnm.snmjournals.org/content/55/Supplement_2/56S.long
  52. Positron emission tomography in oncology [Internet]. Ottawa: CADTH; 2009 Aug. [cited 2016 Feb 1]. (Policy forum: health technology policy information). Available from: https://www.cadth.ca/collaboration-and-outreach/advisory-bodies/policy-forum/policy-information-documents/positron-emission-tomography-oncology
  53. Delso G, Voert ET, Barbosa FG, Veit-Haibach P. Pitfalls and limitations in simultaneous PET/MRI. Semin Nucl Med. 2015 Nov;45(6):552-9.


APPENDICES

APPENDICES

 

 

 

 

 

We would like your feedback