Signs of Inequitable Access: Users of Private Physiotherapy Services Do Not Reflect the Urban Population in Winnipeg, Manitoba

Abstract

Purpose: Both private and public funding cover outpatient physiotherapy (PT) in Canada. Knowledge is lacking in who does and does not access PT services, which limits the ability to identify health/access inequities created by current financing structures. This study characterizes the individuals accessing private PT in Winnipeg to better understand whether inequities exist, given the very limited publicly financed PT. Methods: Patients attending PT in 32 private businesses, sampled for geographic variation, completed a survey online or on paper. We compared the sample’s demographic characteristics with Winnipeg population data using chi-square goodness-of-fit tests. Results: In total, 665 adults accessing PT participated. Respondents were older and had higher levels of income and education compared to Winnipeg census data (p < 0.001). Our sample included higher proportions of female and White individuals, and lower proportions of Indigenous persons, newcomers, and people from visible minorities (p < 0.001). Conclusions: There are signs that inequities exist in access to PT in Winnipeg; the cohort who access private PT services does not reflect the wider population, which suggests that some segments of the population are not receiving care.

Physiotherapy (PT) can positively affect general health outcomes by reducing pain and disability, improving physical functioning, and increasing quality of life; however, some Canadians experience difficulties accessing outpatient rehabilitation services.^1–5 For example, a study of 97 hospitals in Quebec from 2014 to 2017 found that 57% of sites reported an increase in PT referrals, while 36% observed a decrease in PT staff, and 38% noted an overall reduction in access.⁴ When the demand for publicly funded outpatient PT consistently exceeds resources,⁶ prolonged wait times that negatively affect clinical outcomes result.⁷ Affordability is another important factor affecting access, as physiotherapists in Canada provide services in both publicly and privately financed settings. This may mean that Canadians with limited income and/or without insurance could experience difficulties accessing PT services. Accessibility of services relates to health equity: “that all people can reach their full health potential and should not be disadvantaged from attaining it because of their race, ethnicity, religion, gender, age, social class, socio-economic status or other socially determined circumstances.”^8(p.2)

Access to PT is influenced by both geographic and socio-demographic factors.^9–11 There are variations in the distribution of physiotherapists relative to PT use across Canada, creating unequal access in rural and remote regions.¹¹ Intra-urban access to PT is also not uniform. A recent study conducted in Saskatoon, Saskatchewan, found that some socio-economic factors were associated with access.¹⁰ For example, more newcomers lived in areas with fewer outpatient physiotherapists than in other neighbourhoods.¹⁰ Similarly, a greater proportion of people with no post-secondary education lived in areas with less access. This was not true for Indigenous peoples or those from low-income households; they were equally likely to live in areas with relatively low or relatively high access to community-based PT.

Facing a need to reduce costs as mandated by the provincial budget, the Winnipeg Regional Health Authority (population 784,441)¹³, closed outpatient PT departments at seven city hospitals and eliminated two-thirds of the publicly funded PT positions at the remaining hospitals in 2017.^14,15 In juxtaposition with this action, a 2018 position statement on health equity acknowledged the Manitoba government’s responsibility to “improve the availability, accessibility, acceptability and quality of health services with a focus on structurally disadvantaged populations.”¹² It specifically noted that Indigenous peoples, newcomers and refugees, people living with disabilities, and people with low income (among others) experience poor health outcomes. Currently, adults only qualify for PT at the one hospital providing service for the region if they require rehabilitation for a very short list of complex upper extremity, spinal, and joint revision surgeries.^15,16 People with common musculoskeletal conditions (e.g., neck or back pain, arthritis, soft tissue and overuse injuries) or those who have undergone common surgeries, such as hip or knee arthroplasty, are not eligible to receive service.¹⁷ Alternative publicly funded outpatient PT is available through a limited number of Access Centres (primary health clinics) and a small number of charitably funded services;¹⁷ however, most of these practices offer only consultations and focus on home programmes for treatment. As well, almost all have either diagnostic or geographic limitations (e.g., for people rostered to specific primary care clinicians or with neurological conditions) and many have lengthy wait times.¹⁷ When similar PT cuts were pursued in Ontario in 2005, people who lost access were those with lower levels of self-reported health; those who accessed PT did so by paying out of pocket, via private insurance, or by accessing the remaining publicly financed services.¹⁸

Limiting access to PT affects the individual, the health care system, and society more broadly. Numerous recent systematic reviews demonstrate that treatments provided through outpatient PT are effective in reducing pain and improving function for people with a range of conditions including adhesive capsulitis¹⁹ and knee osteoarthritis (OA).²⁰ There is support for physiotherapists helping patients manage pain without the use of opioids.²¹ Other research supports the cost-effectiveness of PT to health care systems and society in general (when evaluating total costs, including lost productivity) for conditions such as neck pain, chronic low back pain, OA, intermittent claudication, and Parkinson disease.²²

With the 2017 public system cuts in the Winnipeg Regional Health Authority, the government claimed PT services would transfer to the private sector; however, this assertion ignores potential barriers to access. Despite promises made to evaluate the effects of these cuts, the government has not released any analyses or reports on the topic. The purpose of this study was to characterize individuals accessing private PT in Winnipeg to better understand who does and does not receive PT through the private system. In the absence of a coordinated centralized data repository that collects information about patients receiving care, we compared the demographics of individuals accessing private PT to population data.

Methods

Study design

We employed a cross-sectional survey that was administered primarily using the online tool SurveyMonkey (SurveyMonkey, San Mateo, CA). Ethics approval was obtained from the Health Research Ethics Board at the University of Manitoba in Winnipeg.

Recruitment

PT service providers

To recruit patients, we first required consent from private service providers to approach their patients. Publicly available information regarding service providers in Winnipeg is listed on the College of Physiotherapists of Manitoba Web site.²³ We categorized the 158 businesses according to five geographical areas of the city plus one category for mobile services (see Table 1). For practical reasons, we anticipated that we would be able to collect data from 20% (n = 32) of the providers. We used a random number generator to assign an identification number to each service provider. We then sorted the service providers in each area from the smallest to largest identification numbers and contacted them (in this order) to invite participation. We aimed for a proportional representation of service providers from each of the six categories. During the conversation with each clinic, we confirmed eligibility by making sure that the clinic had at least two physiotherapists seeing patients during the designated data collection time period. This inclusion criterion was used to maximize data collection during the time research assistants (RAs) were available to attend the clinics. Clinic owners were approached by phone/email until 32 service providers agreed to participate and provided written consent.

Table 1 .

Distribution of Eligible and Participating Clinics

	Central	Northeast	Northwest	Southeast	Southwest	Mobile	Total
No. (%)	13 (8.2)	30 (19)	21 (13.3)	27 (17.1)	47 (29.7)	20 (12.7)	158 (100)
Ineligible	9	8	6	11	23	7	64
Did not respond - unknown eligibility	2	5	2	5	7	4	25
Eligible - lost contact	0	1	1	0	1	3	6
Eligible - declined to participate	0	3	4	3	2	2	14
Eligible - participated	2	7	3	4	13	3	32
Clinics not contacted because they were lower on the list	0	6	5	1	0	0	12
Not contacted because owner declined affiliated clinic(s) participating	0	0	0	2	0	0	2

Patients attending private PT clinics

An RA attended each clinic (between October and December 2019) for five hours on two consecutive days, when possible, for a total of 10 hours of data collection per clinic. The RA confirmed patients were there to see a physiotherapist and were at least 18 years of age, briefly described the study, and asked for their participation. If interested, the patient filled out the survey on the RA’s iPad, used their own device via the QR code link, or completed it on paper. Clinic staff provided information about the total number of PT patients who attended during the data collection period.

Patients using mobile PT services

We delivered study information sheets, paper surveys, and the survey QR link to the owners of mobile PT businesses, who then distributed the information to their physiotherapists and patients.

Survey instrument

Three PT researchers developed the survey using questions and response categories from the 2016 Canadian census.²⁴ Pilot testing ensured question clarity and determined the time commitment for the survey (5–10 min). The final survey had 17 questions that focused on demographics (see online Appendix 1). We also collected information about the reason for accessing PT services, method of payment, and self-reported health. All responses were anonymous.

Data and statistical analysis

We exported SurveyMonkey data to Microsoft Excel 2016 (Microsoft Corporation, Redmond, WA) and added the survey data from the responses on paper. Data were reviewed for consistency and any questions that allowed an “other, please specify” response were reviewed and placed in one of the existing categories, if appropriate. Five respondents reported receiving PT services in a personal care home, which we categorized as “at home.” Modes were imputed for missing demographic information²⁵ (age, sex, visible minorities/identity, immigration, language, education, income) and the health status question. All values for these variables were calculated out of the total number, 665. All variables had less than 7% missing data. We used the most current Canada Post map of Winnipeg to determine whether each patient’s home postal code (first three digits) was in the same region as the clinic they attended.

Using SigmaPlot, version 12.5 (Systat Software, San Jose, CA) and IBM SPSS Statistics, version 25 (IBM Corporation, Armonk, NY), we examined data for normality, and generated descriptive statistics (mean ± SD; median and inter-quartile range [IQR]; proportions/percentages). We used chi-square goodness-of-fit tests to compare the sample’s demographic characteristics and health status with Winnipeg population data.^24,26 Also, we calculated standardized residuals when the chi-square result was significant (p < 0.05) and considered standardized residuals that were greater than 1.96 to indicate important differences.^{27(p. 419)} In secondary analyses, we used Mann-Whitney U-tests to compare the number of PT visits among respondents who received PT in a private clinic vs. through a mobile service.

Results

Of the 665 surveys received, 641 were submitted electronically, and 24 came from paper copies. Approximately 1,188 patients attended an appointment with a participating PT provider during the data collection period, resulting in a response rate of about 56%. The proportion of patient responses generally reflected the distribution of clinics in Winnipeg (Central, 6.8 vs. 8.2%; Northeast, 19.5 vs. 19.0%; Northwest, 9.3 vs. 13.3%; Southeast, 15.2 vs. 17.1%) except for the Southwest region, which was overrepresented (40.8 vs. 29.7%), and mobile services, which was underrepresented (5.7 vs. 12.7%). In total, 94% of the 663 respondents who specified clinic vs. mobile PT visits received PT care in a clinic, and 6% accessed a mobile service. One-hundred and fifty-six respondents attended a clinic within the postal code region where they resided, 177 participants travelled to an adjacent region, and 238 respondents attended a clinic even farther from home.

In terms of demographics, our sample included more female (58%) and less male (42%) respondents compared to census data (49% female, 51% male) ( ${\text{χ}}_1^2$ = 24.5, p < 0.001). Patients receiving private PT services largely self-identified as White (87%) and less often identified as a visible minority (11%), or Indigenous (2%). These findings were significantly different from Winnipeg census data (60% White, 28% visible minority, 12% Indigenous) ( ${\text{χ}}_2^2$ = 210.4, p < 0.001). Three per cent of our survey respondents reported immigrating to Canada in the last five years, compared to 8% in census data ( ${\text{χ}}_1^2$ = 27.5, p < 0.001). In general, the participants were older (Figure 1), had higher education levels (Figure 2), and reported higher incomes (Figure 3) than population data.

Figure 1 .

Percentages of respondents in age groups compared to Winnipeg census data.

* = comparison with standardized residual > |1.96| and considered an important difference.^{27(p. 419)}

Figure 2 .

Percentages of respondents’ education levels compared to Winnipeg census data.

* = comparison with standardized residual > |1.96| and considered an important difference.^{27(p. 419)}

Figure 3 .

Percentages of respondents’ household income before tax compared to Winnipeg census data.

*comparison with standardized residual > |1.96| and considered an important difference.^{27(p. 419)}

The sources of payment for the PT visits varied. The majority of those surveyed had insurance to cover costs. PT for 67 survey respondents was covered by the Workers Compensation Board of Manitoba (WCB); 51 respondents were covered by Manitoba Public Insurance (MPI) due to a motor vehicle accident; and 396 used private insurance. Almost one-fifth of the participants paid out of pocket (117/631 respondents who provided payment data). Most people paying out of pocket either did not have private insurance (n = 62) or their private insurance had run out (n = 47). A small number reported they were saving their insurance for other services. About one in 10 participants reported not having any private insurance. The number of PT visits in the past year was variable (1–156 visits) but was higher for those paying out of pocket compared to participants who had insurance (WCB, MPI, private) (median [IQR] was 8 [11] vs. 5 [8] respectively, U 24,027.5, p = 0.002).

Because the census data did not include a measure of self-reported health, we compared self-rated health to that documented in the Winnipeg Health Region in the 2015–2016 Canadian Community Health Survey (CCHS).^{26(p. 156)} Analysis of standardized residuals revealed that all categories were significantly different ( ${\text{χ}}_3^2$ = 44.3, p < 0.001), with a smaller proportion of survey participants reporting “excellent health” (13% of 665 respondents vs. 22% from CCHS data); a larger proportion answering with “very good or good health” (77% vs. 65%); and a smaller proportion stating they were in “fair or poor health” (9% vs. 12%).

Twenty-four individuals in our sample came from low-income households (< $36,084/year)²⁴ and paid out of pocket for PT services. As such, they represented 21% of the 117 people in our study who paid out of pocket. In contrast, only 11% of people who used insurance (WCB, MPI, private) to pay for PT services (n = 503) were from low-income households.

Most individuals were seeking PT services to treat a “bone, joint, muscle, ligament or tendon problem” or for rehabilitation after surgery (Table 2). Users of mobile PT services (n = 38) reported a much higher number of treatment sessions in the previous year compared to those who attended a private clinic (median [IQR] was 30 [36], vs. 5 [8] respectively (U 3241.5, p < 0.001)]. Greater proportions of mobile service users paid out of pocket (33%, 12/36) and reported lower income (44%, 12/27) compared to the reference group of clinic users (18%, 105/593 and 11%, 68/595) respectively, ${\text{χ}}_1^2$ = 6.0, p = 0.014 and ${\text{χ}}_1^2$ = 29.1, p < 0.001).

Table 2 .

Reasons for Seeking Physiotherapy Care (N = 665)

Reason for PT	No. (%)*
Rehabilitation after surgery	101(15.2)
Bone, joint, muscle, ligament, or tendon condition	499(75.0)
Neurological condition (e.g., stroke, multiple sclerosis)	28(4.2)
Cardiovascular or respiratory condition (e.g., emphysema, asthma)	1(0.2)
Other	23(3.5)

^*Does not total 100% because 23 individuals chose more than 1 response option, and 36 surveys were missing data.

Discussion

Respondents who accessed private PT services in Winnipeg during our data collection period were generally older and had higher levels of income and education compared to the distributions reported in census data. Our sample also included higher proportions of female and White individuals, and lower proportions of Indigenous persons, newcomers, and people from visible minorities. An analysis of users of ambulatory PT and occupational therapy in the United States similarly found that the typical patient was a White female from a middle- or high-income household.²⁸ For the reasons we outline here, we interpret this pattern in our data as suggestive of inequities in access to private PT.

Shah and colleagues¹⁰ identified inequities in the distribution of private and publicly funded outpatient PT clinics by neighbourhood for newcomers and people with no post-secondary education in Saskatoon, Saskatchewan.¹⁰ Almost half of all private PT clinics in Winnipeg are located in the south end of the city,²³ where there are fewer areas of poverty compared to the city centre, as identified by the City of Winnipeg.²⁹ While data describing specific PT needs do not exist, the 2019 Winnipeg Health Region Community Health Assessment reported higher rates of chronic diseases, such as OA and diabetes, in people living near the city centre compared to those living in the south end.^{26(p. 278, 310)} The relative lack of clinics in the city centre, where the greatest proportion of high poverty areas exist, may have contributed to fewer people of lower socio-economic status attending PT in our study. This reduced availability in lower income neighbourhoods may or may not be a geographic access issue. Only 25% of PT clients sought services within the same postal code region as their home, so proximity to services may have only a small influence. However, people with lower incomes are more likely to face transportation barriers and may find it easier to access clinics in their neighbourhood versus travelling to other parts of the city.

A smaller percentage of our survey respondents reported excellent health compared to that recorded in Winnipeg Health Region data (13% vs. 22%). This is not unexpected, given that patients were presumably seeking PT to address a health issue. However, our sample also included a larger proportion of individuals reporting “very good” or “good” health, and a smaller proportion of individuals reporting “fair” or “poor” health. This may reflect inequities in who is able to access PT care through the private system. This resonates with Landry and colleagues’ study in Ontario,¹⁸ which found that people with the poorest self-reported health were the least likely to access PT after publicly funded centres were closed.

Availability of services and patients’ ability to reach and pay for health care are among the dimensions that influence access.³⁰ The cost of a private PT visit, coupled with expenses related to transportation, may be too much for individuals with lower incomes. As of 2019, a “customary” PT assessment ranged from $77 to $81 in Manitoba, and each subsequent treatment session cost $62 to $66,³¹ although fees in many of the clinics in Winnipeg were higher. Private PT services are for-profit businesses designed to allow physiotherapists to earn a living while providing treatment to patients. In a 2016 literature review of ethical issues in private practice PT, conflicts of interest about what is best for the financial welfare of the business versus what is most appropriate for patients was the most common concern that therapists raised.³² It is interesting to note that while people with lower incomes were under-represented in our sample of those who attended private practice PT, CCHS data indicated that income was not associated with difficulty accessing publicly funded first-contact and specialized health care services.³³ This discrepancy supports the idea that affordability may be a barrier for some, and that private services cannot and do not exactly replace those that are publicly funded.

Although the number of PT visits attended in the past year by individual respondents was highly variable in our sample, people paying out of pocket reported attending more frequently compared to participants who had any type of insurance (8 vs. 5 visits). This indicates that people are willing to pay privately when they need PT services, and it may also be indicative of the limited number of PT visits covered by many insurance plans. One unexpected finding was that individuals who used mobile PT services reported having a much greater number of visits within the previous 12 months compared to those who attended a private clinic (30 vs. 5 visits). In exploring this further, we discovered a greater proportion of mobile service users paid out of pocket and were in the low-income category compared to people attending a clinic. This may suggest that these individuals find PT essential for them to maintain function and live independently. These individuals and/or their families found a way to pay for services despite their low income. Further research using qualitative methodology is needed to learn more about these people, and to understand how and why they are paying for extensive PT treatment.

Our study results support the concern that there may be inequity in access to private PT services, with a subsequent worsening of health disparities, particularly in the absence of comparable publicly or charitably funded services as found in a recent environmental scan.¹⁷Research supports the provision of PT treatment to improve function and reduce pain in individuals with specific conditions^19,20,34,35 and to provide cost-effective care in our health care systems.²² In addition to lobbying for revisions to provincial funding models, physiotherapists and their professional organizations can lobby third-party payers to increase insurance benefits and employers to provide reasonably priced insurance plans for their employees. Physiotherapists in private practice should consider strategies to increase access to their services; for example, by offering reduced rates on group exercise and education classes, and tele-rehabilitation to reduce patients’ transportation costs and time lost from work. As health care professionals, all physiotherapists should be aware of issues related to equity and be familiar with initiatives and organizations in their province that support quality health care.

Our study had several limitations. The survey was only available in English, and only adult patients were included. We chose the latter because PT services for those under 18 years of age are offered through multiple publicly funded channels, including schools and children’s rehabilitation centres. Because of our delay in requesting approval from the First Nations Health and Social Secretariat of Manitoba, our inability to ask a specific question about Indigenous identity may have resulted in some inaccuracy in that variable. Clinic and patient participation were voluntary; therefore, response bias may be an issue. Due to the cross-sectional nature of our study, we do not know if the characteristics of patients accessing private PT services have changed since the cuts to publicly funded services in 2017. We also do not know if patients using publicly funded outpatient services before the cuts differed from census data. To our knowledge, comprehensive data repositories specific to PT services do not exist. This lack of data hampers research and the ability of health regions to plan for the provision of effective, equitable health care. Our data were collected in Winnipeg, and findings may not be generalizable to other urban centres, depending on population characteristics and existing funding models.

Conclusion

It is considered an essential competency for physiotherapists to demonstrate leadership and “advocate for accessibility and sustainability of physiotherapy” (Essential Competency 5.1).^36(p.15) This study provides information about segments of an urban population who are not accessing outpatient PT through private businesses, in a city where only very limited publicly funded services are available. People who do not identify as White, are newcomers to Canada, and those with lower education and/or income do not access private PT services on par with their representation in the community.

These inequities in access may negatively affect health, if not addressed by publicly funded services in the health care system. Despite position statements recognizing government responsibility to address health equity, in some jurisdictions in Canada, publicly funded outpatient PT services have been actively reduced. It is essential to continue addressing access challenges through research, lobbying, and implementing the strategies known to improve patient-centred accessibility to PT care.

Key Messages

What is already known on this topic?

Reductions to publicly funded PT services have occurred in recent years, with the assumption that those needing care can access private providers. Affordability of services is a known barrier to accessing PT, and the removal of publicly funded care may result in some segments of the population being unable to access PT, deepening existing health inequities.

What this study adds

Individuals accessing private PT services do not reflect the general population. Users were more likely to be White, have a higher education, and report higher incomes compared to the population of the overall health region. Reliance on private PT services may mean inequitable access to care for some groups of people in the absence of a robust public system offering comparable services.