Perceived barriers to primary care among western Canadians with chronic conditions

Paul E. Ronksley, Claudia Sanmartin, David J.T. Campbell, Robert G. Weaver, G. Michael Allan, Kerry A. McBrien, Marcello Tonelli, Braden J. Manns, Deirdre Hennessy and Brenda R. Hemmelgarn

About a third of Canadians have chronic conditions such as diabetes, hypertension and heart disease. The care of patients with chronic conditions has become a major focus in health services research and public health policy,Note1-3 and  the direct costs of managing these conditions exceed $40 billion a year.Note4-8 Even so, not all Canadians with chronic conditions receive the care they require.Note9-14

Previous research suggests that approximately one in seven adults with a chronic condition reports  an unmet health care need—a common indicator of problems with access to services.Note15-21 Information about the nature of these unmet needs is limited, particularly whether they pertain specifically to chronic disease care and whether they are associated with access to primary care physicians. Given that many chronic conditions are managed largely in outpatient settings, identifying perceived obstacles to primary care is important in determining which barriers are potentially modifiable.

Based on the results of a population-based survey conducted in the four western provinces, this analysis examines the prevalence of barriers to the receipt of care from primary care physicians and to other health services among adults aged 40 or older who reported having been diagnosed with diabetes, heart disease, hypertension and/or stroke. Socio-demographic factors associated with barriers to care are also explored.

Data and methods

Data sources

The data are from the Barriers to Care for People with Chronic Health Conditions (BCPCHC) survey, administered by Statistics Canada and designed by the Interdisciplinary Chronic Disease Collaboration (www.ICDC.ca).Note22,Note23 People aged  40 or older residing in the four western Canadian provinces (Manitoba, Saskatchewan, Alberta and British Columbia) who responded to the 2011 Canadian Community Health Survey (CCHS), and who reported having been diagnosed with diabetes, heart disease, hypertension and /or stroke, were eligible for inclusion. Members of the Canadian Forces, First Nations people living on reserves, and people in institutions are not covered by the CCHS, and thus, were not eligible for the BCPCHC. Computer-assisted telephone interviews were conducted by Statistics Canada in 2012. Of the 2,316 individuals selected for the BCPCHC, 1,849 (80%) completed the survey. With permission, their responses were linked to their 2011 CCHS data, which provided detailed demographic and health-related information.

Variables

Respondents were asked if, during the previous 12 months, they had needed the services of a family physician or general practitioner for their chronic condition(s). Those who replied affirmatively were asked if they had difficulty obtaining these services, and if so, what kind of difficulty they encountered. Difficulties were classified into four categories: initiating care (contacting a physician/getting an appointment); waiting for an appointment; in-office waiting; and other barriers  (Text Table 1).

Respondents were also asked if there had been a time in the previous 12 months when they needed care for their chronic condition(s) but did not receive it. Those who reported a general unmet need for chronic disease care (which could include services outside a family physician’s office, such as specialist care) were asked why they did not receive care. Responses were classified into two types of barrier: system level (resource availability) or patient level (accessibility, acceptability, or personal choice)Note17 (Text Table 2). Respondents reporting an unmet need for chronic condition care were asked where they had sought care (doctor’s office versus facility such as hospital, out-patient clinic or emergency department), and if they felt their health worsened because this need was not met. Although the present analysis focuses on barriers to primary care, the data about barriers to care in general provide additional information about where patients with chronic conditions seek care and the obstacles they confront.

Socio-demographic and health behaviour variables included in this study were based on the Health Behavior Model,Note24 a framework of determinants of health service use that encompasses predisposing factors, enabling factors, personal health factors, and health care system/environmental factors. The variables in the present analysis are: age, sex, marital status, education, household income, ethnicity, province of residence, rural/urban location, having a regular medical doctor, perceived health status, and body mass index (BMI). BMI categories were calculated from self-reported height and weight, using an adjustment for self-report.Note25

Analysis

The percentages of respondents reporting key variables were calculated and stratified by the type (diabetes versus other) and number (one versus two or more) of chronic conditions. All descriptive statistics were weighted to reflect the population aged 40 or older in the four western provinces, who had the selected chronic conditions. These frequency weights were calculated by Statistics Canada based on weights from the 2011 CCHS and were readjusted for non-response by province, chronic condition, and age group to represent the target population. Bootstrapping techniques with 500 replications were used to calculate standard errors and confidence intervals (CI) around the estimates. To determine the reliability of the reported percentages, the coefficient of variation (CV) was calculated. Consistent with Statistics Canada guidelines, all estimates with a CV greater than 33.3% were considered unreliable and were not reported.Note26 Log-binomial regression was used to determine associations between socio-demographic/patient factors and difficulty getting primary care. Unadjusted and age-adjusted prevalence rate ratios (PRR) were calculated for each covariate. For all statistical tests, p< 0.05 was considered significant. All analyses were conducted at the Prairie Regional Data Centre in Calgary, Alberta, using STATA 11.0 (Statacorp, College Station, Texas). This study was approved by the Conjoint Health Research Ethics Board of the University of Calgary and the Health Research Ethics Board of the University of Alberta.

Results

Study population

Approximately half (51%) of the 1,849 people who completed the survey were aged 65 or older (Table 1). Respondents were equally divided among men and women. Almost all respondents (95%) had a regular medical doctor. Two-thirds of respondents had only one chronic condition; the remaining third had at least two. Hypertension was reported by  82% of respondents; almost two-thirds of this group did not have any of the other conditions examined in this analysis. Compared with respondents who had only one condition, those with multiple conditions tended to be older, to be obese, and to have less than secondary school graduation. Even though all respondents had at least one of the four chronic conditions examined in this study, 37% reported their health as excellent or very good.

Reported need for and barriers to primary physician care

More than three-quarters (78%) of respondents reported that, because of their chronic condition(s), they had required the services of a primary care physician in the past year (Table 2). The percentage was higher among those with multiple conditions and among those who had diabetes (both 85%). Overall, 10% of the respondents who accessed primary care reported difficulty obtaining these services; for those with diabetes, the percentage was higher at 16%.

Among the 10% of respondents who reported difficulty getting primary care, the barriers mentioned most frequently were related to the initiation of care (54%) and waiting to get an appointment (38%).  Barriers related to initiating care were more common among respondents with only one chronic condition (60%), whereas waiting to get an appointment was more common among those with multiple chronic conditions (49%).

Factors associated with difficulty getting primary physician care

Being younger than 65 and having diabetes were significantly associated with difficulty getting primary physician care (Table 3). Patients younger than age 65 were almost three times as likely to report a barrier (PRR 2.9; 95% CI: 1.7-5.0), compared with seniors.  When age was taken into account, respondents with diabetes were twice as likely as respondents reporting any of the three other selected conditions to report a barrier to primary care (PRR 2.2; 95% CI: 1.3-3.9).

Unmet needs

Approximately 5% of respondents reported having  an unmet need for care in general related to their chronic condition(s) in the previous 12 months (Table 4). This type of unmet need was most prevalent among respondents with multiple conditions (7%) and among those with diabetes (8%). Half of respondents who reported an unmet need stated that it was associated with seeking care in a doctor’s office.

While respondents with unmet needs gave a range of personal reasons for not receiving care (for example, too busy, family responsibilities), more than 60% reported a system-level barrier (service not available when required or in the area, or long wait times).

Most (61%) respondents who reported an unmet need felt that their chronic condition became worse as a result of the delayed care. Given the small number of respondents who reported an unmet need in general, it was not possible to identify socio-demographic/patient factors associated with this outcome.

Discussion

According to the results of the Barriers to Care for People with Chronic Health Conditions survey, the majority of adults in western Canada who had at least one chronic condition required the services of a primary care physician in the previous year and had no difficulty receiving it. Nonetheless, approximately one in 10 reported a barrier. Barriers were reported most frequently by respondents with diabetes, and tended to be related to initiating care or waiting too long to receive it. As well, about 5% of respondents reported an unmet need for care in general, and more than half of this group considered that their condition had worsened because of  this unmet need.

This analysis complements earlier studies using population-based survey data.Note15,Note17-21 Prior estimates have reported that somewhat higher percentages—12% to 15%—of Canadians with chronic conditions had an unmet health care need in the past year.Note15,Note18,Note19,Note21,Note27,Note28 However, those studies did not distinguish between the need for primary versus specialist care, whereas the present analysis focuses on primary care physicians.

In this study, the barriers most commonly cited involved systemic factors such as wait times and service availability. However, these results are not limited to the services of primary care physicians, but rather, refer to any component or service involved in chronic disease care. Previous research indicates that barriers to specialist care may explain why long wait times and service availability issues are most prevalent among chronic disease groups.Note27,Note29 Even so, the present analysis also reveals perceptions of barriers to care from primary care physicians, especially among patients with more than one chronic condition. A study published in 2010 suggested that the delivery of health care by multiple providers in multiple locations may partially explain why access-related barriers are most commonly reported by patients with more than one chronic condition.Note30

Only a small percentage of adults with chronic conditions reported an unmet health care need. Nonetheless, earlier studies have shown that unmet health care needs related to resource availability are associated with an increased risk of all-cause hospitalization and higher rates of emergency department visits.Note15,Note18,Note31,Note32

Limitations

This study should be interpreted in light of its limitations. First, as is the case with all self-reported data, issues related to the reliability and validity of the responses can arise. Second, the small sample limited the extent to which data could be stratified by socio-demographic variables, and the wide confidence intervals surrounding some estimates make interpretation difficult. Third, it was not possible to measure the influence of local health care availability (for example, the ratio of primary care physicians to population) on reported barriers to primary care. Because the analysis was restricted to four chronic conditions, the generalizability of the findings could be limited. Respondents may have had other chronic conditions that differentially affect reported barriers. Finally, it is possible that the results are confounded by differing expectations of what should be available from the health care system and may depend on the type of barrier experienced. Additional studies are needed to explore how respondents interpret this survey question and to validate the findings.

Conclusion

The findings of this analysis should be considered to be hypothesis-generating and as input for subsequent  studies. The majority of people aged 40 or older with one or more of the four selected chronic conditions (diabetes,  heart disease, hypertension and stroke) reported that they had required the services of a primary care physician during the past year and had no difficulty receiving this care. A small percentage reported barriers, notably, problems related to contacting a physician or waiting too long to receive care. Future research might examine specific interventions developed to address these barriers, such as the expansion of primary care hours and increased numbers of after-hours clinics, and whether they are associated with improved outcomes. As well, further work might be conducted to determine if the findings of this study can be extrapolated to patients with other common chronic conditions. An examination of how patients who perceive barriers to care navigate the health care system would also provide insight into the association between reported gaps in care and health outcomes.

Acknowledgements

The Interdisciplinary Chronic Disease Collaboration (ICDC) is funded through the Alberta Heritage Foundation for Medical Research (AHFMR) Interdisciplinary Team Grants Program. AHFMR is now Alberta Innovates – Health Solutions (AI-HS). This research was supported by an interdisciplinary team grant from AI-HS, the Interdisciplinary Chronic Disease Collaboration. Paul E. Ronksley is supported by a Postdoctoral Fellowship award from the Canadian Institutes of Health Research. David J.T. Campbell is supported by an AI-HS Clinician Fellowship award. Braden J. Manns and Brenda R. Hemmelgarn are supported by AI-HS salary awards. Brenda R. Hemmelgarn is also supported by the Roy and Vi Baay Chair in Kidney Research. G. Michael Allan is supported by a knowledge translation grant from the Alberta College of Family Physicians. Marcello Tonelli is supported by a Canada Research Chair. The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or writing the manuscript. G. Michael Allan, Kerry A. McBrien, Marcello Tonelli, Braden J. Manns and Brenda R. Hemmelgarn were supported by an alternative funding plan from the Government of Alberta and the Universities of Alberta and Calgary.

Date modified: