Room for improvement

Abstract

OBJECTIVE

To investigate variations in accessibility, continuity of care, and coordination of services as experienced by patients in Quebec on the eve of major reforms, and to provide baseline information against which reforms could be measured.

DESIGN

Multilevel cross-sectional survey of practice.

SETTING

One hundred primary health care settings were randomly selected in urban, suburban, rural, and remote locations in 5 health regions in Quebec.

PARTICIPANTS

In each clinic, we chose up to 4 physicians and 20 consecutive patients consulting each physician.

MAIN OUTCOME MEASURES

Patients’ responses to a self-administered questionnaire, the Primary Care Assessment Tool, that assessed patient-provider affiliation, accessibility, relational continuity, coordination of primary and specialty care, and whether patients received health promotion and preventive services.

RESULTS

A total of 3441 patients participated (87% acceptance rate) in 100 clinics (64% response rate). Timely access was difficult; only 10% expressed confidence they could be seen by their regular doctors within a day if they became suddenly ill. Average waiting time for a doctor’s appointment was 24 days. Coordination of care with specialists was at minimally acceptable levels. Patients with family physicians recalled them addressing only 56% of the health promotion and preventive issues appropriate for their age and sex, and patients without family physicians recalled physicians addressing substantially fewer (38%). Most patients reported they were highly confident that their physicians knew them well and would manage their care beyond clinical encounters (relational continuity). The exception was the 16% of patients overall who did not have family physicians (34% of patients at walk-in clinics).

CONCLUSION

This survey highlights serious problems with accessibility. Improvement is needed urgently to avoid deterioration of patients’ confidence in the healthsystem even though patients rate their relationships with their physician highly. Health promotion, preventive services, and coordination with specialists also needed to be improved, and careful thought must be given to the plight of those without family physicians.

In 2000, the Commission d’étude sur les services de santé et services sociaux¹ found Quebec’s health care system deficient with regard to accessibility, continuity, and coordination of care. Like other such commissions, this one identified better organization and delivery of primary care as the key solution.^1–5 The First Ministers’ Health Accord in 2000 allocated $800 million over 5 years to accelerate change in primary care,⁶ and in 2002, Quebec’s Ministry of Health and Social Services created and accredited family medicine groups (FMGs) to strengthen primary care.

Quebec is divided into 170 subregions, each of which is associated with a Centre local de services commun-autaires (CLSC), or local community health centre, that provides health and social services to a defined area. The scope of services varies with the needs of the population, and many metropolitan CLSCs do not offer general medical services. Approximately 15% of primary care services are provided through CLSCs and the rest are provided through physician-run clinics and private practices that are the norm elsewhere in Canada.

Although Statistics Canada has provided information on access to health services and the proportion of Canadians who have family physicians,^7,8 the Canadian Community Health Survey did not provide information in enough detail to permit assessment of primary care overall. The purpose of our study was to survey primary care sites in Quebec to investigate patients’ experiences with primary care before reforms in order to establish a baseline against which reforms could be measured.

METHODS

We conducted a cross-sectional, multilevel survey of primary care settings, physicians, and patients from December 2001 to October 2002 in 5 health regions of Quebec: Montreal, Montérégie, Bas-Saint-Laurent, Côte-Nord, and Gaspésie. We used a listing of primary care clinics compiled in 2000⁹ and the 170 CLSCs to classify sites into urban, suburban, rural, and remote geographic areas. Then we used systematic random sampling to ensure that within each geographic area our study sample had 10% solo practices, 65% group practices, and 15% CLSCs.

Eligible sites were those that offered general medical services to an undifferentiated clientele and had not undergone major changes in the last year. An additional eligibility criterion, added after data collection began, was that at least two thirds of clinic patients could respond in English or French. Within each clinic, depending on size, we selected up to 4 physicians who had practised general medicine there at least 1 day a week for at least 1 year and recruited 20 consecutive patients consulting each physician on a certain clinic day. We chose representative days; for example, if a clinic offered walk-in and scheduled care, we sampled patients from both. Data collection sometimes spanned more than 1 day; collection was stopped after 5 half-days if 20 patients were not recruited.

Eligible patients were those able to complete a self-administered questionnaire in English or French (translation validated). Parents or guardians of children younger than 18 years responded to questionnaires about their children’s care. Research technicians recruited patients on-site and administered the questionnaire face-to-face if patients had difficulty filling it in by themselves (n = 216, 6%). Most patients completed the questionnaires in the waiting room, but some (37%) opted to mail back their responses later. Sensitivity analysis showed no systematic differences in patients’ reported experiences by method or timing of administration.

Patients’ experiences were assessed principally through the Primary Care Assessment Tool (PCAT),¹⁰ which measures strength of patient-provider affiliation, degree of confidence in being seen within a day for a new problem (first-contact accessibility), confidence that providers know the medical history and personal situation and will manage care on an ongoing basis (relational continuity) and work well with specialists (coordination of care), and patients’ recall of having received preventive care. Responses on the PCAT can be valid even if patients do not have regular providers. The questionnaire included the scale for organizational accessibility from the Primary Care Assessment Survey.¹¹ All questions related to patients’ regular care providers or, for those without regular providers, physicians or clinics being consulted that day. Conceptual and operational definitions of the constructs are listed in Table 1.^12,13

Table 1.

Conceptual and operational definitions of minimum expectations for core dimensions of primary health care

DEFINITION OF CONCEPT	OPERATIONAL DEFINITIONS OF MINIMUM EXPECTATIONS
Accessibility: The ability of a person to obtain needed care, including advice and support, from his or her practitioner of choice within a time frame appropriate to the urgency of the problem12	First-contact accessibility: Confidence that patients could get advice or orientation within 1 day from their regular providers if they suddenly got sick and needed care at various times of day (4 items rated on a scale of 1 to 4, expected—3) Organizational accessibility: Rating of office location, hours of operation, wait times for appointments, and ease of contacting the clinic and the regular physician (6 items rated on scale of 1 to 6, expected—4.5)
Relational continuity: A therapeutic relationship with a provider that spans more than 1 episode of care and leads to a sense of clinical responsibility in the provider and accumulated knowledge of a patient’s personal and medical context13	Physician-patient affiliation: Ability to identify a regular physician who knows the patient and takes principal responsibility for his or her health care (3 items rated on a scale of 0 to 4, expected—3) Accumulated knowledge: Confidence that a physician knows a patient’s medical history and personal situation (7 items rated on a scale of 1 to 4, expected—3) Care management: Confidence that a physician will manage care beyond the current clinical encounter (6 items rated on a scale of 1 to 4, expected—3)
Coordination and continuity: Delivery of services by different providers in a timely and complementary manner such that patients’ care is connected and coherent13	Coordination with specialists: Confidence that primary care physician and specialist will collaborate and communicate in patient’s care (8 items rated on a scale of 1 to 4, expected—3)
Comprehensive care: Provision of a full range of services to meet patients’ health care needs, including health promotion and disease prevention and management of common conditions12	Health promotion and preventive care: Recall of physician having addressed at least two thirds of the recommended age- and sex-appropriate strategies to maintain health (list varied for age and sex, expected—67% of strategies “probably” or “definitely” addressed)

Responses to each PCAT item were ratings on a scale of 1 to 4: 1—definitely not, 2—probably not, 3—probably, and 4—definitely. For each dimension scale we calculated the mean of the items. To facilitate interpretation of results, we established the quality standard as a mean of 3.0 (“probably”) for each item as the minimum level that should be expected for each attribute of care. The minimum level represented a quantitative expression of patients’ expectations of care.^14,15

Physicians completed the provider version of the PCAT, which assessed their usual first-contact accessibility, accumulated knowledge of patients, case management responsibility, coordination of care, and provision of preventive services. Physicians’ secretaries provided dates of their third-next available appointments.¹⁶

All descriptive results were weighted (inverse of the sampling fraction for clinics) to be representative of the regions in which clinics were sampled. We present the mean response to each attribute and the proportion of patients who rated it above or below the expected minimum level. We conducted subanalyses to look for the effects of various patient characteristics.

The study received ethical approval from the University of Montreal’s Hospital Research Centre and from several CLSCs that required local approval.

RESULTS

The final participation rate among eligible clinics was 61% (100 of 164) (Figure 1). The study sample included 10 clinics that identified themselves as stand-alone walk-in clinics. A total of 221 physicians contributed an average of 18 (median 20) patients in clinics and 12 patients (median 12) in CLSCs. Of the 4764 patients contacted, 93% were eligible and 89% consented to participate. Only 69% of the 1333 who chose to mail in their questionnaires did so. The final patient response rate among eligible patients was 87% (n = 3441). Table 2 shows the distribution of clinics and characteristics of patients. Patients’ characteristics were typical of those seen in primary care.

Figure 1.

Selection of study clinics

Table 2.

Characteristics of clinics and patients included in the study

CHARACTERISTICS	%
CLINICS (N = 100)
Type of clinic	57
• Group practice
• Solo practice	16
• Stand-alone walk-in clinic	10
• CLSC	17
Geographic location	38
• Urban
• Suburban	22
• Rural	24
• Remote	16
PATIENTS* (N = 3441)
Women	65
Language spoken
• French	76
• English	13
• Many or other	11
Education level
• Primary	16
• Secondary incomplete	26
• Secondary complete	16
• Postsecondary	42
Self-perceived health status
• Poor	5
• Fair	22
• Good	40
• Very good or excellent	33

CLSC—Centre local de services communautaires

^*Mean age of patients was 48.9 years (SD ±18.7 years).

About 77% of patients were consulting their regular providers on the study day. Overall, 16% did not have regular physicians, but in Montreal 22% did not; in walk-in clinics 34% did not. Surprisingly, 44% of walk-in clinic patients identified clinic doctors as the physicians responsible for most of their care; in most cases (85%), these doctors shared that perception.

Table 3 shows mean patient ratings of experiences relative to minimum expected levels of care for the 7 dimensions assessed. Patients’ confidence in first-contact accessibility falls strikingly below the minimum expected level. Only 10.4% expressed confidence they could be seen within a day by their regular providers. Chances were better if patients were sick during clinics’ open hours (67% were confident they would be seen).

Table 3.

Mean level of patients’ experience of core dimensions of primary care

ASPECT OF CARE	MEAN RATING (SD)	% WHO RATED IT BELOW EXPECTED LEVEL
Accessibility
• First-contact accessibility (rated on a scale of 1 to 4, expected—3)	2.21 (0.61)	89.9
• Organizational accessibility (rated on a scale of 1 to 6, expected—4.5)	4.61 (0.89)	38.6
Relational continuity
• Physician-patient affiliation (rated on a scale of 0 to 4, expected—3)	3.44 (1.19)	16.2
• Accumulated knowledge (rated on a scale of 1 to 4, expected—3)	3.28 (0.70)	26.5
• Care management (rated on a scale of 1 to 4, expected—3)	3.26 (0.52)	23.7
Coordination
• Coordination with specialists (rated on a scale of 1 to 4, expected—3)	3.22 (0.83)	30.0
Comprehensive care (health promotion)
• Health promotion and preventive care (each strategy rated on a scale of 1 to 4, expected—67%* of age- and sex- appropriate strategies rated as probably [3] or definitely [4] addressed)	56.4* (29.6)	57.4

^*Mean and expected ratings are expressed as the proportion of age- and sex-appropriate strategies rated as probably (3) or definitely (4) addressed.

Ratings of organizational accessibility were better, but ratings of individual components varied greatly. While 85% rated convenience of office location as very good or excellent, only 48% rated waiting time for an appointment as very good or excellent, and only 46% rated ease of contacting a physician by telephone as very good or excellent. Average wait to the third-next available appointment was 24 days (SD 22 days, median wait time 19 days, range 0 to 167 days); only 19% of physicians were available within 7 days. These statistics exclude the 5% of physicians never available by appointment and the 5% where delays exceeded 6 months. Study physicians largely agreed with patients’ perceptions of accessibility: only 12% reported being able to see patients within a day for sudden illness.

Relational continuity, accumulated knowledge, and care management were well within acceptable levels. Not surprisingly, patients without regular physicians had significantly less confidence in these aspects of care than those with regular physicians did (that physicians knew their personal and medical context 2.5 vs 3.4 on a scale of 4—definitely to 1—definitely not, t = 64, P < .001, or would manage their care beyond the clinical encounter 2.9 vs 3.4, t = 16.9, P < .001).

The 60% of patients who had seen specialists within the last 2 years rated coordination of care at minimum acceptable levels. Patients whose primary care providers helped make specialist appointments (almost half) rated coordination of care higher than those whose providers did not (3.4 vs 2.4 on a scale of 1 to 4, t = 34, P < .001). For 26% of patients, the last specialist visit had not been referred by their regular providers, but about two thirds of these patients were nonetheless optimistic that their providers knew of the visit and its results.

On average, patients recalled that their physicians had addressed 56% (SD 30%) of the health promotion and preventive care issues appropriate for their age and sex (eg, nutrition, vaccinations, and mental health for all; safety for children; exercise, smoking, and alcohol for adults; osteoporosis for women; and prevention of falls for the elderly).¹⁷ Among those without regular physicians, only 38% of health issues had been addressed; among those with regular physicians about 64% had been addressed (t = 13, P < .001). By contrast, physicians reported “usually” addressing each issue; only a few indicated they addressed them “occasionally.”

Finally, we found significant differences by geographic area (Figure 2). Rural patients reported significantly better experiences than urban patients did for all attributes of care. While there were also differences by type of clinic, geographic differences were more consistent. For instance, while urban and suburban CLSCs were similar to urban practices with respect to accessibility, CLSCs in remote areas were rated highly accessible.

Figure 2.

Mean level of patients’ experience: Comparison of urban and rural patients’ ratings relative to minimum expected and maximum scores (all differences were statistically significant).

DISCUSSION

Before this study, policy makers and the public perceived primary care in Quebec as unresponsive, fragmented, and uncoordinated with other levels of the health care system. Our study corroborates perceptions that primary care is unresponsive to the need for timely access and that coordination and health promotion barely meet minimum expectations. Patients expressed high levels of confidence in their relationships with their regular physicians, however, and these relationships translated into better coordination and comprehensiveness of care.

The large number of patients without family physicians is very worrisome (approximately 1 in 4 patients in Montreal). Patients without family physicians are less likely to get recommended preventive services and experience considerably poorer access to health care and continuity and coordination of care.^18,19 The plight of those without regular physicians is critical and should be a major policy concern, especially in Quebec.

Patients are not confident they could be seen by their regular physicians or clinics within a day for sudden illness, especially at night or on weekends. While a 1-day wait might seem unreasonable to some physicians, certainly an average wait of 24 days for an appointment is not acceptable. Although most physicians have alternate arrangements for urgent care, patients depend on doctors’ secretaries for information on these, and this information varies greatly across practices.

Walk-in clinics, surprisingly, provide continuity of care for a large proportion of patients. Our results might not be generalizable, however, as some walk-in clinics were excluded from our sample because none of the rotating physicians worked a full day and others were inaccessible by telephone. Nonetheless, these clinics meet a need for patients without regular doctors, many of whom have figured out how to use them to achieve continuity of care. Studies have shown that the technical quality of care in walk-in clinics is as good or better than that in family medicine clinics.²⁰

Has primary care changed since reforms in Quebec?

How much is primary care likely to have changed since reforms were implemented in Quebec? Family medicine groups have extended clinic hours and explicit agreements with other institutions for smooth transitions between various levels of care; accessibility and continuity are expected to improve. As of March 2006, there were 103 accredited FMGs (personal communication from Dr Jean-Guy Émond, President of the FMG Implementation Support Group), but the vast majority of Quebec’s population is not served by FMGs. It is unlikely the portrait of care we found in 2002 has changed substantially since then.

How different is Quebec’s health care system from systems in the rest of Canada? The PCAT questionnaire was also used in 4 rural communities in Nova Scotia in 1999. When asked whether they could be seen within a day if they got sick, Nova Scotians’ mean response was 2.5 (“maybe”), compared with 1.9 (“probably not”) in Quebec, according to the Nova Scotia Primary Health Care Department. Overall, the picture was more positive in Nova Scotia than in Quebec, and this could be related to physician-patient affiliation. According to the 2003 Canadian Community Health Survey,⁷ 94% of Nova Scotians had regular doctors, while only 76% of people had them in Quebec (86% had them in Canada as a whole).

Access to health care is similar across Canada. According to the 2003 Canadian Community Health Survey,⁷ 24% of Canadians needing immediate care for minor health problems reported having difficulty accessing care.⁷ Difficulty accessing care was reported by 26% of patients in Quebec, one of the highest rates, although similar to other provinces, with only Saskatchewan (17%), Alberta (22%), and British Columbia (21%) reporting lower rates. As in our survey, the most common problems across the country were difficulty contacting physicians or nurses and long waiting times for appointments.

A 2004 survey of primary care and patients’ perceptions of health systems carried out in 5 countries found that while Canada was second only to the United Kingdom in removing financial barriers, Canadians faced the greatest organizational barriers, with 25% of adult patients waiting longer than 6 days for care compared with 2% in New Zealand, 7% in Australia, 13% in the United Kingdom, and 19% in the United States.²¹ Only in the United States did a higher percentage of respondents believe it was very or somewhat difficult to get care on nights, weekends, or holidays without going to emergency departments (63% vs 59% in Canada).

Limitations

With a clinic response rate of 61%, there might be important biases in our sample. Most likely the study sample is biased toward good care, since clinic directors who were more secure in the quality of their services likely were more willing to participate. The actual experiences of patients might be worse that those reported here. Also, patients’ reports might not be accurate. Indeed, patients’ reports on, say, accessibility varied considerably within the same clinic, and patients with regular physicians consistently rated all aspects of services more positively. It appears that patients with good relationships with their physicians are more tolerant, so their reports might be positively biased. Finally, it is critical to keep in mind that inferences drawn from the results of the study apply only to service users, not to the population in general.

Conclusion

This survey highlights the fact that there is much room for improvement in accessibility to primary care services. People with regular primary care physicians have good continuity of care, experience acceptable coordination with specialists, and receive preventive care. Those without regular physicians have considerably worse experiences of care.

EDITOR’S KEY POINTS

• This study addresses a very important step in health care reform, as it is crucial to have an understanding of the baseline in order to make the right changes and determine whether those changes have had the desired effects.

• This study found that, in the Quebec health care settings studied, there was much room for improvement, particularly in the areas of accessibility to health care and provision of health promotion and preventive services. Patients who had regular family physicians fared better than those who did not.

POINTS DE REPÈRE DU RÉDACTEUR

• Cette étude concerne un point très important de la réforme des soins de santé, soit la nécessité d’avoir une vue claire de la situation actuelle si on veut effectuer les bons changements et vérifier si ces changements apportent les effets désirés.

• Les résultats de l’étude indiquent que dans l’organisation des soins de santé au Québec, il y a place à beaucoup d’amélioration, en ce qui concerne notamment l’accessibilité aux soins, la promotion de la santé et les services préventifs.