Association of Advanced Access With Primary Care Performance: A Systematic Review

Abstract

PURPOSE

Timely access to primary care remains a major challenge. Advanced access, an organizational model for appointment management, aims to improve accessibility by allowing patients to consult their health care professional in a timely manner. This systematic review evaluated its association with several key dimensions of primary care performance.

METHODS

We conducted a comprehensive search in MEDLINE, Embase, Web of Science, CINAHL, and the Cochrane Library from inception to May 10, 2025. Randomized intervention studies, nonrandomized intervention studies, and observational studies comparing advanced access with traditional scheduling systems among patients in primary care were eligible. Outcomes were the wait time to an appointment, continuity of care, emergency department (ED) use, and patient satisfaction. Two independent reviewers selected studies, extracted data, and assessed risk of bias. We performed a narrative synthesis, separately analyzing universal and non-universal health care systems.

RESULTS

Among 7,595 unique records identified, 29 studies were included. Of these, 23 evaluated appointment wait time, 13 examined continuity of care, 3 assessed ED use, and 13 evaluated patient satisfaction. Appointment wait time decreased in all 23 studies assessing this outcome, with 13 reporting statistically significant reductions. Continuity of care improved in 11 of 13 studies, of which 7 reported statistically significant improvements. All 3 studies evaluating ED use reported reductions in ED visits, although none demonstrated statistical significance. Patient satisfaction improved in 8 studies, with 3 reporting statistically significant improvements. Overall, findings were consistent across universal and non-universal health care systems.

CONCLUSIONS

Advanced access appears to improve appointment timeliness, continuity of care, and patient satisfaction, and may help reduce avoidable ED use across diverse health care systems.

INTRODUCTION

Access to primary care is a fundamental pillar of high-performing health systems and is associated with better population health outcomes.^1,2 Even in high-income countries, however, many patients face difficulties obtaining timely medical appointments.³ According to a 2023 international survey by the Commonwealth Fund, the possibility of seeing a clinician on the same day or the following day remains limited and is reportedly declining.⁴ Across the countries surveyed, the proportion of adults reporting timely access fell on average from 60% in 2016 to 42% in 2023, with the Netherlands ranking highest (54%) and Canada lowest (26%). The survey also found that access to care outside regular office hours, without resorting to an emergency department (ED) visit, remains constrained. This lack of accessibility in primary care directly contributes to the overcrowding of EDs, which are often used for nonurgent medical issues.^5-8

To address these challenges, several reforms and strategies have been proposed, among which the advanced access model has emerged as a potential solution. Developed in the early 2000s in the United States, this model aims to restructure appointment scheduling and clinical practice to ensure patients can see their health care clinician in a timely manner. Patient-centered in its design, it is based on the simple principle of “doing today’s work today.”⁹ In practice, this involves keeping a substantial proportion of appointment slots open each day to accommodate same-day demand, thereby promoting proactive management of supply and demand.^9,10 This model stands in contrast to traditional approaches, often characterized by rigid scheduling systems, overbooked calendars, and backlogs that result in extended wait times for patients.¹⁰ To improve both access to and quality of care, many health systems, including Canada’s, have begun implementing the advanced access model.^10-13

Although advanced access has generated increasing interest, evidence regarding its effects remains inconsistent and the most recent comprehensive synthesis is now outdated.¹⁴ At a time when improving timely access, continuity of care, and the efficient use of primary care resources is more crucial than ever, it is necessary to reassess the effectiveness of this model using all currently available empirical data. We therefore undertook a systematic review to evaluate, among patients receiving primary care, the effect of advanced access compared with traditional appointment scheduling on appointment wait times, continuity of care, ED use, and patient satisfaction.

METHODS

This systematic review was conducted in accordance with Cochrane methodological standards,¹⁵ and the results are reported following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.¹⁶ The protocol was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42024511882.

Eligibility Criteria

Studies were eligible for inclusion if they were randomized intervention studies, nonrandomized intervention studies, or observational studies conducted among patients in primary care, defined as first-contact care typically delivered in ambulatory settings by frontline clinicians.¹⁷ Studies had to compare the advanced access model with traditional appointment scheduling and report at least 1 of the following outcomes: wait time for an appointment, continuity of care, ED use, and patient satisfaction.

We did not apply any restrictions regarding language or year of publication. Literature reviews, editorials, correspondence, conference abstracts, conceptual papers, and reports were excluded to ensure methodologic rigor and comparability of the included evidence.

Search Strategy and Study Selection

We developed a comprehensive search strategy (Supplemental Table 1) in collaboration with a health sciences librarian (F.B.). The search was conducted in MEDLINE (Ovid), Embase (www.Embase.com), Web of Science (all indexes), CINAHL (with full text), and the Cochrane Library, from inception to May 10, 2025. All identified references were imported into Covidence software (Veritas Health Innovation Ltd)¹⁸ for study selection, and duplicate records were automatically removed. Two independent reviewers (J.A. and J.G.) screened titles and abstracts to identify potentially eligible studies, and thereafter conducted a full-text review to confirm eligibility. Any disagreements between reviewers were resolved through discussion or, when necessary, arbitration by a third reviewer (S.B.).

Data Extraction and Risk of Bias Assessment

We performed data extraction using a standardized form developed in Excel (Microsoft Corp),¹⁹ which was pilot-tested on 5 studies to ensure clarity and consistency. Two independent reviewers (J.A. and S.L.) extracted the data, and any discrepancies were resolved through discussion or, if needed, by arbitration from a third reviewer (S.B.). Extracted data included general study identifiers; methodological characteristics such as study design and setting; inclusion and exclusion criteria; numbers of clinics, practitioners, participants, and visits; study duration; details on the intervention and comparator; and all reported results for each outcome of interest.

Risk of bias was assessed for each outcome using the Risk of Bias in Non-Randomized Studies of Interventions (ROBINS-I) tool,²⁰ by the same reviewers (J.A. and S.L.) and by arbitration from a third reviewer (S.B.) when needed. On the basis of this assessment, studies were categorized into 4 levels of risk of bias: critical, serious, moderate, or low. Judgments were summarized in separate tables for each outcome analyzed.

Data Analysis and Synthesis

We did not undertake a meta-analysis because of substantial methodologic heterogeneity across the included studies, including differences in outcome definitions and measurement methods and follow-up durations, as well as incomplete or inconsistent reporting of data required for quantitative synthesis. Consequently, a narrative synthesis was conducted. We first summarized results across all included studies and then described results according to type of health system—universal coverage (United Kingdom and Canada) vs non-universal, multipayer coverage (United States)—without conducting any formal between-group comparisons.

RESULTS

Study Characteristics

A total of 9,268 records were identified through database searches, 7,595 of which were unique (Figure 1). After screening the records and applying the eligibility criteria, we ultimately included in the systematic review 29 articles that reported studies of advanced access published from 2001 to 2018.^21-49

Figure 1.

PRISMA Flow Diagram of Study Selection

PRISMA = Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Among the 29 included studies, 23 studies were conducted in the United States, 4 in the United Kingdom, and 2 in Canada (Table 1). Regarding study design, 2 studies were observational, whereas 27 were quasi-experimental before-and-after studies, including 22 without a comparison group. Ten studies were conducted exclusively in academic practice settings. The follow-up duration ranged from 3 to 48 months, with 10 studies reporting a follow-up of less than 12 months. With respect to outcomes, 23 studies assessed wait time for an appointment, 13 evaluated continuity of care, 3 examined ED use, and 13 assessed patient satisfaction.

Table 1.

Characteristics of Included Studies Evaluating the Effects of Advanced Access in Primary Care (N = 29 Studies)

Study authors, year	Country	Study design	Clinic type	No. of clinics	No. of cliniciansa	No. of patients/visits	Follow-up, mo	Outcomes studied
Radel et al,35 2001	United States	Uncontrolled before-and-after study	Family medicine practices	2	6	NR	12-15	Appointment wait time, patient satisfaction
Kennedy and Hsu,36 2003	United States	Uncontrolled before-and-after study	Academic family medicine center	1	12.8 FTE	NR	5	Appointment wait time, patient satisfaction
Meyers,21 2003	United States	Uncontrolled before-and-after study	Naval medical clinic	1	9	NR	8	Appointment wait time, continuity of care
Belardi et al,22 2004	United States	Controlled before-and-after study	Academic family medicine clinic	1	10 (5 per team)	NR	12	Appointment wait time, continuity of care, patient satisfaction
Mallard et al,37 2004	United States	Uncontrolled before-and-after study	Pediatric clinic	1	2	Before: 4,063 appointments After: 3,531 appointments	6	Appointment wait time
Pickin et al,23 2004	United Kingdom	Uncontrolled before-and-after study	General practices (141 academic)	352	NR	NR	8-17	Appointment wait time
Solberg et al,24 2004	United States	Uncontrolled before-and-after study	Community-based multidisciplinary group practices	17	550	Before: 22,000 visits After: 17,000 visits	20	Appointment wait time
Solberg et al,45 2004	United States	Uncontrolled before-and-after study	Community-based multidisciplinary group practices	17	NR	Patients with CHD   Before: 3,555   After: 3,802 Patients with diabetes   Before: 6,741   After: 7,238 Patients with depression   Before: 5,803   After: 6,336	24	Continuity of care, ED use
Bundy et al,25 2005	United States	Uncontrolled before-and-after study	Family and pediatric clinics	4	41.3 FTE	1,270 patients; >11,000 visits	9	Appointment wait time; continuity of care; patient satisfaction
Parente et al,38 2005	United States	Uncontrolled before-and-after study	Pediatric clinic	1	4	2,791 patients	3	Appointment wait time; continuity of care; patient satisfaction
Armstrong et al,39 2006	United States	Uncontrolled before-and-after study	VHA general medicine clinics	1,300	NR	NR	48	Appointment wait time
Dixon et al,26 2006	United Kingdom	Uncontrolled before-and-after study	General practices (141 academic)	352	NR	NR	8-17	Appointment wait time
Lewandowski et al,48 2006	United States	Uncontrolled before-and-after study	Multidisciplinary medical group (family medicine, internal medicine, pediatric)	18	99.6 FTE	NR	48	Patient satisfaction
Solberg et al,27 2006	United States	Uncontrolled before-and-after study	Community-based multidisciplinary group practices	17	NR	Before: 6,609 patients After: 7,284 patients	24	Appointment wait time, continuity of care
Steinbauer et al,28 2006	United States	Uncontrolled before-and-after study	Academic family medicine clinic	1	6 FTE	NR	30	Appointment wait time; patient satisfaction
Cherniack et al,40 2007	United States	Uncontrolled before-and-after study	Academic geriatric clinic (VHA)	1	14	NR	12	Appointment wait time
Salisbury et al,29 2007	United Kingdom	Controlled before-and-after study	General practices (21 academic)	47 (24 AA, 23 TS)	NR	Wait time: NR Continuity: 9,400 patients	12	Appointment wait time; continuity of care
Salisbury et al,41 2007	United Kingdom	Cross-sectional cohort study	General practices (21 academic)	47 (24 AA, 23 TS)	NR	10,821 patients	NA	Appointment wait time; continuity of care; patient satisfaction
Mehrotra et al,30 2008	United States	Uncontrolled before-and-after study	Family medicine, general medicine, and community care clinics	5	45 clinicians (35.3 FTE)	NR	32	Appointment wait time; patient satisfaction
Pomerantz et al,42 2008	United States	Uncontrolled before-and-after study	Primary mental health clinic in veterans medical center	1	2	987 patients	6	Appointment wait time
Sperl-Hillen et al,31 2008	United States	Uncontrolled before-and-after study	Community-based multidisciplinary group practices	17	NR	Before: 6,741 patients After: 7,238 patients	24	Appointment wait time; continuity of care
Bennett and Baxley,32 2009	United States	Uncontrolled before-and-after study	Academic family medicine clinic	1	49	43,349 visits	15	Appointment wait time; continuity of care
Phan and Brown,46 2009	United States	Uncontrolled before-and-after study	Academic family medicine clinic	1	32	Before: 475 patients After: 375 patients	12	Continuity of care
Subramanian et al,44 2009	United States	Controlled before-and-after study	Academic primary care group practices	12 (6 AA, 6 TS)	AA: 51.5 FTE TS: 44.6 FTE	AA: 3,147 patients TS: 913 patients	12	ED use
Cameron et al,33 2010	Canada	Uncontrolled before-and-after study	Academic family medicine clinics	2	NR	Before: 21,838 visits After: 21,819 visits	9	Appointment wait time
Hudec et al,47 2010	Canada	Controlled before-and-after	Solo family medicine clinic	4 (1 AA, 3 TS)	NR	ED use: NR Satisfaction:   AA: 117 patients   TS: 207 patients	3	ED use; patient satisfaction
Tuli et al,43 2010	United States	Controlled before-and-after study	Academic pediatric clinic	1	31 (19 AA, 12 TS)	Wait time, continuity: NR Satisfaction:   Before: 66 patients   After: 73 patients	4	Appointment wait time; continuity of care; patient satisfaction
Tseng et al,34 2015	United States	Uncontrolled before-and-after study	Community academic family med clinic	1	NR	NR	4.5 (18 weeks)	Appointment wait time; continuity of care; patient satisfaction
Duhigg et al,49 2018	United States	Case-control study	Primary and specialty care clinics	1,422 (711 case, 711 control)	1,422	23,292 patients	NA	Patient satisfaction

AA = advanced access; CHD = coronary heart disease; ED = emergency department; FTE = full-time equivalent; mo = months; NA = not applicable; NR = not reported; TS = traditional scheduling; VHA = Veterans Health Administration.

^aDefined as physicians and/or nurse practitioners providing patient care.

The risk of bias assessment, conducted using the ROBINS-I tool, revealed a high proportion of studies with a serious risk of bias, which varied by outcome and was most often attributable to a lack of control for confounding factors (Supplemental Table 2). Specifically, 87.0% of studies assessing wait time (Supplemental Table 3), 70.0% of those assessing continuity of care (Supplemental Table 4), 33.3% of those assessing ED use (Supplemental Table 5), and 85.0% of studies assessing patient satisfaction (Supplmemental Table 6) were judged to be at serious risk of bias.

Study Outcomes

Wait Time For an Appointment

Twenty-three studies evaluated the association of advanced access with appointment wait times (Table 2).^21-43 Two indicators were used: the time to the third next available appointment, reported in 14 studies,^21-34 and the interval between the appointment request and the scheduled date, assessed in 10 studies.^29,35-43 One study examined both indicators simultaneously.²⁹ The third next available appointment time is defined as the number of days between the request date and the third available appointment in a physician’s schedule, excluding slots reserved for specific populations or procedures (eg, obstetrics, pediatrics, minor surgery).⁵⁰

Table 2.

Association of Advanced Access With Appointment Wait Time (N = 23 Studies)

Study authors, year	Wait time before advanced access, days	Wait time after advanced access, days	Difference (95% CI), days	Change from baseline, %	P value
Time to third next available appointment
United Kingdom/Canada
Pickin et al,23 2004	3.6	1.9	−1.7 (−2 to −1.4)	−47.2	NR
Dixon et al,26 2006	3.6	1.5	−2.1	−58.3	NR
Salisbury et al,29 2007	2.87	1.61	−1.14 (−2.23 to −0.05)	−39.7	.04
Cameron et al,33 2010	13.7 (1.93)a	3.6 (0.78)a	−10.1	−73.7	<.001
United States
Meyers,21 2003	2.2	1	−1.2	−54.5	NR
Belardi et al,22 2004	21	4-7	−17 to −14	−81 to −66	<.001
Solberg et al,24 2004	17.8	3.9	−13.9	−78.1	NR
Bundy et al,25 2005	36	4	−32 (−44 to −20)	−88.9	NR
Solberg et al,27 2006	19.4	4.5	−14.9	−76.8	<.001
Steinbauer et al,28 2006	17	1	−16	−94.1	NR
Mehrotra et al,30 2008	21	11	−10	−47.6	<.001
Sperl-Hillen et al,31 2008	21.6	4.2	−17.4	−80.6	<.001
Bennett and Baxley,32 2009	30.7	9	−21.7	−70.7	<.001
Tseng et al,34 2015	10.5b	2b	−8	−76.2	NR
Interval between request and appointment
United Kingdom/Canada
Salisbury et al,29 2007	1.87	1	−0.75 (−1.51 to 0.004)c	−40.1	.05
Salisbury et al,41 2007	NR	NR	2.32 (1.51 to 3.57)d	NA	<.001
United States
Radel et al,35 2001	Site A: 11	Site A: 2	Site A: −9	Site A: −81.8	NR
	Site B: 59	Site B: 1	Site B: −58	Site B: −98.3
Kennedy and Hsu,36 2003	NR	NR	Reduction in wait time	NA	NR
Mallard et al,37 2004	46	15	−31	−67.4	<.001
Parente et al,38 2005	18.7	11.8	−6.9	−36.9	<.001
Armstrong et al,39 2006	42.9	15.7	−27.2	−63.4	NR
Cherniack et al,40 2007	>30	<30	NR	NA	NR
Pomerantz et al,42 2008	33	0	−33	−100.0	NR
Tuli et al,43 2010	AA group: 35.7	AA group: 20.3	AA group: −15.1	AA group: −42.3	<.05
	TS group: 35.7	TS group: 32.9	TS group: −2.8	TS group: −7.8

AA = advanced access; NA = not applicable; NR = not reported; TS = traditional scheduling.

Note: Values are means unless otherwise noted.

^aMean (SD).

^bMedian.

^cAdjusted mean difference (95% CI).

^dAdjusted odds ratio (95% CI), AA vs TS for timely access to care.

All 14 studies assessing the third next available appointment reported a reduction after the implementation of advanced access. Before implementation, values ranged from 2.2 to 36 days, compared with 1 to 11 days after implementation. Percentage reductions ranged from 39.7% to 94.1%. Six studies^{21,23,26,28,29,34} reported postimplementation values of 2 days or fewer, while only 2 studies^21,28 reported a time of 1 day. Among the 14 studies, 9 studies^{22,23,25,27,29-33} showed statistically significant improvements.

For the interval between the appointment request and the scheduled date, all 10 studies evaluating this indicator reported reductions.^29,35-43 Pre-implementation wait times ranged from 1.87 to 59 days, compared with 0 to 20.3 days after implementation, corresponding to percentage reductions between 36.9% and 100%. Among the 10 studies, 5 reported P values, with 4 of them demonstrating statistically significant reductions.^37,38,41,43

In summary, irrespective of the indicator used or the setting, all studies assessing appointment wait time showed a reduction with advanced access, though only slightly more than one-half (13 out of 23)^{22,23,25,27,29-33,37,38,41,43} provided statistical confirmation.

Continuity of Care

Thirteen studies assessed continuity of care using various methodological approaches (Table 3).^{21,22,25,27,29,31,32,34,38,41,43,45,46} More than one-half of the studies (7 out of 13)^{21,22,25,32,34,38,41} used direct measures, such as the percentage of patients seen by their assigned or preferred clinician, or the percentage of visits made to that clinician relative to all visits. Four studies^27,29,31,45 applied the continuity of care (COC) index (also known as the Bice-Boxerman index),⁵¹ which captures both the frequency of visits to each clinician and the distribution of visits across multiple clinicians. Phan and Brown⁴⁶ assessed continuity with 2 indices: the usual provider of care (UPC) index⁵² and the modified modified continuity index (MMCI).⁵³ The UPC index is the proportion of visits made to the assigned or most frequently seen clinician, while the MMCI combines the number of visits and the number of clinicians consulted to account for the dispersion of care across multiple clinicians.^52,53 Finally, Tuli et al⁴³ adopted an indirect approach, examining the percentage of well-child visits conducted in acute care clinics, under the assumption that stronger continuity would reduce such nonurgent consultations through more consistent follow-up with the child’s designated physician.

Table 3.

Association of Advanced Access With Continuity of Care (N = 13 Studies)

Study authors, year	Indicator	Before advanced access	After advanced access	Difference (95% CI)	Change from baseline, %	P value
United Kingdom/Canada
Salisbury et al,29 2007	COC index	AA group: 0.32 (0.31)a TS group: 0.32 (0.30)a	AA group: 0.28 (0.27)a TS group: 0.34 (0.30)a	0.006 (−0.07 to 0.08)b	1.9	.88
Salisbury et al,41 2007	Percentage of patients reporting usually seeing or speaking to their preferred GP (GPAQ)	NR	NR	1.20 (0.91 to 1.57)c	NA	NR
United States
Meyers,21 2003	Percentage of patients seen by their assigned PCP	37%	45%	8%	21.6	NR
Belardi et al,22 2004	Percentage of visits with the patient’s PCP	NR	>90%	NR	NA	<.02
Bundy et al,25 2005	Percentage of patients reporting seeing their preferred clinician	76%	89%	13% (−7% to 32%)	17.1	NSd
Parente et al,38 2005	Percentage of visits with the patient’s PCP	69.8%	91.4%	21.6%	30.9	<.001
Solberg et al,27 2006	Percentage of patients for whom >50% of visits were with 1 clinician	67.3%	74.0%	6.7%	9.9	<.001
Bennett and Baxley,32 2009	Percentage of visits with the patient’s PCP	64.0%	68.2%	4.2%	6.6	<.001
Tseng et al,34 2015	Percentage of patients seen by their assigned PCP	57%	55%	−2%	−3.5	NR
Solberg et al,45 2004	COC index	CHD: 0.66 Diabetes: 0.68 Depression: 0.60	CHD: 0.72 Diabetes: 0.73 Depression: 0.63	CHD: 0.06 Diabetes: 0.05 Depression: 0.03	CHD: 9.1 Diabetes: 7.4 Depression: 5.0	CHD: <.001 Diabetes: < .001 Depression: < .001
Sperl-Hillen et al,31 2008	COC index	0.681	0.725	0.044	5.9	<.001
Phan and Brown,46 2009	UPC index	0.59 (0.23)a	0.55 (0.22)a	−0.04	−6.8	.01
	MMCI	0.51 (0.30)a	0.44 (0.29)	−0.07	−13.7	<.001
Tuli et al,43 2010	Percentage of wellchild checks seen in acute clinic	4%	1.4% to 1.5%	−2.6% to −2.5% 0.3 (0.2 to 0.5)e	NA	<.001

AA = advanced access; CHD = coronary heart disease; COC = continuity of care (COC index, also called Bice-Boxerman index, is calculated as the extent to which a patient’s visits are concentrated with the same clinician⁵¹; possible values range from 0 [each visit was to a unique clinician] to 1 [all visits were to a single clinician], with higher values indicating greater continuity); GP = general practitioner; GPAQ = General Practice Assessment Questionnaire (for continuity, the item used was the proportion of patients reporting that they usually see their preferred/usual clinician⁶³); MMCI = modified modified continuity index (accounts for the dispersion of care among multiple clinicians⁵³; possible values range from 0 to 1, with higher values indicating greater continuity); NR = not reported; NS = not significant; PCP = primary care provider; TS = traditional scheduling; UPC = usual provider of care (UPC index is proportion of visits conducted with the usual or preferred clinician⁵²; possible values range from 0 to 1, with higher values indicating greater continuity).

^aMean (SD).

^bAdjusted mean difference (95% CI).

^cAdjusted odds ratio (95% CI), AA vs TS for continuity of care.

^dP value not reported.

^eCrude odds ratio (95% CI).

Among the 13 studies assessing continuity of care, 11 reported improvements after implementation of advanced access.^{21,22,25,27,29,31,32,38,41,43,45} Of these, 7 studies demonstrated statistically significant improvements.^{22,27,31,32,38,43,45} Reported gains relative to baseline ranged from 1.9% to 30.9%, suggesting a generally positive impact of advanced access, although the magnitude of improvement varied by setting and context.

In contrast, 2 studies^34,46 observed declines in continuity of care after implementation of advanced access, 1 of which⁴⁶ reached statistical significance. Both were conducted in academic clinics, where care is predominantly delivered by resident physicians under faculty supervision.

ED Use

Three studies evaluated the association of advanced access with ED use (Table 4).^44,45,47 All 3 reported decreases in use of the ED or urgent care, with reductions ranging from 3% to 28%; however, in 2 studies^45,47 statistical significance was not reported, and in the third,⁴⁴ the observed reduction did not reach statistical significance. Taken together, these findings demonstrate a consistent trend suggesting that the implementation of advanced access may contribute, to some extent, to reducing avoidable ED use, defined as visits for nonurgent conditions or for issues that could be appropriately managed in primary care.

Table 4.

Association of Advanced Access With Emergency Department Use (N = 3 Studies)

Study authors, year	Before advanced access	After advanced access	Difference (95% CI)	P value
United Kingdom/Canada
Hudec et al,47 2010	NR	NR	−28%	NR
United States
Solberg et al,45 2004	CHD: 65.0%a	CHD: 59.5%a	CHD: − 5.5%a	NR
	Diabetes: 31.9%a	Diabetes: 27.5%a	Diabetes: −4.4%a
	Depression: 46.7%a	Depression: 39.7%a	Depression: −7.0%a
Subramanian et al,44 2009	AA group: 1.1 (1.9)b	AA group: 1.1 (2.1)b	0.97 (0.92-1.02)c	NS
	TS group: 0.9 (2.0)b	TS group: 0.9 (1.9)b

AA = advanced access; CHD = coronary heart disease; NR = not reported; NS = not significant; TS = traditional scheduling.

^aPercentage of patients with at least 1 urgent care visit or emergency department visit annually.

^bMean (SD) number of urgent care and/or emergency department visits per patient annually.

^cAdjusted rate ratio (95% CI).

Patient Satisfaction

Thirteen studies assessed the association of advanced access with patient satisfaction (Table 5).^{22,25,28,30,34-36,38,41,43,47-49} Two dimensions were explored: overall patient satisfaction, examined in 10 studies,^{22,25,34-36,38,41,43,47,48} and satisfaction with the appointment system, assessed in 6 studies^{22,25,28,30,47,49}; 3 studies^22,25,47 measured both dimensions simultaneously.

Table 5.

Association of Advanced Access With Patient Satisfaction (N = 13 Studies)

Study authors, year	Before advanced access	After advanced access	Difference (95% CI)	Change from baseline, %	P value
Overall satisfaction
United Kingdom/Canada
Salisbury et al,41 2007	NR	NR	0.93 (0.67 to 1.28)a	NA	NSb
Hudec et al,47 2010	NR	NR	+1 satisfaction point after vs before	NA	Significantb
United States
Radel et al,35 2001	72%	95%	23%	31.9	NR
Kennedy and Hsu,36 2003	NR	NR	Improvement in satisfaction	NA	NR
Belardi et al,22 2004	NR	NR	No significant improvement in satisfaction score	NA	NR
Bundy et al,25 2005	45%	61%	16% (0.2% to 30%)	35.6	Significantb
Parente et al,38 2005	6.21 out of 7	6.08 out of 7	−0.13 out of 7	−2.1	NSb
Lewandowski et al,48 2006	84%	87%	3%	3.6	NSb
Tuli et al,43 2010	63%	90%	27%	42.9	<.001
Tseng et al,34 2015c	79%	78%	−1%	−1.3	NSb
Satisfaction with appointment system
United Kingdom/Canada
Hudec et al,47 2010	NR	NR	0.26d	NA	Significantb
United States
Belardi et al,22 2004	NR	NR	No significant improvement	NA	NR
Bundy et al,25 2005	37%	47%	10% (−9% to 29%)	27.0	NSb
Steinbauer et al,28 2006	NR	NR	Improvement in satisfaction	NA	NR
Mehrotra et al,30 2008	53%	51%	−2%	−3.8	NR
Duhigg et al,49 2018	NR	NR	0.05e	NA	<.10

NA = not applicable; NR = not reported; NS = not significant.

Note: Values are percentages of patients who reported being satisfied (according to study criteria) or satisfaction scores on Likert scales.

^aAdjusted odds ratio.

^bP value not reported.

^cMean percentile score.

^dCrude mean difference (Likert scale points); 5-point Likert scale with 5 indicating highest satisfaction.

^eAdjusted mean difference (Likert scale points); 6-point Likert scale with 6 indicating highest satisfaction.

Regarding overall patient satisfaction, 6 studies^{25,35,36,43,47,48} of the 10 studies assessing this outcome reported improvements after the implementation of advanced access compared with baseline, although only 3 of them^25,43,47 demonstrated statistically significant gains. In contrast, 4 studies reported no improvement, with satisfaction remaining stable in 1 study²² and showing small, nonsignificant declines in 3 others.^34,38,41

With respect to satisfaction with the appointment system, 4 studies^25,28,47,49 out of the 6 studies assessing this outcome reported improvements, including 1 study⁴⁷ with a statistically significant result. Two studies^22,30 found no improvement, 1 of which³⁰ reported a nonsignificant decrease in satisfaction after implementation.

In summary, although findings were heterogeneous and not consistently statistically significant, most studies (8 out of 13)^{25,28,35,36,43,47-49} reported improvements in patient satisfaction after the implementation of advanced access, with 3 studies^25,43,47 showing statistically significant gains. Studies that did not report improvements^{22,30,34,38,41} generally indicated stable satisfaction levels or minor, nonsignificant changes, reinforcing the notion of a generally favorable impact of advanced access on the perceived care experience.

Narrative Synthesis Stratified by Health System Type

Our separate examination of studies conducted in the United States and studies conducted in the United Kingdom and Canada revealed broadly convergent trends between non-universal and universal health care systems (Tables 2-5).

Most included studies (23 out of 29) were carried out in the United States, providing a more extensive picture for a country with a non-universal health care system. Wait times decreased across all 23 studies, with 9 studies^{22,25,27,30-32,37,38,43} of 18 reporting statistically significant reductions. Three studies^21,28,34 achieved a third next available appointment time of 2 days or less, including 2 studies^21,28 reporting a time of 1 day. Continuity of care improved in most studies (9 out of 11),^{21,22,25,27,31,32,38,43,45} with statistically significant improvements reported in 7 of them.^{22,27,31,32,38,43,45} ED use decreased in both studies^44,45 assessing this outcome, although neither demonstrated statistical significance. Patient satisfaction improved in 7 studies^{25,28,35,36,43,48,49} of the 11 assessing this outcome, with statistically significant gains reported in 2 of them.^25,43

In the United Kingdom and Canada, countries with universal health care systems, fewer studies (6 studies) were available but similar patterns were observed. Wait times decreased in all studies, with statistically significant reductions reported in 4 studies^23,29,33,41 of 5 assessing this outcome. Among the 4 studies that measured the time to third next available appointment, 3 studies^23,26,29 succeeded in reducing it to 2 days or less, although none reached a 1-day threshold. Statistically significant improvements in patient satisfaction were reported in 1 study⁴⁷ of 2 assessing this outcome.^41,47 Improvements in continuity of care were observed in both studies^29,41 that assessed this outcome, and 1 study⁴⁷ reported a reduction in nonurgent ED visits; however, none of these changes reached statistical significance.

Overall, the observed effects appeared similar when comparing universal and non-universal systems, suggesting that the advanced access model may contribute to improving the organization of primary care across diverse health care settings.

DISCUSSION

This systematic review evaluated the impact of advanced access on key dimensions of primary care performance. Across 29 studies, the model was generally associated with shorter wait times to appointments, improved continuity of care, enhanced patient satisfaction, and, to some extent, reduced use of ED services, regardless of the type of health system in which it was implemented.

Findings in Context

Wait time was the most frequently assessed outcome (Table 2). All studies evaluating this indicator reported a reduction after implementation of advanced access, although fewer than one-half provided robust statistical evidence. Fourteen studies^21-34 measured the time to the third next available appointment, widely recognized as the reference indicator for timely access in primary care, notably because it is less sensitive to last-minute cancellations.^12,44,54 From an operational perspective, a value of 0 to 1 day is commonly considered the threshold at which access becomes a stable property of the system, enabling rapid response to routine care needs and limiting delays in care delivery.^10,55 In our review, only 2 studies, both conducted in the United States, reported a postimplementation time to third next available appointment of 1 day.^21,28 It is noteworthy, however, that studies conducted in the United Kingdom reported postimplementation values very close to this threshold, although none formally reached a 1-day time. These findings suggest that, although strict achievement of a 1-day third next available appointment time remains challenging, very high levels of access can nevertheless be attained across diverse organizational contexts. Moreover, the substantial variability observed across studies in waiting times is likely multifactorial. It may reflect differences in pre-implementation wait times as well as variations in the fidelity of advanced access implementation. In some settings, practices maintained extended booking windows or faced structural constraints, such as understaffing, absenteeism, or restrictive organizational policies, which may have attenuated or masked the full effect of the advanced access model.^28,34

Most studies also reported improvements in continuity of care, although findings were more mixed in academic settings (Table 3).^34,46 In such environments, care is often provided by resident physicians and faculty members with fragmented schedules and multiple responsibilities. Frequent resident rotations, academic obligations, and intermittent presence hinder the establishment of long-term patient-clinician relationships.⁵⁶ Yet, relational continuity is a cornerstone of primary care: it is associated with better treatment adherence, increased receipt of preventive care, reduced ED use, and fewer avoidable hospitalizations.^57,58 Evidence also shows a dose-response relationship between continuity and clinical outcomes, with significant benefits observed beyond a 50% threshold, and optimal effects around 80%, potentially leading to a 15% reduction in overall health care costs.^59-61

The association of advanced access with ED use was less frequently evaluated, with only 3 studies^44,45,47 addressing this outcome. All of them reported declines in visits to the ED or urgent care clinics among primary care patients generally or among patients with common chronic conditions specifically (Table 4). Although the statistical evidence was not sufficiently available or robust, this finding likely reflects indirect benefits of improved access and continuity, consistent with broader evidence linking poor same-day access to increased ED use in less responsive primary care systems.⁶² Advanced access may thus contribute to alleviating ED overcrowding by meeting more first-line care demands.

Patient satisfaction was another extensively studied outcome (Table 5). Nearly two-thirds of the studies evaluating satisfaction, whether overall^{25,35,36,43,47,48} or specifically with the appointment system,^25,28,47,49 reported improvements. Nevertheless, relatively few studies demonstrated statistically significant improvements, and substantial heterogeneity in measurement instruments and outcome definitions limited cross-study comparisons. A consistent pattern nonetheless emerged: patients tended to report higher satisfaction when timely access was combined with continuity of care with their preferred clinician. As Bermingham et al⁶⁰ have noted, removing access barriers and enabling continuity with a known clinician not only reduces care-seeking behavior but also enhances patient satisfaction.

Consistency Across Health Systems

The impact of implementing advanced access could potentially differ across health systems having different coverage, organization, procedures, and payment models. In our included studies, far fewer were conducted in the United Kingdom and Canada than in the United States, yet the overall trends were consistent for these 2 broad groups. This consistency suggests that the effects of advanced access are not limited to a specific health system design but may be generalizable across universal and non-universal, multipayer contexts.

Strengths and Limitations

Our findings are consistent with those of a systematic review by Rose et al,¹⁴ which included studies published up to 2010 and reported similar trends across the 4 outcomes examined. By including a larger and more recent body of evidence, our review updates the available data and, to some extent, strengthens the evidence base while reducing the uncertainties highlighted in that earlier review.

The main strengths of our review lie in its comprehensive coverage of more than 2 decades of literature, its updated synthesis, and its outcome-specific analyses enriched by exploration of impact in the diverse contexts in which advanced access has been implemented. Conducted in accordance with Cochrane methodologic standards¹⁵ and reported following PRISMA guidelines,¹⁶ we also ensured transparency and rigor throughout the process.

Nonetheless, certain limitations must be considered when interpreting our findings. Many included studies had serious risk of bias, often relying on uncontrolled before-and-after designs and descriptive analyses, sometimes with partial adjustment for confounding, and limited use of rigorous inferential statistical methods. These factors restrict causal inference and internal validity. Considerable heterogeneity was also present, reflecting differences in health system organization, clinical settings, and outcome measurement. Essential data were sometimes missing, further limiting comparability. As a result, meta-analysis was not feasible, and separate analyses by health system type remained purely qualitative, without statistical testing, making it difficult to draw strong evidence-based conclusions. In addition, some studies did not report P values or CIs, which limited the statistical robustness of the findings and led us to interpret the results primarily in terms of trends.

Short follow-up durations were another important limitation: 10 of 29 studies had less than 12 months of follow-up, insufficient to capture long-term sustainability or seasonal variations in service demand. It is therefore possible that observed improvements reflect short-term dynamics related to early implementation rather than enduring changes. Publication bias cannot be excluded; however, the inclusion of several studies reporting neutral or slightly negative results suggests that its overall impact on our conclusions is likely limited. Finally, the geographic concentration of evidence in 3 high-income countries (United States, United Kingdom, and Canada) highlights a gap in the international literature and may limit the generalizability of our findings to health systems with different economic and organizational contexts.

Conclusions

The findings of this review suggest that advanced access is likely to improve key dimensions of primary care delivery and may represent a promising approach to enhancing efficiency and patient-centeredness in primary care access across both universal and non-universal, multipayer health care systems. Studies from a broader range of countries will be needed to confirm the generalizability of our results.