Review of patient-reported experience within Patient-Centered Medical Homes: insights for Australian Health Care Homes

Additional keywords: healthcare evaluation, health services research, quality of health care.

Managing the healthcare needs of people living with chronic conditions is an ongoing challenge, with many people suffering from multiple conditions. A coordinated approach to care management is needed, although for people already disadvantaged by ill-health, the challenge of navigating a complex healthcare system increases the burden on individuals, their families and carers. Derived from the Chronic Care Model (Green et al. 2012), the concept of patients having a Medical Home to manage their primary healthcare needs has been widely supported across the United States through the Patient-Centered Medical Home (PCMH) model (Peikes et al. 2012). Australia’s Health Care Home model has been adapted from the PCMH to address the complex healthcare needs of people living with chronic conditions within a different fiscal environment (Commonwealth of Australia 2016).

Key attributes of the PCMH model include: (i) each patient has a primary care physician responsible for fostering a supportive relationship to deliver whole-person, coordinated health care; (ii) the physician is located as part of a wider practice team working collaboratively to support the primary physician–patient partnership; (iii) care is coordinated, with a focus on using technology to support health information exchange; (iv) care is available and accessible as required; and (v) patients are actively encouraged to participate in healthcare decisions (American Academy of Family Physicians 2008). Fig. 1 shows the PCMH model. Although paediatric Medical Homes were developed in the 1960s, Medical Homes for adult populations have only been widely established after 2007, following endorsement of the PCMH by leading American physician organisations (Baird et al. 2014).

Fig. 1.  Patient-Centered Medical Home (PCMH) model of care.

Both the PCMH and Health Care Home models aim to deliver sustainable health care and improve patient outcomes and experience (American Academy of Family Physicians 2008; Commonwealth of Australia 2016). Since endorsement in 2007, evaluations of the PCMH have primarily been instigated by insurers to appraise service funding or by practice managers as part of quality improvement strategies. Although understanding patient experience is a crucial part of the continuous quality improvement cycle of PCMHs, very few studies have specifically focussed on examining this in detail (Aysola et al. 2015). There have been some evaluations in paediatric populations; however, the experience of adults living with chronic conditions differs from this cohort.

A search of PROSPERO in May 2016 for systematic reviews of patient experience in the PCMH identified nil results. Searches using Medline, CINAHL and Scopus identified several reviews of the implementation of the PCMH, some of which examined patient experience as part of a range of influencing factors, but none of which specifically scrutinised the adult patient experience in the Medical Home. Our review was conducted to address this knowledge gap, by assessing the scope of literature in which adult patients have reported their experiences of using a PCMH model of care.

A five-step methodological framework was used for this scoping review (Arksey and O’Malley 2005; Levac et al. 2010; The Joanna Briggs Institute 2015).

The primary research question was: what is the adult patient-reported experience of using the Patient Centered Medical Home model?

The strategy for article inclusion is outlined in Box 1. Using these strategies, 631 articles were identified for review. Fig. 2 outlines the selection process for articles.

Fig. 2.  Flow diagram illustrating process for article inclusion.

Data were extracted for each study using a charting table. Information recorded included: author(s); year of publication; study aim; population of interest; methods including study design, sample size, data collection instrument; study outcome; and, study strengths and limitations.

Results were collated and summarised to address the research question, then examined to determine if the individual PCMH model attributes had been measured or described. A thematic approach derived from individual PCMH model attributes was used to describe patient-reported experience within PCMHs.

The majority of studies were classified as quantitative (n = 29), followed by qualitative (n = 8) and mixed methods (n = 2) (Table 1). Quantitative studies collected data from patients using a mix of validated (n = 24) and non-validated survey tools (n = 5) (Table 1). Validated tools included the Consumer Assessment of Healthcare Providers and Systems Clinician & Group (CAHPS-CG) (Agency for Healthcare Research and Quality 2017) and Press Ganey surveys (Press Ganey 2017). The quantitative studies included a mix of descriptive only and comparative study designs (Table 1), with comparisons being made pre-post PCMH implementation (n = 3) (Coleman et al. 2010; Kern et al. 2013; Carrillo et al. 2014); between PCMH and non-PCMH sites (n = 3) (Christensen et al. 2013; Maeng et al. 2013); and both pre-post implementation, PCMH and non-PCMH sites, using a quasi-experimental study design (n = 6) (Reid et al. 2009, 2010; Jaén et al. 2010a; Nutting et al. 2010; Fishman et al. 2012; Heyworth et al. 2014). Analyses from existing large-scale survey datasets were also undertaken (n = 6) (Beal et al. 2009; Solberg et al. 2011; Thygeson et al. 2012; Lebrun-Harris et al. 2013; Nelson et al. 2014; Reddy et al. 2015). There were 23 independent data sources identified, as some articles used the same research investigation technique, specifically, studies that examined the Group Health Medical Home pilot (n = 4) (Reid et al. 2009, 2010; Coleman et al. 2010; Fishman et al. 2012); evaluations conducted as part of the National Demonstration Project (n = 2) (Jaén et al. 2010a; Nutting et al. 2010); studies that examined Medical Home implementation in Florida (n = 2) (Cook et al. 2015, 2016); and studies from Health Partners Medical Group (n = 2) (Solberg et al. 2011; Thygeson et al. 2012). The majority of quantitative studies examined patient experience as part of a broader evaluation of PCMH implementation.

Table 1.  Summary characteristics of studies included in this review (n = 39)
PCMH, Patient-Centered Medical Home; MEPS, Medical Expenditure Panel Survey; RN, Registered Nurse; CAHPS-CG, Consumer Assessment of Healthcare Providers and Systems Clinician & Group; PACIC, Patient Assessment of Chronic Illness Care; PCAS, Primary Care Assessment Survey; ACES-SF, Ambulatory Care Experiences Survey; ACGME, American Council for Graduate Medical Education; SHEP, Survey of Healthcare Experiences of Patients; CPCI, Components of Primary Care Index; DDSM-QM, Diabetes Disease State Management Questionnaire

Attributes derived from the Joint Principles of the Patient-Centered Medical Home model were identified within each study. Table 2 identifies each attribute and outlines the study designs used to measure patient experience of the attribute.

Table 2.  Investigation of Patient-Centered Medical Home (PCMH) model attributes, by study type
Data are presented as n (%)

Although a significant portion of studies investigated the patient–provider relationship in the PCMH (Table 2), the results were mixed. Four studies reported a slightly higher level of satisfaction among patients regarding the care they received from their physician following implementation of the Medical Home model (Solberg et al. 2011; Carrillo et al. 2014; Hall et al. 2014; Heyworth et al. 2014). One study reported an improvement in patient satisfaction after 12 (Reid et al. 2009) and 24 months (Reid et al. 2010) of PCMH care; however, this improvement in doctor–patient interaction diminished over time. Three studies that examined improvement in patients’ perception of their relationship with a primary care physician found no change following implementation of the PCMH model (Jaén et al. 2010a; Kern et al. 2013; Reddy et al. 2015).

Other studies reported a high level of patient satisfaction with their provider in the PCMH model, although these studies reported on satisfaction with care at only one point in time (Day et al. 2013; Kennedy et al. 2013; Lebrun-Harris et al. 2013; Cook et al. 2015, 2016; Wagner et al. 2015) with no comparison to a non-PCMH site or to any change in the care model over time.

Two studies comparing the experience of patients in a PCMH with traditional care sites reported a slightly better patient experience of provider communication in the Medical Home (Beal et al. 2009; Christensen et al. 2013). A study of the Geisinger Health System, ProvenHealth Navigator, PCMH model found no improvement in the patient–provider relationship between the PCMH and non-PCMH sites, although this was measured using a non-validated survey tool (Maeng et al. 2013).

Some investigations that examined influencing factors in the patient–physician relationship found that personal physician engagement and communication with patients significantly improved post implementation (Christensen et al. 2013; Heyworth et al. 2014; Cook et al. 2015). Contrastingly, another study found that patient–provider communication did not improve and overall physician rating did not improve with implementation of the PCMH model (Kern et al. 2013). Patients reported an improvement in their perception of the time spent in consultation with the physician in the PCMH model (Kern et al. 2013), and a positive correlation was observed in several studies for physician continuity and patient satisfaction in the PCMH (Fishman et al. 2012; Takane and Hunt 2012; Day et al. 2013; Wagner et al. 2015).

The importance of the patient–provider relationship was explored in qualitative studies, with a range of positive experiences reported (Takane and Hunt 2012; Fix et al. 2014; Aysola et al. 2015; Wagner et al. 2015). If patients had a positive relationship with their physician, this was seen to be of greater importance than any challenges encountered when accessing care (Aysola et al. 2015).

There was insufficient evidence to determine if implementing the PCMH model improves patient experience with practice staff. Two studies demonstrated an increase in patient satisfaction with office staff within the PCMH model compared with the traditional model of care (Christensen et al. 2013; Hall et al. 2014). Other researchers described a high level of positive patient experience in the PCMH when it came to practice staff providing respectful, helpful care (Cook et al. 2015, 2016) and in terms of friendly, helpful staff (Kennedy et al. 2015). Contrastingly, two independent cross-sectional studies, using the traditional care model as the comparative group, found no difference in the helpfulness of practice staff between the groups (Reid et al. 2009; Maeng et al. 2013). In a study that measured the change in patient experience over time, the perceived helpfulness of office staff improved; however, across a 15-month timeframe, overall patient experience with office staff did not improve (Kern et al. 2013).

Current evidence indicates that the PCMH may improve care coordination, although the results are mixed. Several studies reported an improvement in the patient experience of care coordination in PCMHs (Reid et al. 2009, 2010; Maeng et al. 2013; Schmidt et al. 2013; Carrillo et al. 2014; Hall et al. 2014). However, it is worth noting that two of these studies did not use validated survey tools.

In contrast, five other studies identified no improvement in patient satisfaction with care coordination in the PCMH model (Jaén et al. 2010a; Nutting et al. 2010; Kern et al. 2013; Heyworth et al. 2014; Reddy et al. 2015). There was no improvement for the patient experience of follow up of test results in the PCMH following implementation (Kern et al. 2013) or when compared with a non-PCMH site (Maeng et al. 2013).

For studies that described patient experience, one study reported no patient concern relating to care coordination in the PCMH (Aysola et al. 2015), whereas other studies had mixed results for care coordination (Fix et al. 2014; Wagner et al. 2015). Patients reported very positive experiences in the PCMH model for provision of reminders and administration (87 and 93.7% agreement) (Cook et al. 2015, 2016) and for support provided by health navigators (Janiszewski et al. 2015). A more moderate level of positive experience was reported for test follow up (83.9 and 78.6% agreement) (Solberg et al. 2011; Cook et al. 2016). Patients with chronic conditions that required more than three visits per year reported better coordination and service experience than patients who had fewer visits (Cook et al. 2016).

Accessing service in the Medical Home is a priority for patients (Janiszewski et al. 2015) and there was some evidence to suggest that the PCMH model can improve patient access to care. Along with care coordination, access to care was the most commonly investigated attribute of the PCMH model across the range of studies included in this review (Table 2). Kern et al. (2013) observed that access to care had the least patient satisfaction at baseline and the most potential for improvement under the PCMH model.

A range of studies reported improvement in patient perceived access to care in the PCMH (Reid et al. 2009, 2010; Solberg et al. 2011; Fishman et al. 2012; Christensen et al. 2013; Kern et al. 2013; Schmidt et al. 2013; Carrillo et al. 2014; Hall et al. 2014; Jubelt et al. 2014). Specific areas of improvement were: ease of appointment scheduling; access to routine appointments; ability to obtain urgent appointments; and, reduced in-office waiting time. In the Group Health studies, better access to care was observed at 12 months (Reid et al. 2009), with continuing improvement in access to care at 24 months (Reid et al. 2010). This is in direct contrast with the patient–provider relationship model attribute, which in these studies, was shown to improve, albeit at a diminishing rate over time. Another study reported overall positive ratings (63%) for patient experience in accessing care across 26 safety-net clinics. Safety-net clinics deliver care to vulnerable populations, and there was a positive association for increased access to care to small- and medium-sized clinics when compared with larger clinic sites (Schmidt et al. 2013).

Although a range of studies demonstrated that the PCMH care model can enhance patient access to care, there were also a collection of studies that found no significant improvement (Jaén et al. 2010a; Nutting et al. 2010; Solberg et al. 2011; Maeng et al. 2013; Schmidt et al. 2013; Heyworth et al. 2014; Aysola et al. 2015; Reddy et al. 2015). Two studies of PCMHs in Florida using the same cohort reported limited access to care, both in and out of hours (Cook et al. 2015, 2016). Patients reported poor satisfaction for in-clinic waiting time (Day et al. 2013); getting an appointment (Kennedy et al. 2013, 2015); and no improvement in patient satisfaction for post-appointment access to care in the Medical Home (Solberg et al. 2011). Patients provided mixed results for timely access to care in the Veterans Woman’s PCMH (Wagner et al. 2015).

Descriptive studies examined a range of characteristics related to access to care in the PCMH. Access to care was an important component of improving patient satisfaction and patient perception of care quality (Lebrun-Harris et al. 2013). Patients identified that improvements in appointment scheduling and reduced in-clinic wait time would improve their experience in the PCMH model (Kennedy et al. 2013, 2015).

There was limited investigation into the patient engagement, activation and shared decision-making model attribute. A military population study that compared PCMH with non-PCMH sites found a higher level of patient activation in the PCMH (Christensen et al. 2013). The Group Health PCMH evaluation identified improvements in patient activation, involvement and goal setting at 12 months (Reid et al. 2009). At 24 months, the improvement continued for patient activation and goal setting (Reid et al. 2010), and although patient involvement was still improving, it was at a diminishing rate (Reid et al. 2010). Senior patients in the Group Health’s PCMH study reported an improved experience with shared decision-making, when compared with controls (Fishman et al. 2012).

By contrast, most patients in 24 safety-net clinics did not identify that patient activation improved under the PCMH care model; however, for the cohort of patients experiencing the poorest level of health, there was an association between an increased uptake of the PCMH model and perceived clinic support for patient activation (Nocon et al. 2014). This result is important as it signals the potential for the PCMH model to promote patient activation in underserved minority groups.

Aspects of patient activation were explored in a survey of patients enrolled in five Florida PCMHs, and a moderate level of positive experience was described for patient goal setting. Very few patients, however, reported that they received recommendations on education to improve their own health (23.6%) (Cook et al. 2015). Patients were satisfied with their opportunities for shared decision-making in the Veterans Woman’s PCMH (Wagner et al. 2015). In a qualitative investigation, most participants identified the importance of a supportive patient–doctor relationship to promote shared decision-making (Aysola et al. 2015).

Overall, this review found mixed evidence that the PCMH model improves adult patient-reported experience across the five attributes described in the ‘Joint Principles of the Patient-Centered Medical Home’ (American Academy of Family Physicians 2008). The importance of the primary patient–physician relationship was supported, but the extent to which PCMH implementation affects this relationship is unclear. Evidence suggests that some aspects of care coordination and access may improve for patients in the PCMH. Results for all model attributes are limited by the scope of existing evidence, with the patient–practice relationship and patient engagement, activation and shared decision-making attributes being the least investigated.

A lack of discernible effect on patient experience following PCMH implementation may be attributable to the model structure. Some approaches, such as improvements in care coordination, are in the background to service delivery. These strategies may not directly affect patient’s perceptions of their experience of care. Further, patients who currently utilise practices with high levels of service delivery may not be notably affected by changes resulting from PCMH implementation (Maeng et al. 2013). This observation has the potential to affect patient experience evaluation of the Health Care Home, as practice site participation is voluntary, indicating a willingness by the practice to participate in strategies that aim to improve quality of service.

This review was conducted using a structured framework reflecting a leading methodological approach; a comprehensive search strategy was used and references were checked in the identified literature. Given, however, that ‘patient satisfaction’ and ‘patient experience’ are terms not clearly defined, there is potential for literature to exist and not have been included in this review. Similarly, although the search strategy for the ‘Patient-Centered Medical Home’ was detailed, there is the potential for derivatives of the terminology to have been missed.

Research that examines the experience of patients as they interact with healthcare services is difficult to assess with consistency. The measurement of patient experience is subject to potential bias, as it is based on a perception of care not an objective measure of care delivery, generating ongoing debate on ways to measure patient experience (Berkowitz 2016). Further, our study found there was an absence of quantity and rigor when evaluating the patient experience in the PCMH. A significant portion of quantitative studies used descriptive, single-measure designs generating commentary, but without the ability to determine the effect of model implementation. Identifying a paucity of investigation into patient experience in the PCMH is consistent with previous studies (Nocon et al. 2014; Aysola et al. 2015), although this is the first review to specifically quantify the evidence for individual PCMH model attributes.

Using validated measuring tools enables comparison across populations and within populations and has the potential to promote consistency in evaluation. In Australia, the validated Patient Partnership in Care (PPiC) tool (Powell et al. 2009), which incorporates patient-reported experience and outcome, is indicated to evaluate trials of the Health Care Home. Examining patient experience within the Australian primary healthcare context is challenged, however, by a lack of publicly available survey instruments, the limited publication of survey responses and a corresponding absence of independent review (Gardner et al. 2016). Australian policymakers have the opportunity to learn from international experience. In the United States, patient experience is measured as part of the quality improvement cycle of PCMH accreditation (Quigley et al. 2015). A standard survey tool used to measure patient experience is the freely available Consumer Assessment of Healthcare Providers and Systems (CAHPS) instrument, which includes a subset of PCMH-specific questions (Agency for Healthcare Research and Quality 2017). Results from the CAHPS surveys are publicly available, enabling practices to benchmark their performance and providing the opportunity for comparative evaluation. In the United Kingdom, patient experience is measured annually by the large-scale GP Survey, with the results being utilised to inform patient decision-making through an easily accessed consumer website (NHS England, see https://gp-patient.co.uk/practices-search, accessed 19 July 2017).

It is worth appraising survey measures to ensure they are population-appropriate and that variability between practices is considered. Given that the Health Care Home trial sites include practices in metropolitan, regional and remote communities, as well as Aboriginal Community-Controlled Health Services, there is a need to tailor the evaluation to include a diverse range of patient experiences. Several study authors have espoused the use of mixed-methods approaches to measure patient experience in the Medical Home (Jaén et al. 2010b; Goldman et al. 2015), combining qualitative investigation, to determine contextual detail from the distinctive patient group, with quantitative investigation, using rigorous, validated survey methods to promote generalisability of results to the wider population.

Improving patient experience has been identified as one of the key reasons to implement the PCMH care model by primary care physicians in the United States. This is the first study to explore the patient-reported evidence for each attribute of the PCMH model.

Our results suggest that the patient experience of their relationship with providers and access to care in the Medical Home were the most commonly investigated model attributes, with some positive findings for implementation of the care model. Patient engagement, activation and shared decision-making, along with patient experience with practice staff and other team-care health professionals were model attributes that had a significantly limited scope of existing evidence. Generally, all model attributes lacked rigorous, detailed investigation, and an increased research agenda is proposed to determine whether implementation of the Health Care Home model can improve the patient experience of health service.

The authors declare that they have no conflicts of interest.