FAMILY medicine is associated with superior health care delivery in countries in which it is best-implemented (Starfield et al., 2005). Implementing family medicine research-based solutions into practice and policy is still reported as a challenge and the principles of family medicine are affected consequently (Alison & Hill, 2011 ; Mangione-Smith et al., 2007 ; McGlynn et al., 2003). Patient-centered medical homes (PCMH) emerged in the United States as a set of standards of best research conclusions to improve care in family medicine with increasing reports of its effectiveness (Annis Emeott et al., 2013 ; Hoff et al., 2012). Nevertheless, few reports indicated the contrary (Jackson et al., 2013) and some debated for the need of a newer model of presentation of PCMH that simplifies the complexity of its implementation (Wasson, 2017).
United Arab Emirates since early 1970s had provided for its citizens a free access to network of primary health care centers. All resourced with multidisciplinary teams of physicians, nurses, and pharmacists. The network covers all UAE distributed by catchment area for the population. In 2008, the Abu Dhabi Health Services Company (SEHA) centralized the management of all 47 centers in the Emirate of Abu Dhabi, the largest Emirate in the UAE, as Ambulatory Healthcare Services (AHS). Unified electronic medical records (EMRs) were concomitantly implemented across AHS and the 6 government hospitals that provided informational continuity and facilitated care coordination. A major change in the same period was the introduction of an advanced payment system through nationwide health insurance. The SEHA manages all government health institutions, including hospitals and AHS clinics under one governance structure with unified policies and regulations that are monitored and tracked through a dashboard of numerous clinical and operational key performance indicators (KPIs).
In a strategic plan to provide best care for the Abu Dhabi population, the National Commission for Quality Assurance (NCQA) PCMH standards were thought to provide evidence-based care strategies and provide corrective measures for existing gaps in structure, operation, and care outcome. Essentially, this study seeks to determine whether the implementation of a more consistent care process within the Abu Dhabi AHS improved quality, health promotion, and access to care and whether PCMH implementation was beneficial, neutral, or harmful to health care within the first year.
Pragmatic, quantitative, before and after quasi-experimental study investigating health services' intervention assessment.
Study timeline: The experience started on January 2014 and the 3 pilot centers outcome was compared with other non–pilot centers. For another 3 AHS centers, it was planned to start PCMH implementation 6 months after the first group and therefore to best show impact of the intervention it was separated in the analysis. Rural centers were excluded from the control to decrease variations in the control group.
Data sources and outcome measures
Quality of care
As part of SEHA, AHS clinics have established a tracking system for KPIs, which includes clinical and operational criteria. The introduction and definition of criteria and KPI targets is performed by the SEHA central-quality department, while data extraction, reporting, and presentation are conducted by the AHS quality department. Clinical KPIs reported in this study include diabetes, hypertension, asthma, and breast and cervical cancer screenings. The following were the inclusion criteria: for patients, treated at the center for at least 2 visits for a condition tracked by a KPI in the last year, and for preventive KPIs, any woman meeting screening criteria who visited the center. These data were extracted from the EMR-generated reports. These reports are available to the centers so that they can extract data and identify gaps in care. Then it is recommended that patients be called to improve KPIs. No sample size calculation was needed, as data for all patients who met these criteria were extracted.
The patient experience (PE) program operated by SEHA distributes quarterly surveys to monitor patient satisfaction. Two new questionnaires, the Patient Assessment of Chronic Illness Care (PACIC) and the Customer Assessment of Healthcare Providers and Systems (CAHPS) for adults and children were translated, reverse translated, and validated at the pilot centers before they were used to collect data by the PE team at each center. The PE team was trained to use the questionnaires before October-November 2014. The response rates for the AHS PE questionnaires are usually above 80%, although no specific response rate was tracked for these new questionnaires. Data were collected for the pilot centers and 6 other centers, which were thought to have very similar characteristics to the pilot centers. For the PACIC, 113 interviews were conducted. A sample of 106 was needed for confidence interval of 9.5 and confidence level of 95% to survey the chronic disease population of these centers.
With regard to the CAHPS 160 for the CAHPS child and 281 for CAHPS adult were collected. Patient experience personnel were trained to administer the questionnaires, which were completed by interviews and entered by the patient experience personnel within 4 weeks.
Emergency room (ER) and hospital admission data were extracted from EMR charts. Data were extracted for 2013 and 2014 for the pilot centers and all other urban AHS centers.
Staff satisfaction was assessed on a small scale using direct questions about the introduced changes based on the NCQA PCMH standards. Five subscales were measured: access to care, communication with patients, communication with other providers, tracking data, and care management and quality improvement. Reponses were collected through self-administered questionnaire. To assess burnout, a self-administered questionnaire based on previously validated Maslach Burnout Inventory Human Services Survey (MBI-HSS) questionnaire was used as well. One hundred fifteen completed the survey.
The 3 centers were chosen by the steering committee, as they were academic centers with which the implementation team was working. It was thought that this would accelerate the implementation process. All Abu Dhabi AHS centers are distributed to cover the catchment areas of the population served. All population have close access to an AHS center that is fully resourced. Rural centers have in addition 24/7 access doctors, nurses, and pharmacists. No difference existed between pilot and non–pilot centers at the start of the project (Table 1).
The implementation team met weekly for the first year, and all 3 centers relied on the same framework and care improvement strategies but differed in their timelines, with centers starting within a week or 2 of each other depending on their settings and preparedness levels.
The project timeline, the intervention components, and detailed operational implementation plan are shown in Table 2. This was an organization-wide adaptation involving all departments supporting the project and the pilot centers. The areas were based on the NCQA PCMH standards and included patient-centered access and continuity, team-based care and practice organization, knowing and managing patients, care management and support, care coordination and care transitions, and performance measurement and quality improvement. The project was facilitated by strong health informatics, resourceful organizations, and the government population management programs implemented by the centers over many years, including the national vaccination program; the well child program; and the adult preventive program, Weqaya (the national cardiovascular screening and prevention program) (Hajat et al., 2012); premarital, preemployment, and school health screenings; and travel health, mental health, and chronic kidney disease programs. These programs aided in implementing the PCMH, as they required the management of various identified populations, and the teams already considered population management one of the centers' functions. All government programs are endorsed by decree, and the organizational initiatives for chronic diseases were approved policies by the AHS in 2010; therefore, centers cannot choose to provide these services. The latter are based on the chronic disease model (Bodenheimer et al., 2002) successfully implemented by the Al Ain centers since 2006 (Baynouna et al., 2010) and rolled out to all the Emirate centers since 2010.
Table 2 summaries all interventions that were piloted in the 3 centers and very few were shared among all centers as empanelment, which was felt already to be a necessity not in need of piloting and quality tracking and reporting which promote competitiveness of the centers to reach KPI targets.
Analysis was done using SPSS v21. Frequencies and cross-tabulation were used to summarize measures. T tests were used to detect significant improvements in the outcome of P value of less than .05. Power was adequate, as all patients with the KPI diagnosis and definition were included. Linear regression was used to control for confounders of ER utilization and hospital admissions.
The pilot centers showed better performance overall. For the clinical KPIs, all 3 pilot centers had showed significant differences in the number of KPIs met compared with all other centers only 3 months after starting the project. At 6 months, the difference between groups was marked (Figure). Most importantly, the pilot centers maintained their lead in the number of KPIs, meeting their targets even after 18 months, although the other centers showed improvement as all new initiatives were implemented. The drop in Q4 in all centers was due to new KPI added that needed time to be improved and the pilot started at lower level and then led as better performing early 2015. However, no difference was noted among those with very poorly controlled diabetes, hypertension, or lipids.
With regard to patient satisfaction, the PACIC questionnaires showed a clear distinction between the pilot and the non–pilot centers, and the difference between them was as high as 15%. In responding to the statement “hospital care was preferred over the center,” approximately 70% of patients disagreed in the pilot centers compared with 50% in other centers.
There is an indication of better utilization of ERs in the pilot centers. Before implementing the project (2013), non–pilot centers had less patients admitted to the hospital and more patients with ER visits: P = .004 (odds ratio [OR] = 0.014 [0.005-0.024]) and P = .005 [OR = −0.051 (−0.086 to −0.016)], respectively. After the implementation of the project (2014), the pilot centers had less ER visits: P = .036 (OR = −0.037 [−0.072 to −0.002]), and there was no difference in hospital admission rate: P = .68 (OR = 0.008 [−0.001 to −0.017]). Nevertheless, the change is small and the large population size would have affected this result as significant. In reporting outcome measures, it is worth mentioning that there were fewer incidents reported through the organization incident reporting system, fewer complaints, and patient safety notification, which declined in the pilot centers more than in the other AHS centers.
In process of care measure, collectively auditing the pilot centers based on the NCQA 2011 criteria for PCMH accreditation yielded a score of 84% in April 2015, which represents an increase of 23% since an external audit was conducted in 2012. Adherence to the standards was evident through objective measures such as empanelment rates. More than 80% to 90% of patients visiting AHS centers during the last quarter of 2013 were assigned to a Primary Care Physician (PCP), with all 3 pilot centers reporting rates of 90%. Continuity of care was monitored using the Continuity Index (CI) of the PCP (total number of visits to the PCP/total center visits) and the CI of the most frequently seen doctor (total number of visits to the most frequently seen doctor/total center visits). Data entered regarding PCP allocation were later found to require cleaning, as a large number of patients had not been allocated to their PCP and had been seen in the last 6 months more frequently by another physician.
An important process measure was medication reconciliation, which was significantly better in the pilot centers than in all other AHS centers. All pilot centers achieved rates above 90% (92, 94.9, and 100). By comparison, among non–pilot centers, only 1 was above 90%. The remainder reported rates of less than 90%, with the lowest rate at 1.56%.
Another important process measure was test result tracking, which reached 100% in all pilot centers; manual audits of the physicians' use of EMR message centers in the 3 pilot centers indicated that more than 90% of doctors cleared their message centers within 3 days. It is worth mentioning that test result tracking was not an activated function of the EMR system before the project implementation; thus, the use of EMR message centers was very minimal and variable among doctors before the project, as it was a personal choice.
With regard to staff satisfaction, high subscale score were found in the following patient-centered medical domain: access, communication skill, tracking data, care management, and quality improvement.
Burnout was reported to be 21.7% for high emotional exhaustion and 23.5% for moderate emotional exhaustion. Burnout increased in middle age group, married female, staff with less than 5 years of experience, and staff working in urban area. Burnout decreased in the staff with more than 5 years of experience.
This project started from significant system strengths: free access to health services, same-day appointments, Emirates-wide health information systems and EMRs, leadership commitment, teaching centers whose faculty members were experienced in data management, and well-established preventive and chronic disease programs. Chronic disease programs had already been implemented in the centers; thus, for most Abu Dhabi centers, part of PCMH implementation was not completely new, as the chronic diseases programs share many requirements (Baynouna et al., 2010). Similar to our experience, a study of the implementation of the Chronic Care Model and PCMH through regional Breakthrough Series learning collaboratives, supported by Improving Performance in Practice (IPIP) coaches, with required monthly quality reporting enhanced by multipayer infrastructure payments, produced significant improvements that were evident from the first year (Gabbay et al., 2011). Nevertheless, initiatives targeted at closing PCMH standards gaps, such as empanelment, and optimized Health Information System (HIS) for coordinating, tracking, monitoring, prescribing safety, and managing population initiatives added to the existing programs and produced significant improvements over other centers. Moreover, the concept of teamlets endorsed team members' empowerment. Note that all AHS centers have been continuously working on the mentioned clinical, operational, and financial KPIs since 2009 and that these are reviewed quarterly in high-level meetings with the senior management team, the CEO, and the executive committee. The distinctive improvements seen in the pilot centers showed the effects of the changes and new initiatives. However, this is not the case with utilization, CAHPS, or PACIC, which all were measured for the first time. As no center was targeting these measures, they show the genuine impact of the project.
Improvements with PCMH standards implementation have been commonly reported in the literature, which has made PCMH a well-recognized solution for transforming primary care (Kevin Grumbach et al., 2009). Nevertheless, a few contradicting reports emerged, such as a study conducted on Rhode Island's chronic care sustainability initiative pilot program. Rosenthal et al. (2013) reported that for more than 2 years the substantial improvements in medical home recognition scores did not significantly reduce emergency department visits or inpatient admissions, nor were clinical KPIs improved by the pilot program compared with control practices in that study (Rosenthal et al. 2013). This variability in the outcomes of PCMH standards implementation in the United States has been explored (Annis Emeott et al., 2013 ; Hoff et al., 2012). A recent systematic review (Janamian et al., 2014) identified challenges based on the US experience: with transformation and change management in adopting a PCMH model, difficulties arose with electronic health records, funding and payment models, insufficient practice resources and infrastructure, inadequate measures of performance, and inconsistent accreditation and standards. In Abu Dhabi, the first year revealed major challenges: insufficient staffing of doctors and nurses, as well as the need for new roles such as medical assistants and for more care coordinators and self-management nurses. As we start to see panel size increases and plans to incorporate complexity when determining the size (Murray et al., 2007 ; Schaefer et al., 2010), shortages of physicians and nurses may appear, which is anticipated by both management and physicians.
In Singapore, the initial report on implementing PCMH outside of the United States has demonstrated its potential to improve patient access to medical care and to reduce hospital-based care, but no detailed results have been reported (Koh et al., 2013).
Regarding the use of NCQA PCMH standards audit scores, there are standards whose scores will be high, such as test result tracking, medication safety, disease and population management, and quality monitoring, as previously explained. Some health system deficiencies cannot be solved immediately, such as off-hours coverage provided by the same center staff, although there is informational continuity if patients visit any government hospital. However, some standards are more mature than others. For empanelment, optimized HIS enabled the team to assess intervention fidelity by measuring the difference between the PCP assignment and the most visited doctor, which was found to be large. Consequently, it was decided to begin cleaning the PCP data, beginning 1 year after empanelment was started as a corrective action that was expected to impact the project outcomes tremendously. Because empanelment is fundamental to PCMH implementation and continuity, panel and population management are dependent on it, with all the responsibilities and care accountability to patients' care that entails. Although empanelment is new in the AHS and surrounding health care systems, its status may not be far from the US experience. In a recent review, only one-third of the family physicians knew the size of their panel, indicating the value of the focused measurement of what is expected from an intervention as a process rather than only as an outcome and therefore validating the anticipated link between both.
A similar area is self-management, which reflects human relationships and capacity building, but at the same time, a plateau can be reached, with teams exerting more effort without the resources or required support. This may impact care between visits and provide crucial support for the patients. In addition, although the role of the care coordinator is a new full-time job, 2 were hired for each center. However, they rapidly found their hands full, and there was a need to increase their numbers. This need is due to their role in key project components, such as test result and referral tracking, preplanned visits, and huddles, which made care coordinators essential members of the teamlets. Care coordinators and self-management personnel roles were essential in the US experience, and in many reported experiences, it was the main external support (Biernacki et al., 2015 ; Bodenheimer et al., 2014).
Attributes of success
It is thought that the structured care approach to providing broad population management, which is part of the network design of care, worked best for the Abu Dhabi community. Although staffing shortages may have affected the full implementation of the model and resulted in overlap, in the pilot centers, structured care with colocation of teamlets for the different populations and strict scheduling increased the transparency of patient flows and controlled and facilitated continuity. The doctor sees his or her panel of patients within differently prepared and empowered teamlets based on the population to which they are assigned. Otherwise, they are seen in general family practice, which is a challenge in that the vast majority of patients walk in and do not have appointments. In Abu Dhabi, this is a reality for a number of factors: patients usually prefer a PCP of their gender, centers are open for 2 equal shifts, and doctors have to work 1 to 2 evening shifts per week. In addition, preventive programs are not currently under the PCP but are scheduled separately. With PCP cleanup and as the culture becomes accustomed to the PCP concept, it is hoped that continuity will improve. Although characteristics may not be fully transferable, they may be informative to other systems. The improved outcomes in the pilot centers are largely explained by this system, and population management is known as a means to meet the Triple Aim of improving quality, cost, and patient satisfaction (Berwick et al., 2008). For example, it was estimated that in the case of chronic disease, 50% of the cost of health care is for 5% of the population. Therefore, population management became a prominent method to improve this situation (Care Continuum Alliance, 2012), especially with advanced EMR functionality (Cusack et al., 2010). In 1998, Rivo et al. described a 4-step approach to starting population management: choose common conditions, identify practice patients with the conditions, choose measurable outcomes that reflect the best evidence-based medical practices, and regularly measure and try to improve these outcomes (Rivo et al., 1998). The structured care approach is best implemented in chronic disease programs with colocated teamlets supported with ongoing training and continuous optimization of EMRs and health informatics, which had strong influences on KPI improvement, which were mainly related to diabetes and hypertension.
Electronic medical record
The EMR reports provided timely feedback on implementation, which resulted in ongoing responses to emerging results and accelerated the project. However, the AHS EMR system lagged behind in 2 areas. The first was the lack of register functionality, which was substituted by a snapshot of the list of patients with the required information and reflected on frequently. The other area was the lack of patient portals. Both of these are currently in the process of being added by SEHA.
Finally, Bleser et al. (2014) found that effective communication, internal PCMH campaigns, effective resource utilization, and creation of a team environment were essential to achieve (Bleser et al., 2014) which was felt strongly in this project implementation.
Perceived challenges to implementation include the diverse backgrounds of health care professionals, where less than 20% of practicing doctors are family physicians. A high percentage of nurses are non-Arabic speaking, with relatively high staff turnover. Another perceived challenge was patient acceptance. However, there were opportunities for well-established EMR and payment systems, as well as health insurance. Although the latter improved access, it threatened continuity via consumers' perceptions of their right to access care at any health care organization, resulting less coordination of care by the primary health care provider.
Limitations of this study include that it is an initial report, with major changes being introduced and their maximum effect needing years to materialize. Similar to many studies in health services' research, it is conducted just over a year and the anticipated effects and changes may take decades to produce results because health services interventions depend on sustained relationships among providers and between providers and patients.
Nevertheless, the reported results are very promising. This study contributes detail about how PCMH can be performed in settings beyond those for which it was originally developed. Longer-term studies or analyses are required to ascertain the effects of PCMH uniform standards on the quality, efficiency, and equitability of care.
This study highlights preliminary clinical experiences in the implementation of PCMH criteria within a well-established care delivery system by sharing the barriers for implementation and operational strategies to enhance care delivery in a pragmatic method providing a more solid path to meeting the Institute for Healthcare Improvement's Triple Aims. This article presents preliminary work, with forthcoming research to more rigorously track metrics and efficiency and quality data and examine staffing roles. The PCMH metrics are dynamic and subject to interpretation and considerable revision in the United States; hence, over time, conclusions based on past metrics may need to be refined or reexamined. The observed improvements in utilization, as well as operational and clinical KPI, justify further work, such as strengthening self-management, off-hours coverage, and coordination with other levels of care and within the community.
Alison P., Hill M. B. (2011). Promoting continuity of care in general practice. RCGP Policy Paper. London, UK: The Royal College of General Practitioners.
Annis Emeott A., Markovitz A., Mason M. H., Rajt L., Share D. A., Paustian M., El Reda D. (2013). Four-year evolution of a large, state-wide patient-centered medical home designation program in Michigan. Medical Care, 51, 846–853.
Baynouna L. M., Shamsan A. I., Ali T. A., Al Mukini L. A., Al Kuwiti M. H., Al Ameri T. A., Omar A. O. (2010). A successful chronic care program in Al Ain-United Arab Emirates. BMC Health Services Research, 10, 47.
Berwick D. M., Nolan T. W., Whittington J. (2008). The triple aim: Care, health, and cost. Health Affairs (Millwood), 27, 759–769.
Biernacki P. J., Champagne M. T., Peng S., Maizel D. R., Turner B. S. (2015). Transformation of care: Integrating the registered nurse care coordinator into the patient-centered medical Home. Population Health Management, 18(5), 330–336.
Bleser W. K., Miller-Day M., Naughton D., Bricker P. L., Cronholm P. F., Gabbay R. A. (2014). Strategies for achieving whole-practice engagement and buy-in to the patient-centered medical home. Annals of Family Medicine, 12, 37–45.
Bodenheimer T., Ghorob A., Willard-Grace R., Grumbach K. (2014). The 10 building blocks of high-performing primary care. Annals of Family Medicine, 12, 166–171.
Bodenheimer T., Wagner E. H., Grumbach K. (2002). Improving primary care for patients with chronic illness. The Journal of the American Medical Association, 288, 1775–1779.
Care Continuum Alliance (2012, October). Implementation and evaluation: A population health guide for primary care models. Washington, DC: Author. Retrieved from http://www.carecontinuumalliance.org
Cusack C. M., Knudson A. D., Kronstadt J. L., Singer R. F., Brown A. L. (2010, July). Practice-based population health: Information technology to support transformation to proactive primary care. Rockville, MD: Agency for Healthcare Research and Quality
. AHRQ Publication No. 10-0092-EF.
Gabbay R. A., Bailit M. H., Mauger D. T., Wagner E. H., Siminerio L. (2011). Multipayer patient-centered medical home implementation guided by the chronic care model. The Joint Commission Journal on Quality
and Patient Safety, 37, 265–273.
Hajat C., Harrison O., Al Siksek Z. (2012). Weqaya: A population-wide cardiovascular screening program in Abu Dhabi, United Arab Emirates. American Journal of Public Health, 102, 909–914.
Hoff T., Weller W., DePuccio M. (2012). The patient-centered medical home: a review of recent research. Medical Care Research and Review, 69, 619–644.
Jackson G. L., Powers B. J., Chatterjee R., Bettger J. P., Kemper A. R., Hasselblad V., Williams J. W. (2013). Improving patient care. The patient centered medical home. A systematic review. Annals of Internal Medicine, 158, 169–178.
Janamian T., Jackson C. L., Glasson N., Nicholson C. (2014). A systematic review of the challenges to implementation of the patient-centred medical home: lessons for Australia. The Medical Journal of Australia, 201, S69–S73.
Kevin Grumbach M. D., Thomas Bodenheimer M. D., Paul Grundy M. D. (2009). The outcomes of implementing patient-centered medical home interventions: A review of the evidence on quality
, access and costs from recent prospective evaluation studies. Washington, DC: Patient-Centered Primary Care Collaborative.
Koh TW., Tai E. S., Wee H. L., Tham T. Y., Sim K. P., Khoo Y. H. E., Lim Y. W. (2013). Piloting the patient-centered medical home concept: Singapore's initial experience. International Journal of Integrated Care, 13(8).
Mangione-Smith R., DeCristofaro A. H., Setodji C. M., Keesey J., Klein D. J., Adams J. L., McGlynn E. A. (2007). The quality
of ambulatory care delivered to children in the United States. The New England Journal of Medicine, 357, 1515–1523.
McGlynn E. A., Asch S. M., Adams J., Keesey J., Hicks J., DeCristofaro A., Kerr E. A. (2003). The quality
of health care delivered to adults in the United States. The New England Journal of Medicine, 348, 2635–2645.
Murray M., Davies M., Boushon B. (2007). Panel size: How many patients can one doctor manage? Family Practice Management, 14, 44–51.
Rivo M. L. (1998). It's time to start practicing population-based health care. Family Practice Management, 5, 37–46.
Rosenthal M. B., Friedberg M. W., Singer S. J., Eastman D., Li Z., Schneider E. C. (2013). Effect of a multipayer patient-centered medical home on health care utilization and quality
: The Rhode Island chronic care sustainability initiative pilot program. JAMA Internal Medicine, 173, 1907–1913.
Schaefer J., Van Borkulo N., Morales L. (2010, December). Safety Net Medical Home Initiative. In Burton T. (Ed.), Patient-centered interactions implementation guide part 2: Engaging patients in their health and healthcare (1st ed.). Seattle, WA: The MacColl Center for Health Care Innovation at the Group Health Research Institute and Qualis Health.
Starfield B., Shi L., Macinko J. (2005). Contribution of primary care to health systems and health. The Milbank Quarterly, 83, 457–502.
Wasson J. (2017). A troubled asset relief program for the patient-centered medical home. The Journal of Ambulatory Care Management, 40, 89–100.