This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
There is a call for bold and innovative action to transform the current care systems to meet the needs of an increasing population of frail multimorbid elderly. International health organizations propose complex transformations toward digitally supported (1) Person-centered, (2) Integrated, and (3) Proactive care (Digi-PIP care). However, uncertainty regarding both the design and effects of such care transformations remain. Previous reviews have found favorable but unstable impacts of each key element, but the maturity and synergies of the combination of elements are unexplored.
This study aimed to describe how the literature on whole system complex transformations directed at frail multimorbid elderly reflects (1) operationalization of intervention, (2) maturity, (3) evaluation methodology, and (4) effect on outcomes.
We performed a systematic health service and electronic health literature review of care transformations targeting frail multimorbid elderly. Papers including (1) Person-centered, integrated, and proactive (PIP) care; (2) at least 1 digital support element; and (3) an effect evaluation of patient health and/ or cost outcomes were eligible. We used a previously published ideal for the quality of care to structure descriptions of each intervention. In a secondary deductive-inductive analysis, we collated the descriptions to create an outline of the generic elements of a Digi-PIP care model. The authors then reviewed each intervention regarding the presence of critical elements, study design quality, and intervention effects.
Out of 927 potentially eligible papers, 10 papers fulfilled the inclusion criteria. All interventions idealized Person-centered care, but only one intervention made
Despite a strong common-sense belief that the Digi-PIP ingredients are key to sustainable care in the face of the silver tsunami, research has failed to produce evidence for this. We found that interventions reflect a reductionist paradigm, which forces care workers into standardized narrowly focused interventions for complex problems. There is a paucity of studies that meet complex needs with digitally supported flexible and adaptive teamwork. We predict that consistent results from care transformations for frail multimorbid elderly hinges on an individual care pathway, which reflects a synergetic PIP approach enabled by digital support.
The combination of increased longevity, more sensitive diagnostics, and improved treatment contribute to the increasing prevalence of multimorbidity [
There is a call for bold and innovative action to reform the current analog, episodic, single disease, profession-centric, and reactive care system to meet the new needs of the population. Health care organizations and researchers in the United States, the World Health Organization, and the European Union are developing roadmaps to deal with the silver tsunami [
However, the agreement and shared understanding of what
There is an urgent need to determine whether health services that take a synergistic Digi-PIP care approach to meet the silver tsunami have a documented effect on the triple aims of population health, patient experience, and cost-effectiveness. As care systems vary significantly in the extent they support and operationalize Digi-PIP-care, the review of the documentation is only meaningful if we can get a grasp on the degree of fidelity and maturity of any given implementation.
Our research questions are as follows:
How are Digi-PIP care interventions operationalized?
How can we capture the maturity of a Digi-PIP intervention?
What is the Digi-PIP study methodological quality, and which effects are reported?
Does intervention maturity matter for effects?
Understanding whether each of the Digi-PIP elements are present or not, and to what degree they are present, is essential to be able to evaluate their effectiveness. In a
However, the individual PIP elements are all poorly defined in the literature. “The term ‘person-centered care’ is used to refer to many different principles and activities, and there is no single agreed definition of the concept.” [
Instead of trying to harmonize definitions from the literature, or inviting professionals to discuss their way to consensus, we have chosen the patient perspective to be our guiding principle. In a previous paper, we show that to patients with complex long-term needs, the essence of care quality was that the care supported their long-term goals, linked to the question:
Below, we describe each of the active elements in PIP care regarding their key characteristics, their care component, and relevant digital support. See also
The Person-centered integrated care quality framework. The walls, foundation, and roof symbolize the structural resources. The cyclical care process in the house center consists of exploring what matters to the person and translating this into relevant and realistic goals for care, which feed into a multi-professional care plan. The care team delivers care according to the plan, which is continuously adjusted according to a patient and professional joint evaluation of goal attainment. See text for further explanation. (illustration inspired by House of care by Angela Coulter).
In summary, Digi-PIP care seeks to reduce the risk of clinical crises that are costly in both human and economic terms, mainly through a digitally supported proactive dimension. However, it cannot be stressed enough that Proactive care will fail, if it is not well enough supported by the people who produce the patient pathway: the patient, his or her significant others, and the care professionals. Person-centered care is necessary to understand and integrate the personal agenda into the care plan. The Integrated care elements are necessary to create the conditions in which the relevant competencies are brought together to design the whole person care plan. Only when these 2 elements are in place, can proactive care across conditions and personal agenda be integrated successfully into the care plan activities. The digital support functions as a change agent and is essential to scalability. The triple aim is a product of the synergies of the 3 PIP elements and their digital support.
This paper employs a methodological combination of a scoping review and a systematic intervention review of health service and eHealth research literature. The research literature is the most formal arena for new ideas, discussions, and evaluations of current work toward improved care quality for persons with complex long-term needs.
A scoping study approach helps rapidly identify gaps in existing literature relative to a predefined set of expectations, such as a model or acknowledged challenges in the field and points out areas worthy of further attention [
The search strategy was set out in a protocol document, outlining the intention of the review, the inclusion and exclusion criteria, and the selection and data extraction methods. We give a more detailed outline of our methods, including exact search terms in
Target population: must include at least a subset of elderly over 60 years with complex care needs. An author’s description of the study population as being frail, multimorbid, or having complex needs was considered sufficient for fulfilling this criterion. We understand frailty in this context as a state of increased biological vulnerability resulting in a reduced ability to cope with stressors [
Intervention: includes all 3 elements defined as:
Person-centered care: Paper describes person-centered care as an ideal for care, no definition required.
Integrated care: Either a shared care plan or a multidisciplinary team responsible for the cohesive planning and delivery of care or both.
Proactive care: early identification of risk, or prevention of risk development at the population or individual level, including self-management support.
Digital support: any digital technology, supportive of the above intervention components beyond basic electronic health record (EHR) functionality, published after 2000 (to avoid outdated technology).
Outcome: patient or professional qualitative experience or quantitative measures of the triple aim: (1) patient experience, (2) health outcomes, or (3) cost or benefit ratio.
Study design: any qualitative or quantitative design, which includes comparisons between situations with or without access to the intervention in either a before-after design or a group comparison.
Exclusion criteria were as follows:
Papers that were not original research or had no comparative elements.
The study population did not include patients with complex long-term needs.
The technology support did not target the Person-centered, Integrated, or Proactive elements. We excluded interventions using a phone, documentation in a general EHR, or shared paper records.
We finalized the search in November 2017. After removal of duplicates, the search identified 927 potentially relevant publications. Moreover, 2 reviewers (KN and KL) independently identified papers that matched the inclusion criteria based on title and abstract. We included 65 publications in the full-text review conducted by FS and GB, and we resolved conflicts between reviewers through discussion until consensus. The study flowchart in
The studies included consist of a sociotechnical intervention expected to improve the individual patient pathway and their corresponding evaluation. Each study, which is our unit of observation, may be described by several papers as shown in
In a secondary, deductive-inductive analysis ad modum Tjora [
Flowchart of a systematic search and inclusions and exclusions of studies of digitally supported person-centered, integrated, and proactive care (Digi-PIP care) for frail multimorbid elderly. Search finalized in November 2017.
Included papers’ authors, publication year | Supporting papers | Acronym or short name | Study context | Study population characteristics |
Bleijenberg 2016 [ |
Bleijenberg [ |
U-PROFIT (Utrecht PROactive Frailty Intervention Trial) | Netherland, primary care | Randomized controlled trial (RCT) of frail individuals aged >60 years, using screening tools |
Blom 2016 [ |
—a | ISCOPE (Integrated Systematic Care for Older People) | Netherland, primary care | RCT of Individuals aged >75 years reporting issues in at least 3 of 4 domains in a screening questionnaire |
Martin 2012 [ |
— | PaJR (Patient Journey Record system) | Ireland, primary care | At least 1 chronic condition and high health care use last year randomly allocated in intervention and control groups |
Boult 2013 [ |
Giddens [ |
Guided care | United States, primary care | Cluster randomized trial of persons > 65 years of age and identified as potential high resource users in screening |
Council 2012 [ |
— | PCCP (Person-Centered Care Plan) | United States, Dartmouth, clinic with both primary care and hospital services | 5 heuristically selected patients with complex care needs. Before after comparison. |
Nelson 2014 [ |
Rosland [ |
PCMH-VA (Person-centered Medical Home, Veterans Health Administration) | United States, Veterans Health Admin, both primary care and hospital services | Observational study of all patients in the Veterans Health Admin system, with subanalyses for persons with chronic conditions |
Liss 2011 [ |
Reid [ |
PCMH-GH (Person-centered Medical Home, Group Health) | United States, group health, primary care | Adults with diabetes, hypertension, and/or coronary heart disease at a Patient Centered Medical Home prototype site compared with other sites in Group Health |
aThe paper had no supporting references relevant to this study.
In an explorative exercise, we used the key components attributed to the 3 active ingredients and their digital support to score each included study regarding maturity. The process of scoring revealed that the interpretation of key components was challenging. The high-level terminology was interpreted differently across authors. Although we were able to arrive at a consensus through dialogue and reflection, it is clear that our scoring system will probably meet the same interpretational challenges in other contexts. Our scores are presented, somewhat tentatively, in
In the next stage, we created the matrix framework, where we translated key system components into system capabilities. A capability is a system’s ability to achieve a desired goal or result and does not specify how the task is solved. These capabilities were then mapped to patient-pathways stages so that it would become more evident what we expect from the system at each stage. We hope this matrix may prompt other researchers to engage in reflection and dialogue about the system capabilities necessary to support a PIP care system.
We used a best-evidence synthesis approach ad modum de Bruin [
Although most of the outcomes were negative, to be parsimonious in our presentation, we have chosen to present all primary outcomes and any secondary outcomes that show a significant effect of the intervention. We present only analyses that are adequately adjusted for baseline biases. We did not attempt to perform a meta-analysis due to the heterogeneity of the contexts, interventions, methodologies, and outcomes. If the intervention reported effects in more than 1 paper, we used data from the latest study.
A short version of the review is presented in
This paper is a literature review. It includes no original information on patients, and there are, therefore, no privacy or ethics concerns requiring board review.
Approximately 1.0% (10/927) of the papers identified in the digital search (10 papers representing 7 interventions of 927 potential papers) were included in the final review. We present the intervention context and study population in
All studies claimed to adhere to and acknowledge the long-term aspect of the care process. The degree to which they document that personal goals make an impact on care decisions varied from the mention of Person-centered care as an ideal for care at one end of the scale to a documented impact on care plans and evaluation at the other.
Only the Person-Centered Care Plan (PCCP) study reached the highest possible maturity score on Person-centered maturity. They document how they understand person-centeredness as a journey undertaken by the patient and the surrounding team. The goal in the PCCP is “...to create negotiated goals that incorporate the values of the patient and the healthcare team into a mutually agreed upon explicit action plan.” Council demonstrates how the PCCP identifies the care team and distributes responsibilities for goals and tasks, including the goals for which the patient is responsible. The care plan outlines what to do in an emergency. The PCCP is digital, which means it is interactive, updated, and shared among those providers that have access to the same EHR [
Integrated Systematic Care for Older People (ISCOPE) mentions goal setting together with the patient: “...care should be provided proactively to set and prioritize goals together with the patient and to empower the patient to reach these goals.” [
The other 5 studies described Person-centered care in terms of patient involvement and engagement. However, they did not document how
All the papers highlight fragmentation of care as a significant challenge and explain how their intervention supports seamless care. Evidence-based and shared care plans are core to the integrated care effort [
In addition to a care plan, 5 of the studies dedicated extra resources to the development, delivery, recruitment of external resources, and follow-up of the care plan [
Effect of a care plan is contingent on its actual delivery, but no system monitored care plan delivery systematically. Moreover, 2 studies observed patient progress "by monthly monitoring of symptoms and adherence [
There are 2 main approaches to proactive care: self-management support and risk and emergency management.
Self-management support improves the person’s capacity to maintain health and well-being, and it simultaneously strengthens self-agency. Both the PCCP and the Guided Care studies describe how the care plan includes those actions that the patient is responsible for [
A total of 3 papers did not report any activities on self-management support [
Risk assessment strategies identify and act upon early impactable risk, rather than wait for the clinical crisis. Risk assessment at the individual level was an element in 6 of the 7 studies. The Patient Journey Record system (PaJR) study used lay care guides to keep in touch with patients at high risk for hospitalization. After each conversation, a natural language analysis of the written synopsis would estimate the risk of hospitalization [
The risk identification was linked to a range of
Only the PCCP plan summarized the emergency measures agreed by both patient and providers [
We have conceptualized Digi-PIP care as a set of characteristics systematically included in the individual Patient Pathway in support of the 4 generic stages of a patient pathway [
Although all the included papers addressed the 3 active ingredients and some form of digital support, it was clear that none of the interventions succeeded in giving equal focus to all elements.
The digital support, particularly, was marginal and far less advanced than what is considered state of the art in research projects addressing only 1 of the PIP digital axis. PCMH-VA had the highest total digital score, with a patient portal, low-threshold e-visits, a shared care plan, and telemonitoring services. They were far from sporting a full suite of eHealth services that would both leverage and scale their PCMH approach. All other studies lacked digital support in at least 1 PIP elements. We believe that a genuinely sociotechnical design of PIP care, where technology supports and replaces analog services, is still somewhere in the future.
Even when the care packages are complex, the studies do not explicitly acknowledge the interwoven dependencies between the PIP components. There is some understanding that all Digi-PIP ingredients are essential. For instance, in U-PROFIT study, the main focus was on integrated care, whereas the person-centered and proactive components were considered integral parts of the overarching comprehensive care plan [
However, the focus on the unbroken chain of care events that lead from intervention to the desired outcome is not always present. For example, care outcomes require loyalty in care delivery to the care plan. Impressive as the service redesign efforts in the Guided Care study is, the Guided care nurse is
Key care and digital components, described in terms of the capabilities they offer in support of the person-centered, integrated, and proactive care. We have mapped each PIP-element to statements of care system capability for each of the 4 generic stages of an individualized patient pathway.
Care components | Goals | Plans | Delivery | Evaluation | |
Care | ...declares Person-centered care as an ideal and explores “what matters to me?” and “patient values, needs, and preferences.” | ...uses “What matters to me?” to negotiate realistic goals and create a care plan. | ...includes patient capabilities aligned with “What matters to me?” in care delivery. | ...asks for patient feedback/ PROMsa | |
Digital support | ...offers access to digital health information/ electronic health record and supports the formulation of “what matters to me?” | ...offers digital sharing of: What my carers should know about me. | ...includes the patient in virtual care delivery and team exchanges. | ...encourages digital feedback from patients, including PROMs. | |
Care delivery | ...identify condition- or function-specific goals that support “what matters.” | ...combines condition/function-specific pathways into a whole person care plan for all conditions. | ...allocates resources to care plan, to show who does what when. | ...follow up to identify needs for adjustment of care plans or delivery. | |
Digital support | ...digitally identifies potential professionals to contribute to care plan development aligned with “what matters to me?” | ...provide tools to build a personalized digital evidence-based care plan, with workflow optimization to show: who does what when. | ...shares the care plan digitally across providers and offers tools for virtual team communication (video, messages, and chat). | ...triggers an alarm in case of gaps in critical care delivery. | |
Care delivery | ...identifies high-risk subpopulations, their individual high-risk scenarios over time and aligns focus on risk with ”What matters to me?“ | ...supports risk monitoring, self-managed or professional follow-up, in alignment with ”What matters to me?“ | ...offers low threshold response (self-managed, office or home visits) to uncertainties, emergencies, and alarms. | ...learns and adjusts goals and plans in light of undesired events and ”What matters to me?“ | |
Digital support | ...offers an algorithm-based risk-stratification tool to identify high-risk populations and their individual risk scenarios over time. | ...offers personal digital health apps and sensors that monitor risk and provide digital contingency plans in case of uncertainty, emergencies, or alarms. | ...provides digital decision support and low-threshold e-visits in case of uncertainty, emergencies, or alarms. | ...is a learning health care system improves prediction and action plans in light of undesired events. |
aPROMs: patient-reported outcome measures.
In compliance with de Bruin’s methodology [
The U-PROFIT, ISCOPE, Guided Care, and PCMH-VA studies present 8, 3, 7, and 6 outcome analyses, respectively, 24 in all. Outcomes lie within areas of patient satisfaction, quality of life, function, disease process quality, health care utilization, mortality, and staff burnout. Of these, only the emergency department visits in Nelson’s study from the VA and the home-care visit frequency in Guided Care showed clear and clinically meaningful significant positive effects (see
This study cannot answer the question of whether intervention maturity matters, because the number of high-quality studies is not large enough to support a correlation analysis between maturity and outcomes.
Selected outcomes in 4 high-quality studies of
Paper | Outcome measure | Patient or clinics | N | Effect intervention | Effect control | Ratio Intervention/ Control | |
Utrecht PROactive Frailty Intervention Trial [ |
Katz 15 scores at 6 months. Range 0-15, lower score is better | Pa | 2754 | 1.7 | 1.7 | 0.97 | Not significant |
Katz 15 scores at 12 months. Range 0-15, lower score is better | P | 2489 | 1.9 | 2.0 | 0.92 | .03 | |
Integrated Systematic Care for Older People [ |
12 months follow-up, change in quality of life, Cantril’s ladder (range 0-10, higher is better) | P | 842 | −0.2 | −0.2 | 1.00 | .82 |
12 months follow-up, Delta Groningen Activities Restriction Scale (range 18-72, lower score is better). | P | 842 | 2.9 | 3.5 | 0.83 | .30 | |
Guided Care [ |
Functional health Short Form 36, higher score is better | P | 477 | 36.1 | 37.5 | 0.96 | Not significant |
Home health care episodes | P | 477 | 0.9 | 1.3 | 0.71 | <.05 | |
Person-centered Medical Home-Veterans Health Administration [ |
Emergency Department visits per 1000 patients per year (secondary outcome) | Clinic | 913 | 188 | 245 | 0.77 | <.001 |
aP: Patient.
We identified 927 potentially eligible papers, but after full-text review, we included only 10 papers.
The PIP elements were supported to a varying degree. Person-centeredness was an ideal for care, but only one intervention made
The most prominent digital support of the PIP elements was risk stratification tools. Second, 3 providers supported the sharing of care plans in the EHR.
The chain of care is only as strong as its weakest link. The maturity matrix made it possible to identify several potential breaks or weaknesses in the chain of success. The most common weaknesses were:
A lack of documentation that
That planned care plan was actually delivered.
That risk identification schemes were coupled with adequate responses from the care system.
Finally, there are feedback loops that support learning and adjustment of the PIP dimensions of care.
We included 4 studies in our summary of effects after methodological quality assessment. Moreover, 2 of the 24 analyses in 4 studies reported modest positive outcomes with reductions in emergency care utilization and home health visits.
The specialization of health care remains both its biggest asset and weakness. Systems theory has shown long ago that when systems grow, they tend to specialize. If specialization is not coupled with centralized coordination, the organization’s ability to deliver its end product will be crippled [
We structured this review according to the Digi-PIP care framework and a maturity evaluation. As noted, there is an abundance of literature and reviews supporting each of the PIP elements alone [
Scientists are taught to narrow down and examine
Complexity theory [
Instead of placing all faith in method X, the intervention will need to include a way to detect and manage challenges as professionals become aware of them. In complex adaptive systems, one moves toward a goal first and foremost by creating a good enough vision of what the goal is. In Digi-PIP care, we believe that vision is a negotiated and realistic set of personalized goals aligned with
The researchers who work in improving outcomes in complex adaptive systems will be working with questions such as How can we help formulate good enough goals and subgoals, that are also observable, and provide process guidance?; What are areas of knowledge, skills, and tools essential to make available to enable professionals to invent proper processes?; Are there standardized components that can be plugged into and tailored to the individual pathway?; and How can we help frontline professionals hypothesize and modulate the chain of events that will lead to success in each case?
We have done systematic searches in the 3 largest literature databases covering the health and eHealth field. A librarian trained in building complex searches conducted the search. Pairs or triplets of authors performed all the steps in the screening of papers, data extraction, and grading of papers. Author-pairs discussed disagreements until they reached consensus, or, if it concerned a matter of principle, the entire author group addressed the issue. We defined rather broad inclusion criteria so that we should not inadvertently exclude studies that might bring forth new knowledge. The authors were, with one exception, senior researchers at the professor or associate professor level with long track records in the health and eHealth fields. These are the strengths of the study, which ensure that we have indeed identified the current published knowledge base regarding Digi-PIP care.
Given the extensive activity in this field and the many large enterprises underway in this area, we were surprised by the meager catch and by how old many of the papers were. We believe that the small number of articles indicates that this is an area that many researchers find too complicated to bring into a viable research model. Those organizations that are making progress in this field do so without publishing their interventions, the digital solutions, and their results. There may well be significant experiential knowledge in the field that we do not capture in this review.
The research literature is permeated with
It is not enough to provide a care plan that seems sensible to the providers. It must also be
We believe that a reductionist scientific methodology is blocking the way forward. We need to embrace the problem-solving methods suited for the improvement of outcomes in complex adaptive systems. Researchers need to embrace questions such as How do we formulate good enough process guiding goals?; How does a professional formulate a rationale for adjustment of a process?; and What are the generalizable components of
We predict that research will not show consistent results from care transformation for persons with complex long-term needs until all 3 PIP care elements are successfully and flexibly implemented with digital support. We need a chain of success thinking in the work of creating patient pathways. The art of high-quality care is to invent a road as it is being walked, toward
More detail on methods, and literature review search strings.
The short version: The evidence base for an ideal care pathway for frail multi-morbid elderly: A combined scoping and systematic intervention review.
Digitally supported Person-centered, Integrated, and Proactive care
electronic health
electronic health record
Integrated Systematic Care for Older People
Patient Journey Record system
Person-Centered Care Plan
Person-Centered Medical Home, Veterans Health Administration
Person-centered, Integrated, and Proactive
patient-reported outcome measures
randomized controlled trial
Utrecht PROactive Frailty Intervention Trial
Veterans Health Administration
The authors are grateful for the competent and timely help of the research librarian Karianne Lind (KL) in performing database searches and sifting through titles and abstracts in search for eligible papers.
The study was funded by 2 sources: (1) The Norwegian Research Council through the m&e Health project (contract 269884/H10) and (2) The Norwegian Centre for E-health Research. This work has been undertaken in complete independence from funders. The publication charges for this study have been funded by a grant from the publication fund of UiT The Arctic University of Norway.
The authors represented a set of multiprofessional competencies: technological/eHealth (FS, MR, RF, and BS), medical (GB), and nursing (BS, KN) disciplines, and they are all experienced health service researchers. Some of the authors also had patient experience. GB and FS were responsible for writing the protocol. KN and our research librarian (KL—see Acknowledgments) performed searches and initial screening. The first author was responsible for drafting the main text. All authors took part in the reading of original literature, extraction of data and data analysis, critical review of drafts, and approved the last version of the manuscript.
None declared.