The Krijtmolen Alliantie and their stakeholders: performance intelligence needs for the integrated care network in the Amsterdam Noord district
At the time of this study the Krijtmolen Alliantie had 13 member organizations representing health and social care provision organizations active in the Amsterdam Noord district. Selected stakeholders for interviews (n = 12) include the health insurer Zilveren Kruis, and the municipality of Amsterdam (both funders in health and social care in the Amsterdam Noord district), patient and client representation, the Dutch Healthcare Authority (NZa), general practitioners, and the ambulant pharmacy of the local hospital (see appendix 1 for details on interviewees and catchment areas). KMA members reported working with a variety of catchment areas ranging from provision at neighbourhood level within the district, to province level. The KMA members and stakeholders serve different proportions of the population of Amsterdam Noord either due to the nature of their provision (for example: welfare aid to those in financial problems) and/or due to sharing the market in the district (for example with competing elderly care providers Amstelring, Cordaan and Evean). Multiple KMA members report overlapping service provision, stating potential for competition, as well as beneficial alignment, between providers. KMA members and stakeholders showed aligned needs for performance intelligence. The most commonly expressed needs for performance intelligence among interviewees were related to: 1) population data and outcomes adjusted to their catchment area, 2) information on the alignment between providers, 3) outcomes of (multiple-provider) interventions, and 4) an overview of health and social care information of and for the citizen/patient.
Population data and outcomes adjusted to catchment areas
Most health and social care providers report a need for information about the population they serve, either to be able to better define and support their target group, or as an outcome indicator for their work. Only few referred to the availability of population data in the Amsterdam Noord region, and the ones that did, found it hard to align the information with their practice as it is presented on an aggregated level and not tailored to their catchment areas.
Information on the alignment between providers
Most interviewees reported a lack of information on the alignment between providers. KMA members report that they are not always aware if other providers are involved with their patient/client. Most interviewees mention there is a barrier between social care information and health care information either on micro (individual patient), meso (between organizations) or macro (alignment social and health budgets) level. Some interviewees emphasized the importance of the correlation between the two, stating that coordinating health and social care in an integrated way can change decisions for individual treatment and adjust service provision on a meso and macro level.
Information on outcomes of (multiple-provider) interventions
Interviewees report a need for insights into the efficacy and efficiency of implemented multi-provider interventions. This performance intelligence can generate a learning curve on these interventions (mostly stated by providers of care) or can be used for accountability purposes (mostly stated by financers and patient representatives).
Overview of health and social care information for the citizen/patient
The interviewees representing patients in the region reported a need for an overview of health and social care data. They describe having limited information on possible health and social care providers in their region as well as lacking an overview of their own health and social care usage. Available information is compartmentalized per provider in, among other things, bilateral digital patient portals and limited in content (e.g. summarized letters or appointment dates). The interviewees representing patients in the region report that their main source for health and social care information comes from direct contact with their doctor, general practitioner (GP) or specialist, and experiences of friends and family.
Available data (infrastructure) on health and social care in the Amsterdam Noord district and existing exchanges of data
Data collected in health and social care organizations
Interviewees describe data collection within their organizations that can be categorized as being used for five different purposes: 1) Identification and contact details of clients (name, address, personal identifier); 2) Treatment support (what are the needs of the client, what treatment is given). Every provider makes their own comprehensive analysis of the client, even after referral from other health and social care providers. Reasons mentioned to do this were incomplete information from referral in order to start health or social care provision, trust in referral information, and to encompass changes over time that might have occurred; 3) General management information for the organization, e.g. wellbeing of employees (e.g. absence due to illness of employees), quality (e.g. near misses, complaints), service (e.g. time until phone is being answered), production (e.g. amount of treatments done or clients seen), and finances (e.g. costs and benefits); 4) Financial accountability (different per stakeholder varying from minute writing of professionals to complex coded accounting); and 5) Quality of care outcome measurements (e.g. routine outcome measurements, patient reported outcome measures). When done, providers collect their own outcome measures, aimed at particular interventions within the institution and its target population. Many state to obtain most information, even in the case of a referral, from the patient themselves.
Data exchange between health and social care organizations
All health and social care providers exchange information mostly on an aggregated (organizational) meso-level with their funders and their relevant overseeing authorities (e.g. the Dutch Health Authority and/or Inspectorate of Health and Youth (IGJ)) in line with Dutch legislation. Structural exchange of information between health and social care providers (e.g. including patient identifier) is limited. The general practitioners, functioning as gate keepers, exchange most information with other providers via referrals (e.g. the referral from general practitioners to medical specialists through a referral data exchange programme (Zorgdomein)). However, general practitioners state an incomplete exchange of information with elderly care providers and social care providers, noting this is not standardized and only available circumstantially or on the initiative of certain care professionals. Health insurance companies have the possibility to integrate data on patient-level from different providers, however strictly monitored by privacy laws, and only for the percentage of patients that they finance (their market share). A couple of providers of those interviewed have agreements to exchange anonymized aggregated data with academic networks to contribute to national-level data for performance measurement of the overall health system, for example exchanges with the Academic Collaborative Centre for Elderly Care, the Academic Collaborative Centre for General Practice and the Netherlands Institute for Health Services Research (NIVEL). There is potential to exchange health and social care data on citizen level as most providers use personal identifiers in their registration.
The patient is the only point where all data for an individual comes together, however, this information is fragmented in bilateral communication between patient and providers and often summarized in referral letters or discharge letters formats. Patients can choose between one of the, at the time of this study, 29 accredited providers which all offer a ‘personal health environment’ (persoonlijke gezondheidsomgeving; PGO), to gather all of their individual healthcare data. Furthermore, there is a national healthcare data exchange platform (Landelijk Schakelpunt; LSP). This platform gives an overview to the patient what data (for example medication list) has been requested by providers. This platform, however, does not contain content data on care provided. Most health and social care providers report the existence of a patient portal or information exchange platform to communicate and/or share data bilaterally. All health and social care providers report internal data collection supporting their care provision. As providers do their own purchasing for digital support software, there is a great variety in digital support software and suppliers used by health and social care providers. An example being that there are eight different systems used by general practitioners in the district.
Available performance intelligence to support Triple Aim goals
On population health
The descriptive data in the ‘Gemeentezorgspiegel’ and public data from the municipality of Amsterdam and the Public Health Service of Amsterdam (GGD) include more than 100 indicators on population health and welfare available about the Amsterdam Noord district. Only few of the interviewed KMA members mentioned this information and the ones that did, expressed to experience difficulties to use this information. Only one social care provider mentioned the use of neighbourhood-level data on income from the publicly available data combined with their own data gives them actionable insights. Multiple interviewees refer to the reports made by independent research institutes or the municipality. These reports give insights, but there remains a gap towards actionability, not aligning to the catchment areas or information needs of providers.
On costs per capita
Financial and usage data on health and social care are siloed and costs per patient or inhabitant over the whole spectrum of care are not integrated on patient identifier level due to privacy concerns, reducing the value for use. Costs for long-term care (Wet Langdurige Zorg; WLZ), medical care (Zorgverzekeringswet; Zvw) and social support (Wet Maatschappelijke ondersteuning; WMO) can be shown separately with breakdowns (e.g. age groups) and benchmarks (e.g. municipality vs national). The absence of integration of costs on personal identifier level makes it difficult to construct one indicator on per capita costs for governing integrated care delivery purposes.
Using either cost per inhabitant or cost per patient as an indicator for costs in the district, measure different things and show different outcomes. In the feasibility analysis of the Health Insurance Act claims, which includes the reimbursement of hospital care, GP’s, pharmaceuticals and mental healthcare the benchmark between Amsterdam Noord and Amsterdam excluding the Noord district, showed relative less costs per patient, but more cost per inhabitant. At the time of the study there is still a lag of about one year in reporting based on this database. To support governance of integrated care networks, information will have to be timelier.
On care experiences
Experience of care is documented in bilateral relations between KMA members and patients through patient satisfaction surveys, requests to (anonymously) review care givers or organizations and compliments and complaints reporting systems. There is a national system for reviewing healthcare providers called ‘Zorgkaart Nederland’(25), however this system also measures patient experience per siloed provider (organization or specific professional). There is no data collection that gives insight in integrated patient experience in the Amsterdam Noord district.
Governing an integrated health and social care district – priority domains for improvement of performance intelligence
In the reflection meeting with the board of the KMA, 7 members were represented. Reflecting on the findings, the fragmented data collection mirrors the fragmented accountability and financing siloes currently applied in the Netherlands. As one member put it: “die fragmentatie, ook in verantwoording, nekt ons”, loosely translated: “this fragmentation, also in accountability, is our downfall”. Governing to incentivize the Triple Aim in an integrated health and social care provision network requires an integrated data infrastructure that aligns with the governing structure. Board members emphasise that there is a willingness to integrate care provision, however this must be aligned with external expectations in accountability and financing. Board members reflect that to counteract the individual accountability and financing siloes, boldness is needed to step away from the norms and set up a performance intelligence dashboard that reflects the governing ambition. Multi-year contracts of involved parties are needed to create such movement space to innovate. Making the best use of available data is the first step towards building performance intelligence. The areas that were highlighted in the reflection meeting as potential domains for improving performance intelligence to govern the integrated care network were: costs, effectiveness & safety, patient experiences & outcomes, usage and process of care provision, cooperation & capacity in the network, evaluation of interventions, and population outcomes. There were some important reflections on these domains. The analytical capacity needed to process data into pieces of information and performance intelligence for a person centred approach, outreaches the capacity of one organization. The fragmentation in bilateral patient portals and patient experience measures per organization does not reflect a person centred approach. The current Dutch funding system is making use of budgets per organization, which does not allow for changing capacity needs within an integrated care network to carry over funding from one organization to the other. The timeliness of Dutch administrative reporting systems for care, needs to improve in order to create indicators that are actionable for governance. The reflection meeting defined costs (combined accountability through bundling financial incentives) and patient experiences & outcomes (measured in an integrated approach) as priority domains to improve the performance intelligence of their integrated care network.