Measuring populations: In search of methods to evaluate health and experienced quality of care

(1)

Tilburg University

Measuring populations

Hendrikx, R.J.P.

Publication date:

2019

Document Version

Publisher's PDF, also known as Version of record

Link to publication in Tilburg University Research Portal

Citation for published version (APA):

Hendrikx, R. J. P. (2019). Measuring populations: In search of methods to evaluate health and experienced

quality of care. Ipskamp.

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.

(2)

Measuring

Populations

_{In search of methods to evaluate}

health and experienced

quality of care

Roy Hendrikx

Measur

ing

Populations

R

oy

Hendr

ikx

UITNODIGING

Voor het bijwonen van

de openbare verdediging

van het proefschrift

Measuring

Populations

In search of methods

to evaluate health

and experienced quality

of care

Op vrijdag 21 juni 2019

om 10 uur in de aula

van Tilburg University,

Cobbenhagen gebouw,

Warandelaan 2, Tilburg

(3)

(4)

531422-L-bw-Hendrikx 531422-L-bw-Hendrikx 531422-L-bw-Hendrikx 531422-L-bw-Hendrikx Processed on: 22-5-2019 Processed on: 22-5-2019 Processed on: 22-5-2019

Processed on: 22-5-2019 PDF page: 1PDF page: 1PDF page: 1PDF page: 1

Measuring

Populations

_{In search of methods to evaluate}

health and experienced

quality of care

(5)

Colofon

Measuring Populations. In search of methods to evaluate health and experienced quality of care

All rights reserved. No part of this thesis may be reproduced, stored or transmitted in any way or by any means without the permission of the author, or when applicable, of the publishers of the scientific papers.

The research reported in this dissertation was financially supported by the National Institute for Public Health and the Environment (RIVM, SPR program PreCare) and Tilburg University.

Layout and design by Legatron Electronic Publishing, Rotterdam Cover by Roy Hendrikx

(6)

Measuring populations

In search of methods to evaluate health and experienced quality of care

Proefschrift

ter verkrijging van de graad van doctor

aan Tilburg University op gezag van prof. dr. G.M. Duijsters, als tijdelijk waarnemer van de functie rector magnificus

en uit dien hoofde vervangend voorzitter van het college voor promoties, in het openbaar te verdedigen

ten overstaan van een door het college voor promoties aangewezen commissie in de Aula van de Universiteit op vrijdag

21 juni 2019 om 10.00 uur door

(7)

Promotores

Prof. dr. C. A. Baan Prof. dr. D. Ruwaard

Copromotores

Dr. H. W. Drewes Dr. M.D. Spreeuwenberg

Promotiecommissie

(8)

Chapter 1

General Introduction

1

Chapter 2

Which Triple Aim Related Measures are Being Used to Evaluate

11

Population Management Initiatives? An International

Comparative Analysis

Chapter 3

How to Measure Population Health: an Exploration Towards an

43

Integration of Valid and Reliable Instruments

Chapter 4

Comparing the Health of Populations: Methods to Evaluate

69

and Tailor Population Management Initiatives in the Netherlands

Chapter 5

Measuring Population Health from a Broader Perspective:

95

Assessing the My Quality of Life Questionnaire

Chapter 6

Harvesting the Wisdom of the Crowd: Using Online Ratings

117

to Explore Care Experiences in Regions

Chapter 7

Measuring Regional Quality of Care Using Unsolicited Online

135

Data: Creating More Detailed Insight Using Text Analyses

Chapter 8

General Discussion

159

Summary

173

Nederlandse samenvatting

177

Dankwoord

181

List of publications

185

(9)

(10)

Chapter 1

(11)

2

In healthcare, many utter phrases such as “care is too expensive”, “the demand for care has changed” and “the quality of care should be improved”. Each depicts a pressing issue. First, the phrase: “care is too expensive”. Over the past century, healthcare has become much more effective against communicable and infectious diseases. The invention of vaccines and antibiotics as well as improvements in nutrition and hygiene has drastically decreased the incidence or has even eradicated several infectious diseases [1]. These improvements led to an increase in average life expectancy of more than ten years in developed countries since 1970 [2]. However, this also led to a change in population characteristics, there are now more elderly that require care more often. Additionally, care is provided using new technologies, such as Magnetic Resonance Imaging and certain drugs, that are effective, but also expensive [3]. These developments have driven up the price of care of the average patient [2] and are strongly related to the second issue.

Second, “the demand for care has changed”. Together with the reduction of infectious diseases, the prevalence of chronic diseases has increased. The prevalence of chronic diseases has increased so much that they became the dominant health issue in Western society and led to a shift in care demand. Infectious diseases required acute short-term care, while chronic diseases call for long-term care as patients often remain patients for the rest of their life [1]. The change from predominantly acute to long-term care also asks caregivers to change their approach. Caregivers are challenged to cooperate more closely with each other and align their services. They also gained a supportive role to help patients manage their own care and well-being from home [4, 5].

Third, “the quality of care should be improved”. Variation in services between care providers and the underuse of effective services indicate that healthcare has not kept up with science to ensure the use of the latest evidence-based practices [6-8]. On the one hand, protocols and clinical guidelines have been introduced to overcome specific underuse issues, but more effort is needed to improve the uptake of effective interventions and make them affordable [9]. On the other hand, the overuse of services is still prevalent and causes its own problems [10]. Doing ‘more’ in healthcare does not always lead to better results and is associated with patient harm and excess costs [11]. From the patient’s perspective, differences in experienced quality can also be seen. How a patient experiences treatment does not necessarily reflect the clinical effectiveness of a treatment, but is considered an integral part of healthcare and varies greatly between care providers [12, 13].

Shifting focus towards regional care in order to achieve the Triple Aim

(12)

3

1

quality and efficiency, as well as prevention. As a result, reforms designated as population (health) management (PM) are becoming more and more widespread in health policy. Even though different definitions exist [16], PM initiatives generally focus on the health needs of a specified population across the continuum of health and well-being by introducing multiple interventions that organize services related to health and social care, as well as prevention and welfare [17, 18]. Well known international examples include the American accountable health communities, the English NHS Vanguards and Gesundes Kinzigtal in Germany. American accountable health communities aim to reduce care use and costs by identifying and addressing insurance beneficiaries’ health-related social needs through screening, referral, and community navigation services [19]. The NHS Vanguards use several models that aim to reduce hospital visits by delivering care as close to patients’ homes as possible [20]. Gesundes Kinzigtal uses a single health management company that is responsible for the healthcare in the region of Kinzigtal with the intention to promote the integration of services [21].

(13)

4

Figure 1 | Conceptual framework for population management with components influencing health system improvements

[23].

An emphasis on evaluation

(14)

5

1 The conceptual struggles of population health and quality of care

Despite a generally accepted definition of population health by Kindig and Stoddart [29], the discussion around the concept has been going on for years [30, 31]. Kindig and Stoddart consider population health as the health of an entire population in a geographical area. The part of this definition that causes discussions is not ‘population’, which is generally accepted as “a body of persons or individuals having a quality or characteristic in common’’ or ‘‘the organisms inhabiting a particular locality” [32]. However, health has demonstrated to be not as easy to define [30]. The still dominant, most well-known definition was created by the World Health Organization back in 1948. It states: “health is a state of complete physical, mental and social well-being and not merely the absence of disease or infirmity [33].” However, this definition is highly debated and many suggestions for change/expansion exist [34-36]. Most of these suggestions include more social and meaningful living aspects of health and focus on the capabilities of the individual instead of the limitations. This new way of thinking can also be seen in the methods used to measure health. Objective instruments used to be the focus, e.g. life expectancy, but subjective measures have gotten more attention. Examples include national health monitors that assess the individual health status as experienced by that individual.

Quality of care, just like health, has changed over the years in both perspective and methodology. In the past, it was mainly approached with an objective clinical perspective. Does a treatment adequately control a patient’s blood pressure? Do the HbA1c values of a diabetic patient stay within the acceptable range? Even though these questions are vital, the patient’s subjective experience of care is now thought to be an integral part of the quality of care. For many, aspects such as the communication with the care provider and integrating the individual preferences of patients became essential to delivering high quality care [37, 38]. The measures used to evaluate quality of care also show this change in attitude, as Patient Reported Outcome Measures (PROMs) and Patient Reported Experience Measures (PREMs) have gained a lot of traction. For example, patients are asked about the responsiveness of staff and quietness in the hospital. From a policy standpoint, the focus on experience requires tailoring of care and personal assessments at the individual level. However, when changing from individual to population-focused policies, this asks for an alternate approach. Existing instruments need to be processed differently or new instruments need to be explored to create insight in care experiences at the population level. It might be possible to use individual-based data for this, but it might also require the use of alternate data sources.

Aim

(15)

6

be assessed. To fill this gap, this dissertation studies the potential of existing instruments and explores new datasets that can be used by PM initiatives in their quest to achieve the Triple Aim. The aim is to find out how population health and experienced quality of care can be evaluated by initiatives that focus on the general (regional) population. The following questions will therefore be answered:

1. Which instruments are used by PM initiatives to evaluate the Triple Aim? What do they measure and why have they been selected? [Chapter 2]

2. How do existing health instruments perform when used to evaluate and compare health at a population level? [Chapter 3, 4 & 5]

3. Which insights can be gained from unsolicited online data into quality of care for PM initiatives? [Chapter 6 & 7]

The Dutch setting

The Dutch PM setting was used to answer the research questions, with the exception of the first one as this question was answered with an international review. PM initiatives have been introduced in the Netherlands. Nine of these initiatives have been selected by the Dutch Ministry of Health, Welfare and Sport to be closely monitored by the National Institute for Public Health and the Environment (RIVM). Since 2013, the RIVM has been following these so-called pioneer sites using a combination of qualitative and quantitative techniques in the National Monitor Population management (NMP). The qualitative focus was on the implementation process these pioneer sites go through. Interviews were conducted on a regular basis to keep up with their experiences. The goal was to uncover the mechanisms that explain how a certain strategy leads to a specific outcome. The quantitative analyses used a survey that was sent out by the RIVM as well as existing national health datasets, such as claims and health data. This dissertation utilizes the quantitative data from the NMP survey and builds on this with additional in-depth studies.

Outline

(16)

7

1

(17)

8

References

1. van der Lucht, F. and Polder, J.J., Van Gezond naar Beter, in Volksgezondheid Toekomst Verkenning.

2010, RIVM: Bilthoven.

2. OECD, Health at a Glance 2017: OECD Indicators. 2017, OECD: Paris.

3. OECD, Fiscal Sustainability of Health Systems: Bridging Health and Finance Perspectives. 2015: Paris.

4. Shrivastava, S.R., Shrivastava, P.S., and Ramasamy, J., Role of self-care in management of diabetes

mellitus. Journal of Diabetes and Metabolic Disorders, 2013. 12: p. 14-14.

5. McCorkle, R., Ercolano, E., Lazenby, M., Schulman-Green, D., Schilling, L.S., Lorig, K., and Wagner,

E.H., Self-Management: Enabling and empowering patients living with cancer as a chronic illness. CA: a cancer journal for clinicians, 2011. 61(1): p. 50-62.

6. Birkmeyer, J.D., Reames, B.N., McCulloch, P., Carr, A.J., Campbell, W.B., and Wennberg, J.E.,

Understanding regional variation in the use of surgery. Lancet, 2013. 382(9898): p. 1121-1129.

7. Casati, A., Sedefov, R., and Pfeiffer-Gerschel, T., Misuse of Medicines in the European Union: A

Systematic Review of the Literature. European Addiction Research, 2012. 18(5): p. 228-245.

8. AHRQ, Improving Health Care Quality, in Fact Sheet. 2002, Agency for Healthcare Research and

Quality: Rockville.

9. Glasziou, P., Straus, S., Brownlee, S., Trevena, L., Dans, L., Guyatt, G., Elshaug, A.G., Janett, R., and

Saini, V., Evidence for underuse of effective medical services around the world. The Lancet, 2017. 390(10090): p. 169-177.

10. Berwick, D.M., Avoiding overuse—the next quality frontier. The Lancet, 2017. 390(10090): p. 102-104. 11. Lipitz-Snyderman, A. and Bach, P.B., Overuse: when less is more… more or less. JAMA internal

medicine, 2013. 173(14): p. 1277-1278.

12. Tsai, T.C., Orav, E.J., and Jha, A.K., Patient Satisfaction and Quality of Surgical Care in US Hospitals. Annals of surgery, 2015. 261(1): p. 2-8.

13. Solomon, L.S., Zaslavsky, A.M., Landon, B.E., and Cleary, P.D., Variation in Patient-Reported Quality

Among Health Care Organizations. Health Care Financing Review, 2002. 23(4): p. 85-100.

14. Stiefel, M. and Nolan, K., Measuring the Triple Aim: A Call for Action. Population Health Management, 2013. 16(4): p. 219-220.

15. Kindig, D.A., Isham, G.J., and Siemering, K.Q., The Business Role in Improving Health: Beyond Social

Responsibility, in NAM Perspectives. 2013, National Academy of Medicine: Washington, DC.

16. Steenkamer, B.M., Drewes, H.W., Heijink, R., Baan, C.A., and Struijs, J.N., Defining Population Health

Management: A Scoping Review of the literature. Population Health Management, 2017. 20(1).

17. Struijs, J.N., Drewes, H.W., Heijink, R., and Baan, C.A., How to evaluate population management?

Transforming the Care Continuum Alliance Population Health Guide into a broadly applicable analytical framework. Health Policy, 2015. 119: p. 522-529.

18. Steenkamer, B.M., Drewes, H.W., Heijink, R., Baan, C.A., and Struijs, J.N., Defining Population Health

Management: A Scoping Review of the Literature. Population Health Management, 2017. 20(1): p.

74-85.

19. Corrigan, J.M. and Fisher, E.S., Accountable Health Communities: Insights from State Health Reform

Initiatives. 2014, The Dartmouth Institute for Health Policy & Clinical Practice: Lebanon.

20. Lewis, G., Why population health analytics will be vital for the vanguards, in Blog. 2016, NHS England. 21. Hildebrandt, H., Schulte, T., and Stunder, B., Triple Aim in Kinzigtal, Germancy: improving population

health, integrating health care and reducing costs of care- lessons for the UK? Journal of Integrated

Care, 2013. 20(4): p. 205-222.

22. Berwick, D.M., Nolan, T.W., and Whittington, J., The triple aim: care, health, and cost. Health Affairs, 2008. 27(3): p. 759-69.

23. Steenkamer, B.M., Baan, C.A., Putters, K., van Oers, H., and Drewes, H.W., Population health

management guiding principles to stimulate collaboration and improve pharmaceutical care.

(18)

9

1

24. Üstün, B. and Jakob, R. Re-defining ‘Health’. Bulletin of the World Health Organization 2005 [cited

2015 21-12-2015]; Available from: http://www.who.int/bulletin/bulletin_board/83/ustun11051/ en/#.

25. Campbell, S.M., Roland, M.O., and Buetow, S.A., Defining quality of care. Social Science & Medicine, 2000. 51(11): p. 1611-25.

26. Derose, S.F. and Petitti, D.B., Measuring Quality of Care and Performance from a Population Health

Care Perspective. Annual Review of Public Health, 2003. 24(1): p. 363-384.

27. Donabedian, A., Evaluating the Quality of Medical Care. The Milbank Quarterly, 2005. 83(4): p. 691-729.

28. Jacobson, D.M. and Teutsch, S. An Environmental Scan of Integrated Approaches for Defining

and Measuring Total Population Health by the Clinical Care System, the Government Public Health System, and Stakeholder Organization. 2012 21-12-2015]; Available from: http://www.

improvingpopulationhealth.org/PopHealthPhaseIICommissionedPaper.pdf.

29. Kindig, D. and Stoddart, G., What is population health? American Journal of Public Health, 2003. 93(3): p. 380-3.

30. Etches, V., Frank, J., Di Ruggiero, E., and Manuel, D., Measuring Population Health: A Review of

Indicators. Annual Review of Public Health, 2006. 27: p. 29-55.

31. Vuik, S., Siegel, S., and Darzi, A. How Should We Measure The Distribution Of Health In A Population? Health Affairs Blog 2017 22-03-2017]; Available from: http://healthaffairs.org/blog/2017/03/17/ how-should-we-measure-the-distribution-of-health-in-a-population/.

32. Harris, D., Puskarz, K., and Golab, C., Population Health: Curriculum Framework for an Emerging

Discipline. Population Health Management, 2016. 19(1): p. 39-45.

33. World Health Organization. The Constitution. in International Health Conference. 1948. New York. 34. Huber, M., Knottnerus, J.A., Green, L., et al., How should we define health? BMJ, 2011. 343: p. d4163. 35. Walburg, J.A., Positieve Gezondheid: Naar een bloeiende samenleving. 1 ed. 2015, Houten: Bohn

Stafleu van Loghum.

36. The Lancet, What is health? The ability to adapt. The Lancet, 2009. 373(9666): p. 781.

37. Institute of Medicine. Crossing the Quality Chasm: A New Health System for the 21st Century 2001 02-03-2016]; Available from: http://www.nap.edu/html/quality_chasm/reportbrief.pdf.

38. Lavallee, D.C., Chenok, K.E., Love, R.M., Petersen, C., Holve, E., Segal, C., and Franklin, P.D.,

Incorporating Patient-Reported Outcomes Into Health Care To Engage Patients And Enhance Care.

Health Affairs, 2016. 35(4): p. 575-582.

39. Stiefel, M. and Nolan, K., A guide to measuring the Triple Aim, in Innovation Series. 2012, Institute for Healthcare Improvements: Cambridge.

40. Care Continuum Alliance, Implementation and Evaluation: A Population Health Guide for Primary

Care Models. 2012, Care Continuum Alliance: Washington DC.

41. Blumenthal, D., Malphrus, E., and McGinnis, J.M. Vital Signs: Core Metrics for Health and Health Care

Progress. 2015 21-12-2015]; Available from: https://iom.nationalacademies.org/~/media/Files/

(19)

(20)

Chapter 2 Which Triple Aim Related Measures are Being

Used to Evaluate Population Management

Initiatives? An International Comparative

Analysis

(21)

12

Abstract

Introduction

Population Management (PM) initiatives are introduced in order to create sustainable health care systems. These initiatives should focus on the continuum of health and well-being of a population by introducing interventions that integrate various services. To be successful they should pursue the Triple Aim, i.e. simultaneously improve population health and quality of care while reducing costs per capita. This study explores how PM initiatives measure the Triple Aim in practice.

Methods

An exploratory search was combined with expert consultations to identify relevant PM initiatives. These were analyzed based on general characteristics, utilized measures and related selection criteria.

Results

In total 865 measures were used by 20 PM initiatives. All quality of care domains were included by at least 11 PM initiatives, while most domains of population health and costs were included by less than 7 PM initiatives. Although their goals showed substantial overlap, the measures applied showed few similarities between PM initiatives and were predominantly selected based on local priority areas and data availability.

Conclusions

(22)

2

13

Introduction

Health care systems around the world are being challenged to reform by rising costs and disparities in the provided quality of care [1]. In order to realize sustainable and higher quality health care systems, so-called population (health) management (PM) initiatives are being introduced. These initiatives aim to achieve this goal by focusing on the health needs of a specified population across the continuum of health and well-being by introducing multiple interventions that integrate services related to health and social care, as well as prevention and welfare [2]. This approach addresses the current need for preventing or postponing chronic diseases as well as the push away from fee-for-service towards accountable care [3]. In order to realize sustainable and higher quality health care systems, PM initiatives should pursue the Triple Aim, i.e. simultaneously strive to improve population health and quality of care while reducing cost growth [4]. Hence, evaluations of the Triple Aim dimensions (population health, quality of care and cost) are needed to adapt and improve PM initiatives. Evaluating the three dimensions of the Triple Aim appears to be difficult in practice since the concepts of (population) health, quality of care and costs are not unanimously defined and measures for these concepts are under construction [5-10]. For example, Kindig and Stoddart (2003) define population health as “the health outcomes of a group of individuals, including the distribution of such outcomes within the group [11]”, while Young (2005) regards population health as “a conceptual framework for thinking about why some people are healthier than others and the policy development, research agenda and resource allocation that flow from this [12]”. Further adding to this complexity is the introduction of new concepts regarding health and quality of care [13, 14] as well as the rise of new types of measures, such as patient reported outcome measures (PROMs) and patient reported experience measures (PREMs) [15]. Several papers provide guidance on how to measure population health, quality of care and costs [2, 16-18]. Frameworks suggested by these papers provide many possible measures, potentially implying a large measurement burden and lack of comparability between PM initiatives. To explore how to best deal with the many possibilities, it is of interest to have insight into the currently applied measures for evaluating PM initiatives. Recently, an overview of applied health and health care performance measures was given by the Institute of Medicine (IOM) [18]. The IOM studied current measures used in the United States and found that a large number of various measures are utilized to evaluate health and health care. However, it is unclear whether these results are in line with the applied measures used in PM initiatives in and outside the United States. This is due to the IOM’s focus on general health care rather than PM and the differences between the United States and other OECD countries in health care performance and organization [19-21].

(23)

14

and well-being [22, 23]. Hence, this study explored which measures are used in practice to evaluate PM outcomes within the general population reflecting population health, quality of care and costs, and looked for emerging patterns and outliers.

Material and Methods

Search Strategy

Initially, literature searches were performed in order to explore the value of a systematic review (Appendix 1). This showed that current PM initiatives’ evaluations were not (yet) published in Medline. Therefore, PM initiatives were identified using a two-step exploratory search strategy that was performed during the period March to August 2015. The first step was to consult websites of research institutions involved in PM research (such as King’s Fund, Commonwealth Fund, Nuffield Trust, World Health Organization and the Institute of Health care Improvement (IHI)) for publications related to PM and the Triple Aim. Next, a manual search on the Internet was performed using the search terms ‘population health’, ‘population health management’ , ‘population management’ and ‘integrated care’. A list of relevant PM initiatives was compiled, which was subsequently evaluated by all authors in order to add missing known PM initiatives.

In the second step, the list of PM initiatives was sent to eight experts in the field of PM. These experts were asked to review the list to see if any relevant PM initiatives were missing. The suggestions provided by five experts (Appendix 2) were explored to create the final list of potential PM initiatives before scoring.

For analysis, information of included PM initiatives was collected by consulting websites of associated institutions and organizations. All available information related to the selected PM initiatives was screened, including documents, articles, webpages and presentations. If this did not provide the necessary information, published papers were searched using search terms related to the PM initiative (e.g. affiliated authors). Initiatives that did not publicly provide all information needed for scoring were asked to provide additional information by email. Finally, the quality of the initiatives’ (public) reporting was assessed based on the standards created by Nothacker et al. (2016) [24]. The found sources of each initiative were searched for the presence of the following seven standards: description of the measures development process, measures appraisal, measures specification, description of the intended use of the measures, measures testing/validation, measures review/re-evaluation, and composition of the measures developmental team.

Inclusion criteria

(24)

2

15

First, PM initiatives had to focus on a general, non-disease-specific population. Second, the initiatives had to either a) use interventions that covered at least two areas of care (prevention, health care, social care and/or welfare) and/or b) try to achieve Triple Aim. Criterion 2b was added due to the studies’ focus on the Triple Aim, hereby labeling initiatives improving the Triple Aim within at least one area of health care as PM for the purpose of this study. The third criteria stated that the country in which the initiative took place must have been classified as an OECD high-income country by the World Bank list of economies [25]. By focusing on high-income countries, the comparability between initiatives improved. Fourth, the initiative had to be evaluated and a description of this evaluation, including the measures used, population and the goal of the initiative, had to be available. Fifth, needed information had to be available in English, Dutch and/or German, and sixth, the evaluation had to be conducted at least once in the last five years.

Review process

The review process consisted out of three steps and was conducted by one researcher (RH) and checked by a second (HD). Any disagreements were resolved by consensus.

First, general features from the included PM initiatives were subtracted, including year of implementation, country, involved actors, target population (criteria and size), goals, included areas of care (health care, social care, prevention and/or welfare), organizational structure and implemented programs/interventions.

(25)

16

Framework Dimension Domain Subdomain Measure

Triple aim

Population health

Health outcomes e.g. Mortality e.g. Life expectancy at birth Disease burden e.g. Disease_prevalence e.g. Percentage of _{insured with}

cancer Behavioral/

physiological factors

e.g. Tobacco use e.g. Percentage of smokers Partcipation e.g. Informal care persons that provide e.g. Percentage of

informal care Functioning/

quality of life e.g. Quality of life e.g. Average quality of life (SF-12)

Quality of care

Patient safety e.g. Avoidable care avoidable hospital e.g. Number of admissions Effectivity e.g. Clinical_outcomes e.g. HbA1c control _{(<8 mmol/L)}

Responsiveness e.g. Communication between patient and caregiver

e.g. provision of safety instructions

for medication Timeliness needed in hospital e.g. (Waiting) time

proces

e.g. Average waiting time in ER Support e.g. Evaluation of

care giver

e.g. Percentage of carers getting an

assessment Accessibility e.g. Access to _{medical care} e.g. Ease of getting

an appointment

Direct costs

Costs of care e.g. Average cost

of care e.g. Per capita costs Volume e.g. Care

utilization e.g. Number of hospital admissions Costs PM organization e.g. Improvement

process e.g. Start-up costs Indirect costs Productivity losses e.g. Personnel

(26)

2

17

grouped into subdomains to compress the number of measures and create an overview for analyses. The relation between dimensions, domains, subdomains and measures is visualized in Figure 1. Measures were further characterized and clustered by measure type; these included process, outcome and structure as defined by Donabedian [26], and patient reported outcome measures (PROMs) or patient reported experience measures (PREMs) [15].

Results

Search results

The search strategy produced 62 potential PM initiatives; 20 were found in grey literature [27-30], 6 in published articles [31, 32], 13 originated from author discussions and 23 from experts’ suggestions. These 23 suggestions also included PM initiatives found in literature provided by experts [23]. A total of twenty PM initiatives met the six inclusion criteria and were included in the study (Appendix 2). The 42 excluded initiatives were mostly rejected because they did not have a specified population or an available evaluation strategy (Appendix 2).

Description of PM initiatives

(27)

18

Table 1 |

Description of

the included P

opulation Management initiatives

Initiative [including refer

ences] Country , Y ear of implementation Actors Tar get P opulation Goals

Design (Organizational structur

e and implemented pr

ogr

ams)

Alberta He

alth

Services - Edmonton Zone (AHS)

[33-35] Canada, 2012 Private or ganizations, he alth car e pr oviders Specific gr

oups in the Edmonton

zone,

selected based on car

e

use and housing situation

Advance the

Triple

Aim for high need

and high cost patients.

AHS focuses on sever

al populations; the older

, tri-morbid

adults,

fr

ail older adults,

young adults with addictions and

mental he alth concerns, child-be aring women, high-needs childr en and comple

x infants and toddlers.

Five pr

ojects aimed

at e

ach of

the subpopulations wer

e intr oduced. AHS is part of the IHI Triple Aim cohort.

Alternative Quality Contr

act (A QC) [36-38] United S tates, 2009 Insur ance company , private or ganizations Insur

ed population that uses a

primary car e physician (PCP) who participates in an AQC gr oup The

AQC pursues the

Triple Aim by enabling car egivers to pr ovide the right car

e at the right time in the most

appr opriate set ting, and by r educing expense tr

ends for participating

or

ganizations.

Car

e pr

oviders can opt-in to participate in

AQCs.

The

AQC is a

global payment model that combines set per

-patient payments

with shar

ed savings.

The global budget entails all services and

costs for a particular patient.

Bellin He alth of Gr een Bay [23, 39, 40] United S tates, 2002 Private or ganizations Employees of Bellin He alth

and their spouses

Achieve the

Triple

Aim with a focus on

costs.

The model used by Bellin He

alth is called the

Total He

alth

Model.

Employees and their spouses r

eceive he alth insur ance benefits, he alth car e coaching, and an annual he alth risk appr aisal (HRA). Bidasoa Integr ated He alth Or ganization (IHO) [41-43] Spain, 2011 Government, he alth car e pr oviders Population of Bidasoa Impr ove quality of car e using evidence based pr

actice and coor

dination

between levels,

and impr

ove efficiency

while being patient-center

ed.

IHOs consist out of

a hospital matched with primary car

e

or

ganizations,

based on their geogr

aphical ar ea. IHOs have a single sour ce of funding,

common goals and risk str

atification

for primary and secondary car

e.

Bidasoa is based on cultur

e,

clinical pr

actice and governance.

Implementations such as Continuity of C ar e Units, Technical Boar ds and shar ed electr onic recor ds ar e used.

Canterbury District Health Boar

d (DHB) [44] New Z ealand, 2011 Government Population of the C anterbury District Achieve integr ated car e to do the right

thing for the patient and the system by letting people tak

e r

esponsibility for

their own he

alth,

pr

oviding continuity and

impr

oving r

esponsivenes

s to

episodic events.

Canterbury DHB focuses on home car

e,

services in the

community

, r

egional collabor

ations and network

s and

managed specialization.

P

rojects could be clas

sified by pr ojects connecting he alth systems, supporting vulner able populations

and delivering clinically

, and financially viable he

(28)

2

19 Initiative [including refer

e and implemented pr ogr ams) Clinical Commis sioning Gr oups (C CG’ s) [45-47] England, 2012

Government, autonomous governmental organizations

Population of a specific geogr aphic ar ea Incr

easing focus on patient- and

public-center ed car e and impr oving outcomes, autonomy , accountability , and efficiency , while cut ting bur eaucr acy . Gener al pr

actitioners (GP) become members of

C

CGS; they

cooper

ate to plan and design local he

alth services,

govern the

CC

Gs,

and they ar

e an independent statutory body

. T

hey contr

ol

the budget and buy local he

alth car e services on behalf of the local population. Ther e ar e curr ently 211 C CGS is England with varying appr oaches. Coor dinated C ar ed Model (C CO) [32, 48-51] United S tates, 2012 Government, private or ganizations Medicaid clients in Or egon Impr ove he alth, quality , r eliability ,

availability and continuity of

car e CC Os ar e r egional entities r

esponsible for managing a global

budget and coor

dinating he

alth services for Medicaid in

Or

egon.

C

COs try to achieve the goals by integr

ating,

local

accountability

, and standar

ds for safe and ef

fective car e. Most CC Os have intr oduced community he alth work ers or community

agencies into their network.

Counties Manuk au District He alth Boar d (DHB) [52-55] New Z ealand, 2011 Government Populations of the local authorities in Auckland, W aik

ato and Haur

aki District

Become the best he

alth car e system in Austr alasia in 2015 by balancing ex

cellence and sustainability

. A four -ye ar str ategic fr amework was cr

eated in which six

str

ategies wer

e described based on the domains of

the Triple Aim. For e xample, bet ter he

alth outcomes for all,

system integr

ation and ensuring financial sustainability

. T

he

accompanying implementations wer

e electr

onic medication

conciliation,

fall and infection pr

evention and the 20.000 day

campaign.

Gesundes Kinzigtal [56-58]

Germany

, 2006

Local physicians network,

he alth car e management company , insur ance companies People insur ed by a specific insur

ance company in the

Kinzigtal r egion Achieve the Triple Aim; impr ove insur er ’s contribution mar

gin and the he

alth

status as well as the quality of

life of

the

population.

One or

ganization is r

esponsible for the spending and car

e of two sicknes s funds populations. They ar e r ewar ded by using shar ed savings and ar

e financially accountable for all members

of the sicknes s funds. Since 2006, thirty car e and pr eventive pr ogr ams wer e intr oduced. He alth car e Mark etplace Collabor ative [28, 59, 60] United S tates, 2009 Private or ganizations

Intel employees in the United States and their dependents

Pr

ovide bet

ter

, faster and mor

e af for dable car e. Intel hir ed Vir ginia Mason to tr ain personnel in le an and best pr actice pr oces ses. This way pr oven value str eams fr om Vir

ginia Mason wer

e adapted and implemented in other he

alth

car

e systems for patient scr

eening,

illnes

ses and six medical

(29)

e and implemented pr ogr ams) He althP artners Best Car e [28, 61, 62] United S tates, 2004 Insur ance company , he alth car e pr oviders Members of the insur ance company Achieve the Triple Aim by impr oving reliability , customization/tailoring of car e and by ensuring e asy acces s and coor dination Examples of implementations wer e the intr oduction of an online IT

and information-sharing system,

evidence-based car

e,

generic pr

escribing,

best pr

actice for MRI,

support of patients af ter dischar ge, same-day mammogr

aphy and phone-based

24-hours nurse car

e. He althy Shelby [63, 64] United S tates, 2007 Private or ganizations, he alth car e pr oviders Residents of Memphis and Shelby C ounty Achieve the Triple

Aim and incr

ease busines s gr owth, quality of life and community he alth, and r educe he alth car e bur den The Memphis F orwar

d Initiative is focused on collective action

by the dif

fer

ent actors in the community

. He

althy Shelby is part

of

this initiative and decided to focus on thr

ee ar

eas: infant

mortality

, chr

onic dise

ase and living well/dying well.

P rojects wer e implemented for e ach of the ar eas of focus. Medicar e Shar ed Savings P rogr am

(MSSP) (including Advanced payment and P

ioneer model) [31, 65-67] United S tates, 2011 Government, private or ganizations Populations as signed to Accountable C ar e Or ganizations (A COs) Achieve the Triple Aim by pr omoting accountability

, incenting higher value

car e, impr oving coor dination and infr astructur e, and r

edesign the car

e

pr

oces

ses.

AC

Os can choose to participate in the MSSP

; the shar

ed savings

ar

e link

ed to quality and savings r

equir

ements.

The

AC

Os can

choose their own implementations to fulfill these r

equir ements, which ar e based on quality me asur es. Minnesota He alth Model [32, 68-71] United S tates, 2013 Government, private or ganizations Populations as signed to AC Os in Minnesota To ensur e patients r eceive patient-center ed primary car e and pr oviders

include the community

, participate in

accountable car

e and payment models

to tak

e r

esponsibility for the populations’

he

alth.

The model is based on (primary) he

alth car e homes, community car e tr ansitions and

Accountable communities for He

alth, that need to include an AC O. Local stak

eholder get fle

xibility on how

they implement these building block

s National He alth Service (NHS) Highland [72-75] Scotland, 2010 Government Population of Highland r egion Pr ovide safe, sustainable, compas sionate, ef fective, af for

dable and efficient delivery

of

he

alth and social car

e and impr ove population he alth, while r educing he alth inequalities

NHS Highland uses the He

alth car e S ystem Appr oach focused on person-center ed car e while r

educing waste and harm,

and managing variation. One of the implemented pr ojects was the C ar e P athways, a r

oute capturing all activities for patients

from start to finish.

Others include lifestyle interventions and

(30)

2

e and implemented pr ogr ams) NHS K ernow England, 2013 Government Residents of C

ornwall and Isled

of

Scilly

Impr

ove he

alth,

wellbeing and people’

s

experience of

car

e,

and r

educe the costs

of car e (T riple Aim). Ten locality gr oups commis sion and as ses s the he alth car e needs of the population. They r

eceive an annual budget fr

om

the NHS England.

Multiple pr

ojects wer

e aimed at including the

community

, while others included bet

ter diagnosis by the GP to

pr

event avoidable admis

sions, among others. Pueblo T riple Aim Coalition (P TA C) [76-81] United S tates, 2010 Non-pr ofit or ganization, private or ganizations Population of P ueblo C ounty Achieve a set of

specific goals based on

the Triple Aim; lower ye ars of potential life lost, incr

ease the clinical acces

s and quality r anking, and r educe illnes s bur den and pr

eventable hospital stays.

Thr

ee dominant pr

ogr

ams that wer

e implemented ar

e

obesity r

eduction (activity and food),

lower teen unintended

pr

egnancies (youth development) and smoking ces

sation

(counseling and aid).

Regional C ar e Collabor ative Or ganizations (RC CO) [32, 82, 83] United S tates, 2011 Government Accountable C ar e C ollabor ative enr ollees Ensur e acces s to car e, coor dinate medical

and non-medical car

e,

impr

ove member

experience,

and pr

ovide the neces

sary

data to support these goals.

RC COs ar e contr acted by Medicaid to cr eate network s of PCP s

they have to coor

dinate and to integr

ate car

e for Medicaid

enr

ollees.

RC

COs use varying str

ategies.

Some e

xamples

include embedded car

e managers and funding pools at

community level. St. Charles He alth System [84-86] United S tates, 2008 Private or ganizations Population of centr al Or egon “Cr eate America ’s he althiest community , together ” thr ough bet ter he alth, bet ter car e and bet ter value.

Each goal entails a set of

pr ojects. Bet ter he alth, for e xample, has pr

ojects aimed at integr

ating data,

partnerships with the

community and patient-center

ed medical homes.

In 2013 S

t.

Charles He

alth S

ystem started a partnership with the Institute of

He

althcar

e Impr

ovement focusing on high risk patients.

Torbay and Southern Devon He

alth and Car e NHS T rust [87-89] England, 2000 Government Population of

Torbay and South

Devon

De

al with the rising demand and the

diminishing budget while ensuring people r

eceive the right car

e,

in the right

place,

at the right time

The ‘Mrs.

Smith’ model is r

evolves ar

ound the an imaginary

patient that helps visualize the needs of

actual (futur

e) patients.

This model led among others to the implementation of

‘joined-up car

e’ between he

alth and social car

e. The C CG, in cooper ation with the Trust,

has also implemented virtual war

ds,

a method of

pr

eventive risk str

(31)

22

Measurement of the dimensions and (sub)domains

The 20 PM initiatives combined, used 865 measures of which 103 measures related to population health, 585 to quality of care and 177 to costs. An overview of the number of initiatives found per domain can be seen in Figure 2.

0 2 4 6 8 10 12 14 16 18 Health Outcome s Disease burden

Behavioral and physiologica

l factors Partcipation Functioning/quality of life Patient safety Effectivit y Responsiveness Support Timelines s Accessibilit y Costs of care Volum e Costs of PM organization Productivity losses

Population health Costs

Initiatives

Quality of care Figure 2 | Overview of number of initiatives found per domain

Population health

The most often evaluated domain was ‘health outcomes’ (n=13), in which the subdomains ‘mortality’ (n=8) and ‘overall health status’ (n=7) were most frequently measured (Table 2). Ten PM initiatives evaluated the ‘behavioral and physiological factors’ domain. Nine of these considered ‘lifestyle characteristics of the population’ (e.g. smoking status) and four included the ‘weight’ subdomain. The domains ‘disease burden’ (n=5), ‘participation’ (n=2) and ‘functioning/quality of life’ (n=5) were less prevalent. The number of measures used, varied greatly within each subdomain, ranging from 1 to 33 measures.

Quality of care

(32)

2

23

‘communication between patient and caregiver’ (n=9). Thirteen PM initiatives evaluated the ‘timeliness’ domain and both ‘accessibility’ and ‘support’ were evaluated by eleven PM initiatives. Dominant subdomains within ‘timeliness’ and ‘accessibility’ were ‘access to medical care’ (n=8) and ‘(waiting) time needed in hospital process’ (n=9). The number of measures per subdomain varied between a single measure and 69 measures.

Table 2 | Domains, subdomains and measures used for population health

Domain Subdomains # of initiatives evaluating subdomain

# of measures in subdomain

Example of used measures in subdomain

Health Outcomes

Measured by 13 initiatives

Mortality 8 11 Under 75 mortality form cardiovascular

diseases

Overall health status 7 8 Change in health

Life expectations 2 2 Life expectancy at birth

Medication utilization 1 1 Share of the population with

multi-medication Disease burden

Disease prevalence 4 19 Proportion of patients with diabetes

diagnosis

Disease incidence 3 3 Chlamydia rate per 100,000 residents

Care utilization 1 1 Illness burden (measures relative health

of a group based upon the number and types of health care services used)

Comorbidity 1 3 Proportion of insured with

multi-morbidity Behavioral and physiological

factors

Lifestyle characteristics of the population

9 33 The percentage of the population (15+)

who smoke

Weight 5 9 Obesity prevalence

Mortality 1 1 Percentage of driving deaths with alcohol

involvement

Provision of preventive care 1 1 MMR vaccine uptake rate

Participation

Dependent patient prevalence

1 1 Proportion of insured with working

disability

Social inclusion 1 1 Social inclusion (of mental health and

addiction clients) Functioning/quality of life

Quality of life 4 6 Health-related quality of life for people

with long term conditions Dependent patient

prevalence

2 3 Proportion of insured with aid- and safety

(33)

24

Table 3 | Domains, subdomains and measures used for quality of care

Example of used measures in subdomain

Patient safety

Complications 11 63 Number of Clostridium Difficile Health Care

Associated Infections

Protocols and guidelines 4 5 Rate of compliance with good hand hygiene

practice Care instruction provided

to patient

3 10 Percent of patients who reported that staff

“Always” explained about medicines before giving it to them

(Un)safe use of medication2 8 Proportion of insured with Benzodiazepines

(>20 DDD)

Care utilization 2 2 Early elective delivery

Mortality 2 5 Heart bypass surgery mortality

Provision of correct treatment

2 2 % of patients that receive treatment in

accordance with validated research Provision of preventive

screening

2 2 % of inpatients (aged 75+) who received a

falls assessment Communication between

health care professionals

1 1 Safe use of abbreviations

Effectivity

Follow-up care 10 35 Diabetes eye exam

Provision of preventive screening

9 28 Proportion of women aged 50-69 years who

have had a breast screen in the last 24 months

Clinical outcomes 8 21 Percent of beneficiaries with diabetes

whose HbA1c in poor control (>9 percent) Hemoglobin A1c Control HbA1c)

Medication utilization 6 19 Proportion of patients with CHD and beta

blockers

Mortality 6 14 Pneumonia mortality rate per discharge

Avoidable care prevalence

4 8 Percentage of hospital readmissions within 30

days of discharge Patient receiving

self-directed support

4 7 (Provision of) patient assessment for

self-management Provision of lifestyle

interventions

4 13 % Opiate or Crack Users in effective drug

treatment

Care utilization 3 10 Proportion of patients who had at least one

hospitalization

Optimal care 3 11 Optimal diabetes care

Recovery 3 5 Depression remission at six months

Pain control 2 2 Percent of patients who reported that their

pain was “Always” well controlled Provision of correct

treatment

2 2 Proportion of Counties Manukau residents

(34)

2

25

Effectivity (CONTINUED)

Absence due to sickness 1 1 Proportion of back pain patients incapable to

work because of low back pain for more than 14 days

Composite measure (Functioning and Quality of life)

1 3 Functional status and quality of life outcome:

total knee replacement Delivery of ineffective

care

1 2 Ineffective pharmacotherapy in Alzheimer’s

disease

Disease survival rate 1 2 One year survival from all cancers

End-of-life care 1 1 End-of-life measure (percentage of time spend

at home during the last six months of life)

Integration of services 1 3 % of patients that stated that hospital and

primary care coordination was good/very good Lifestyle characteristics

of the population

1 1 % of babies exclusively breastfeeding on

hospital discharge

Patient knowledge 1 2 Knowledge of the patient (Consumer

Assessment of Health care Providers and Systems)

Responsiveness

Patient satisfaction 13 24 Patients who reported YES they would

definitely recommend the hospital Communication between

patient and caregiver

9 29 Percentage of surveyed patients that were

‘very satisfied’ with communication and coordination of experience

Participation of patient in

the care process

4 6 Shared decision making

Integration of services 3 6 (Referring/) Connecting the patient to the

community resources Patient experience of

care environment

3 5 Percent of patients who reported that their

room and bathroom were “Always” clean (Waiting) time needed

in hospital process

2 2 Percent of patients who reported that they

“Always” received help as soon as they wanted Change rate of insurance

company

1 1 Change rate (number of people changing

insurance company)

Complaints about care 1 2 Complaints: total number received – hospital

1 1 Preventive care reminders and data

Support

Communication between health care professionals

5 21 Documentation send prior to or within 60 after

patient ED transfer

Improvement process 5 29 Number of caregivers that complete the Soul

and Science of Caring program

(35)

26

Evaluation of care giver 3 4 % of carers receiving an assessment

Absence due to sickness 2 2 Sickness absence of staff

Integration of services 2 2 Engagement with community based providers

Staff capacity 2 6 Staff turnover rate

Communication between patient and caregiver

1 1 Innovation in interaction between patients and

professionals

Environmental impact 1 2 Carbon emissions of initiative

Protocols and guidelines 1 1 (Agreement on) protocolization

Workforce diversity 1 1 Improved workforce diversity as a percentage

by ethnicity compared to population percentage by ethnicity

Timeliness

(Waiting) time needed in hospital process

9 69 Waiting in the exam room

Follow-up care 6 11 Proportion of diabetes patients with

ophthalmologist contact in two years Provision of preventive

screening

4 6 Proportion of patients referred urgently

with high suspicion of cancer to first cancer treatment within 62 days

Medicalization utilization 2 7 Aspirin at arrival at emergency department

Provision of lifestyle interventions

2 4 % safeguarding referrals (reporting abuse)

conferences within 30 days Patient receiving

1 1 People with diabetes diagnosed less than a

year referred to structured education Accessibility

Access to medical care 8 21 Getting an appointment for routine care

Care utilization 4 16 % of adults (20-64) accessing to specialist

mental health services

Spread of caregivers 2 4 Ratio of population to primary care physicians

in a county Access to non-medical

facilities

1 2 Percentage of the population who live

reasonably close to locations for physical activity, including parks or recreational facilities

Communication between health care professionals

1 1 % of admissions into assessment, treatment

& rehabilitation made by direct community referral

Insurance coverage 1 2 Percentage of the population younger than

age 65 without health insurance Participation of patient in

care process

(36)

2

27

Costs

In the costs dimension (Table 4) the most prevalent domain was ‘volume’ (n=17), in which thirteen PM initiatives evaluate the ‘care utilization’ (e.g. number of GP visits per inhabitant). The domains ‘costs of care’ (n=7) and ‘costs of PM organizations’ (n=6) were less prevalent. The most often used subdomains within these two domains were ‘average costs of care’ (n=6) in ‘total costs’ and ‘financial performance’ (n=6) in ‘costs of PM organization’. For each of the subdomains, 1 to 24 measures were used. The ‘productivity losses’ domain was not measured by any PM initiative.

Table 4 | Domains, subdomains and measures used for costs

Example of used measures in subdomain

Costs of care

Average costs of care 6 11 Average costs of inpatient stay

Substitution 2 2 Specialized care vs. primary care

expenditure (%)

Volume

Care utilization 13 57 Number of ED visits of population

Avoidable care prevalence 11 24 Acute hospital readmissions

11 36 Mammography screening

Provision of lifestyle interventions

6 10 ‘Appetite for life’ nutrition courses

provided in the community

Medication utilization 4 14 Regulation scope for generics

Claims send to insurance company

1 3 Total medical claims

Integration of services 1 1 Session with mental health & addiction

counsellors Patients receiving

1 7 People supported by district nursing

services Costs PM organization

Financial performance 6 9 Achievement of 1% surplus of health

commissioned spend

Improvement process 1 3 Delivery of Cost Improvement Program

savings plan Productivity losses

Measuring populations: In search of methods to evaluate health and experienced quality of care

Tilburg University

Measuring populations

Hendrikx, R.J.P.

Publication date:

2019

Document Version

Publisher's PDF, also known as Version of record

Link to publication in Tilburg University Research Portal

Citation for published version (APA):

Hendrikx, R. J. P. (2019). Measuring populations: In search of methods to evaluate health and experienced

quality of care. Ipskamp.

Measuring

Populations

In search of methods to evaluate

health and experienced

quality of care

Roy Hendrikx

Measur

ing

Populations

R

oy

Hendr

ikx

UITNODIGING

Voor het bijwonen van

de openbare verdediging

van het proefschrift

Measuring

Populations

In search of methods

to evaluate health

and experienced quality

of care

Op vrijdag 21 juni 2019

om 10 uur in de aula

van Tilburg University,

Cobbenhagen gebouw,

Warandelaan 2, Tilburg

Measuring

Populations

In search of methods to evaluate

health and experienced

quality of care

Colofon

Measuring populations

Promotores

Copromotores

Promotiecommissie

Contents

Chapter 1

1

Chapter 2

11

Chapter 3

43

Chapter 4

69

Chapter 5

95

Chapter 6

117

Chapter 7

135

Chapter 8

159

173

177

181

185

Chapter 1

Shifting focus towards regional care in order to achieve the Triple Aim

1

An emphasis on evaluation

1

The conceptual struggles of population health and quality of care

Aim

The Dutch setting

Outline

_{In search of methods to evaluate}

_{In search of methods to evaluate}