2012-1350 1..8

By Geraint Lewis, Heather Kirkham, Ian Duncan, and Rhema Vaithianathan

How Health Systems Could Avert

‘Triple Fail’ Events That Are

Harmful, Are Costly, And Result

In Poor Patient Satisfaction

ABSTRACT

Health care systems in many countries are using the “Triple

Aim”—to improve patients’ experience of care, to advance population

health, and to lower per capita costs—as a focus for improving quality.

Population strategies for addressing the Triple Aim are becoming

increasingly prevalent in developed countries, but ultimately success will

also require targeting specific subgroups and individuals. Certain events,

which we call “Triple Fail” events, constitute a simultaneous failure to

meet all three Triple Aim goals. The risk of experiencing different Triple

Fail events varies widely across people. We argue that by stratifying

populations according to each person’s risk and anticipated response to

an intervention, health systems could more effectively target different

preventive interventions at particular risk strata. In this article we

describe how such an approach could be planned and operationalized.

Policy makers should consider using this stratified approach to reduce

the incidence of Triple Fail events, thereby improving outcomes,

enhancing patient experience, and lowering costs.

he “Triple Aim” of health care is to

improve individual patients’ expe-

riences of care, advance population

health, and reduce per capita

health care costs.

A central tenet

of the Triple Aim is to restructure care in ways

that lead to improvements across all three of

these goals.

The Institute for Healthcare Improvement has

worked with organizations in many countries to

implement populationwide interventions to

foster the Triple Aim.

Examples are programs

that encourage self-management of chronic con-

ditions,

promote e-mail communication be-

tween patients and physicians,

and encourage

greater use of primary care.

Other organizations have adopted a more

targeted approach to achieving the Triple Aim.

For example, a Commonwealth Fund case study

found examples of organizations that were

focusing on improving access and care for

individual patients who had low incomes, were

uninsured, or had complex chronic conditions.

Indeed, several authors have argued that success

will require both population health and in-

dividually focused strategies, such as those

employed by Genesys Health System in Flint,

Michigan.

1,3,6,7

For example, Genesys increased

its primary care capacity (that is, a population

approach) and offered health navigators to its

high-risk patients (that is, a targeted approach).

The objectives of this article are to propose a

third, stratified approach to tackling the Triple

Aim and to explore some of the ethical chal-

lenges that this new approach presents. The

stratified approach to the Triple Aim involves

identifying and prioritizing subpopulations ac-

cording to their risk of experiencing health

encounter failures—what we call “Triple Fail”

events—and according to their likelihood of

benefiting from preventive care.

6,8

We define a Triple Fail event as a health outcome

doi: 10.1377/hlthaff.2012.1350

HEALTH AFFAIRS 32,

NO. 4 (2013): –

The People-to-People H ealth

Foundation, Inc.

Geraint Lewis (geraint.lewis@

nhs.net) is chief data officer

of the National Health

Service, in London, England.

Heather Kirkham is a manager

in the Clinical Outcomes and

Analytics Departm ent at

Walgreens, in Deerf ield,

Illinois.

Ian Duncan is the vic e

president in the Clinical

Outcomes and Analytics

Departmen t at Walgreens.

Rhema Vaithianathan is a

senior research fellow at Sim

Ki Boon Insti tu te, Singa por e

Management University, and

director of the Centre for

Applied Research in

Economics, University of

Auckland, in New Zealand.

April 2013 32:4 Health Affairs 1

Quality

Governance

that is recorded in administrative data and arises

from the health care process. Such events simul-

taneously have three failures: They are costly,

represent a suboptimal health outcome, and

are a poor patient experience. To generate the

list of Triple Fail events in Exhibit 1, we applied

our definition to the published literature to con-

firm each example’s failure on all three Triple

Aim goals.

Triple Fail Events

Triple Fail events include untimely nursing

home admissions,

unplanned hospital re-

admissions,

10–12

inappropriate initiations of

hemodialysis,

13–15

and surgeries for low back pain

in patients not offered decision support.

For

example, nursing home admission was cited as

a common fear among older people,

was often

preceded by poor physical and emotional

health,

and was estimated to cost $203–

$243 billion annually in the United States in

2009.

Other Triple Fail events are more controver-

sial, such as overmedicalized death, which may

be defined as receiving “life-sustaining treat-

ments, such as mechanical ventilation, that most

beneficiaries indicate they prefer to avoid when

faced with less than a year to live.”

Exhibit 1

Examples Of Triple Fail Events

Event Quality of care Patient experience Cost

Unplanned hospital

readmission within

30 days

A readmission may indicate complications,

a premature discharge, a failure to

coordinate and reconcile medications,

inadequate communication, or poor

discharge planning

Higher 30-day risk-

standardized hospital

readmission rates are

associated with lower

patient satisfaction

A 2009 study found that 30-day

rehospitalizations cost Medicare

$17.4 billion annually

Nursing home admission Predictors of nursing home admission

include low self-rated health status and

functional and cognitive impairment

Loss of independence and

nursing home admission

are two of the major fears

of older people

The cost of long-term care in the

United States in 2009 was

estimated to be $203–$243 billion

Inappropriate initiation of

hemodialysis

Peritoneal dialysis patients experienced

a lower adjusted relative risk of death

compared with those beginning

hemodialysis;

late and early dialysis

initiation appear to be associated with

equal outcomes

Peritoneal dialysis patients

reported better quality of

life and better satisfaction

with dialysis care

Median annual health care costs in

2004 were $43,510 higher for

hemodialysis patients than

peritoneal dialysis patients

Wrong-site surgery Wrong-site surgery is one of the National

Quality Forum’s “never events”

Wrong-site surgery can be

a devastating experience

for the patient

A 2010 study found that the average

compensation paid to victims of

wrong-site surgery was $47,216

Intentional injury or

maltreatment of a child

Child maltreatment involving physical

abuse is the leading cause of infant

death from injury

Child abuse has been

associated with a wide

range of psychological

symptoms in the victim

A 2012 study found the annual acute

medical costs and annual societal

costs of childhood abuse in the

United States were $2.9 billion and

$80 billion, respectively

Overly invasive treatment

for a preference-sensitive

condition

Patients offered a decision aid for a

preference-sensitive condition reported

significantly improved outcomes

Patients offered a decision

aid reported significantly

greater satisfaction with

their selected treatment

Patients offered a decision aid were

21–44% less likely to choose

costly, aggressive surgery

SOURCE Authors’ analysis. Sources are listed by individual entries in this exhibit. NOTE Apreference-sensitiveconditionhasmultipletreatmentoptionswithroughly

equivalent risks and benefits, which me ans that th e patient’spreferenceshoulddeterminewhichoptionisselected.Anexampleislowbackpain,whichcanbe

treated by sur gery, p hysical therapy, or analge sics.

See Note 10 in text.

See Note 11 in text.

See Note 12 in text.

See Note 17 in text.

See Note 9 in text.

See

Note 18 in text.

Mehrotra R, Chiu YW, Kalantar-Zadeh K, Bargman J, Vonesh E. Similar o utcomes with hemodialysis and peritoneal dialysis in patients with end-

stage renal disease. Arch Intern Med. 20 11;171(2):110 –8.

See Note 13 in text.

See Note 14 in text.

See Note 15 in text.

Michaels RK, Makary MA, Dahab Y,

Frassica FJ, Heitmiller E, Rowen LC, et al. Achieving the National Quality Forum’s “Never Events”:preventionofwrongsite,wrongprocedure,andwrongpatient

operations. Ann Surg. 2007;245(4):526–32.

Mulloy DF, Hughes RG. Wrong-site surgery: a preventable medical error. Chapter 36 in: Hughes RG, editor. Patient

safety and qualit y: an evi dence-base d handbook for nurses. Rockville (MD): Agency for Healthcar e Research and Quality ; 2008 M ar. p . 381–94.

Stahel PF, Sabel

AL, Victoroff MS, Varnell J, Lembitz A, Boyle DJ, et al. Wrong- site and wrong-patient proced ures in the universal protoc ol era: analysis of a prospective databa se of

physician self-reported occurrences. Arch Surg. 2010;145(10):978–84.

Overpeck MD, Brenner RA, Trumble AC, Trifiletti LB, Berendes HW. Risk factors for infant

homicide in the U nited States. N Engl J Med. 1998;33 9(17):1211– 6.

Gross AB, Keller HR. Long-term consequences of childhood physical and p sychological

maltreatment. A ggress Beh av. 2006;18( 3):171–85.

Gelles RJ, Perlman S. Estimated annual cost of child abuse and neglect [Internet]. Chicago (IL): Prevent Child

Abuse America; 2012 Apr [cited 2013 Feb 27]. Available from: http://www.preventchildabuse.org /downloads/PC AA_Cost_Report_2012 _Gelles_Per lman_final. pdf.

See

Note 16 in text.

Quality

Governance

2 Health Affairs April 2013 32:4

Approaches To Preventive Care

Population And Targeted Approaches The

two leading approaches to preventive care—the

population strategy and the targeted (high-risk)

strategy—were described in a seminal article by

the British epidemiologist Geoffrey Rose.

The

population strategy seeks to shift the distribu-

tion of risk within an entire population toward a

lower range—for example, by decreasing the

amount of salt in the typical diet to reduce the

population’s average blood pressure. The tar-

geted strategy aims to truncate the risk distri-

bution by identifying high-risk individuals and

offering them interventions to reduce their indi-

vidual susceptibility—for example, by screening

blood pressure among the population and offer-

ing medication to people with hypertension.

An important advantage of the population

strategy is its potential to make widespread im-

provements in public health.

However, the ben-

efit to each individual is relatively small using

this approach. Most people experience no par-

ticular improvement in their health—a result

known as the “prevention paradox.”

20,21

In con-

trast, the targeted strategy provides customized

care that maximizes outcomes for individual pa-

tients. But such customized care will rarely be

cost-effective for all patients, and hence it should

not be offered universally.

The choice of preventive strategies should de-

pend on the cost and effectiveness of the pro-

posed intervention. Both of Rose’s strategies

may be cost-effective when the intervention cost

is low. But populationwide approaches tend to be

more appropriate when the intervention’s effect

is large, while targeted approaches are more suit-

able when accurate predictive models are avail-

able and an optimal risk threshold for interven-

tion is used. For example, treatment may be

determined by a risk score cut-off point, which

is based on a simulation of the costs and benefits

and is designed to maximize the cost-effective-

ness of the intervention.

Rose’s hypothesis was formulated prior to the

development of accurate multivariable risk-

prediction tools.

More recent analysis of pop-

ulation data suggests that targeted, high-risk

approaches may be particularly advantageous

under certain circumstances, such as when the

intervention has a degree of disutility—for exam-

ple, if the intervention is considered a poor pa-

tient experience, which might include cost, lost

time, or an adverse lifestyle change.

The Stratified Approach Our novel, strati-

fied approach to the Triple Aim could generate

additional value by combining some of the ad-

vantages of the other two approaches.

This third

approach is best adopted by organizations with

responsibility for a population’s health, such as

accountable care organizations. It involves ana-

lyzing medical claims, pharmacy claims, elec-

tronic health record information, and other ad-

ministrative data to predict individuals’ risks of

different Triple Fail events. The organization

would next estimate each person’s likely re-

sponse to a range of preventive programs and

then assign people to different interventions ac-

cording to their likely benefit.

Compared with the population at large, the

subpopulation of each risk stratum would be

relatively homogeneous, allowing the interven-

tion to be customized to meet the needs of the

patients in that stratum. For example, a program

aimed at preventing hospital readmission might

offer case management, telephonic care, and re-

mote monitoring, with more intensive interven-

tions offered to patients in higher risk strata.We

recommend that an ethics committee approve

any algorithms used for targeting interventions

prior to implementation and that excluded pa-

tients be considered as part of a feedback loop for

program evaluation and improvement.

Within a high risk stratum, the average per-

son’s risk of experiencing the Triple Fail event

will be higher than the average population risk.

As a result, a higher proportion of individuals in

these groups could benefit from preventive care.

In other words, this stratum would have a higher

positive predictive value, which in turn would

increase the cost-effectiveness of the preventive

intervention, all other things being equal. How-

ever, the stratified approach is beset by a number

of challenges, principally those relating to the

ethical aspects of risk stratification, which are

described below.

The stratified approach to the Triple Aim de-

scribed in this article includes three phases. A

planning phase would involve conducting an op-

portunity analysis, developing predictive models

and impactibility (also known as intervenability)

models. The latter are models that seek to iden-

tify subgroups of high-risk people who are most

likely to engage with and respond to various

preventive interventions, such as case manage-

ment. The planning phase would also include an

ethical review to ensure its compliance with our

adaptation of James Wilson and Gunner

Jungner’s prerequisites, described below.

An operational phase would use the predictive

models and impactibility models to identify

high-opportunity patients—those who are both

at risk and amenable to an intervention—and

offer them preventive interventions. An ongoing

feedback phase would refine the predictive mod-

els and impactibility models—for example, by

prioritizing patients with characteristics similar

to those of patients who responded well to the

intervention.

April 2013 32:4 Health Affairs 3

Implementing The Stratified

Approach

Opportunity Analysi s The starting point for a

health system interested in pursuing the strati-

fied approach to the Triple Aim is to undertake a

detailed analysis of where the greatest opportu-

nities exist for improving care. Known in the

strategic management literature as opportunity

analysis, this process would, in the case of the

Triple Aim, involve analyzing historical popula-

tion data to identify Triple Fail events and gaug-

ing how responsive each such event might have

been to a cost-effective preventive intervention

identified from the literature.

Once a high-opportunity subpopulation of

patients has been identified in historical data,

people in the population with these character-

istics need to be identified prospectively to de-

termine who should be offered an intervention.

Therefore, a key requirement for the stratified

approach is the ability to identify patients who

are at risk of future Triple Fail events.

Predictive Modeling Predictive risk models

are statistical algorithms based on relationships

in historical population data. They may be ap-

plied in an automated fashion to routinely col-

lected data to estimate the probability that a

person will experience a Triple Fail event in a

specified future time period.

25,26

For example, the

combined predictive model is used by National

Health Service organizations in England to cal-

culate a person’s risk of unplanned hospitali-

zation in the next twelve months, according to

factors recorded in the previous two years’ worth

of primary care electronic health record data and

hospital claims data.

A predictive model for identifying vulnerable

people could potentially be built for any type of

Triple Fail event whose occurrence is recorded in

routine data. For example, researchers in New

Zealand have developed a model that predicts

the risk within five years of intentional injury

or maltreatment to an individual child.

A model

developed in the United Kingdom predicts the

risk of admission to a nursing home within

twelve months.

And Canada uses a model to

predict hospital readmissions within thirty

days.

Impactibility Modeling A recognized limita-

tion of using predictive risk models to organize

care is that some of the patients identified as

being at high risk may not be amenable to the

proposed preventive intervention.

In response,

many organizations have developed impactibil-

ity models that seek to identify the subgroups of

high-risk people who are most likely to engage

with and respond to various preventive inter-

ventions.

31,32

This additional filter is intended

to improve the cost-effectiveness of preventive

programs.

Several approaches to impactibility modeling

have been described elsewhere.

First, health

care organizations may prioritize people with

conditions known to be responsive to preventive

care, such as patients with an ambulatory

care–sensitive condition, who may be particu-

larly likely to respond to hospital avoidance

interventions.

Second, organizations may prioritize patients

whose care appears suboptimal, such as patients

with multiple “gaps in care.” An example of such

a gap would be not giving beta-blocker therapy to

a patient with heart failure.

Third, some organizations report that they

place lower priority on patients who are expected

to respond poorly to preventive care, such as

people with cognitive or other mental health

disabilities and those who have language bar-

riers. Or an organization may exclude all of

the very highest-risk patients, because such pa-

tients are sometimes regarded as being less ame-

nable than others to preventive care.

The first two approaches to impactibility mod-

eling may help reduce health care disparities,

because both suboptimal care

and ambulatory

care–sensitive conditions

tend to be more

common in people with low incomes. In con-

trast, the third approach—placing lower priority

on those less likely to benefit from preventive

care—raises serious ethical concerns. This ap-

proach would probably exacerbate health care

disparities, and it may be illegal in some coun-

tries.

Finally, because patients in very high risk

strata have such a high propensity for Triple Fail

events, expending resources to identify the few

who can be affected is usually worth the effort.

Ethics Of Screening Using Predictive

Risk Models

As Rose noted, the individual approach to health

improvement is hampered by the “difficulties

and costs of screening.”

20(p36)

A screening test

seeks to identify people who are at sufficiently

high risk of an adverse outcome to warrant offer-

ing them a diagnostic test or recommending a

prophylactic treatment. The stratified approach

to the Triple Aim likewise requires screening a

population to identify subpopulations that are at

sufficiently high risk of a Triple Fail event and

sufficiently amenable to a preventive interven-

tion to justify further action.

Any screening test has the potential to cause

more harm than good, such as by exposing pa-

tients to false positive and false negative results.

Therefore, strict ethical guidelines are required

to safeguard against the inappropriate use of

screening.

The World Health Organization

Quality

Governance

4 Health Affairs April 2013 32:4

published ten prerequisites, proposed by Wilson

and Jungner, that should be met by any ethical

screening program.

38,39

Among these pre-

requisites are that the condition being screened

for should be an important health problem; that

there should be a detectable early stage when

treatment would be of more benefit than it would

be later; and that the risks, both physical and

psychological, should be less than the benefits.

Because the stratified approach to the Triple

Aim involves population screening using routine

data, we suggest that equivalent caveats should

apply as part of the planning phase. Adapting

Wilson and Jungner’s prerequisites,

we pro-

pose the following ethical criteria for stratifying

populations according to risk for Triple Fail

events.

Prerequisites For Stratification The

event being predicted should be an important

health problem. There should be an intervention

that can mitigate the risk of the event; resources

and systems for timely risk stratification and

preventive interventions; sufficient time for in-

tervention between stratification and the occur-

rence of the event; a sufficiently accurate predic-

tive risk model for the event, which—together

with the impactibility model—is acceptable to

the population at large; and an accepted policy

about who should be offered the preventive

intervention.

In addition, the natural history—that is, the

practices and processes that typically lead to this

type of Triple Fail event—should be adequately

understood by the organization offering the pre-

ventive intervention. The cost of stratification

should be “economically balanced,” meaning

that it should not be excessive relative to the cost

of the program as a whole. And stratification

should be a continuous process, not just a “once

and for all” occurrence.

Access For Certain High-Risk Populations

Another important ethical concern relates to the

use of impactibility modeling. Although certain

subpopulations are at high risk, they may be

denied preventive care because they are not ex-

pected to respond to it. For instance, people with

personality disorders or alcohol dependency

might not be amenable to programs aimed at

preventing hospital readmission. The question

is whether such people should be denied preven-

tive care on this basis, which corresponds to

Wilson and Jungner’s requirement that there

be an accepted policy about who should be of-

fered the intervention.

Nancy Kass argued that for reasons of distribu-

tive justice, programs should not exclude indi-

viduals on the basis of nonclinical characteristics

such as race and sex.

However, Andrew Smart,

Paul Martin, and Michael Parker argued that

such discrimination may be justified on the prin-

ciple that it is permissible to treat people differ-

ently if there is some ethical justification. For

example, programs that target low-income or

uninsured people because of social justice con-

siderations rightly treat different people

differently.

Discussion

Establishing A Feedback Loop A feedback

loop is necessary for assessing the impact of

the preventive program on outcomes.

This step

may be valuable both for evaluating the program

and for refining the impactibility model.

For example, a regression analysis might show

that patients with certain characteristics were

more likely than others to respond to an inter-

vention. This knowledge could then be used to

adjust the impactibility model to ensure that pa-

tients with these characteristics were prioritized

in the future, unless that adjustment violated

ethical considerations.

Current Policy Context The Triple Aim is

becoming increasingly important to policy mak-

ers in developed countries as their populations

age, chronic diseases increase in prevalence,

and funding constraints become pressing.

Predictive modeling is now widely accepted in

the United States,

but the use of impactibility

models is less extensive.

New financial and quality rules are giving

hospitals and accountable care organizations in-

centives to prevent Triple Fail events such as

avoidable readmissions. However, many such

organizations are relying on population strate-

gies to achieve these goals, including better care

coordination and improvements in information

technology.

Ultimately, success will probably

require targeting specific subpopulations as

well.

Recommendations The following recommen-

dations could promote the appropriate use of the

stratified approach to the Triple Aim.

▸

USE PILOTS

: First, Triple Aim pilots should

be established in which demonstration sites are

required to compare population, individual, and

stratified approaches; conduct opportunity

analyses; and apply predictive risk and impacti-

bility models.

▸

REDUCE DATA LAGS

: Second, the use of pre-

dictive modeling should be promoted by reduc-

ing lags in the availability of information from

the Medicare limited data set as well as files

available for specific uses, such as those for

accountable care organizations. The Medicare

limited data set lags by more than a year, and

the specific-use files lag by several months. Since

some factors recorded in routine data are

April 2013 32:4 Health Affairs 5

strongly predictive of an imminent Triple Fail

event, reducing the time lag between the occur-

rence of such factors and their availability in

the data improves the accuracy of the predictive

model.

▸

CONDUCT AN ETHICAL REVIEW

: Third, there

should be appropriate ethical reviews of both the

design and implementation of predictive and

impactibility models.

▸

RECORD MORE EVENTS

: Finally, the use of

database indicators should be expanded to cover

more potentially adverse events. For example,

the Hospital Episode Statistics database in

England records whether each admission was

elective or not, thereby indicating whether the

event was a likely Triple Fail event.

Focus Of Fu ture Wo rk There is a growing

body of evidence for the ability of predictive mod-

els to identify and classify adverse outcomes.

However, relatively few studies have assessed

the amenability of people to respond to different

interventions.

Mathematical simulation mod-

els may help clinicians and administrators

choose between different approaches for

addressing the Triple Aim.

Given the current

focus on comparative effectiveness research, it

might be helpful if future trials of interventions

designed to prevent Triple Fail events examined

the effect of impactibility models on population

outcomes and on disparities.

Conclusion The Triple Aim is being used in

many countries to improve the quality, experi-

ence, and cost-effectiveness of health care. A

stratified approach to the Triple Aim offers a

number of potential advantages but requires

careful planning, monitoring, and adaptation.

By increasing the use of predictive modeling to

identify Triple Fail events before they occur, this

approach can contribute to improved individual

and population health in a cost-conscious health

care environment.

▪

Geraint L ewis , He ather Kirkham, and Ian

Duncan were employees of Walgreens at

the time this artic le was written.

Walgreens uses some of the techniques

described in thi s article, such as

supporting the accou ntable care

organizations of which the company is a

member. Le wis received research

funding from the Commonwealth Fund,

the UK Department for Communities

and Local Government, the UK

Department of Health, the Nuffield

Trust, the UK National Institute for

Health Research, the University of

Auckland Faculty Research Fund, and

the Basque Institute for Healthcare

Innovatio n. Dun can hol ds an ownershi p

share in SCIOinspire Corp., a privately

held company that performs predictive

modeling for clients on a consulting

basis, of 1.0–1.5 p ercent. Rhema

Vaithianathan received research funding

from the University of Auckland Faculty

Research Fund, the UK National

Institute for Health Research, and

variou s Re gio nal Distr ict H eal th Board s

in N ew Zealand. Lewis received

reimbursement of travel expenses to

speak at conferences relating to

predictive modeling. The funding came

from the Commonwealth Fund, the

Basque Institute for Healthcare

Innovation, the US Department of

Veterans Affairs, the University of

Auckland, the Germa n Sickness Fund

AOK, the Scott ish governm ent , th e

Royal C ollege of Physicians of London,

the Society of Physicia ns in Wales, the

Portugue se Na tio nal S cho ol o f Pub lic

Health, the Borås Health Region in

Sweden, and two of his previous

employers: the Nuffield Trust and

Walgreens. The authors are grateful to

Bradford Gray and Rob in Osborn, who

provided helpful comments on a

previous draft. The views expressed in

this article are those of the authors

alone and do not necessar ily represent

the views of their employing

organizations.

NOTES

1 Berwick DM, Nolan TW,Whittington

J. The Triple Aim: care, health, and

cost. Health Aff (Millwood). 2008;

27(3):759–69.

2 Institute for Healthcare

Improvement. The Triple Aim:

project prospectus for the United

Kingdom and Europe [Internet].

Cambridge (MA): The Institute;

2010 Apr [cited 2013 Feb 27].

Available from: http://www.ltu.se/

cms_fs/1.64419!/ta%20

international%20prospectus.pdf

3 McCarthy D, Klein S. The Triple Aim

journey: improving population

health and patients’ experience of

care, while reducing costs [Internet].

New York (NY): Commonwealth

Fund; 2010 Jul 22 [cited 2013

Feb 27]. Available from: http://

www.commonwealthfund.org/

Publications/Case-Studies/2010/

Jul/Triple-Aim-Improving-

Population-Health.aspx#citation

4 Zhou YY, Kanter MH, Wang JJ,

Garrido T. Improved quality at

Kaiser Permanente through e-mail

between physicians and patients.

Health Aff (Millwood). 2010;

29(7):1370–5.

5 Sweeney S, Bazemore A, Phillips RL

Jr., Etz RS, Stange KC. A re-emerging

political space for linking person

and community through primary

health care. Am J Prev Med. 2012;

42(6 S2):S184–90.

6 Beasley C. The Triple Aim. Healthc

Exec. 2009;24(1):64–6.

7 Martinez-Vidal E, Gauthier AK,

DiVincenzo A. Helping states en-

hance health care quality through

technical assistance. Health Aff

(Millwood). 2010;29(3):558–62.

8 Manuel DG, Lim J, Tanuseputro P,

Anderson GM, Alter DA, Laupacis A,

et al. Revisiting Rose: strategies for

reducing coronary heart disease.

BMJ. 2006;332(7542):659–62.

9 Quine S, Morrell S. Fear of loss of

independence and nursing home

admission in older Australians.

Health Soc Care Community.

2007;15(3):212–20.

10 Berenson RA, Paulus RA, Kalman

NS. Medicare’s readmissions-

reduction program—a positive al-

ternative. N Engl J Med. 2012;

366(15):1364–6.

11 Boulding W, Glickman SW, Manary

MP, Schulman KA, Staelin R.

Relationship between patient satis-

faction with inpatient care and hos-

pital readmission within 30 days. Am

J Manag Care. 2011;17(1):41–8.

12 Jencks SF, Williams MV, Coleman

EA. Rehospitalizations among pa-

tients in the Medicare fee-for-service

program. N Engl J Med. 2009;

360(14):1418–28.

13 Cooper BA, Branley P, Bulfone L,

Collins JF, Craig JC, Fraenkel MB,

et al. A randomized, controlled trial

of early versus late initiation of

dialysis. N Engl J Med. 2010;

363(7):609–19.

Quality

Governance

6 Health Affairs April 2013 32:4

14 Kutner NG, Zhang R, Barnhart H,

Collins AJ. Health status and quality

of life reported by incident patients

after 1 year on haemodialysis or

peritoneal dialysis. Nephrol Dial

Transplant. 2005;20(10):2159–67.

15 Berger A, Edelsberg J, Inglese GW,

Bhattacharyya SK, Oster G. Cost

comparison of peritoneal dialysis

versus hemodialysis in end-stage

renal disease. Am J Manag Care.

2009;15(8):509–18.

16 O’ Connor AM, Llewellyn-Thomas

HA, Flood AB. Modifying unwar-

ranted variations in health care:

shared decision making using pa-

tient decision aids. Health Aff

(Millwood). 2004;23:var-63–72.

DOI: 10.1377/hlthaff.var.63.

17 Luppa M, Luck T, Weyerer S, König

H-H, Brähler E, Riedel-Heller SG.

Prediction of institutionalization in

the elderly: a systematic review. Age

Ageing. 2010;39(1):31–8.

18 Frank RG. Long-term care financing

in the United States: sources and

institutions. Appl Econ Perspect

Policy. 2012;34(2):333–45.

19 Barnato AE, Farrell MH, Chang CC,

Lave JR, Roberts MS, Angus DC.

Development and validation of hos-

pital “end-of-life” treatment inten-

sity measures. Med Care. 2009;

47(10):1104.

20 Rose G. Sick individuals and sick

populations. Int J Epidemiol.

1985;14(1):32–8.

21 Rose G. High-risk and population

strategies of prevention: ethical

considerations. Ann Med. 1989;

21(6):409–13.

22 Ahern J, Jones MR, Bakshis E, Galea

S. Revisiting Rose: comparing the

benefits and costs of population-

wide and targeted interventions.

Milbank Q. 2008;86(4):581–600.

23 Zulman DM, Vijan S, Omenn GS,

Hayward RA. The relative merits of

population-based and targeted pre-

vention strategies. Milbank Q. 2008;

86(4):557–80.

24 Duncan I. Opportunity analysis.

Unpublished paper.

25 Cousins MS, Shickle LM, Bander JA.

An introduction to predictive mod-

eling for disease management risk

stratification. Dis Manag. 2002;

5(3):157–67.

26 Duncan I. Healthcare risk adjust-

ment and predictive modeling.

Winsted (CT): Actex Publications;

2011.

27 Wennberg D, Siegel M, Darin B,

Filipova N, Russell R, Kenney L, et al.

Combined predictive model: final

report [Internet]. London: King’s

Fund; 2006 Dec [cited 2013 Feb 28].

Available from: http://www

.kingsfund.org.uk/sites/files/kf/

field/field_document/PARR-

combined-predictive-model-final-

report-dec06.pdf

28 Vaithianathan R, Maloney T, Jiang

N, De Haan I, Dale C, Putnam-

Hornstein E, et al. Vulnerable chil-

dren: can administrative data be

used to identify children at risk of

adverse outcomes? [Internet].

Auckland: University of Auckland

Centre for Applied Research in

Economics; 2012 Sep [cited 2013

Mar 12]. Available from: http://

www.msd.govt.nz/documents/

about-msd-and-our-work/

publications-resources/research/

vulnerable-children/auckland-

university-can-administrative-data-

be-used-to-identify-children-at-risk-

of-adverse-outcome.pdf

29 Bardsley M, Billings J, Dixon J,

Georghiou T, Lewis GH, Steventon

A. Predicting who will use intensive

social care: case finding tools based

on linked health and social care data.

Age Ageing. 2011;40(2):265–70.

30 Van Walraven C, Dhalla IA, Bell C,

Etchells E, Stiell IG, Zarnke K, et al.

Derivation and validation of an in-

dex to predict early death or un-

planned readmission after discharge

from hospital to the community.

CMAJ. 2010;182(6):551–7.

31 Meek JA. Affordable Care Act: pre-

dictive modeling challenges and op-

portunities for case management.

Prof Case Manag. 2012;17(1):15–21.

32 Lewis GH. “Impactibility models”:

identifying the subgroup of high-risk

patients most amenable to hospital-

avoidance programs. Milbank Q.

2010;88(2):240–55.

33 Billings J, Dixon J, Mijanovich T,

Wennberg D. Case finding for pa-

tients at risk of readmission to hos-

pital: development of algorithm to

identify high risk patients. BMJ.

2006;333(7563):327.

34 Weber C, Neeser K. Using individu-

alized predictive disease modeling to

identify patients with the potential

to benefit from a disease manage-

ment program for diabetes mellitus.

Dis Manag. 2006;9(4):242–56.

35 Hart JT. The inverse care law .

Lancet. 1971;1(7696):405–12.

36 Billings J, Mijanovich T, Dixon J,

Curry N, Wennberg D, Darin B, et al.

Case findings algorithms for patients

at risk of re-hospitalisation: PARR 1

and PARR 2 [Internet]. London:

King’s Fund; [updated 2006 Feb 22;

cited 2013 Feb 28]. Available from:

http://www.kingsfund.org.uk/sites/

files/kf/field/field_document/

PARR-case-finding-algorithms-

feb06.pdf

37 National Health Service. Equality

and diversity in the NHS: what the

Equality Act 2010 means for you

[Internet]. London: NHS; 2012; [last

reviewed 2012 Sep 25; cited 2013

Feb 28]. Available from: http://www

.nhs.uk/NHSEngland/thenhs/

equality-and-diversity/Pages/

equality-and-diversity-in-the-NHS

.aspx

38 Andermann A, Blancquaert I,

Beauchamp S, Déry V. Revisiting

Wilson and Jungner in the genomic

age: a review of screening criteria

over the past 40 years. Bull World

Health Organ. 2008;86(4):317–9.

39 Wilson J, Jungner G. Principles and

practice of screening. Geneva: World

Health Organization; 1968.

40 Kass NE. An ethics framework for

public health. Am J Public Health.

2001;91(11):1776–82.

41 Smart A, Martin P, Parker M.

Tailored medicine: whom will it fit?

The ethics of patient and disease

stratification. Bioethics. 2004;18(4):

322–42.

42 Epstein L, Gofin J, Gofin R,

Neumark Y. The Jerusalem experi-

ence: three decades of service, re-

search, and training in community-

oriented primary care. Am J Public

Health. 2002;92(11):1717–21.

Martini EM, Garrett N, Lindquist T,

Isham GJ. The boomers are coming:

a total cost of care model of the im-

pact of population aging on health

care costs in the United States by

Major Practice Category. Health Serv

Res. 2007;42(1 Pt 1):201–18.

44 Feder JL. Predictive modeling and

team care for high-need patients at

HealthCare Partners. Health Aff

(Millwood). 2011;30(3):416–8.

45 Burns LR, Pauly MV. Accountable

care organizations may have diffi-

culty avoiding the failures of inte-

grated delivery networks of the

1990s. Health Aff (Millwood). 2012;

31(11):2407–16.

46 Farley TA, Dalal MA, Mostashari F,

Frieden TR. Deaths preventable in

the U.S. by improvements in use of

clinical preventive services. Am J

Prev Med. 2010;38(6):600–9.

April 2013 32:4 Health Affairs 7

ABOUT THE AUTHORS: GERAINT LEWIS, HEATHER KIRKHAM,

IAN DUNCAN

RHEMA VAITHIANATHAN

Geraint Lewis is

chief data officer

of the National

Health Service in

England.

In this month’s Health Affairs,

Geraint Lewis and coauthors

propose a new rubric for improving

health care: predicting and

preventing so-called Triple Fail

events, which simultaneously

refl ect poor pati ent expe ri enc e and

high costs wh ile constituting

inferior patient care—the opposite

of the Triple Aim. They note that

the risk of experiencing different

Triple Fail events varies widely

across individuals, and they argue

that by stratifying populations

according to each person’sriskand

anticipated response to a health

intervention, health systems could

more effectively target different

preventive services, such as case

management. The authors also

discuss how the approach of

stratifying the population and

tailoring interventions could be

planned and operationalized.

Lewis, now chief data officer of

the National Health Service in

England, was senior director of

clinical outcomes and analytics at

Walgreens. He is a fellow of the

Royal College of Physicians and a

fellow of the UK Faculty of Public

Health. Lewis holds a master’s

degree in public health from the

London School of Hygiene and

Tropical Medicine and a medical

degree from the University of

Cambridge.

Heather Kirkham is

amanagerinthe

Clinical Outcomes

and Analytics

Department at

Walgreens.

Heather Kirkham is a manager in

the Clinical Outcomes and

Analytics Department at Walgreens.

She is responsible for research

involving Walgreens’ health system

partners. Kirkham earned a

master’sdegreeinpublichealth

from the George Washington

University and a doctorate in

public health epidemiology from

Walden University.

Ian Duncan is the

vice presid ent in

the Clinical

Outcomes and

Analytics

Department at

Walgreens.

Ian Duncan is the vice president

in the Clinical Outcomes and

Analytics Department at Walgreens.

He is responsible for Walgreens’

outcomes, research and

publications, and custo m analytics.

He is also an adjunct professor of

actuarial statistics at the University

of California, Santa Barbara, and

an adjunct research professor of

health care administration at

Georgetown University. Duncan

founded Solucia Consulting, a

provider of analytical and

consulting services to the health

care financing industry. He holds a

degree in economics from Balliol

College, Oxford.

Rhema

Vaithianathan is a

senior research

fellow at Sim Ki

Boon Institute,

Singapore

Management

University.

Rhema Vaithianathan is a senior

resea rch fell ow at th e Sim Ki Boon

Institute, Singapore Management

University, and director of the

Centre for Applied Research in

Economics and an associate

professor of economics at the

University of Auckland, New

Zealand. She has a doctorate in

economics from the University of

Auckland.

Quality

Governance

8 Health Affairs April 2013 32:4