PHM Glossary: T

Targeted Population

Targeted population refers to a subset of the identified population who meet requirements for program qualifications and are targeted for wellness or care management program intervention.

See also population.


The use of telecommunications to deliver health care services and information that support patient care, administrative activities and health education. The technology is a means to improve access to care, while reducing cost of transportation and increasing convenience to patient care.

There is no widely accepted, single definition for telehealth. In general, telehealth is a term used to describe the delivery of health care from a distance or via the integration of several technologies. These may include information exchange, medical/health care delivery, telecommunication, and human-machine interface. The goal of telehealth is to facilitate the delivery of health care or promotion of health, including health education, disease prevention and management, clinical treatment and monitoring, data collection and analysis, and health care referrals.


The Centers for Medicare & Medicaid Services’ definition of telehealth includes the following: telehealth (or telemonitoring) is the use of telecommunications and information technology to provide access to health assessment, diagnosis, intervention, consultation, supervision and information across distance.

Telehealth includes such technologies as telephones, facsimile machines, electronic mail systems, and remote patient monitoring devices which are used to collect and transmit patient data for monitoring and interpretation. While they do not meet the Medicaid definition of telemedicine they are often considered under the broad umbrella of telehealth services. Even though such technologies are not considered "telemedicine," they may nevertheless be covered and reimbursed as part of a Medicaid coverable service under section 1905(a) of the Social Security Act such as laboratory service, x-ray service or physician services.


Telehealth applies telecommunication technologies to better exchange health information and provide health care services across geographic, time, and social/cultural barriers.

The definition of telehealth is closely related to telemedicine as it includes remote health care but it may not necessarily involve clinical services. Compared with telemedicine, telehealth is a relatively newer and broader term. Telehealth generally describes a service that is delivered or information that is exchanged in another time and/or place using any form of information technology. Bauer and Ringel state that:

Telemedicine is the combined use of telecommunications and computer technologies to improve the efficiency and effectiveness of health care services by liberating caregivers from traditional constraints of place and time and by empowering consumers to make informed choices in a competitive marketplace.

Per National Telecommunications and Information Administration’s Telemedicine Report to Congress, 1997:

The terms telemedicine and telehealth are often confused and used interchangeably. Telemedicine is the use of telecommunications and information technologies for the provision and support of clinical care to individuals at a distance and the transmission of information needed to provide that care. Telehealth includes clinical care but additionally encompasses the related areas of health professionals education, consumer health education, public health, research, and administration of health services.

Telehealth may incorporate videoconferencing, transmission of still images, e-health (including patient portals), remote monitoring of vital signs, continuing medical education, and nursing call centers. Some of these are also considered part of telemedicine. Examples of telehealth include:

  • Remote patient monitoring, which utilizes specialized devices or monitors to collect and transmit data to a centralized location for use in managing the patient—weight, clinical symptoms, blood glucose levels, and heart electrocardiogram outputs are examples of the types of information being transmitted by various devices and mechanisms in order to enable remote telehealth;
  • Education provided over the telephone by a case manager; and
  • Electronic communications with consumers and patients to transmit medical and health information.

Federal BIPA legislation in 2000 expanded Medicare coverage for telehealth beyond telemedicine demonstration projects:


Section 223 establishes revised payment provisions, effective no later than October 1, 2001, for services that are provided via a telecommunications system by a physician or practitioner to an eligible beneficiary in a rural area. The Secretary is required to make payments for telehealth services to the physician or practitioner at the distant site in an amount equal to the amount that would have been paid to such a physician or practitioner if the service had been furnished to the beneficiary without the use of a telecommunications system.

A facility fee will be paid to the originating site (allowable originating sites include: a physician/practitioner office, a critical access hospital, a rural health clinic, a federally qualified health center or a hospital). Section 223 requires the Secretary of Health and Human Services to conduct a study, and submit recommendations to Congress, that identify additional settings, sites, practitioners, and geographic areas that would be appropriate for telehealth services. Entities participating in federal demonstration projects approved by, or receiving funding from, the Secretary as of December 31, 2000, would be qualified sites.

Telehealth can be integrated into all aspects of population health management, as illustrated below:

Telehealth 1


Centers for Medicare & Medicaid Services.

A Glossary of Health Care Terms, 3rd ed. Boston: Gouston & Storrs (1997).

ACNP Home Page. American College of Nurse Practitioners. 23 Feb. 2006

American Nurses Association. “Telehealth: A Tool for Nursing Practice. In Nursing Trends & Issues.” ANA Policy Series. (1997).

American Nurses Association. “Telehealth: Issues for Nursing. In Nursing Trends & Issues, ANA Policy Series.” (1996).

Bauer, J. Not What the Doctor Ordered: How to End the Medical Monopoly in Pursuit of Managed Care. McGraw-Hill 1998.

Bauer, J., and Ringel, M. Telemedicine and the Reinvention of Healthcare: The Seventh Revolution in Medicine. New York: McGraw-Hill (1999).

Coiera, E. Guide to Medical Informatics, the Internet, and Telemedicine. London: Chapman and Hall Medical (1997).

CTEC Home Page. California Telemedicine and eHealth Center. 23 Feb. 2006

Grigsby, J. et al. “Analytic Framework for Evaluation of Telemedicine.” Telemedicine Journal, 1 (1995): 31-39.

Lindbergh, D. In testimony to the US House of Representatives, Committee on Science, Space, and Technology; Subcommittee on Oversight and Investigations, (May 2, 1994.) Medicare, Medicaid and SCHIP Benefits Improvement and Protection Act of 2000 (BIPA). Available as “Overview of Telemedicine.” ATA News and Resources. 2005. American Telemedicine Association. 23 Feb. 2006

Preston, J. The Telemedicine Handbook: Improving Health Care with Interactive Video. Austin: Telemedical Interactive Consultative Services (1993).

“Telemedicine.” 2006. American Academy of Family Physicians. 23 Feb. 2006

“Telemedicine Report to Congress.” 31 Jan. 1997. 23 Feb. 2006

U.S. Department of Commerce, National Telecommunications and Information Administration. “A Nation of Opportunity: Realizing the Promise of the Information Superhighway.” US Government Printing Office 1996.

World Health Association. A Health Telematics Policy: Report of WHO Group Consultation on Health Telematics.Geneva: (1997).


Telemedicine refers to the use of electronic communications and information technologies to provide clinical services when participants are at different locations. Closely associated with telemedicine is the term telehealth. Local state regulations may apply for reimbursement and scope for services.

Telemedicine involves the exchange of medical information from the location where a patient is treated or seen to another location by using electronic transmission of clinical variables, test results, etc., for the remote provider(s) to assist in the diagnosis and/or treatment of the patient’s medical conditions.


For purposes of Medicaid, telemedicine is the use of medical information exchanged from one site to another via electronic communications to improve a patient's health. Electronic communication means the use of interactive telecommunications equipment that includes, at a minimum, audio and video equipment permitting two-way, real time interactive communication between the patient, and the physician or practitioner at the distant site. Telemedicine is viewed as a cost-effective alternative to the more traditional face-to-face way of providing medical care (e.g., face-to-face consultations or examinations between provider and patient) that states may choose to cover. This definition is modeled on Medicare's definition of telehealth services located at 42 CFR 410.78. Note that the Federal Medicaid statute (Title XIX of the Social Security Act) does not recognize telemedicine as a distinct service.


The goal of telemedicine is to foster connectivity between providers and thereby improve efficiency in the delivery of care to improve the health of patients. It is typically associated with clinical care that is provided in something other than a face­-to-face situation. According to the American Telemedicine Association, telemedicine brings together various types of programs and providers to deliver medical services to patients. Generally, there is an assumption that the remote providers’ skills, resources, and/or proficiency surpass those of the local providers, thereby allowing better care to be delivered more efficiently.

Telemedicine is not a separate medical specialty, and reimbursement mechanisms rarely make a distinction made between services provided onsite and those provided through telemedicine. Typically, no separate codes are required for reimbursement of remote services. Telemedicine may involve:

  1. Remote consults with a specialist, including the transmission of diagnostic images and/or video along with patient data for the specialist to view and analyze at a later time.
  2. Remote, live patient consultations that rely on audio, video, and data exchange tools between a patient and a care provider (e.g., primary care or specialty physician) to make a diagnosis and develop a treatment plan.

Telemedicine can be thought of as a subset of telehealth. It extends to a variety of medical specialties (telepsychiatry) and subspecialties (teleradiology and telecardiology).


Arizona’s House Bill on Telemedicine (#2224) defines telemedicine as the practice of health care delivery, diagnosis, consultation, treatment, transfer of medical data, and education through interactive audio, video, or data communications.

The Texas House Bill (#2033) takes a more conservative view by defining telemedicine more narrowly as the remote practice of health care delivery.


Telehealth Legislation. 1997. 23 Feb. 2006

Telemedicine Program Home Page. Arizona. 16 Feb. 2006

Third party administrator (TPA)

A third party administrator (TPA) is an organization that processes insurance claims or certain aspects of employee benefit plans for a separate entity, for example self-insured employers, without assuming financial risk. The TPA often affiliates with a network administrator to provide clients with access to discounted contracts with hospitals, physicians, and other providers.


Nash DB, Reifsnyder J, Fabius RJ, Pracilio VP. Glossary of Population Health: Creating a Culture of Wellness. Sudbury, MA: Jones & Bartlett Learning; 2011.

Public Broadcasting Service. Healthcare Crisis: Who’s at Risk?

Threshold Model (Predictive Models)

A threshold model uses one or more criteria (e.g., demographic, clinical, functional, utilization, or economic)—alone or in combination—to identify individuals likely to benefit from an intervention. While straightforward and clinically or economically reasonable, threshold models are very prone to fail to accurately classify individuals because of selection bias.


All predictive models strive to accurately classify individuals as being appropriate for an intervention. Appropriateness in this context means achieving an undesired outcome—one that the chronic care (or case) management intervention hopes to avoid. Also called rules- or criterion-based models, threshold models use one or multiple criteria alone or in combination to attempt to predict who will have the undesired outcome. Such outcomes may be clinical (a heart attack or stroke), functional (loss of ability to function independently), utilization (emergency room visits or hospitalizations), or financial (having a health care bill over $10,000, in the upper 10% for the insured population, or for the population with a disease of interest).

For example, a heart failure program might want to find people who carry a diagnosis of heart failure and who are likely to be hospitalized in the next 6 months, or to otherwise clinically deteriorate. The program could use thresholds such as one hospital admission for heart failure in the past year, one emergency room visit for heart failure in the past 6 months, an ejection fraction of <35%, a history of myocardial infarction, and claims (covered charges) > $5,000 in the past 12 months. The program could invite anyone who meets one of these criteria (single threshold) into a chronic care management program or they might assign points to the criteria and assign those with 3 or more points to the highest-impact intervention, those with 1-2 points to the next level of intervention, and heart failure patients with no points to an intervention consisting of educational brochures and a self-help Web site.

Technically, a threshold model is not a predictive model because it is not a model that gives the amount or probability of an outcome as a dependent variable. However, it is possible to build predictive models using thresholds. For example, one could regress age, gender, ejection fraction, and whether the patient is taking an ACE inhibitor against the likelihood of being rehospitalized for heart failure in the next 12 months. Generally, threshold models are not used this way; rather they are used to identify candidates for interventions such as case management (all patients with emergency room visits or admissions for certain diseases in the past 3 months) or to offer a more intensive level of health screening (e.g., if the health risk assessment score is above 15, have the nurse contact the patient for a more detailed screening).


Threshold-based case-finding systems have the advantages of being easy to understand (the criteria are explicit, usually binary, and clinically or economically reasonable). They are easy to apply: someone is identified based on a single threshold, having a combination of thresholds, or using a point system.

However, threshold-based case-finding systems have two significant disadvantages. The first is generic to all expert (rules-based) systems. Some studies have shown that true predictive models based on statistical modeling (or neural net) of actual data are more accurate in predicting who will or will not achieve a given outcome. This leads to misclassification of individuals – consuming resources offering interventions to those who don’t need them and missing those who do.

The second disadvantage is due to threshold systems’ inherent selection bias. This is the tendency to find individuals who already have a problem (such as a high HbA1c, a recent hospitalization, or high costs). Such individuals are less likely to have the same type of problem in the next measurement period. This tendency for people with outlier measurements to have subsequent measurements closer to the average for the entire group is called regression to the mean. Suppose, for example, that an asthma management program identifies people to target for an intervention based on one of two criteria: (a) having an emergency room or hospital stay for asthma in the past 6 months, or (b) having >$5,000 in covered charges in the past year. Experience has shown that these outliers will tend to have lower costs and lower asthma-related utilization (as a group) in the next year, while in the next year, many high-cost or high-utilization asthmatics will come from those who were below the thresholds last year.

Compared with true predictive modeling (which is based on statistical and other methods of prediction that do not incorporate selection bias), threshold-based modeling might both find fewer people of interest (because they haven’t had the criterion event yet) and also find too many people who turn out not to be of interest because they regress toward the mean. Other methods of prediction, such as regression and neural networks, are free of selection bias and would be expected to have greater ability to accurately classify people by whether they ended up achieving the (undesirable) outcome of interest.

Nevertheless, threshold case-selection methods can be useful in reducing the likelihood of some undesirable outcomes, such as rehospitalization, for example, by contacting those hospitalized for a condition with a reasonably high likelihood of rehospitalization, and where hospitalization is frequently avoidable by understanding the factors leading to the hospitalization and eliminating or modifying them.


Cousins, M., Shickle, L., and Bander, J. “An introduction to Predictive Modeling for Disease Management Risk Stratification.” Disease Management (2002);5:157-167.

Transaction Systems

Transaction systems support granular business or clinical processes, typically involving one patient at a time. A transaction system provides specific views into the data, as opposed to disease registries and data warehouses which aggregate data for multiple patients and provide population-level views. Examples include patient registration, appointment scheduling, and visit documentation.

For example, point of care systems enable connectivity from within and between clinical systems to other technologies and transaction systems, including hardware, software and diagnostic devices such as electronic health records (EHRs), telemedicine interactions and testing data; often used in the care of individual patients, in real time, in the environment where care is provided such as home, office, and outpatient environments. Such systems provide the opportunity to embed clinical decision support tools into the care process.


Clinical and Laboratory Standards Institute (CLSI). (2006). Point-of-care connectivity; approved standard—second edition. Wayne, PA: CLSI.

Nash DB, Reifsnyder J, Fabius RJ, Pracilio VP. Glossary of Population Health: Creating a Culture of Wellness. Sudbury, MA: Jones & Bartlett Learning; 2011.

Transitions of Care (TOC)/Care Transitions

Transitions of carerefers to the movement of patients between health care locations, providers, or different levels of care within the same location as their conditions and care needs change.

Specifically, they can occur:

  • Within settings (e.g., primary care to specialty care, or intensive care unit (ICU) to ward);
  • Between settings (e.g., hospital to sub-acute care, or ambulatory clinic to senior center);
  • Across health states (e.g., curative care to palliative care or hospice, or personal residence to assisted living); or
  • Between providers (e.g., generalist to a specialist practitioner, or acute care provider to a palliative care specialist).

A recent position statement from the American Geriatrics Society defines transitional care as follows:

For the purpose of this position statement, transitional care is defined as a set of actions designed to ensure the coordination and continuity of health care as patients transfer between different locations or different levels of care within the same location. Representative locations include (but are not limited to) hospitals, sub-acute and post-acute nursing facilities, the patient's home, primary and specialty care offices, and long-term care facilities. Transitional care is based on a comprehensive plan of care and the availability of health care practitioners who are well-trained in chronic care and have current information about the patient's goals, preferences, and clinical status. It includes logistical arrangements, education of the patient and family, and coordination among the health professionals involved in the transition. Transitional care, which encompasses both the sending and the receiving aspects of the transfer, is essential for persons with complex care needs. In effect, transitions of care are a subpart of the broader concept of care coordination.

See also care coordination.


Coleman, Eric A. Commissioned Paper: Transitional Care Performance Measurement. Performance Measurement Report, Institute of Medicine, 2006. Appendix I, pp. 250-276.

National Transitions of Care Coalition ( and

The Care Transitions Program (

Trend (Health Care Cost Trend)

Trend is generally defined as the tendency to show movement toward something or in a particular direction. In health care, the trend is viewed from the actuarial perspective as the rate of increase in a health care measure per unit of exposure over a period, generally 1 year. Specific to health care, the health care cost Trend (identified with a capital “T” to differentiate from the general trend definition) is the year-over-year rate of increase in health care costs per member.  


The health care cost Trend is the term applied to the empirical observation that most health care measures (utilization, unit cost, per member per month costs, etc.) tend to change over time. Where Trend results in increases in health care measures, unit cost and per member per month Trends are influenced by many factors: underlying cost pressures, such as general inflation; increases in intensity of services; introduction of new and higher-cost procedures; actions taken as a result of cost-shifting by some payers; provider contract changes; leveraging of plan design, etc.

Additionally, utilization Trend is influenced by intensity of services, the propensity of demand for services to be affected by supply of providers, and changes in medical practice or regulation (for example, the change to more defensive medical practice resulting in increased testing, or regulations that mandate minimum maternity stays). In some circumstances Trend may also reflect the aging and worsening health status of a population, although when making comparisons between populations it is usual to try to allow for these factors directly (so that the Trend that applies should be that which applies to a stable population basis).


Health care Trend is used in some chronic care management savings calculations (for example, in the historical or actuarially adjusted historical control design). For the purpose of the actuarially-adjusted historical control design, the appropriate Trend to be used is that which applies to a stable population. The appropriate Trend to apply in these circumstances would be one that is neutral with respect to demographic and health status effects. Health care cost trends may vary from market to market, depending on the level of provider and health plan competition and the regional economy.

Before using an external estimate of Trend, it is important to understand how that Trend estimate was constructed and to what it applies. The examples (below) include an example of a basic Trend calculation, as well as examples of situations in which external Trend estimates need to be carefully understood, and that may require adjustment before being applied to a specific calculation.

Certain applications of the actuarially adjusted historical control methodology apply a single Trend to baseline costs. To correct for factors such as deductible leveraging or contract changes, one must make a refinement to the simple single Trend approach that decomposes the calculation into service categories and further decomposes Trend into its utilization, unit cost, and cost-sharing components. While this decomposition complicates a calculation, it allows for identification of anomalous Trends (for example, due to the effect of contracting or new technology) that may then be adjusted for by mutual agreement with the client.


The other use of Trend in chronic care management is in savings projections. For those situations in which projections of future savings are required (either dollar or percentage savings), the incorporation of a Trend assumption is necessary.

For correctly estimating future claims costs (or savings) from a claims base, very specific calculations are necessary. An example is provided below.

  • Baseline period: 10/1/01 to 9/30/02
  • Intervention period: 1/1/03 to 12/31/03
  • Annual Trend, baseline period: 5%
  • Baseline claims: $1,000
  • Baseline claims projected to intervention period: $1,000* 1.0515/12 = $1,063
  • Annual Trend factor: 1.05

The 15-month annual Trend factor moves the Trend experience forward from 10/1/02 – 9/30/03 to 1/1/03 through 12/31/03.


In this example, the basic Trend (Year 2 on Year 1) is: 10.64/10.06 or 5.7%.

  Year 1 Year 2
Month No. of Members Medical Claims Claims PMPM No. of Members Medical Claims Claims PMPM
1 10,000 $100,000 $10.00 10,300 $106,000 $10.29
2 10,200 $102,000 $10.01 10,000 $108,500 $10.85
3 9,800 $101,450 $10.35 9,875 $109,000 $11.04
4 9,950 $102,000 $10.25 9,700 $109,600 $11.30
5 10,100 $104,000 $10.30 $9,950 $108,500 $10.90
6 11,000 $105,000 $9.55 10,000 $107,500 $10.75
7 10,900 $106,000 $9.72 10,600 $110,000 $10.38
8 10,800 $108,000 $10.00 11,500 $117,500 $10.22
9 10,850 $109,500 $10.09 11,700 $120,000 $10.26
10 10,700 $110,000 $10.28 11,800 $125,000 $10.59
11 10,600 $108,500 $10.24 12,000 $128,500 $10.71
12 10,500 $105,600 $10.06 12,500 $132,000 $10.56
Total 125,400 $1,262,150 $10.06 129,925 $1,382,100 $10.64

Note: PMPM = per member per month.


Year 1 allowed charges: $1,000
Year 1 deductible: $100
Net paid claims: $900
Trend in allowed charge: 5%
Year 2 allowed charges:  $1,050
Year 2 deductible: $100
Net paid claims:  $950
Observed Trend in net paid claims: 5.6%

*In these examples, the effect of population changes such as an increase in average age or morbidity is assumed to be zero.


Assume that charges increase 3% for each year-of-age, and that there is no cost-sharing. Underlying Trend is 5%.

Year 1 charges are $1,000. Year 2 charges will be $1,082, reflecting the combined effect of aging and other Trend factors. Unless the change in the underlying population is explicitly allowed for in the Trend calculation, the observed Trend will be 8.2%.


Year 1 charges: $1,000
Year 1 deductible: $100
Net paid claims: $900
Year 2 charges: $1,082
Year 2 deductible: $100
Net paid claims:  $982
Observed Trend: 982/900 = 9.2%


“Behind the numbers: Medical cost trends for 2010. PriceWaterhouseCoopers’ Health Research Institute. 2010.

Cookson, J. and Reilly, P. “Medical Claim Cost Trend Analysis.” Group Insurance, Winsted, CT: 4th ed. Actex Publishers (2003).

Trigger Event

A trigger event is a clinical or other indicator that alerts a health care provider of a potential undesirable event or outcome. The use of "triggers," or clues, to identify adverse events (AEs) is an effective method for measuring the overall level of harm from medical care in a health care organization. A trigger event acts as a stimulus to initiate or precipitate a reaction or series of reactions. Trigger events can be thought of as red flags indicating outliers or gaps, including such trigger events as missed appointments, poor and/or unexpected outcomes from treatments or interventions, laboratory results that exceed certain thresholds or limits, and frequent use of health care services such as the emergency room.

Trigger events can be captured and reported for further analysis by use of a trigger tool. Trigger tools specific to adverse events such as medications, intense care delivery, and surgical intervention are available for use. Once a trigger event occurs, it must be evaluated, most often against clinical criteria, to determine what, if any, action is warranted.


Classen DC, Lloyd RC, Provost L, Griffin FA, Resar R. Development and evaluation of the Institute for Healthcare Improvement Global Trigger Tool. Journal of Patient Safety. 2008 Sep;4(3):169-177.