Concept: Measuring Continuity of Care (Continuity of Care Index)

Introduction

This concept defines and discusses the measure of continuity of care, describes the methods used at MCHP to operationalize this concept and provides links to MCHP research that describe and discuss the research findings related to continuity of care.

Defining Continuity of Care

Continuity of Care (COC) measures the extent to which an individual patient sees a given provider over a specified period of time. This involves the creation of a Continuity of Care Index (COCI) that identifies the number of providers providing service to a patient and the percentage of care provided by each provider. Possible scores range from just greater than 0 (where visits are made to different providers) to 1 (all visits made to the same provider). In MCHP research, the focus of this measure is on ambulatory visits to a provider. Continuity of Care can be measured for one physician, physician groups, primary care provider, or clinics, depending on the focus and scope of the research.

Continuity of care is a complex construct, but the basic idea is that a patient has an on-going relationship with a provider, a physician group practice, primary care provider, or a clinic. For "continuity" to exist, two criteria must be met:


1. a patient receives care from their provider(s);
   
2. the care continues over time.

"The fact that a patient returns to see the same physician repeatedly allows the physician to develop and implement a care plan over time. When a patient sees different physicians, this opportunity does not exist and a comprehensive care plan addressing a variety of issues is less likely to be implemented" (Katz et al., 2004).

General Approach to Measuring Continuity of Care

The general steps to developing the continuity of care measure involve:


1. allocate patients to an individual physician, physician group, primary care provider, or clinic. For more detailed information on how this can be done, please read the Patient Allocation: Assigning Patients to Physicians, Physician Groups or Clinics concept.
   
   
2. determine the exclusion level for Low Users. This identifies the minimum number of visits within a specific time frame for inclusion in your analyses. For more detailed information on this, please read the section titled Consideration of Low Users in the Patient Allocation: Assigning Patients to Physicians, Physician Groups or Clinics concept.
   
   
3. run the appropriate SAS code to calculate the continuity of care measures for your project. See the example SAS code below - Continuity of Care SAS Macro - (_concare) Original Version for more information.

The Continuity of Care Index (COCI)

The Continuity of Care Index (COCI) identifies the number of providers providing service to a patient and the percentage of care provided by each physician. The index is created for each patient and is calculated by taking the number of visits to each individual physician divided by the total number of visits the patient had overall. This index weights both the frequency of ambulatory visits to each physician and the dispersion of visits between physicians. Index values range from just greater than 0 (visits made to a number of different physicians) to 1 (all visits made to the same physician).

For more information on the COCI, please read the methodology section and discussion below under Frohlich et al. (2006).

MCHP Methods of Measuring Continuity of Care

Over time, MCHP has used different methods for measuring continuity of care. Some of these methods are described below.

1 - Roos et al. (1998)

Roos et al. (1998) compared three ways of measuring the continuity of ambulatory care that a patient receives over a one-year period:


1. SCC_R: usual provider continuity standardized. It is the proportion of visits made to the most frequently seen provider standardized to a mean of 0 and variance of 1.
   
2. COC_R: sensitive to changes in the total number of visits and in their distribution across different providers. It ranges from 0 to 1, where 0 occurs when each visit is to a different provider and 1 occurs when all visits are to one provider.
   
3. SECON_R: measures the sequential nature of provider continuity. It is the fraction of sequential visit pairs at which the same provider is seen. Like COC_R, it ranges from 0 to 1, but in contrast is dependent on the sequential order of visits. A patient who alternates between two providers will have a score of 0.

Normally, the 'provider' of the visit is the individual physician who appears on the medical claim; however, in this study, two situations were adjusted for:


1. The visit is a referral - in this case the provider was the referring physician.
   
2. MD practice - the physician group practice was considered one 'provider'.

The correlation matrix in Table 1 shows different versions of the continuity measures accounting for one or both of these conditions. After reviewing Steinwachs D (1979), Roos et. al. (1998) decided to use SECON_R as the measure of continuity because it is easier to interpret and it is sensitive to the sequence of visits. The matrix in Table 2 shows the strong correlation among all three measures of continuity in ambulatory care.

Calculating SCC / COC / SECON

For the following equations, let:


N = total number of visits
n _i = the number of visits to the 'i'th provider, where i = 1,2, ... M
M = the number of potentially available providers
si = 1, if same the provider is seen at sequential visits, otherwise si = 0.


Or SCC_R is UPC standardized to mean 0 and var 1



Where n _i is the number of visits to usual provider.

Notes:
• This process is part of the _concare SAS macro. It can be used to calculate all these measures. (see the _concare macro section in the concept titled MCHP SAS Macros for information).

NOTE: This section is based on a memo from David Friesen to Management and Programmers (April 3rd, 1997).

2 - Katz et al. (2004)

In the deliverable Using Administrative Data to Develop Indicators of Quality in Family Practice, Katz et al. (2004) developed a continuity of care measure that counted the number of ambulatory visits for a patient to each physician. In the case of ties, expenditures / payments to GPs were used to determine the most responsible physician. Each GP would have for every patient a ratio indicating a fraction of the visits dollars that Manitoba Health paid to this physician for the care of that particular patient.

"The continuity of care variable was developed to reflect the proportion of ambulatory care provided by primary care physicians to a patient by any one particular physician. The continuity of care score for each physician represents an average of the proportion of care (measured by the cost of real visits) that a physician provided to all the patients who accessed them for care compared to other physicians who provided care for those same patients. Possible scores range from just greater than 0 to 1; thus, a practitioner who was a patient's only primary care physician (and who provided care during the study year) was allocated a score of 1 for that patient. If a patient accessed two physicians for equal proportions of their care, each of those physicians were allocated a score of 0.5" (Katz et al., 2004).

In this study, continuity of care was used as an explanatory variable in the two models of disease prevention and care management. Table 7 in Katz et al. (2004) shows the distribution of physician practice characteristics in Winnipeg, Brandon and Non-Urban areas in Manitoba, including the Continuity of Care scores.

3 - Frohlich et al. (2006)

In the deliverable Profiling Primary Care Physician Practice in Manitoba, Frohlich et al. (2006) they investigated continuity of care as an indicator of quality care. The COC Index for a patient was calculated as the number of visits to the most frequent physician divided by the total number of visits that patient had overall.

For more information, please see section 3.3.1 Continuity of Care Index (COCI) in Frohlich et al. (2006).

4 - Brownell et al. (2007)

In the deliverable Next Steps in the Provincial Evaluation of the BabyFirst Program: Measuring Early Impacts on Outcomes Associated with Child Maltreatment, Brownell et al. (2007) calculated continuity of care as the proportion of visits made to the most frequent provider, with a provider defined either as a General Practitioner or Generalist Pediatrician. This was used to measure a families' connection with community resources.

For more information, please see Continuity/Lack of Continuity of Care in Brownell et al. (2007).

5 - Katz et al. (2013)

In the deliverable Understanding the Health System Use of Ambulatory Care Patients, Katz et al. (2013) investigated continuity of care for Manitobans with at least one of six chronic conditions. This group was chosen as the focus of the study because those with a chronic condition tend to use the healthcare system more frequently and they are more likely to benefit from continuity of care and high quality primary care services.

For more information, please read the discussion on Patterns of Care in Katz et al. (2013).

6 - Katz et al. (2014)

In the deliverable Physician Integrated Network: A Second Look, Katz et al. (2014) defined continuity of care (COC) as "... the extent to which an individual sees a particular physician over a specified period of time. It is typically defined as an ongoing relationship between a patient and a single physician outside of a specific incident of illness, and it is often seen as a core value of patient care in primary care medicine (Dreiher et al., 2012). COC encourages "improved communications, trust and a sense of continuous responsibility" (Dreiher et al., 2012). Individuals seeing the same primary care physician over time may have improved health outcomes as a result of having one person managing their healthcare. This indicator uses an index that weights both the frequency of ambulatory visits to each family physician and the dispersion of ambulatory visits between family physicians. The index values range from zero (each visit made to a different physician) to one (all visits made to a single physician). We present analyses by Physician Integrated Network (PIN) clinic rather than by individual physician. A value of zero represents all visits made to a different clinic and a value of one represents all visits made to the PIN clinic. We were not able to compare COC by clinic with shadow practices, because the shadow practices were constructed from a variety of different clinics. People who had fewer than three ambulatory visits to a family physician in a two-year period were excluded from this analysis."

For more information, please see Continuity of Care in Katz et al. (2014).

7 - Fransoo et al. (2019)

In The 2019 RHA Indicators Atlas by Fransoo et al. (2019), they measured the COCI for primary care providers (family physicians and nurse practitioners). In this research, residents with fewer than three ambulatory visits over a three-year period were excluded from analysis.

For more information, please see section 5.6 Continuity of Care Index in the online deliverable.

8 - Katz et al. (2019)

In the deliverable The Health Status of and Access to Healthcare by Registered First Nation Peoples in Manitoba, Katz et al. (2019) defined continuity of care as "The Continuity of Care Index is an indicator that weighs both the frequency of ambulatory visits to primary care providers (mdbloc=11 , 200) and the dispersion of ambulatory visits between family physicians and nurse practitioners. The possible index values range from just greater than zero (where all visits are made to different physicians or NPs) to one (all visits made to the same physician or NP)."

For more information, please see Continuity of Care in Katz et al. (2019) deliverable and Technical Definitions of Indicators and Drug Codes from online supplement material from Katz et al. (2019) deliverable.

Notes and Cautions

The term "Continuity of Care" has been used in many MCHP research projects. In several of these projects, the term "continuity of care" was really a measure of Majority of Care. For more information, please read the Measuring Majority of Care concept.

Important Considerations

There are some very important considerations when calculating continuity of care values. These include:


• What Types of Physicians Will Be Included?
  
• What Types of Physician Services Will Be Included?
  
• Identifying Unique Physicians
  
• Identifying Physician Groups / Clinics
  
• Consideration of Low Users

For more information on these considerations, please read the section titled General Approach and Consideration for Patient Allocation in the Patient Allocation: Assigning Patients to Physicians, Physician Groups or Clinics concept.