How Should We Measure Retention in HIV Care?

Thomas P. Giordano, MD, MPH

Disclosures

August 15, 2012

Introduction

Retention in HIV care is a critical determinant of clinical outcomes, including viral suppression and survival.[1,2,3] Although there are many ways to measure retention in care, there is currently no gold-standard measure, and the commonly used measures differ in several characteristics.[4,5]

With the increased attention on retention in care, clinicians, administrators, and policymakers need to understand the nature of each measure as well as its strengths and limitations. This review will describe the commonly used measures of retention, provide an overview of their characteristics and limitations, and highlight various situations in which they might be useful.

Distinguishing Between Adherence and Constancy Measures

There are generally 2 types of measures of retention in care: adherence measures and constancy measures. Adherence measures focus on whether scheduled visits were in fact attended or missed. They might simply count missed visits over a time period, or calculate the percentage of scheduled visits that were attended.

Constancy measures rely on attended visits only, ignoring missed visits, and consider whether a patient attended a visit during a certain time period -- for example, whether the patient attended at least 1 visit every 6 months in the past year. Gap measures are constancy measures that count the number of days between 2 contiguous visits and determine the length of the longest gap in calculations, or determine whether that gap exceeds a certain threshold (eg, 6 months).

Adherence measures have the advantage of being conceptually similar to measures of adherence to medications. They also might better take into account the number of times a person should have been seen over the measurement period, because a person with more advanced disease or with comorbidities would be scheduled more visits than would a healthier person. On the other hand, adherence measures are less meaningful for persons who are so poorly retained in care that they have no or few scheduled visits to miss.

Constancy measures have the advantage of not requiring data on missed visits, which are not always available, and the measures can be calculated for patients who have no scheduled visits. However, they set minimum standards of retention and might not be sensitive to disease severity. For example, the minimum standard of 1 visit in each 6-month period might not be sufficient for persons with unsuppressed HIV.

Commonly used adherence and constancy measures are briefly described in the Table.

Table. Commonly Used Adherence and Constancy Measures

Measure Type of Output;
Calculation Overview
Are
Missed
Visit Data
Needed?
Ease of
Calculating
Observation Time
Needed To
Calculate
Cautionsa
Missed visit Dichotomous; were there any missed visits in the interval?
Count; number of missed visits in the interval
Yes Easy At least 6 months If no visits are scheduled, will be falsely low; if rescheduling is automatic, will be falsely high
Appointment adherence Continuous; attended appointments divided by (attended appointments plus missed appointments) Yes Moderate Patient: at least 1 year; clinic: as short as 1 day If no visits are scheduled, will be falsely high; if rescheduling is automatic, will be falsely low
No-show rate Continuous; 1 minus appointment adherence Yes Moderate Patient: at least 1 year; clinic: as short as 1 day If no visits are scheduled, will be falsely high; if rescheduling is automatic, will be falsely low
Constancy, 3- or 4-month intervals Categorical; number of 3- or 4-month intervals with at least 1 attended visit No Moderate At least 6-8 months Will underestimate retention for patients needing less frequent visits
Constancy, 6-month intervals Categorical; Number of 6-month intervals with at least 1 attended visit No Moderate At least 1 year Will overestimate retention for patients needing more frequent visits
Constancy, 6-month intervals, longer-term Dichotomous; At least 1 attended visit in each 6-month interval with at least 60 days between visits No Moderate to difficult At least 2 years Will overestimate retention for patients needing more frequent visits
Constancy, HAB Dichotomous; At least 2 attended visits in 12 months, separated by at least 90 days No Moderate to difficult At least 1 year Will overestimate retention for patients needing more frequent visits
Gaps Dichotomous; did the time between 2 contiguous attended visits exceed a threshold (eg, 6 months)? No Dichotomous: easy At least 1 year Difficult to calculate gap when there is no recent attended visit; continuous data will not be normally distributed
Continuous; what is the longest duration of time between 2 contiguous attended visits? Continuous: moderate

HAB = HIV/AIDS Bureau of the Health Resources and Services Administration
aAll measures have the potential to be misinterpreted if patients have unknowingly transferred care elsewhere, been incarcerated, or died.

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....