Identifying Indicators to Measure Recovery

As a highly individualized experience, recovery from addiction is difficult to conceptualize in a way that resonates with different people. While there are many definitions of recovery, and research identifying elements that define recovery, there is no validated tool for measuring recovery.


In order to properly create such a measure, Neale and colleagues developed a list of indicators of recovery that were informed by both addiction treatment professionals and current and former drug and alcohol treatment services users. The current study aims to identify which indicators are important for measuring recovery in the eyes of a diverse set of stakeholders working in the addiction field.



The initial stages of this research began by consulting with addiction psychiatrics, residential rehabilitation staff, and inpatient detoxification staff based in the United Kingdom to generate a list of indicators of recovery (see here). The 76 indicators from this study were then discussed in a series of focus groups with current and former drug and alcohol treatment service users to identify which indicators were inappropriate or contradictory and should be removed or revised (see here). A revised list was then further tested and revised among service users, resulting in the list of 28 indicators used in the current study.

This study used the Delphi method, a technique where an anonymous panel is asked a set of questions, and aggregated responses are shared and then used to inform responses for the next round (and so on for a predetermined number of rounds). The ultimate goal is to arrive at a “correct” answer or consensus. Authors did not seek a complete consensus in this study, however, as they were interested in determining agreement or disagreement with the list of indicators. This study recruited people whose job involves working with (e.g., treating) individuals who have problems with drugs or alcohol. Participants completed a three-round online Delphi group. One hundred forty six people registered and completed at least one of the three rounds. Twenty-six percent (n = 38) were personally in recovery.


The graph describes the jobs or roles of participants.

Participants were asked to rate each of the 28 indicators (below) on their importance to the recovery concept using a scale of 1 to 10 with 10 indicating highest level of importance.

Recovery Indicators:


  • Not drinking too much
  • Not using street drugs
  • Not experiencing cravings
  • Taking care of mental health
  • Coping with problems without turning to drugs/alcohol
  • Feeling emotionally stable and secure
  • Feeling like a worthwhile person
  • Taking care of physical health
  • Managing pains and ill-health without misusing drugs/alcohol
  • Taking care of appearance
  • Eating a good diet
  • Sleeping well
  • Getting on well with people
  • Feeling supported by people
  • Having stable housing
  • Having a regular income (from benefits, work, other legal sources)
  • Managing money well
  • Having a good daily routine
  • Going to appointments
  • Spending free time on hobbies/interest that do not involve drugs/alcohol
  • Participation in education, training or work (paid or voluntary)
  • Feeling happy with overall quality of life
  • Feeling positive
  • Having realistic hopes and goals for oneself
  • Being treated with respect and consideration by other people
  • Treating others with respect and consideration
  • Being honest and law-abiding
  • Trying to help and support other people


The authors’ hypothesis that scores of each of the indicators would be high in the first round was correct; median scores (i.e., those in the middle of the group of scores) for all indicators ranged from 7 to 10 and average scores ranged from 6.15 for “not experiencing cravings” to 9.43 for “coping with problems without turning to drugs/alcohol”. Each indicator had a range of scores from participants. For example, the range for “coping with problems without turning to drugs/alcohol” was 5 since the minimum score from any respondent was 5 and the maximum score given was 10.

Following the underlying philosophy of the Delphi method, the authors hypothesized that providing stakeholders with the average score for each indicator in Round 1 would result in a greater consensus in Round 2. This was confirmed by a narrowing total range score from 191 points to 184 points. However, the total range score increased in Round 3, indicating that complete consensus was not achieved.

One hypothesis from Round 3 was that all score indicators would have an average score greater than or equal to 7 (suggesting high importance for all indicators). This was true for 27 of the 28 indicators but the mean score for “not experience cravings” remained below 7. The authors also hypothesized that scores would not vary significantly by characteristics of the stakeholders. However, there were some statistically significant differences with higher scores on some indicators for those who had worked in addiction longer, non-practitioners, and people who were in recovery themselves compared to people working for less time, practitioners, and people not in recovery, respectively.

For example people in recovery rated “not using street drugs”, “coping with problems without turning to drugs/alcohol”, “treating others with respect and consideration” and “being honest and law-abiding” as more important than people who were not in recovery. However, indicators generally earned a 7 or higher in terms of importance to the recovery concept despite differences between these groups.


This study shows that through an extensive process with several focus groups and revisions with individuals in a variety of roles (e.g., providers and patients), it may be possible to create a measure that captures the multi-dimensional nature of recovery.

One of the highlights of the development process described in this study is that people in recovery contributed to its development. It is likely that developing a measure of recovery without input from individuals in recovery themselves would produce a very different outcome that may not adequately measure the concept.

Given the varieties of recovery experiences, it may be impossible to reach a complete consensus on a list of recovery indicators. Instead, a working consensus was achieved such that 27 of the 28 indicators were consistently ranked as important for most of people most of the time by a diverse group of stakeholders from the addiction field. Thus, these 27 behaviors may be a method to conceptualize and capture recovery in a way that represents varied perspectives in the field of addiction treatment.


Since there is currently no validated measure specifically for recovery,  studies tend to rely on scales of addiction such as the Addiction Severity Index, quality of life scales such as the WHOQOL to measure positive changes that may be indicative of recovery, and the Assessment of Recovery Capital scale for assessing progress in recovery.


Tthere is still a need for a measure that captures the multi-dimensional nature of recovery and incorporates the viewpoints of the people who experience this process and stakeholders working in the field. This study sought to broaden the theory of addiction recovery and represents a critical step toward developing a self-report questionnaire to measure it.

The indicators used in this study included items that were directly and indirectly related to drug/alcohol use in addition to indicators related to health and wellbeing. The indicator “not experiencing cravings” consistently received the lowers scores. This may be due to the fact that the actual experience of having a craving cannot be prevented and it is more important for a person in recovery to know how to control or manage these cravings than to avoid them altogether. People in recovery had systematically different viewpoints on some domains such as “not using street drugs” and “coping with problems without turning to drugs/alcohol” which shows the value of including people in recovery when trying to capture this complex concept.

  1. This study was conducted among people from the United Kingdom, and what recovery means to people in one setting could vary by location.
  2. Formal validation of this measure is also needed to determine if these indicators adequately capture the underlying concept of recovery.


Next steps include psychometric testing to determine if this measure is valid and reliable. This study may also be repeated in other locations and among different stakeholders such as politicians or the general public to see if these results are reproducible.


  • For individuals & families seeking recovery: Recovery is difficult to measure and may have different meanings for different people. This study showed, however, that behaviors such as coping with problems without turning to drugs/alcohol are likely to be part of how individuals in the UK define recovery.
  • For scientists: The authors have used a rigorous and extensive series of qualitative studies to develop and refine their measure of recovery. Once finalized, psychometric testing is needed to determine its validity and reliability in clinical settings.
  • For policy makers: This study showed that stakeholders in the addiction field including policy makers agreed that a majority of the recovery indicators were important for most people in recovery. Once it is finalized, this measure could be one tool to evaluate the benefit individuals derive from recovery support services such as recovery community centers.
  • For treatment professionals and treatment systems: This study showed that stakeholders in the addiction field including clinical practitioners agreed that a majority of the recovery indicators were important for most people in recovery. Once it is finalized, this measure could be one tool to evaluate treatment progress.


Neale, J., Panebianco, D., Finch, E., Marsden, J., Mitcheson, L., Rose, D., . . . Wykes, T. (2015). Emerging consensus on measuring addiction recovery: Findings from a multi-stakeholder consultation exercise. Drugs: Education, Prevention and Policy, 1-10. doi:10.3109/09687637.2015.1100587