Can Service Dogs Improve Activity and Quality of Life in Veterans With PTSD?: Complete Statistical Analysis

1. Trial at a Glance

This page covers the registry results for Can Service Dogs Improve Activity and Quality of Life in Veterans With PTSD? (NCT02039843), a randomized na study. The primary outcome is World Health Organization Disability Assessment Scale II (WHO-DAS 2.0) Total Score.

2. Clinical Question

The WHO-DAS 2.0 assessment was scheduled at baseline and every 3 months after dog pairing up to 18 months post-pairing. The registry record describes the outcome as a mean score with standard deviation. It does not report a single primary comparison across the follow-up period, so the direction and precision of a service-dog versus emotional-support-dog difference cannot be stated from the record.

3. Trial Design

This was a randomized, parallel-group trial with no masking. Participants were paired with either a Service Dog or an Emotional Support Dog. The registry record lists enrollment of 287 and identifies the study as completed. The sponsor was the VA Office of Research and Development. The recorded study dates are 2014-12-15 for the start and 2019-06-30 for primary completion. The phase is listed as NA.

The two dog groups differed in their stated training and qualification requirements. Emotional Support Dogs were required to pass the American Kennel Club Canine Good Citizen and Community Canine tests. Service Dogs were required to pass the American Kennel Club Canine Good Citizen and Assistance Dogs International Public Access tests, and were trained for five PTSD-specific tasks. These descriptions establish the intended distinction between the groups; they do not establish how consistently each dog met those requirements during the trial.

The main outcome was the WHO-DAS 2.0 total score, collected at baseline and every 3 months after pairing up to 18 months. The registry record says that the number analyzed at each follow-up reflects participants still participating and completing the assessment at that time. It does not report the randomization method, allocation concealment, masking of outcome assessment, or detailed analysis plan. It also does not report the reasons participants may have stopped participating or missed an assessment.

Other registered outcomes include sleep quality, PTSD symptoms, depression, anger, suicidal behavior or ideation, and use of non-VA health services. The record lists 22 outcome measures and 13 analyses. Those counts do not reveal the order of testing or which comparisons were treated as confirmatory.

4. Participant Flow and Accounting

The registry reports how participants moved through each stage of the study. All counts below are exactly as posted; missing counts are not inferred from other stages.

Before assignment: Sixty participants of 287 consented participants were found to be ineligible for study randomization per study inclusion/exclusion criteria, resulting in 227 participants randomized.

Randomization and Observation Period

Reasons for not completing this period:

Pairing With a Study Dog

Reasons for not completing this period:

Analysis population for the primary outcome, as defined in the registry: Participants randomized and paired with an emotional support dog or a service dog. The number of participants analyzed at each 3-month follow-up time point represents the number of individuals still participating in the study at that time point and completing the assessment. As this was a longitudinal study, attrition was expected and seen over time. Reasons for attrition are noted in the Participant Flow section under the 'Pairing With a Study Dog' Period.

5. Baseline Characteristics

The table reproduces the baseline characteristics posted in the registry, which describe the participants analysed at baseline and help show whether the randomized groups were comparable.

Population: Randomized participants paired with a study dog.

6. Primary Endpoint

The registry lists the primary outcome as World Health Organization Disability Assessment Scale II (WHO-DAS 2.0) Total Score, measured over Baseline and every 3 months after dog pairing up to 18 months post-pairing.

The primary endpoint is the World Health Organization Disability Assessment Scale II (WHO-DAS 2.0) total score. It is a standardized measure intended to assess health and disability. The registered version is a 36-item, self-administered questionnaire. Participants rate the difficulty experienced with each item using the response options None, Mild, Moderate, Severe, or Extreme/cannot do. The resulting standardized total score ranges from 0, meaning no disability, to 100, meaning full disability.

The endpoint is relevant to the study’s stated focus on activity and quality of life because it measures reported difficulty across areas of functioning. Its total score offers a continuous summary rather than a count of events. The registry record specifies that the measure is summarized as a mean and standard deviation. In general, a mean describes the average score in a group, while a standard deviation describes how spread out individual scores are around that average. Neither summary alone describes every participant’s experience.

The assessment schedule was baseline and every 3 months after dog pairing up to 18 months post-pairing. Repeated scores can describe how disability changes over time, but interpretation depends on the comparison being made, the participants contributing at each time point, and the statistical approach. The registry record does not identify a single time-point contrast or provide a primary-outcome estimate with its confidence interval. Therefore, it does not support a conclusion about the size or precision of a difference between the dog groups on this endpoint.

7. Why No Re-analysis Is Shown

8. Statistical Interpretation

9. Statistical Methods Explained

What does the primary measure represent?

The WHO-DAS 2.0 total score is a continuous outcome: it can take values across a scale rather than being limited to categories such as yes or no. The registered score ranges from 0, no disability, to 100, full disability. In general, continuous outcomes are often described by a group mean and standard deviation. The mean gives a central summary, and the standard deviation describes variability among individual scores. A group mean does not mean that every participant had that score, and a standard deviation is not an uncertainty interval for the mean.

The registry record names the mean as the outcome parameter and standard deviation as the dispersion measure. It does not provide the group means or standard deviations at the scheduled assessments. Nor does it report an estimated difference between groups. Without those values, the scale’s definition can be explained, but the trial’s observed score pattern cannot be quantified.

Why does follow-up structure matter?

The WHO-DAS 2.0 was scheduled at baseline and every 3 months after pairing up to 18 months. This makes the outcome longitudinal: the same participants may contribute scores at more than one assessment. In general, repeated measurements can reveal whether a pattern is sustained, changes over time, or differs between groups only at particular assessments. They also require care because measurements from the same participant are related rather than independent.

There are several possible summaries of a longitudinal comparison. A selected follow-up contrast compares the groups at that assessment. A change-from-baseline contrast compares the extent of change in each group. A model of repeated scores can estimate a broader pattern while accounting for the relationship among measurements from the same participant. These approaches answer related but not identical questions. The registry record does not report which approach was used for the primary analysis, the model specification, or the primary contrast.

The registry record says the number analyzed at each follow-up represents people still participating and completing the assessment at that time. In general, follow-up counts can change when participants discontinue, miss assessments, or otherwise lack a score. The record does not state why scores are absent, how missing observations were handled, or whether the analysis used all available measurements or a restricted set of completed assessments. Those details matter because an analysis based only on observed scores can be affected if missingness is related to participants’ health or outcomes.

How should a group difference and confidence interval be read?

In general, a group difference expresses how far apart the groups’ outcome summaries are on the scale being analyzed. Its interpretation depends on the direction of subtraction and on whether the estimate concerns a follow-up score, change from baseline, or a longitudinal pattern. The registry record does not give the primary estimate or define a direction for such a contrast, so no numerical interpretation can be made.

In general, a confidence interval describes the statistical uncertainty around an estimated effect under the analysis model. A narrower interval indicates greater precision than a wider interval, while the interval’s location describes which effect values are compatible with the estimate and method. A confidence interval is not a range containing a specified proportion of individual participants’ scores. No primary-outcome confidence interval is reported in the registry record.

What can randomization and lack of masking tell us?

Random allocation is intended, in general, to make treatment groups comparable in expectation at the point of assignment. It does not guarantee that every characteristic is balanced in a particular trial, and it does not resolve problems caused by incomplete follow-up or outcome measurement. The registry record identifies the trial as randomized but does not describe the sequence-generation or allocation-concealment procedures.

The record states that masking was none. In general, knowledge of group assignment can influence participants’ reports, expectations, behavior, or assessment. That possibility is relevant to a self-administered measure such as the WHO-DAS 2.0, but the record does not establish whether or how such influence occurred. It also does not report whether outcome assessment or data analysis was masked.

What is not available about the statistical plan?

The registry record lists 22 outcome measures and 13 analyses, but those counts do not specify the analysis hierarchy. The record does not report the statistical model, adjustment variables, test sidedness for the primary result, multiplicity procedure, or sample-size justification. It also does not supply a primary-outcome result to evaluate. These omissions limit the ability to assess how the posted results were generated and how uncertainty was handled.

In general, a statistical analysis plan clarifies which outcomes and comparisons are primary, how repeated observations and missing data will be treated, and whether tests are adjusted when many outcomes are examined. Without those details, it is not possible to determine from the registry record whether the analysis followed a prespecified approach or how the reported outcome findings should be weighed against multiple testing.

10. Multiplicity, Missing Data and Analysis Populations

The registry record lists 22 outcome measures and 13 analyses. The primary outcome is assessed repeatedly, at baseline and every 3 months after dog pairing up to 18 months. Considering many outcomes, time points, or contrasts can increase the chance of observing an apparently notable result by chance. In general, multiplicity procedures or a prespecified hierarchy can help control how such results are interpreted. The registry record does not report a multiplicity adjustment, an outcome hierarchy, or which comparisons were designated as confirmatory.

The primary outcome has several time points and is not reduced to a single comparison in the record. The record does not identify a primary follow-up contrast, report an overall longitudinal test, or provide the primary outcome estimates and uncertainty. The 13 listed analyses cannot by themselves establish which findings were planned or how analyses were selected.

Missing-data methods are also not reported. The registry description says that the number analyzed at each follow-up reflects participants still participating and completing the assessment at that time. It does not explain why assessments may be unavailable, whether partial questionnaires were used, or how missing scores entered the analysis. In general, missingness can affect interpretation when the participants with observed outcomes differ from those without them. The record does not report a sensitivity analysis to examine that possibility.

The analysis population is not described in enough detail to identify whether the primary analysis followed randomized assignment, included only participants who were paired with a dog, or used another population definition. The registry record also does not report a prespecified analysis plan or sample-size justification. These gaps limit assessment of multiplicity, missing-data assumptions, and the relationship between the planned and reported analyses.

11. Safety Results

The registry reports serious adverse events as affected participants divided by participants at risk.

Deaths

Other (non-serious) adverse events

Participants with at least one non-serious event at or above the registry's 0% reporting threshold.

Most frequent serious adverse event terms

Most frequent non-serious adverse event terms

Serious events with the highest arm-level rate: suicidal ideation (6.2%), transient ischaemic attack (3.1%), coronary artery disease (2.4%), chest pain (2.4%), pneumonia (2.1%).

The registry record reports serious adverse events affecting 21 of 84 participants in the Emotional Support Dog group and 18 of 97 participants in the Service Dog group. For other adverse events, it reports 5 of 84 in the Emotional Support Dog group and none of 97 in the Service Dog group. It lists one death in the Emotional Support Dog group and none in the Service Dog group. These figures are descriptive counts from the registry record; they do not establish that a dog assignment caused an event or that the groups differed statistically.

The record also reports a separate “Prior to Dog Pairing” category, listing serious adverse events affecting 18 of 287 and other adverse events affecting one of 287, with one death. This is a different period from the post-pairing groups. It should not be treated as a directly comparable treatment group or combined with the post-pairing counts.

The registry lists serious-event terms including suicidal ideation (6.2%), transient ischaemic attack (3.1%), coronary artery disease (2.4%), chest pain (2.4%), pneumonia (2.1%), and dehydration (2.1%). These are reported percentages for listed terms; the record does not provide enough detail to interpret them as comparative risks or attribute them to a study group. Event timing, duration of exposure, attribution, and statistical comparisons are not reported in the registry record.

12. What the Result Does, and Does Not, Mean

The trial addresses a clinically important and specific question: whether disability scores differ over time for Veterans with PTSD paired with Service Dogs compared with those paired with Emotional Support Dogs. Randomization is relevant to that comparison because it is intended to reduce systematic differences in group assignment. The registry also defines the dogs’ qualification and training requirements and specifies a repeated assessment schedule using the WHO-DAS 2.0.

However, the registry record does not provide a single primary-outcome comparison or the estimate, confidence interval, and test needed to interpret the observed difference. It therefore does not show, in the information available for this analysis, whether the groups’ disability scores differed, how large any difference was, or how precisely it was estimated. The absence of a reported comparison is not evidence that the groups were equal, and it is not evidence that a difference was present.

Even when a difference is reported, statistical interpretation requires attention to its scale, uncertainty, timing, and clinical meaning. A difference at one follow-up can tell a different story from a pattern across repeated assessments. In general, a confidence interval helps show the precision of an estimate, while the score’s scale helps explain its direction. Neither is available for the primary comparison in the registry record.

The registered secondary outcomes broaden the assessment to sleep, PTSD symptoms, depression, anger, suicidal behavior or ideation, and non-VA healthcare use. They are separate outcomes and should not be treated as interchangeable with the disability score. The record’s listing of these outcomes does not establish their results, whether all were analyzed, or whether any were adjusted for multiple testing.

The adverse-event information describes counts and selected serious-event terms, but it does not provide causal attribution or a statistical comparison. Those figures should be read as safety reporting, not as proof that an intervention caused or prevented an event. Overall, the registry record establishes the design and measurement framework, while leaving the primary efficacy comparison and much of the statistical interpretation not reported in the registry record.

13. Why This Trial Matters Statistically

This trial is statistically notable for comparing two dog-pairing groups rather than comparing a dog intervention with no dog. The distinction matters: the study’s question concerns the relative outcomes of pairing with a Service Dog versus pairing with an Emotional Support Dog. The registry descriptions specify different training and qualification requirements for the two groups, so the comparison is about those defined group assignments, not an isolated training task or a dog-versus-no-dog contrast.

The primary measure is repeated over time. Longitudinal outcomes can describe whether group differences emerge, persist, or change during follow-up, but the interpretation depends on the chosen contrast and statistical model. The record gives an assessment schedule from baseline through 18 months after pairing, yet does not identify one primary comparison or report its estimate. This makes the distinction between a rich measurement schedule and a clearly interpretable primary result especially important.

The outcome is also self-administered, and the record states that masking was none. In general, participant-reported measures can capture experiences that may not be represented by service-use counts, while awareness of group assignment can also influence reporting. These are considerations for interpretation, not evidence that reporting bias occurred in this trial.

Finally, the registry lists numerous outcomes and analyses, which makes transparency about outcome hierarchy, multiplicity, and missing data important. The record does not report those statistical details. For clinical researchers and biostatisticians, the case illustrates why trial registration and outcome definitions are useful but do not replace complete reporting of estimates, uncertainty, analysis populations, and methods.

14. Limitations

• The posted summary statistics are the only data these results rest on; patient-level data, full protocols and statistical analysis plans are not available.
• Recalculated results are simple, unadjusted calculations from posted summaries. They do not replace the sponsor's prespecified analysis.

The primary limitation for interpreting this record is that it does not report a single primary-outcome comparison for the WHO-DAS 2.0. Group means, a between-group estimate, a confidence interval, and a primary significance test are not reported in the registry record. The outcome is measured at multiple time points, but the record does not specify which contrast was primary. This prevents assessment of the direction, magnitude, and precision of the primary finding.

The statistical methods are incompletely described. The registry record does not report the primary model, adjustment variables, multiplicity approach, missing-data method, or detailed analysis population. Although it describes the number analyzed at each follow-up as reflecting participants still participating and completing assessments, it does not give reasons for missing scores or explain how those scores were handled. These omissions limit evaluation of bias and reproducibility.

The record identifies randomization and no masking but does not describe sequence generation, allocation concealment, or masking of outcome assessment and analysis. The primary measure is self-administered, so lack of masking is relevant as a possible source of influence; the record does not show that such influence occurred.

Adverse-event reporting is descriptive and includes a pre-pairing category separate from the post-pairing groups. The registry record does not provide event attribution, comparative testing, or sufficient detail to interpret the listed percentages as treatment-related risks. The record also does not report whether the stated dog requirements were met consistently over follow-up. These limitations concern what can be concluded from the registry record; they do not establish that the trial itself lacked additional analyses or documentation.

15. Related Tutorials

16. Related Statistical Calculators

17. Sources

• ClinicalTrials.gov: Can Service Dogs Improve Activity and Quality of Life in Veterans With PTSD? — NCT02039843.
• Registry record last updated 2023-12-01; status last verified 2023-02.
• Results first posted 2023-12-01.
• Data retrieved from ClinicalTrials.gov on 2026-11-09.
• Group sizes, means, event counts and sponsor-reported p-values are reproduced as posted. Effect estimates, confidence intervals and p-values not attributed to the sponsor are independently calculated from the posted summary statistics.

18. Record Summary

NCT02039843, titled “Can Service Dogs Improve Activity and Quality of Life in Veterans With PTSD?”, is a completed randomized, parallel-group trial with no masking. The registry record lists enrollment of 287 and identifies the sponsor as the VA Office of Research and Development. Participants were paired with a Service Dog or an Emotional Support Dog.

The primary outcome was WHO-DAS 2.0 total score, assessed at baseline and every 3 months after pairing up to 18 months. The measure ranges from 0, no disability, to 100, full disability, and is summarized as a mean with standard deviation. The registry record lists additional outcomes covering sleep, PTSD symptoms, depression, anger, suicidal behavior or ideation, and non-VA healthcare use.

The record does not report a single primary-outcome comparison, its estimate, or its confidence interval. It also does not report the primary statistical model, missing-data approach, or multiplicity procedure. Serious adverse-event counts and selected terms are listed, but they do not establish cause or a comparative treatment effect.