Investigation into the value of trained glycaemia alert dogs to clients with type I diabetes

Abstract

Previous studies have suggested that some pet dogs respond to their owners' hypoglycaemic state. Here, we show that trained glycaemia alert dogs placed with clients living with diabetes afford significant improvements to owner well-being. We investigated whether trained dogs reliably respond to their owners' hypoglycaemic state, and whether owners experience facilitated tightened glycaemic control, and wider psychosocial benefits. Since obtaining their dog, all seventeen clients studied reported positive effects including reduced paramedic call outs, decreased unconscious episodes and improved independence. Owner-recorded data showed that dogs alerted their owners, with significant, though variable, accuracy at times of low and high blood sugar. Eight out of the ten dogs (for which owners provided adequate records) responded consistently more often when their owner's blood sugars were reported to be outside, than within, target range. Comparison of nine clients' routine records showed significant overall change after obtaining their dogs, with seven clients recording a significantly higher proportion of routine tests within target range after obtaining a dog. HbA1C showed a small, non significant reduction after dog allocation. Based on owner-reported data we have shown, for the first time, that trained detection dogs perform above chance level. This study points to the potential value of alert dogs, for increasing glycaemic control, client independence and consequent quality of life and even reducing the costs of long-term health care.

Conflict of interest statement

Competing Interests: The authors can confirm that although the work was funded by a commercial source, The Company of Animals, this does not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials.

Figures

Figure 1. Odds ratios (Alert/Routine) for samples with glucose concentrations outside each client’s target range in alert samples and routine samples after the acquisition of their dog.

Horizontal lines indicate 95% confidence intervals and any that do not span 1 so are significant at p

Figure 2. Percentage of routine samples with glucose concentrations above and below each client’s target, taken before and after acquisition of a trained dog.

Clients presented in descending order of % low before acquisition and p values for test for significant change in proportion of readings below (left) and above target range (right).

Figure 3. Percentage of nights in which blood glucose readings were recorded to be below clients’ target range before and after dog acquisition.

Figure 4. Mean % HbA1c (glycated haemoglobin), from six-monthly medical records before and after acquisition of trained alert dogs.

Shaded bars represent means and white boxes, the 95% Confidence Intervals.