This month the Journal of Medical Internet Research published a study of users of Patients Like Me, a patient-directed networking site supporting people with conditions ranging from HIV to fibromyalgia to Parkinsons and multiple sclerosis.
As I read it I paid particular attention to bias, since it comes from the website owners. The authors did a respectable job describing study limitations, including likely response bias among those more engaged in their health, and confiding they were looking for positive results.
Among 6825 website “members” presented the survey, 1323 (19%) responded. Responders said the tool helped most in learning about symptoms or medication side effects, recording symptoms or understanding treatment effects. Users gained insight into their condition and increased their coping abilities. Among those with HIV, 71% gained interest in their lab results. Given the degree of social support with such a site, user benefit of self-care is hardly surprising.
What responders said about using Patients Like Me in the context of clinical care was more intriguing.
About 1 out of 3 printed and shared summaries of their symptoms, treatments and outcomes with their clinician. Because of their participation, 2 out of 3 said it was easier to communicate with clinicians (21% a lot) and felt more involved in treatment decisions.
A small but too-big-to-ignore number said they changed providers as a result of website use. It’s unclear what this actually represents. Perhaps users or survey responders have lower access to care; perhaps it does indicate empowerment and raised expectations. As a cross sectional survey, the relationship between usage and behavior cannot be discerned. The authors recognize this, stating
…further research will be needed to elucidate the causal chain: is it that engaged users find more benefit, or that patients who have benefited from the site come back to use more of its features?
Indeed, more research is needed now. Studying the interaction between online support and clinical care, using longitudinal cohort designs, should be supported. Finding an association between site usage and changes in patient (and clinician) behaviors and outcomes has implications for promotion and clinical integration – in particular, linking valuable sites to patient portals and personal health records.
Partnerships wanted: clinical centers, researchers and robust patient online networking sites.
What’s your idea for research and evaluation?