Can text messages be used to monitor health?

text-messageMobile phones are possibly one of the most ubiquitous pieces of technology in the 21st century.  In 2003 the Office of National Statistics reported that 75% of all adults in the UK owned or used a mobile phone, and Wikipedia estimates that the UK now has more mobile phones than people.

Text messaging has been the surprise success of mobile phones – who would have thought that people would prefer to write out a message to their friends rather than simply give them a quick call?  The latest statistics record that 4.7 million messages are being sent every hour in Britain, with 72% of women and 70% of men texting regularly.  These days, however, text messages are being put to much worthier use than simply catching up on last night’s gossip.

Earlier this month the BMJ reported that the ability to send a text message was a sure sign of recovery from a faint or a panic attack.  The authors of the study had spent several years working for the charity Festival Medical Services, helping out in the first aid tents at large British music festivals.  They noticed that teenagers who were brought in for medical attention would send a text message to their friends as soon as they were feeling better.  Given that sending a text message requires “a Glasgow coma scale score of 15 (fully conscious), an adequately functioning ‘executive system’ in the frontal lobes, and a high degree of manual dexterity and psychomotor coordination”, they concluded that this ability was reasonable grounds for discharge and successfully employed ‘the text test’ during busy times at this year’s Reading Festival.

Text messages have also been employed by heathcare trusts to monitor the medical status of patients.  Oxfordshire and Buckinghamshire Mental Health and Oxford University’s Department of Psychiatry have been using a daily text message to monitor the mood of mental health patients.  Patients receive a text message every morning and  respond with a letter in the range of A to E, which corresponds with mood on a scale from depression and euphoria.  This technique allows doctors to to plot mood swings in real time, monitor medication efficacy and assess whether a face-to-face appointment is needed, and removes the need for unreliable retrospective assessments.  The system is current being used with 150 patients and is due to be rolled out across Oxfordshire and Buckinghamshire in the new year.

In a similar vein, London’s Hammersmith & Fulham Primary Care Trust has been using text messages to establish the current smoking status of patients and, accordingly, their risk of developing cardiovascular disease.  As many as 49% of patients responded to texts enquiring whether they were active smokers, a much higher response rate than other methods of communication previously used.  Doctors were then able to quickly and efficiently update medical records.  The system that the Trust used, iPlato, began as a means of reducing missed GP appointments within the NHS and has now been adopted across 40 NHS organisations, including 36 Primary Care Trusts.  As well as monitoring smoking status, iPlato also uses text messages to establish methadone compliance, keep tabs on patients on weight management plans and helps patients stay up to date with prescriptions.

And it’s not just in the UK that healthcare providers are exploiting text messages to monitor patients.  The Infectious Disease Clinic at the University of Virginia is piloting a text messaging system to improve contact with HIV patients living in rural parts of the state. In the programme, patients are given a mobile phone that only receives texts about medication regimens and pending medical appointments, and can only call emergency and medical contacts.  Organizers hope that this initiative will help patients adhere to their treatment for longer than 6 months, the current drop off point.

Stanford University in California and the Centers for Disease Control and Prevention even held a conference on text messages and healthcare in February this year – Texting 4 Health.  The event “showcased applications, ideas, and insights on using mobile text messaging to improve personal and public health”.  Speakers from as far afield as Denmark and Scotland presented the text messaging strategies that are currently in use in their regions – HandyMentor for asthma self management and SweetTalk for diabetes reminders if you’re wondering – and techniques to provide health information and instigate healthy lifestyles via text messaging were presented (the full conference program and slides can be found here).

Given the widespread use of mobile phones, text messages could certainly be a cheap and far reaching way to monitor patients and disseminate public health information.  What do you think?  Are text messages the future of health management?  Are text messaging programmes going to be widely adopted for public health purposes in 2009?  Or are text messages good only for personal affairs and not for important medical information?

Continue Reading

Drinkers don’t take their medications often enough

alcoholA recent study by Bryson et al. has found that moderate to severe alcohol misuse increases the likelihood that patients won’t take their medication properly.

Many patients do not take their medications as often as they should – i.e. on at least 80% of the days they are supposed to.  In fact, a recent study found that over the space of a year, 40% of patients taking cardiovascular or diabetes medications didn’t take their medications often enough.  Such ‘medication nonadherence’ is associated with worsening of disease, increased health care costs, and even death.

Bryson et al. looked at more than 20,000 patients who were receiving treatment for high blood pressure, high cholesterol levels, or diabetes.  All participants undertook a three-point questionnaire to evaluate their alcohol use on the basis of frequency and typical quantity of drinking during the past year, and the frequency of heavy episodic drinking (at least 6 drinks per occasion).  Medication adherence over the space of a year was measured by how often patients went back to their pharmacy for a refill.

Among patients taking medication for high blood pressure or high cholesterol levels, nonadherence increased as the severity of alcohol misuse increased.  Compared with patients who did not drink, the proportion of patients who did not stick to their cardiovascular medications was significantly higher among those who moderately or severely misused alcohol.  Interestingly, there was no difference in adherence to diabetes medications between diabetic patients who did not drink and those who did.

This research might seem like it’s straight from the department of the obvious: “Of course people with alcohol problems don’t take their medications properly!”  There are a couple of key findings that are important to bear in mind though.  For one, the authors were able to assess alcohol consumption with a brief questionnaire , unlike previous studies on this subject that used lengthy, time consuming interviews.  Thus the approach used in this study could be used easily in clinical practice.

In addition, a fair few studies have examined the problem of medication nonadherence, but most have found that the factors responsible for nonadherence are ones that would be very difficult to modify, such as older age (over 80 years old) or low socioeconomic status.  The study by Bryson et al. is important because it identifies a modifiable factor responsible for medication nonadherence.  Counsel a patient to cut their drinking and, in theory, they should be more likely to take their medication properly, which would keep their condition in check and enable them to get on with their life unhindered.

————————————————————————————————-
Bryson CL et al. (2008) Alcohol screening scores and medication nonadherence. Ann Intern Med 149 (11): 795-803. PMID: 19047026

Continue Reading