This post was precipitated by reading two papers and following references and links in the papers. The post draws attention to the serendipity that can occur on reading a paper, and the enormous changes that have occurred over 40 years. On a practical level, a paper on ‘gamification’ led me to social network measures, and a paper on patient preferences led me to goal setting and goal attainment scaling. In half an hour I looked at and read papers of interest that would have taken weeks or longer to find, copy and read when I started. Follow my trail.

I, like many people, receive the Table of Contents of journals by email. The JAMA Network Open notification on 24th March 2021 had two papers that interested me.

The first was: “Effect of Behaviorally Designed Gamification With a Social Support Partner to Increase Mobility After Hospital Discharge: A Randomized Clinical Trial“
and the second was: “Outcome Goals and Health Care Preferences of Older Adults With Multiple Chronic Conditions“

Paper one

This is an interesting paper which starts from the hypothesis that social peers could be effective ancillary agents to changing behaviour. For example, someone trying to loose weight, stop drinking alcohol, or stop smoking is more likely to succeed if family and/or friends provide support. The investigators wished to test whether or not having a ‘named partner’, selected by the patient from people known to the patient, would amplify the effect of an intervention designed to increase exercise. The patients were people who had been discharged from a medical or oncology ward, and could walk.

The intervention was, as in many rehabilitation trials, multifactorial with three components. The first was using an accelerometer device which fed back a patient’s step count at the end of each day. The second, important one was a game that incentivised doing more walking, and gave a reward if achieved. The support partner provided input into achieving this, through monitoring achievement, reporting it back to the research team and encouraging the patient to achieve and congratulating the patient on achievement. The third was that the experimental group had a target amount of improvement set each week – an increase of 10% in daily step count.

The comparison group patients also had feedback of daily step count, but had no goals set externally and no specific social support.

In the trial (n = 232 patients; outcome assessed at 12 weeks), there was no statistically detected difference between the two groups in step counts, function, or hospital admissions.

The research team noticed that many of the patients had quite sparse, thin social networks. For example, in the discussion they wrote: “some [participants] were unable to name 1 person whom they felt could be relied on to receive information on the participant’s progress and send messages for reinforcement and encouragement to the participant“.

So they then looked at the data relating to the 76 people who had larger social networks, and it seemed that in this sub-group analysis, people with stronger social networks did benefit. This is currently an unconfirmed hypothesis.

I am interested in social functioning, including social networks and social roles, and how they can be measured easily. So I followed this up.

The research team measured each patient’s social network, using a short assessment – the “abbreviated version of the Lubben Social Network Scale” – and they gave a reference which can be accessed here. The paper had been published in 2006, but simply activating the link immediately showed me the paper, and the paper showed the abbreviated scale. The scale is indeed short and would be easy to use.

In relation to their unconfirmed hypothesis, they made two strong points within the 2021 paper. The authors refer to systematic reviews of studies of patients with diabetes and heart failure that provided persuasive evidence that higher levels of social support was associated with better health outcomes. The authors also point out that two published trials that did show positive effects for similar social support programmes recruited patients who had or were likely to have significant social support.

It does, therefore, seem likely that this would be a beneficial intervention in people who have an existing reasonable social support network. Although this helps those people, they are already more likely to be more healthy. The challenge, as always, is to help the more disadvantaged sector of the population.

The abbreviated scale could easily be used almost as a routine, just as medically one may ask about chest pain when undertaking ‘a system review’. This would be a social review set of questions, picking up who is likely to be lonely. This might possibly be a more important piece of information. (here)

Paper two

I looked at the second paper for several reasons. It had, in its title, ‘goals’ and ‘health care preferences’, both aspects of rehabilitation that interest me. The first author was Tinetti, a name I recognised. And it was the next paper in the list, so I could not overlook it.

The study involved a population who will be involved with rehabilitation – every one of the 163 patients had three or more long-term conditions, and were on multiple medication. Each participant underwent a 30-40 minute interview to establish their health care priorities. They were then asked to describe their goals in relation to healthcare, and what they found burdensome and what they found helpful about healthcare.

The range of goals are shown in a large Table (two); I doubt that many of these would feature on a rehabilitation plan. Another table also show what patients find burdensome; I doubt that this list would surprise anyone, except we should be surprised that we (healthcare professionals) still ask patients to do these things!

My main interest in this paper was the introduction. It was a short but helpful overview of some concepts involved when considering setting goals and articulating preferences. There were many helpful references, some of which I looked up and could access easily.

For example, a qualitative study published in 2017 explored “what is most important to” people with multiple conditions. This is likely to apply certainly to most people with a disability, and indeed probably to most people. The results identified six domains (these are all direct quotations from the paper (here)):

• “PRINCIPLES consist of beliefs and standards to live by, such as ideals, virtues, and aspirations. 
• RELATIONSHIPS consist of social connections with family, friends, coworkers, neighbors, or community groups.
• EMOTIONS refer to feelings, mood, or states of being that are personal, embodied, and experiential.
• ACTIVITIES are important pursuits, such as reading, gardening, learning, and medical self-care. 
• ABILITIES include physical or mental capacities or skills, such as mental sharpness, vision, mobility, and problem-solving skills. 
• POSSESSIONS include tangible objects and important spaces.“

There was a second reference that caught my eye; a non-randomised comparison of goal-attainment scaling with using individualised patient reported outcome measures to evaluate outcome. (here) The study found, contrary to expectation, that more clinicians chose goal attainment scaling even though it was a new technique for most of the clinicians.

In terms of their study question, they found no difference in outcome between the group using goal attainment scaling and the group using existing measures to define specific patient goals and to measure specific patient outcome. The discussion is interesting because it outlines the not-inconsiderable risks associated with using goal attainment scaling (or any similar means) as a way to define success from a payment perspective.

These experiences led me to two reflective ideas.

The first is that it reminded me of my first serious (at the time) research in my third year as an undergraduate; I happened to be experimenting in fertilising mouse oocytes. I had to write a dissertation, and I discovered the pleasure of following trails from a paper. A paper makes a point and gives a reference which you look up, and the next paper does the same, and you look that up, If the reference journal is available in the library for the requisite year, it is quick but so often it was a matter of asking and then waiting days or weeks.

I still get the same pleasure from following trails, even if they are unrelated to the original question.

The second is that it is now so much easier. Papers often give internet links (DOIs) for references; medical data-base search engines and Google are always available. Though one cannot access every wanted reference immediately, enough can be to increase the enjoyment.

I hope that this will encourage readers to request notification of Tables of Contents and/or newly published papers from a few journals, important now that going to libraries and looking at journals seem rare. I hope this will lead to you reading papers you might never have looked for or heard of. And that may then lead you to areas which are of interest but not necessarily easy to search.