At uni, the campus is slowly starting to wake up after the summer hiatus. The student accommodation is all ready, there are more people in the library and the cafeteria. Outside the university, my kids are back to school. I did the obligatory back to school photo as my kids were running off to their first day with their shiny new shoes and non-scuffed bags. My older son has grown over the summer and is now nearly as tall as me. A real wake-up call that time is passing! On everyone’s lips is talk of instilling better habits, of decluttering houses, doing more exercise, and eating more healthily. The gym is noticeably busier than normal right now.
When are people most likely to try to change their behaviour? Dai, Milkman and Riis examined this question and found convincing evidence for a ‘fresh start effect’; people aspire to change their behaviour at moments which feel like a new beginning, so-called ‘temporal landmarks’. They propose that temporal landmarks enable people to open new ‘mental accounting periods’, putting past failures behind them, providing a broader life perspective and motivating a desire for behaviour change.
Evidence for this theory is provided using three different analyses: Google searches for ‘diet’, university gym attendance records, and commitments made on the goal-setting website StickK. The largest increase for each of these behaviours was seen at the start of January, consistent with people making New Year’s resolutions. Intriguingly, however, they also found increases at other temporal landmarks. The start of each week, each month and each term -for the university gym- saw spikes in both aspirational or actual behaviour. Moreover, personal landmarks were also important, as gym attendance increased dramatically after people’s birthdays (curiously, with the exception of 21st birthdays, attributed by the authors to being the minimum legal drinking age in that state).
Ayers and colleagues used a similar Google search analysis on smoking cessation to examine weekly rhythms in search behaviour across different countries. They found that in general, Monday saw the most number of searches for smoking cessation, with a decline throughout the week followed by a rebound on Sunday. A similar pattern was found for US Google searches classified as health-related, with searches peaking on Monday and Tuesday, declining during the week and rebounding modestly on Sunday. There seems to be a fresh start effect operating on timescales as short as a week, and also as long as a year.
Although this is a relatively new area of inquiry, initial results suggest this ‘fresh start’ effect seems relatively robust across a range of health behaviours (or health-related aspirations, in the case of the indirect evidence such as the Google searches). Understanding people’s self-initiated aspirations to change has clear implications and applications, for example in taking advantage of variations in aspiration to help engage people in healthy behaviours, and in managing resources to take account of these aspirations.
The custom of setting New Year’s resolutions may have its origin in Roman times. January is named after Janus, the two-faced Roman deity who simultaneously faces both backwards into the old year and forwards into the new. The new year symbolises a time of reflection on past events and hopes and plans for the future, perhaps the reason why the ‘fresh start effect’ seems to have its biggest impact in January.
Given the huge amount of research into health behaviour change, the literature is remarkably quiet on the subject of New Year’s resolutions: a time when a large proportion of people pause to reflect on their health and resolve to improve it. The reasons for this relative silence are difficult to understand. Is it because New Year’s resolutions are considered trivial? Not worthy of research effort? Or is it because of the ‘common-sense’, taken-for-granted understanding of New Year’s resolutions as being quickly and flippantly set, and as easily and flippantly discarded? Whether due to these or other reasons unknown, it seems unwise to dismiss New Year’s resolutions so thoughtlessly, given the large number of people who attempt to change their health behaviour at this time of year.
In one of the first studies on New Years’ Resolutions, Marlatt and Kaplan (1972) studied college students’ success at making resolutions. Participants were asked to report on ANY resolutions they made. 34% of participants made at least one, with a mean number of 2.9 resolutions each in categories such as weight loss, health, smoking, and relationships. Participants were followed for 15 weeks, with the average duration of resolutions being approximately 40 days before being broken. Only 25% of resolutions were reported as being broken after 15 weeks, which the authors admit was probably exaggerated, and attributed to the self-report nature of the questionnaire. Another explanation for this inflated success rate is that no guidance was given on the format resolutions should take, so something vaguely formulated such as ‘be nicer to my boyfriend’ could be liberally interpreted as having been successfully implemented than a more specific resolution such as ‘lose 2kg’ or ‘exercise 3 times a week’.
An interesting aspect of this study was that the number of ‘health-related’ (other than weight loss and smoking cessation) resolutions was startlingly small (11 out of 166 total resolutions). This category included not only exercise but also ‘desired changes in physique’. Potential explanations include the student sample (who may have been less concerned about their health than older adults given their relative youth) or due to the study being done over 40 years ago. Physical activity levels globally have declined markedly in recent decades, therefore it could have been the case that these students felt they were physically active enough (and may indeed have been so).
It is also likely that there have been cultural shifts over this time in the recognition of the need for physical activity for health. In fact, a more recent analysis of New Year’s resolutions in the US by YouGov (2013) found that ‘get fitter/do more exercise’ was the second most popular resolution after ‘lose weight’, with 28% of resolvers aiming for more physical activity. This seems a far cry from Marlatt’s results, although in the YouGov survey 35-49 year olds were more likely to make an exercise resolution (35%) than 18-34 year olds (25%). Whilst people towards the bottom end of the lower age bracket (most similar in age to Marlatt’s college student participants) might be less likely to make exercise resolutions than people towards the top of the bracket, it seems unlikely they will approach the very low levels of exercise resolving seen in the Marlatt study.
As an aside, 33% of the survey participants made at least one New Year’s resolution: a very similar rate to that given in Marlatt (1972), although there were large differences between age brackets. 48% of younger participants (18-34) and 37% of 35-49 year olds made at least one New Year’s resolution, but only 26% in the 50+ bracket. Although not a longitudinal study, the YouGov survey was carried out 6 days after New Year’s Day and found that 11% had already broken at least one resolution and 13% were ‘struggling’.
So about the same number of people seem to be making resolutions as 40 years ago, resolutions to increase physical activity levels has rocketed in popularity, and yet there is almost no research on physical activity resolutions. This seems like a massive research opportunity to me! Even if most people don’t stick with their New Year’s resolutions, the number making them are so large that a small proportion changing their behaviour for the longer term could mean huge public health benefits. And importantly, finding out when and how people are most likely to take up exercise just seems like common sense. You can think of immediate applications: starting a running group on Monday rather than a Friday evening, contacting people with health information early in the week or on a Sunday evening, not launching an exercise initiative in July, targeting people just after their birthday, the list goes on.
But anyway, I have to go now. It’s time to make my Back to Uni Resolutions!