For a while I’ve been asking what constitutes ‘self management’ for chronic pain. On the one hand there are a group of people who firmly believe that regular medical treatment (including injections every three months or so) is a perfectly legitimate way to maintain a normal life. On the other hand there are people who equally firmly believe that self management means as little contact with a health care provider as possible, and certainly no hands-on or injection therapy.
This study is an old one, dating from 1999, but there’s no reason to believe the findings differ too much from what we’d find today. It was conducted by Saunders, VonKorff, Pruitt and Moore, based mainly in Seattle, and looks at the extent to which specific patient attitudes and beliefs about medical care and self-care for back pain predict future healthcare use.
The study design involved a five year follow-up of patients who had attended Group Health Cooperative of Puget Sound. All patients were primary care patients, so unlike many studies, these people were not necessarily seen in a tertiary hospital setting.
There were three groups of patients: the first group of around 1200 were recruited in 1989 and 1990, and were adult primary care patients with back pain, interviewed by telephone up to 5 years after recruitment. In 1994 and 1995, over 1000 patients were enrolled in a long-term study in which participants at one month after an initial consultation for low back pain were given questionnaires that included self-care vs. medical care orientation questions – these were recruited from a Pain Register. The final group were recruited from a large study of people invited to participate in a four-session psychoeducational programme for low back pain. A small number of participants were studied (120).
The 5-year follow-up and Pain Register Baseline samples
allowed the evaluation of 11 questions to assess self-care vs.
medical care orientation and their ability to predict back
pain-related healthcare use. The Randomized Trial sample
allowed the examination of the impact of a back pain selfmanagement
intervention on attitudes and beliefs shown to
predict future healthcare use.
Three main measures were used in the study – a specially designed questionnaire of attitudes and beliefs toward self-care and medical input (this questionnaire is published in the article – useful future reference!); pain intensity and interference was also measured using VAS and categorical ratings; and healthcare use data was obtained from the healthcare services healthcare visit and pharmacy databases.
Descriptive statistics were calculated and differences
between groups were examined using chi-square tests. Internal
consistency reliability was estimated using Cronbach’s
a-coefficient and factor analysis was conducted using a
varimax rotation requesting a 2-, 3-, and 4-factor solution.
Multiple regression analysis was used to test potential
differences between group means, while controlling for
pain intensity and pain interference.
Cutting to the results I was interested in (you can read the descriptive statistics and details in the article itself), what predicted higher health care use?
– higher back pain intensity scores
– Agreement with the item, `pain requires ongoing attention from a physician’ consistently predicted `higher’ utilizer
status and higher back pain visit rates
– Belief in the ability to care for back pain on one’s own predicted lower likelihood of frequent back pain visits in both samples
– Agreement with the statement that `prescription pain relievers are necessary to control pain when it is really bad,’ and disagreement with the statement, `rarely go to a physician for back pain because they can’t do much for the problem,‘ also predicted higher use of back pain visits in both samples
The authors state: It is worth noting that items involving the concept of physicians `curing’ back pain or finding the `cause’ of back pain did not predict future healthcare utilization.
The use of opioid medications was specifically studied (I’m noticing an increase in the number of people referred for back pain to the PMC taking opioids)
Patients in both samples who agreed with the statement `prescription pain medications are necessary when pain is really bad’ had roughly three times the rate of opioid prescriptions than did patients who disagreed with the statement. Similar results were observed for the item `I avoid prescription medications even when the pain is severe.’
Agreement with the statement, `pain requires ongoing attention from a physician’ was also a strong predictor of
use of opioid medications. Similarly, agreement with the statement `you feel able to care for your back problem on
your own’ was negatively associated with mean number of opioid prescriptions filled in both samples.
Very similar findings were obtained for the use of hypnotic drugs for sleep.
The effect of the four-session self management programme found that at 3 month follow-up, patients who were assigned to the self-care groups were significantly more likely than those randomly assigned to usual care to agree with the statement `things done on my own are more helpful than medical treatments’ (60.2 vs. 46.7%, P , 0:05). At 6 month follow-up, intervention patients were significantly more likely than usual care patients to agree that they felt able to care for pain on their own (69 vs. 51 %, P , 0:01).
The psychoeducational intervention appeared to have no effect on the two medication-specific items.
What can we conclude from this?
It seems clear that pain intensity alone does not predice use of health care – in fact, it’s much more about the attitudes and beliefs that individuals have toward their back pain, and the value they place on medication and medical care that influence health care seeking.
Interestingly, it’s not apparent from this study, that people who sought medical care were looking for either a cure or a cause – they simply wanted less pain and thought that medication was more effective than their own efforts. What’s scary is that opioids are not often the best medication for low back pain, especially in relatively young people.
What’s even more scary is this finding: While our results indicate that multiple physician visits for back pain are atypical (Pain Register Baseline patients had an average of 1.9 back pain visits over a 2-year period), 10% of this patient sample had five or more physician visits for back pain over this same interval. And, patients with more visits expressed attitudes less favorable toward self-care for back pain management.
There are several limitations to this study – the data used for health care use was drawn from automated records and didn’t include work accident events, the population included are generally well-educated, and the study looked only at patient attitudes and didn’t include health care provider attitudes – several recent studies suggest that provider attitudes are just as likely to influence patient behaviour as patient’s attitudes.
Food for thought however. It would be interesting to see whether these same questions used in general practice, primary physiotherapy practice, or even as part of ACC here in New Zealand, could improve our management of the people we see returning for yet more medical input.
Why worry about it? Cost alone: but more importantly, there doesn’t seem to be much evidence that using opioids improves function, and the ‘innoculation’ effect against self management seen here in people seeking more medical input suggests that self management may be difficult to initiate in this group, yet this is the approach given to people only after they’ve failed to respond to medical care. Hmmm, maybe we have the mix back to front?
Saunders KW, Von Korff M, Pruitt SD, & Moore JE (1999). Prediction of physician visits and prescription medicine use for back pain. Pain, 83 (2), 369-77 PMID: 10534610