The Epidemiology of Pain: An Australian Study

The Results of a Telephone Survey in Brisbane
Simon Strauss M.D.*
Fiona H Guthrie**
Fred Nicolosi**

Introduction

The epidemiology of pain prevalence rates in western communities is beginning to be extensively studied. The Nuprin Pain Report (1) was a telephone survey of 1254 Americans completed in 1985. This represented a cross-section of the adult population in the United States. In Canada, Crook et al (2) interviewed the inhabitants of 500 households chosen from the roster of a group family practice and in Sweden, Brattberg et al (3) conducted a combination of a postal and telephone survey of 1009 randomly chosen individuals.

This study investigates pain prevalence rates in a randomly selected sample of the Australian population.

Method

A random telephone survey of 265 households in the Central Brisbane Telephone Zone and the Outer Metropolitan Zones was undertaken in October/November 1987. This sample was based on an estimated prevalence rate of self reported pain of 20%.

The telephone numbers were drawn from the Brisbane telephone directory. The sampling units (households) were selected by choosing the 10th telephone number of the middle column on every third page, starting at a page chosen by a number from a random number table. Obvious business telephone numbers were excluded, and the next private telephone number substituted. This substitution introduces some bias in the probability of selection. One follow up phone call was made to any number which was not answered initially.

Those people who co-operated in the survey were asked whether they or any other member of their household over 15 years of age "were currently experiencing pain or regularly troubled with pain". If a negative answer was received the interviewer asked for brief demographic data only (sex and age of each person in the household over fifteen years of age) and the survey terminated.

If a positive response was received the full pain questionnaire was completed for each person in the household currently experiencing pain or regularly troubled with pain. Whenever possible attempts were made to obtain the data directly from the person with the pain complaint.

Respondents were asked to identify the location(s) of the pain, its intensity, its frequency, whether it had been experienced in the last two weeks, the circumstances surrounding onset, the length of time since onset, whether they currently visited a health professional for the paincondition or used self or no treatment, and whether the treatment sought was effective.

The results were tabulated to address: household and individual pain prevalence rates; a comparison of the demographic characteristics of the pain and the no pain populations; demographic characteristics of the pain population; pain characteristics; and the treatment sought.

Results

The household and individual pain prevalence rates are presented in Figure 1. Of the 265 households included in the survey, 94 households had at least one person who was currently experiencing pain or was regularly troubled with pain, amounting to a household pain prevalence rate of 355 per 1000 head of population (35.5%). If it is assumed that all of the households that refused to participate in the survey or could not be reached were free of pain then the minimum household pain prevalence rate would be 188 per 1000 head of population (18.8%).

Of the 94 households in which at least one person was currently experiencing pain or regularly troubled with pain, there were 84 households in which one or more persons reported that a pain condition had been experienced within the two weeks preceding the survey. This amounted to a household pain prevalence rate for pain in the two weeks preceding the survey of 317 per 1000 head of population (31.7%). The corresponding minimum pain prevalence rate (defined as above) is 168 per 1000 head of population

A total of 614 individuals participated in the survey. Of these 117 individuals reported that they were currently experiencing pain or regularly troubled with pain amounting to an individual pain prevalence rate of 191 per 1000 head of population (19.1%). A minimum individual pain prevalence rate can be calculated by extrapolating the number of individuals in the co-operating households to the number of households contacted overall. This provides an estimated number of individuals of 1158 and a minimum pain prevalence rate of 101 per 1000 head of population (10.1%).Of the 117 individuals reporting that they were currently experiencing pain or regularly troubled with pain, 101 had experienced this pain within the two weeks preceding the survey amounting to an individual pain prevalence rate of 164 per 1000 head of population (16.4%). The corresponding minimum pain prevalence rate was 87 per 1000 head of population (8.7%).

The demographic characteristics of the survey population were compared using the chi square statistic the variable of sex was not statistically significant in discriminating between the pain and no pain groups. However the pain and no pain groups differed significantly on the variables of age (p<0.001) and household size (p<0.001). The pain population tended to be older and live in smaller households in comparison to the no pain population.

For all persons surveyed pain prevalence rates increase with age. Females have higher pain prevalence rates across all age groups except for the 15-30 year interval where the rate is equal. The overall responses to the pain characteristics investigated are reported below.

The location of most severe pain for the majority of respondents was the back (33%), followed by head and neck (24%), and leg (22%). The scale used in the question about pain intensity was based on the Short Form McGill pain questionnaire. The majority of respondents described their pain condition as discomforting. For 45 percent of respondents their pain condition was distressing or worse.

The circumstances surrounding the onset of the pain condition were of unknown or spontaneous origin for the majority of respondents (34%). Work related injuries accounted for 21 percent of responses; nearly 85 percent of respondents had experienced their pain condition for longer than 12 months and 67 percent had experienced their pain condition for longer than 3 years; and the frequency of pain occurrences was continuous for 23 percent of respondents and daily for 30 percent of respondents. The category of "other" includes respondents with random or seasonal occurrences of pain.

The majority of respondents sought the assistance of health professionals for the treatment of their pain condition (70%). Of the remaining 30 percent, equal numbers used self or no treatment. Doctors were the most commonly consulted health professionals (84%).

Discussion

It was not always possible for the pain questionnaire to be answered by the person with the pain condition. In about alf the cases the pain questionnaire was answered by the person with the pain. Cross tabulations show that this percentage varied across age groups with the lowest percentage of persons reporting their own pain in the 15-30 year age group. This fact introduces some bias and is a commonly encountered problem in telephone health surveys(4). However there is some evidence that proxies can provide useful estimates of subjective experiences such as pain when they live with the sufferer (5).

The Nuprin Pain Report (1) does not report overall pain prevalence rates. The study investigated seven specific types of pain and their prevalence in the preceding 12 months. The most common type of pain was headache (73%) followed by backache (56%). These were also the most common pain conditions in this study. The magnitude of the prevalence rates differ however; as the Nuprin study included pain which had been experienced for even one day in the preceding 12 months.

The study by Brattberg et al (3) asked respondents "do you have/have you had any pain or discomfort in any part of your body?". Sixty six percent of respondents had experienced such pain or discomfort. It is difficult to relate this result to the present study as again the question asked is quite different. However the direction of other results is similar to those found in this study. present pain or discomfort which had lasted for more than 6 months was reported more frequently than pain which had persisted for less than 6 months. Of those respondents whose pain had persisted for longer than 1 month at least 75 percent reported pain intensity as "like being stiff after exercise" or worse.

The initial question posed by Crook et al (2) was "are you or any member of your family ... often troubled with pain?". ‘his was similar to the initial question posed in this survey of "are you or any member of your household currently experiencing pain or regularly troubled with pain?" The results obtained by Crook et al are very similar to those obtained in this study. The household pain prevalence rate reported by Crook et al for the two weeks preceding the survey was 35.2% and the minimum household pain prevalence rate was 26.2%. The corresponding results in this study were 31.7% and 16.8%. The individual pain prevalence rate reported by Crook et al for the two weeks preceding the survey was 16.1% and in this study was 16.4%. The various prevalence rates in the studies cited above and in the present paper indicate large variations. These variations relate to the type of questions asked and the survey technique employed. Nevertheless all studies show high prevalence rates which must be cause for concern. Notwithstanding any possible biases due to the limitations naturally imposed by a telephone survey, this study reveals an alarmingly high prevalence of pain in an Australian community. Even an at worst result calculation indicates a household pain prevalence rate of 18% and an individual pain prevalence rate of 10%. In human terms a number of people in the community have pain which affects their quality of life. The economic cost to all of us must be considerable.

  1. Sternberg RA. Survey of pain in the United States: the Nuprin pain report. The Clinical Journal of Pain 1986; 2: 49-53.
  2. Crook J, Rideout E, Browne G. The prevalence of pain
    complaints in a general population. pain 1984; 18: 299-314.
  3. Brattberg G, Thorslund M, Wikman A. The prevalence of pain in a general population. The results of a postal survey in a county of Sweden. pain 1989; 37: 215-222.
  4. Raj D. The design of sample surveys. Sydney: McGraw-Hill Book Company, 1972: 260-262.
  5. O'Brien J, Francis A. The use of next-of-kin to estimate pain in cancer patients. pain 1988; 35: 171-178.

SUMMARY OF FINDINGS

This report presents the results of a telephone survey on the prevalence rate of pain, the distribution of pain rates and the characteristics of pain according to selected demographic variables, and the treatment undertaken by persons experiencing pain.

The key findings are as follows:


* 31 Charlton St. Southport 4216 and ** 31 Charlton St. Southport 4216. At the time this research was undertaken both these authors were candidates for the degree of Master of Business Administration at the University of Queensland. Correspondence to Dr. Strauss.