The treatment of back and neck pain. Cost and utility

Abstract

Aims To analyze the costs and effects of treatments used to return persons sick-listed due to back or neck problems to work, and to perform a cost-utility analysis of such treatments.Study population and Methods Included were men and women between the ages of 18 and 59 years who were employed but sick-listed (100 %) for at least 28 days due to a low back or neck diagnosis. The subjects received questionnaires which pertained to back function, back pain and quality of life, all of which were answered after 28 days, 90 days, one and two years. The subject's working status was obtained through insurance registers. Main outcome measures were working/not working, back function, and pain. Paper I. Subjects sick-listed due to low back pain for a minimum of 90 days at entry were recruited in Denmark, Germany, Israel, the Netherlands, Sweden and the United States. The questionnaires, which included background factors and treatments were sent to the subjects for the first time after 90 days. Paper II. A bottom-up approach was used in a cost-of-illness analysis for individuals with low back or neck problems in Sweden. Each individual's healthcare consumption was prospectively documented in detail through diaries and questionnaires from the first day of sick-listing, in a cohort of 1822 persons sick-listed 28 days or longer. The costs for this consumption and the loss in productivity due to sick-listing, and permanent disability benefits were determined individually as were the total annual costs for back and neck problems in Sweden. Paper III. The responses to ten different health questionnaires by 1575 Swedish subjects sick-listed for more than 28 days for back or neck problems were analysed for their ability to predict return to work (RTW) or not (NRTW). A statistical diagnostic test was used to study how well the subjects' values on these questionnaires would predict RTW or NRTW within 90 days, one and two years. Paper IV. Ninety-two individuals in a cohort of 1146 Swedish persons sick-listed for more than 28 days due to back problems underwent lumbar disc herniation surgery during the study period. Each person operated on was individually matched with one treated conservatively. The costs and effects of the treatments were determined individually. By estimating quality of life before and after the treatment (EQ-5D), the number of quality-adjusted-life-years (QALY) gained with and without surgery was calculated.Results Paper I. The questionnaires were completed by 2080 subjects in the six countries. There were great similarities in the number of appointments, examinations, and treatments in the different countries. Back surgery rates varied, from 6% in Sweden to 32% in the United States during the first 90 days of the study. Almost no noticeable positive effects on work resumption, pain, or back function were found. Back surgery in Sweden was an exception. Work resumption ranged from 32% in Denmark to 73% in the Netherlands during year one. Paper II. The costs for all medical services during the two year study (direct costs) were 6.9% of the total costs. Surgery was the single most expensive item. Nationally, the total annual Swedish costs for back and neck problems were at least 1% of the GNP. Paper III. A general improvement occurred from day 28 up until two years, and was most pronounced for men with low back problems. RTW and NRTW were easier to predict within 90 days than one or two years, and EQ-5D was best to predict both. Paper IV. The total costs were lower for those operated than for those not operated for a disc herniation, in spite of the relatively high costs for surgery. Those operated had lower recurrence rates for sick-listing episodes and permanent disability benefits. The gain in QALY was ten times higher among those operated, and surgical treatment had a better cost utility. Conclusions Paper I. Long-term back problems received the same type of treatment in the European countries, as well as in the US. No treatment in any of the six participating countries improved pain, back function or work ability, except for back surgery which improved pain in three cohorts. Paper II. The relative direct and indirect costs in Sweden for back and neck pain amounted to 6.9% and 93.1%, respectively. The Swedish annual cost of illness for back and neck problems was at least 1% of the nations GNP (2001). Paper III. The single most powerful predictor of RTW and NRTW was the quality of life instrument EQ-5D (EuroQol). Paper IV. Surgery for herniated lumbar disc disease was a cost-effective procedure for appropriately selected patients. The total costs for disc surgery patients were lower than for non-operated subjects, and there was a large improvement of utility.