Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The cost‐effectiveness of fluoridating water supplies in New Zealand

The cost‐effectiveness of fluoridating water supplies in New Zealand Objective: Tooth decay has been substantially reduced in New Zealand, and the difference in caries rates between fluoridated and non-fluoridated areas has narrowed. We investigated whether it is cost-effective to fluoridate water supplies that are now non-fluoridated. Janice C. Wright Health economist, Wellington, New Zealand Michael N. Bates Institute of Environmental Science and Research Ltd, New Zealand Terry Cutress Wellington School of Medicine, New Zealand Methods: The net cost of fluoridation was based on the cost of fluoridating a water supply minus the averted costs of treating decay. A range of population sizes was considered. The main analysis was conducted from a societal perspective, using a real discount rate of 5%. Fluoridation was assumed to occur continuously between the years 2000 and 2030. Other assumptions were a Maori population proportion of 15%, no new decay after age 34, and no further dental cost savings after age 45. Information on averted decay in 4 to 12 year old New Zealand children (29,000 receiving fluoridated water and 47,000 receiving nonfluoridated water) was available; information on averted decay in adults was obtained from a study in the United States. Sensitivity analyses investigated the effects of varying the Maori population proportion, the discount rate, and the number of fluoride injection sites. Martin Lee School and Community Dental Service, Canterbury District Health Board, New Zealand Results: Fluoridation was cost-saving (dental cost savings exceeded fluoridation costs) for communities above about a thousand people. The true break-even community size may be lower. For smaller communities, fluoridation may be considered cost-effective depending on the non-monetised value assigned to an averted decayed surface. Conclusions and implications: Fluoridation remains very cost-effective, and is particularly so for communities with high proportions of children, Maori, or people of low socio-economic status. (Aust N Z J Public Health 2001; 25: 170-8) luoridation of water supplies began in New Zealand in 1954. By 1999, fluoridated supplies served about 57% of the population (Personal communication: Alan Ferguson, ESR). Over about the past two decades, tooth decay in deciduous and permanent teeth of New Zealand children has dramatically reduced. The difference in dental caries rates between fluoridated and non-fluoridated areas has narrowed.1 The caries reduction is attributable to widespread use of fluoride-containing toothpaste and water from fluoridated supplies. The ‘halo effect’, where foods and beverages from fluoridated areas are consumed in non-fluoridated areas, has probably also played a role. The general reduction in caries raises the question of whether it is economic to fluoridate presently non-fluoridated New Zealand water supplies, or to replace existing fluoridation equipment. This study addressed this issue. It assumes that optimally fluoridated water supplies do not cause adverse health effects.2-5 A key aim of this cost-effectiveness analysis (CEA) was to estimate the minimum population for which oral health benefits from the introduction of fluoridation to a water supply would be greater than water treatment costs. However, population size is only one factor that influences the cost-effectiveness of this intervention. The sensitivity of the cost-effectiveness of fluoridation to two other ‘community’ factors – the number of fluoride injection sites and the demographic composition of the population served by the fluoridated supply – is explored in this paper. Methods Economic methodology A CEA differs from a full cost-benefit analysis because benefits are measured in appropriate ‘natural’ units, rather than in dollars. In this study, one averted newly decayed tooth surface in a permanent tooth was the unit used to measure the benefits (effectiveness) of water fluoridation. A societal perspective, taking account of all benefits and costs wherever they fall, was adopted. Thus, the net cost of fluoridation is the cost of fluoridating the water supply minus the averted dental costs. Nonmonetised benefits, such as reduced pain and improved social interactions, are represented by the proxy of averted decayed surfaces. Therefore, cost-effectiveness ratios are measured in units of dollars per averted decayed surface. Submitted: June 2000 Revision requested: January 2001 Accepted: March 2001 Correspondence to: Dr Michael Bates, School of Public Health, 140 Warren Hall, University of California, Berkeley, CA 94720, United States. Fax: +1 510 843 5539; e-mail: m_bates@uclink.berkeley.edu AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2001 VOL. 25 NO. 2 Health Promotion Cost-effectiveness of fluoridating water supplies Fluoridation of a water supply incurs a stream of costs and provides streams of averted decay and associated dental cost savings stretching into the future. Both net costs and effectiveness (number of averted decayed surfaces) were expressed in terms of their present value (PV), using a real discount rate of 5%. A real discount rate reflects the decision maker’s preference for present benefits and costs over future benefits and costs, without the need to adjust for inflation.6 All costs were measured in 1999 dollars. Data and assumptions Four sets of data were required – demography (community size and age structure), averted decay, cost savings from averted decay, and the costs of fluoridation. Demography Five assumptions were used in constructing the demographic component of the model. 1. Continuous fluoridation of a supply between the years 2000 and 2030, with one fluoridation plant replacement. Discounting renders insignificant costs and benefits more than 30 years into the future. 2. No new averted decay after age 34. This and the next assumption were a consequence of the source of adult dental data (see below). 3. No dental cost savings after age 45. 4. From 2000 to 2030 there is no mortality in the birth cohorts receiving fluoridated water. 5. Out-migration in the cohort is exactly counterbalanced by inmigration. The first three assumptions would cause under-estimation of fluoridation benefits, and the fourth over-estimation of benefits. The third assumption would be dominant, as dental care becomes increasingly complex and expensive with age. The benefits depend on the year of birth. Those born in 2000 or after would benefit from fluoridation from their birth year to 2030; those aged 15 or less in the year 2000 would benefit from 2000 to 2030; and those aged over 15 in the year 2000 would benefit until age 45. Population growth from 2000 to 2030 was taken as the ‘average’ projection from the 1996 New Zealand Census.7 We assumed fluoridated communities maintained the same projected age and ethnic structure as the total NZ population in the year 2000 – 15% Maori, 85% non-Maori. The percentage of Maori in the total population is actually projected to increase to 19% in 2030. Estimates of averted decay Estimates of averted decay were taken from two sources. For children aged from 4 to 13, New Zealand data were available. For 14 to 34 year-olds, results of a study from the United States were used. Ages 4 to 13 years: In New Zealand, dental treatment is publicly funded until age 17. Although there are few routinely collected data for adolescents, data are collected on the dental status of 5 to 13-year-old children by school dental service providers. Wellington and Canterbury providers use the same software for 2001 VOL. 25 NO. 2 recording dental data for each child. Whether the child lives in a fluoridated area is also recorded. All records for 1996, for children aged 4 to 13, were extracted from the Wellington and Canterbury databases. The resulting dataset contained 29,097 children receiving fluoridated water and 46,825 receiving non-fluoridated water. Data on treatment for deciduous teeth were restricted to ages 4 to 10 and, for permanent teeth, to ages 6 to 13. Data for extractions of deciduous teeth, but not permanent teeth, were available. Permanent tooth extractions are not usually performed by School Dental Services. An extraction was counted as equivalent to five decayed tooth surfaces. Our natural unit of effectiveness, one averted decayed surface in a permanent tooth, does not capture the benefits of averted decay in deciduous teeth. This understates the case for fluoridation, although the value of averted decay in deciduous teeth is less than for permanent teeth because deciduous teeth are exfoliated by late childhood. However, the averted cost of filling deciduous teeth was incorporated into the analysis. Overall, then, averted decay in deciduous teeth is taken into account in the calculation of costs, but not in the estimation of effectiveness. Ages 14 to 34 years: Because the effect of fluoridation on teeth is largely topical, it benefits people of all ages.8-12 However, there are no New Zealand data on adult dental benefits associated with fluoridation. Therefore, for those aged 14 and above, data from a study in the United States were used.13 The study population was 10,628 Washington State employees and their spouses, aged 20 to 34. After controlling for various factors, including other fluoride exposure sources, the study found that each year of fluoridation exposure reduced decayed and filled surfaces by 0.29 surfaces (95% CI 0.19-0.39 surfaces). We used this estimate of 0.29 decayed and filled surfaces averted annually for non-Maori aged 14 through 34, on the assumption that the benefit accrues in the year of exposure. We increased the 0.29 surfaces to 0.59 surfaces for effectiveness in Maori teenagers and young adults, by scaling it by the ratio of the Maori to non-Maori effectiveness at age 13. Thus, for the base-case analysis, in which the Maori population proportion was 15%, average annual effectiveness was 0.33 averted decayed surfaces. Cost savings from averted decay Using only the averted cost of the initial filling greatly underestimates the lifetime flow of costs triggered by the initial decay of a tooth surface. More than 60% of all dental restorative work involves replacement of restorations.14 Continued replacement often entails an escalation in complexity and treatment costs. An averted decayed surface, rather than an averted decayed tooth, was used as the unit for the effectiveness of fluoridation, since it is a more sensitive measure. This excludes downstream averted dental costs, like crowns and root fillings. We assumed: • All restorations are single surface amalgams. • Amalgam restorations require replacement every eight years. • No restorations after age 45. This is a consequence of the time horizon of the analysis and the assumption of no averted decay after age 34. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Wright et al. Article Table 1: Costs of establishing and operating a water fluoridation plant. Population served 1,000 5,000 500 250 5,000 2,000 24,000 31,750 2,500 2,000 2,000 913 7,413 10,000 500 250 5,000 2,000 24,000 31,750 2,500 2,000 2,000 1,825 8,325 25,000 500 250 5,000 2,000 24,000 31,750 2,500 2,000 2,000 4,562 11,062 50,000 2,000 250 10,000 2,000 24,000 38,250 2,500 2,000 2,000 9,124 15,624 75,000 2,000 250 10,000 2,000 24,000 38,250 2,500 2,000 2,000 13,688 20,188 100,000 2,300 250 10,000 2,000 24,000 38,550 2,500 2,000 2,000 18,245 24,745 200,000 2,300 250 15,000 2,000 24,000 43,550 2,500 2,000 2,000 36,490 42,990 300,000 2,300 250 15,000 2,000 24,000 43,550 2,500 2,000 2,000 54,736 61,236 Capital investment ($) Fluoridation pump 500 Pipework 250 Tank and bunded area 5,000 Installation 2,000 Monitoring equipment 24,000 Total investment 31,750 Annual costs ($) Operating costs Maintenance costs Monitoring costs Hydrofluosilicic acid Total annual costs 2,500 2,000 2,000 183 6,683 Notes: 1. 375 litres of water per day, per person. 2. Hydrofluosilicic acid (HFA) (liquid) used, rather than the alternatives of sodium silicofluoride or sodium fluoride. HFA costs about $200 per tonne. 3. Fluoridated at 1 g of fluoride per cubic metre of water. HFA has 15% available fluoride. 4. Labour costed at $50 per hour. 5. Equipment lifetime – 15 years. 6. Housed with other water treatment equipment (i.e. no additional building costs). 7. Only one injection site. 8. No planning costs included. Finally, the cost of a simple restoration is required. The average General Dental Benefit fee paid nationally for a single surface amalgam for an adolescent is currently $24; the Benefit for an extraction is $40. For adults, dental charges vary widely. A survey gave the average target hourly rate for dentists in March 1998 as $257.15 Insertion of a single surface amalgam filling typically takes about 15 minutes. One quarter of the target rate, increased by 3% to allow for inflation into 1999 dollars, gives $66. This figure was used for the averted cost of a single surface amalgam filling for those aged 18 and above. Costs of fluoridating Information on the costs of establishing and operating a fluoridation plant was obtained by consulting equipment providers and operators of fluoridation systems. Estimates of capital and annual operating costs for a range of population sizes are presented in Table 1. For this analysis, the fluoridation system used hydrofluosilicic acid, containing 15% available fluoride. Fluoridation systems based on hydrofluosilicic acid serve a larger proportion of the New Zealand population than the alternatives. Fluoridation of a supply was assumed to occur at a single point Figure 1: Fluoridation costs – economies of scale. 70,000 60,000 Annual costs 50,000 Capital costs Costs ($s) 40,000 30,000 20,000 10,000 Population ('000s) AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2001 VOL. 25 NO. 2 Health Promotion Cost-effectiveness of fluoridating water supplies with one delivery pump. Once operating satisfactorily, fluoridation pumps should require minimal maintenance. Labour costs are based on $50 per hour. The major operating cost is for hydrofluosilicic acid, at about $200 per tonne. A national population-weighted average water use of 375 litres per person per day was calculated using flow data on all New Zealand water supplies. A requirement of 1 gram of fluoride per cubic metre of water (to achieve 1 ppm of fluoride) gave an annual cost of hydrofluosilicic acid per person of about 18 cents. Water supplies with higher natural levels of fluoride would require less added hydrofluosilicic acid. Economies of scale are shown in Figure 1. Capital investment is largely independent of the population served. Annual costs are dominated by hydrofluosilicic acid, and rise linearly with population increase. It was assumed that equipment would be replaced halfway through the 30-year time frame. these data exclude decayed but untreated teeth, they may underestimate dental caries incidence. The base case results (see Table 3) show that, for populations from 1,000 to 300,000, the net cost of fluoridation is negative, since treatment savings exceed fluoridation costs. Figure 2 shows net costs saved as a function of population size. The relationship is essentially linear above a population of about 5,000, when water fluoridation costs become insignificant. The net costs saved are then approximately equal to the dental cost savings, which are proportional to population size. The financial break-even point occurs for a population between 800 and 900 people. Sensitivity analyses Maori communities Results The results of the base case analysis are followed by results of three sensitivity analyses. Base case Table 2 summarises the dental data for deciduous and permanent teeth. For both deciduous and permanent teeth, children living in fluoridated areas had fewer fillings than children living in non-fluoridated areas, at every age from 4 to 13 years. The difference for Maori is approximately twice that for non-Maori. Since Table 4 gives economic analysis results for communities that are 100% Maori. While there are few, if any, 100% Maori communities, and certainly none with large populations, comparison of Table 4 with Table 3 shows a rapidly diverging difference. This can also be seen in Figure 3. Fluoridation of the water supply of a Maori community of 1,000 would generate a net cost saving of over $300,000 compared with the net cost saving of $17,000 for the base case community of 1000, including 15% Maori. Discount rate Table 5 shows results for a discount rate of 10%. The net cost is negative for populations 5,000 and above, although about half that for a discount rate of 5% (see Table 3). Figure 3 also shows Table 2: Mean numbers of averted decayed surfaces in children from fluoridated and non-fluoridated areas. A. Non-Maori Deciduous teeth surfaces Fluoridated Not fluoridated Age Restorations Extractions Restorations Extractions Averted deciduous teeth costs ($s) Permanent teeth surfaces Fluoridated Restorations Not fluoridated Restorations Averted decayed surfaces in permanent teeth B. Maori Deciduous teeth surfaces Fluoridated Not fluoridated Age Restorations Extractions Restorations Extractions Averted deciduous teeth costs ($s) Permanent teeth surfaces Fluoridated Restorations Not fluoridated Restorations Averted decayed surfaces in permanent teeth Note: Costs associated with decay of a surface in a deciduous tooth – $24 for restoration, $50 for extraction. 2001 VOL. 25 NO. 2 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Wright et al. Article Table 3: Base case results: fluoridation from 2000-2030 (15% Maori, discount rate of 5%, societal perspective, one injection site). Population Fluoridation costs (dollars) 156,000 168,000 183,000 228,000 312,000 387,000 462,000 768,000 1,070,000 Dental cost savings (dollars) 174,000 870,000 1,740,000 4,350,000 8,700,000 13,000,000 17,400,000 34,800,000 52,200,000 Net cost (dollars) -17,500 -701,000 -1,560,000 -4,120,000 -8,390,000 -12,700,000 -16,900,000 -34,000,000 -51,100,000 Averted decay (permanent tooth surfaces) 1,480 7,420 14,800 37,100 74,200 111,000 148,000 297,000 445,000 Cost-effectiveness (dollars per averted decayed surface) N/A N/A N/A N/A N/A N/A N/A N/A N/A 1,000 5,000 10,000 25,000 50,000 75,000 100,000 200,000 300,000 Notes: 1. All costs and averted decay are discounted back to present value in 2000. 2. Averted decay is measured as averted decayed surfaces in permanent teeth. 3. Estimates of costs and averted decay are rounded to three significant figures. Table 4: Sensitivity analysis: Maori Communities. Population Fluoridation costs (dollars) 156,000 168,000 183,000 228,000 312,000 387,000 462,000 768,000 1,070,000 Dental cost savings (dollars) 484,000 2,420,000 4,840,000 12,100,000 24,200,000 36,300,000 48,400,000 96,700,000 145,000,000 Net cost (dollars) -327,000 -2,250,000 -4,650,000 -11,900,000 -23,900,000 -35,900,000 -47,900,000 -96,000,000 -144,000,000 Averted decay (permanent tooth surfaces) 4,310 21,600 43,100 108,000 216,000 323,000 431,000 862,000 1,290,000 Cost-effectiveness (dollars per averted decayed surface) N/A N/A N/A N/A N/A N/A N/A N/A N/A 1,000 5,000 10,000 25,000 50,000 75,000 100,000 200,000 300,000 Notes: 1. All costs and averted decay are discounted back to present value in the year 2000. 2. Averted decay is measured as averted decayed surfaces in permanent teeth. 3. Estimates of costs and averted decay are rounded to three significant figures. 4. There are, of course, no totally Maori large communities, but we present these results so as to be consistent. Table 5: Sensitivity analysis: discount rate = 10%. Population Fluoridation costs (dollars) 109,000 117,000 126,000 155,000 210,000 258,000 306,000 502,000 692,000 Dental cost savings (dollars) 82,300 412,000 823,000 2,060,000 4,120,000 6,170,000 8,230,000 16,500,000 24,700,000 Net cost (dollars) 26,700 -295,000 -697,000 -1,900,000 -3,900,000 -5,910,000 -7,920,000 -16,000,000 -24,000,000 Averted decay (permanent tooth surfaces) 837 4,180 8,370 20,900 41,800 62,800 83,700 167,000 251,000 Cost-effectiveness (dollars per averted decayed surface) 32 N/A N/A N/A N/A N/A N/A N/A N/A 1,000 5,000 10,000 25,000 50,000 75,000 100,000 200,000 300,000 Notes: 1. All costs and averted decay are discounted back to present value in 2000. 2. Averted decay is measured as averted decayed surfaces in permanent teeth. 3. Estimates of costs and averted decay are rounded to three significant figures. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2001 VOL. 25 NO. 2 Health Promotion Cost-effectiveness of fluoridating water supplies Averted Costs minus Fluoridation Costs ($'000s) Figure 2: Base case – net costs saved. 60,000 50,000 40,000 30,000 20,000 10,000 Population ('000s) the comparison. Even for a population of 1,000, fluoridation is worthwhile if the non-monetised benefits of an averted decayed surface are considered to be greater than its net cost of $32. Number of fluoride injection sites The number of injection sites was varied from one to five. All costs, except that of the hydrofluosilicic acid, were increased in direct proportion to the number of sites. For populations of 10,000 or more, net cost savings are little affected by the number of injection sites because the fluoridation cost stream is dominated by the cost of the hydrofluosilicic acid (see Table 6). For a population of 1,000, there is a positive net cost associated with more than one injection site, and the cost-effectiveness ratio ranges from $92 per averted decayed surface for two sites to $402 for five sites. Discussion This analysis shows fluoridation is cost-saving for communities of 1,000 people and above. The major underlying assumptions concern adult dental benefits. In the absence of data for older people, we truncated the effectiveness of fluoridation at age 34. Since the benefits of fluoridation extend over the whole of life, this limitation would have resulted in an under-estimation of the benefits. In using data from Washington State to assess benefits in younger adults, we probably understated the case for fluoridation, for two reasons: First, the adults in the US sample were all employees (or their spouses) of the State. As such, unlike the situation in New Zealand, they would all have had dental insurance, and the condition of their teeth would probably be better than that of the average Table 6: Sensitivity analysis: varying number of injection sites. Population 1 Net cost (dollars) 1,000 5,000 10,000 25,000 50,000 75,000 100,000 200,000 300,000 -17,500 -701,000 -1,560,000 -4,120,000 -8,390,000 -12,700,000 -16,900,000 -34,000,000 -51,100,000 Number of injection sites 2 3 Net cost (dollars) Net cost (dollars) 136,000 -548,000 -1,400,000 -3,970,000 -8,220,000 -12,500,000 -16,800,000 -33,900,000 -50,900,000 289,000 -395,000 -1,250,000 -3,810,000 -8,060,000 -12,300,000 -16,600,000 -33,700,000 -50,800,000 4 Net cost (dollars) 443,000 -241,000 -1,100,000 -3,660,000 -7,900,000 -12,200,000 -16,400,000 -33,500,000 -50,600,000 5 Net cost (dollars) 596,000 -87,600 -942,000 -3,510,000 -7,730,000 -12,000,000 -16,300,000 -33,300,000 -50,400,000 Notes: 1. All costs are discounted back to present value in the year 2000. 2. Estimates of costs are rounded to three significant figures. 2001 VOL. 25 NO. 2 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Wright et al. Article Averted Costs minus Fluoridation Costs ($'000s) Figure 3: Sensitivity analysis compared with the base case – net costs saved. 50,000 40,000 Maori 30,000 20,000 Base case 10,000 Discount at 10% 0 0 -10,000 20 40 60 80 100 120 Population ('000s) New Zealander. Our data show that the effectiveness of fluoridation is higher for lower socio-economic groups (see Appendix). This confirms results found elsewhere.16,17 Second, we have assumed the same effectiveness for teenagers (14 to 19) as for young adults (20 to 34). Our data show average annual effectiveness at age 13 to be 0.66 filled surfaces averted. It is unlikely that a year older, it has dropped to 0.29. The assumption that all averted dental treatments are simple restorations would also under-estimate averted costs in the teenage and young adult years. Many restorations involve approximal surfaces, which are more costly to restore and require more frequent repair than single surface amalgam fillings. There is an implicit assumption that costs of fluoridation and costs of dental restorations experience the same rates of inflation from 2000 to 2030. We have no information on whether the costs of water fluoridation are rising faster than general inflation. However, dentistry costs are reported to have increased by 9%, up to 25% ahead of inflation, in the past four years.18 Again, the bias in our analysis would be towards understatement of the economic case for fluoridation. The assumption of an unchanging 15% Maori proportion biases our results towards under-estimation of the benefits of fluoridation. On the other hand, the conservative assumption that the current differential between the effectiveness of fluoridation for Maori and non-Maori will not narrow over time biases our results towards over-estimation of the benefits of fluoridation. The resulting net direction of the bias is unclear. There is a greater rate of extractions of deciduous teeth in fluoridated areas than in non-fluoridated areas in our data. This anomaly, based on a small number of extractions, is probably due to differences in extraction practices between the two main areas for which data were obtained. Not all deciduous tooth extractions are performed because of dental caries. Many extractions are to hasten the natural exfoliation process for teeth that have become loose. Dental providers vary in their willingness to undertake these elective extractions. Since extractions of deciduous teeth were more common in fluoridated than in non-fluoridated areas, this contributes to an understatement of the case for fluoridation. There are also uncertainties in the costing of fluoridation. Reasons why the costs of fluoridating water supply systems may differ from those in this analysis include: • A different set-up of fluoridation equipment. • A different water flow rate. • Appreciable naturally-occurring fluoride in the water. • Multiple fluoride injection points. Potentially, the last of these would have the greatest effect. Our sensitivity analyses showed the break-even point for five injection sites was for a community of about 10,000 people. Fortunately, it is larger communities that are likely to require multiple fluoride injection sites. We also carried out additional sensitivity analyses (not shown in Results) to check whether lowering the cost of an adult amalgam filling ($66) by 20%, or by using the lower confidence bound (0.19) for the number of decayed and filled surfaces reduced annually by fluoridation in adults, taken from a study of Washington State employees. The break-even points for these two sensitivity analyses were populations of 1,071 and 1,130, respectively. These differ little from the results of our main analysis. The reason is that most of the effectiveness occurs before age 18. Is it cost-effective to fluoridate communities with less than about 1,000 people? First, as shown, our analysis is conservative and probably under-estimates the cost-effectiveness of fluoridation. Irrespective of this, if the cost-effectiveness ratio of fluoridation for a community is $X per averted decayed surface, it is a value 2001 VOL. 25 NO. 2 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Health Promotion Cost-effectiveness of fluoridating water supplies judgement as to whether the non-monetised benefits of an averted decayed surface are worth more than its net cost of $X. If not, then the intervention would not be considered cost-effective. This cost-effectiveness analysis has particular strengths. It uses data reflecting recent decay rate reductions in both fluoridated and non-fluoridated areas, and the narrowing of the differences between these areas. We are aware of no other cost-effectiveness analysis that has done this. Dental treatment savings are included as a negative cost. Other cost-effectiveness analyses have been based solely on the cost of fluoridating water supplies.19 Further strengths of the analysis are the substantial time horizon (30 years), and the incorporation of estimated benefits to adults (to age 45). Other cost-effectiveness analyses have had shorter time horizons and been based on benefits to children only. We conclude that the installation of water fluoridation facilities for presently non-fluoridated supplies remains very cost-effective. Where a community has a substantial proportion of Maori, a low socio-economic status, or a high proportion of children, the case is particularly strong. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Australian and New Zealand Journal of Public Health Wiley

The cost‐effectiveness of fluoridating water supplies in New Zealand

Loading next page...
 
/lp/wiley/the-cost-effectiveness-of-fluoridating-water-supplies-in-new-zealand-QreX1dLWPX

References (22)

Publisher
Wiley
Copyright
2001 The Public Health Association of Australia Inc
ISSN
1326-0200
eISSN
1753-6405
DOI
10.1111/j.1753-6405.2001.tb01841.x
Publisher site
See Article on Publisher Site

Abstract

Objective: Tooth decay has been substantially reduced in New Zealand, and the difference in caries rates between fluoridated and non-fluoridated areas has narrowed. We investigated whether it is cost-effective to fluoridate water supplies that are now non-fluoridated. Janice C. Wright Health economist, Wellington, New Zealand Michael N. Bates Institute of Environmental Science and Research Ltd, New Zealand Terry Cutress Wellington School of Medicine, New Zealand Methods: The net cost of fluoridation was based on the cost of fluoridating a water supply minus the averted costs of treating decay. A range of population sizes was considered. The main analysis was conducted from a societal perspective, using a real discount rate of 5%. Fluoridation was assumed to occur continuously between the years 2000 and 2030. Other assumptions were a Maori population proportion of 15%, no new decay after age 34, and no further dental cost savings after age 45. Information on averted decay in 4 to 12 year old New Zealand children (29,000 receiving fluoridated water and 47,000 receiving nonfluoridated water) was available; information on averted decay in adults was obtained from a study in the United States. Sensitivity analyses investigated the effects of varying the Maori population proportion, the discount rate, and the number of fluoride injection sites. Martin Lee School and Community Dental Service, Canterbury District Health Board, New Zealand Results: Fluoridation was cost-saving (dental cost savings exceeded fluoridation costs) for communities above about a thousand people. The true break-even community size may be lower. For smaller communities, fluoridation may be considered cost-effective depending on the non-monetised value assigned to an averted decayed surface. Conclusions and implications: Fluoridation remains very cost-effective, and is particularly so for communities with high proportions of children, Maori, or people of low socio-economic status. (Aust N Z J Public Health 2001; 25: 170-8) luoridation of water supplies began in New Zealand in 1954. By 1999, fluoridated supplies served about 57% of the population (Personal communication: Alan Ferguson, ESR). Over about the past two decades, tooth decay in deciduous and permanent teeth of New Zealand children has dramatically reduced. The difference in dental caries rates between fluoridated and non-fluoridated areas has narrowed.1 The caries reduction is attributable to widespread use of fluoride-containing toothpaste and water from fluoridated supplies. The ‘halo effect’, where foods and beverages from fluoridated areas are consumed in non-fluoridated areas, has probably also played a role. The general reduction in caries raises the question of whether it is economic to fluoridate presently non-fluoridated New Zealand water supplies, or to replace existing fluoridation equipment. This study addressed this issue. It assumes that optimally fluoridated water supplies do not cause adverse health effects.2-5 A key aim of this cost-effectiveness analysis (CEA) was to estimate the minimum population for which oral health benefits from the introduction of fluoridation to a water supply would be greater than water treatment costs. However, population size is only one factor that influences the cost-effectiveness of this intervention. The sensitivity of the cost-effectiveness of fluoridation to two other ‘community’ factors – the number of fluoride injection sites and the demographic composition of the population served by the fluoridated supply – is explored in this paper. Methods Economic methodology A CEA differs from a full cost-benefit analysis because benefits are measured in appropriate ‘natural’ units, rather than in dollars. In this study, one averted newly decayed tooth surface in a permanent tooth was the unit used to measure the benefits (effectiveness) of water fluoridation. A societal perspective, taking account of all benefits and costs wherever they fall, was adopted. Thus, the net cost of fluoridation is the cost of fluoridating the water supply minus the averted dental costs. Nonmonetised benefits, such as reduced pain and improved social interactions, are represented by the proxy of averted decayed surfaces. Therefore, cost-effectiveness ratios are measured in units of dollars per averted decayed surface. Submitted: June 2000 Revision requested: January 2001 Accepted: March 2001 Correspondence to: Dr Michael Bates, School of Public Health, 140 Warren Hall, University of California, Berkeley, CA 94720, United States. Fax: +1 510 843 5539; e-mail: m_bates@uclink.berkeley.edu AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2001 VOL. 25 NO. 2 Health Promotion Cost-effectiveness of fluoridating water supplies Fluoridation of a water supply incurs a stream of costs and provides streams of averted decay and associated dental cost savings stretching into the future. Both net costs and effectiveness (number of averted decayed surfaces) were expressed in terms of their present value (PV), using a real discount rate of 5%. A real discount rate reflects the decision maker’s preference for present benefits and costs over future benefits and costs, without the need to adjust for inflation.6 All costs were measured in 1999 dollars. Data and assumptions Four sets of data were required – demography (community size and age structure), averted decay, cost savings from averted decay, and the costs of fluoridation. Demography Five assumptions were used in constructing the demographic component of the model. 1. Continuous fluoridation of a supply between the years 2000 and 2030, with one fluoridation plant replacement. Discounting renders insignificant costs and benefits more than 30 years into the future. 2. No new averted decay after age 34. This and the next assumption were a consequence of the source of adult dental data (see below). 3. No dental cost savings after age 45. 4. From 2000 to 2030 there is no mortality in the birth cohorts receiving fluoridated water. 5. Out-migration in the cohort is exactly counterbalanced by inmigration. The first three assumptions would cause under-estimation of fluoridation benefits, and the fourth over-estimation of benefits. The third assumption would be dominant, as dental care becomes increasingly complex and expensive with age. The benefits depend on the year of birth. Those born in 2000 or after would benefit from fluoridation from their birth year to 2030; those aged 15 or less in the year 2000 would benefit from 2000 to 2030; and those aged over 15 in the year 2000 would benefit until age 45. Population growth from 2000 to 2030 was taken as the ‘average’ projection from the 1996 New Zealand Census.7 We assumed fluoridated communities maintained the same projected age and ethnic structure as the total NZ population in the year 2000 – 15% Maori, 85% non-Maori. The percentage of Maori in the total population is actually projected to increase to 19% in 2030. Estimates of averted decay Estimates of averted decay were taken from two sources. For children aged from 4 to 13, New Zealand data were available. For 14 to 34 year-olds, results of a study from the United States were used. Ages 4 to 13 years: In New Zealand, dental treatment is publicly funded until age 17. Although there are few routinely collected data for adolescents, data are collected on the dental status of 5 to 13-year-old children by school dental service providers. Wellington and Canterbury providers use the same software for 2001 VOL. 25 NO. 2 recording dental data for each child. Whether the child lives in a fluoridated area is also recorded. All records for 1996, for children aged 4 to 13, were extracted from the Wellington and Canterbury databases. The resulting dataset contained 29,097 children receiving fluoridated water and 46,825 receiving non-fluoridated water. Data on treatment for deciduous teeth were restricted to ages 4 to 10 and, for permanent teeth, to ages 6 to 13. Data for extractions of deciduous teeth, but not permanent teeth, were available. Permanent tooth extractions are not usually performed by School Dental Services. An extraction was counted as equivalent to five decayed tooth surfaces. Our natural unit of effectiveness, one averted decayed surface in a permanent tooth, does not capture the benefits of averted decay in deciduous teeth. This understates the case for fluoridation, although the value of averted decay in deciduous teeth is less than for permanent teeth because deciduous teeth are exfoliated by late childhood. However, the averted cost of filling deciduous teeth was incorporated into the analysis. Overall, then, averted decay in deciduous teeth is taken into account in the calculation of costs, but not in the estimation of effectiveness. Ages 14 to 34 years: Because the effect of fluoridation on teeth is largely topical, it benefits people of all ages.8-12 However, there are no New Zealand data on adult dental benefits associated with fluoridation. Therefore, for those aged 14 and above, data from a study in the United States were used.13 The study population was 10,628 Washington State employees and their spouses, aged 20 to 34. After controlling for various factors, including other fluoride exposure sources, the study found that each year of fluoridation exposure reduced decayed and filled surfaces by 0.29 surfaces (95% CI 0.19-0.39 surfaces). We used this estimate of 0.29 decayed and filled surfaces averted annually for non-Maori aged 14 through 34, on the assumption that the benefit accrues in the year of exposure. We increased the 0.29 surfaces to 0.59 surfaces for effectiveness in Maori teenagers and young adults, by scaling it by the ratio of the Maori to non-Maori effectiveness at age 13. Thus, for the base-case analysis, in which the Maori population proportion was 15%, average annual effectiveness was 0.33 averted decayed surfaces. Cost savings from averted decay Using only the averted cost of the initial filling greatly underestimates the lifetime flow of costs triggered by the initial decay of a tooth surface. More than 60% of all dental restorative work involves replacement of restorations.14 Continued replacement often entails an escalation in complexity and treatment costs. An averted decayed surface, rather than an averted decayed tooth, was used as the unit for the effectiveness of fluoridation, since it is a more sensitive measure. This excludes downstream averted dental costs, like crowns and root fillings. We assumed: • All restorations are single surface amalgams. • Amalgam restorations require replacement every eight years. • No restorations after age 45. This is a consequence of the time horizon of the analysis and the assumption of no averted decay after age 34. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Wright et al. Article Table 1: Costs of establishing and operating a water fluoridation plant. Population served 1,000 5,000 500 250 5,000 2,000 24,000 31,750 2,500 2,000 2,000 913 7,413 10,000 500 250 5,000 2,000 24,000 31,750 2,500 2,000 2,000 1,825 8,325 25,000 500 250 5,000 2,000 24,000 31,750 2,500 2,000 2,000 4,562 11,062 50,000 2,000 250 10,000 2,000 24,000 38,250 2,500 2,000 2,000 9,124 15,624 75,000 2,000 250 10,000 2,000 24,000 38,250 2,500 2,000 2,000 13,688 20,188 100,000 2,300 250 10,000 2,000 24,000 38,550 2,500 2,000 2,000 18,245 24,745 200,000 2,300 250 15,000 2,000 24,000 43,550 2,500 2,000 2,000 36,490 42,990 300,000 2,300 250 15,000 2,000 24,000 43,550 2,500 2,000 2,000 54,736 61,236 Capital investment ($) Fluoridation pump 500 Pipework 250 Tank and bunded area 5,000 Installation 2,000 Monitoring equipment 24,000 Total investment 31,750 Annual costs ($) Operating costs Maintenance costs Monitoring costs Hydrofluosilicic acid Total annual costs 2,500 2,000 2,000 183 6,683 Notes: 1. 375 litres of water per day, per person. 2. Hydrofluosilicic acid (HFA) (liquid) used, rather than the alternatives of sodium silicofluoride or sodium fluoride. HFA costs about $200 per tonne. 3. Fluoridated at 1 g of fluoride per cubic metre of water. HFA has 15% available fluoride. 4. Labour costed at $50 per hour. 5. Equipment lifetime – 15 years. 6. Housed with other water treatment equipment (i.e. no additional building costs). 7. Only one injection site. 8. No planning costs included. Finally, the cost of a simple restoration is required. The average General Dental Benefit fee paid nationally for a single surface amalgam for an adolescent is currently $24; the Benefit for an extraction is $40. For adults, dental charges vary widely. A survey gave the average target hourly rate for dentists in March 1998 as $257.15 Insertion of a single surface amalgam filling typically takes about 15 minutes. One quarter of the target rate, increased by 3% to allow for inflation into 1999 dollars, gives $66. This figure was used for the averted cost of a single surface amalgam filling for those aged 18 and above. Costs of fluoridating Information on the costs of establishing and operating a fluoridation plant was obtained by consulting equipment providers and operators of fluoridation systems. Estimates of capital and annual operating costs for a range of population sizes are presented in Table 1. For this analysis, the fluoridation system used hydrofluosilicic acid, containing 15% available fluoride. Fluoridation systems based on hydrofluosilicic acid serve a larger proportion of the New Zealand population than the alternatives. Fluoridation of a supply was assumed to occur at a single point Figure 1: Fluoridation costs – economies of scale. 70,000 60,000 Annual costs 50,000 Capital costs Costs ($s) 40,000 30,000 20,000 10,000 Population ('000s) AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2001 VOL. 25 NO. 2 Health Promotion Cost-effectiveness of fluoridating water supplies with one delivery pump. Once operating satisfactorily, fluoridation pumps should require minimal maintenance. Labour costs are based on $50 per hour. The major operating cost is for hydrofluosilicic acid, at about $200 per tonne. A national population-weighted average water use of 375 litres per person per day was calculated using flow data on all New Zealand water supplies. A requirement of 1 gram of fluoride per cubic metre of water (to achieve 1 ppm of fluoride) gave an annual cost of hydrofluosilicic acid per person of about 18 cents. Water supplies with higher natural levels of fluoride would require less added hydrofluosilicic acid. Economies of scale are shown in Figure 1. Capital investment is largely independent of the population served. Annual costs are dominated by hydrofluosilicic acid, and rise linearly with population increase. It was assumed that equipment would be replaced halfway through the 30-year time frame. these data exclude decayed but untreated teeth, they may underestimate dental caries incidence. The base case results (see Table 3) show that, for populations from 1,000 to 300,000, the net cost of fluoridation is negative, since treatment savings exceed fluoridation costs. Figure 2 shows net costs saved as a function of population size. The relationship is essentially linear above a population of about 5,000, when water fluoridation costs become insignificant. The net costs saved are then approximately equal to the dental cost savings, which are proportional to population size. The financial break-even point occurs for a population between 800 and 900 people. Sensitivity analyses Maori communities Results The results of the base case analysis are followed by results of three sensitivity analyses. Base case Table 2 summarises the dental data for deciduous and permanent teeth. For both deciduous and permanent teeth, children living in fluoridated areas had fewer fillings than children living in non-fluoridated areas, at every age from 4 to 13 years. The difference for Maori is approximately twice that for non-Maori. Since Table 4 gives economic analysis results for communities that are 100% Maori. While there are few, if any, 100% Maori communities, and certainly none with large populations, comparison of Table 4 with Table 3 shows a rapidly diverging difference. This can also be seen in Figure 3. Fluoridation of the water supply of a Maori community of 1,000 would generate a net cost saving of over $300,000 compared with the net cost saving of $17,000 for the base case community of 1000, including 15% Maori. Discount rate Table 5 shows results for a discount rate of 10%. The net cost is negative for populations 5,000 and above, although about half that for a discount rate of 5% (see Table 3). Figure 3 also shows Table 2: Mean numbers of averted decayed surfaces in children from fluoridated and non-fluoridated areas. A. Non-Maori Deciduous teeth surfaces Fluoridated Not fluoridated Age Restorations Extractions Restorations Extractions Averted deciduous teeth costs ($s) Permanent teeth surfaces Fluoridated Restorations Not fluoridated Restorations Averted decayed surfaces in permanent teeth B. Maori Deciduous teeth surfaces Fluoridated Not fluoridated Age Restorations Extractions Restorations Extractions Averted deciduous teeth costs ($s) Permanent teeth surfaces Fluoridated Restorations Not fluoridated Restorations Averted decayed surfaces in permanent teeth Note: Costs associated with decay of a surface in a deciduous tooth – $24 for restoration, $50 for extraction. 2001 VOL. 25 NO. 2 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Wright et al. Article Table 3: Base case results: fluoridation from 2000-2030 (15% Maori, discount rate of 5%, societal perspective, one injection site). Population Fluoridation costs (dollars) 156,000 168,000 183,000 228,000 312,000 387,000 462,000 768,000 1,070,000 Dental cost savings (dollars) 174,000 870,000 1,740,000 4,350,000 8,700,000 13,000,000 17,400,000 34,800,000 52,200,000 Net cost (dollars) -17,500 -701,000 -1,560,000 -4,120,000 -8,390,000 -12,700,000 -16,900,000 -34,000,000 -51,100,000 Averted decay (permanent tooth surfaces) 1,480 7,420 14,800 37,100 74,200 111,000 148,000 297,000 445,000 Cost-effectiveness (dollars per averted decayed surface) N/A N/A N/A N/A N/A N/A N/A N/A N/A 1,000 5,000 10,000 25,000 50,000 75,000 100,000 200,000 300,000 Notes: 1. All costs and averted decay are discounted back to present value in 2000. 2. Averted decay is measured as averted decayed surfaces in permanent teeth. 3. Estimates of costs and averted decay are rounded to three significant figures. Table 4: Sensitivity analysis: Maori Communities. Population Fluoridation costs (dollars) 156,000 168,000 183,000 228,000 312,000 387,000 462,000 768,000 1,070,000 Dental cost savings (dollars) 484,000 2,420,000 4,840,000 12,100,000 24,200,000 36,300,000 48,400,000 96,700,000 145,000,000 Net cost (dollars) -327,000 -2,250,000 -4,650,000 -11,900,000 -23,900,000 -35,900,000 -47,900,000 -96,000,000 -144,000,000 Averted decay (permanent tooth surfaces) 4,310 21,600 43,100 108,000 216,000 323,000 431,000 862,000 1,290,000 Cost-effectiveness (dollars per averted decayed surface) N/A N/A N/A N/A N/A N/A N/A N/A N/A 1,000 5,000 10,000 25,000 50,000 75,000 100,000 200,000 300,000 Notes: 1. All costs and averted decay are discounted back to present value in the year 2000. 2. Averted decay is measured as averted decayed surfaces in permanent teeth. 3. Estimates of costs and averted decay are rounded to three significant figures. 4. There are, of course, no totally Maori large communities, but we present these results so as to be consistent. Table 5: Sensitivity analysis: discount rate = 10%. Population Fluoridation costs (dollars) 109,000 117,000 126,000 155,000 210,000 258,000 306,000 502,000 692,000 Dental cost savings (dollars) 82,300 412,000 823,000 2,060,000 4,120,000 6,170,000 8,230,000 16,500,000 24,700,000 Net cost (dollars) 26,700 -295,000 -697,000 -1,900,000 -3,900,000 -5,910,000 -7,920,000 -16,000,000 -24,000,000 Averted decay (permanent tooth surfaces) 837 4,180 8,370 20,900 41,800 62,800 83,700 167,000 251,000 Cost-effectiveness (dollars per averted decayed surface) 32 N/A N/A N/A N/A N/A N/A N/A N/A 1,000 5,000 10,000 25,000 50,000 75,000 100,000 200,000 300,000 Notes: 1. All costs and averted decay are discounted back to present value in 2000. 2. Averted decay is measured as averted decayed surfaces in permanent teeth. 3. Estimates of costs and averted decay are rounded to three significant figures. AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH 2001 VOL. 25 NO. 2 Health Promotion Cost-effectiveness of fluoridating water supplies Averted Costs minus Fluoridation Costs ($'000s) Figure 2: Base case – net costs saved. 60,000 50,000 40,000 30,000 20,000 10,000 Population ('000s) the comparison. Even for a population of 1,000, fluoridation is worthwhile if the non-monetised benefits of an averted decayed surface are considered to be greater than its net cost of $32. Number of fluoride injection sites The number of injection sites was varied from one to five. All costs, except that of the hydrofluosilicic acid, were increased in direct proportion to the number of sites. For populations of 10,000 or more, net cost savings are little affected by the number of injection sites because the fluoridation cost stream is dominated by the cost of the hydrofluosilicic acid (see Table 6). For a population of 1,000, there is a positive net cost associated with more than one injection site, and the cost-effectiveness ratio ranges from $92 per averted decayed surface for two sites to $402 for five sites. Discussion This analysis shows fluoridation is cost-saving for communities of 1,000 people and above. The major underlying assumptions concern adult dental benefits. In the absence of data for older people, we truncated the effectiveness of fluoridation at age 34. Since the benefits of fluoridation extend over the whole of life, this limitation would have resulted in an under-estimation of the benefits. In using data from Washington State to assess benefits in younger adults, we probably understated the case for fluoridation, for two reasons: First, the adults in the US sample were all employees (or their spouses) of the State. As such, unlike the situation in New Zealand, they would all have had dental insurance, and the condition of their teeth would probably be better than that of the average Table 6: Sensitivity analysis: varying number of injection sites. Population 1 Net cost (dollars) 1,000 5,000 10,000 25,000 50,000 75,000 100,000 200,000 300,000 -17,500 -701,000 -1,560,000 -4,120,000 -8,390,000 -12,700,000 -16,900,000 -34,000,000 -51,100,000 Number of injection sites 2 3 Net cost (dollars) Net cost (dollars) 136,000 -548,000 -1,400,000 -3,970,000 -8,220,000 -12,500,000 -16,800,000 -33,900,000 -50,900,000 289,000 -395,000 -1,250,000 -3,810,000 -8,060,000 -12,300,000 -16,600,000 -33,700,000 -50,800,000 4 Net cost (dollars) 443,000 -241,000 -1,100,000 -3,660,000 -7,900,000 -12,200,000 -16,400,000 -33,500,000 -50,600,000 5 Net cost (dollars) 596,000 -87,600 -942,000 -3,510,000 -7,730,000 -12,000,000 -16,300,000 -33,300,000 -50,400,000 Notes: 1. All costs are discounted back to present value in the year 2000. 2. Estimates of costs are rounded to three significant figures. 2001 VOL. 25 NO. 2 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Wright et al. Article Averted Costs minus Fluoridation Costs ($'000s) Figure 3: Sensitivity analysis compared with the base case – net costs saved. 50,000 40,000 Maori 30,000 20,000 Base case 10,000 Discount at 10% 0 0 -10,000 20 40 60 80 100 120 Population ('000s) New Zealander. Our data show that the effectiveness of fluoridation is higher for lower socio-economic groups (see Appendix). This confirms results found elsewhere.16,17 Second, we have assumed the same effectiveness for teenagers (14 to 19) as for young adults (20 to 34). Our data show average annual effectiveness at age 13 to be 0.66 filled surfaces averted. It is unlikely that a year older, it has dropped to 0.29. The assumption that all averted dental treatments are simple restorations would also under-estimate averted costs in the teenage and young adult years. Many restorations involve approximal surfaces, which are more costly to restore and require more frequent repair than single surface amalgam fillings. There is an implicit assumption that costs of fluoridation and costs of dental restorations experience the same rates of inflation from 2000 to 2030. We have no information on whether the costs of water fluoridation are rising faster than general inflation. However, dentistry costs are reported to have increased by 9%, up to 25% ahead of inflation, in the past four years.18 Again, the bias in our analysis would be towards understatement of the economic case for fluoridation. The assumption of an unchanging 15% Maori proportion biases our results towards under-estimation of the benefits of fluoridation. On the other hand, the conservative assumption that the current differential between the effectiveness of fluoridation for Maori and non-Maori will not narrow over time biases our results towards over-estimation of the benefits of fluoridation. The resulting net direction of the bias is unclear. There is a greater rate of extractions of deciduous teeth in fluoridated areas than in non-fluoridated areas in our data. This anomaly, based on a small number of extractions, is probably due to differences in extraction practices between the two main areas for which data were obtained. Not all deciduous tooth extractions are performed because of dental caries. Many extractions are to hasten the natural exfoliation process for teeth that have become loose. Dental providers vary in their willingness to undertake these elective extractions. Since extractions of deciduous teeth were more common in fluoridated than in non-fluoridated areas, this contributes to an understatement of the case for fluoridation. There are also uncertainties in the costing of fluoridation. Reasons why the costs of fluoridating water supply systems may differ from those in this analysis include: • A different set-up of fluoridation equipment. • A different water flow rate. • Appreciable naturally-occurring fluoride in the water. • Multiple fluoride injection points. Potentially, the last of these would have the greatest effect. Our sensitivity analyses showed the break-even point for five injection sites was for a community of about 10,000 people. Fortunately, it is larger communities that are likely to require multiple fluoride injection sites. We also carried out additional sensitivity analyses (not shown in Results) to check whether lowering the cost of an adult amalgam filling ($66) by 20%, or by using the lower confidence bound (0.19) for the number of decayed and filled surfaces reduced annually by fluoridation in adults, taken from a study of Washington State employees. The break-even points for these two sensitivity analyses were populations of 1,071 and 1,130, respectively. These differ little from the results of our main analysis. The reason is that most of the effectiveness occurs before age 18. Is it cost-effective to fluoridate communities with less than about 1,000 people? First, as shown, our analysis is conservative and probably under-estimates the cost-effectiveness of fluoridation. Irrespective of this, if the cost-effectiveness ratio of fluoridation for a community is $X per averted decayed surface, it is a value 2001 VOL. 25 NO. 2 AUSTRALIAN AND NEW ZEALAND JOURNAL OF PUBLIC HEALTH Health Promotion Cost-effectiveness of fluoridating water supplies judgement as to whether the non-monetised benefits of an averted decayed surface are worth more than its net cost of $X. If not, then the intervention would not be considered cost-effective. This cost-effectiveness analysis has particular strengths. It uses data reflecting recent decay rate reductions in both fluoridated and non-fluoridated areas, and the narrowing of the differences between these areas. We are aware of no other cost-effectiveness analysis that has done this. Dental treatment savings are included as a negative cost. Other cost-effectiveness analyses have been based solely on the cost of fluoridating water supplies.19 Further strengths of the analysis are the substantial time horizon (30 years), and the incorporation of estimated benefits to adults (to age 45). Other cost-effectiveness analyses have had shorter time horizons and been based on benefits to children only. We conclude that the installation of water fluoridation facilities for presently non-fluoridated supplies remains very cost-effective. Where a community has a substantial proportion of Maori, a low socio-economic status, or a high proportion of children, the case is particularly strong.

Journal

Australian and New Zealand Journal of Public HealthWiley

Published: Apr 1, 2001

There are no references for this article.